EUROPEAN COMMISSION

Integrated Pollution Prevention and Control

Reference Document on
Best Available Techniques in the Smitheries and
Foundries Industry
May 2005
This document is one of a series of reference documents foreseen to be produced in the EIPPCB
work plan as below (at the time of writing, not all the documents have been drafted):

Full title BREF code
Reference Document on Best Available Techniques for Intensive Rearing of Poultry and Pigs ILF
Reference Document on the General Principles of Monitoring MON
Reference Document on Best Available Techniques for the Tanning of Hides and Skins TAN
Reference Document on Best Available Techniques in the Glass Manufacturing Industry GLS
Reference Document on Best Available Techniques in the Pulp and Paper Industry PP
Reference Document on Best Available Techniques on the Production of Iron and Steel I&S
Reference Document on Best Available Techniques in the Cement and Lime Manufacturing Industries CL
Reference Document on the Application of Best Available Techniques to Industrial Cooling Systems CV
Reference Document on Best Available Techniques in the Chlor Alkali Manufacturing Industry CAK
Reference Document on Best Available Techniques in the Ferrous Metals Processing Industry FMP
Reference Document on Best Available Techniques in the Non Ferrous Metals Industries NFM
Reference Document on Best Available Techniques for the Textiles Industry TXT
Reference Document on Best Available Techniques for Mineral Oil and Gas Refineries REF
Reference Document on Best Available Techniques in the Large Volume Organic Chemical Industry LVOC
Reference Document on Best Available Techniques in the Waste Water and Waste Gas
Treatment/Management Systems in the Chemical Sector
CWW
Reference Document on Best Available Techniques in the Food, Drink and Milk Industry FM
Reference Document on Best Available Techniques in the Smitheries and Foundries Industry SF
Reference Document on Best Available Techniques on Emissions from Storage ESB
Reference Document on Best Available Techniques on Economics and Cross-Media Effects ECM
Reference Document on Best Available Techniques for Large Combustion Plants LCP
Reference Document on Best Available Techniques in the Slaughterhouses and Animals By-products
Industries
SA
Reference Document on Best Available Techniques for Management of Tailings and Waste-Rock in
Mining Activities
MTWR
Reference Document on Best Available Techniques for the Surface Treatment of Metals STM
Reference Document on Best Available Techniques for the Waste Treatments Industries WT
Reference Document on Best Available Techniques for the Manufacture of Large Volume Inorganic
Chemicals (Ammonia, Acids and Fertilisers)
LVIC-AAF
Reference Document on Best Available Techniques for Waste Incineration WI
Reference Document on Best Available Techniques for Manufacture of Polymers POL
Reference Document on Energy Efficiency Techniques ENE
Reference Document on Best Available Techniques for the Manufacture of Organic Fine Chemicals OFC
Reference Document on Best Available Techniques for the Manufacture of Specialty Inorganic
Chemicals
SIC
Reference Document on Best Available Techniques for Surface Treatment Using Solvents STS
Reference Document on Best Available Techniques for the Manufacture of Large Volume Inorganic
Chemicals (Solids and Others)
LVIC-S
Reference Document on Best Available Techniques in Ceramic Manufacturing Industry CER
Executive Summary
Smitheries and Foundries Industry i
EXECUTIVE SUMMARY

The Smitheries and Foundries BREF (Best Available Techniques reference document) reflects
an information exchange carried out under Article 16(2) of Council Directive 96/61/EC. This
executive summary is intended to be read in conjunction with the BREFs Preface, which
explains the structure of the document, its objectives, usage and legal terms. The executive
summary describes the main findings, the principal BAT conclusions and the associated
emission/consumption levels. It can be read and understood as a stand-alone document but, as a
summary, it does not present all the complexities of the full BREF text. It is therefore not
intended as a substitute for the full BREF text in BAT decision making.

Scope of this BREF

This document reflects the exchange of information on the activities covered by Annex I,
categories 2.3 (b), 2.4 and 2.5 (b) of the IPPC Directive, i.e.

2.3. Installations for the processing of ferrous metals:
(b) smitheries with hammers the energy of which exceeds 50 kilojoule per hammer, where
the calorific power used exceeds 20 MW
2.4. Ferrous metal foundries with a production capacity exceeding 20 tonnes per day
2.5. Installations
(b) for the smelting, including the alloyage, of non-ferrous metals, including recovered
products, (refining, foundry casting, etc.) with a melting capacity exceeding 4 tonnes
per day for lead and cadmium or 20 tonnes per day for all other metals.

After comparing the above descriptions to the actual capacitites of existing installations in
Europe, the TWG outlined a working scope, which covered the following:

- the casting of ferrous materials, e.g. lamellar cast iron, malleable and nodular iron, steel
- the casting of non-ferrous materials, e.g. aluminium, magnesium, copper, zinc, lead and
their alloys.

Smitheries were excluded from this documents scope since no European smitheries were
reported which met the conditions stated in Annex I 2.3.(b). This document therefore only
discusses foundry processes. Cadmium, titanium and precious metals foundries, as well as bell
casting and art casting foundries were also excluded on capacity grounds. Continuous casting
(into sheets and slabs) has already been covered in the BREF documents related to iron and
steel production and non-ferrous metal industries, and therefore, it is not dealt with in this
document. In covering non-ferrous metals in this document, the process is considered to start
with the melting of ingots and internal scrap or with liquid metal.

From a process point of view, the following foundry process steps are covered in this document:

- pattern making
- raw materials storage and handling
- melting and metal treatment
- mould and core production, and moulding techniques
- casting or pouring and cooling
- shake-out
- finishing
- heat treatment.

Executive Summary
ii Smitheries and Foundries Industry
The foundry industry

Foundries melt ferrous and non-ferrous metals and alloys and reshape them into products at or
near their finished shape through the pouring and solidification of the molten metal or alloy into
a mould. The foundry industry is a differentiated and diverse industry. It consists of a wide
range of installations, from small to very large; each with a combination of technologies and
unit operations selected to suit the input, size of series and types of product produced by the
specific installation. The organisation within the sector is based on the type of metal input, with
the main distinction being made between ferrous and non-ferrous foundries. Since castings in
general are semi-finished products, foundries are located close to their customers.

The European foundry industry is the third largest in the world for ferrous castings and second
largest for non-ferrous. The annual production of castings in the enlarged European Union
amounts to 11.7 million tonnes of ferrous and 2.8 million tonnes of non-ferrous castings.
Germany, France and Italy are the top three production countries in Europe, with a total annual
production of over two million tonnes of castings each. In recent years, Spain has taken over the
fourth position from Great Britain, with both having a production of over one million tonnes of
castings. Together, the top five countries produce more than 80 % of the total European
production. Although the production volume has remained relatively stable over the past few
years, there has been a decline in the total number of foundries (now totalling around 3000
units), which is also reflected in the employment numbers (now totalling around 260000
people). This can be explained by progressive upscaling and automation in the foundry units.
However, the foundry industry is predominantly still an SME industry, with 80 % of companies
employing less than 250 people.

The main markets served by the foundry industry are the automotive (50 % of market share),
general engineering (30 %) and construction (10 %) sectors. A growing shift of the automotive
industry towards lighter vehicles has been reflected in a growth in the market for aluminium and
magnesium castings. While iron castings mostly (i.e. >60 %) go to the automotive sector, steel
castings find their market in the construction, machinery and valve making industries.

The foundry process

A general flow chart of the foundry process is depicted in the figure below. The process can be
divided into the following major activities:
- melting and metal treatment: the melting shop
- preparation of moulds and cores: the moulding shop
- casting of the molten metal into the mould, cooling for solidification and removing the
casting from the mould: the casting shop
- finishing of the raw casting: the finishing shop.

Various process options can be taken, depending on the type of metal, size of series and type of
product. Generally, the main division within the sector is based on the type of metal (ferrous or
non-ferrous) and the type of moulding used (lost moulds or permanent moulds). While any
combination is possible, typically ferrous foundries largely use lost moulds (i.e. sand moulding)
and non-ferrous foundries mainly use permanent moulds (i.e. die-casting). Within each of these
basic process options a variety of techniques exist according to the type of furnace used, the
moulding and core-making system (green sand or various chemical binders) applied, and the
casting system and finishing techniques applied. Each of those have their own technical,
economic and environmental properties, advantages and disadvantages.

Chapters 2, 3 and 4 of this document follow a process-flow approach to describe the various
operations, from pattern making to finishing and heat treatment. Applied techniques are
described, emission and consumption levels given and techniques to minimise the
environmental impact are discussed. The structure of Chapter 5 is based on a distinction
between the type of metal and the type of moulding.
Executive Summary
Smitheries and Foundries Industry iii
Melting
- ferrous:
cupola
induction
electric arc
rotary
- non-ferrous:
induction
shaft
crucible
reverberatory
Metal treatment
Permanent
mould
maintenance
Lost mould making
- sand mould
- sand core
- inserts
- pattern
wood, plastic,
metal
- lost model
resin, wax
Manual moulding
Moulding automate
Casting
- gravitational pouring - high-pressure
- tilt pouring - centrifugal
- low-pressure - continuous
Cooling
Shake-out / Take-out
Sand preparation
Sand regeneration
Finishing
-removal of casting system
-shot blasting
-deburring
-thermal treatment
Raw materials, chemicals, patterns, dies
Finished casting
Sand Legend:
The foundry process

Key environmental issues

The foundry industry is a major player in the recycling of metals. Steel, cast iron and aluminium
scrap is remelted into new products. Most possible negative environmental effects of foundries
are related to the presence of a thermal process and the use of mineral additives. Environmental
effects therefore are mainly related to the exhaust and off-gases and to the re-use or disposal of
mineral residues.

Emissions to air are the key environmental concern. The foundry process generates (metal-
laden) mineral dusts, acidifying compounds, products of incomplete combustion and volatile
organic carbons. Dust is a major issue, since it is generated in all process steps, in varying types
and compositions. Dust is emitted from metal melting, sand moulding, casting and finishing.
Any dust generated may contain metal and metal oxides.

Executive Summary
iv Smitheries and Foundries Industry
The use of cokes as fuels, or the heating of crucibles and furnaces with gas or oil-fired burners
can cause emissions of combustion products, such as NO
x
and SO
2
. Additionally, the use of
cokes and the presence of impurities (e.g. oil, paint, ) in scrap can cause the production of
some products of incomplete combustion or recombination (such as PCDD/F) and dust.

In the making of moulds and cores, various additives are used to bind the sand. In binding the
sand and pouring the metal, reaction and decomposition products are generated. These include
inorganic and organic (e.g. amines, VOCs) compounds. The generation of decomposition
products (mainly VOCs) further continues during the casting cooling and de-moulding
operations. These products can also cause an odour nuisance.

In the foundry process, emissions to air will typically not be limited to one (or several) fixed
point(s). The process involves various emission sources (e.g. from hot castings, sand, hot
metal). A key issue in emission reduction is not only to treat the exhaust and off-gas flow, but
also to capture it.

Sand moulding involves the use of large sand volumes, with sand-to-liquid-metal weight ratios
generally ranging from 1:1 up to 20:1. The used sand can be regenerated, re-used or disposed of.
Additional mineral residues such as slag and dross are generated in the melting stage when
removing impurities from the melt. These can be either re-used or disposed of.

Since foundries deal with a thermal process, energy efficiency and management of the generated
heat are important environmental aspects. However, due to the high amount of transport and
handling of the heat carrier (i.e. the metal) and its slow cooling, the recovery of heat is not
always straightforward.

Foundries may have a high water consumption e.g. for cooling and quenching operations. In
most foundries, water management involves an internal circulation of water, with a major part
of the water evaporating. The water is generally used in the cooling systems of electric furnaces
(induction or arc) or cupola furnaces. In general, the final volume of waste water is very small.
Nevertheless, when wet dedusting techniques are used, the generated waste water requires
special attention. In (high) pressure die-casting, a waste water stream is formed, which needs
treatment to remove organic (phenol, oil) compounds before its disposal.

Consumption and emission levels

A general overview of the inputs and outputs of the foundry process is given in the figure
below. The Casting step mentioned in the centre of the picture covers also all necessary
moulding operations. The major input streams are metal, energy, binders and water. The key
emissions are dust, amines and VOCs, and for specific furnace types also SO
2
, dioxins and NO
x
.
The melting stage uses 40 60 % of the energy input. For a certain metal type, the energy use is
dependent on the type of furnace used. The melting energy input ranges from 500 to
1200 kWh/t metal charge for ferrous metals and from 400 to 1200 kWh/t metal charge for
aluminium.

The amounts and types of binders, chemicals and sand used are very dependent on the type of
casting made, especially regarding its size and shape, as well as whether serial or batch wise
production is used.

Water consumption is largely dependent on the type of furnace used, the type of flue-gas
cleaning applied and the casting method applied.

Dust is generated in each of the process steps, albeit with different levels of mineral oxides,
metals and metal oxides. Dust levels for metal melting range from below detection limit, for
certain non-ferrous metals, to above 10 kg/tonne, for the cupola melting of cast iron. The high
Executive Summary
Smitheries and Foundries Industry v
amount of sand used in lost mould casting results in dust emissions during the various moulding
stages.

Amines are used as a catalyst in the most commonly used core-making system. This results in
guided emissions from the core-shooting machines and diffuse emissions from core-handling.

Volatile organic compounds emissions (mainly solvents, BTEX and to a lesser extent phenol,
formaldehyde, etc.) result from the use of e.g. resins, organic solvents, or organic-based
coatings in moulding and core-making. Organic compounds are thermally decomposed during
metal pouring and are emitted further during shake-out and cooling. Emission levels between
0.1 1.5 kg/tonne of casting are presented in this document.

Foundry Process
Melting - Casting - Finishing
Metal
scrap/ingots
Castings
Energy
- fuel
- electricity
- heat
Water
- cooling
- scrubbing
Sand and Chemicals
- additives melting and metal treatment
- binders/release agents
- oils and lubricants
Noise
Water
- organics
- metal dust
Solid
- dust
- used sand
- dross
- slag
Energy
- heat
- steam
- hot water
Air
- combustion and reaction products
- particulate matter
- pyrolysis and evaporation products
Odour
Mass stream overview of the foundry process

Techniques to consider in the determination of BAT

The minimisation of emissions, efficient raw material and energy usage, optimum process
chemical utilisation, the recovering and recycling of waste and the substitution of harmful
substances are all important principles of the IPPC Directive. For foundries the focal points are
air emissions, the efficient use of raw materials and energy, and waste reduction, in conjunction
with any recycling and re-use options.

The environmental issues as mentioned above are addressed by using a variety of process-
integrated and end-of-pipe techniques. Over 100 techniques for pollution prevention and control
are presented in this document, ordered under the following 12 thematic headings, which are
largely based on the process flow:
Executive Summary
vi Smitheries and Foundries Industry
1. Raw material storage and handling: Materials storage and handling techniques aim at the
prevention of soil and water pollution and optimisation of the internal recycling of scrap
metal.
2. Metal melting and molten metal treatment: For each furnace type, different techniques may
be considered for optimisation of the furnace efficiency and minimisation of any residue
production. These mainly involve in-process measures. Environmental considerations can
also be taken into account in the selection of the furnace type. Special attention is paid to
cleaning of the aluminium melt and the melting of magnesium, due to the high pollution
potential of the products used until recently (HCE and SF
6
).
3. Mould- and core-making, including sand preparation: Best practice measures and
techniques for minimising the consumption may be applied for each type of binder system
and for die-casting release agents. For the reduction of VOCs and odour emissions from lost
mould systems, water-based coatings and inorganic solvents may be considered. While
water-based coatings are commonly used, the applicability of inorganic solvents in core-
making is still limited. Another approach is the use of different moulding methods.
However, those techniques are used only in specific fields of application.
4. Metal casting: In order to improve the efficiency of the casting process, measures aimed at
increasing the metal yield (i.e. the mass ratio of molten metal over finished casting) may be
considered.
5. Fume, flue-gas and exhaust air capture and treatment: Dealing with the emissions to air in
all the different foundry stages requires an adequate capture and treatment system to be in
place. According to the unit operation, various techniques may be considered, depending on
the type of compounds emitted, the off-gas volume and the ease of capture. Techniques
applied for off-gas capture play an important role in the reduction of fugitive emissions.
Additionally, for fugitive emissions, good practice measures may be considered.
6. Waste water prevention and treatment: In many cases, waste water can be prevented or
minimised by taking in-process measures. Waste water that cannot be prevented, will
contain mineral or metal dust, amines, suphates, oil or lubricants, depending on its source
within the process. The applicable treatment techniques differ for each of these compounds.
7. Energy efficiency: Metal melting consumes 40 60 % of the energy input of a foundry.
Energy efficiency measures should therefore take both the melting and the other processes
(e.g. air compression, plant actuation, hydraulics) into account. The need for furnace and
off-gas cooling generates a hot water or hot air stream, which may allow an internal or
external utilisation of the heat.
8. Sand: regeneration, recycling, re-use and disposal: Since foundries make intensive use of
sand as an inert primary material, the regeneration or re-use of this sand is an important
point of consideration as part of its environmental performance. Various techniques are
applied for regeneration of the sand (i.e. treatment and internal re-use as moulding sand),
the selection of which depends on the binder type and the sand flow compostion. If sand is
not regenerated, then external re-use may be considered in order to prevent the need for its
disposal. Its application in various areas has been demonstrated.
9. Dust and solid residues: treatment and re-use: In-process techniques and operational
measures may be considered for the minimisation of dust and residues. The collected dusts,
slags and other solid residues may find an internal or external re-use.
10. Noise reduction: Various foundry activities are point sources of noise. For foundries near
housing, this may cause a nuisance to the neighbours. The setting up and implementation of
a noise reduction plan, covering both general and source-specific measures, may therefore
be considered.
11. Decommissioning: The IPPC Directive requests attention be paid to consideration of the
possible pollution upon decommissioning of the plant. Foundries present a specific risk for
soil pollution at this stage. There are a number of general measures, which apply more
widely than just to foundries, which may be considered to prevent pollution at the
decommissioning stage.
12. Environmental management tools: Environmental management systems are a useful tool to
aid the prevention of pollution from industrial activities in general. Their presentation is
therefore a standard part of each BREF document.

Executive Summary
Smitheries and Foundries Industry vii
BAT for foundries

The BAT chapter (Chapter 5) identifies those techniques that the TWG considered to be BAT in
a general sense for the foundry industry, based upon the information in Chapter 4 and taking
account of the Article 2(11) definition of best available techniques and the considerations
listed in Annex IV to the Directive. The BAT chapter does not set or propose emission limit
values but suggests emission levels that are associated with the use of BAT.

During the information exchange by the TWG, many issues were raised and discussed. A
selection of these are highlighted in this summary.The following paragraphs summarise the key
BAT conclusions relating to the most relevant environmental issues.

The BAT elements will need to be adapted to the foundry type. A foundry basically consists of
a melting shop and a casting shop, both with their own supply chain. For lost mould casting this
supply chain includes all activities related to moulding and core-making. In the BAT chapter, a
distinction is made between the melting of either ferrous or non-ferrous metal, and for casting in
either lost or permanent moulds. Each foundry may be classified as a combination of a
particular melting with an associated moulding class. BAT is presented for each class. Generic
BAT, which are common to all foundries, are also presented.

Generic BAT
Some BAT elements are generic and apply for all foundries, regardless of the processes they
utilise and the type of products they produce. These concern material flows, finishing of
castings, noise, waste water, environmental management and decommissioning.

BAT is to optimise the management and control of internal flows, in order to prevent pollution,
prevent deterioration, provide adequate input quality, allow recycling and re-use and to improve
the process efficiency. The BREF refers to storage and handling techniques covered in the
Storage BREF, but also adds some foundry specific BAT related to storage and handling, such
as the storage of scrap on an impermeable surface with a drainage and collection system
(although applying a roof can reduce the need for such a system), the separate storage of
incoming materials and residues, the use of recyclable containers, optimisation of the metal
yield, and good practice measures for molten metal transfer and ladle handling.

BAT are given for finishing techniques that generate dust and for heat treatment techniques. For
abrasive cutting, shot blasting and fettling, BAT is to collect and treat the finishing off-gas using
a wet or dry system. For heat treatment, BAT is to use clean fuels (i.e. natural gas or low-level
sulphur content fuel), automated furnace operation and burner/heater control and also to capture
and evacuate the exhaust gas from heat treatment furnaces.

Concerning noise reduction, BAT is to develop and implement a noise reduction strategy, with
general and source-specific measures, being applicable, such as using enclosure systems for
high-noise unit operations such as shake-out and using additional measures depending on and
according to local conditions.

BAT for waste water management includes prevention, separation of the waste water types,
maximising internal recycling and applying an adequate treatment for each final flow. This
includes techniques utilising e.g. oil interceptors, filtration or sedimentation.

Fugitive emissions arise from uncontained sources (transfers, storage, spills) and the incomplete
evacuation of contained sources. BAT is to apply a combination of measures concerning
material handling and transport and to optimise exhaust gas capture and cleaning through one or
more capture techniques. Preference is given to collection of the fume nearest to the source.

BAT is to implement and adhere to an Environmental Management System (EMS) that
incorporates, as appropriate to individual circumstances, features concerning e.g. the
commitment of top management, planning, establishing and implementing procedures,
performance checking with corrective actions and reviews.
Executive Summary
viii Smitheries and Foundries Industry
BAT is to apply all necessary measures to prevent pollution upon decommissioning. These
include minimisation of the risks during the design stage, implementation of an improvement
programme for existing installations and development and utilisation of a site closure plan for
new and existing installations. In these measures, at least the following process parts are
considered: tanks, vessels, pipework, insulation, lagoons and landfills.

Ferrous metal melting
For the operation of cupola furnaces, BAT includes techniques which can give increased
efficiency, such as divided blast operation, oxygen enrichment, continuous blowing or long
campaign operation, good melting practice measures, and control of the coke quality. BAT is to
collect, cool and dedust the off-gas, and to apply post combustion and heat recovery under
specific conditions. Several dedusting systems are BAT but wet dedusting is preferable when
melting with basic slag and in some cases as one of the measures to prevent and minimise
dioxin and furan emissions. Industry has expressed doubts on the implementation of secondary
measures for dioxin and furan abatement that have only been proven in other sectors and in
particular questions their applicability for smaller foundries. For cupolas, BAT for residue
management includes minimising slag forming, pretreating the slags in order to allow their
external re-use and collecting and recycling coke breeze.

For the operation of electric arc furnaces, BAT includes applying reliable and efficient process
controls to shorten the melting and treatment time, using the foamy slag practice, efficiently
capturing the furnace off-gas, cooling the furnace off-gas and dedusting using a bag filter. BAT
is to recycle the filter dust into the EAF furnace.

For the operation of induction furnaces, BAT is to melt clean scrap; use good practice measures
for the charging and operation; use medium frequency power and, when installing a new
furnace, to change any mains frequency furnace to medium frequency; to evaluate the
possibility of waste heat recuperation and under specific conditions to implement a heat
recovery system. For exhaust capture and treatment from induction furnaces, BAT is to use a
hood, lip extraction or cover extraction on each induction furnace to capture the furnace off-gas
and maximise the off-gas collection during the full working cycle; to use dry flue-gas cleaning;
and to keep dust emissions below 0.2 kg/tonne molten iron.

For the operation of rotary furnaces, BAT is to implement a combination of measures to
optimise furnace yield and to use an oxyburner. BAT is to collect the off-gas close to the
furnace exit, apply post combustion, cool it using a heat-exchanger and then to apply dry
dedusting. For the prevention and minimisation of dioxin and furan emissions, BAT is to use a
combination of specified measures. Similarly to the situation with cupola furnaces, industry has
expressed doubts on the implementation of secondary measures for dioxin and for an abatement
that have only been proven in other sectors and in particular questions their applicability for
smaller foundries.

The actual metal treatment applied depends on the type of product made. BAT is to collect the
exhaust gas from AOD converters using a roof canopy and to collect and treat the off-gas from
nodularisation, using a bag filter. BAT is also to make the MgO-dust available for recycling.

Non-ferrous metal melting
For the operation of induction furnaces for melting aluminium, copper, lead and zinc, BAT is to
use good practice measures for the charging and operation; to use medium frequency power
and, when installing a new furnace, change any mains frequency furnace to medium frequency;
to evaluate the possibility of waste heat recuperation and under specific conditions to implement
a heat recovery system. For exhaust capture from these furnaces, BAT is to minimise emissions
and if needed to collect the off-gas, maximising the off-gas collection during the full working
cycle and applying dry flue-gas cleaning.

For the other furnace types, BAT mainly focuses on the efficient collection of furnace off-gas
and/or the reduction of fugitive emissions.
Executive Summary
Smitheries and Foundries Industry ix
For non-ferrous metal treatment, BAT is to use an impeller station for the degassing and
cleaning of aluminium. BAT is to use SO
2
as a covering gas for magnesium melting in
installations with an annual output of 500 tonnes and more. For small plants (<500 tonnes Mg
parts output/year) BAT is to use SO
2
or to minimise the use of SF
6
. In the case where SF
6
is
used, the BAT associated consumption level is <0.9 kg/tonne casting for sand casting and <1.5
kg/tonne casting for pressure die-casting.

Lost mould casting
Lost mould casting involves moulding, core-making, pouring, cooling and shake-out. This
includes the production of green sand or chemically-bonded sand moulds and chemically-
bonded sand cores. BAT elements are presented in three categories: green sand moulding,
chemical sand moulding and pouring/cooling/shake-out.

For green sand preparation, BAT items deal with exhaust capture and cleaning and the internal
or external recycling of the captured dust. In line with the goal of minimising waste for disposal,
BAT is to apply a primary regeneration of green sand. Regeneration ratios of 98 % (monosand)
or 90 94 % (green sand with incompatible cores) are associated with the use of BAT.

For chemically-bonded sand, the proposed BAT cover a variety of techniques, and deal with a
broad range of environmental issues. BAT is to minimise the binder and resin consumption and
sand losses, to minimise fugitive VOC emissions by capturing the exhaust gas from core-
making and core handling, and to use water-based coatings. The use of alcohol-based coatings
is BAT in a limited number of applications, where water-based coatings cannot be applied. In
this case, the exhaust should be captured at the coating stand, whenever this is feasible. A
specific BAT is given for amine-hardened urethane-bonded (i.e. cold-box) core preparation, to
minimise amine emissions and optimise amine recovery. For these systems both aromatic and
non-aromatic solvents are BAT. BAT is to minimise the amount of sand going to disposal,
primarily by adopting a strategy of regeneration and/or re-use of chemically-bonded sand (as
mixed or monosand). In the case of regeneration, the BAT conditions are given in the table
below. Regenerated sand is re-used only in compatible sand systems.

Sand type Technique Regeneration ratio
1
(%)
Cold setting monosand Simple mechanical regeneration 75 80
Silicate monosand Heating and pneumatic treatment 45 85
Monosands of cold-box, SO
2
,
hot-box, croning
Mixed organic sands
Cold mechanical or thermal regeneration in cores: 40 100
in moulds: 90 100
Mixed green and organic sand Mechanical-thermal-mechanical
treatment, grinding or pneumatic chafing
in cores: 40 100
in moulds: 90 100
(1) mass of regenerated sand/total mass of sand used
BAT for chemically-bonded sand regeneration (mixed and monosand)

Alternative moulding methods and inorganic binders are considered to have a promising
potential for minimisation of the environmental impacts of moulding and casting processes.
Pouring, cooling and shake-out generate emissions of dust, VOCs and other organic products.
BAT is to enclose pouring and cooling lines and provide exhaust extraction, for serial pouring
lines, and to enclose the shake-out equipment, and treat the exhaust gas using wet or dry
dedusting.

Executive Summary
x Smitheries and Foundries Industry
Permanent mould casting
Due to the different nature of the process, the environmental issues for permanent mould casting
require a different focus than those for lost mould techniques, with water as a more prominent
item. Emissions to air are in the form of an oil mist, rather than the dust and combustion
products encountered in the other processes. BAT therefore focusses on prevention measures
involving minimisation of the water and release agent consumption. BAT is to collect and treat
run-off water and leakage water, using oil interceptors and distillation, vacuum evaporation or
biological degradation. If oil mist prevention measures do not allow a foundry to reach the BAT
associated emission level, BAT is to use hooding and electrostatic precipitation for the exhaust
of HPDC machines.

BAT for chemically-bonded sand preparation is analogous to the elements mentioned for lost
mould casting. BAT for used sand management is to enclose the de-coring unit and to treat the
exhaust gas using wet or dry dedusting. If a local market exists, BAT is to make the sand from
de-coring available for recycling.

BAT associated emission levels
The following emission levels are associated to the BAT measures stated above.

Activity Type Parameter Emission level
(mg/Nm)
Finishing of castings Dust 5 20
Ferrous metal melting General Dust
(1)
5 20
PCDD/PCDF 0.1 ng TEQ/Nm
3
Hot Blast Cupola CO 20 1000
SO
2
20 100
NO
x
10 200
Cold Blast Cupola SO
2
100 400
NO
x
20 70
NM - VOC 10 20
Cokeless Cupola NO
x
160 400
Electric Arc Furnace NO
x
10 50
CO 200
Rotary Furnace SO
2
70 130
NO
x
50 250
CO 20 30
Non-ferrous metal melting General Dust 1 20
Aluminium melting Chlorine 3
Shaft furnace for Al SO
2
30 50
NO
x
120
CO 150
VOC 100 150
Hearth type furnace for Al SO
2
15
NO
x
50
CO 5
TOC 5
Moulding and casting using
lost moulds
General Dust 5 20
Core shop Amine 5
Regeneration units SO
2
120
NO
x
150
Permanent mould casting General Dust 5 - 20
Oil mist,
measured as
total C
5 - 10
(1) the emission level of dust depends on the dust components, such as heavy metals, dioxins, and its mass
flow.
Emissions to air associated with the use of BAT for the various foundry activities

Executive Summary
Smitheries and Foundries Industry xi
All associated emission levels are quoted as an average over the practicable measuring period.
Whenever continuous monitoring is practicable, a daily average value is used. Emissions to air
are based on standard conditions, i.e. 273 K, 101.3 kPa and dry gas.

While the BAT reference documents do not set legally binding standards, they are meant to give
information for the guidance of industry, Member States and the public on achievable emission
and consumption levels when using specified techniques. The appropriate limit values for any
specific case will need to be determined taking into account the objectives of the IPPC Directive
and the local considerations.

Emerging techniques

Some new techniques for minimisation of the environmental impacts are currently still in the
research and development phase or are only just beginning to enter the market these are
considered to be emerging techniques. Five of these techniques are discussed in Chapter 6,
namely: the use of low combustible materials in cupola melting, the recycling of metal-bearing
filter dust, amine recovery by waste-gas permeation, the separate spraying of release agent and
water in aluminium die-casting, and inorganic binder material for core-making. The latter
technique was especially pointed out by the TWG as promising, although the current limited
scale of testing and implementation does not allow it to be yet incorporated as a technique to
consider in the selection of BAT.

Concluding remarks on the exchange of information

Information exchange
The BREF document is based on more than 250 sources of information. Foundry research
institutes provided an important share of this information and played an active role in the
information exchange. Local BAT notes from various Member States gave the information
exchange a firm basis. The majority of the documents provided in the information exchange
dealt with processes and techniques as applied in ferrous foundries. Throughout the writing of
the BREF the non-ferrous foundry processes have been underrepresented. This is reflected in a
lower level of detail in the BAT conclusions for non-ferrous foundries.

Level of consensus
A good general level of consensus was reached on the conclusions and no split views were
recorded. The industry representation added a comment, expressing their doubt on the ease of
implementation of secondary measures for dioxin abatement.

Recommendations for future work
The information exchange and the result of this exchange, i.e. this document, present an
important step forward in achieving the integrated prevention and control of pollution from the
foundry industry. Future work could continue this purpose by focusing on the collection and
assessment of information that was not provided during this information exchange. In particular,
in future work should cover in more detail the following topics:

- Techniques for VOC abatement: There is a need for data and information on methods
applied for the efficient capture and treatment of VOC-laden exhaust gases from foundries.
The use of alternative binder and coating materials may prove to be an important prevention
measure in this respect
- Waste water treatment: There is a need for data from a broad range of water treatment
systems in foundries, this should also show emission levels in relation to the inputs and
treatment techniques applied

Executive Summary
xii Smitheries and Foundries Industry
- Melting of non-ferrous metals: Emission data for non-ferrous foundries are presented in this
document only for some specific installations. There is a need for more complete
information on both guided and fugitive emissions from non-ferrous metal melting in
foundries. This should be based on operational practice and expressed both as emission
levels and mass flows..
- Economic data for BAT techniques: There is a lack of economic information for many of the
techniques presented in Chapter 4. This information needs to be collected from projects
dealing with the implementation of the presented techniques.

Suggested topics for R&D projects
The information exchange has also exposed some areas where additional useful knowledge
could be gained from research and development projects. These relate to the following subjects:

- Dioxin monitoring and abatement: There is a need for a better understanding of the
influence of process parameters on the formation of dioxins. This requires the monitoring of
dioxin emissions for various installations and under varying conditions. Additionally, there
is a need for research on the use and effectiveness of secondary measures for dioxin
abatement in the foundry industry
- Mercury emissions: The high volatility of mercury may cause gaseous emissions, which are
not related to dust. In view of the implementation of a European policy on mercury
emissions, there is a need for research into the emissions of mercury from melting processes
in general and from (non-ferrous) foundries in particular
- Oxygas burners and their use in cupola furnaces: The TWG reported that new applications
have been set-up as a result of ongoing research. There is scope for further research and
development here, to bring this technique up to a development level that allows its further
dissemination.
- Alternative replacement gases for SF
6
in magnesium melting: Alternative cover gases to
replace SF
6
such as HFC-134a and Novec612 have been developed and successfully tested
but have not found industrial implementation. These gases may present an alternative for
the substitution by SO
2
. There is a need for research and demonstration projects that provide
knowledge on the protective properties, decomposition behaviour and emission pattern of
these compounds. This should allow a clearer indication of the applicability and may
support industrial implementation.

The EC is launching and supporting, through its RTD programmes, a series of projects dealing
with clean technologies, emerging effluent treatment and recycling technologies and
management strategies. Potentially these projects could provide a useful contribution to future
BREF reviews. Readers are therefore invited to inform the EIPPCB of any research results
which are relevant to the scope of this document (see also the preface of this document).

Preface
Smitheries and Foundries Industry xiii
PREFACE

1. Status of this document
Unless otherwise stated, references to the Directive in this document means the Council
Directive 96/61/EC on integrated pollution prevention and control. As the Directive applies
without prejudice to Community provisions on health and safety at the workplace, so does this
document.

This document forms part of a series presenting the results of an exchange of information
between EU Member States and industries concerned on best available technique (BAT),
associated monitoring, and developments in them. It is published by the European Commission
pursuant to Article 16(2) of the Directive, and must therefore be taken into account in
accordance with Annex IV of the Directive when determining best available techniques.

2. Relevant legal obligations of the IPPC Directive and the definition of BAT

In order to help the reader understand the legal context in which this document has been drafted,
some of the most relevant provisions of the IPPC Directive, including the definition of the term
best available techniques, are described in this preface. This description is inevitably
incomplete and is given for information only. It has no legal value and does not in any way alter
or prejudice the actual provisions of the Directive.

The purpose of the Directive is to achieve integrated prevention and control of pollution arising
from the activities listed in its Annex I, leading to a high level of protection of the environment
as a whole. The legal basis of the Directive relates to environmental protection. Its
implementation should also take account of other Community objectives such as the
competitiveness of the Communitys industry thereby contributing to sustainable development.

More specifically, it provides for a permitting system for certain categories of industrial
installations requiring both operators and regulators to take an integrated, overall look at the
polluting and consuming potential of the installation. The overall aim of such an integrated
approach must be to improve the management and control of industrial processes so as to ensure
a high level of protection for the environment as a whole. Central to this approach is the general
principle given in Article 3 that operators should take all appropriate preventative measures
against pollution, in particular through the application of best available techniques enabling
them to improve their environmental performance.

The term best available techniques is defined in Article 2(11) of the Directive as the most
effective and advanced stage in the development of activities and their methods of operation
which indicate the practical suitability of particular techniques for providing in principle the
basis for emission limit values designed to prevent and, where that is not practicable, generally
to reduce emissions and the impact on the environment as a whole. Article 2(11) goes on to
clarify further this definition as follows:

techniques includes both the technology used and the way in which the installation is
designed, built, maintained, operated and decommissioned;

available techniques are those developed on a scale which allows implementation in the
relevant industrial sector, under economically and technically viable conditions, taking into
consideration the costs and advantages, whether or not the techniques are used or produced
inside the Member State in question, as long as they are reasonably accessible to the operator;

best means most effective in achieving a high general level of protection of the environment
as a whole.

Preface
xiv Smitheries and Foundries Industry
Furthermore, Annex IV of the Directive contains a list of considerations to be taken into
account generally or in specific cases when determining best available techniques bearing in
mind the likely costs and benefits of a measure and the principles of precaution and prevention.
These considerations include the information published by the Commission pursuant to
Article 16(2).

Competent authorities responsible for issuing permits are required to take account of the general
principles set out in Article 3 when determining the conditions of the permit. These conditions
must include emission limit values, supplemented or replaced where appropriate by equivalent
parameters or technical measures. According to Article 9(4) of the Directive, these emission
limit values, equivalent parameters and technical measures must, without prejudice to
compliance with environmental quality standards, be based on the best available techniques,
without prescribing the use of any technique or specific technology, but taking into account the
technical characteristics of the installation concerned, its geographical location and the local
environmental conditions. In all circumstances, the conditions of the permit must include
provisions on the minimisation of long-distance or transboundary pollution and must ensure a
high level of protection for the environment as a whole.

Member States have the obligation, according to Article 11 of the Directive, to ensure that
competent authorities follow or are informed of developments in best available techniques.

3. Objective of this Document

Article 16(2) of the Directive requires the Commission to organise an exchange of information
between Member States and the industries concerned on best available techniques, associated
monitoring and developments in them, and to publish the results of the exchange.

The purpose of the information exchange is given in recital 25 of the Directive, which states that
the development and exchange of information at Community level about best available
techniques will help to redress the technological imbalances in the Community, will promote
the worldwide dissemination of limit values and techniques used in the Community and will
help the Member States in the efficient implementation of this Directive.

The Commission (Environment DG) established an information exchange forum (IEF) to assist
the work under Article 16(2) and a number of technical working groups have been established
under the umbrella of the IEF. Both IEF and the technical working groups include
representation from Member States and industry as required in Article 16(2).

The aim of this series of documents is to reflect accurately the exchange of information which
has taken place as required by Article 16(2) and to provide reference information for the
permitting authority to take into account when determining permit conditions. By providing
relevant information concerning best available techniques, these documents should act as
valuable tools to drive environmental performance.

4. Information Sources

This document represents a summary of information collected from a number of sources,
including in particular the expertise of the groups established to assist the Commission in its
work, and verified by the Commission services. All contributions are gratefully acknowledged.

5. How to understand and use this document

The information provided in this document is intended to be used as an input to the
determination of BAT in specific cases. When determining BAT and setting BAT-based permit
conditions, account should always be taken of the overall goal to achieve a high level of
protection for the environment as a whole.

Preface
Smitheries and Foundries Industry xv
The rest of this section describes the type of information that is provided in each chapter of the
document.

Chapters 1 and 2 provide general information on the industrial sector concerned and on the
industrial processes used within the sector.

Chapter 3 provides data and information concerning current emission and consumption levels,
reflecting the situation in existing installations at the time of writing.

Chapter 4 describes in more detail the emission reduction and other techniques that are
considered to be most relevant for determining BAT and BAT-based permit conditions. This
information includes the consumption and emission levels considered achievable by using the
technique, some idea of the costs and the cross-media issues associated with the technique, and
the extent to which the technique is applicable to the range of installations requiring IPPC
permits, for example new, existing, large or small installations. Techniques that are generally
seen as obsolete are not included.

Chapter 5 presents the techniques and the emission and consumption levels that are considered
to be compatible with BAT in a general sense. The purpose is thus to provide general
indications regarding the emission and consumption levels that can be considered as an
appropriate reference point to assist in the determination of BAT-based permit conditions or for
the establishment of general binding rules under Article 9(8). It should be stressed, however,
that this document does not propose emission limit values. The determination of appropriate
permit conditions will involve taking account of local, site-specific factors such as the technical
characteristics of the installation concerned, its geographical location and the local
environmental conditions. For existing installations, the economic and technical viability of
upgrading them also needs to be taken into account. Even the single objective of ensuring a high
level of protection for the environment as a whole will often involve making trade-off
judgements between different types of environmental impact, and these judgements will often
be influenced by local considerations.

Although an attempt is made to address some of these issues, it is not possible for them to be
considered fully in this document. The techniques and levels presented in Chapter 5 will
therefore not necessarily be appropriate for all installations. On the other hand, the obligation to
ensure a high level of environmental protection, including the minimisation of long-distance or
transboundary pollution, implies that permit conditions cannot be set on the basis of purely local
considerations. It is therefore of the utmost importance that the information contained in this
document is fully taken into account by permitting authorities.

Since the best available techniques change over time, this document will be reviewed and
updated as appropriate. All comments and suggestions should be made to the European IPPC
Bureau at the Institute for Prospective Technological Studies at the following address:

Edificio Expo, c/ Inca Garcilaso, s/n, E-41092 Seville, Spain
Telephone: +34 95 4488 284
Fax: +34 95 4488 426
e-mail: JRC-IPTS-EIPPCB@cec.eu.int
Internet: http://eippcb.jrc.es

Smitheries and Foundries Industry xvii
Best Available Techniques Reference Document on
Smitheries and Foundries
EXECUTIVE SUMMARY.........................................................................................................................I
PREFACE..............................................................................................................................................XIII
SCOPE................................................................................................................................................ XXXI
1 GENERAL INFORMATION ON FOUNDRIES............................................................................ 1
1.1 Sector overview.......................................................................................................................... 1
1.1.1 Foundry industry................................................................................................................. 1
1.1.2 Foundry markets ................................................................................................................. 8
1.1.3 Foundry types ................................................................................................................... 10
1.2 Environmental issues ............................................................................................................... 11
1.2.1 Air ..................................................................................................................................... 11
1.2.2 Residues............................................................................................................................ 11
1.2.3 Energy............................................................................................................................... 11
1.2.4 Water ................................................................................................................................ 11
2 APPLIED PROCESSES AND TECHNIQUES IN FOUNDRIES............................................... 13
2.1 Overview.................................................................................................................................. 13
2.1.1 The foundry process.......................................................................................................... 13
2.1.2 Iron casting ....................................................................................................................... 15
2.1.3 Steel casting...................................................................................................................... 17
2.1.4 Aluminium casting............................................................................................................ 19
2.1.5 Magnesium casting ........................................................................................................... 19
2.1.6 Copper casting .................................................................................................................. 20
2.1.7 Zinc casting....................................................................................................................... 21
2.1.8 Lead casting...................................................................................................................... 21
2.1.9 Casting of superalloys....................................................................................................... 21
2.2 Pattern making ......................................................................................................................... 22
2.2.1 General pattern making..................................................................................................... 22
2.2.2 Rapid prototyping (RP)..................................................................................................... 23
2.3 Raw materials and raw material handling ................................................................................ 25
2.4 Melting and metal treatment .................................................................................................... 28
2.4.1 Cupola furnaces ................................................................................................................ 29
2.4.1.1 Cold blast cupola furnace............................................................................................ 29
2.4.1.1.1 Description.......................................................................................................... 29
2.4.1.1.2 Maintenance........................................................................................................ 30
2.4.1.1.3 Advantages: ........................................................................................................ 30
2.4.1.1.4 Disadvantages: .................................................................................................... 30
2.4.1.2 Hot blast cupola furnace.............................................................................................. 30
2.4.1.2.1 Description.......................................................................................................... 30
2.4.1.2.2 Advantages: ........................................................................................................ 32
2.4.1.2.3 Disadvantages: .................................................................................................... 32
2.4.1.3 Long campaign cupola ................................................................................................ 32
2.4.1.4 Nature of atmospheric emissions ................................................................................ 33
2.4.2 Electric arc furnace (EAF) ................................................................................................ 34
2.4.2.1 Description.................................................................................................................. 34
2.4.2.2 Melting and refining with the acidic lined EAF.......................................................... 35
2.4.2.3 Melting and refining with the basic lined EAF ........................................................... 35
2.4.2.4 Nature of atmospheric emissions ................................................................................ 35
2.4.3 Induction furnace (IF) ....................................................................................................... 36
2.4.3.1 Coreless induction furnace .......................................................................................... 36
2.4.3.1.1 Description.......................................................................................................... 36
2.4.3.1.2 Melting practice .................................................................................................. 38
2.4.3.1.3 Advantages: ........................................................................................................ 39
2.4.3.1.4 Disadvantages: .................................................................................................... 39
2.4.3.2 Channel induction furnace .......................................................................................... 40
2.4.3.2.1 Description.......................................................................................................... 40
2.4.3.2.2 Advantages: ........................................................................................................ 41
xviii Smitheries and Foundries Industry
2.4.3.2.3 Disadvantages: ....................................................................................................41
2.4.3.3 Nature of emissions .....................................................................................................41
2.4.4 Radiant-roof furnace (resistance heated) ...........................................................................42
2.4.5 Rotary furnace ...................................................................................................................43
2.4.5.1 Description...................................................................................................................43
2.4.5.2 Melting practice...........................................................................................................43
2.4.5.3 Metallurgy ...................................................................................................................43
2.4.5.4 Application ..................................................................................................................44
2.4.5.5 Advantages: .................................................................................................................44
2.4.5.6 Disadvantages: .............................................................................................................44
2.4.6 Hearth type furnace ...........................................................................................................44
2.4.7 Shaft furnace .....................................................................................................................45
2.4.7.1 Description...................................................................................................................45
2.4.7.2 Advantages: .................................................................................................................46
2.4.7.3 Disadvantages: .............................................................................................................46
2.4.8 Crucible furnace ................................................................................................................46
2.4.8.1 Description...................................................................................................................46
2.4.8.2 Melting practice...........................................................................................................47
2.4.8.3 Advantages: .................................................................................................................47
2.4.8.4 Disadvantages: .............................................................................................................47
2.4.9 Argon Oxygen Decarburisation (AOD) converter for steel refining.................................48
2.4.10 Vacuum Oxygen Decarburisation converter (VODC) for steel refining...........................48
2.4.11 Metal treatment of steel .....................................................................................................49
2.4.12 Cast iron treatment ............................................................................................................50
2.4.12.1 Alloying.......................................................................................................................50
2.4.12.2 Homogenisation...........................................................................................................50
2.4.12.3 Desulphurisation and recarburisation of cupola melted iron .......................................50
2.4.12.4 Nodularisation treatment of the melt ...........................................................................51
2.4.12.5 Inoculation of the melt .................................................................................................52
2.4.13 Non-ferrous metal treatment..............................................................................................52
2.5 Mould and core production.......................................................................................................53
2.5.1 Raw materials....................................................................................................................54
2.5.1.1 Refractory materials.....................................................................................................54
2.5.1.1.1 Silica sand ...........................................................................................................55
2.5.1.1.2 Chromite sand .....................................................................................................56
2.5.1.1.3 Zircon sand..........................................................................................................56
2.5.1.1.4 Olivine sand ........................................................................................................56
2.5.1.2 Binders and other chemicals ........................................................................................57
2.5.1.2.1 Bentonite .............................................................................................................57
2.5.1.2.2 Resins ..................................................................................................................57
2.5.1.2.3 Coal dust..............................................................................................................58
2.5.1.2.4 Cereal binders......................................................................................................59
2.5.1.2.5 Iron oxide ............................................................................................................59
2.5.1.3 Running, gating, feeding and filtration........................................................................59
2.5.2 Sand preparation (transport, sieving, cooling, mixing) .....................................................60
2.5.2.1 Sand conditioning for green sand moulding ................................................................60
2.5.3 Moulding with natural sand...............................................................................................62
2.5.4 Moulding with clay-bonded sand (green sand moulding) .................................................62
2.5.5 Moulding with unbonded sand (V-process) ......................................................................63
2.5.6 Moulding and core-making with chemically-bonded sand................................................65
2.5.6.1 Cold-setting processes .................................................................................................65
2.5.6.1.1 Phenolic, acid catalysed ......................................................................................65
2.5.6.1.2 Furan, acid catalysed ...........................................................................................66
2.5.6.1.3 Polyurethane (phenolic isocyanate).....................................................................66
2.5.6.1.4 Resol ester (alkaline phenolic ester hardened) .................................................67
2.5.6.1.5 Alkyd oil, unbaked ..............................................................................................67
2.5.6.1.6 Ester silicate ........................................................................................................67
2.5.6.1.7 Cement ................................................................................................................67
2.5.6.2 Gas-hardened processes...............................................................................................67
2.5.6.2.1 Cold-box (amine hardened phenolic urethane)....................................................68
2.5.6.2.2 Resol ester (alkaline phenolics methyl formate hardened) ...............................68
2.5.6.2.3 SO
2
hardened furan resins ...................................................................................69
2.5.6.2.4 SO
2
hardened epoxy/acrylic (free radical curing) ...............................................69
Smitheries and Foundries Industry xix
2.5.6.2.5 CO
2
hardened sodium silicate (water glass)........................................................ 69
2.5.6.2.6 CO
2
hardened alkaline phenolic.......................................................................... 70
2.5.6.3 Hot curing processes ................................................................................................... 70
2.5.6.3.1 Hot-box, phenolic and/or furan based................................................................. 70
2.5.6.3.2 Warm-box........................................................................................................... 71
2.5.6.3.3 Shell (Croning) ................................................................................................... 71
2.5.6.3.4 Linseed oil .......................................................................................................... 72
2.5.6.3.5 Alkyd oil, baked.................................................................................................. 72
2.5.6.4 Coating of chemically-bonded sand moulds and cores ............................................... 73
2.5.6.4.1 Composition of coatings ..................................................................................... 73
2.5.6.4.2 Coating process................................................................................................... 73
2.5.7 Expendable pattern casting ............................................................................................... 74
2.5.7.1 Unbonded sand Lost Foam process.......................................................................... 74
2.5.7.2 Chemically-bonded sand Full mould process........................................................... 76
2.5.8 Permanent (metal) moulds preparation............................................................................. 77
2.5.9 Investment casting and ceramic shell................................................................................ 77
2.6 Casting ..................................................................................................................................... 79
2.6.1 Casting in lost moulds....................................................................................................... 79
2.6.1.1 Pouring........................................................................................................................ 79
2.6.1.2 Solidification (1
st
cooling)........................................................................................... 81
2.6.1.3 Shake-out .................................................................................................................... 82
2.6.1.4 Casting cooling (2
nd
cooling) ...................................................................................... 82
2.6.2 Casting in permanent moulds............................................................................................ 82
2.6.2.1 Gravity and low-pressure die-casting.......................................................................... 82
2.6.2.2 High-pressure die-casting............................................................................................ 84
2.6.2.3 Centrifugal casting ...................................................................................................... 86
2.6.2.4 Continuous casting...................................................................................................... 86
2.7 Finishing and post casting operations ...................................................................................... 87
2.7.1 Removal of the running system........................................................................................ 88
2.7.2 Sand removal .................................................................................................................... 88
2.7.3 Removal of burrs .............................................................................................................. 89
2.8 Heat treatment .......................................................................................................................... 90
2.8.1 Introduction....................................................................................................................... 90
2.8.2 Heat treatment furnaces .................................................................................................... 90
2.8.2.1 Chamber furnaces........................................................................................................ 90
2.8.2.2 Shaft furnaces.............................................................................................................. 91
2.8.2.3 Annealing furnaces...................................................................................................... 91
2.8.3 Quenching......................................................................................................................... 91
2.8.4 Heat treatment of ductile iron (SG iron) ........................................................................... 91
2.8.4.1 Stress relief.................................................................................................................. 92
2.8.4.2 Breakdown of carbides................................................................................................ 92
2.8.4.3 Annealing to produce a ferritic matrix ........................................................................ 92
2.8.4.4 Normalising to produce a pearlitic matrix................................................................... 92
2.8.4.5 Producing hardened and tempered structures.............................................................. 92
2.8.4.6 Austempered ductile iron (ADI).................................................................................. 92
2.8.5 Heat treatment of steel ...................................................................................................... 93
2.8.6 Heat treatment of aluminium............................................................................................ 94
2.8.6.1 Stress relieving and annealing..................................................................................... 94
2.8.6.2 Solution treatment and quenching............................................................................... 94
2.8.6.3 Precipitation treatment ................................................................................................ 94
2.8.6.4 Artificial ageing .......................................................................................................... 94
2.9 Quality control ......................................................................................................................... 95
3 CURRENT EMISSION AND CONSUMPTION LEVELS IN FOUNDRIES ........................... 97
3.1 Mass stream overview.............................................................................................................. 97
3.1.1 Introduction....................................................................................................................... 97
3.2 Melting and the metal treatment of ferrous metals................................................................... 97
3.2.1 Properties of melting furnaces for steel and cast iron ....................................................... 97
3.2.2 Cupola furnaces ................................................................................................................ 99
3.2.2.1 Coke and energy consumption .................................................................................... 99
3.2.2.2 Particulate matter ........................................................................................................ 99
3.2.2.3 Waste gases ............................................................................................................... 100
3.2.2.4 Cupola slag................................................................................................................ 102
xx Smitheries and Foundries Industry
3.2.2.5 Waste refractory.........................................................................................................102
3.2.3 Electric arc furnace..........................................................................................................103
3.2.3.1 Input...........................................................................................................................103
3.2.3.2 Particulate matter .......................................................................................................103
3.2.3.3 Visible fumes .............................................................................................................104
3.2.3.4 Waste gases................................................................................................................104
3.2.3.5 Slags ..........................................................................................................................105
3.2.4 Induction furnace.............................................................................................................105
3.2.4.1 Coreless induction furnace.........................................................................................105
3.2.4.1.1 Energy input ......................................................................................................105
3.2.4.1.2 Particulate matter...............................................................................................106
3.2.4.1.3 Waste gases .......................................................................................................107
3.2.4.1.4 Slags ..................................................................................................................107
3.2.4.2 Channel induction furnace .........................................................................................108
3.2.5 Rotary furnace .................................................................................................................108
3.2.5.1 Input...........................................................................................................................109
3.2.5.2 Particulate Matter.......................................................................................................109
3.2.5.3 Waste gases................................................................................................................109
3.2.6 Argon Oxygen Decarburisation (AOD) converter ..........................................................111
3.2.6.1 Input...........................................................................................................................111
3.2.6.2 Output ........................................................................................................................111
3.2.6.3 Waste gases................................................................................................................111
3.2.6.4 Slags ..........................................................................................................................112
3.2.7 Vacuum Oxygen Decarburisation Converter (VODC)....................................................112
3.2.8 Steel refining and treatment.............................................................................................112
3.2.9 Cast iron treatment ..........................................................................................................112
3.2.9.1 Nodularisation............................................................................................................112
3.3 Melting and the metal treatment of aluminium.......................................................................113
3.3.1 Survey of melting furnaces for aluminium......................................................................113
3.3.2 Shaft furnace ...................................................................................................................115
3.3.3 Induction furnace.............................................................................................................116
3.3.4 Radiant roof furnace (resistance heated) .........................................................................116
3.3.5 Hearth type furnace .........................................................................................................116
3.3.6 Crucible furnace (fuel and resistance heated)..................................................................117
3.3.7 Aluminium melt treatment ..............................................................................................117
3.4 Melting and casting of magnesium and magnesium alloys ....................................................118
3.4.1 Magnesium melt protection.............................................................................................118
3.4.2 Magnesium melt treatment ..............................................................................................119
3.4.3 Magnesium scrap.............................................................................................................119
3.5 Melting and casting of copper and copper alloys ...................................................................120
3.5.1 Melting and casting units ................................................................................................120
3.5.2 Copper and copper alloy melt treatment..........................................................................121
3.6 Melting and casting of zinc and zinc alloys............................................................................121
3.7 Melting and casting of lead.....................................................................................................122
3.8 Waste gas cleaning .................................................................................................................122
3.8.1 Abatement systems..........................................................................................................122
3.8.2 Dioxins ............................................................................................................................122
3.9 Mould and core production.....................................................................................................125
3.9.1 Introduction .....................................................................................................................125
3.9.2 Moulding with clay-bonded sand (green sand moulding) ...............................................127
3.9.3 Moulding with unbonded sand (V-process) ....................................................................127
3.9.4 Moulding and core-making with chemically-bonded sand..............................................128
3.9.4.1 Chemicals consumption levels...................................................................................128
3.9.4.2 Emission factors.........................................................................................................129
3.9.4.3 Cold-setting processes emissions...............................................................................129
3.9.4.4 Gas-hardened processes emissions ............................................................................130
3.9.4.5 Hot curing processes emissions .................................................................................130
3.9.5 Coating of chemically-bonded sand moulds and cores ...................................................131
3.9.6 Expendable pattern casting (Lost Foam/full mould casting) ...........................................131
3.9.6.1 Unbonded sand Lost Foam.....................................................................................131
3.9.6.2 Chemically-bonded sand full mould.......................................................................132
3.10 Casting....................................................................................................................................133
3.10.1 Casting, cooling and shake-out, using lost moulds..........................................................133
Smitheries and Foundries Industry xxi
3.10.1.1 Emission levels.......................................................................................................... 133
3.10.1.2 Sand-to-liquid metal ratios ........................................................................................ 137
3.10.1.3 Metal yield ................................................................................................................ 139
3.10.1.4 Used foundry sand..................................................................................................... 140
3.10.2 Casting in permanent moulds.......................................................................................... 141
3.11 Finishing/post-casting operations........................................................................................... 142
3.11.1 Slide grinding.................................................................................................................. 142
3.11.2 Shot blasting ................................................................................................................... 143
3.11.3 Fettling............................................................................................................................ 143
3.11.4 Finishing operations in steel foundries ........................................................................... 143
3.12 Heat treatment ........................................................................................................................ 144
3.13 Waste water............................................................................................................................ 144
3.13.1 Sources of waste water ................................................................................................... 144
3.13.2 Waste water from scrap storage...................................................................................... 145
3.13.3 Waste water from wet scrubbers used in cupola melting................................................ 145
3.13.4 Waste water from the casting, cooling and shake-out area and from mould
production/sand preparation............................................................................................ 146
3.13.5 Waste water from core-making....................................................................................... 146
4 TECHNIQUES TO CONSIDER IN THE DETERMINATION OF BAT FOR FOUNDRIES
......................................................................................................................................................... 147
4.1 Raw material storage and handling ........................................................................................ 148
4.1.1 Introduction..................................................................................................................... 148
4.1.2 Covered storage and impermeable scrap yard ................................................................ 148
4.1.3 Storage measures for binder chemicals........................................................................... 149
4.1.4 Use of clean scrap for melting and removal of sand from the return material ................ 150
4.1.5 Internal recycling of scrap ferrous metal ........................................................................ 151
4.1.6 Internal recycling of magnesium scrap ........................................................................... 152
4.1.7 Recycling of used containers .......................................................................................... 155
4.2 Metal melting and molten metal treatment............................................................................. 155
4.2.1 Cupola furnaces .............................................................................................................. 155
4.2.1.1 Optimisation of furnace operation............................................................................. 155
4.2.1.2 Quality control of coke input..................................................................................... 156
4.2.1.3 Operation with acidic or basic slag ........................................................................... 158
4.2.1.4 Increasing shaft height of CBC................................................................................. 158
4.2.1.5 Installation of secondary row of tuyres for a cold blast cupola (CBC).................... 159
4.2.1.6 Oxygen enrichment of blast supply........................................................................... 161
4.2.1.7 Superheating of HBC blast air................................................................................... 163
4.2.1.8 Minimal blast shut-off periods for HBC ................................................................... 164
4.2.1.9 Cokeless cupola furnace............................................................................................ 165
4.2.1.10 Gas-fired coke-cupola ............................................................................................... 169
4.2.2 Electric arc furnace ......................................................................................................... 171
4.2.2.1 Shortening melting and treatment time ..................................................................... 171
4.2.2.2 Foamy slag practice................................................................................................... 172
4.2.3 Induction furnace ............................................................................................................ 173
4.2.3.1 Process optimisation: optimising charge materials, charging and operation............. 173
4.2.3.2 Change from mains frequency to medium frequency furnaces ................................. 174
4.2.4 Rotary furnace................................................................................................................. 175
4.2.4.1 Increasing the furnace efficiency .............................................................................. 175
4.2.4.2 Use of an oxyburner .................................................................................................. 176
4.2.5 Selection of cupola vs. induction or rotary furnace for cast iron melting ....................... 177
4.2.6 Hearth type furnace......................................................................................................... 181
4.2.6.1 Use of an oxyburner .................................................................................................. 181
4.2.7 Other furnace types......................................................................................................... 181
4.2.7.1 Alternatives to the use of SF
6
as a cover gas for Mg-melting ................................... 181
4.2.8 Non-ferrous metal treatment ........................................................................................... 184
4.2.8.1 Aluminium degassing and cleaning using an impeller station .................................. 184
4.3 Mould- and core-making, including sand preparation............................................................ 185
4.3.1 Selection of the mould type ............................................................................................ 185
4.3.2 Moulding with clay-bonded sand (green sand moulding)............................................... 186
4.3.2.1 Preparation of clay-bonded sand by vacuum mixing and cooling............................. 186
4.3.3 Moulding and core-making with chemically-bonded sand ............................................. 188
4.3.3.1 Minimisation of binder and resin consumption......................................................... 188
xxii Smitheries and Foundries Industry
4.3.3.2 Minimisation of mould and core sand losses .............................................................190
4.3.3.3 Best practice for cold-setting processes.....................................................................191
4.3.3.4 Best practice for gas-hardened processes...................................................................191
4.3.3.5 Replacement of alcohol-based coatings with water-based coatings ..........................192
4.3.3.6 Microwave drying for water-based coatings..............................................................195
4.3.3.7 Use of non-aromatic solvents for cold-box core production......................................197
4.3.4 Alternative moulding/core-making methods ...................................................................200
4.3.4.1 Lost Foam casting......................................................................................................200
4.3.4.2 Ceramic shell moulding.............................................................................................202
4.3.5 Permanent (metal) moulds and pressure die preparation.................................................203
4.3.5.1 Minimisation of release agent and water consumption..............................................203
4.3.5.2 Closed-mould application of release agent ................................................................203
4.4 Metal casting...........................................................................................................................204
4.4.1 Improvement of metal yield ............................................................................................204
4.5 Fume, flue-gas and exhaust air capture and treatment ............................................................206
4.5.1 General principles............................................................................................................206
4.5.1.1 Reduction of fugitive emissions.................................................................................207
4.5.1.2 Use of a multi-flue stack............................................................................................209
4.5.1.3 Abatement techniques................................................................................................209
4.5.1.4 Dioxin prevention and abatement ..............................................................................213
4.5.1.5 Odour abatement........................................................................................................216
4.5.2 Cupola furnace ................................................................................................................216
4.5.2.1 Gas collection, cleaning and cooling .........................................................................216
4.5.2.2 Post combustion in an HBC combustion chamber.....................................................221
4.5.2.3 Post combustion in the cupola shaft...........................................................................224
4.5.3 EAF .................................................................................................................................227
4.5.3.1 Off-gas collection ......................................................................................................227
4.5.3.2 Exhaust gas cleaning..................................................................................................229
4.5.4 Induction furnace.............................................................................................................230
4.5.4.1 Off-gas collection ......................................................................................................230
4.5.4.2 Exhaust gas cleaning..................................................................................................232
4.5.5 Rotary furnace .................................................................................................................234
4.5.5.1 Off-gas collection and waste gas cleaning.................................................................234
4.5.6 Crucible, hearth type and radiant roof furnaces...............................................................235
4.5.6.1 Prevention of visible and fugitive emissions during melting and metal treatment ....235
4.5.7 Metal treatment ...............................................................................................................236
4.5.7.1 AOD Converter: exhaust gas capture and treatment..................................................236
4.5.7.2 Nodularisation: off-gas collection and dedusting ......................................................236
4.5.8 Mould and core-making ..................................................................................................238
4.5.8.1 Exhaust capture of dust from green sand preparation (sand plant) and dedusting.....238
4.5.8.2 Exhaust capture of dust from the green sand moulding shop work area and from
dedusting....................................................................................................................239
4.5.8.3 Exhaust capture from core-making general considerations ....................................239
4.5.8.4 Cold-box: capture of amine vapour and exhaust treatment........................................240
4.5.8.5 VOC-collection and removal .....................................................................................241
4.5.8.6 Exhaust gas cleaning using biofilter ..........................................................................242
4.5.8.7 Permanent moulding: collection of release agent emissions......................................245
4.5.9 Casting/Cooling/Shake-out..............................................................................................245
4.5.9.1 Introduction ...............................................................................................................245
4.5.9.2 Pouring and cooling lines encapsulation....................................................................246
4.5.9.3 Exhaust capture and treatment from shake-out..........................................................246
4.5.9.4 Off-gas cleaning using biofilter .................................................................................248
4.5.10 Finishing/post-casting operations: off-gas collection and treatment ...............................249
4.5.10.1 Finishing off-gas collection .......................................................................................249
4.5.10.2 Exhaust air cleaning techniques.................................................................................250
4.5.11 Heat treatment waste gas collection and cleaning ...........................................................251
4.5.11.1 Use of clean fuels in burner-heated heat treatment furnaces .....................................251
4.5.11.2 Quenching baths ........................................................................................................252
4.6 Waste water prevention and treatment....................................................................................253
4.6.1 Measures to prevent the generation of waste water.........................................................253
4.6.2 Treatment of scrubbing water and other waste water flows ............................................255
4.6.3 Example plant for waste water prevention and reduction................................................256
4.6.4 Oil interceptors................................................................................................................260
Smitheries and Foundries Industry xxiii
4.6.5 Recovery of amine from scrubbing water ....................................................................... 260
4.6.6 Reduction of glycols in die-casting waste water streams................................................ 262
4.7 Energy efficiency ................................................................................................................... 262
4.7.1 Introduction..................................................................................................................... 262
4.7.2 Induction furnace: waste heat utilisation......................................................................... 263
4.7.3 Cupola furnace: waste heat utilisation ............................................................................ 265
4.7.4 Reduce energy loss/improve preheating practice of ladles ............................................. 267
4.8 Sand: regeneration, recycling, re-use and disposal ................................................................ 268
4.8.1 Introduction..................................................................................................................... 268
4.8.2 Regeneration of green sand under optimised conditions (primary regeneration) ........... 272
4.8.3 Simple mechanical regeneration of cold-setting sand..................................................... 274
4.8.4 Cold mechanical regeneration using a grinding wheel unit ............................................ 275
4.8.5 Cold mechanical regeneration using an impact drum..................................................... 278
4.8.6 Cold regeneration using a pneumatic system.................................................................. 279
4.8.7 Thermal regeneration...................................................................................................... 281
4.8.8 Combined regeneration (mechanicalthermal-mechanical) for mixed organic-bentonite
sands ............................................................................................................................... 285
4.8.9 Wet sand regeneration .................................................................................................... 288
4.8.10 Regeneration of water glass sand using pneumatic systems ........................................... 289
4.8.11 Internal re-use of uncured core sand............................................................................... 291
4.8.12 Re-use of dusts from the green sand circuit in mould making........................................ 292
4.8.13 External re-use of used sand and the undersize from the sand circuit and regeneration
processes ......................................................................................................................... 293
4.9 Dust and solid residues: treatment and re-use ........................................................................ 294
4.9.1 Introduction..................................................................................................................... 294
4.9.2 (Pretreatment for) external re-use of solid residues ........................................................ 295
4.9.3 Minimisation of slag forming ......................................................................................... 297
4.9.4 Cupola furnace................................................................................................................ 298
4.9.4.1 Collection and recycling of coke breeze ................................................................... 298
4.9.4.2 Recirculation of filter dust into cupola furnace......................................................... 298
4.9.5 EAF................................................................................................................................. 302
4.9.5.1 Recycling of filter dust and sludge (from EAF melting) ........................................... 302
4.9.5.2 Recycling of Al slags and residues............................................................................ 303
4.10 Noise reduction ...................................................................................................................... 303
4.11 Decommissioning................................................................................................................... 305
4.12 Environmental management tools.......................................................................................... 305
5 BEST AVAILABLE TECHNIQUES FOR FOUNDRIES ......................................................... 313
5.1 Generic BAT (for the foundry industry) ................................................................................ 314
5.2 Ferrous metal melting ............................................................................................................ 317
5.3 Non-ferrous metal melting ..................................................................................................... 319
5.4 Lost mould casting ................................................................................................................. 321
5.5 Permanent mould casting ....................................................................................................... 323
6 EMERGING TECHNIQUES FOR FOUNDRIES ..................................................................... 325
6.1 Use of low cost combustible materials in cupola melting...................................................... 325
6.2 Recycling of metal-bearing filter dust (ferrous metals).......................................................... 325
6.3 Amine recovery from the core-making waste gas by gas permeation.................................... 326
6.4 Separate spraying of release agent and water in aluminium die-casting................................ 327
6.5 Inorganic binder material for core-making............................................................................. 328
7 CONCLUDING REMARKS ........................................................................................................ 331
7.1 Timing of the work................................................................................................................. 331
7.2 Sources of information........................................................................................................... 331
7.3 Degree of consensus reached ................................................................................................. 332
7.4 Recommendations for future work......................................................................................... 333
7.5 Suggested topics for future R&D projects ............................................................................. 333
8 REFERENCES .............................................................................................................................. 335
9 GLOSSARY................................................................................................................................... 343
9.1 Abbreviations ......................................................................................................................... 343
9.2 Terms ..................................................................................................................................... 345
9.3 Lexicon................................................................................................................................... 348
10 ANNEXES ...................................................................................................................................... 351
xxiv Smitheries and Foundries Industry
10.1 Annex 1 Overview of data for cast iron melting in various cupola types and in the induction
furnace ....................................................................................................................................351
10.2 Annex 2 Dust control equipment at a large-scale automotive foundry..................................361
Smitheries and Foundries Industry xxv
List of figures

Figure 1.1: Ferrous foundry productivity data for various European countries ........................................... 8
Figure 1.2: Relative sectoral market shares.................................................................................................. 9
Figure 1.3: Market shares for iron castings (data for Spanish market)......................................................... 9
Figure 1.4: Market shares for steel castings (data for Spanish market)...................................................... 10
Figure 2.1: The foundry process ................................................................................................................ 13
Figure 2.2: Process flow diagrams for the melting and metal treatment of cast iron ................................. 16
Figure 2.3: Process flow diagrams for the melting and metal treatment of steel........................................ 18
Figure 2.4: Wooden pattern........................................................................................................................ 23
Figure 2.5: Coreboxes ................................................................................................................................ 23
Figure 2.6: Thermopolymer (wax) patterns in an RP machine .................................................................. 24
Figure 2.7: Aluminium scrap(l.) and ingots (r.) ......................................................................................... 25
Figure 2.8: Pneumatic conveyors and silos for powdered materials .......................................................... 25
Figure 2.9: Schematic outline and miniature model of a (cold blast) cupola furnace ................................ 29
Figure 2.10: Effect of air preheating on blast furnace efficiency............................................................... 31
Figure 2.11: Schematic representation of a long campaign cupola ............................................................ 32
Figure 2.12: EAF furnace........................................................................................................................... 34
Figure 2.13: General arrangement of coreless induction furnace............................................................... 37
Figure 2.14: Coreless induction furnaces ................................................................................................... 38
Figure 2.15: Typical construction of a bath channel furnace ..................................................................... 40
Figure 2.16: Channel induction furnace ..................................................................................................... 41
Figure 2.17: Radiant-roof furnace.............................................................................................................. 42
Figure 2.18: Cross-section of a hearth type furnace................................................................................... 44
Figure 2.19: Shaft furnace.......................................................................................................................... 46
Figure 2.20: Crucible furnaces ................................................................................................................... 47
Figure 2.21: AOD converter....................................................................................................................... 48
Figure 2.22: Deoxidation, using an aluminium wire.................................................................................. 49
Figure 2.23: The sandwich method for nodularisation............................................................................... 51
Figure 2.24: Inoculation during pouring .................................................................................................... 52
Figure 2.25: Typical grain size distribution for silica foundry sand........................................................... 55
Figure 2.26: Components of the running and gating system...................................................................... 60
Figure 2.27: Various types of filters........................................................................................................... 60
Figure 2.28: Flow diagram for a typical green sand plant.......................................................................... 61
Figure 2.29: Various types of sand mixers................................................................................................. 61
Figure 2.30: Vacuum moulding ................................................................................................................. 64
Figure 2.31: Furan cores............................................................................................................................. 66
Figure 2.32: Cold-box core ........................................................................................................................ 68
Figure 2.33: Croning cores (top) and moulds (bottom).............................................................................. 72
Figure 2.34: Flow coating with alcohol-based coating, followed by firing or drying................................ 73
Figure 2.35: Dip-coating with water-based coatings, followed by drying in a hot-air furnace .................. 74
Figure 2.36: The Lost Foam process .......................................................................................................... 75
Figure 2.37: Expendable polystyrene pattern (r.) for the Lost Foam process and an example of a casting
produced by the Lost Foam method (l.).................................................................................. 76
Figure 2.38: Investment casting process .................................................................................................... 78
Figure 2.39: Lip pour ladle for molten metal pouring................................................................................ 80
Figure 2.40: Teapot ladle for molten metal pouring................................................................................... 80
Figure 2.41: Bottom pour ladle for molten metal pouring ......................................................................... 81
Figure 2.42: Pouring furnace...................................................................................................................... 81
Figure 2.43: Shake-out at the end of an automated moulding line............................................................. 82
Figure 2.44: The principle of a low-pressure die-casting machine............................................................. 83
Figure 2.45: Low-pressure die-casting machine ........................................................................................ 83
Figure 2.46: Cold-chamber and hot-chamber high-pressure die-casting devices....................................... 84
Figure 2.47: Schematic representation of a centrifugal casting machine ................................................... 86
Figure 2.48: Schematic representation of a thread casting machine used for continuous casting.............. 87
Figure 2.49: Casting with running and gating system................................................................................ 88
Figure 2.50: Castings before (l., m.) and after (r.) sand removal by blasting............................................. 88
Figure 2.51: Burrs (l.) and their removal using a grinding stone (m.) and slide grinding (r.) .................... 89
Figure 2.52: Quenching of a hot casting shortly after heat treatment ........................................................ 91
Figure 2.53: Typical austempering heat treatment stages .......................................................................... 93
Figure 3.1: Mass stream overview for the foundry process........................................................................ 97
xxvi Smitheries and Foundries Industry
Figure 3.2: Typical energy losses from a mains frequency coreless induction furnace............................106
Figure 3.3: Energy consumption (kWh/tonne) as a function of transferred tonnage and loss of temperature
for channel induction furnaces ..............................................................................................108
Figure 3.4: Mg die-casting production in tonnes/yr by EU states and by cover gas (SO
2
or SF
6
) ............119
Figure 3.5: Waste gas load upon pouring in full and hollow moulds .......................................................133
Figure 3.6: Emission levels of pollutants upon pouring, cooling and shake-out for a green sand iron
foundry..................................................................................................................................136
Figure 3.7: Emission levels for particulate matter upon pouring, cooling and shake-out for a green sand
iron foundry ..........................................................................................................................137
Figure 3.8: Green sand-to-liquid metal ratios in iron foundries................................................................138
Figure 3.9: Green sand-to-liquid metal ratios in copper foundries ...........................................................138
Figure 3.10: Total mixed sand-to-liquid metal ratios in UK foundries.....................................................139
Figure 4.1: Mass flow diagram for the fluxless re-melting of magnesium class 1 scrap..........................153
Figure 4.2: Mass flow diagram for the salt re-melting of class 1 magnesium scrap.................................154
Figure 4.3: Various methods of oxygen injection.....................................................................................162
Figure 4.4: Effect of blast shut-off periods on tapping temperature in a post-blast cupola ......................164
Figure 4.5: Cokeless cupola furnace in duplex operation.........................................................................166
Figure 4.6: Moulding-sand preparation plant with vacuum mixer-cooler ................................................187
Figure 4.7: Thermal balance of a stove for core drying, working at 72 % capacity .................................194
Figure 4.8: Thermal balance for the microwave drying of cores ..............................................................196
Figure 4.9: Total carbon emissions from various process steps, using aromatic- and vegetable-based
solvents .................................................................................................................................199
Figure 4.10: Bag filter units; inside (l.) and outside (m., r.) views ...........................................................210
Figure 4.11: Operational conditions for wet and dry dedusting of hot blast cupola off-gas .....................212
Figure 4.12: Flow sheet of cold blast cupola with heat recovery, cooling and a bag filter .......................218
Figure 4.13: Hot blast cupola with a recuperator and wet scrubber..........................................................221
Figure 4.14: Hot blast cupola with a recuperator and bag filter................................................................222
Figure 4.15: Principle of post combustion and off-gas cooling in cold blast cupolas...............................225
Figure 4.16: Principle drawings of (a) a roof mounted hood, (b) a side-draught hood, and (c) direct
evacuation through a fourth-hole ..........................................................................................227
Figure 4.17: Doghouse cover for a shake-out grate, for large-size batch castings....................................248
Figure 4.18: Treatment of waste water and sludge from the wet de-dusting system of a cupola furnace.256
Figure 4.19: Water circuit for the wet dedusting system of a cupola furnace...........................................258
Figure 4.20: Treatment system for sludge from the wet dedusting of cupola flue-gas .............................259
Figure 4.21: Amine recycling plant ..........................................................................................................261
Figure 4.22: Use of waste heat for scrap drying .......................................................................................264
Figure 4.23: Schematic representation of a hot blast cupola with a steam boiler, turbine and generator .266
Figure 4.24: Process flow diagram of a hot blast cupola with heat recovery............................................267
Figure 4.25: Sand balance diagram for a thermal/mechanical regeneration system.................................271
Figure 4.26: Cold mechanical regeneration using grinding......................................................................275
Figure 4.27: Operational data on the sand circuit of a Dutch green sand foundry....................................276
Figure 4.28: Fixed costs in euros per tonne of regenerated sand, for the mechanical regeneration of sand
with cold-setting binders.......................................................................................................277
Figure 4.29: Cold mechanical regeneration using a pneumatic system....................................................280
Figure 4.30: Fixed costs in euros per tonne of regenerated sand for thermal regeneration and high level
mechanical regeneration of mixed sands ..............................................................................284
Figure 4.31: Mechanical-thermal-mechanical sand regeneration unit ......................................................285
Figure 4.32: Rotary mechanical regeneration unit....................................................................................290
Figure 4.33: Zinc distribution for tuyre injection, upon injection during 20 days ..................................300
Figure 4.34: Zinc distribution for recirculation through the charge, upon charging during 11 days ........301
Figure 6.1: Spray head with separate nozzles for water and release agent ...............................................327
Figure 10.1: Input-output balance for various technical modifications of cupola melting .......................353
Figure 10.2: Cost of raw materials (France: 1993 to 2003) ......................................................................360
Figure 10.3: Monitoring data for dust emissions from 4 locations in the sand plant (24 hours continuous
monitoring) ...........................................................................................................................362
Figure 10.4: Monitoring data for dust emissions from 4 locations in the sand plant (2 weeks continuous
monitoring) ...........................................................................................................................362
Figure 10.5: Monitoring data for dust emissions from 4 locations in the sand plant (30 days continuous
monitoring) ...........................................................................................................................363
Smitheries and Foundries Industry xxvii
List of tables

Table 1.1: European production data for ferrous castings, i.e. iron, steel and malleable iron castings (in
kilotonnes) ................................................................................................................................ 2
Table 1.2: European production data for non-ferrous metal castings (in kilotonnes) .................................. 3
Table 1.3: Number of foundries (production units) for iron, steel and malleable iron casting..................... 4
Table 1.4: Number of foundries (production units) for non-ferrous metal casting ...................................... 5
Table 1.5: Employment in the foundry industry for iron, steel and malleable iron casting ......................... 6
Table 1.6: Employment in the foundry industry for non-ferrous casting ..................................................... 7
Table 2.1: Relative shares of applied casting techniques for Al................................................................. 19
Table 2.2: Most common zinc alloys, contents in %.................................................................................. 21
Table 2.3: Description of rapid prototyping techniques ............................................................................. 24
Table 2.4: Potential soil contamination from ferrous foundry raw materials............................................. 27
Table 2.5: Applicability of furnace types, for melting (m) and holding (h) ............................................... 28
Table 2.6: Field of application of available induction furnaces as related to their frequency.................... 38
Table 2.7: Survey of various resin types and their applicability ................................................................ 58
Table 2.8: Composition of natural sand ..................................................................................................... 62
Table 2.9: Composition of additive mix for green sand preparation (excl. water) ..................................... 63
Table 2.10: Core-making processes used in 48 automobile foundries in Germany, 1991 ......................... 65
Table 3.1: Typical melting furnace properties and emission data.............................................................. 98
Table 3.2: Average energy consumption for off-gas treatment and holding.............................................. 99
Table 3.3: Dust emission levels (raw cupola off-gas) from various cupola types...................................... 99
Table 3.4: Cupola dust emission levels and PM size distribution............................................................ 100
Table 3.5: Typical composition of cupola dust, data in weight per cent .................................................. 100
Table 3.6: Unburned top gas composition for cupola furnace.................................................................. 101
Table 3.7: Emission data for hot blast and cold blast cupolas working with different set-ups................. 102
Table 3.8: Typical cupola slag composition............................................................................................. 102
Table 3.9: Chemical composition of EAF dust from steel foundries ....................................................... 103
Table 3.10: Typical emission values for EAF furnaces............................................................................ 104
Table 3.11: Chemical composition of EAF slag ...................................................................................... 105
Table 3.12: IF dust emission level and PM size distribution.................................................................... 106
Table 3.13: Chemical composition of induction furnace dust in cast iron foundries ............................... 107
Table 3.14: Typical emission data for induction furnace melting in a ferrous foundry ........................... 107
Table 3.15: Typical composition of induction furnace slag ..................................................................... 108
Table 3.16: Chemical composition of rotary furnace dust in ferrous melting.......................................... 109
Table 3.17: Measured emissions for rotary furnace, melting cast iron, without flue-gas cleaning
equipment ............................................................................................................................. 110
Table 3.18: Consumption levels for the AOD treatment of steel ............................................................. 111
Table 3.19: AOD slag composition.......................................................................................................... 112
Table 3.20: Comparison of various nodularisation procedures ................................................................ 112
Table 3.21: Typical furnace properties and emission data for aluminium melting .................................. 114
Table 3.22: Input and output for aluminium melting in a shaft furnace................................................... 115
Table 3.23: Results from a 6-day trial melt in a 2 tonne melting capacity shaft furnace for Al............... 115
Table 3.24: Emission data for an oil-fired hearth type furnace melting Al .............................................. 117
Table 3.25: Consumption and emission data for the crucible melting of aluminium............................... 117
Table 3.26: The use of SF
6
in pressure die-casting operations under various operating conditions......... 118
Table 3.27: Use of SF
6
in gravity die-casting operations ......................................................................... 118
Table 3.28: Magnesium scrap grade classes............................................................................................. 119
Table 3.29: Mass balance data for a brass low-pressure die-casting foundry (melting + die-casting
operations) ............................................................................................................................ 120
Table 3.30: Particulate emissions from the crucible melting of copper alloys......................................... 121
Table 3.31: Typical mass and energy data for zinc pressure die-casting ................................................. 121
Table 3.32: Example of current emisions from some dust abatement applications.................................. 122
Table 3.33: Dioxin emission data for various foundry types ................................................................... 123
Table 3.34: Cupola furnace dioxin emissions .......................................................................................... 124
Table 3.35: Environmental impact from binder systems.......................................................................... 126
Table 3.36: Emission values and emission factors for dust emissions from moulding shops after exhaust
cleaning ................................................................................................................................ 126
Table 3.37: Example dust emissions and particle size for mould and core-making................................. 126
Table 3.38: Typical properties of green sand, measured on 105 sand samples from 105 iron foundries . 127
Table 3.39: Consumption levels of various resins, catalysts, hardeners and additives for chemically-
bonded sand preparation....................................................................................................... 128
xxviii Smitheries and Foundries Industry
Table 3.40: Emission factors for mould making.......................................................................................129
Table 3.41: Emission factors for core-making..........................................................................................129
Table 3.42: Mass balance data for brass foundry core using a hot-box process .......................................131
Table 3.43: Comparative data for residue production from Lost Foam and green sand aluminium casting
..............................................................................................................................................131
Table 3.44: Stack emission data for Lost Foam and green sand aluminium foundries.............................132
Table 3.45: Environmental impact from binder systems upon pouring, shake-out and cooling...............135
Table 3.46: Emission factors for iron casting in cold-setting resin-bonded sand moulds.........................135
Table 3.47: Emission values and emission factors for dust emission from shake-out after flue-gas cleaning
..............................................................................................................................................137
Table 3.48: Metal yield in the main alloy sectors .....................................................................................140
Table 3.49: Results from analyses of used foundry sands from several sources ......................................140
Table 3.50: PAH and phenol contents of used foundry sands ..................................................................141
Table 3.51: Mass balance data, exhaust and effluent composition for aluminium presure die-casting units
..............................................................................................................................................141
Table 3.52: Indicative emission factors for casting of bronze, brass and zamac ......................................142
Table 3.53: Input-output balance for the slide grinding of aluminium castings .......................................142
Table 3.54: Emission data for the exhaust gas from shot blasting, using various dedusting techniques ..143
Table 3.55: Emission data for the exhaust gas from fettling, using various dedusting techniques, ..........143
Table 3.56: Average dust emission from steel finishing...........................................................................143
Table 3.57: Possible impurities for specified scrap types .........................................................................145
Table 3.58: Pollutant concentrations in filter effluent from dewatering the sludge from the wet scrubbers
of cupola furnaces .................................................................................................................146
Table 4.1: Information contained in the discussion of each technique included in Chapter 4..................147
Table 4.2: Problems arising from an incorrect storage of liquid binder chemicals...................................149
Table 4.3: Inputs and outputs for a magnesium scrap treatment plant......................................................153
Table 4.4: Typical properties of foundry cokes ........................................................................................157
Table 4.5: Shaft height requirements ........................................................................................................158
Table 4.6: Example data for the change in coke consumption upon shaft height increases .....................159
Table 4.7: Typical operational data for a cokeless cupola furnace ...........................................................167
Table 4.8: Emission data for cokeless cupola and hot blast cupola ..........................................................168
Table 4.9: Operational costs for cokeless cupola, relative to hot blast cupola (set at 100%) ...................169
Table 4.10: Effect of oxygas burners on the flue-gas composition for a hot blast cupola ........................170
Table 4.11: Operational costs for cold blast cupola with and without oxygas burners.............................171
Table 4.12: Energy and temperature data for EAF furnace melting with normal slag and foamy slag ....172
Table 4.13: Table of energy consumptions (minimum melt)....................................................................177
Table 4.14: Technical choice of the melting equipment for cast iron melting..........................................179
Table 4.15: Example plant cost data for replacement of an existing cupola with a rotary or induction
furnace ..................................................................................................................................180
Table 4.16: Cost comparison between SO
2
and SF
6
used for protective gas atmospheres .......................182
Table 4.17: Operational and cost data for a new die-casting plant of 1000 tonne/yr Mg output ..............183
Table 4.18: Investment costs for an impeller station ................................................................................184
Table 4.19: Applicability of the various moulding types..........................................................................185
Table 4.20: Technical properties of the various moulding types ..............................................................186
Table 4.21: Process variables and how they indicate mixer performance ................................................189
Table 4.22: Annual savings, cost and payback for example plant mixer control......................................190
Table 4.23: Results of test campaigns on core drying, using hot air and microwave furnaces.................196
Table 4.24: Emission data for cold-box core systems in an aluminium foundry, measured at the shake-out
grate and in the exhaust-air stack..........................................................................................198
Table 4.25: Emissions (%) of selected compounds from vegetable-based cold-box core systems...........198
Table 4.26: Operational data for the production of a similar cast iron compressor casing, using various
methods.................................................................................................................................201
Table 4.27: Economic data for an aluminium Lost Foam foundry...........................................................201
Table 4.28: Relative weight reduction for ceramic shell casting of valves as compared to sand casting .202
Table 4.29: Typical metal yields for different iron casting types .............................................................205
Table 4.30: Weighted average of yield performances in 82 UK ferrous foundries, 1981 - 1987 .............205
Table 4.31: Survey of air emissions from different ferrous foundry stages..............................................207
Table 4.32: Properties of wet and dry abatement systems for foundries ..................................................212
Table 4.33: Investment and energy consumption data for various abatement systems.............................213
Table 4.34: Investment cost estimation for carbon injection installation, added to HBC.........................215
Table 4.35: Total costs for carbon injection installation, added to HBC..................................................216
Table 4.36: Operational data of cold blast cupola furnaces with a bag filter for dust abatement .............220
Smitheries and Foundries Industry xxix
Table 4.37: Operational data of hot blast cupola furnaces using a bag filter and a disintegrator for dust
capture .................................................................................................................................. 223
Table 4.38: Energy consumption of hot blast cupolas.............................................................................. 224
Table 4.39: Operational data of two example configurations using post combustion with water cooling (I),
and air cooling (J)................................................................................................................. 226
Table 4.40: Analytical results and applicable (French) legislation, for post combustion in a cold blast
cupola with water cooling (I), and the air cooling (J) of flue-gases ..................................... 226
Table 4.41: Typical exhaust flowrates (in m/s) and particulate removal efficiency of EAF exhaust capture
systems ................................................................................................................................. 228
Table 4.42: Cost data for EAF flue-gas capture and cleaning equipment ................................................ 230
Table 4.43: Emission data for a cast iron foundry, using induction melting and a centralised exhaust
system with a bag filter......................................................................................................... 233
Table 4.44: Investment cost and power consumption for a bag filter unit on induction furnaces, with
varying final dust emission levels......................................................................................... 233
Table 4.45: A cost estimate for installation of an abatement system for visible fumes during charging and
tapping.................................................................................................................................. 235
Table 4.46: Cost calculation for MgO fume abatement ........................................................................... 238
Table 4.47: Dust emission values from moulding and post casting shops ............................................... 239
Table 4.48: Emission data for a cold-box core-making shop, using an acid scrubber.............................. 241
Table 4.49: Specification and investment cost for an amine scrubber on cold-box core-shop exhaust ... 241
Table 4.50: Operational data for the biofiltration of a cold-box core-making off-gas ............................. 243
Table 4.51: Emission data for the biofiltration of exhaust from green sand pouring and a cooling line.. 244
Table 4.52: Emission data from EP-cleaned HPDC off-gas..................................................................... 245
Table 4.53: Applicability of dust capture techniques for various finishing operations ............................ 250
Table 4.54: Emission levels from a heat treatment furnace before and after conversion to natural gas
burners system...................................................................................................................... 252
Table 4.55: Applicability of fume collection techniques for quenching baths ......................................... 253
Table 4.56: Production and water consumption data (on yearly basis) for example foundry .................. 257
Table 4.57: Disposal costs for the extension of a waste water system with sludge treatment.................. 259
Table 4.58: Typical energy use in a non-ferrous and EAF steel foundry................................................. 263
Table 4.59: Fields of application of different regeneration systems for monosands ................................ 270
Table 4.60: Fields of application of different regeneration systems for mixed sands .............................. 270
Table 4.61: Applicability of various sand regeneration techniques to various sand types ....................... 271
Table 4.62: Compatibility of regenerated sources sands with various binders......................................... 272
Table 4.63: Cost benefits from primary reclamation................................................................................ 274
Table 4.64: Operational data of 3 German reference foundries applying thermal sand regeneration ...... 283
Table 4.65: Operational data of a mechanical-thermal-mechanical treatment system (pneumatic - fluid bed
- pneumatic).......................................................................................................................... 286
Table 4.66: Operational data of a silicate sand regeneration unit............................................................. 290
Table 4.67: Necessary treatment and possible limitations for external re-use of solid residues .............. 295
Table 4.68: Summary of external re-use applications for solid foundry residues (status 1999)............... 296
Table 4.69: Properties of injection techniques for recirculation of cupola furnace dust .......................... 299
Table 5.1: Emission to air associated with the use of BAT for ferrous metal melting and treatment ...... 319
Table 5.2: Emissions to air associated with the use of BAT for the cupola melting of ferrous metals .... 319
Table 5.3: Emissions to air associated with the use of BAT for the EAF melting of ferrous metals ....... 319
Table 5.4: Emissions to air associated with the use of BAT for the rotary melting of ferrous metals ..... 319
Table 5.5: Emissions to air associated with the use of BAT in the melting of aluminium....................... 321
Table 5.6: Emissions to air associated with the use of BAT for moulding and casting using lost moulds
.............................................................................................................................................. 323
Table 5.7: Emissions to air associated with the use of BAT for permanent mould casting (incl. HPDC)324
Table 10.1: Operational data of a cold blast cupola (10 tonne/h)............................................................. 354
Table 10.2: Operational data of a hot blast long campaign cupola (10 tonne/h) ...................................... 355
Table 10.3: Operational data of a cokeless cupola and hot blast cupola with air preheating ................... 356
Table 10.4: Comparison of liquid cast iron costs -June 2003. Mass production 10 t/h......................... 357
Table 10.5: Price of lamellar cast iron: Metallic charge + energy............................................................ 358
Table 10.6: Price of nodular cast iron: Metallic charge + energy ............................................................ 359
Scope
Smitheries and Foundries Industry xxxi
SCOPE

This document reflects the exchange of information on the activities covered by Annex I
categories 2.3 (b), 2.4 and 2.5 (b) of the IPPC Directive, i.e.

2.3. Installations for the processing of ferrous metals:
(b) smitheries with hammers the energy of which exceeds 50 kilojoule per hammer, where
the calorific power used exceeds 20 MW

2.4. Ferrous metal foundries with a production capacity exceeding 20 tonnes per day

2.5. Installations
(b) for the smelting, including the alloyage, of non-ferrous metals, including recovered
products, (refining, foundry casting, etc.) with a melting capacity exceeding 4 tonnes
per day for lead and cadmium or 20 tonnes per day for all other metals.

In setting up the working scope for this document, the TWG considered the possible
interpretation of the threshold values. The TWG compared the pre-defined IPPC scope to the
actual existence of installations that meet the above descriptions, in the European Union and in
the Candidate Countries. This comparison resulted in a working scope which covers the
following:

- the casting of ferrous materials, e.g. grey cast iron, malleable and ductile iron, steel
- the casting of non-ferrous materials, e.g. aluminium, magnesium, copper, zinc, lead and
their alloys.

Smitheries were excluded from the scope of the document since no European smitheries were
reported which meet the conditions stated in Annex I 2.3 (b), i.e. Smitheries with hammers the
energy of which exceeds 50 kJ per hammer, (and) where the calorific power used exceeds 20
MW. Accordingly, cadmium, titanium and precious metals foundries, as well as bell and art
casting foundries were excluded on capacity grounds.

Continuous casting (into sheets and slabs) has already been covered in the BREF documents on
iron and steel production and on non-ferrous metal industries. Therefore it is not dealt with in
this document.

Smelting, alloying and the refining of non-ferrous metals are covered in the BREF document on
non-ferrous metal industries and will therefore be excluded from the scope of this document. In
covering non-ferrous metals in this document, the process is considered to start with the melting
of ingots and internal scrap or with liquid metal.

From a process point of view, the following foundry process steps are covered by the scope of
this document:

- pattern making
- raw materials storage and handling
- melting and metal treatment
- mould and core production, and moulding techniques
- casting or pouring and cooling
- shake-out
- finishing
- heat treatment.

Chapter 1
Smitheries and Foundries Industry 1
1 GENERAL INFORMATION ON FOUNDRIES

1.1 Sector overview

1.1.1 Foundry industry

Foundries melt ferrous and non-ferrous metals and alloys and reshape them into products at or
near their finished shape through the pouring and solidification of the molten metal or alloy into
a mould. The foundry industry is a differentiated and diverse industry. It consists of a wide
range of installations, from small to very large; each with a combination of technologies and
unit operations selected to suit the input, size of series and types of product produced in the
specific installation. The organisation within the sector is based on the type of metal input, with
the main distinction being made between ferrous and non-ferrous foundries.

The European foundry industry is the third largest in the world for ferrous castings and the
second largest for non-ferrous. The total production of castings in various European countries
are given in Table 1.1 and Table 1.2. Data for Northern Ireland, Luxemburg and some of the
Accession Countries have not been provided, however it is known that activity in these regions
is low compared to the listed regions. Germany, France and Italy are the top three production
countries in Europe, with a total annual production of over two million tonnes of castings each.
In recent years Spain has taken over the fourth position from Great Britain, with both having a
production of over one million tonnes of castings. Together, the top five countries produce more
than 80 % of the total European production.

The total European production tonnage of ferrous castings has been stable over the past five
years, although some fluctuations have occurred for individual countries. For instance, the
figures for Great Britain indicate a general declining trend in production output, whereas the
trend for Spain is one of growth. The non-ferrous foundry sector has undergone steady growth
since 1998. The total figure for 2001 is obscured by the lack of data from Great Britain. In
general, it can be seen from Table 1.2 that in most countries production has risen. This holds not
just for the major producing countries but also for those countries with low amounts of
production.

Chapter 1
2 Smitheries and Foundries Industry
Country 1998 1999 2000 2001 2002 2000:2001 2001:2002
% change
Austria 190.1 181.7 191.4 192.4 181.2 0.5 -5.8
Belgium 144.4 149.3 149.8 149.5 143.7 -0.2 -3.9
Czech
Republic 493 379.1 390.3 415.3 381.6 6.4 -8.1
Denmark 85.8 86 96.4 85.7 87.3 -11.1 1.9
Estonia n.d n.d 0.94 1.07 1.1 13.8 2.8
Finland 122.6 109 117.6 119.5 112.5 1.6 -5.8
France 2250.8 2146.6 2283.1 2147.4 2128.6 -5.9 -0.9
Germany 3662.9 3555.2 3758.2 3801.4 3749.7 1.1 -1.4
Great Britain 1076.3
a
949.2
a
968.2
a
906.3
a
886.3
a
-6.4 -2.2
Hungary 78.1 68.7 74.8 62.8 67.9 -16.0 8.2
Ireland 450 480 520 275 n.d -47.1
Italy 1508.4 1492.6 1516.4 1433.3 1460.9 -5.5 1.9
Netherlands 140.6 121 136 132.3 123.7 -2.7 -6.5
Norway 65.3 67.7 70.1 73.4 67.3 4.7 -8.3
Poland 675 610.2 671.2 673 598.0 0.3 -11.1
Portugal 98.6 97.7 102.3 100 96.7 -2.2 -3.3
Slovakia n.d n.d n.d 47.5 n.d
Slovenia 89.8 81.9 86.9 96.3 n.d 10.8
Spain 706.6 759.3 950.5 955.7 992.9 0.5 3.9
Sweden 264.2 253.2 266.7 244.7 234.6 -8.2 -4.1
Switzerland 122.8
b
122
b
119.9
b
105.5
b
81.8
b
-12.0 -22.5
Total 12225 11710 12471 12018 11396
Full total
c
12018 11815
a) Without steel castings
b) Without steel and malleable iron castings
c) Full total calculated by taking most recent available number for the years for which no data are given
Table 1.1: European production data for ferrous castings, i.e. iron, steel and malleable iron castings
(in kilotonnes)
[168, CAEF, 2002], [202, TWG, 2002]

For ferrous foundries, the progress of material substitution in recent years has caused the share
of iron castings in the output total to decline slightly, dropping from 58.9 % in 2001 to 58.2 %
in 2002. At the same time, producers of nodular-iron castings held a share of 34.3 % in the
production total in 2002, marking an increase of 0.5 percentage points compared to 2001.
Producers of malleable castings were able to expand their share from 1.1 % in 2001 to 1.3 % in
2002, while the share of steel castings in the output total ranged around 5.8 % in 2002 (5.9 % in
2001).

Chapter 1
Smitheries and Foundries Industry 3
Country 1998 1999 2000 2001 2002 2000:2001 2001:2002
% change
Austria 90.4 92.4 105.9 113.3 116.2 7.0 2.6
Belgium 25.3 23.8 27.2 26.3 26.7 -3.3 1.6
Czech Republic 44.8 48 57.7 58.1 59.6 0.7 2.6
Denmark 1.7
a
4 4 4.8 4.6 20.0 -3.2
Estonia n.d n.d 0 0 0
Finland 10.5 10 10 10 9.7 0.0 -3.3
France 338.2 343.8 373.9 394.7 390.3 5.6 -1.1
Germany 783.9 777 842.1 849.6 845.8 0.9 -0.4
Great Britain 121
a
n.d n.d n.d n.d
Hungary 24.8 35 44.8 58.4 68.3 30.4 16.9
Ireland
b
25.8 25.8 26 26.6 n.d 2.3
Italy 832.3 832.1 959.1 960 979.7 0.1 2.1
Netherlands n.d n.d n.d n.d n.d
Norway 22.2 25.2 26.4 30.9 26.7 17.0 -13.5
Poland 66.5 84 84 72.2 76.3 -14.0 5.7
Portugal 17.5 21.2 22.6 25.4 25.6 12.4 0.6
Slovakia n.d n.d n.d 7.6 n.d
Slovenia 15.9 17.3 23.6 24.8 n.d 5.1
Spain 140.7 153.6 121.1 142.1 149.9 17.3 5.5
Sweden 51.8 55.7 58.5 53.3 52.9 -8.9 -0.8
Switzerland 22.3 22.9 25.1 24.1 21.1 -4.0 -12.3
Total 2636 2572 2812 2481 2853
Full total
c
2602 3033
a) Only aluminium
b) Only lead sheet production
c) Full total calculated by taking most recent available number for the years for which no data are given
Table 1.2: European production data for non-ferrous metal castings (in kilotonnes)
[168, CAEF, 2002], [202, TWG, 2002]

The output of non-ferrous metal alloys is still dominated by light metal castings at a share of
75.1 %, despite a decline by 3.5 percentage points compared to the year before. The share of
copper alloys went down from 10.1 to 9.8 %, and the share held by the producers of zinc alloys
similarly shrank from 8.7 to 7.3 %. The difference was absorbed by miscellaneous non-ferrous
metals and processes that are not detailed in the statistics.

Data on the number of foundries are given in Table 1.3 and Table 1.4. These data show that
there has been a general decline in the number of foundries since 1998, with the loss of about
5 % of the existing foundries each year. This decline is also reflected in the employment
numbers, as given in Table 1.5 and Table 1.6.

Chapter 1
4 Smitheries and Foundries Industry
Country 1998 1999 2000 2001 2002 2000:2001 2001:2002
% change
Austria 28 26 25 24 41 -4.0 70.8
Belgium 40
a
25 24 21 21 -12.5 0.0
Czech Rep. n.d n.d n.d 140 143 2.1
Denmark n.d n.d 12 12 12 0.0 0.0
Estonia n.d n.d 1 1 1 0.0 0.0
Finland 19 23 20 19 19 -5.0 0.0
France 169 167 167 163 159 -2.4 -2.5
Germany 324 310 299 288 273 -3.7 -5.2
Great Britain 228
b
214
b
198
b
188
b
179
b
-5.1 -4.8
Hungary 34 34 32 33 n.d 3.1
Ireland n.d n.d n.d 1 n.d
Italy 310 307 293 291 281 -0.7 -3.4
Netherlands 22 n.d n.d 28 n.d
Norway 12 12 12 11 11 -8.3 0.0
Poland 234 230 230 220 190 -4.3 -13.6
Portugal 62 61 61 61 61 0.0 0.0
Slovakia n.d n.d n.d 12 n.d
Slovenia n.d n.d n.d n.d n.d
Spain 224 221 105 102 98 -2.9 -3.9
Sweden 48 49 49 49 50 0.0 2.0
Switzerland 23 23 23 22 20 -4.3 -9.1
Total 1777 1702 1551 1686 1559
Full total
c
1732 1686 1633
a) Only members
b) Without steel castings
c) Full total calculated by taking most recent available number for the years for which no data are given
Table 1.3: Number of foundries (production units) for iron, steel and malleable iron casting
[168, CAEF, 2002], [202, TWG, 2002]

Chapter 1
Smitheries and Foundries Industry 5
Country Total Pressure Die Casting Other Light casting Other Heavy
metal alloy casting
2001 2002 2001 2002 2001 2002 2001 2002
Austria 63 61 20 21 28 25 15 15
Belgium 12 10 3 3 6 5 3 2
Czech Rep. 58 63 39
a
40 n.d n.d n.d 23
Denmark 8 8 n.d n.d n.d n.d n.d
Estonia 0 0 0 0 0 0 0 0
Finland 22 25 4 6 11 12 7 7
France 288 283 n.d n.d n.d n.d n.d n.d
Germany 414 400 n.d n.d n.d n.d n.d n.d
Great Britain n.d n.d n.d n.d n.d n.d n.d n.d
Hungary 78 n.d 23 n.d 35 n.d 20 n.d
Italy n.d n.d n.d n.d n.d n.d n.d n.d
Netherlands n.d n.d n.d n.d n.d n.d n.d n.d
Norway 10 13 3 3 7 6 4
Poland 290 280 n.d n.d n.d n.d n.d n.d
Portugal 67 54 38 32 12 9 17 13
Slovakia 7 n.d n.d n.d n.d n.d n.d n.d
Slovenia n.d n.d n.d n.d n.d n.d n.d n.d
Spain 55
b
57 n.d n.d n.d n.d n.d n.d
Sweden 84 84 43 n.d 29 n.d 12 n.d
Switzerland 49 48 15 14 23 23 11 11
Total 1505 1386 149 119 151 80 85 75
a) Incl. all light casting
b) Only members
Table 1.4: Number of foundries (production units) for non-ferrous metal casting
[168, CAEF, 2002], [202, TWG, 2002]

Chapter 1
6 Smitheries and Foundries Industry
Country 1998 1999 2000 2001 2002 2000:2001 2001:2002
% change
Austria 3465 3314 3342 3936 3067 17.8 -22.1
Belgium 2823 2299 3260 1847 1936 -43.3 4.8
Czech Republic n.d n.d n.d 17536 14847 -15.3
Denmark n.d n.d 1481 1393 1290 -5.9 -7.4
Estonia n.d n.d 125 133 129 6.4 -3.0
Finland 2326 2058 2027 2090 2045 3.1 -2.2
France 26407
a
25714 25613 24871 24651 -2.9 -0.9
Germany 46944 45157 44896 44796 42748 -0.2 -4.6
Great Britain 24000
b
20000
b
18000
b
16500
b
15900
b
-8.3 -3.6
Hungary 3485 3285 3175 2734 n.d -13.9
Ireland 502 503 509 309 n.d -39.3
Italy 22050 22200 22100 21400 20630 -3.2 -3.6
Netherlands 2462 2122 2119 2148 1830 1.4 -14.8
Norway 1864 1706 1730 1754 1564 1.4 -10.8
Poland 33600 28500 26800 26370 24500 -1.6 -7.1
Portugal 2649 2800 2782 2780 2710 -0.1 -2.5
Slovakia n.d n.d n.d 1925 n.d
Slovenia n.d n.d n.d n.d n.d
Spain 13860 14040 11803 11006 11385 -6.8 3.4
Sweden 3650 3650 3650 3800 3800 4.1 0.0
Switzerland 2400 2300 2400 2400 1930 0.0 -19.6
Total 192487 179648 175812 189728 174962
Full total
c
195273 189728 179930
a) Break in continuity of series
b) Without steel castings
c) Full total calculated by taking most recent available number for the years for which no data are given
Table 1.5: Employment in the foundry industry for iron, steel and malleable iron casting
[168, CAEF, 2002], [202, TWG, 2002]

Chapter 1
Smitheries and Foundries Industry 7
Country 1998 1999 2000 2001 2002 2000:2001 2001:2002
% change
Austria 4029 4179 4349 4585 4398 5.4 -4.1
Belgium 1824 803 800 n.d 558
Czech Republic n.d n.d n.d 5083 5374 5.7
Denmark n.d n.d 377 372 349 -1.3 -6.2
Estonia n.d n.d 0 0 0
Finland 708 744 884 718 730 -18.8 1.7
France 17926 17821 17651 17932 17720 1.6 -1.2
Germany 32000 33000 33000 34500 34390 4.5 -0.3
Great Britain n.d n.d n.d n.d n.d
Hungary 3208 3941 5503 4702 n.d -14.6
Ireland
a
70 70 70 70 n.d 0.0
Italy n.d n.d n.d n.d n.d
Netherlands n.d n.d n.d n.d n.d
Norway 1271 1411 1483 1491 1307 0.5 -12.3
Poland 4433 6500 6200 4130 4100 -33.4 -0.7
Portugal 1200 1230 1280 1380 1350 7.8 -2.2
Slovakia n.d n.d n.d 845 n.d
Slovenia n.d n.d n.d n.d n.d
Spain 5650 5620 4810 5034 4994 4.7 -0.8
Sweden 3700 3700 3700 3700 3700 0.0 0.0
Switzerland 1900 2000 2100 2200 1900 4.8 -13.6
Total 77919 81019 82207 86742 80870
Full Total
b
88135 87300 86487
a) Only lead
b) Full total calculated by taking most recent available number for the years for which no data are given
Table 1.6: Employment in the foundry industry for non-ferrous casting
[168, CAEF, 2002], [202, TWG, 2002]

The tables show that European production levels are relatively stable or are slightly rising but
that this production now results from fewer units and less employees. This can be explained by
progressive upscaling and automation in the foundry units. The relationship between unit size,
production and employment is well illustrated in Figure 1.1. This shows that the larger West-
European producers (Germany, France) are attaining higher productivities with fewer people.
The more labour-intensive units are found in the Eastern and Southern part of Europe (Poland,
Hungary, Portugal).

Chapter 1
8 Smitheries and Foundries Industry
Production (1 Mt)
0
20
40
60
80
100
120
140
160
180
0 20 40 60 80 100
Average production per employee (t)
A
v
e
r
a
g
e
n
u
m
b
e
r
o
f
e
m
p
l
o
y
e
e
s
p
e
r
f
o
u
n
d
r
y
Austria
Norway
Poland
Hungary
Germany
France
Italy
Sweden
Spain
Portugal
Great Britain
Switzerland
Finland
Belgium
Figure 1.1: Ferrous foundry productivity data for various European countries
the size of the circle represents the total production in the specified country

The casting of metal is an ancient activity, dating back to more than 3000 BC. The development
of the European foundry industry is linked with the development of both the metal and the
automotive industries. Present foundries often have a history dating back to the beginning of the
20
th
century. Originally they were usually located on the outskirts of towns, but as villages and
cities have grown around them they are now often surrounded by habitation. The foundry
industry is basically an SME industry, with 80 % of companies employing less than 250 people.
Since castings in general are semi-finished products, foundries are located close to their
customers.

1.1.2 Foundry markets

The main markets served by the foundry industry are the automotive, general engineering and
construction sectors. The relative shares of these sectors as markets for the foundry industry are
given in Figure 1.2. The high dependence on the automotive sector has a major influence on
activities in the foundry sector, and concerns various aspects, such as economy, location, quality
standards, environmental standards, new developments, etc. One example of this dependency is
that the automotive industrys shift towards lighter vehicles, is reflected in the foundry industry
by an increased demand (and thus market) for aluminium and magnesium casting, thus enabling
the growth of these sectors.

Chapter 1
Smitheries and Foundries Industry 9
General
engineering
30%
Construction,
public utilities
10%
Automotive
50%
Others
(aeronautics,
electronics,
medicine,...)
10%
Figure 1.2: Relative sectoral market shares

The market shares differ according to the type of metal. This is illustrated by data from the
Spanish foundry market, as shown in Figure 1.3 and Figure 1.4. The automotive sector takes up
more than 60 % of all iron castings from Spanish foundries. Steel castings on the other hand
(including low alloyed as well as stainless and other alloys) are used for machine parts and in
valve making, and therefore serve a broader range of sectors. Indeed, the largest share of the
market for valve making is taken up by stainless castings.

64%
5% 5% 6%
4%
3%
6%
7%
0%
10%
20%
30%
40%
50%
60%
70%
M
a
r
k
e
t
s
h
a
r
e
A
u
t
o
m
o
t
i
v
e
C
o
n
s
t
r
u
c
t
i
o
n
a
n
d
c
e
m
e
n
t
M
a
c
h
i
n
e
t
o
o
l
s
F
a
r
m
i
n
g
m
a
c
h
i
n
e
r
y
E
l
e
c
t
r
i
c
a
l
m
a
c
h
i
n
e
r
y
D
i
e
m
a
k
i
n
g
V
a
l
v
e
m
a
k
i
n
g
O
t
h
e
r
s
Figure 1.3: Market shares for iron castings (data for Spanish market)
[108, FEAF, 1999]

Chapter 1
10 Smitheries and Foundries Industry
9%
18%
2%
11%
3%
20%
5%
3%
3%
2%
19%
5%
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
M
a
r
k
e
t
S
h
a
r
e
A
u
t
o
m
o
t
i
v
e
C
o
n
s
t
r
u
c
t
i
o
n
a
n
d
c
e
m
e
n
t
E
l
e
c
t
r
i
c
a
l
e
n
e
r
g
y
R
a
i
l
w
a
y
C
h
e
m
i
c
a
l
I
n
d
u
s
t
r
y
M
a
c
h
i
n
e
r
y
f
o
r
P
u
b
l
i
c
W
o
r
k
s
M
i
n
i
n
g
M
a
r
i
t
i
m
e
S
i
d
e
r
u
r
g
y
D
i
e
m
a
k
i
n
g
V
a
l
v
e
m
a
k
i
n
g
O
t
h
e
r
s
Figure 1.4: Market shares for steel castings (data for Spanish market)
[108, FEAF, 1999]

The opening of Europe towards the East has led to the big European producers showing a
growing interest in the existing foundry activity in countries such as Poland, the Czech Republic
and Hungary. Some of the big European companies have invested in these regions. For the East
European countries, the opening of their markets combined with inward foreign investment has
allowed the implementation of new techniques, thereby increasing productivity and reducing
their effects on the environment. Due to the low labour costs in these countries, their
competitive strength lies in jobbing foundries, mainly producing large castings, and in foundries
producing a broad range of products. To compete on the world market, West-European
foundries now focus on their technological skills, selecting niche markets which require
complex castings with high precision, specific quality requirements, or those that require quick
or just-in-time delivery.

1.1.3 Foundry types

Besides the metal type (i.e. ferrous/non-ferrous) the foundry layout is largely dependent on the
size of castings and the series size. A small series foundry is termed a jobbing foundry and a
large series one is termed a series foundry. Foundries may be also be classified according to
the type of metal manufactured, i.e. either a ferrous or a non-ferrous foundry. There is a large
difference between a zinc foundry producing a large series of frames for toy cars and a cast iron
foundry producing rotor housings for wind turbines. Foundries apply different degrees of
automation according to their series size and the repeatability of the work. Concerning applied
techniques, the main distinction is made by the type of melting furnace used (e.g. cupola,
electrical, rotary, ) and the type of mould (e.g. sand moulding, die-casting). These will be
described and discussed further in Chapter 2.

Chapter 1
Smitheries and Foundries Industry 11
1.2 Environmental issues

The foundry industry is a major player in the recycling of metals. Steel, cast iron and aluminium
scrap can all be remelted into new products. The possible negative environmental effects of
foundries result from the presence of a thermal process and the use of mineral additives. The
environmental effects of a foundry process therefore mainly relate to the exhaust and off-gases
and to the re-use or disposal of mineral residues.

1.2.1 Air

Noxious emissions from the melting and treatment of metals are generally related to the use of
additives and fuels or to impurities in the feed. The use of cokes as fuels or the heating of
crucibles with gas or oil-fired burners can cause emissions of combustion products. Also the
application of additives in metal treatment processes generates reaction products. The presence
of impurities (e.g. oil, paint, ) in scrap used for re-melting can potentially cause the
production of the products of incomplete combustion or recombination and dust. Also any dust
generated may contain metal and metal oxides. The evaporation of elements with a high vapour
pressure occurs during melting and small particles of metal escape from the bath. Metallic
particles are also generated during grinding and finishing operations.

In the making of moulds and cores, various additives are used to bind the sand. In the binding of
the sand and pouring of the metal, reaction and decomposition products are generated. These
include both inorganic and organic products. The generation of decomposition products further
continues during the casting cooling and de-moulding operations.

Dust and particles releases are a general issue in all stages of the foundry process, and for all
processes used. Dust is generated in the production and processing of sand moulds and cores, as
well as in the finishing of the castings (both from lost moulds and permanent moulds).

In the foundry process, emissions to air are not limited to one (or several) fixed point(s). The
process involves various emission sources (e.g. from hot castings, sand, hot metal). A key issue
in emission reduction is not only to treat the exhaust and off-gas flow, but also to capture it.

1.2.2 Residues

Sand moulding involves the use of large sand volumes, with sand-to-liquid-metal weight ratios
generally ranging from 1:1 up to 20:1. At the end of the moulding process the used sand can be
regenerated, re-used or disposed off. Additional mineral residues such as slag and dross are
generated in the melting stage when removing impurities from the melt. These should also be
considered for either re-use or disposal.

1.2.3 Energy

Since foundries deal with a thermal process, energy efficiency and management of the generated
heat are important environmental aspects. However, due to the high amount of transport and
handling of the heat carrier (i.e. the metal) and due to its slow cooling, the recovery of heat is
not always straightforward.

1.2.4 Water

In most foundries, water management involves an internal circulation of water, but a major part
of the water still evaporates. The water is generally used in the cooling systems of electric
furnaces (induction or arc) and cupola furnaces. In general, the outgoing waste water stream is
therefore very small. For (high) pressure die-casting, a waste water stream is formed, which
needs treatment to remove organic (phenol, oil) compounds before disposal.
Chapter 2
Smitheries and Foundries Industry 13
2 APPLIED PROCESSES AND TECHNIQUES IN FOUNDRIES

2.1 Overview

2.1.1 The foundry process

A general flow chart of the foundry process is depicted in Figure 2.1. The process can be
divided into the following major activities:
- melting and metal treatment: the melting shop
- preparation of moulds: the moulding shop
- casting of the molten metal into the mould, cooling for solidification and removing the
casting from the mould: the casting shop
- finishing of the raw casting: the finishing shop.

Melting
- ferrous:
cupola
induction
electric arc
rotary
- non-ferrous:
induction
shaft
crucible
reverberatory
Metal treatment
Permanent
mould
maintenance
Lost mould making
- sand mould
- sand core
- inserts
- pattern
wood, plastic,
metal
- lost model
resin, wax
Manual moulding
Moulding automate
Casting
- gravitational pouring - high-pressure
- tilt pouring - centrifugal
- low-pressure - continuous
Cooling
Shake-out / Take-out
Sand preparation
Sand regeneration
Finishing
-removal of casting system
-shot blasting
-deburring
-thermal treatment
Raw materials, chemicals, patterns, dies
Finished casting
Sand Legend:
Figure 2.1: The foundry process

Chapter 2
14 Smitheries and Foundries Industry
Starting from foundry scrap (selected scrap according to a certain chemical composition) or
ingots, the foundry produces finished castings. Usually these are components which will require
further treatment or assembly to yield a final product.

On the moulding side, a basic distinction is made between permanent and lost moulds.
Foundries casting in permanent moulds, buy these metal moulds (dies) externally, but typically
operate an in-house mould repair and maintenance shop. Foundries casting in lost moulds, often
buy wooden, metal or plastic patterns (for their mould design) and operate an in-house pattern
maintenance and repair shop. Moulds, cores and lost models are generally produced as part of
the foundry process.

Traditionally in the foundry sector the main distinction made is between ferrous and non-ferrous
foundries. This is mainly because the applied processes in both sectors differ. Non-ferrous
foundries often apply die-casting techniques. These allow a better surface finish, which is
important for many of the aluminium and brass applications. Due to the high cooling rate,
castings with a high mechanical strength are produced. However, this technique does not allow
the production of massive or large pieces, which require sand casting techniques. Sand casting
techniques are applied in non-ferrous foundries for those products that are not produced in large
series. The non-ferrous metals (and their alloys) discussed in this document are:
- aluminium
- magnesium
- copper
- zinc
- lead.

Ferrous foundries generally apply the lost mould techniques. Due to their greater stiffness and
strength, ferrous alloys are used in different applications to non-ferrous alloys. The size of the
products that can be produced is almost unlimited. Ferrous metals have a higher melting point
and therefore require different melting techniques. The ferrous metals and alloys discussed in
this document are the various types of cast iron (which may be classified according to their
properties or by the graphite type) and cast steel.

Superalloys with a high content of alloying elements, such as nickel, will also be discussed.

Foundries utilise mechanisation and automation depending on the need for reproductivity and
on the series sizes. The most flexible installation is typically the jobbing foundry. This produces
a variety of products in small numbers (<100). In general, this type of foundry applies manual
moulding techniques with resin-bonded sand moulds. The melting furnace works batch wise to
allow an easy change of alloy. This implies the use of induction or rotary furnaces.

For medium-sized series (<1000 parts), mechanised moulding and casting lines are used. Lost
mould foundries utilise mould making machines. This implies the use of green sand, which
allows fast mould making. The size of the mould making machine limits the maximum size of
the castings. Casting can be performed manually or by using a pouring machine. Auxiliary side
processes, such as sand preparation, are operated in a semi-automated way with remote control.
Both continuous furnaces (cupola, shaft) and batch furnaces are used. For non-ferrous alloys,
die-casting techniques are applied.

Large series of small castings are made in flaskless green sand moulding. For specific
applications, die-casting also can be used in ferrous foundries if the final casting quality requires
it, although in reality the technique finds only limited implementation. The main difference for
medium-sized series is the further automation of the finishing, the quality control and the mould
assembly. For die-casting in non-ferrous alloy facilities, further automation is often applied, this
is especially the case in pressure die-casting shops.

Specific casting techniques, such as full mould casting, centrifugal casting and continuous
casting are applied where the product type requires it.

Chapter 2
Smitheries and Foundries Industry 15
2.1.2 Iron casting

Cast iron is an iron-carbon alloy, containing usually between 2.4 and 4 % carbon. The minimum
carbon content is 1.8 %. Silicon, manganese, sulphur and phosphorus are also present in various
amounts. Special grades of iron are produced which contain various levels of nickel, chrome
and other metals. Due to its high carbon content, cast iron has a low melting point and a good
casting ability as compared to steel. Its ductility is low and does not allow rolling or forging.
Variations in properties can be achieved by varying the ratio of carbon to silicon, by alloying,
and by heat treatment.

Depending on the concentration and form of the carbon (lamellar, spheroidal or compact),
various types of cast iron may be defined:
- lamellar iron: carbon in the form of flakes
- nodular iron: carbon in spheroidal form
- compact graphite iron: carbon in bonded form.

The classification of cast iron is often made according to its material properties:
- grey iron: iron with a grey fracture surface. Although this applies for lamellar, nodular and
compact graphite iron, the term is commonly used as a synonym for lamellar iron
- ductile iron: cast iron with an increased ductility. This is one of the effects caused by
nodularisation, but it also applies to malleable iron. The term is commonly used as a
synonym for nodular iron
- malleable iron: iron that is capable of extension or of being shaped under the hammer. This
property is related to a low carbon content, which leaves most of the carbon in bonded form.

Cast iron can be melted in the cupola furnace, induction furnace (generally of coreless type, but
very occasionally can be the channel type) or in the rotary furnace. The electric arc furnace is
only very rarely used for the preparation of cast iron. Figure 2.2 gives process flow diagrams for
the melting and metal treatment of cast iron in the three different furnace types. The process
generally consists of melting tapping metal treatment pouring. The various aspects of
melting and metal treatment are discussed in the following sections. Metal treatment involves
various steps such as desulphurisation, nodularisation, inoculation and deslagging. The
desulphurisation step in cupola melting may also be incorporated into the nodularisation, e.g. by
using a nodularisation process which simultaneously takes up the sulphur, such as the core
wired process.
Chapter 2
16 Smitheries and Foundries Industry
CAST IRON FOUNDRY
Process scheme with the Cupola furnace
Charging of the furnace
Melting
Tapping
Slag granulation or
ingotting
Pouring
Filled moulds
Moulds
Deslagging of ladles Ingotting of excess
metal
Holding of the melt
Desulphurisation
Deslagging
Nodularisation
Ladle repair
Deslagging
Deslagging
Inoculation
Preheated ladles
Preparation of the metallic charge material
Melting
Deslagging
Adjustment of the
composition
Tapping
Pouring
Filled moulds
Preheated transfer
ladles
Moulds
Ladle repair
Deslagging
Ingotting of
excess metal
Deslagging of ladles
Nodularisation -
Inoculation
Melting (intermittent rotation)
Superheating (continuous
rotation)
Holding of the melt
Tapping
Nodularisation - Inoculation
Pouring
Filled moulds
Moulds
Ladle repair
Deslagging
Ingotting of excess
metal
Preheated transfer
ladles
Deslagging of ladles
Charging of the furnace
CAST IRON FOUNDRY
Process scheme with the electric
coreless induction furnace
Preparation
and weighing
of the metallic
charge
Sieving and
weighing
of the
coke
Weighing
of
limestone
Addition
of
ferroalloys
CAST IRON FOUNDRY
Process scheme with the
rotary furnace
Charging of the furnace
Preparation and
weighing of the
metallic charge
Preparation and
weighing additions
(slagging, carb.agents)
Figure 2.2: Process flow diagrams for the melting and metal treatment of cast iron
[32, CAEF, 1997]
Chapter 2
Smitheries and Foundries Industry 17
The cupola is the leading device for re-melting iron in Europe. It is responsible for some 55 %
of the tonnage of iron castings produced in Western Europe. Nowadays, the cupola is
increasingly facing major challenges to its market domination. This is partially due to its flue-
gas quality, which requires treatment. Faced with the possible financial burden of investing in,
and then depreciating, a stack gas treatment installation, many small and medium sized units
have turned to electric or oxygas melting units. Thus the number of cupolas used in foundries is
falling, but their average size is increasing. There have been major changes in the market for
cupolas in Europe in recent years, particularly due to the restructuring of the coke industry,
leading to a decreased number of suppliers and a need to import coke into Europe. Another
major change is the smaller number of cupola manufacturers, with one German firm having a
quasi-monopoly in the hot blast type.

The majority of repetitive iron castings are made in green sand moulds with resin-bonded cores.
The cold-box amine and hot-box techniques are most widely used. The Croning resin shell
moulding process is used where a high precision and good surface finish are needed. The Lost
Foam process is used to a limited extent, for repetition castings. Castings made in smaller
numbers are made in chemically-bonded sand moulds. Special sand processes, such as vacuum
moulding and full moulding are used for certain iron castings. There are also a few permanent
moulding (die-casting) foundries making iron castings, but the short die-life of a mould limiting
it to making only a few thousand components has restricted the use of ferrous die-casting.
[156, Godinot, 2001], [174, Brown, 2000]

2.1.3 Steel casting

Steel is a material of which the (mass) content of iron is bigger than that of any other element,
with a carbon content generally lower than 2%, and which also usually contains other elements.
A limited number of chromium steel types may contain over 2 % carbon, but 2 % is the usual
cut-off limit used to distinguish steel from cast iron [201, CEN, 2000]. One particularly useful
aspect of steel is that it can be hot worked. Low alloy cast steel contains elements such as Mn,
Cr, Ni, and Mo in amounts less than 5 %. High alloy steel includes more than 5 % of alloying
elements, e.g. 12 % Cr and 8 % Ni. Special steel grades are produced with enhanced properties,
such as higher strength; higher magnetic permeability; better resistance to corrosion, fatigue or
wear; and improved behaviour during welding or at high or low temperatures.

Cast steel is normally melted in electric arc furnaces (EAF) or in coreless induction furnaces
(IF). Once melted, the liquid metal can be refined (i.e. removal of carbon, silicon, sulphur and or
phosphorus) and deoxidised (i.e. reduction of metallic oxides), depending on the base material
and the quality requirement of the finished product . Figure 2.3 gives process flow diagrams for
the melting and metal treatment of cast steel in the different furnace types.
[32, CAEF, 1997]
Chapter 2
18 Smitheries and Foundries Industry
STEEL FOUNDRY
Process scheme with the
acid lined electric arc furnace
Preparation of the charge material
Charging of the furnace
Melting
Decarburisation (oxygen injection)
Adjustment of the composition
Pouring
Filled moulds
Preheated transfer ladles
Moulds
Ladle repair
Deslagging
Ingotting of excess metal
AOD refining
Deslagging of ladles
Adjustment of the composition
STEEL FOUNDRY
Process scheme with the
basic lined electric arc furnace
Dephosphorisation
Decarburisation
Deslagging
Pouring
Filled moulds
Preheated transfer
ladles
Moulds
Ladle repair
Deslagging
Ingotting of
excess metal
Desulphurisation
Deslagging of ladles
Tapping and deoxidation
Preparation of the metallic charge material
Charging of the furnace
Melting
Deslagging
Tapping
and deoxidation
Pouring
Filled moulds
Preheated transfer
ladles
Moulds
Ladle repair
Deslagging
Ingotting of excess
metal
Deslagging of ladles
AOD/VODC refining Vacuum refining
Tapping and deoxidation
STEEL FOUNDRY
Process scheme with the
electric coreless induction furnace
Preparation of the charge material
Charging of the furnace
Melting
Adjustment of the composition
Adjustment of the composition
AOD/VODC refining
Figure 2.3: Process flow diagrams for the melting and metal treatment of steel
[32, CAEF, 1997]
Chapter 2
Smitheries and Foundries Industry 19
2.1.4 Aluminium casting

About two-thirds of all aluminium castings are used in the automotive industry, e.g. in cars,
buses, lorries, trains and aircraft. The need to reduce vehicle fuel consumption and weight has
increased the interest in aluminium. The total mass of aluminium in a European car roughly
doubled between 1990 and 2000. This growing use of aluminium in its major user sector clearly
has an effect on the overall number of castings produced.

Aluminium is mainly cast into permanent moulds. The relative shares of applied casting
techniques for Al are given in Table 2.1:

Casting type Relative share
(%)
Pressure die-casting 59
Low-pressure die-casting & gravity casting 37
Sand casting 3
Others 1
Table 2.1: Relative shares of applied casting techniques for Al
[143, Inasmet and CTIF, 2002], [225, TWG, 2003]

Many different types of melting furnaces are used in aluminium foundries the choice depending
on individual requirements. Directly and indirectly heated furnaces, using fuel and electricity,
are applied. The fossil fuels currently used are natural gas, liquid petroleum gas (LPG) and oil.
Natural gas is favoured by most foundries on convenience grounds. Electrical heating may be
provided by either resistance elements or by induction. Capacity is one of the most important
parameters for melting and holding furnaces. Today induction furnaces are normally used when
a high melting capacity, e.g. above 10 tonnes/hour, is needed. Shaft melting and holding
furnaces, as well as crucible furnaces, are often used when the melting capacity is less than five
tonnes/hour. Small and medium crucible furnaces are often used when it might be necessary to
be able to change the alloy easily or if the production rate is low.

For holding, electric furnaces have the advantage of not producing burner off-gases and being
able to sustain a homogeneous temperature over the whole molten volume, at a relatively low
energy expense.

Aluminium melting in foundries generally uses alloyed ingots as a starting material, although in
some cases the metal is delivered already as a liquid. The secondary melting of aluminium scrap
is usually not performed in foundries and falls outside the scope of this document. It is
discussed in the BAT reference document for the non-ferrous metals industries.
[48, ETSU, 1994] [148, Eurofine, 2002], [155, European IPPC Bureau, 2001]

2.1.5 Magnesium casting

Magnesium alloy castings are used for aerospace, automotive and electronic applications. The
main advantage for using them is their light weight; typically magnesium alloys have a density
of 1.8 g/ml compared with 2.7 g/ml for aluminium alloys. Aluminium is the principle alloying
constituent of magnesium-based casting alloys, with zinc and manganese also present in small
amounts. Pressure die-casting is the most commonly used casting process, due to the low
casting temperature (650 700 C); both hot chamber and cold chamber die-casting machines
are used. Sand moulding is applied to a lesser extent. Magnesium die-castings can be made with
thinner walls than aluminium, but their use is limited by stiffness problems. The thinner walls
allow the overall weight of the components to be substantially reduced, thus compensating for
the higher alloy cost per kilogram. Gravity die-casting and sand casting are also used,
particularly for more highly stressed castings. The use of magnesium alloy die-castings in
automotive components is growing rapidly, with some vehicles already containing 10 20 kg of
Mg components. The most popular parts made at present are instrument panel substrates, cross
car beams, wheeldrives and seat frames.
Chapter 2
20 Smitheries and Foundries Industry
Molten magnesium alloys attack firebrick and refractory furnace linings, resulting in harmful
silicon contamination. Therefore steel crucibles are used. Iron is also slightly soluble in
magnesium but it has a much less harmful effect than silicon. Scrap is usually cleaned and if
possible shot blasted to remove any adhering sand as a further precaution against silicon pick-
up. To eliminate ladling, the molten alloy is, if possible, poured direct from the melting pot.

Due to its very easy oxidation, magnesium alloys are melted under a cover using a cleansing
flux or cover gas, to avoid oxidation losses and to prevent inclusions. Inhibitor powders are used
to cover any exposed metal during holding and pouring, and are added to moulding-sand to
prevent chemical reactions. The fluxless melting of Mg alloys requires another form of melt
protection. For this purpose, sulphur hexafluoride (SF
6
) is used, as it promotes the formation of
a protective film on liquid magnesium, which prevents oxidation. It is used at low concentration
(<0.3 vol %) in a mix with air or air/CO
2
. SF
6
is a greenhouse gas, considered harmful to the
atmosphere and falls under the Kyoto protocol, which requires its use to be minimised. Austria
and Denmark have issued regulations to ban the use of SF
6
by 2003 and 2006, respectively.

Magnesium alloys benefit from grain refinement, which is carried out by inoculation with
carbonaceous materials. This used to be done with hexachloroethane, but since 1 July 2003 this
product is banned in Europe generally to ensure environmental protection and for health and
safety reasons. This applies both for magnesium and aluminium alloys.
[175, Brown, 1999], [225, TWG, 2003]

2.1.6 Copper casting

Copper is cast in the form of various groups of alloys, each having copper as the main element.
Short descriptions of some of these are given below:

- High conductivity coppers: These are used mainly for their high electrical and thermal
conductivities. Applications include tuyres for blast furnaces and hot blast cupolas, water-
cooled electrode clamps, switchgear, etc.
- Brasses: Cu-Zn-alloys, where zinc is the major alloying element. These are easy to cast,
with excellent machinability and good resistance to corrosion in air and fresh water. They
are widely used for plumbing fittings. High tensile brasses are more highly alloyed and find
uses in marine engineering. Brasses are cast both in sand and in permanent moulds
- Tin bronzes: Cu-Sn-alloys, where tin is the major alloying element. With tin contents of
10 12 %, tin bronze castings are more expensive than brass. They have high corrosion
resistance and are suitable for handling acidic waters, boiler feed-waters, etc. High tin alloys
are also used in wear-resistant applications. Their applied casting techniques are sand and
centrifugal casting
- Phosphor bronzes: Cu-Sn-alloys, with an addition of about 0.4 1.0 % P. These are harder
than tin bronzes but have lower ductility. They are used for bearings where loads and
running speeds are high and for gears such as worm wheels
- Lead bronzes: Cu-Sn-Pb-alloys. These are used almost exclusively for bearings, where
loads and speeds are moderate
- Gunmetals: Cu-Sn-Zn-Pb-alloys. These are the optimal alloys for sand casting. They have a
good combination of castability, machinability and strength, and good corrosion resistance.
They are used for intricate, pressure-tight castings, such as valves and pumps. They are also
used for bearings, where loads and speeds are moderate
- Aluminium bronzes: Cu-Al-alloys, where Al is the major alloying element. These combine a
high strength with high resistance to corrosion. Their applications range from decorative
architectural features to highly stressed engineering components. They have many marine
uses, including propellers, pumps, valves. They are also used for the manufacture of non-
sparking tools. Al casting techniques are applied
- Copper-Nickels: Cu-Ni-alloys, where Ni is the major alloying element. These are used for
e.g. pipework for marine applications in severe conditions
Chapter 2
Smitheries and Foundries Industry 21
- Copper-beryllium alloys: Beryllium is cast as a copper-beryllium alloy for the production of
parts that require resistance to corrosion and very high mechanical characteristics. These
include plunger tips for die-casting machines, precision parts for the electrical and
mechanics industry, in watchmaking, for tooling, and for measurements instruments. Two
alloy types are used: a copper-beryllium alloy with 2 % Be, and a copper-cobalt-beryllium
alloy with 0.5 % Be. There is a tendency to reduce or exclude beryllium in alloys due to its
known carcinogenic character. Casting is done in permanent moulds using pressure or
gravity die-casting. For precision parts casting, the investment casting technique is used.
[175, Brown, 1999]

2.1.7 Zinc casting

Zinc casting almost exclusively uses the pressure die-casting technique. In the EU, there are
mainly two alloys in use, their compositions are given in Table 2.2. They are also referred to as
Zamac, which in origin is a trade name. The basis of these alloys is pure zinc.

Symbol Alloy number Al Cu Mg
ZnAl4Cu1 ZP0410 3.7 4.3 0.7 1.2 0.025 0.06
ZnAl4 ZP0400 3.7 4.3 0.25 0.025 0.06
Table 2.2: Most common zinc alloys, contents in %

The zinc alloy is almost exclusively melted in a casting machine, of the hot chamber type. In
rare cases, and only when high production capacity is needed, centralised melting may be
applied.

Zinc alloys have comparable material properties to aluminium. The main differences are the
lower melting point and the higher density of the zinc alloys (6.7 g/cm versus 2.6 2.7 g/cm).
They are mostly used for small pieces requiring high precision and a low wall thickness. They
also allow a higher casting speed to be utilised and result in a 10 times longer die life (800000 to
1200000 shots), which makes them more suitable for large series (of small pieces). The zinc
alloys are melted in an electrically or fuel heated cast iron crucible and cast using hot-chamber
die-casting machines. The products are used in e.g. automotive and electronics components and
in machine construction applications.

2.1.8 Lead casting

Lead is a low melting (melting point 327 C) heavy metal. Lead is relatively soft,
corrosion-resistant and has good self-lubricating properties. The uses of lead castings include
accumulator sheets, shielding material for X-rays and nuclear applications, and in ballast and
counterweight materials. Mainly pressure and gravity die-casting techniques are applied.

2.1.9 Casting of superalloys

The superalloys are typically Ni, Ni-Fe, and Co based alloys with Cr, Ti, W, Al additions. They
were originally used for high temperature applications (over 810 C) or in severe corrosive
media. Superalloys can be distinguished from high alloyed steels (see definition in Section
2.1.3). Since iron is not the major compound (as defined in [201, CEN, 2000]), they are
considered non-ferrous materials. The casting of superalloys may occur in certain investment
casting foundries, as well as partly in foundries which specialise in high alloyed steel qualities.

The nickel base alloys are produced from a group of alloys which have chemical compositions
generally over 50 % nickel and less than 10 % iron. They are mainly strengthened by
intermetallic precipitation in an austenitic matrix. The cobalt base alloys have a high Co content
(40 to 70 %), high Cr (over 20 %), high W (7 to 15 %) and they are strengthened by a
combination of carbides and solid solution hardeners.
Chapter 2
22 Smitheries and Foundries Industry
Some superalloys, particularly Ni-Fe and Co based alloys, are directly melted in electric
furnaces by classical methods usually applicable to stainless steels. However for Ni and special
Ni-Fe superalloys, vacuum induction melting is required in order to reduce the content of
interstitial gases (O, H, N) to a very low level. This enables foundries to achieve high and
controlled contents of oxidisable elements such as Ti or Al.

The control of interstitial gases and oxidisable elements is very important for the products
mechanical properties, the corrosion resistance and its reliability. In general, superalloys are cast
into complex final shapes where machining is not possible. Therefore, they are mainly produced
by investment casting (i.e. using a ceramic mould). This casting process produces a product of
very precise dimensions with a very smooth surface. Additional processes, such as HIP (hot
isostatic pressing), can be used to eliminate the internal porosity that can appear in large
castings. In aircraft gas turbine manifolds, directional casting technology is commonly applied.
This technology eliminates the grain boundaries and greatly increases the strength of the
material.

Initially superalloys were developed for high temperature applications. However, their field of
application continues to expand and now covers areas such as cryogenic temperature appliances
and orthopaedic and dental prostheses. In general, superalloys are mainly used in aircraft and
industrial gas turbines, in nuclear reactors, in aircraft and spacecraft structures, in petrochemical
production and in medical applications.
[202, TWG, 2002]

2.2 Pattern making

2.2.1 General pattern making

Pattern making, or foundry tooling as it is also called, requires a high level of skill to achieve
the close tolerances required of the patterns and coreboxes. This step is critical in the casting
process since the castings produced can be no better than the patterns used to make them.
Patterns are made by means of hand tools, universal machines, or by a CAD/CAM system on
computer-numerical-controlled (CNC) machines. In some pattern making shops, computer-
aided design (CAD) is used in the design of patterns. Cutter tool paths are designed with
computer-aided manufacturing (CAM). The numerical output from these computers is conveyed
to CNC machine tools, which then cut the production patterns to shape. Such computer-aided
systems have better dimensional accuracy and consistency than manual methods.

Patterns (Figure 2.4) and corebox materials (Figure 2.5) are typically metal, plastic, wood or
plaster. Wax and polystyrene are used in the investment and Lost Foam casting processes,
respectively. Pattern makers have a wide range of tools available to them, including
woodworking and metal machining tools. Mechanical connectors and glues are used to join
pattern pieces together. Wax, plastic or polyester putty is used as a filler to fill or round the
inside of square corners.
[42, US EPA, 1998]

Chapter 2
Smitheries and Foundries Industry 23
Figure 2.4: Wooden pattern
[237, HUT, 2003]

Figure 2.5: Coreboxes
[237, HUT, 2003]

2.2.2 Rapid prototyping (RP)

Rapid prototyping is a technique to pass very quickly from a product concept to a cast
prototype. The term Rapid Prototyping includes all technical and organisational measures
from the formulation of the concept of a product to the manufacture of the product. Rapid
prototyping can be used for every stage of product development, i.e. for concept models,
geometrical prototypes, functional prototypes or for technical or sales prototypes. All the
currently available techniques allow the fabrication of a prototype part from a three-dimensional
drawing. They are also used for the direct sintering of sands for the production of moulds or
cores. The technique consists of building an object to a design pattern by joining particles or
layers of raw material such as polymer resin, wax, paper or ceramic powder. An example
picture of a wax pattern is given in Figure 2.6

Chapter 2
24 Smitheries and Foundries Industry
Figure 2.6: Thermopolymer (wax) patterns in an RP machine
[237, HUT, 2003]

There are 4 basic steps to rapid prototyping:
1. the creation of a 3D-CAD model
2. making the interface between the 3D-CAD system and the rapid prototyping system. For
example the CAD data may be converted to the STL (Standard transformation format) file
format
3. slicing the STL file into thin crosssectional layers
4. realisation of the RP-model.

The advantages of Rapid Prototyping, particularly for complex forms, include:
- shorter development time and the possibility of rapid modifications during the development
- saving of costs, materials and time
- early error detection.
[202, TWG, 2002], [203, Linxe, 2002]

A survey of available techniques is given in Table 2.3.

Process Principle Materials Special features
Liquid solid Polymerisation by local UV
exposure of a photosensitive
resin
Photosensitive resins
Acrylates, epoxydes
- shrinkage and
deformations
- model needs
supports
Solid solid Contours of a layer are cut out
from a foil
Paper, metals, polymers - no supports
- consumption of
base material
Deposition of a material
showing thermal fusion
ABS, wax, elastomers - model needs
supports
Material jet Waxes, polymers - model needs
supports
Powder solid Agglomeration of powder by
sintering
Metals, ceramics, sand,
polystyrene, nylon,
polyamide, wax
- no supports
- porosity
- shrinkage
Agglomeration of powder by
binder projection
Alumina - no supports
- porosity
Table 2.3: Description of rapid prototyping techniques
[203, Linxe, 2002]
Chapter 2
Smitheries and Foundries Industry 25
2.3 Raw materials and raw material handling

The main flows of raw materials entering the foundry are metal ingots, foundry scrap and sand.
A distinction can be made between ferrous and non-ferrous foundries. Non-ferrous foundries
generally melt only internal return material and alloy ingots (Figure 2.7). The re-melting of
external scrap is generally considered a separate activity, usually forming part of the secondary
metal production. If external scrap is acquired, it is first subjected to a spectroscopy analysis in
order to determine the alloy type. Ferrous foundries use pig iron and selected iron and steel
scrap as starting materials, besides internal return material. The various qualities of metal feeds
are stored in separate areas in order to allow the controlled feeding of the melting furnace.

Figure 2.7: Aluminium scrap(l.) and ingots (r.)
[237, HUT, 2003]

Raw materials, including fluxes in lump and powder form; foundry cokes for cupola furnaces;
deoxidants; and refractories are normally stored under cover. Following delivery, handling is
kept to a minimum. Powdered materials may be stored in sealed silos and conveyed
pneumatically or kept and handled in sealed bags, as displayed in Figure 2.8.

Figure 2.8: Pneumatic conveyors and silos for powdered materials
[237, HUT, 2003]

Sand is normally delivered in bulk and discharged directly to a silo via a pneumatic conveyor,
conveyor belt or grab. Specialist sands may arrive in bags or by tanker. Used sands are stored in
silos for regeneration and in silos or heaps for transport for external re-use or disposal.

Liquid binders and oil products are delivered in drums, by bulk container or by road tanker.
They are stored in their delivery containers or, in the case of rail tankers, discharged direct into
a dedicated storage. The containers are connected by pipe directly to the sand/resin/catalyst-
mixing unit. Some catalysts and co-reactants are used in a gaseous form, but these are also
delivered as liquids and handled in a similar fashion before being vaporised and mixed with a
carrier gas. Evaporation is enclosed and may be carried out by a variety of methods.
Chapter 2
26 Smitheries and Foundries Industry
Refractories, release agents and other minor deliveries are stored indoors.

Coarse solid residues, such as used refractories and slags, are stored on separate heaps, in a
subdivided storage area or in boxes. They are moved and handled using small lift trucks. Fine
solid residues are collected at the filter unit into big bags or containers, which may be stored
intermediately before transport for disposal. In order to avoid soil contamination from the
various classes of materials, specific measures are taken. Potential impacts are listed in Table
2.5.

The metal charged in the melting furnace is carefully selected and weighed to ensure the correct
composition. The charge composition is calculated based on the average chemical composition
of each component, the oxidation losses during melting and the required final composition of
the casting. The different charge elements are combined into a charging device (e.g. drop
bottom skip, vibrating feeder, skip hoist) with a tilting magnet, usually equipped with a
weighing system, to allow collection of the correct amount.

Additional alloying elements can be added to the charge as ferro-alloys such as FeSi, FeMn,
FeCr, or pure such as Cu, C, Ni. However most alloys are added to the molten metal to prevent
metallic losses due to oxidation. Alloying elements are usually present in the foundry in small
quantities and are always stored inside the building, preferably close to the melting installation.
[32, CAEF, 1997].

Chapter 2
Smitheries and Foundries Industry 27
Activities Contaminants of Concern Potential Impact
Raw materials storage
Loose materials from scrap - metals
and coatings
Localised contamination of
soil surface
Oils - may include:
- PAH from combustion engine oils
- PCBs from capacitors (mainly in
shredded scrap, unless PCB phase-out
has taken place successfully)
Scrap storage external, often
on soil
Cutting fluids - may be chlorinated
Leaching into groundwater
and nearby surface waters
Liquids in underground tanks Petroleum products, e.g. fuel oil
diesel
Leakage into soil with
leaching into ground and
surface waters
Raw materials handling, transfer and use
Chemical binder products
accidental spillage or leaks
Phenolic resins Free phenol content
of 0.5 to 5 %
Water-soluble; can leach
into groundwater.
Rapid degradation if less
than 400 ppm.
Concentrated larger volume
spills will be slower to
degrade due to toxicity to
bacteria
Solvent carriers
in resins
e.g. methanol, up
to 20 % volume
Mostly water-soluble, can
leach into groundwater
Urethane resins Aromatic solvents
1
Solvents could leach into
groundwater
Furan hardeners Toluene-, Xylene-,
Benzene-sulphonic
acids
Could leach into
groundwater or alter soil
properties, e.g.
mobilisation of metals
Urea-based
resins
Ammonia Can leach into ground and
surface waters
Silicate resins Alkaline pH Could leach into
groundwater or alter soil
properties, e.g.
mobilisation of metals
Mould coatings Isopropyl alcohol (IPA) Water-soluble, can leach
into groundwater
Fuels, maintenance products:
- accidental spillage or leaks
(particularly around filling
points)
- deliberate disposal to ground
Fuel oils, diesel, petrol with PAH &
SO
4
2-
post-combustion;
Lubrication & hydraulic oils;
Quench oils;
Transformer oils (potential for PCBs)
Oils can leach into ground
and surface waters.
Air deposition of PAH and
SO
4
2-
from the combustion
of fuel oils.
SO
4
2-
will leach down soil
profile.
PAH, PCBs will tend to
absorb onto soils
1
Former formulations contained PAH (mainly naphtalene), but naphtalene depleted formulations
have been developed, which have eliminated this contaminant.
Table 2.4: Potential soil contamination from ferrous foundry raw materials
[140, EU Thematic Network Foundry Wastes, 2001], [225, TWG, 2003]

Chapter 2
28 Smitheries and Foundries Industry
2.4 Melting and metal treatment

The selection of a melting furnace is an important aspect in the setting up of a foundry process.
Each furnace type has its own properties concerning feed requirements and alloying
possibilities, which in turn will have repercussions on the full foundry process. On the other
hand, the type of metal to be melted determines which furnace may or may not be used. The
applicability of the various furnace types is given in Table 2.5.

Cupola Electric arc Channel
induction
Coreless
induction
Rotary Hearth type Shaft Crucible/
Ladle

Iron m m* h m, h m* m h
Steel m m h
Aluminium m, h m, h m m m m, h
Magnesium m, h
Copper h m, h m m, h
Lead m, h m, h
Zinc m, h m, h
* Less common
Table 2.5: Applicability of furnace types, for melting (m) and holding (h)

Iron foundries require metal of a controlled composition and temperature, supplied at a rate
sufficient to match the varying demand of the moulding line. The metallic charge to be melted
usually consists of foundry returns, iron and steel foundry scrap and pig iron, with alloying
additions such as e.g. ferrosilicon, ferrophosphorus or ferromanganese. The charge is usually
melted in a cupola or in an electric induction furnace. Induction furnaces are gradually gaining
higher market preference compared to the cupola type. Coreless induction furnaces are used for
melting. Channel induction furnaces are only used for holding, their main application being in
combination with the cupola furnace, in the so-called duplex configuration. Gas-fired and oil-
fired rotary furnaces can also be used, although their use is less common. Short-term holding,
transport and metal treatment are performed in ladles.

Steel is melted in both electric arc and induction furnaces. Large steel foundries may use electric
arc furnaces, but induction furnaces are more commonly used. Arc furnaces are capable of using
low cost scrap charges, since refining takes place in the furnace. However, they have the
limitation that there is always some carbon pick-up from the graphite electrodes, so very low
carbon stainless steels (<0.03 % C) cannot be made. In the induction furnace, refining is not
possible, so a carefully selected charge must be used. However, any type of steel may be melted.
Short-term holding, transport and metal treatment are performed in ladles.

The melting furnace used in non-ferrous melting is dependent on the foundry size. Non-ferrous
foundries often use a variety of different alloys and/or have a limited melting capacity. Melting
is done in small volume furnaces, for which the crucible furnace is most suited. Additionally,
die-casting is the major casting technique. In this case, there often is no need for a centralised
melting, as the melting (and holding) furnace is integrated into the casting machine. Non-ferrous
foundries with a higher capacity and a need (or reason) for centralised melting typically use
induction, hearth type or shaft furnaces for melting, and then distribute the molten metal to
holding furnaces and casting crucibles.
[174, Brown, 2000], [225, TWG, 2003]

Chapter 2
Smitheries and Foundries Industry 29
2.4.1 Cupola furnaces

2.4.1.1 Cold blast cupola furnace

2.4.1.1.1 Description

The cupola is a refractory lined shaft furnace where the metal charge is heated by the
combustion of coke, which takes place in the lower part of the shaft (the hearth). Combustion
air, supplied by fans, is injected in the hearth through a number of exhaust nozzles (tuyres).
A ring and control valve allows a controlled and evenly distributed flow of combustion air
through the tuyres. The metal (pig iron, steel scrap, scrap iron, foundry returns), coke, alloying
elements (e.g. FeSi, SiC), slag forming (SiO
2
) and fluxing agents (e.g. CaCO
3
) are added to the
shaft through a charging door at the upper part of the shaft. The combustion gases move upward
from the hearth and exchange heat with the charge, before leaving the furnace through the
cupola stack.

When the preheated charge reaches the combustion zone, the metallic parts melt due to the high
temperatures, and the charged coke starts to burn in the presence of oxygen. The molten metal
droplets run through the coke bed and gather in the zone called the well, which is below the
combustion zone. All the impurities are trapped in the slag, which is mostly formed by SiO
2
,
CaO, Al
2
O
3
and FeO. Fluxing agents lower the melting point and the viscosity of the slag. Due
to its lower density, the slag floats on the molten metal in the well. Once the liquid metal in the
well has reached a certain level, a tap-hole is opened. The metal flows discontinuously through
the tap hole, via a refractory lined channel or launder into a separate collection vessel or ladle.
Alternatively, the molten metal can be continuously directed to a holding furnace.

The slag is tapped separately by means of a dam and a slag spout placed at a higher level. It is
collected discontinuously in pots, or continuously granulated in a water stream, or in a special
installation for dry granulation.

In its basic configuration, the cupola is called a cold blast cupola (CBC). This is a cupola
which uses the blast at atmospheric pressure and at normal environmental temperature.
Figure 2.9: Schematic outline and miniature model of a (cold blast) cupola furnace
[44, ETSU, 1993], [237, HUT, 2003]
Chapter 2
30 Smitheries and Foundries Industry
In order to reduce the CO-emission and increase the energy efficiency, the conversion of cold
blast furnaces to hot blast furnaces may be considered (see Section 4.5.2). However, for
medium-sized foundries handling less than 2000 tonnes/month of good castings, the hot blast
cupola will be difficult to consider because of the large investment it requires. The use of a hot
blast cupola is also restricted to continuous melting, otherwise the running costs are too high.
The cold blast cupola prevails for some types of production, e.g. enamelled iron castings,
kitchenware, heating appliances, counterweights. These types of castings allow a high level of
old cast irons to be used in the melting bed; a charge material still plentiful in the countries of
the old industrial Europe and one that is well suited to the cold blast cupola.
[44, ETSU, 1993], [156, Godinot, 2001], [202, TWG, 2002]

2.4.1.1.2 Maintenance

One specific feature of the cupola furnace is that the lining material (quartz-clay-mixture) of the
furnace at the melting and heating zone only lasts for one melting campaign. The intense heat
and the presence of slag results in a chemical dissolution and mechanical wearing of the lining,
which consequently converts it into slag. Cupola furnaces are therefore normally constructed in
pairs. While one furnace is melting, the second one can be lined with new refractory material,
with the operation being switched around the next day of use.
[110, Vito, 2001]

2.4.1.1.3 Advantages:

- the investment cost is EUR 125000 150000 per tonne installed per hour, inclusive of a
dedusting installation
- different kinds of (cheap) scrap can be used due to the cleaning and carburising operation
- the thermal efficiency is acceptable if appropriate measures are utilised.

2.4.1.1.4 Disadvantages:

- the production regime is not flexible
- difficult production management because of the slowness of the system
- expensive charge with pig iron, and little steel scrap
- no quick alloy change possible
- sulphur pick-up/take-up in the cupola
- an environmental burden is caused by the foundry; i.e. lots of dust, slag and refractory
lining
- big dedusting installations are needed because of the high flue-gas rate.
[110, Vito, 2001]

2.4.1.2 Hot blast cupola furnace

2.4.1.2.1 Description

In order to optimise the efficiency of the cupola furnace, the combustion air may be preheated.
This principle is used in the hot blast cupola (HBC), i.e. a cupola which uses a preheated blast.

The advantages of the hot-blast operation may be summarised as follows:
- reduced coke consumption
- increased metal temperature
- increased melting rate
- reduced sulphur pick-up
- reduced melting losses
- increased carbon pick-up and hence the ability to substitute steel scrap for pig iron in the
furnace charge.
Chapter 2
Smitheries and Foundries Industry 31
It should be noted that not all these advantages can be attained at the same time. For example,
an increase in the proportion of steel scrap in the furnace charge may require an increase in the
proportion of coke for recarburisation; this in turn will reduce the melting rate and increase the
sulphur pick-up.

Two methods of heating are:
- Recuperative heating: This involves the transfer of the residual (latent) heat of the flue-
gases to the combustion air. The flue-gases are collected at the top of the furnace, mixed
with sufficient air and then burned in a post combustion unit. This provokes the exothermic
oxidation of CO. The burnt gases are led through a heat-exchanger (recuperator) where the
heat is transferred to the combustion air. Typically the blast air is heated at temperatures of
500 to 600 C. Above these temperatures, problems arise with the sintering of furnace dust
on the surface of the recuperator
- External heating: Here the combustion air is heated by some external means, e.g. by a gas
or fuel burner, by electrical resistance or by a plasma torch.

The combination of these two heating methods, permits the superheating of the blast air up to
1000 C. These high temperatures, however, require the use of more expensive refractory
materials and may cause too high a melt temperature.

Recuperative systems offer increased energy and thermal efficiencies. The effect of air
preheating on thermal efficiency and coke use is depicted in Figure 2.10. It should be noted that
the coke quality may affect the overall blast efficiency.

Figure 2.10: Effect of air preheating on blast furnace efficiency
[44, ETSU, 1993]

For medium-sized foundries producing up to 2000 tonnes/month of good castings, the hot blast
cupola is difficult to consider, in particular because of the large investment it requires. In these
instances, the cold blast cupola prevails for some types of production. The hot blast cupola
remains the most widely applied melting device for mass production foundries, e.g. for parts for
the automobile industry, centrifugal casting, road accessories.
Hot blast cupolas are normally set up for long campaign operation, in order to minimise process
switch overs and maintenance time and effort.
[32, CAEF, 1997], [44, ETSU, 1993], [156, Godinot, 2001]

Chapter 2
32 Smitheries and Foundries Industry
2.4.1.2.2 Advantages:

- reduced coke consumption
- high tap temperature
- high melting capacity
- less sulphur pick-up in the cupola
- possibility to use different kinds of cheap lower grade ferrous scrap
- more scrap steel can be used because of the higher pick-up of the carbon by pig iron.

2.4.1.2.3 Disadvantages:

- very expensive investment, due to the additional environmental measures required
- production regime is not flexible
- difficult production management because of the slowness of the system
- limited to iron alloys only, no quick alloy change possible
- environmental burden caused by the foundry, e.g. lots of dust, slag and refractory lining
- big dedusting installations are needed because of the high flue-gas rate.
[110, Vito, 2001]

2.4.1.3 Long campaign cupola

A long campaign cupola is usually a water-cooled refractory lined cupola that may be hot blast
or cold blast. Such cupolas are operated daily for one, two or three shifts and are very often used
only as a single unit. This form has a campaign life of several weeks or months. The liningless
cupola allows a much longer campaign life but heat losses through the furnace shell can be
significant. Further developments in refractories and operating practices are continually being
made which improve the life and cost effectiveness of the long campaign cupola. A schematic
representation of a long campaign cupola with an in-shaft afterburner is given in Figure 2.11.

Figure 2.11: Schematic representation of a long campaign cupola
[150, ETSU, 1998]

Chapter 2
Smitheries and Foundries Industry 33
To allow for long campaign operation the following measures need to be taken:
- apply a more persistent refractory lining of the shaft, bottom and the hearth
- apply water cooling of the furnace wall: this keeps the wall temperature low and thus
prevents rapid wearing of the hearth lining
- use water-cooled blasting pipes that penetrate deeper into the furnace shaft. Here, the
combustion zone is not in direct contact with the furnace lining. Water-cooling has also
been used for reasons other than only reducing the consumption of refractories, such as:
to extend the duration of the melt
to enable the internal diameter of the furnace to be increased, thus allowing a higher
melting rate.

In the liningless operation, the charge is in direct contact with the water-cooled steel furnace
shell. Liningless operation is only used on relatively large capacity cupolas, melting for long
campaigns. This results in the following advantages:
- only one cupola is required
- the cupola only requires internal repairs after one, or several, weeks operation
- daily bed-coke consumption is reduced
- monitoring is easier throughout the campaign, because of the constant diameter and the
more consistent melting conditions.

A specific type of long campaign cupola is the cokeless cupola. This is discussed in Section
4.2.1.8.
[32, CAEF, 1997], [44, ETSU, 1993], [110, Vito, 2001], [150, ETSU, 1998]

2.4.1.4 Nature of atmospheric emissions

Cupolas can be charged with a wide range of materials, many of which may contain loose
particles such as rust, sand and non-ferrous materials. The metallurgical-coke can break and
produce small pieces, as can the added fluxing materials. Breakage and mechanical abrasion
during charge preparation, as well as during charging itself, generate particles, some of which
are immediately emitted.

During melting, abrasion of the charge against the refractory lining will also generate dust.

A third source of particulate matter is coke ash, generated in the melting zone, which is not
trapped by the slag phase.

Particulate matter of various sources, if light enough, can be entrained in the combustion gases
of the cupola. Under certain conditions metallurgical-fume may be generated from the melting
zone, leading to a visible plume from the cupola stack. The smoke particles consist of
submicron agglomerates of spherical soot particles and metallic oxides, such as ZnO, PbO, etc.,
if the metals are present in the charged steel or iron scrap, such as in galvanised or painted
scrap. The smoke emission will increase with the proportion of coke and contaminants in the
charge, the blast temperature, and the oxygen injection rate.

Carbonaceous smoke is airborne matter formed by the incomplete combustion of organic matter
in the cupola. Scrap contaminants such as oil and grease, wood, textiles and, rubber will form
oily vapours in the stack gases. Vapours and partially burnt organic matter may carry unpleasant
smells.

Again, scrap cleanliness and its nature significantly effect the nature of the emissions. The
burning of coke creates odorous gas emissions containing CO
2
, CO and SO
2
. Decreasing the
proportion of the coke charge (by increasing the thermal efficiency) or (partial or complete)
substitution of the coke by natural gas can help reduce the levels of these substances.
Chapter 2
34 Smitheries and Foundries Industry
2.4.2 Electric arc furnace (EAF)

2.4.2.1 Description

The EAF is a batch-melting furnace consisting of a large bowl shaped refractory lined body
with a dish shaped hearth. The wide furnace shape allows the handling of bulky charge material
and leads to efficient reactions between the slag and metal. Typically the shell diameter is 2 to
4 m. As shown in Figure 2.12, the furnace is covered by a refractory roof, which has ports for
three graphite electrodes. The electrodes are supported by arms, which allow movement up and
down. Most furnaces use roof charging: by moving the roof and electrodes aside, the furnace
can be charged using a drop bottom charging bucket or a magnet. The metal charge is heated by
an electric arc, which is created by a three phase alternate electrical current between the three
graphite electrodes. These are positioned above the charge, which itself acts as the neutral.

Figure 2.12: EAF furnace
[237, HUT, 2003]

The furnace is tapped by tilting it, forcing the metal to flow out through the spout. Opposite the
spout, an operable door allows deslagging and sampling operations to be carried out prior to
tapping.

The lining of the furnace may be acidic (SiO
2
based refractory) or basic (MgO based refractory).
A basic lining allows the use of virtually all kinds of steel scrap. The furnace can also be used
for the production of high alloy and manganese steels. If scrap with high phosphor or sulphur
conent is used, lime and limestone which are usually added for dephosphorisation and
desulphurisation. An acidic lining would be attacked by these compounds. Therefore, the acidic
type refractory is used for melting scrap with a low sulphur or phosphorus content only.

Electric arc furnaces are almost exclusively used for the melting of steel. Only in a few cases are
they used for cast iron production, which requires an addition of coal dust to the melt. Electric
arc furnaces designed for steel foundries' purposes usually range from 2 to 50 tonnes capacities.
They can be run intermittently and are suitable for a wide range of steel analyses. They can
provide steel at high temperatures, with typical meltdown times of about one to two hours,
while achieving high thermal efficiencies of up to 80 %
1
. Power consumption varies from 500 to
800 kWh/tonne of molten steel, depending on the furnace capacity, the hot metal consumption,
and the refining techniques, tapping temperature and pollution control equipment applied. The
total melt time is typically 1 to 4 hours.

1
Without taking into account the efficiency of electric power generation.
Chapter 2
Smitheries and Foundries Industry 35
2.4.2.2 Melting and refining with the acidic lined EAF

Due to the chemical nature of the acid lining (SiO
2
), the refining abilities of this type of furnace
are restricted to decarburisation. Consequently considerable care has to be exercised in selecting
the charge for acceptable sulphur and phosphorus levels, as these cannot be removed in the acid
melting practice. The charge consists of balanced quantities of pig iron, foundry returns and
purchased scrap. The carbon in the charge is held at a high enough level so that after melting it
is 0.2 to 0.4 % above the final level.

The decarburisation starts by injecting oxygen into the metal bath. This creates a strong stirring
action during which carbon in the melt is burned. At the same time the boiling burns out Si,
and flushes out H
2
and N
2
from the metal bath. All impurities (oxides) are trapped in the slag.
During melting, sand (SiO
2
) may be added to bring the slag to the proper consistency. When the
carbon reaches the proper concentration, oxygen injection is stopped, and silicon and
manganese are added to halt the boiling reaction.

After slag removal the metal composition is controlled and adjusted if necessary. Finally the
metal is deoxidised by adding aluminium or other agents into the melt stream during tapping, to
prevent the formation of CO bubbles during solidification. Additional desulphurisation and or
dephosphorisation refining can be executed in an AOD or VODC converter (see Section 2.4.9
and 2.4.10).
[32, CAEF, 1997], [161, UK Environment Agency, 2002]

2.4.2.3 Melting and refining with the basic lined EAF

The alkaline MgO based lining of this EAF makes it possible to refine the metal in the furnace
itself. Therefore it is possible to charge the furnace with virtually any combination of scrap and
foundry returns. The basic lining practice is used when the purchased scrap contains higher
phosphorus and/or sulphur levels than desirable.

Dephosphorisation of the melt is performed by periodic additions of lime during meltdown.
Upon the injection of oxygen in the bath, phosphorous oxide is formed and trapped in the slag,
together with other metallic oxides and impurities. The lime keeps the slag very basic, which
stabilises the phosphorous oxide. At the same time carbon is burnt out. After sufficient reaction
time, the oxygen injection is stopped and the slag is fully removed.

Desulphurisation takes place in a second stage, in a similar way but at a higher temperature.
Again, lime or limestone is added to the melt, reacting with sulphur to form insoluble CaS, that
is trapped by the slag. Periodic additions of carbon, aluminium, or FeSi reduce the metallic
oxides (e.g. manganese-, chromium oxides), and thereby minimise losses of these elements
from the metal bath. All other impurities (oxides) are trapped in the slag and removed during the
final deslagging operation.

After refining the metal composition is controlled and adjusted if necessary. Finally the metal is
deoxidised by the addition of aluminium or other agents in the metal bath prior to tapping, to
prevent the formation of CO bubbles during solidification. Further metal treatment, using an
AOD or VODC converter (see Section 2.4.9 and 2.4.10) may be applied if the final alloy
composition requires it.
[32, CAEF, 1997], [161, UK Environment Agency, 2002]

2.4.2.4 Nature of atmospheric emissions

The emissions from EAFs originate from the charging, melting, and refining operations and
during tapping of the furnace.

Chapter 2
36 Smitheries and Foundries Industry
During charging, dust and dirt will be emitted from the open furnace body. When charging a hot
furnace (for instance when melting with a molten heel in particular), any combustibles such as
grease, paint or oil ignite and give rise to smoke plumes of burnt and partially burnt organic
material and dust particles. The mechanical abrasion of the furnace lining also generates
additional dust.

During melting, heating of the scrap generates metal oxide fumes that significantly increase
during the decarburisation treatment. The injection of oxygen gas into the molten metal
develops significant quantities of iron oxide fumes, which leave the furnace as red clouds. The
addition of slag forming materials increases the furnace emission, but only in small quantities
and only for a short time.

Minor emissions occur during the transfer of the molten metal into a ladle or holding furnace.

Emissions from the melting operation itself are referred to as primary emissions. Secondary
emissions are the fumes and dust originating from the charging and tapping.

In considering the nature and quantities of the emissions, the different sources of dust and fume
emissions, as described above, demonstrate that large differences exist between foundries,
depending on the cleanliness of the charged material, the applied charging procedure, the
composition of the charge, the refining treatments and on the additions to the melt. Since no
coal or fuel is burned in the furnace the emissions solely depend on these parameters.

2.4.3 Induction furnace (IF)

Induction furnaces are used to melt both ferrous and non-ferrous metals. There are several types
of induction furnaces available, but all operate by utilising a strong magnetic field created by
passing an electric current through a coil wrapped around the furnace. The magnetic field in turn
creates a voltage across, and subsequently an electric current through, the metal to be melted.
The electrical resistance of the metal produces heat, which in turn melts the metal. Induction
furnaces are made in a wide range of sizes. Because there is no contact between the charge and
the energy-carrier, the induction furnace is suited for the melting of steel, cast iron and non-
ferrous metals, so long as a suitable lining material can be found.

Proper functioning of the water cooling circuits is crucial to prevent the coil from overheating.
The water cooling systems are therefore designed to provide the highest level of reliability, and
thus incorporate various thermostats and flow-meters.
[42, US EPA, 1998], [47, ETSU, 1992], [110, Vito, 2001]

2.4.3.1 Coreless induction furnace

2.4.3.1.1 Description

The coreless IF is a batch-melting furnace containing a water-cooled copper coil, the inside of
which is internally refractory lined. The outside is insulated and enclosed in a steel shell. The
furnace body is mounted in a frame equipped with a tilting mechanism. A coreless induction
furnace is normally a refractory-lined bucket-shape refractory, the top of which is open for
charging and deslagging operations (see Figure 2.13).

Chapter 2
Smitheries and Foundries Industry 37
Figure 2.13: General arrangement of coreless induction furnace
[47, ETSU, 1992]

According to the capacity, the furnace is charged by a lifting magnet, bucket skips, a vibrating
conveyor or manually. A large number of foundries use induction furnaces for producing
relatively small lots in a large variety of compositions. Furnace capacities range from 10 kg up
to 30 tonnes. The core is fed by mains, medium or high frequency alternating current (50, 250
or 1000 Hz, respectively).

Depending on the installed power density and the melting practice the thermal efficiency can
exceed 80 %
2
, but usually ranges from 60 to 70 %. If the efficiency of the electric power
generation is taken into account, an overall efficiency of 15 20 % results, which is rather low
in comparison with other furnace types.

The coreless induction furnace can be designed to operate at any frequency from 50 Hz
upwards. The induction heating of liquid metal causes a stirring effect. The lower the frequency
of the primary current, the more intense is the stirring. Therefore, in a mains frequency furnace
working at 50 or 60 Hz, the turbulence is greater than in one operating at higher frequency.
Because of the high turbulence, the power input to a mains frequency furnace is restricted to
around 250 kW per tonne of capacity. With higher frequencies, the power density can be
increased to three or four times this level.

The frequency of operation also affects the current penetration. The higher the frequency, the
lesser the penetration depth. This affects the minimal charge piece size and the effective furnace
size. 50 Hz furnaces are not practicable at capacities below 750 kg. At 10 kHz, charge pieces
less than 10 mm in diameter can be heated, so furnaces as small as 5 kg capacity can be used.
The availability of reliable frequency converters has allowed the development of application
specific units, as given in Table 2.6. Frequencies are usually limited to 250 to 350 Hz (in the
case of variable frequencies), as at higher frequencies metal homogenisation becomes
insufficient. Higher frequencies are used in special cases, such as with very small furnaces.
Figure 2.14 shows coreless induction furnaces of various sizes.

2
Without taking into account the efficiency for electric power generation.
Chapter 2
38 Smitheries and Foundries Industry
Frequency (Hz) Application
70 Brass swarf
100 Aluminium scalpings
100 150 Cast iron borings
250 Aluminium extrusion scrap
250 Iron from foundry returns and steel scrap
500 and 1000 Steel melting and melting of wet cast iron borings
1000 and 3000 Wide variety of copper alloys
3000 Investment casting
10000 Jewellery trade
Table 2.6: Field of application of available induction furnaces as related to their frequency
[174, Brown, 2000]

Figure 2.14: Coreless induction furnaces
[237, HUT, 2003]

Water cooling systems are essential for operation of the coreless induction furnace. Cooling the
coil protects both the coil and the insulation from thermal damage, not only during normal
operation but also during the cool-down period when the power has been switched off and the
furnace emptied. Several types of cooling are available, using closed circuits with heat-
exchangers or open evaporative systems. The availability of a cooling system opens up the
possibilities for internal energy recovery.
[32, CAEF, 1997], [47, ETSU, 1992], [110, Vito, 2001], [174, Brown, 2000], [176, ETSU,
1998], [202, TWG, 2002]

2.4.3.1.2 Melting practice

The coreless induction furnace is used for melting but cannot be used for refining. Therefore, in
steel foundries the induction furnace has to be charged with raw materials of the correct
chemical composition, i.e. that corresponding to the required composition of the melt; hence
steel scrap is mainly used. If necessary, the metal can be refined after melting in an AOD
converter or in special treatment ladles (see Section 2.4.12).

High powered furnaces allow melting following the tap and charge method. Here the furnace
is tapped completely empty and charged with cold material to start the next melting cycle. The
molten heel method is used on low powered (mains frequency) furnaces where approximately
one third of the melt is tapped before the cold charge material is added. Because of the better
electromagnetic coupling between the melt liquid and the coil compared to a less dense solid
(cold) charge, the production rate increases significantly when using the latter method on low
powered furnaces.

Chapter 2
Smitheries and Foundries Industry 39
Steel grades containing more than 0.2 % of reactive elements such as Al, Ti and Zr cannot be
melted in an oxidising environment such as air. They require an inert atmosphere or a vacuum
melting and casting method. This is obtained by placing an induction furnace in a vacuum or
airtight chamber. The application of a vacuum ensures very good degassing of the melt. Highly
oxidisable elements are added in vacuum or after backfilling with an inert gas.

Induction furnaces are excellent melting units, but in general they are less efficient holders.
When they are used for melting only, the molten metal is mostly moved to an efficient holding
furnace as soon as it has reached the desired temperature. Many types of coreless induction
furnaces are available, with fixed or removable crucibles. For aluminium, both channel and
coreless induction furnaces are available for melting and holding. However, the channel type is
seldom used, due to difficulties in keeping the channel open and due to the need to maintain a
molten heel at all times.
[32, CAEF, 1997], [48, ETSU, 1994]

2.4.3.1.3 Advantages:

Due to its multiple advantages, the induction furnace is increasingly being implemented. Its
main advantages are:
- higher flexibility in alloys and melting regime. As such it is an ideal melting technique for
jobbing foundries and special alloys
- short meltdown times
- lower environmental burden caused by the foundry
- little maintenance, depending on the lifetime of the refractory lining
- good process control: computer support and fully automatic operation is possible, which
allows optimal temperature control
- maximum thermal efficiency is possible, if process parameters are locally calculated and
set-up
- intense stirring in the bath makes the melt homogeneous
- charging, sampling and deslagging of holding the liquid metal, though typically with low
efficiencies, although high holding efficiencies have been reported for copper and
aluminium.

2.4.3.1.4 Disadvantages:

- because of the monopoly of the local electricity supplier, the operator is fully dependent on
the connection conditions of the local electricity grid, the energy costs and any possible
extra costs (peak control, etc.)
- the energy costs are more expensive than the costs of using fossil fuels
- the cleaning action of the induction furnace on the melt is limited because of the small
amount of slag and the relatively small contact area between the slag and melt. This requires
the use of a high quality, and thus, more expensive charge than cupola or EAF furnaces
- the installation requires a high investment, although the operator can save on additional
environmental investments. The net cost per tonne installed furnace content is around
EUR 375000
- other melting techniques are more suitable for capacities >15 tonnes per hour. Depending
on the intended alloy, the hot cupola or electric arc furnace may be considered
- it has a low efficiency during holding because of the heat losses in the water-cooled
induction coil.
[110, Vito, 2001]

Chapter 2
40 Smitheries and Foundries Industry
2.4.3.2 Channel induction furnace

2.4.3.2.1 Description

This type of furnace is mainly used for holding purposes, though it may also be used as a
combined melting- and holding aggregate.

The channel induction furnace consists of a big, thermally isolated bucket, equipped with an
isolated top-lid for introduction of the charge (Figure 2.15). The bottom is equipped with one or
more U-channels. Around these channels is a water-cooled induction coil, which heats and
circulates the metal. The induction current has the frequency of the local electricity grid. The
furnace is generally mounted in an hydraulically tiltable frame for tapping or maintenance
operations. Pressurised and non-tilting furnaces are also in use.

Figure 2.15: Typical construction of a bath channel furnace
[55, ETSU, 1993]

In order to allow operation, a minimum amount of molten metal needs to stay inside the crucible
and channel. The crucible needs to stay filled up to one third of its capacity. Two values are
usually quoted when the capacity of a holding furnace is defined, these are total capacity and
useful capacity; for example 60/35 tonnes. The difference between these two values represents
the amount of metal that must be retained in the furnace.

Thermal losses through the cooling water and the furnace wall are low compared to those in the
coreless induction furnace. Equipping the furnace with pouring channels according to the
teapot principle allows for a reduced oxidation of the melt and wearing of the refractory.

Due to the large content of the crucible, any changes in the melt composition are flattened out.
This principle however counteracts the flexibility of the furnace, as changing to another melt
composition requires a long transition period. In practice, the melt composition is therefore kept
fairly constant.

Chapter 2
Smitheries and Foundries Industry 41
Figure 2.16: Channel induction furnace
[237, HUT, 2003]

The channel induction furnace finds its main application as a holding furnace in iron foundries.
An example picture is given in Figure 2.16. It is the furnace of choice for duplex operation with
cupola furnaces. The capacity varies between 5 and more than 100 tonnes. The holding furnace
serves as a buffer between the melting and casting shop. It is important to assess foundry and
production requirements fully when considering whether to use a holding furnace. There may be
more cost-effective and energy efficient methods of achieving the requirements, and it is
advisable to investigate thoroughly all possible solutions before a decision is made. For
aluminium, the channel type is seldom used, due to difficulties in keeping the channel open and
due to the need to maintain a molten heel at all times.

Output requirements are an important consideration when deciding the size of a channel
furnace. The choice of a smaller furnace may be advantageous. Although a smaller furnace is
less efficient, the loss in efficiency will be offset by its reduced annual power consumption
compared with larger furnaces.
[48, ETSU, 1994], [55, ETSU, 1993], [110, Vito, 2001],

2.4.3.2.2 Advantages:

- high thermal efficiency as a holding furnace
- minimal burn-out of the alloy elements
- little maintenance.

2.4.3.2.3 Disadvantages:

- a minimum amount of molten iron, but which may be a substantial part of the furnace
capacity, has to be maintained in the furnace body in order to guarantee proper electrical
functioning
- the furnace cannot perform a cold start because of the limited power density that can be
achieved in the loop
- difficult to monitor the channel wear
- potential danger of accidental leaks, because of the induction coil on the bottom
- contact between the cooling water and metal is difficult to prevent.
[110, Vito, 2001]

2.4.3.3 Nature of emissions

Since no coal or fuel is burned in the induction furnace and no refining procedures are executed,
the emissions solely depend on the cleanliness and the composition of the charged material.
Two major categories of emissions can be distinguished. The first, and major, category relates
to the charge cleanliness, e.g. rust, dirt, foundry sand, paint, oil, galvanised or soldered metal,
all of which are elements which give rise to the emission of dust and fumes (organic or
Chapter 2
42 Smitheries and Foundries Industry
metallic). The second category relates to chemical reactions at high temperatures, (e.g. while
holding or adjusting the metal composition), which can give rise to metallurgical fume due to
oxidation.

Additionally the refractory lining (acid SiO
2
based, neutral Al
2
O
3
based, or basic MgO based)
may add a small amount of dust particles to the emission.

It is difficult to obtain average emission data since the charge cleanliness, which is the dominant
contributor to emissions, varies from foundry to foundry.

2.4.4 Radiant-roof furnace (resistance heated)

The radiant-roof furnace is a low-energy holding furnace with a heavily insulated box design
with banks of resistance elements in a hinged, insulated roof. They are mainly used in non-
ferrous (aluminium) pressure die-casting shops with centralised melting facilities. Typical units
have a capacity of 250 1000 kg with a 5 kW to 12 kW connected load. Bale-out and charge
wells are separated from the main bath by refractory walls with connectors at the bottom to
allow clean metal to pass from one area to another. Figure 2.17 gives a schematic view of the
radiant-roof principle.

Figure 2.17: Radiant-roof furnace
[48, ETSU, 1994]

The advantages of radiant-roof furnaces are:
- no crucible required
- very low energy costs
- close temperature control
- clean, cool, silent working conditions.

Although most low-energy holding furnaces serve individual casting machines, some larger
ones are used as buffers between bulk melters and machine furnaces. This latter use allows
better use of the melter, which is seldom an efficient holder.

Some foundries use large radiant-roof furnaces with higher power as melters, for example to
generate molten metal for low-pressure die-casting. Several manufacturers build versions of
radiant-roof furnaces that can be fully sealed and pressurised by gas, in order to function as
dosing furnaces. Dosing furnaces provide precise shots of metal for pressure or gravity die-
casting and compete, to some extent, with mechanical ladle systems as automatic pouring
systems. Provided that they are carefully maintained and used, radiant-roof furnaces can provide
a valuable control of both temperature and shot weight and can improve yield.
[48, ETSU, 1994]
Chapter 2
Smitheries and Foundries Industry 43
2.4.5 Rotary furnace

2.4.5.1 Description

The rotary furnace consists of a horizontal cylindrical vessel, in which the metallic charge is
heated by a burner located at one side of the furnace. The flue-gases leave the oven through the
opposite side. To generate the required heat, fuel or natural gas is used combined with air or
pure oxygen.

A tilting mechanism allows the furnace to be lifted to a certain angle or into a vertical position.
This position is used for charging of the furnace with a drop bottom bucket or a vibrating chute,
and for lining repair or renewal. During heating and melting the furnace is rotated slowly to
allow the heat transfer and distribution. The furnace atmosphere is controlled by the air
(oxygen)/fuel ratio.

Once the metal is melted, and after a composition check and adjustment, a tap-hole in front of
the furnace is opened and the melt in the furnace is discharged into ladles. Because of its lower
density, the slag floats on the metal bath in the furnace and is finally collected through the tap-
hole into slag pots.

A melting cycle spans 1 to several hours. For continuous molten metal production, foundries
install 2 to 4 rotary furnaces, which are operated consecutively. The thermal efficiency of the
rotary furnace is very high, i.e. at 50 to 65 %
3
, depending on the capacity. This high yield is
achieved by using pure oxygen instead of air as the combustion medium.
[32, CAEF, 1997]

2.4.5.2 Melting practice

For cast iron melting, the furnace is charged with pig iron, foundry returns, steel scrap, slagging
(e.g. sand, lime), and carburisation agents (e.g. graphite). The melting cycle starts with a slightly
oxidising and short flame (air factor, [ = 1.03), which gives the highest energy input. The
furnace is rotated stepwise through 90 and the direction of the rotation is changed from time to
time. This way, the furnace walls can exchange heat with the charge by convection. As soon as
the charge is melted, the flame is reduced to prevent excessive oxidation of the alloying
elements. During overheating and holding, a long and reducing flame is applied ([ = 0.9) and
the furnace movement is changed to full and continuous rotation. The slag layer provides
thermal insulation and prevents the burning-off of the alloying elements. After controlling and
adjusting the melt composition and temperature, the tap-hole is opened and the metal is tapped
into ladles. The slag floats on the metal bath and is collected separately after the metal is
removed.

The lifespan of the refractory is largely dependent on the overheating temperature and the
charge composition. In the charging operation, mechanical shocks and cold start-ups need to be
prevented. The furnace atmosphere, the holding time, rotational speed and the burner position
also affect the refractory life. In normal conditions the refractory life is 100 to 300 melting
cycles.
[32, CAEF, 1997], [110, Vito, 2001]

2.4.5.3 Metallurgy

The rotary furnace has been used in non-ferrous melting for many years. In this application
traditional oil-air burners can provide the relatively low melting temperatures. The development
of oxygen-air burners has enabled the introduction of cast iron production, using a higher
relative amount of steel scrap and applying graphite for carburisation.

3
That is providing the energy for oxygen production is not taken into account. With oxygen production, the efficiency will be 10
to 15 % less.
Chapter 2
44 Smitheries and Foundries Industry
A significant disadvantage of the rotary furnace is that it also burns Fe, C, Si, Mn and S. These
losses have to be compensated for by the addition of alloying elements before or after melting.
The efficiency of uptake of these elements is usually rather low. Concentration gradients may
occur between the front and the back of the metal bath due to the absence of axial motion and
due to inhomogeneities in radiation and the atmosphere above the wide bath surface.
[110, Vito, 2001]

2.4.5.4 Application

Due to its batch character, the rotary furnace provides an equal flexibility as the coreless
induction furnace in the cast iron foundry. The investment costs however are lower. A 5 tonne
furnace costs EUR 500000 600000, of which 30 % are for the exhaust system and dedusting.
The rotary furnace is also a good alternative for the small-scale cold blast cupola, due to its
higher flexibility and lower environmental costs. Rotary furnaces are used for melting volumes
of 2 to 20 tonnes, with production capacities of 1 to 6 tonnes per hour.
[110, Vito, 2001]

2.4.5.5 Advantages:

- quick change of alloy possible
- melting without contamination, e.g. without sulphur pick-up
- low investment costs
- small dedusting system because of low flue-gas rate
- easy to maintain.

2.4.5.6 Disadvantages:

- easy burn-out of C, Si, Mn
- gas and oxygen use can be high if not operated continuously
- the energy consumption increases if more steel is added to the charge.
[110, Vito, 2001]

2.4.6 Hearth type furnace

The hearth type furnace is also known as a reverberatory or bale-out furnace. It is a static
furnace with direct heating. Hot air and combustion gases from oil or gas burners are blown
over the metal (melt) and exhausted out of the furnace. The hearth type furnace finds its main
application in non-ferrous metal melting. A typical furnace design is given in Figure 2.18.

Figure 2.18: Cross-section of a hearth type furnace
[175, Brown, 1999]

Chapter 2
Smitheries and Foundries Industry 45
This is a refractory lined, rectangular or circular bath furnace that is fired by wall or roof-
mounted burners. Varieties of fuels are used and additionally oxy-fuel burners can be used to
increase the melting rate. Extraction and treatment of the combustion gases is typically carried
out and for this the furnaces are partially sealed. Hoods and covered launders provide extraction
during tapping and charging. The furnace can be constructed in a number of configurations
depending on the particular metal and application, variations include the provision of sloping
hearths and side wells for specific melting purposes, tuyres and lances for the addition of
gases. The furnaces can sometimes be tilted to pour or blow metal.

The melting efficiency of a hearth type furnace is usually not great because of the poor heat
transfer from the burner. The efficiency has been improved in practice by the use of oxygen
enrichment or by using combinations of gas and solid fuel to extend the flame length. These
furnaces are used for batch melting, refining and for holding a variety of metals. Hearth type
furnaces are mainly used to melt large quantities of non-ferrous metals [155, European IPPC
Bureau, 2001].

Large hearth type furnaces give rapid melting and can handle bulky charge materials, but the
direct contact between the flame and charge may lead to high metal losses, gas pick-up and to
considerable oxide contamination. Temperature control can also be difficult. These difficulties
can be overcome to some extent by good practice. For example, dross can be removed by
applying suitable fluxing practice, and applying modern burners reduces problems with the
temperature control. After tapping, filtering and degassing, further treatments can be applied as
the metal is transferred to holding furnaces.
[42, US EPA, 1998], [48, ETSU, 1994], [155, European IPPC Bureau, 2001], [175, Brown,
1999]

2.4.7 Shaft furnace

2.4.7.1 Description

This is a simple vertical furnace with a collecting hearth (inside or outside the furnace) and
burner system at the lower end and a material charging system at the top. The burners are
usually gas-fired. Metal is fed into the top of the furnace and is melted as it passes down the
shaft. Independent control of the fuel/air ratio is usually provided for each burner. Continuous
CO or hydrogen monitoring is also provided for each row of burners and monitor the
combustion gases from each burner in turn. Combustion gases are usually extracted and
cleaned. An afterburner is sometimes used to decompose any carbon monoxide, oil, VOCs or
dioxins that are produced. The addition of oxygen above the melting zone has been used to
provide afterburning in the upper levels of shaft or blast furnaces.

The furnace is used to melt pure metal, but occasionally metal that is contaminated with organic
material may be used. If oily metal is fed into the furnace, it passes through a temperature
gradient existing between the charging area and the burners. The low temperature can produce a
mist of partially burned organic material. The shaft furnace is also used to preheat the charge
material before smelting. A typical representation of this furnace is given in Figure 2.19.

This furnace type is only used for non-ferrous metal melting, mainly for aluminium. Due to the
complex construction and difficult renewal of the refractory, the furnace is only used for metals
with low melting points. Therefore, the maintenance requirements of the furnace lining are
rather limited. Typical refractory lifetimes are 4 to 8 years.

Modern furnace types with computer controlled burner systems reach an energy use of
650 kWh/tonne of melted Al. (at 720 C). The theoretical energy needed is 320 kWh/tonne. The
thermal efficiency therefore is 50 %.

Chapter 2
46 Smitheries and Foundries Industry
The shaft furnace is a continuous melting furnace with high capacity, ranging from 0.5 to
5 tonnes per hour and with a holding capacity of up to 50 tonnes. Due to its holding function, a
change of alloy is difficult.
[48, ETSU, 1994], [155, European IPPC Bureau, 2001]

Figure 2.19: Shaft furnace
[48, ETSU, 1994]

2.4.7.2 Advantages:

- due to the long preheating, the charge is very well dried before the melting starts. This
makes the furnace well suited for Al, due to the reduced risk of hydrogen uptake
- relatively low investment and operating costs. Operating costs are kept low by the effective
preheating, automatic control and long refractory life
- technical advantages are: low gas pick-up, excellent temperature control and low metal loss.

2.4.7.3 Disadvantages:

- no flexibility regarding a possible change of alloy.
[110, Vito, 2001]

2.4.8 Crucible furnace

2.4.8.1 Description

These are simple crucibles that are heated externally by the combustion gases from gas or oil
combustion, by electricity or, for lower temperatures, by thermal fluid. Contact with a direct
flame is avoided to prevent there being local hot spots at the base of the crucible and so that
good temperature control can be maintained in the melt, to prevent oxidation and vaporisation
of the metal.

Chapter 2
Smitheries and Foundries Industry 47
This furnace type is only used for non-ferrous metal melting. Due to the indirect heating
(through the crucible wall) no burn-off or gas take-up can take place. These furnaces are used
for the production of small amounts of molten metal (less than 500 kg per batch) and for low
production capacities. Example furnaces are displayed in Figure 2.20.

Figure 2.20: Crucible furnaces
[237, HUT, 2003]

The crucible is tilted manually, with a crane, or automatically to pour the molten metal into the
mould. For copper based materials, only graphite or carborundum (silicon carbide) crucibles are
used, whereas aluminium can also be melted in cast iron crucibles.

Crucibles used for holding, transport and metal treatment in ferrous foundries are called ladles.
[110, Vito, 2001], [126, Teknologisk, 2000], [225, TWG, 2003]

2.4.8.2 Melting practice

The cold load is brought into the crucible, and heating is started at full power in order to melt
the load. At 50 to 100 C below the melting temperature, the power is switched off, and the load
is heated further by the thermal inertia of the crucible. Afterwards, the temperature is levelled
off using a control system. After deslagging, metal treatment can be performed. This involves
oxygen removal, degassing, grain refining and the adjustment of volatile metals such as zinc
and magnesium. The slag is then removed once more before casting takes place.

Besides regular renewal of the crucible, no maintenance is needed. The renewal time is
primarily dependent on the alloy to be melted. Multiple alloy changes cause a more rapid
crucible wear. SiC-crucibles, used for low melting alloys, have a useful life of 130 to 150
charges. For high melting alloys, the useful life is 50 to 80 charges.

Thermal efficiencies are 750 3000 kWh per tonne of aluminium, i.e. 15 - 30 % efficiency.
[34, Binninger, 1994], [110, Vito, 2001], [126, Teknologisk, 2000]

2.4.8.3 Advantages:

- simple technology
- low maintenance
- flexibility regarding change of alloy.

2.4.8.4 Disadvantages:

- low efficiency and production capacity.
[110, Vito, 2001]
Chapter 2
48 Smitheries and Foundries Industry
2.4.9 Argon Oxygen Decarburisation (AOD) converter for steel
refining

The AOD converter is a special vessel for refining steel. The molten metal is transferred directly
from the melting furnace (generally electric arc or induction) into the converter. As shown in
Figure 2.21, oxygen (for the removal of carbon, silicon) and argon gas (for the stirring action)
can be injected into the converter by means of tuyres positioned in its lower part, to refine the
metal. The converter is equipped with a tilting mechanism in order to allow filling and
emptying. Steel with a precise chemistry can be produced using AOD, but at a high cost. This
system is not widely used in steel foundries outside the US.

Figure 2.21: AOD converter
[237, HUT, 2003]

The first phase of the refining process consists of a decarburisation, through the injection of
oxygen into the converter. This is a refining operation that holds the carbon content of the
charge material within a specific range, as required. Decarburisation is initiated by injecting
oxygen into the metal bath. This creates a strong stirring action during which carbon in the melt
is burned. At the same time the boiling burns out silicon, and flushes out hydrogen and
nitrogen from the metal bath. All impurities (oxides) are trapped in the slag. During melting,
lime may be added to bring the slag to the proper basicity. When the carbon reaches the required
concentration, oxygen injection is stopped and silicon and manganese are added to halt the
boiling reaction, especially in the production of stainless steels. A dilution of the oxygen with
argon or nitrogen assists the oxidation of carbon in preference to metallic elements such as iron,
chromium, etc., leading to very good metallic yields. Consequently Al or Si and limestone are
added to the metal and argon is injected in order to remove sulphur. The continuous gas
injection causes a violent stirring action and an intimate mixing of slag and metal which can
lower sulphur values to below 0.005 %. The residual gas content of the treated metal (hydrogen
and nitrogen) is very low. All impurities are trapped in the slag and removed with it. After
temperature and alloy adjustment the metal is transferred to ladles for pouring.
[174, Brown, 2000], [32, CAEF, 1997]

2.4.10 Vacuum Oxygen Decarburisation converter (VODC) for steel
refining

The VODC (Vacuum Oxygen Decarburisation Converter) operates in such a way that molten
steel is contained in the converter, which is attached to vacuum pumps, steam ejectors and an
argon gas source. Two distinct processes are carried out in the vacuum treatment of stainless
steel to obtain a higher quality level.
Chapter 2
Smitheries and Foundries Industry 49
First, the steel is decarburised by introducing oxygen into the melt. At the same time, some
argon is injected from the converter base. The converter uses vacuum pumps to lower the partial
pressure of carbon monoxide, to such an extent that effective decarburisation can be carried out
without oxidising too much of the chromium. This first process is similar to AOD
decarburisation, but less argon is needed thanks to the lower total gas pressure and the process is
much more efficient. The oxidised chromium is reduced back to liquid steel with aluminium.

The second process involves degassing. The converter is brought under deep vacuum
(1 - 5 mbar) using a water ring pump and steam ejectors. Slight argon bubbling is continued in
order to maintain an effective steel movement. At the very low pressure, gaseous impurities
such as hydrogen and nitrogen are eliminated effectively. At the same time, the total oxygen
content and the sulphur content are drastically decreased, which is beneficial for the mechanical
properties of the end steel.

Low alloyed steels are normally only degassed.

The VODC process provides specific steel qualities, which cannot be obtained by other
methods. The total oxygen content is also lower than what is typical for electric arc melted and
AOD-treated steel, because a high level of oxide inclusions are removed form the melt during
the VODC process, and most of the dissolved oxygen is further removed during the degassing
phase.
[202, TWG, 2002]

2.4.11 Metal treatment of steel

In order to guarantee a good casting quality, cast steel needs further treatment to remove
impurities and the possible causes of defects, i.e.:

- Deoxidation: Oxygen dissolves in liquid steel, in the form of FeO. During solidification, the
oxygen can then combine with C in the steel to form CO. This process can thus change the
composition of the steel and generate porosity. Deoxidation is therefore always necessary.
Deoxidation is performed with an element that preferentially binds the oxygen. Silicon,
calcium silicide, titanium, zirconium and aluminium are possible deoxidants, with
aluminium being the most powerful and the one that is most generally used. Aluminium is
added in the form of a stick or wire (see Figure 2.22). The treatment is usually performed in
the furnace and in the ladle. The produced aluminium oxide is insoluble in the melt and
mixes with the slag.

Figure 2.22: Deoxidation, using an aluminium wire
[237, HUT, 2003]
Chapter 2
50 Smitheries and Foundries Industry
- Sulphide formation: As the tensile strength of steel increases, the harmful effect of sulphur
increases. Sulphur is soluble in liquid steel, but on solidification it precipitates as MnS.
Precipitates can take various forms and have different effects. The form of the sulphides is
related to the residual aluminium content after deoxidation. The residual amount of Al
should allow the formation of type III sulphides.
[110, Vito, 2001], [174, Brown, 2000]

2.4.12 Cast iron treatment

2.4.12.1 Alloying

During melting, some elements in the melt oxidise and are lost to the slag. At the end of the
melting period a correction of the composition needs to be performed, in order to guarantee the
appropriate final quality. For unalloyed cast iron, this is needed for: C, Si, Mn, S and P. If
special properties are desired, specific alloying elements can be added, ranging from aluminium
to zirconium, in concentrations of less than 1 % to more than 30 %. In general these are added
as ferrous alloy blocks or grains. The additions are made to the liquid iron, since this reduces the
risk of oxidation losses.

The addition is performed in the furnace, in the molten metal flow during pouring, or by putting
the additives in the transport ladle before pouring the molten metal into it.
[110, Vito, 2001]

2.4.12.2 Homogenisation

The addition of alloying elements can introduce impurities such as oxides, sulphides or carbides
into the iron. In order to reduce the negative effects from these compounds, the metal is
overheated at 1480 C to 1500 C. Nevertheless overheating can affect graphite precipitation
during solidification. Homogenisation has the following positive effects:

- reduction of oxides (FeO, SiO
2
, MnO) by C, generating CO-bubbles. On their way through
the melt these bubbles remove H
2
and N
2
from the melt
- at high temperature and under intense bath movements, impurities coagulate and rise to the
melt surface more quickly, where they are then taken up by the slag.
[110, Vito, 2001]

2.4.12.3 Desulphurisation and recarburisation of cupola melted iron

Due to the intimate contact of the liquid metal with the coke, cupola melted iron shows a
relatively high sulphur content. Sulphur pick-up in the cupola gives rise to a lower viscosity of
the liquid metal, which is an advantage in some cases, e.g. for thin walled grey iron castings.

However, if the cast iron is to be treated with Mg (as described in the next section) to produce a
nodular cast iron quality then the sulphur should be neutralised, to prevent excessive Mg
consumption. This is performed by using one of a number of methods. In the porous plug
method, nitrogen or argon gas is used to agitate the metal and CaC
2
powder is added to the
metal. Upon contact with the liquid metal CaS is formed and removed as slag. The treated metal
then flows to a holding ladle from which it is tapped for further treatment.

Chapter 2
Smitheries and Foundries Industry 51
2.4.12.4 Nodularisation treatment of the melt

Nodular cast iron is obtained by the addition of Mg, either pure or as a FeSiMg or NiMg alloy,
to the molten metal. Prior to this, a desulphurisation treatment may be necessary to guarantee
successful nodularisation. The presence of Mg in the melt provokes a solidification of the
graphite phase in the shape of microscopic spheres. This gives rise to higher mechanical
properties, such as strength and ductility.

There are several techniques that can be applied to introduce Mg into the liquid metal:
- The pour over technique: This is the simplest method, whereby the liquid metal is poured
over the Mg alloy in the bottom of the ladle
- The sandwich method: Here Mg alloy is placed at the bottom of a specially designed ladle
and covered with steel sheets or FeSi, as shown in Figure 2.23. The metal is poured into the
ladle and after melting of the cover the Mg reaction takes place
- The Tundish Cover: This is an improved sandwich technique, whereby the ladle is covered
by a lid after the Mg alloy has been placed at the bottom of the ladle. The metal is poured in
the concave lid and flows through a hole in the ladle, where the reaction takes place
- The plunging method: Using a plunger bell, the Mg alloy is submerged in the liquid metal
until the reaction is finished. During the treatment the ladle lid is closed, to prevent MgO
fume emissions
- The G Fischer converter: This process uses a special ladle that is closed tightly with a lid
after the metal is poured into the converter in the horizontal position. The converter is then
tilted into a vertical position, allowing the Mg alloy to react with the metal. Once the
reaction is finished the converter is turned back into the horizontal position and is tapped
after opening the lid
- The core wire injection: Here very fine powdered Mg alloy is rolled in tin steel sheet,
forming a filled wire. This wire is then injected, by a controlled mechanism, into a slender
shaped ladle, where the Mg is released
- The flow through treatment: Here the metal is poured into a special designed reaction
chamber in which the Mg alloy was put earlier
- The ductilator: In this method the metal is poured into a reaction chamber where the flow is
forced to form a vortex. The Mg alloy is injected into the vortex with an inert carrier gas.
The treatment can be performed on large quantities of metal or during mould filling
- The inmold process: In this process the Mg alloy, shaped as a tablet, is put directly into the
mould cavity (gating system). The reaction takes place during mould filling, ensuring a high
yield.

Figure 2.23: The sandwich method for nodularisation
[237, HUT, 2003]

After treatment, the metal has to be poured into the mould within a specific time, as the Mg
effect tends to fade quickly, necessitating a new treatment if a certain time limit (10 to 15 min)
is exceeded.
Chapter 2
52 Smitheries and Foundries Industry
2.4.12.5 Inoculation of the melt

The presence of coarse shaped graphite in a metal matrix leads to poor mechanical properties of
the material. In order to get a fine crystalline metallurgical structure, inoculation of the melt is
necessary prior to casting. This process introduces seeds for crystal growth into the metal melt.
Usually FeSi alloys are used for this purpose. Ca, Al and rare earth materials are often included
in the inoculant.

Again, several techniques can be used to inoculate the molten metal:
- Injection during tapping: Here the inoculant is injected directly into the metal flow during
tapping
- Injection during pouring: Here the inoculant is injected directly into the metal flow during
pouring of the mould (see Figure 2.24)
- Filled wire injection in the melt (see Section 2.4.12.4)
- Inmold process (see Section 2.4.12.4).

Figure 2.24: Inoculation during pouring
[237, HUT, 2003]

2.4.13 Non-ferrous metal treatment

There are three main metal treatment (or refining) operations carried out in aluminium melting
processes. These are as follows:

- Degassing: Molten aluminium dissolves hydrogen, which is then expelled upon cooling and
can thus lead to porosity in the finished casting. Hydrogen therefore needs to be removed.
This is done by bubbling an inert gas through the melt. Good degassing methods provide
bubbles with high residence time and high surface area. The degassing of aluminium is
mostly performed using an impeller station. This treatment uses a rapid-spinning mixer and
nitrogen injection into the melt. Degassing is often combined with a cleaning of the melt.
Cleaning is performed to remove alkali or alkaline earth metals, such as Ca. Cleaning can be
done by Cl
2
-gas. The use of HCE for this purpose has been banned in the EU (Dir.
97/16/EC). A mixture of nitrogen with 3 % Cl
2
is generally used, for simultaneous
degassing and cleaning. Alternative degassing methods use tablets, a lance with a porous
head, or a porous stone in the holding furnace
- Modification and grain refining: For aluminium alloys this usually involves the addition of
small amounts of metal to the melt. These additions control the grain size and modify the
microstructure of the solidifying metal and thus enhance the casting mechanical properties.
Sodium or strontium are used for the modification, while grain refinement is achieved with
titanium, titanium boride, zircon or carbon. This treatment is mostly performed in
combination with degassing in a dedicated metal treatment station
- Fluxing: This usually involves the addition of solid fluoride based fluxes to the melt, to
remove solid contaminants.
[164, UK Environment Agency, 2002], [175, Brown, 1999], [178, Wenk, 1995]
Chapter 2
Smitheries and Foundries Industry 53
There are four main metal treatment (or refining) operations carried out in magnesium melting
processes. These are as follows:
- Grain modification: Grain modifiers for magnesium alloy processes are usually in the form
of zirconium or formerly hexachloroethane. The use of HCE for this purpose has been
banned in the EU (Dir. 97/16/EC)
- Fluxing: This usually involves the addition of alkali, earth alkali, chloride and fluoride
proprietary fluxes to the melt to remove solid contaminants
- Degassing: For magnesium processes, nitrogen or argon gas sparging may take place for
degassing and oxide removal. The degassing of magnesium alloys may also be effected by a
mixture of argon and chlorine gases, when argon is used as a carrier gas
- Oxidation control: The presence of beryllium within the melt grain modifies and stops
oxidation. Finished magnesium alloys may contain up to 15 ppm beryllium by weight. A
master alloy of aluminium/beryllium, containing up to a nominal 5 % beryllium, may be
added to the molten magnesium alloy to give it this beryllium loading. Oxidation control
may also be achieved by blanketing the surface of the metal with a carrier gas, such as
carbon dioxide or argon, containing up to 4 % sulphur hexafluoride. To date, the only flux-
free alternative to SF
6
is SO
2
. SO
2
is significantly cheaper then SF
6
but its main drawback is
its toxicity, and as a result operations need to be enclosed better. Up to the present, only 1/3
of the foundries in Europe have been using SO
2
. Occasionally oxidation control may be
achieved by sprinkling sulphur powder on the surface of the melt. This issue is discussed in
Section 4.2.7.1.

There are three main metal treatment (or refining) operations carried out in copper melting
processes. Molten copper dissolves oxygen and hydrogen, which may recombine to form water
vapour. This in turn will generate porosity in the casting. Degassing and deoxidation are thus
applied to remove hydrogen and oxygen, respectively. The applicable metal treatment
operations are as follows:

- Deoxidation: Deoxidation is performed by adding a reagent which binds the oxygen and
forms a fluid slag. Care must be taken to prevent the deoxidation products from being
entrained in the solidified casting and the residual deoxidant from adversely affecting the
alloy properties. Phosphorus is the most commonly used deoxidation reagent. Alternatives
are magnesium, manganese, calcium, silicon and boron
- Degassing: Hydrogen is removed from the melt by bubbling an inert gas through it. Both
argon and nitrogen may be used. The technique is comparable to aluminium degassing
- Fluxing: Aluminium in alloys can oxidise and generate oxide skins. These can cause
problems upon casting. In non-aluminium alloys, traces of aluminium can cause defects.
Therefore they need to be removed, using fluxing agents. Fluxes are also used to cover the
surface in order to prevent oxidation, zinc loss and hydrogen pick-up during melting.
Specific fluxing agents exist for each type of treatment.
[165, UK Environment Agency, 2002], [182, Closset, 2002]

2.5 Mould and core production

Moulding consists of making a mould in which the molten metal will be poured. Some moulds
may need to have special properties to produce high quality castings, which, for example:

- reproduce exactly and with a high dimensional precision the shape of the casting model
- give a smooth skin to the casting in order to avoid excessive fettling
- avoid any casting defects such as cracks, worming, pinholes, etc.

Just as the mould defines the outer shape of the casting, the core defines the inner one, or at
least the parts not directly attainable by moulding.

Chapter 2
54 Smitheries and Foundries Industry
Moulds may be classified in two large families:

- lost moulds (single use moulds): These are specially made for each casting and are
destroyed after pouring. The moulds are generally made of sand, and are chemically-
bonded, clay-bonded, or even unbonded. Investment casting can also be included in this
family
- permanent moulds (multi use moulds): These are used for gravity and low-pressure
casting, pressure die-casting, and centrifugal casting. Typically the moulds are metallic.

The cores used for ferrous castings are practically always made of sand. The choice of binder
technology used depends on factors such as the size of the casting, the production rate, the metal
poured, the shake-out properties, etc.

For sand moulding, the mould may be produced by manual or mechanical ramming actions,
such as by jolt, squeeze, air impact, vibration, etc.. When the mould has sufficient strength it is
released from the model, which can then be used to produce a new mould.

Generally, cores are produced by the same techniques as moulds, but small or medium sized
cores are often blown or shot into wooden, plastic or metallic core-boxes.

For non-ferrous casting, about 30 % of copper alloys are cast in sand moulds. Only about 10 %
of light non-ferrous metals are cast in single use moulds.

The production of patterns and dies is generally carried out by specialised external suppliers.
These activities are found in the metal and plastics treatment sectors.
[2, Hoffmeister, et al., 1997], [32, CAEF, 1997], [110, Vito, 2001]

2.5.1 Raw materials

2.5.1.1 Refractory materials

Whatever binder applied, the physical and chemical properties of the refractory material used to
make the moulds or cores affects their characteristics and their behaviour during pouring. This
is not surprising as these materials make up 95 to 99 % of the products used.

The purchase price of each sand type has four main components extraction, preparation,
packaging and transportation. Transportation costs will vary for different regions. The main
factor however in the price is the sand type. The average purchase price for the various types
varies widely. From a 1995 UK questionnaire, the tonnage price of chromite and zircon sand
were respectively a factor 9 and 14 times higher than that of silica sand. In Portugal the cost of
sand varies according to the amount purchased, but sand is generally cheaper in Portugal than in
Spain, France or Italy. Portuguese prices (year 2003) range between EUR 20 - 25 per tonne of
dry sand AFS 55, plus transport. Prices in the Czech Republic (year 2003) for silica sand range
between EUR 10 and 20 for silica sand, with the price depending on the volume, packaging and
treatment of the sand. The price for chromite sand is EUR 250 300/t, and for zircon sand
EUR 250 400/t.
[72, ETSU, 1995], [225, TWG, 2003]

The different types of refractory sands used for foundry purposes are described in the following
sections.

Chapter 2
Smitheries and Foundries Industry 55
2.5.1.1.1 Silica sand

This type of sand is the most commonly used, mostly because of its wide availability and its
relatively low cost. Silica sand is composed of the mineral "quartz" (SiO
2
), which is more or
less pure and clean, depending on its origin. Its dry piece specific weight fluctuates between 2.5
to 2.8 kg/dm. The dry bulk density (or loose volume weight) of silica sand is 1.4 to 1.6 kg/dm.

The thermal expansion of silica sand generates a mould movement upon pouring and cooling.
Specific additives are therefore used, especially when producing cores, in order to prevent
casting errors. These may be wood fluff, iron oxide or cured foundry sand. Sand containing
feldspar has a lower thermal expansion than pure quartz sand and a lower sintering point, but it
is widely used in order to minimise expansion defects in castings.

Silica sand is neutral and is compatible with all binders and normal cast alloys. The respirable
particulate matter (RPM) fraction of quartz is classified by IARC as carcinogenic [233, IARC,
1997]. This is a health and safety issue. Studies are ongoing to determine whether or not an air
pollution issue also exists. The amount of quartz in dust is defined by the content of quartz in
the input materials.

In green sand moulding technology, control of the grain size distribution is very important.
Figure 2.25 shows a typical grain size distribution for silica sand. The grain size distribution is
used to calculate the AFS-number (AFS = American Foundry Society). This gives the overall
fineness of the sand. The higher the AFS-number, the finer the sand. An alternative
classification system is the middle grain number or MK.

A finer sand grade will have more grains per gram, and thus a higher surface area. This requires
more binder addition for an equal mould strength. Operators therefore try to use the coarsest
sand, but one that still gives a good surface finishing aspect. Standard AFS-numbers are 50
60. For very smooth surfaces a fine sand is needed, typically one with an AFS of 90 110 is
used. Fine sands are also used in some instances to replace mould coatings.
[110, Vito, 2001], [202, TWG, 2002], [225, TWG, 2003]

0.53 1.32
7.79
39.70
37.78
11.49
0.99
0.00 0.00 0.00 0.40
0
20
40
60
80
100
0 63 90 125 180 250 355 500 710 1000 1400
Mesh width (Dm)
S
i
e
v
e
r
e
s
i
d
u
e
(
%
)
Figure 2.25: Typical grain size distribution for silica foundry sand
[110, Vito, 2001]

Chapter 2
56 Smitheries and Foundries Industry
2.5.1.1.2 Chromite sand

Chromite is a chrome ore, with the theoretical formula FeO.Cr
2
O
3
, which contains other
components such as magnesium and aluminium oxides. For its use in foundries, its silica
content must be lower than 2 % in order to prevent sintering at low temperatures. Its
characteristics are as follows:

- density: 4.3 to 4.6, cf. 2.65 for silica sand
- theoretical melting point: 2180 C, but the presence of impurities can lower this to 1800 C
- thermal diffusivity: more than 25 % higher than that of silica sand
- thermal expansion: regular, without a transition point, and lower than that of silica sand
- pH: rather basic, from 7 to 10.

Chromite sand is more refractory than silica. It is more thermally stable and has a greater
chilling power. Chromite sand provides a better surface finish on large castings. It is therefore
used for the production of large castings and in areas of the mould where chilling is required.
[32, CAEF, 1997]

2.5.1.1.3 Zircon sand

Zircon is a zirconium silicate, ZrSiO
4
. Zircon is the most widespread zirconium ore. Its
characteristics are as follows.

- density: 4.4 to 4.7, cf. 2.65 for silica sand
- melting point: higher than 2000 C
- thermal diffusivity: more than 30 % higher than that of silica sand
- thermal expansion: regular, without a transition point, and lower than that of silica sand.

The general characteristics of zircon sand are similar to that of chromite, but zircon sand
produces a better finish as a finer grade is used. These physical and thermal properties account
for its use for moulding or core-making in difficult cases, in spite of its very high price.
[32, CAEF, 1997], [72, ETSU, 1995]

2.5.1.1.4 Olivine sand

Olivine sands are a mineral group which includes forsterite, fayalite and others. The
characteristics of olivine sands are as follows:

- melting point: forsterite: 1890 C, fayalite: 1205 C
- density: 3.2 to 3.6
- pH: around 9.

The basic pH makes this type of sand not suitable for use with acid catalysed binder systems.

Olivine sand is produced by crushing natural rocks, which explains its diverse characteristics. It
is usually used for moulding and core-making in manganese steel casting. The presence of
manganese prohibits the presence of silica as these two components react to give a very fusible
compound. The reported purchase price in Spain is EUR 130/tonne (2002).
[32, CAEF, 1997], [210, Martnez de Morentin Ronda, 2002]

Chapter 2
Smitheries and Foundries Industry 57
2.5.1.2 Binders and other chemicals

2.5.1.2.1 Bentonite

Bentonite is a smectite clay, which has a lamellar structure. By adding water, the clay structure
swells due to the adsorption of water molecules. The clay then becomes workable and can be
spread to coat the sand grains during mixing.

Natural calcium bentonites do not swell or gel when mixed with water. They are rarely used at
present, being used only for very special castings. Alternatively, these materials may be
"activated" by treatment with soda ash to give "soda activated bentonites". These are used
extensively in ferrous foundries throughout Europe; their properties approach those of natural
sodium bentonites.

Natural sodium bentonites greatly swell when mixed with water. The main features in green
sand are their high dry strength, good tolerance to water content variation, high resistance to
burn-out and their improved high temperature durability. As they are imported from the US,
where they are in common use, their price generally limits their use to high value steel casting
or in blends with activated calcium bentonites.

Pouring molten metal into a green sand mould subjects the moulding-sand to considerable heat.
This heat removes moisture from the sand and destroys the clay-bonded structure (and
additives). If during pouring and cooling bentonite stays below the deactivation temperature, the
lamellar structure is maintained as well its ability to swell and develop cohesion. The
deactivition temperature varies with the bentonite type.

Prices for bentonite range between EUR 70 250/t, depending on the packaging and product
type (Czech Republic, 2003).
[32, CAEF, 1997], [73, ETSU, 1995], [202, TWG, 2002], [225, TWG, 2003]

2.5.1.2.2 Resins

Over the past few decades a range of chemical binders have been developed. These are single-
or multi-component systems which are mixed with the foundry sand until all the grains are
coated with a thin film. After mixing, a hardening reaction starts, binding the sand grains
together and developing mould strength. Resins can be classified according to the hardening
method:

- cold-setting resins
- gas-hardened resins
- hot-curing resins.

The various resin types will be discussed under Section 2.5.6. In Table 2.7 a survey of the
applicability of the various resins is given.

Chapter 2
58 Smitheries and Foundries Industry
Hardening Resin type
("commercial
name")
Mould
production
Core
production
Curing
temperature
Hardening
time (*)
Metal
types
Size of
series
Cold-setting Furan Medium to
large
Some 10 30 C 10 120
min
Ferrous +
non-
ferrous
Small to
large
Phenolic Large No 10 30 C 10 180
min
Ferrous Small to
large
Polyurethane
(Pepset/Pentex")
Small to
medium
Some 10 30 C 5 60 min Ferrous +
non-
ferrous
Small to
large
Resol - ester
("Alfaset")
Small to
Large
Some 10 30 C 5 400
min
Ferrous +
non-
ferrous
Small to
large
Alkyd oil Large Some 10 30 C 50 min steel Small
Silicate -
Water glass
Medium to
large
No 10 30 C 1 60 min Ferrous +
non-
ferrous
Small to
medium
Gas-hardening Phenolic/Furan
1
("Hardox")
Small Yes 10 30 C <60 s Ferrous +
non-
ferrous
All
Polyurethane
("Cold-box")
Small Yes 10 30 C <60 s Ferrous +
non-
ferrous
All
Resol
("Betaset")
Small Yes 10 30 C <60 s Ferrous +
non-
ferrous
All
Acryl/Epoxy
("Isoset")
No Yes 10 30 C <60 s Ferrous +
non-
ferrous
All
Silicate Small Yes 10 30 C <60 s Ferrous +
non-
ferrous
All
Thermosetting Oil Small Yes 180 240 C 10 60
min
Ferrous Small
"Warm-box" Seldom Yes 150 220 C 20 60 s Ferrous Medium
to large
"Hot-box" Seldom Yes 220 250 C 20 60 s Ferrous +
non-
ferrous
Medium
to large
"Croning" Yes Yes 250 270 C 120 180 s Ferrous +
non-
ferrous
Large
(*) i.e. the release time - the time in which the mould/core has gained sufficient strength to be released from the pattern
1
Not applied for capacities <20 tonnes/day
Table 2.7: Survey of various resin types and their applicability
[110, Vito, 2001]

2.5.1.2.3 Coal dust

Coal dust is usually added to green sand for cast iron moulding. It is used to a limited extent in
some non-ferrous foundries. Coal dust may be mixed with small amounts of resins and oil
products. During casting, the thermal degradation produces lustrous carbon, which improves
the casting surface finish and shake-out properties. Coal dust is added for three reasons:
- to create an inert atmosphere in the mould cavity during pouring, through the combustion of
organic compounds, which in turn slows down the oxidation of metal (slag formation)
- to reduce the penetration of metal in-between the quartz grains, through deposition of a
graphite film, which also creates a flat casting surface
- to reduce the amount of sand remaining on the casting surface upon shake-out.
Chapter 2
Smitheries and Foundries Industry 59
In addition to the black and sticky dust caused by handling, coal dust may contain or generate
polycyclic aromatic hydrocarbons (PAH) during pouring.

For steel castings, coal dust is not used because of carbon pick-up. In this case, it is usually
replaced by cereal binders such as starch or dextrin.

Various types of coal dust replacements exist. These consist of blends of high volatile, high
lustrous carbon materials blended with clays. They are generally more environmentally
acceptable than coal dust, i.e. producing less fume during casting, although some coal dust
replacements will generate more PAH in the sand.
[174, Brown, 2000], [225, TWG, 2003]

2.5.1.2.4 Cereal binders

Cereal binders are used mainly in steel foundries to increase the strength and toughness of the
green sand. There are two main types of cereal binder: starch and dextrin. Starch is the basic
material and is produced from a number of plant materials, with maize starch being the most
commonly used for foundry purposes. Dextrin is a repolymerised form of starch, produced
through a subsequent acid and thermal treatment of starch.

Starches can help to reduce expansion defects, since as they burn out, they allow the sand grains
to deform without deforming the mould. Cereals increase the green strength, dry strength and
toughness but can reduce the flowability. Dextrines improve the flowability and moisture
retention, preventing moulds from drying out and edges becoming friable.

Cereal additions do not improve the erosion resistance of the sand nor its resistance to metal
penetration.
[174, Brown, 2000], [175, Brown, 1999]

2.5.1.2.5 Iron oxide

Iron oxide reacts with quartz at high temperature and forms a low melting compound, fayalite.
This glassy plastic product sinters the grains together. It is mainly used in the production of core
sand, in order to reduce the formation of veins.
[110, Vito, 2001]

2.5.1.3 Running, gating, feeding and filtration

The different parts of the running and gating system are shown in Figure 2.26. The system
carries out the following functions:

- controls the flow of the metal into the mould cavity at the rate needed to avoid cold metal
defects in the casting
- avoids turbulence of metal entering the mould
- prevents slag and dross present in the melt from entering into the mould
- avoids high velocity impingement of the metal stream onto cores or mould surfaces
- encourages thermal gradients within the casting, which help to produce sound castings
- enables the casting to be separated from the running/gating system easily.

Chapter 2
60 Smitheries and Foundries Industry
Figure 2.26: Components of the running and gating system
[237, HUT, 2003]

The design of the running system needs to take account of the varying slag and dross forming
properties of the metal to be poured, e.g.:
- in lamellar iron, some furnace slag may be present but the melt is not subject to inclusions
due to oxidation
- nodular iron contains magnesium silicate and sulphide dross, arising from the nodularisation
treatment
- steel is susceptible to oxidation and slag formation
- aluminium alloys (and aluminium bronzes) are all subject to dross formation, whereby a
film of oxide forms immediately on any metal surface exposed to air.

Inclusions in steel can arise from slag entrapment, the erosion of furnace or ladle linings or
refractories, or from deoxidation. Filtration is now widely used to reduce the presence of
inclusions. The introduction of ceramic filters has even enabled simplifications to be made in
the design of running and gating systems. Various types of filters are shown in Figure 2.27.

Figure 2.27: Various types of filters
[237, HUT, 2003]

2.5.2 Sand preparation (transport, sieving, cooling, mixing)

2.5.2.1 Sand conditioning for green sand moulding

One of the major advantages in using green sand moulding is that the sand from the moulds can
be reconditioned for multiple re-use. The layout of a typical green sand plant is depicted in
Figure 2.28, and some examples of sand mixers are shown in Figure 2.29.

Chapter 2
Smitheries and Foundries Industry 61
Figure 2.28: Flow diagram for a typical green sand plant
[174, Brown, 2000]

Figure 2.29: Various types of sand mixers
[237, HUT, 2003]
Chapter 2
62 Smitheries and Foundries Industry
As the sand usually contains metallic elements such as flashes, pouring drops, pieces of sprue or
even small parts of casting, all these have to be removed, at first by means of magnetic iron
separators. If separation of the magnetic iron castings is not achieved or not possible, separation
may be carried out by Eddy current separators. The residual sand lumps are then broken up. It is
important not to crush the sand, in order to prevent the demixing of the sand and bentonite.

Usually, the sand has to be cooled in order to keep the moisture level of the prepared sand as
constant as possible and to avoid any loss by evaporation. Cooling is often performed in a
fluidised bed, which also allows the sand to be dedusted by removing excessive amounts of
fines.

The sand is then screened to remove the remaining lumps and stored before mixing with the
required amounts of additives, say clay, water, etc. to prepare the green sand for re-use.
[32, CAEF, 1997], [174, Brown, 2000]

2.5.3 Moulding with natural sand

Some foundries use naturally bonded sand. This is sand which contains a natural percentage of
clay. Only water needs to be added in order to activate the binding capacity. If needed, some
additives may be mixed as well. The approximate composition of natural sand is given in Table
2.8.

Compounds Approximate %
Quartz sand 80
Clay 15
Water 5
Table 2.8: Composition of natural sand
[126, Teknologisk, 2000]

Natural sand, which is found ready-mixed in nature, does not have the same high demands for
mixing equipment as synthetic sand. It is mainly used in small-sized non-ferrous (e.g. copper)
foundries and is not used in foundries casting iron and steel.

2.5.4 Moulding with clay-bonded sand (green sand moulding)

Green sand moulding is the most common moulding process. Green sand is generally not used
to make cores. Cores are formed using a chemical binding system. Green sand is the only
process that uses a moist sand mix. The mixture is made up of about 85 to 95 % silica (or
olivine or zircon) sand; 5 to 10 % bentonite clay; 3 to 9 % carbonaceous materials such as
powdered (sea) coal, petroleum products, corn starch or wood flour; and 2 to 5 per cent water.
The clay and water act as the binder, holding the sand grains together. The carbonaceous
materials burn off when the molten metal is poured into the mould, creating a reducing
atmosphere which prevents the metal from oxidising as it solidifies. Table 2.9 gives a survey of
the additives used for various types of metal castings.

Chapter 2
Smitheries and Foundries Industry 63
Metal Additives for green sand preparation
Nodular cast iron casting
Lamellar iron casting
Malleable iron casting
Bentonite
Cereal binder*
Coal dust
Steel casting Bentonite
Cereal binder*
Light metal and aluminium casting Bentonite
Cereal binder*
Magnesium-alloyed-aluminium casting Bentonite
Boric acid
Magnesium casting Bentonite
Powdered sulphur
Boric acid
Heavy metal casting (copper alloys) Bentonite
Cereal binder*
coal dust
* Optional additive
Table 2.9: Composition of additive mix for green sand preparation (excl. water)
[36, Winterhalter, et al., 1992]

Green sand, as demonstrated by its widespread use, has a number of advantages over other
casting methods. The process can be used for both ferrous and non-ferrous metal casting and it
can handle a more diverse range of products than any other casting method. For example, green
sand is used to produce the total range of castings, from small precision castings up to large
castings of up to a tonne. If uniform sand compaction and accurate control of sand properties are
maintained, very close tolerances can be obtained. The process also has the advantage of
requiring only a relatively short time to produce a mould compared to many other processes. In
addition, the relative simplicity of the process makes it ideally suited to a mechanised process.

Although manual moulding is still in use, machine moulding is currently the most widespread.
Two sequential operations must be performed by a moulding machine: the first is ramming the
sand, followed by separating the pattern from the compacted sand. The most commonly used
processes proceed from the working principles described hereafter.

Squeeze moulding machines use pressure to pack the sand, which is applied through a squeeze
head or by a multiple-piston squeeze head. Moulding by squeezing alone becomes less effective
as the depth of the half mould increases. In these cases, jolting the work table greatly increases
the compaction of the sand.

In impact moulding, sand is fed under gravity into the flask and compacted by the instantaneous
release of compressed air through a rapid acting valve. This process gives a high and uniform
densification, particularly in the sand surrounding the pattern.

Flaskless moulding, with both vertical and horizontal parting, achieves impressive production
efficiencies. High mould accuracies can be achieved, but this process requires an effective set-
up and high quality patterns to achieve the desired objective.
[32, CAEF, 1997], [42, US EPA, 1998]

2.5.5 Moulding with unbonded sand (V-process)

This process uses dry sand, rammed by vibration without any binder addition, with the sand
held between two polyethylene sheets by partial vacuum.

Chapter 2
64 Smitheries and Foundries Industry
The production of a half mould is illustrated in Figure 2.30. The successive steps of the process
are as follows:

- the pattern is fixed to an airtight chamber which is connected to a vacuum pump. The
pattern is vented by small diameter holes connecting through the airtight chamber
- a thin film of polyethylene vinyl acetate (PEVA), 75 to 100 microns thick, is heated up to
85 C
- this film, which is expanded under heat, is applied on the pattern and fixed by vacuum
applied through the airtight chamber
- a moulding box, in which a vacuum can be created, is placed on the pattern, and filled with
dry sand
- this sand is compacted by vibration, made even, and then a second film of PEVA is applied
to the sand
- air is extracted from the moulding box and at the same time, the vacuum is released in the
airtight chamber; the vacuum stiffens the sand and the half mould can then be stripped off
- the second half mould is made in the same way and both are assembled and closed, with the
vacuum still being applied
- the metal can now be poured, the two moulding boxes remaining under vacuum until the
casting is cool enough to be shaken out
- shake-out occurs simply by shutting off the vacuum: the sand flows out from the box
through the grid and may be recycled after dedusting and separation of the unburned plastic
sheets.
[174, Brown, 2000]

Figure 2.30: Vacuum moulding
[179, Hoppenstedt, 2002]

Chapter 2
Smitheries and Foundries Industry 65
2.5.6 Moulding and core-making with chemically-bonded sand

For core-making, chemical binding systems are primarily used. Cores require different physical
characteristics than moulds; therefore, the binding systems used to make cores may be different
from those used for moulds. Cores must be able to withstand the strong forces which can occur
when molten metal fills the mould, and often must be later removed from small passages in the
solidified casting. This means that the binding system used must produce strong, hard cores that
will collapse to allow removal after the casting has hardened. Therefore, cores are typically
formed from silica sand (and occasionally olivine, zircon or chromite sand), and strong
chemical binders. The sand and binder mix is placed in a core-box where it hardens into the
desired shape and is removed. Hardening, or curing, is accomplished with a chemical or
catalytic reaction or by heat.

Table 2.10 gives the relative share of the various core-making processes in German automobile
foundries in 1991. This shows that the amine cold-box and hot-box systems dominate the
market. Over 90 % of the automobile foundries use the amine cold-box system. The other
processes (croning, CO
2
-silicate) are mainly used for supplementary use, i.e. making cores with
specific requirements (size, thickness,).
[42, US EPA, 1998], [174, Brown, 2000]

System Number
Amine cold-box 44
Hot-box 10
Shell/croning 9
CO
2
silicate 3
Table 2.10: Core-making processes used in 48 automobile foundries in Germany, 1991
[174, Brown, 2000]

2.5.6.1 Cold-setting processes

The curing of cold-setting sands is effective at ambient temperature. The process begins when
the last component of the formulation has been introduced into the mix. It then continues for a
few minutes to several hours, depending on the process, the amount of binder and the hardening
agent strength.

These processes are more often used to make moulds than cores, especially for medium or large
sized castings.

2.5.6.1.1 Phenolic, acid catalysed

This process has been in use since 1958. As the components are relatively cheap, it is mainly
used for the production of large parts. It is applicable for all alloy types. The hardening of these
resins is more difficult and less regular compared to the furan resins.

The resins are either phenol-formaldehyde (PF), or urea-formaldehyde/phenol-formaldehyde
copolymers (UF/PF), both being "resols", with a formaldehyde/phenol ratio higher than one.
The catalysts are strong sulphonic acids, such as paratoluene, xylene or benzene-sulphonic, with
sometimes an addition of sulphuric acid, usually used in a diluted form.

Chapter 2
66 Smitheries and Foundries Industry
2.5.6.1.2 Furan, acid catalysed

These binders, firstly introduced in foundries in 1958, are commonly used for the moulding and
core-making of medium and large sized parts, for small and medium batch production and for
all alloy types. Only certain types are used for steel casting, as cracks, fins or pinholes may
occur. The process allows good flexibility in application and properties. Furfuryl alcohol (FA)
has the disadvantage of being a (strategic) basic product, which leaves it subject to market price
variations. Furan binders are comparable to phenolic binders in that their setting mechanism and
the acid catalysts used are the same for both processes. Some examples of furan cores are shown
in Figure 2.31.

Figure 2.31: Furan cores
[237, HUT, 2003]

The addition of an acid catalyst to a furan resin causes an exothermic polycondensation, which
hardens the binder. Furan resins are available with different formulations, all of which are based
on furfuryl alcohol:

- furan resin FA
- urea - formaldehyde - furfuryl alcohol UF - FA
- phenol - formaldehyde - furfuryl alcohol PF - FA
- urea - formaldehyde - phenol - furfuryl alcohol UF - PF - FA
- resorcinol - furfuryl alcohol R - FA.

A silane is nearly always added to enhance the resin-sand bond. The catalysts are strong
sulphonic acids, such as paratoluene, xylene or benzene-sulphonic, sometimes with an addition
of sulphuric or phosphoric acid, usually used in a diluted form.
[110, Vito, 2001]

2.5.6.1.3 Polyurethane (phenolic isocyanate)

This process is used to a minor extent for moulding and core-making, with restrictions for steel
casting, as cracks or pinholes may occur. These can be prevented however, by the addition of
iron oxide and by drying the moulds and cores. In some countries (e.g. Sweden) this type of
binder has not been used for 25 years, mainly because of its effect on the working environment.

This process is based on a polyaddition reaction between a phenolic resin and an isocyanate
(mostly MDI), catalysed by a pyridine derivative; with the result being a polyurethane structure.
All the components are in solution in aromatic and/or polar solvent (i.e. aliphatic solvent) with a
high boiling point. Contamination by water must be strictly avoided since water reacts strongly
with the isocyanate.

Chapter 2
Smitheries and Foundries Industry 67
2.5.6.1.4 Resol ester (alkaline phenolic ester hardened)

This process is used for small or medium batch production. It may be used for all alloy types but
is of special interest for lighter alloys, because of the ease of shake-out. The absence of nitrogen
in the compounding reagents is a benefit for steel casting.

The resin is an alkaline phenolic resol solution, which reacts with a liquid ester. The resin and
the ester produce an unstable complex, causing gelification. The complex disintegrates and
causes cross-polymerisation of the resin, producing a salt and an alcohol.

In this process, the curing speed is not adjusted by the addition rate of the hardener, but by using
different grades. The hardening time may vary from a few minutes to more than one hour.
Mechanical properties immediately after hardening are rather poor, but improve upon storage.

2.5.6.1.5 Alkyd oil, unbaked

This process is mostly used for single castings or for small batch production in steel foundries.
It provides the benefit of a good surface finish and good stripping properties. However, it is
expensive.

An oil modified polyester resin is mixed with isocyanate, producing a polyurethane resin, which
hardens slowly. The curing of the mould is accelerated by a catalyst and completed by heating
to 150 C.

2.5.6.1.6 Ester silicate

This process is used mostly in steel foundries, for medium and large sized castings in medium
and small series. It is similar to the unbaked alkyd oil process. However, it has poor shake-out
properties and lower mechanical resistance than organic resin-bonded systems.

The hardening of a silicate - ester sand proceeds through an intermediate step, which consists of
the hydrolysis of the ester by the alkaline silicate solution. This hydrolysis produces glycerol
and acetic acid, which precipitates a silicate gel to form the initial bond. Further strength
develops as the residual silicate dries.

2.5.6.1.7 Cement

This process is only used for very large sized castings. Its application is relevant for public
works applications. It does not generate any emission problems during moulding or core-
making.

2.5.6.2 Gas-hardened processes

In these processes, curing takes place by injecting a catalyst or a hardener in a gaseous form.
The curing speed can be very high, which allows high production rates to be achieved. They are
suitable for moulds and cores of limited size, in medium batch or mass production. Their use
has been constantly expanding over the last few years.

The chemistry behind many of these gas-hardened processes is similar to the cold-setting
processes. Because of the gaseous form of the catalysts, it is sometimes necessary to collect and
treat their emissions.

Chapter 2
68 Smitheries and Foundries Industry
2.5.6.2.1 Cold-box (amine hardened phenolic urethane)

This process is commonly used to make cores of up to 100 kg and more, and small moulds. It
provides a very smooth surface finish, and the dimensional accuracy is very high. The core
removal properties are excellent and the sand can easily be reclaimed. This process is the most
commonly used in core-making. An example cold-box core is shown in Figure 2.32.

Figure 2.32: Cold-box core
[237, HUT, 2003]

The chemistry of this polyurethane based process is very similar to the polyurethane cold-
setting one, e.g.: it involves a phenolic resin and an isocyanate (MDI). Only the catalyst is
different; here a tertiary amine is used, such as triethylamine (TEA), dimethylethylamine
(DMEA), dimethylisopropylamine (DMIA) or dimethylpropylamine (DMPA). The amine is
applied as a vapour, using compressed air, nitrogen or CO
2
as a carrier gas. Both resin and
isocyanate are in solution in aromatic and/or polar solvents that have high boiling points.
Contamination by water must be strictly avoided as it reacts strongly with isocyanate and
weakens the binder.

The amine is delivered either by a generator, in which a supporting gas, preferably inert, is
saturated with amine vapours, or by an injector, which measures out just the right amount of
amine required for the operation, this latter being carried to the sand by compressed air or
gaseous nitrogen. The addition rate for the binder varies from 1.0 to 2.0 % based on the sand
weight, with resin and isocyanate in a 50:50 ratio.

The amine is just a catalyst and is not consumed in the reaction. After curing, it remains in the
sand of the mould or core and has to be purged. The purging period currently lasts 10 to 15
times longer than the injection of the amine. The amount of amine required for curing is about
0.05 % based on the sand weight, but typically 0.10 to 0.15 % is used.

In a variant of the process called the Cold-box Plus process the corebox is heated to 40 80 C
using circulating hot water. This yields cores with even better mechanical properties, but results
in higher interval times.
[110, Vito, 2001]

2.5.6.2.2 Resol ester (alkaline phenolics methyl formate hardened)

This process is relatively new. Its advantages and disadvantages are similar to those of the cold-
setting process from which it is derived. As its price is relatively high, the process is mostly
used for core-making. Even though difficulties in recycling used sand have been reported, this
process is widely used, mainly because of its shake-out properties; the ability to prevent
cracking, veining and pinholes; and because of its low odour emissions. The process can be
used for all types of series and alloys but is mainly used for light and super light alloys (easy
shake-out) and steel (low crack risk).
Chapter 2
Smitheries and Foundries Industry 69
The resin is an alkaline phenolic resol, which reacts with methyl formate to produce methanol
and an alkaline formate. The phenolic resin is precipitated as a gel that binds all compounds
together. Further cross-linking leads to even higher strength on storage.

Methyl formate is liquid at ambient temperature, the boiling point being 32 C, but it is gasified
by air typically heated up to 80 C; which also acts as the carrier in the process. The gassing
period is always followed by an air purge, the purpose of which is to evenly distribute the
methyl formate throughout the mass of sand.

2.5.6.2.3 SO
2
hardened furan resins

This process is no longer in widespread use, although it is suitable for a large range of
applications in the manufacturing of small to medium sized moulds and cores in any alloy. The
sulphur catalyst may cause some metallurgical problems on the surface of ductile iron castings.
The major advantages it provides are long lifetime of the prepared sand, good mechanical
properties and shake-out performance, and the prevention of cracks. Nevertheless its effective
implementation is limited due to the adhesive character of the resin and due to concerns in using
sulphur dioxide as a hardener.

This process uses furan resins, which contain about 80 % furfuryl alcohol. Both resins
polymerise under acid conditions. These resins need to be mixed with the sand and with
oxidising agents, such as organic peroxides or hydrogen peroxide. The reaction between these
peroxides and the injected sulphur dioxide then generates sulphuric acid, which causes a rapid
polymerisation. The gassing period is always followed by a purging period, the purpose of
which is to remove the excess unreacted sulphur dioxide from the mass of sand.
[32, CAEF, 1997], [174, Brown, 2000]

2.5.6.2.4 SO
2
hardened epoxy/acrylic (free radical curing)

This process offers many advantages: good compaction ability; long bench lifetime of the
prepared sand (the mixers or shooting heads do not need to be cleaned); good mechanical
properties; no nitrogen, phenol or formaldehyde; good shake-out performance; and no crack
formation. However, the higher cost of binder constituents is one major disadvantage though.

This process is characterised not by the type of resin, but by the principle of its cross-linking,
which occurs through free radicals. The resin must contain carbon double bonds: polyester -
acrylic, polyester - urethane or polyester - epoxy resins may be used. These resins generally
have a low molecular weight and are diluted with organic solvents at nearly 50 % by weight,
though solvent free types are also available. They are mixed with an organic peroxide, which
acts as a reaction initiator. For curing, sulphur dioxide is carried by an inert gas, such as carbon
dioxide or nitrogen, through the sand.
The gassing period is always followed by purging with the same inert gas that was used for
curing, the purpose of which is to remove the unreacted excess sulphuric dioxide from the sand
mass.

2.5.6.2.5 CO
2
hardened sodium silicate (water glass)

This process presents real advantages: it is cheap, easy to handle and environmentally clean.
Also for workers health and for reliability of operation, this technique has advantages over
organic binders. Its use however, has declined in popularity due to technical reasons, such as
poor compaction ability and problems with collapsibility, poor mechanical strength, sensitivity
to crumbling, moisture pick-up and poor reclaimability. Furthermore, the use of water glass can
lead to increased cleaning costs. The cores reach their full strength only after a drying period.
This reduces their applicability in automated processes. The technique therefore finds its main
application in small scale foundries.
Chapter 2
70 Smitheries and Foundries Industry
The silicate generally used is sodium silicate, defined by its concentration (dry solid content)
and modulus (silica soda ratio SiO
2
/Na
2
O). This modulus ranges from 2.0 to 2.8, the most
common being 2.0 to 2.3. This silicate is mixed with the sand at a concentration of between 2 to
4 %. Shake-out and de-coring additives are commonly used and are usually pre-mixed with the
silicate.

Curing occurs by gassing with carbon dioxide, due to its slightly acidic character. The CO
2
-ratio
should not exceed 1 to 2 % of the mass of sand, with a gassing time in the range of 10 to 60
seconds. The cured moulds and cores do not require purging.
[126, Teknologisk, 2000], [152, Notzon and Heil, 1998]

2.5.6.2.6 CO
2
hardened alkaline phenolic

This process was introduced for the first time in 1989 and has been undergoing improvements
ever since. It is now commercially available and is in use in a few locations, but its use is not
yet widespread.

The resin is an alkaline phenolic one, containing a linking substance stabilised at a high pH,
approximately 14. Curing occurs by gassing with carbon dioxide, which dissolves in the water
solvent of the resin, so lowering its pH and activating the linking substance.

2.5.6.3 Hot curing processes

In these processes, curing takes place by heating the sand-resin mix or, more often, by allowing
it to come into contact with the heated pattern equipment. They all provide a high dimensional
accuracy, which can only be achieved by means of using high quality (metal) patterns, which
can be very expensive. For this reason, hot curing processes are used for the production of cores
of limited size, mostly in mass production processes. Their use, very widespread for years, is
now on the decline, as they are being replaced by gas cured processes.

Hot curing processes are characterised by major emission problems: when heated, the resins and
catalysts emit noxious chemicals including ammonia and formaldehyde that can be the source of
odour nuisance. A number of abatement techniques have been tried to solve these problems,
such as scrubbing, combustion or biological abatement, but none of these has proven to be
effective.

2.5.6.3.1 Hot-box, phenolic and/or furan based

This process can produce cores of a high dimensional accuracy and good mechanical strength,
but to achieve these, the operators need a very good knowledge and control of the production
process. The limits of the process are set by its cost, principally for the resin, energy and pattern
equipment, and by the poor internal working conditions caused. It is currently used in the
manufacturing of small or medium sized cores in mass production.

A resin binder and a heat activated catalyst are pre-mixed with sand and the mixture is blown
into a heated corebox or pattern, where it is cured for approximately 5 to 60 seconds.

A wide range of resins may be employed, such as:

- urea - formaldehyde UF
- urea - formaldehyde - furfuryl alcohol UF - FA
- phenol - formaldehyde PF
- phenol - formaldehyde - furfuryl alcohol PF - FA
- urea - formaldehyde - phenol - formaldehyde UF - PF
- urea - formaldehyde - phenol - formaldehyde - furfuryl alcohol UF - PF FA.
Chapter 2
Smitheries and Foundries Industry 71
The catalysts are ammonium salts of mineral acids, sometimes with urea additions to reduce
free formaldehyde. Additionally, other additives are used, such as silanes, iron oxides,
preservatives and silicon oil.

The addition ratio for the resin varies from 1.2 to 3.0 % based on sand weight, with an average
of about 1.8 %. The addition ratio for the catalyst varies from 10 to 25 % based on resin weight,
with most formulations being optimised at 20 %. The temperature mostly used for the pattern
ranges from 230 C to 290 C, with the optimum range being 220 C to 250 C. If excessive heat
is used in an attempt to accelerate the curing time, the core surface can be burnt, producing
brittleness during pouring.

2.5.6.3.2 Warm-box

This process is very similar to the hot-box process and uses the same production techniques.
Only the type of resin differs, allowing curing at a lower temperature. However, this kind of
resin is significantly more expensive than those in use in the hot-box process. Therefore the
warm-box process, in spite of some real advantages, is not generally find widespread use.

The binder is furfuryl alcohol-based, with a typical composition containing around 70 %
furfuryl alcohol or a low polymer of furfuryl alcohol. The catalysts are copper salts derived
from aromatic sulphonic acids, in water or an alcohol solution. The distinctive feature of these
catalysts is their excellent stability at ambient temperature and their relatively low dissociation
temperature, which is 150 - 170 C. Accordingly, the tool temperature can be held at around
180 C, which leads to major energy savings of about 15 to 25 % compared to the hot-box.

2.5.6.3.3 Shell (Croning)

This process is the only one among all moulding and core-making processes that can use pre-
coated sand, directly available from suppliers and ready for use, although pre-coating of the
sand may also be performed at the foundry.

The sand is cured by heating it in a metallic pattern, producing a hardened surface layer. The
unheated or uncured sand may be discarded by turning the pattern upside down and then re-
used. The cured sand forms a "shell", which has given its name to the process.

This process gives high dimensional accuracy and a good surface finish to the castings, good
shake-out and de-coring properties and allows a nearly unlimited storage time for the pre-coated
sand. The process limits are the price of the pre-coated sand and the pattern equipment costs. Its
use is restricted to the production of small or medium sized moulds and cores in mass
production. Example cores and mould are shown in Figure 2.33.

Chapter 2
72 Smitheries and Foundries Industry
Figure 2.33: Croning cores (top) and moulds (bottom)
[237, HUT, 2003]

The resin used for pre-coating is a phenolic "novolac", with a formaldehyde/phenol ratio lower
than 1. Hexamethylenetetramine is added to the resin as a hardener. Hexa decomposes at
160 C into its 2 basic components: formaldehyde and ammonia. When pre-coated sand contacts
the heated pattern, the hexamethylenetetramine decomposes and the formaldehyde cross-links
the resin binder to form the characteristic strong bond.

2.5.6.3.4 Linseed oil

This ancient process is probably the oldest one processing chemically-bonded sands. It is very
easy to use, does not need sophisticated patterns and has a good resistance to veining and
cracking. It is still in widespread use for specialised small cores.

These sands are bonded with a mixture of drying oils, these usually being to a great extent
linseed oil, often with an addition of dextrin and a few per cent water. The addition rate of oil
varies from 0.8 to 4 %, based on sand weight. Drying agents may also be added. Curing occurs
by cross-linking of the unsaturated fatty acids contained in the drying oils, induced by
atmospheric oxygen and accelerated by heating in a drying-oven at a temperature between of
190 and 260 C for 1 to 2 hours.

2.5.6.3.5 Alkyd oil, baked

This process is similar to the previously described unbaked alkyd oil process, the only
difference being in this case heat treatment of the parts in order to accelerate the curing.

Chapter 2
Smitheries and Foundries Industry 73
2.5.6.4 Coating of chemically-bonded sand moulds and cores

The skill of the founder is to produce high quality castings, without defects, and requiring only
minimal finishing and repair. To achieve this, the ideal is to minimise all the interactions that
may occur between mould, core and metal during pouring. These effects may be generated by
different causes, such as sand expansion, ramming defects, abrasion of sand, metal penetration,
chemical decomposition or by interaction between the binders, etc.

For these reasons, it is often beneficial to coat the mould or the core with a refractory lining, in
order to give a smooth surface finish to the casting and to reduce fettling cost.

2.5.6.4.1 Composition of coatings

Coatings are available as ready-to-use products or as a mass to dilute with water or alcohol.
They usually contain the following components:

- one or more refractory fillings, such as talc, pyrophyllite, mica, zircon, magnesite, quartz,
etc., or blacking
- a make-up liquid, which may be an alcohol (e.g. isopropanol, ethanol) or water
- high temperature bonding agents, such as bentonites, resins, boric acid
- rheological agents, such as bentonites or synthetic polymers
- additives, such as surfactants, foam breakers, fungicides, etc.

2.5.6.4.2 Coating process

The coating may be applied on the mould or on the core by different means:

- by brushing, for small cores or for localised application
- by dipping, for complex shaped cores; this process often being automated
- by spraying, usually airless
- by flow-coating, for large or medium sized moulds or cores.

If alcohol-based coatings are applied, the working area needs to be ventilated in order to prevent
the danger of fire or explosion. The coated moulds and cores are mostly burnt out, which limits
the emissions. If they are not burnt out then drying is carried out under exhaust with the VOC
emissions maintained under control. Flow coating with alcohol-based coatings and subsequent
firing or drying is presented in Figure 2.34.

Figure 2.34: Flow coating with alcohol-based coating, followed by firing or drying
[237, HUT, 2003]

The drying of water-based coatings occurs by heating in a drying furnace using hot air, infrared
or microwaves. The dip-coating of cores and subsequent hot air drying is presented in Figure
2.35. Water-based coatings do not generate emission problems, neither during coating nor
during drying. For these reasons, water coatings are now increasingly replacing alcohol-based
Chapter 2
74 Smitheries and Foundries Industry
ones. However, their application faces technical constraints in terms of the coating quality
continuity and drying constraints. A full discussion of water-based versus alcohol-based
coatings is provided in Section 4.3.3.5.
[143, Inasmet and CTIF, 2002]

Figure 2.35: Dip-coating with water-based coatings, followed by drying in a hot-air furnace
[237, HUT, 2003]

2.5.7 Expendable pattern casting

In expendable pattern casting, the pattern is not removed from the mould before pouring. The
pattern, which is made of expanded material, is a single-use one, which is destroyed when
poured. These expandable patterns may be embedded either in chemically-bonded sands or in
binderless sand, hardened by vibration.

This process, commonly called "Lost Foam casting, was developed 30 years ago and its
commercial growth was initially rather slow. However, during the last 10 to 15 years, it is has
become used more often, primarily for the mass production of automotive parts or similar
products, in spite of significant set-up difficulties.

2.5.7.1 Unbonded sand Lost Foam process

The Lost Foam process (Figure 2.36) starts with the 3D-CAD design and production of a
precision moulded foam pattern, made from expanded polystyrene (EPS) or PMMA, produced
by automated injection moulding machines. These patterns can be made from one piece or by
assembling a number of parts by adhesives. Here usually, a number of patterns, dependent on
size, are fixed to a runner-down gating system, made from the same material as the pattern,
thereby generating a cluster.

Chapter 2
Smitheries and Foundries Industry 75
Figure 2.36: The Lost Foam process
[110, Vito, 2001]

The clusters are dipped in a water-based refractory type coating, which creates a barrier between
the molten metal and the sand during pouring. After drying, the cluster is positioned in a
moulding flask, which is filled with unbonded sand. A low-viscosity resin may be added to the
sand, in order to prevent deformation during compaction. The sand is compacted by triaxial
vibration around the patterns, penetrating the holes and reproducing exactly all the details, thus
making the core and core-making equipment unnecessary. Upon pouring, the molten metal
causes a pyrolysis of the polystyrene and fills the emptied space.

The sand, mostly quartz, needs to be very permeable, in order to ensure the evacuation of the
pyrolysis gases. Sand with an AFS-number of 35 50 is used. Interruption of the pouring may
cause a collapse of the sand mould. Therefore an automatic casting system is often used.

The cast pieces are characterised by a very good dimensional precision, as shown in Figure
2.37. The technique can be used for any type of alloy and is used for medium to large sized
series production. The process is environmentally friendly during the moulding but produces
gaseous emissions during pouring and shake-out, due to the vaporisation of the expendable
patterns. Lost Foam casting enables metal casters to produce complex parts which would often
not be possible using other methods. The process allows designers to develop more complex
shapes, to reduce machining need and to minimise assembly operations. Proper controls must be
exercised in each step of the process to ensure consistent high quality castings. A lack of in-
depth industry-wide knowledge on how to set up proper control measures has slowed adoption
of the Lost Foam casting process.
[32, CAEF, 1997], [110, Vito, 2001], [144, US Dept. of Energy, 1998]

Chapter 2
76 Smitheries and Foundries Industry
Figure 2.37: Expendable polystyrene pattern (r.) for the Lost Foam process and an example of a
casting produced by the Lost Foam method (l.)
[237, HUT, 2003]

2.5.7.2 Chemically-bonded sand Full mould process

The Lost Foam process can also be used to produce single-piece prototypes, thereby permitting
a shortening of the delivery time. Furthermore the technique can be applied for the production
of very big pieces, such as bases for metal processing machines, or presses, etc. in cast iron,
steel or non-ferrous alloys. The wall thickness may range from 5 mm up to 1000 mm. Pieces of
up to 50 tonnes have been cast using this technique. For the production of these bigger pieces,
the application of a (furan) binder is needed in order to attain the necessary mould strength.

The Lost Foam process with chemically-bonded sand, in this case also called the full mould
process, is mostly used for producing middle or large sized castings, in single production or in
small batch series.

The patterns are made from low density expanded materials, such as:
- white, foamed polystyrene, with a density of 16 to 20 g/dm
3
- blue, foamed polystyrene, also called "poresta-blue", with a density of 18 to 22 g/dm
3
- foamed polymethylmetacrylate, also called PMMA, with a density of 25 g/dm
3
.
These materials are cut out as various parts, which are then assembled and stuck together with
hot melt glue, thus creating the final shape of the casting, also taking into consideration any
possible shrinkage of the metal.

The assembled pattern, with its gating and feeding system, has to be coated, mostly with a
water-based coating system, and thoroughly dried before being embedded in the sand. The
binding process of the sand has to be specially selected in order to avoid any moisture pick-up
by the pattern before pouring.

This process, as far as moulding is concerned, is very environmentally friendly, but produces
gaseous emissions during pouring and shake-out, due to the vaporisation of the expandable
patterns and destruction of the sand binder.

As with the unbounded process, uninterrupted casting is very important in order to prevent the
mould form collapsing. For the casting of big pieces, pouring is performed using two (or more)
pouring ladles and two (or more) entry holes simultaneously.

The full mould process has the following advantages:
- high size precision
- production of complex geometries, especially internal cavities
- integration of several parts into one casted piece
- reduction or exclusion of release edges
- possibility of thermal sand regeneration.
Chapter 2
Smitheries and Foundries Industry 77
Despite the fact that the Lost Foam and full mould techniques have been known for a long time,
they are not widely applied in Europe. This is primarily because their optimisation requires a lot
of research and development. The main difficulties are:
- definition of the product range: determining which pieces can be made more easily as
compared to traditional methods
- selection of the compaction technique: application of the sand in order to perfectly fill all
cavities
- choice of coating and sand type: these should have a sufficient gas permeability to allow the
combustion gases to escape.
[110, Vito, 2001]

2.5.8 Permanent (metal) moulds preparation

Permanent moulds are made from metallic elements adapted to the shape of the casting, whose
assembly allows the moulding, pouring and stripping of the casting. These metal moulds are
used for gravitational casting, high pressure die-casting, centrifugal casting, continuous casting
and for low pressure casting. Unlike sand moulds, they can be re-used many times and for this
reason, they are called permanent moulds.

If the shape of the casting cannot be easily made by the use of metallic core pins, sand cores
may be used. For instance, sand cores are used to form the widening-out of centrifugal cast
pipes.

Permanent moulds are generally coated with a "white or black wash"; these water-based release
agents are refractory based in the case of the white one and graphite based for the black one.
Their function is to provide protection of the die, to regulate die cooling by water evaporation,
and to provide lubrication. They also act as parting compounds.

In a few cases, blackening is provided by burning acetylene in the absence of air, producing
acetylene black that partially sticks to the die. The non-adhesive black then has to be collected
and filtered before emission.

Permanent moulds (or dies) are generally not made in the foundry, although die-casting
foundries do have a workshop for die assembly, maintenance and repair. This type of operation,
however, does not generate any environmental effects.

2.5.9 Investment casting and ceramic shell

This process is used to produce intricate, thin section parts with high dimensional accuracy, fine
details, and very smooth surfaces.

The process steps are depicted in Figure 2.38. They begin with the manufacture of expendable
wax patterns, by injecting molten wax into an aluminium or epoxy die to form a pattern that is
virtually an exact replica of the desired casting. The wax may contain fillers. For smaller
castings, several wax patterns are attached to a wax gating system. Water soluble die release
agents are used to facilitate the wax model take-out.

The wax patterns are cleaned with water or organic solvent and coated with a wetting agent,
which helps the ceramic slurry to adhere to the wax. The cluster is then dipped in a liquid
ceramic slurry, stuccoed with granular silica, zircon or alumina/silica refractories and then dried
before the application of the next coat. The coating process is continued until a sufficiently thick
shell is established.

The dried mould is then de-waxed by inserting it into a steam autoclave in which the wax
patterns are melted out, or into a "flash-furnace", in which the wax may be partially burnt out.
Chapter 2
78 Smitheries and Foundries Industry
After this, the shell is fired in a furnace at a high temperature. This burns out the residual wax
and hardens the ceramic, leaving a one piece ceramic shell mould, into which the molten metal
is poured to form the casting.

Figure 2.38: Investment casting process
[110, Vito, 2001]

In this process, air pollution may occur during two operations: investing and de-waxing.

Investing:
The liquid ceramic slurry consists of a binder and a very fine refractory powder, reduced to
slime by permanent mixing. The binder may be colloidal silica, hydrolysed ethyl silicate or
hydrolysed sodium silicate, the choice of the technology applied being determined by technical
reasons. When using ethyl silicate, drying is faster, which allows a higher cycle rate, but ethyl
alcohol vapours are emitted. These vapours, if present in too large an amount, have to be
collected and treated before emission.

De-waxing and firing:
De-waxing by means of a steam autoclave does not produce any significant impact on air.
However, this is not the case if the de-waxing is conducted in a firing oven, and during the
firing of the shell, where wax residues are burned.

Chapter 2
Smitheries and Foundries Industry 79
This burning occurs often in a lack of air, which results in the formation of carbon black
particles. These particles have to be collected and either destroyed by afterburning or removed.
Ceramic filters have been successfully used for this purpose, as they suit the high temperature of
the exhaust gases, but this very recent technology has not yet become widespread.

This technique is used e.g. for precision casting and art casting. It is generally not applied in
large-scale foundries (>20 tonnes/day capacity), and therefore is mentioned here only for
informative reasons.

The ceramic shell process (patented as Replicast

) extends the quality and precision of

investment casting to larger components, using a combination of Lost Foam and investment
casting principles. The technique uses an inert, fired ceramic mould. To create the mould, an
expanded polystyrene replica of the required component is produced, which is dimensionally
precise and of a high surface finish. No parting lines or cores are required, nor draft angles, and
polystyrene replicas can be glued together to create complex geometries. The polystyrene is
burnt out before casting, allowing a wide range of alloys to be cast in the mould - from ultra low
carbon stainless steel to nickel based alloys. This is in contrast to the Lost Foam process where
liquid metal consumes and replaces the polystyrene pattern, making it unsuitable for the vast
majority of steel components (polystyrene is 92 % by weight carbon).
[219, Castings Technology International, 2003]

2.6 Casting

Pouring is the central activity in casting production. The finished mould is filled with the liquid
metal under the effects of gravitational or centrifugal forces or pressure. After pouring the
casting is cooled to allow solidification and is then removed from the mould for further cooling
and treatment.
[32, CAEF, 1997]

2.6.1 Casting in lost moulds

2.6.1.1 Pouring

There are two types of ladle generally used in pouring liquid metal: lip and teapot ladles. A third
type (i.e. the bottom ladle) is more specific to steel.

- Lip pour: In this type of ladle (Figure 2.39) the metal is charged over the lip, with the flow
controlled by tilting the ladle using a geared handwheel. Since the metal flows from the top
of the ladle, the metal surface must be slag-free or a skimmer must be used to prevent slag
entering the mould. Lip pour ladles are used for pouring small steel castings

Chapter 2
80 Smitheries and Foundries Industry
Figure 2.39: Lip pour ladle for molten metal pouring
[237, HUT, 2003]

- Teapot ladles: As depicted in Figure 2.40, a refractory dam before the ladle lip ensures that
the metal is drawn from the bottom of the ladle, so that the stream is slag free. The molten
metal is generally cleaner than from a lip pour ladle. One disadvantage is that the narrow
spout may occasionally permit the liquid steel to freeze if the heat is tapped cold or if
pouring is prolonged

Figure 2.40: Teapot ladle for molten metal pouring
[237, HUT, 2003]

- Bottom pour ladles: The ladle is fitted with a pouring nozzle in its base, closed by a
refractory stopper rod. The metal is drawn from the bottom and is therefore slag-free and
non-metallics, such as deoxidation products, are able to float out of the melt. The metal
stream flows downwards, so that there is no movement of the stream during pouring. The
disadvantage is that the velocity and the rate of flow change during pouring as the
ferrostatic head changes.

Chapter 2
Smitheries and Foundries Industry 81
Figure 2.41: Bottom pour ladle for molten metal pouring
[237, HUT, 2003]

Automatic casting lines are often equipped with a pouring furnace. The working principle
behind one example furnace is depicted in Figure 2.42. The casting line halts when a mould is in
the correct position, i.e. under the pouring exit. Metal is poured during a fixed period by
elevating a stopper. Because the metal level in the casting recipient is held constant, a fixed
volume of molten metal is poured into the mould. The metal level in the furnace is controlled by
a floating device, which controls the gas pressure inside the furnace. The casting furnace is
refilled with metal from the melting furnace at fixed time intervals.
[110, Vito, 2001], [174, Brown, 2000]

Figure 2.42: Pouring furnace
[110, Vito, 2001]

2.6.1.2 Solidification (1
st
cooling)

The poured moulds are transported along the moulding line onto the cooling lines. The length of
the cooling line determines the final temperature of the casting at the point of shake-out. This
temperature must be low enough to provide the casting with sufficient strength during shake-out
and further manipulation.

Big moulds are not moved during cooling. The cooling time can be up to several days.
[110, Vito, 2001]
Chapter 2
82 Smitheries and Foundries Industry
2.6.1.3 Shake-out

In the case of individual moulding, clay or chemically bonded moulds can usually be destroyed
by vibration. In most cases this is accomplished by placing the moulding box on a vibrating
grate, using a crane to do this. As a result of the vibration, the sand is separated from the casting
and the moulding box. The casting and moulding box remain on the grate rods, whereas the
sand falls through and is subsequently reprocessed. The casting is usually transported to a
cooling area for further cooling in ambient air (second cooling).

A similar shake-out procedure is often carried out in mechanised systems and older low-
capacity automated systems (Figure 2.43). Here the moulding boxes are removed from
conveyor belts, by means of hoists or other transfer equipment, and placed on vibrators. Finally,
the castings are left to cool down or placed in a casting cooling device. In many systems, the
sand casting is pressed out of the moulding box and a subsequent controlled cooling of the
castings and sand is achieved in combined or separate cooling devices, such as in cooling
drums, swing pipes, chain conveyors, fluidised bed coolers, etc.

Figure 2.43: Shake-out at the end of an automated moulding line
[237, HUT, 2003]

Vacuum bonded sand moulds are destroyed by releasing the vacuum. The casting bucket or
moulding box, containing the loose sand and the casting, is emptied and then the casting is
cooled by one of the described methods.
[32, CAEF, 1997]

2.6.1.4 Casting cooling (2
nd
cooling)

The controlled cooling of castings and sand is carried out in rotary drums, swing drums or on
oscillating conveyor troughs. The casting is cooled in oscillating conveyors or in cable-car
baskets. In many cases an airflow, which is often directed in the opposite direction to that of the
casting, is used for cooling. In some cases, fine water jets are used in order to increase the
cooling effect.

2.6.2 Casting in permanent moulds

2.6.2.1 Gravity and low-pressure die-casting

Gravity and low-pressure die-casting make use of a permanent steel die into which the melt is
poured under the influence of gravity or a low-pressure gas. Sand cores can be used to form
undercuts and complex interior shapes in the casting. Due to the rapid solidification process,
permanent mould castings have a dense fine-grained structure with good strength
characteristics.

Chapter 2
Smitheries and Foundries Industry 83
The principle of low-pressure die-casting is shown in Figure 2.44. A metal die is mounted above
a sealed furnace containing molten metal. A refractory-lined riser tube extends from the bottom
of the die into the molten metal. When air is introduced into the furnace under low pressure
(15 100 kPa) the molten metal rises up the tube to flow into a die cavity with low turbulence,
the air in the die escaping through vents and the parting lines of the die. When the metal has
solidified, the air pressure is released, allowing the still-molten metal in the riser tube to fall
back into the furnace. After a further cooling time, the die is opened and the casting extracted.
Due to the absence of feeders and risers, the casting yield is exceptionally high, generally over
90 %. A good dimensional accuracy and surface finish are possible and complex castings can be
made using sand cores. This technique is typically used for aluminium castings, e.g. automotive
parts such as wheels, cylinder heads, and electric motor housings, and for domestic kitchenware.
The die must be coated to optimise the casting release and cooling. In general, a die coating is
applied once per shift. The die life is normally around 30000 50000 shots. An example of a
low-presure die-casting machine is given in Figure 2.45.

Figure 2.44: The principle of a low-pressure die-casting machine
[175, Brown, 1999]

Figure 2.45: Low-pressure die-casting machine
[237, HUT, 2003]

In gravity die-casting, the molten metal is poured under gravity into a permanent mould or die.
Gravity die-casting machines range from simple, hand operated rack and pinion die sets,
manually poured, to carousel machines, which usually have a tilting mechanism for filling the
Chapter 2
84 Smitheries and Foundries Industry
die and which are often operated using a pouring robot. The dies are coated with a refractory-
based coating, controll the cooling rate. The time before the casting can be extracted from the
die varies from 4 to 10 minutes depending on the type of casting. The process is therefore
relatively slow, compared to pressure die-casting. To achieve reasonable output rates, a manual
operator will operate 2 4 die sets in sequence, allowing an output of 30 60 castings an hour.
Automatic carousel machines may have 4 6 stations with multiple die sets, allowing
production rates of around 1 casting per minute to be achieved. The process is most widely used
in aluminium casting for series of 1000 to more than 100000 pieces per year for e.g. manifolds,
cylinder heads, water pumps.

Permanent mould coatings are typically formulated using water as a carrier, a high temperature
binder (normally sodium silicate) and a refractory filler or blend of fillers. There are two
categories of coatings:
- insulating: containing blends of insulating minerals such as talc, mica, kieselguhr, titanium
dioxide, alumina, etc.
- lubricating: based on colloidal graphite or boron nitride to aid release of the casting.

Coatings are generally sprayed onto the die. Careful attention to die preparation, coating
preparation and application and the type of coating equipment utilised can yield significant
quality and productivity benefits.
[175, Brown, 1999]

2.6.2.2 High-pressure die-casting

The term die-casting often implies high-pressure die-casting. The process utilises a
permanent die (metal mould) into which molten metal is forced under high pressure. The
application of high pressure causes a high and turbulent metal flow, which allows the
production of castings with high surface areas and low wall thicknesses. Dies are usually made
from two blocks of steel, each containing part of the cavity, which are locked together while the
casting is made. Due to the high metal pressure, the maximum size of the casting is limited by
the maximum locking force of the casting halves. Retractable and removable cores are used to
form internal surfaces. Due to the high metal pressures, only metal cores can be used. This
limits the complexity of the casted piece. The metal is held under pressure until it cools and
solidifies. The die halves are then opened and the casting is removed, usually by means of an
automatic ejection system. Dies are preheated and lubricated before being used, and are either
air- or water-cooled to maintain the desired operating temperature. Two basic types of high-
pressure die-casting (HPDC) machines are used: hot-chamber or cold-chamber (see Figure
2.46).

Figure 2.46: Cold-chamber and hot-chamber high-pressure die-casting devices
[42, US EPA, 1998]

Chapter 2
Smitheries and Foundries Industry 85
Hot-chamber HPDC machines comprise a molten metal reservoir, the die, and a metal-
transferring device, which automatically withdraws molten metal from the reservoir and forces
it under pressure into the die. A steel piston and a cylinder system with a gooseneck is used to
create the necessary pressure within the die. A gooseneck machine utilises a cast iron channel to
transfer the molten metal from the reservoir to the die. Pressures can range from a few bar to
over 350 bar. Hot-chamber techniques are mainly used for zinc alloys and magnesium alloys.

Cold-chamber HPDC machines have molten metal reservoirs separate from the casting machine.
Just enough metal for one casting is ladled by hand, or mechanically, into a small chamber,
from which it is forced into the die under high pressure. Pressure is produced through a
hydraulic system connected to a piston, and is typically in the range of a few hundred bar to
700 bar. In cold chamber machines, the metal is just above the melting point and is in a slush-
like state. Since the metal is in contact with the piston and cylinder for only a short period of
time, the process is mainly applicable to aluminium alloys, and to a lesser extent magnesium
alloys, zinc alloys, and even high melting point alloys such as brasses and bronzes.

Proper lubrication of the dies and plungers is essential for successful high-pressure die-casting.
The die lubrication affects the casting quality, density, and surface finish, the ease of cavity fill,
and the ease of casting ejection. Proper lubrication can also speed up the casting rate, reduce
maintenance requirements, and reduce the build up of material on the die face. Although
specific formulations are proprietary, in general, lubricants (also called release agents) are a
mixture of a lubricant and a carrier material. Formulations may also include additives to inhibit
corrosion, to increase stability during storage, and to resist bacterial degradation. Lubricating
materials are typically mineral oils and waxes in water emulsions. Silicone oils and synthetic
waxes are finding increased use. Both water-based lubricants and solvent-based lubricants are in
use today. Water-based lubricants however dominate the market (95 %). The lubricant is
applied to the open die by spraying, between each shot. Dilute solutions of lubricant are used
(1:20 1:200 release agent:water ratio). Alternative electrostatic powder coatings are currently
being developed.

High-pressure die-casting is not applicable to steel and high melting point alloys. The technique
finds widespread application for aluminium castings. Dies are expensive but can have a life of
150000 shots. The process is therefore most suitable for long runs of castings. One major
advantage of HPDC over other casting methods is that the castings produced can have very
complex shapes. The ability to cast complex shapes often makes it possible to manufacture a
product from a single casting instead of from an assembly of cast components. This can greatly
reduce casting costs, as well as the costs associated with fabrication and machining.
Furthermore, HPDC produces castings which have a high degree of dimensional accuracy and
surface definition compared to other casting methods, which may help to reduce or eliminate
costly machining steps. Finally, castings with relatively thin wall sections can be produced
using the HPDC method. This can result in substantial savings in material costs and reductions
in component weight.

Relatively little waste is generated in the actual HPDC process compared to other metal casting
processes. However, some gaseous and fume emissions occur during metal injection. Metal
oxide fumes are generated as some of the metal vaporises and condenses. Gaseous emissions
can originate from: the molten metal itself; the evolution of chemicals from the lubricant as it is
sprayed onto the hot metal die and as it contacts the molten metal. Emissions to water may
occur from any leakage or spillage of hydraulic oil or heating oil and from cooling water.
[42, US EPA, 1998], [128, IHOBE, 1998], [175, Brown, 1999], [225, TWG, 2003]

Chapter 2
86 Smitheries and Foundries Industry
2.6.2.3 Centrifugal casting

In centrifugal casting, a permanent mould rotates at high speed around its axis as the metal is
poured. The speed of rotation and the metal pouring rate vary with the alloy and the size and
shape being cast. The rotating axis is generally horizontal or under a small angle (Figure 2.47).
Some specific devices apply rotation around a vertical axis. The resulting material has a very
dense structure and properties that cannot be obtained with sand casting.

This technique is used to produce cylindrical products in cast iron, steel, and in alloys of
aluminium, copper and nickel. Typical parts made by this process are pipes, boilers, pressure
vessels, flywheels, cylinder liners and other parts that are axisymmetric.
[179, Hoppenstedt, 2002]

Figure 2.47: Schematic representation of a centrifugal casting machine
[179, Hoppenstedt, 2002]

2.6.2.4 Continuous casting

Continuous casting is a high productivity device for the production of bars, tubes and profiles,
where through rapid cooling, a fine-grained material with good mechanical properties is
obtained. In continuous casting, the molten metal is cast into a water-cooled die, which is open
at the bottom or at the side (Figure 2.48). The die gives the desired form to the product. Through
intensive cooling, the outside of the metal product solidifies, while it is slowly pulled out of the
mould. Through continuous pouring and extraction as the product is pulled out of the mould, the
product gets longer. A burner cuts the product whenever the desired product length is reached.
This technique is used for both ferrous and non-ferrous casting. The technique is used for the
casting of bars, slabs and sheets as a final step in iron, steel and non-ferrous metals production.
Its use in this context is described in the BREF for non-ferrous metal industries [155, European
IPPC Bureau, 2001] and the BREF on iron and steel production [211, European IPPC Bureau,
2000].

Chapter 2
Smitheries and Foundries Industry 87
Figure 2.48: Schematic representation of a thread casting machine used for continuous casting
[179, Hoppenstedt, 2002]

2.7 Finishing and post casting operations

Finishing of the raw castings encompasses all necessary treatments to yield a finished product.
According to the process, various steps may be required, such as:

- removal of the running system
- removal of residual moulding-sand from the surface and core remains in the casting cavities
- removal of pouring burrs
- repair of casting errors
- preparation of the casting for mechanical post-treatment, assembly, thermal treatment,
coating,

In some cases, foundries also perform assembly, surface finishing and coating of the castings.
However, these activities are not discussed in this document. Surface finishing and coating
techniques are discussed in the BREF documents on Surface Treatment of Metals (STM BREF)
and on Surface Treatment Using Solvents (STS BREF).
[110, Vito, 2001], [225, TWG, 2003]

Chapter 2
88 Smitheries and Foundries Industry
2.7.1 Removal of the running system

In the finishing of the raw castings and to remove the running system (see Figure 2.49), the
following operations take place:
- Beating, pressing: In the case of brittle materials such as grey cast iron castings and white
malleable cast iron, chamfers and feeders can generally be knocked off. Hydraulic
equipment is increasingly being used for this task
- Grinding with grinding wheels: These can be handheld, semi-automatic or automatic
- Cutting: For the removal of massive pieces in carbon steel or low alloy steel, an oxygen-
acetylene cutter is used. For cast iron or high alloy steel, oxygen-acetylene-powder or
oxygen-LPG-powder cutters are used
- Sawing: Materials which are sensitive to heat, such as aluminium alloys, are usually sawed.

Figure 2.49: Casting with running and gating system
[237, HUT, 2003]

Through good design of the connection points, the feeding system may even get brokenoff
during shake-out. This is mainly possible for grey iron.
[32, CAEF, 1997], [202, TWG, 2002]

2.7.2 Sand removal

Removal of the sand is performed in blasting cabins. The blasting medium is adapted to the
material to be treated and varies from blasting grit to glass beads. The cleaning of model plates
and permanent moulds is done with glass beads, aluminium beads or CO
2
ice grains. An
example of castings before and after blasting is shown in Figure 2.50.

Figure 2.50: Castings before (l., m.) and after (r.) sand removal by blasting
[237, HUT, 2003]

Various blasting techniques exist. The acceleration of the grit is done with compressed air or
turbine blades. The treatment is carried out in a closed room, with rubber-sealed doors.
Chapter 2
Smitheries and Foundries Industry 89
The castings are suspended on a monorail and move batch wise through the blasting cabin. For
smaller pieces, a specific moving belt is used. Big pieces are blasted manually, using a lance in
an enclosed cabin. In this case, personal safety measures are very important. A helmet with a
dust mask and respiratory equipment are required.

The coarse dust (sand and metal flakes) that is generated by blasting of the workpiece, is
collected together with the grit. It is dedusted, magnetically separated and sieved. The fine
fraction is removed from the exhaust air together with the coarse fraction, using a bag filter.
Cleaning of the grit before internal re-use is of major importance as the presence of sand could
cause a quick wearing of the throwing shovels.

2.7.3 Removal of burrs

Burrs, which occur at the point where mould and core pieces join, at veins and at other surface
irregularities, are removed using grinding wheels and grinding stones. Grinding wheels are
handheld, whilst with grinding stones the casting is pressed against a rotating stone.

Other applied techniques include:
- Slide grinding: Grinding burrs and small amounts of other excess material on the casting
surface can be removed without hand grinding. The parts are revolved in drums or vibrating
containers together with abrasive shapes, causing them to be rubbed against each other and
against the abrasive wheels. In a typical unit the castings are ground using a bed of
pyramidal grinding stones, together with the addition of a water-soap emulsion. The
roughness and the size of the grinding stones vary depending on the size of the castings
- Tumbling: This technique, also called blast removal, is used to remove thin burrs or small
amounts of casting residue. During this process the burrs remove themselves in an airless
blast cleaning process as a result of the parts being impacted together in a rotating drum. At
the same time the edges are rounded-off. Sometimes the process is assisted by a liquid.

Figure 2.51: Burrs (l.) and their removal using a grinding stone (m.) and slide grinding (r.)
[237, HUT, 2003]

Automation of these operations is difficult due to the variability in shape of the burrs and the
need to fix the casting easily and quickly. Nevertheless, automatic grinding machines are
increasingly being used in serial production. Raw casting parts are inserted into such machines
to produce suitably worked parts which do not need any retooling or further manual grinding
work.

Furthermore the following techniques are applied in automated lines:
- Punching: Due to the use of cutting and forming techniques, serial casting parts are often
designed in such a way that the unavoidable burr occurs in predetermined amounts which
are easily accessible as far as processing is concerned. If the series is large enough,
dedicated punching tools can be designed to quickly remove the burr and to provide the
casting with a uniform contour
Chapter 2
90 Smitheries and Foundries Industry
- Milling: With the development of electronic controls for processing machines it has become
increasingly easier to compile programs with which to process individual workpieces. Thus
it is possible to use milling machines for smaller series instead of the punching machines
with their single purpose tools. During such a process, the workpieces are taken up by the
device and passed across a number of different milling machines.

Finally, welding may be carried out in order to join castings, as well as to manufacture and
repair casting flaws. In most of these cases, arc welding is used. Depending on the requirements
and equipment, the work is carried out either with hand operated stick electrodes or with
welding wire and with or without inert gas. Scarfing is used on steel castings, which consists in
applying grooves for stress relieve. This uses a cutting torch with a copper-coated carbon
electrode.
[32, CAEF, 1997], [110, Vito, 2001], [202, TWG, 2002]

2.8 Heat treatment

2.8.1 Introduction

For ferrous castings, there are basically two types of thermal treatments which can be applied:
annealing and hardening. In the case of annealing, the tension caused in the workpiece as a
result of pouring, and the subsequent cooling down of the casting, is reduced and the structure is
evened out. In the case of hardening, the level of heat is raised above the transformation
temperature and the workpiece is subsequently rapidly cooled, in a process known as
quenching. This causes the material properties to change. Different results may be obtained by
the use of water, oil or air quenching. Quench and temper is the name of the procedure in
which the workpiece is reheated to the tempering temperature following quenching, and the
quenching procedure is repeated.

The malleableising of iron is a thermal treatment which differs from annealing and hardening.
Here, the malleable unfinished casting is either transformed to a white or black temperature
casting depending on the procedure.

Many non-ferrous castings are used in an as-cast condition, but certain applications require
higher mechanical properties than as-cast material. Possible treatments are annealing,
controlled cooling, solution heat treatment, artificial ageing and precipitation treatment.
[32, CAEF, 1997], [175, Brown, 1999]

2.8.2 Heat treatment furnaces

2.8.2.1 Chamber furnaces

Chamber furnaces are the most common furnace construction. Their actual design has been
adapted in many sub-forms in order to meet the needs of the differing thermal treatments of
various types of casting and production. Some examples of chamber furnaces are bogie hearth
furnaces, top hat furnaces, hub open-hearth furnaces, etc. Chamber furnaces with continually
running conveyance means are called tunnel furnaces. Heating is provided either electrically, or
with gas or fuel oil.

For non-ferrous materials, some heat treatments are carried out close to the melting point of the
casting, so accurate temperature control is needed. Forced air circulation is used to ensure that
the temperature at all parts of the furnace is constant and equal.
[32, CAEF, 1997], [175, Brown, 1999]

Chapter 2
Smitheries and Foundries Industry 91
2.8.2.2 Shaft furnaces

Pipes, long waves and similar parts are often treated while suspended in a vertical position in
shaft furnaces. In shaft furnaces, the heating is provided by electricity, gas or fuel oil.

2.8.2.3 Annealing furnaces

Chamber, top hat or tunnel furnaces are used to temper unfinished castings. The heating of such
furnaces is carried out via electricity, gas or fuel oil.

2.8.3 Quenching

In thermal treatment processes, chilling is the cooling down of a workpiece at a greater speed
than in calm air. This can be achieved by means of a rapid submersion in water or oil (see
Figure 2.52), as well as by forced air-cooling. Care should be taken that the workpieces are
cooled at a uniform rate. When submersing in fluids, either the parts must be moved or the fluid
must be continually circulated to ensure a full and even cooling of all parts of the workpiece. In
a similar principle, in air quenching the blowing procedure must be arranged so that the air
covers the entire surface.

Figure 2.52: Quenching of a hot casting shortly after heat treatment
[237, HUT, 2003]

2.8.4 Heat treatment of ductile iron (SG iron)

It is obviously desirable to achieve the required metal properties in the as-cast form to save
further treatments being necessary, but this is not always possible because of variations in
section thickness, etc. The heat treatment of the castings can eliminate carbides in thin sections,
produce more consistent matrix structures and for a given structure, often improve its
Chapter 2
92 Smitheries and Foundries Industry
mechanical properties, especially by normalising the grain structure. Where tempered martensite
structures are needed, heat treatment is essential.
[32, CAEF, 1997], [174, Brown, 2000]

2.8.4.1 Stress relief

Stress relief consists of heating the castings at a rate of 50 100 C/h to 600 C (taking care not
to exceed 610 C), followed by soaking them for a minimum of one hour, plus an extra hour for
every 25 mm of section thickness in the thickest section, and then cooling them at a rate of
50 - 100 C/h or less. The castings must be adequately supported in the furnace so that they are
not subjected to stress.

2.8.4.2 Breakdown of carbides

Thin section castings may contain carbides in the as-cast structure. These can be eliminated by
soaking the castings at 900 925 C for 3 to 5 hours.

2.8.4.3 Annealing to produce a ferritic matrix

Annealing involves soaking the castings at 900 925 C for 3 5 hours, followed by a slow
cooling at around 20 35 C/h through the critical temperature range (about 800 710 C), and
finally furnace cooling at, say 50 100 C/h to 200 C.

2.8.4.4 Normalising to produce a pearlitic matrix

For normalising, a soaking of the castings above the critical temperature followed by air cooling
is necessary. Again a soaking temperature of 900 925 C is usually used, to ensure that the
carbides are broken down. Forced air cooling is used to form pearlite. The type of heat treatment
furnace available and the size of the load determines the process cycle that is possible. It may be
necessary to adjust the metal composition with tin or copper to help the formation of fully
pearlitic structures.

2.8.4.5 Producing hardened and tempered structures

Hardened structures are produced by austenitising the casting at 900 920 C, followed by oil
quenching. Tempering is usually carried out at 600 650 C.

2.8.4.6 Austempered ductile iron (ADI)

Austempering is an isothermal heat treatment for producing ausferrite structures. It can double
the strength of ductile iron whilst allowing it to retain good ductility and toughness. Wear
resistance and fatigue properties are excellent, even to such a degree that ADI is comparable
with wrought and hardened-tempered steel.

The ADI heat treatment is a two-stage process, as shown in Figure 2.53. Austenitising is carried
out at 815 930 C to fully transform the matrix to austenite. This is done either in a non-
oxidising atmosphere furnace or in a high temperature salt bath. Temperatures and times are
determined by the chemical composition, the section size and the grade of ADI required; 1 to
1.5 hours is usually adequate. Slow initial heating of the casting is desirable to avoid the risk of
cracking the complex shapes. The castings are then quenched to the required isothermal heat
treatment temperature, usually between 210 and 400 C. This is usually done in a salt bath. The
Chapter 2
Smitheries and Foundries Industry 93
castings are held at this temperature for 1 2 hours to complete the transformation of
austenite to ausferrite. Lower temperatures give high hardness, strength and wear resistance,
while higher heat temperatures result in higher ductility and toughness. After the isothermal
treatment, the castings are cooled to ambient temperature.

Figure 2.53: Typical austempering heat treatment stages
[174, Brown, 2000]

Unalloyed ductile irons may be austempered in sections of up to about 8 mm thickness. Thicker
section castings require the addition of Mo or Ni to increase the hardenability.

Austempered ductile iron is used as a replacement for forged steel components in the
agricultural, railway, automotive and general engineering industries; for example, for plough
tips, digger teeth, spring brackets, rear axle brackets, gears, etc. ADI production is growing but
its use is limited to some extent by the lack of suitable heat-treatment facilities.
[32, CAEF, 1997], [174, Brown, 2000]

2.8.5 Heat treatment of steel

Steel castings are normally subject to a heat treatment before delivery, e.g. normalising, with
this heat treatment leading to a structural change. Additionally, potential casting stresses have to
be reduced (stress relief annealing). Many steel casting qualities have to be additionally
tempered after stress relief anneal (hardening and tempering).

Most steel castings are subjected to a heat treatment to attain the desired mechanical properties
and to a stress relief, to obtain the right corrosion resistance and to avoid the difficulties during
the finishing operations. The heat treatment is defined according to the steel grade. To remove
the chemical and structural segregations, annealing at high temperature is generaly performed.
Carbon and low alloy steels undergo:
- normalising and air cooling, or
- austenitising, quenching and tempering.

Austenitic or duplex stanless steels undergo a solution annealing and water quenching heat-
treatment. Additionnal heat-treatments such as for stress relief or post weld treatment are also
used to eliminate the internal stresses of the material.
[32, CAEF, 1997], [202, TWG, 2002], [225, TWG, 2003]
Chapter 2
94 Smitheries and Foundries Industry
2.8.6 Heat treatment of aluminium

Aluminium castings are heat treated for: homogenisation, stress relief, improved dimensional
stability and machinability, optimised strength, ductility, toughness and corrosion resistance.
Most often the heat treatment is a compromise between varying effects, maximising the
important properties of one often at the expense of others. The heat treatment of aluminium may
involve: annealing, solution heat treatment, quenching, artificial ageing and precipitation
treatment. The type of treatment applied is indicated by a suffix to the alloy designation. This is
called the temper designation. For sand, gravity and low-pressure die-castings all treatments are
possible, though not all are standardised. Pressure die-castings are not solution treated and
quenched in the same way as sand, gravity and low-pressure die-castings. Entrapped gas
bubbles can expand and cause casting defects. Heat treatment is not a common step for high-
pressure die-castings; only about 1 % of this type of castings are subjected to heat treatment. All
die-castings may be quenched from the die, precipitation treated and stress relieved without
suffering any harmful effects. In the low pressure die casting production of aluminium wheels
90 % of the castings are subjected to heat treatment
[225, TWG, 2003].

2.8.6.1 Stress relieving and annealing

Castings with changes of section, or of a complex shape are likely to develop internal stresses.
These can then cause dimensional changes upon machining. To stabilise the casting and to
remove internal stresses, castings are heated to a temperature of 200 C for 5 hours, followed by
slow cooling in the furnace.

2.8.6.2 Solution treatment and quenching

Castings are heated at temperatures just below the melting temperature and held there for a long
time to take the alloying constituents into a homogeneous solid solution. The castings are then
rapidly cooled by quenching to room temperature to retain the elements in solution. Water or
special quenchants are used. The quench tanks are placed close to the furnace to ensure that
rapid cooling is possible. Although sensitivity to the quench interval differs between alloys,
good practice should limit the interval to 5 10 s.

2.8.6.3 Precipitation treatment

The controlled precipitation of alloying constituents is promoted by heating the casting to a
temperature of between 150 C and 200 C. The casting strength and hardness are increased.
The process is therefore also referred to as structural hardening. Each alloy has an optimal heat
treatment cycle.

2.8.6.4 Artificial ageing

Some casting alloys increase in strength and hardness while left standing at room temperature.
The process can take several weeks but can be speeded up by heating above room temperature
and then sustaining this heat over time.
[175, Brown, 1999], [202, TWG, 2002], [213, CTIF and CQRDA, 2002], [212, Zalensas, 1993]

Chapter 2
Smitheries and Foundries Industry 95
2.9 Quality control

During quality control the finished casting is checked for compliance with the product
requirements concerning e.g. dimensions, metal structure defects, surface structure. Depending
on the type of casting and the size of the series, quality control can be achieved by visual
inspection, with the use of measurement tools, or automatically.

In the production of aluminium wheels the castings are controlled via X-ray analysis. A
standard image of a good casting is used for verification by a computer program. If there are any
differences, the image is studied and assessed by a human operator. Random tests of the alloy
are analysed with spectral analysis.

The quality control procedure informs the final decision to reject a casting or to allow its further
shipment to the market. Rejected castings are brought back to the raw material input to be
remelted.
[225, TWG, 2003]

Chapter 3
Smitheries and Foundries Industry 97
3 CURRENT EMISSION AND CONSUMPTION LEVELS IN
FOUNDRIES

3.1 Mass stream overview

3.1.1 Introduction

The general mass stream overview for the foundry process is given in Figure 3.1. This scheme
generally applies for ferrous and non-ferrous foundries. Specific aspects of the various process
steps and types will be worked out below.

Foundry Process
Melting - Casting - Finishing
Metal
scrap/ingots
Castings
Energy
- fuel
- electricity
- heat
Water
- cooling
- scrubbing
Sand and Chemicals
- additives melting and metal treatment
- binders/release agents
- oils and lubricants
Noise
Water
- organics
- metal dust
Solid
- dust
- used sand
- dross
- slag
Energy
- heat
- steam
- hot water
Air
- combustion and reaction products
- particulate matter
- pyrolysis and evaporation products
Odour
Figure 3.1: Mass stream overview for the foundry process

3.2 Melting and the metal treatment of ferrous metals

3.2.1 Properties of melting furnaces for steel and cast iron

Table 3.1 presents an overview of the typical properties of melting furnaces used for steel and
cast iron. The data is then discussed in the sections below.

Chapter 3
98 Smitheries and Foundries Industry
Process Steel Cast Iron
Type ELECTRIC ARC FURNACE INDUCTION F. INDUCTION F. CUPOLA FURNACE ROTARY F.
Subtype Acid lined Basic lined Coreless Coreless Cold blast Hot blast Hot blast
long campaign
Cokeless - duplex
Energy source Electricity Electricity Electricity Electricity Coke Coke Coke Gas/fuel Gas/fuel
Thermal efficiency
1
(%) 60 70 60 70 50 60 50 60 30 40 40 45 35 45 50 60 50 60
Primary thermal
efficiency
2
(%)
21 25 21 25 15 20 15 20 30 40 40 45 35 45 45 50 35 45
3
kWh/tonne metal charge 500 700 500 700 520 800 520 800 950 1200 800 900 810 1100 700 800 600 800
Batch/continuous Batch Batch Batch Batch Continuous Continuous Continuous Continuous Batch
Production rate
4
(tonnes/h)
2 - 10 8 - 70 8 - 70 >5
Furnace capacity
5
(tonnes)
2 50 2 50 0.01 30 0.01 30 1 20
Meltdown time (h) 1 4 1 4 1 2 1 2 2 4
Refining ability Possible Possible No No Yes Yes Yes No No
Capital cost High High High High Medium High High Medium Low
Slag production
(kg/tonne metal charge)
10 40 20 80 10 20 40 80 40 80 40 80 20 60
Dust production
6
(kg/tonne metal charge)
5 8 0.06 1 5 13 4 12 0.8 0.3 2.9
Waste gas
7
emission
(kg/tonne metal charge)
CO
2
8
Depending on power generation Depending on power generation 400 500 350 480 100 120 120
CO 7.5 25 (decarburisation) n.a. Possible
9
0.5 2.5 <10 1.0 1.5
SO
2
<1 Minor 1 2 <1 Fuel dependent 2.5 3.0
NO
x
n.a. n.a. <1 <1 0.5 0.3 0.4
1 Indicated values give an order of magnitude but largely depend on exploitation conditions, such as metal temperature, furnace capacity and production rate
2 Efficiency of electrical power generation assumed to be 35 %
3 Taking into account energy consumption for oxygen production and the raw materials, such as graphite and FeSi, to substitute the oxidised elements during melting
4 For continuous processes only
5 For batch processes only
6 Indicated values are general values found in literature
7 Indicated values are general values found in literature
8 Assuming complete combustion
9 Depending on local exploitation conditions and construction.
Table 3.1: Typical melting furnace properties and emission data
[32, CAEF, 1997], [110, Vito, 2001], [202, TWG, 2002], [225, TWG, 2003] and comments fromCTIF
Chapter 3
Smitheries and Foundries Industry 99
3.2.2 Cupola furnaces

Input Output
ferrous material (iron pigs, sponge iron, steel
scrap, foundry returns )
alloying metal (ferro-alloys )
flux (limestone)
energy (coke, gas, oil, electricity)
oxygen
cooling water
water
metal alloy (cast iron)
dust (metal content)
CO/CO
2,
SO
2
, NO
x
HF
Dioxins, furans
organic pollutants
slag
waste refractory

3.2.2.1 Coke and energy consumption

In cold blast operation, the coke consumption between the charges is generally
90 - 120 kg/tonne metal charge, but can be less than 70 kg/tonne metal charge, e.g. in the case
of counterweights. Accounting for the amount of coke in the bed gives a total coke consumption
of 110 140 kg/tonne metal charge. As the calorific value of European cokes is 8.5 kWh/kg,
this corresponds to a calorific input of 950 1200 kWh/tonne metal charge.

The total coke ratio in a hot blast cupola is generally 110 145 kg/tonne metal charge.
However, as the average steel percentage is 50 %, and the recarburisation consumes about
1.5 %, the real burned coke ratio is 95 130 kg/tonne metal charge, which is 810 to
1100 kWh/tonne metal charge. This corresponds to a thermal efficiency of 35 to 45 %.

Depending on the plant layout, the energy consumed by the fume treatment equipment and the
holding furnace, as given in Table 3.2, must be added. German data indicate a specific
electricity use for the flue-gas cleaning equipment of c. 20 kWh per tonne of good casting. [202,
TWG, 2002]

Energy carrier type Average consumption
kWh/tonne metal charge
Gas for the combustion chamber 40
Electricity for the flue-gas cleaning equipment
(fans, etc.)
40
Electricity for the holding furnace 60
Table 3.2: Average energy consumption for off-gas treatment and holding

3.2.2.2 Particulate matter

The emission range of particulate matter is very wide. Particulate emissions depend primarily
on the cupola type used, as shown in Table 3.3:

Cupola type Dust emission
(kg/tonne metal
charge)
Coke proportion
(kg/tonne metal
charge)
Cold blast 5 13 110 140
Hot blast 4 10 95 130
Liningless hot blast 5 12 115 135
Cokeless 0.8 0
Table 3.3: Dust emission levels (raw cupola off-gas) from various cupola types
[32, CAEF, 1997]

Chapter 3
100 Smitheries and Foundries Industry
Dust emission levels measured at the stack for three German furnaces are given in Table 3.4.

Flue-gas cleaning Volume
(m/h)
Total dust
(mg/m)
PM
10

(%)
PM
2.5

(%)
Above charge hole off-take; bag filter 28500 7 88 47
Cyclone, Venturi, recuperator 16000 68 94 96 88
n.d 6000
(Nm/h dry)
75 100 45 85
Table 3.4: Cupola dust emission levels and PM size distribution
[202, TWG, 2002]

In general, particle sizes range from less than 1 am up to 10 mm, with 50 % less than 100am.
However, 5 to 20 % are smaller than 2 am, which makes the dust collection more difficult.
Cupola dust is primarily made up of coke, silica, rust and limestone, as shown in Table 3.5.

Substance Composition (%)
[32, CAEF, 1997] [158, Charbonnier, et al., 1998]
Iron oxide 30 60 15 25
SiO
2 25 15 30
Coke dust 3 15 n.d
MnO 3 10 2 5
Al
2
O
3
1 3 2 5
MgO 1 3 0 2
CaO <1 5 10
S <2 n.d
ZnO, depending on the charge <3 0 30 Zn*
PbO, depending on the charge <1 0 5 Pb*
* In the form of oxides and silicates; applicable for Zn-enriched dusts
Table 3.5: Typical composition of cupola dust, data in weight per cent
[32, CAEF, 1997] and [158, Charbonnier, et al., 1998]

3.2.2.3 Waste gases

Coke fired cupola gas is composed primarily of N
2
, CO
2
, H
2
O, and CO, with smaller amounts of
SO
2
. On conventional cupolas, where the off-gas is collected above the charging door, a
distinction has to be made between the condition of the flue-gases below and above the charging
door, since ambient air is entrained through the open charge door. This input significantly
changes the total airflow.

If the cupola gases are hot enough, and if there is enough CO present, the gases may burn
spontaneously together with drawn-in air (CO + O
2
2CO
2
) and temperatures may rise to
900 C. Little or no CO will then be left in the exhaust. If no combustion takes place, the air
intake will result in a cooling effect, of between 100 and 300 C, and the CO/CO
2
balance will
remain unchanged. The temperature of the gases just below the charging door is primarily
dependent on the charge height; the input of ambient air is determined by the fan capacity or the
natural draught available.

The flow of the undiluted cupola top gases is proportional to the coke consumption. Increasing
the coke proportion in the charge will decrease the production rate (tonne molten metal/h) if the
same blast airflow is maintained. It may then be necessary to increase the blast to maintain the
production. The metal temperature will also increase. With reference to the combustion rate
(C + O
2
CO
2
), more coke and blast air will result in an increased flow of exhaust gases.

Chapter 3
Smitheries and Foundries Industry 101
For a given furnace, the coke and blast air consumption depend on the targeted melting rate and
the metal temperature, which can vary on an hourly basis. Typical flowrates reported in
literature vary from 600 to 800 Nm
3
/tonne metal charge for cold blast cupolas and 500 to 700
Nm
3
/tonne metal charge for hot blast cupolas. The composition of the unburned top gases may
be as follows (Table 3.6):

Substance Volume %
CO
2
10 18
CO 5 15
H
2
<1
SO
2
<0.05
N
2
Remainder
Table 3.6: Unburned top gas composition for cupola furnace
[32, CAEF, 1997]

Above the charging door, the flowrates of the exhaust gases may be two to five times higher,
depending on the target temperature (which depends on the kind of dust arrestment system
applied) of the diluted gases and, in the case of a recuperative hot blast cupola, the presence of a
post combustion chamber. Typical flowrate values are 3000 to 4000 Nm
3
/tonne metal charge
melted for cold blast cupolas and 900 to 1400 Nm
3
/tonne metal charge when post-combustion is
performed. The composition of the gases is determined by the rate of dilution (natural draught
or fan power), the degree of spontaneous combustion of the CO, or the post combustion itself,
which can be executed on the complete or partial flow.[32, CAEF, 1997]

Emission data for the main combustion gas components are given in Table 3.7. Hot blast and
cold blast cupolas with varying melting capacities are listed. The table also indicates whether
the off-gas is collected under or above the charging door. The hot blast systems show reduced
SO
2
emission levels. The disintegrator shows a better dedusting performance than the venturi
scrubber. The application of post combustion clearly affects both the CO and NO
x
levels, under
optimal working conditions. Analysis of the full measurement campaign data shows that sub-
optimal operation of a hot blast cupola results in increased CO-emissions: 2000 mg/Nm,
compared to 5 20 mg/Nm in optimised conditions. VOC and PAH emissions rise accordingly.
[110, Vito, 2001]

The use of dry dedusting (i.e. using a bag filter) results in dust emission levels 20 mg/Nm.
Inventory data from Italian foundries showed a dust level below 30 mg/Nm (ranging from
0.1 to 32 mg/Nm) when a bag filter was applied. Wet scrubbing gives dust emissions up to
80 mg/Nm (ranging from 5.4 to 78 mg/Nm) [180, Assofond, 2002]. A value of 68 94 mg/m
was reported for a German cupola with wet scrubbing (see Table 3.4).

Chapter 3
102 Smitheries and Foundries Industry
Cupola Capa-
city
Exhaust Flow FGC
equipment
Dust SO
2
CO NO
x
HF O
2
type tonne/h type m/h mg/Nm mg/Nm mg/Nm mg/Nm mg/Nm vol %
Hot blast
1)
19 UC 29000 Venturi
1)
41 21 17798 21 n.d 11
Hot blast 20 UC 40000 Disintegrator 5 57 712 11 n.d 11
Hot blast 24 UC 46445 Bag filter 1.1
1.4
20 14 17 70 75 7 11
Hot blast 25 UC 35000 Venturi 36 28 21 16 n.d 11
Hot blast 60 UC 75000 Disintegrator 5 58 9 7 n.d 11
Cold blast 3.2 AC 12000 Bag filter 10 401 5084 16 1 11
Cold blast 5 AC 23000 Bag filter 6 434 28558 63 n.d 11
Cold blast 8 AC 20000 Bag filter 20 401 936 36 n.d 11
Cold blast 9 UC 22000 Bag filter 4 105 17286 60 n.d 11
1) Working with an old recuperator
UC: Under charge gas collection; AC: above charge gas collection
All data are continuous monitored daily average values
Table 3.7: Emission data for hot blast and cold blast cupolas working with different set-ups
[29, Batz, 1986], [202, TWG, 2002], data recalculated to 11 % O
2
3.2.2.4 Cupola slag

Slags contain oxides that float upon the melt and which arise from impurities in the feed
material, wearing of the furnace refractory and from the ash of coke and from the melting loss
of the metal charge [225, TWG, 2003]. Slags are bonded through the addition of binders, such
as vermiculite. The typical composition of cupola slag is given in Table 3.8. A cupola furnace
typically produces 40 80 kg slag per tonne of liquid iron.

Compound %
SiO
2
45 55
CaO 25 40
Al
2
O
3
8 20
MgO 1 3
MnO 1 4
FeO 1 6
Sulphides <1
TiO
2
<1
ZnO <0.1
Table 3.8: Typical cupola slag composition
[172, Neumann, 1994], [156, Godinot, 2001]

Cupola slag consists of 30 % refractory material, 10 % sand (from internal scrap), 40 % CaO
(flux), 10 % cokes ashes, and 10 % burn-off material.

An important aspect of cupola slag is its high SiO
2
-content. After quenching, the slag has a
vitrified structure. This generates an inert non-leaching material.

3.2.2.5 Waste refractory

The cupola furnace has the specific feature that the lining material (quartz-clay mixture) of the
furnace at the melting zone only lasts for one melting campaign. The main part of the refractory
is converted into slag. The amount to be taken out and disposed of as waste is much smaller
than the applied amount.

Chapter 3
Smitheries and Foundries Industry 103
3.2.3 Electric arc furnace

Input Output
ferrous material (steel scrap, foundry returns
swarf, pig iron )
alloying metal (ferro-alloys )
flux (limestone )
energy (electricity, gas, oil)
oxygen
electrodes
metal alloy (cast steel)
dust (metal content, refractory)
NO
x
, CO
2
, CO
organic air pollutants, HC
metal oxide fumes
slag (CaO, SiO
2
, MgO)
waste refractories
3.2.3.1 Input

500 600 kWh of electricity is used to melt and raise one tonne of steel to its casting
temperature. Furnaces are normally rated at 500 kVA per tonne, giving a melting time of about
1.5 hours.

The electrodes are made of graphite and are consumed during operation through oxidation,
volatilisation and breakage, so they must be replaced as necessary. A three tonne furnace
typically uses electrodes of 200 mm diameter. Electrode consumption is a significant factor in
the cost of arc melting and figures vary from 3 to 10 kg/tonne of steel melted depending on the
type of steel produced and the practice used. [174, Brown, 2000]

3.2.3.2 Particulate matter

Particulate emission rates reported in literature vary from 2 to 20 kg per tonne of iron charged
with an average of 5 to 8 kg per tonne. The highest emission rates are recorded at the beginning
of the melting cycle, during the decarburisation treatment and during back charging [173,
Huelsen, 1985]. Particle sizes range from less than 1 am up to 100 am, with 50 % being less
than 3 to 5 am. During oxygen treatment, one analysis showed almost 90 % of the particles
were smaller than 5 am.

In terms of the chemical composition of the particulate matter, an even bigger range of
measured data can be observed. Table 3.9 gives an overview of the reported figures in the
literature.

Substance Weight per cent (%)
FeO + Fe
2
O
3
30 60
SiO
2
5 35
CaO 1 15
MgO 0 15
ZnO 0 16
Cr
2
O
3
0 8
MnO 2 10
Al
2
O
3
0 5
MoO
3
<1
NiO <1
Pb <1
Cd <0.01
TiO
2
<0.05
V
2
O
5
<0.05
Ignition loss 0 4
Table 3.9: Chemical composition of EAF dust from steel foundries
[32, CAEF, 1997]

Chapter 3
104 Smitheries and Foundries Industry
The composition of the furnace dust primarily depends on the steel grade being produced. For
instance, low alloy steels will not generate emissions containing chromium or nickel, whereas
stainless steel will. Another important factor is scrap quality. Melting galvanised steel scrap
leads to significant emissions of zinc oxide.

The application of an appropriate hooding of the furnace allows the capture of up to 98 % of the
furnace dust. The captured exhaust gas is then cleaned, usually using a bag filter. This reduces
dust emission levels to below 10 mg/Nm.
[29, Batz, 1986]

3.2.3.3 Visible fumes

Visible fumes are emitted during charging of the hot furnace and at the beginning of the melting
cycle. The available literature does not report the quantities and the composition of these
secondary emissions. Again, the nature of the fumes depends on the cleanliness of the charged
material with respect to its oil, grease, paint or other organic matter content.

3.2.3.4 Waste gases

Electric arc furnaces are mainly used for steel melting. They primarily generate dust emissions
and gaseous compounds, such as nitrogen oxides, carbon monoxide and organic compounds.
The organic compounds depend on the type and amount of impurities in the feed metal. The raw
materials used as charge materials in the foundry are selected so that they don't usually contain
materials that can produce dioxins. Organic compounds are decomposed into harmless products,
so long as no scrap preheating without post-combustion is applied. [29, Batz, 1986]

Emission data for the main combustion gas components are given in Table 3.10. Inventory data
from Italian foundries showed a dust level below 10 mg/Nm (ranging from 1.2 8.3 mg/Nm)
when a bag filter is applied. Wet scrubbing gives dust emissions below 25 mg/Nm (ranging
from 12 24.5 mg/Nm). [180, Assofond, 2002]

Capacity Off-gas Flow FGC Dust SO
2
CO NO
x
HF O
2
tonne/
charge
collection m/h equipment mg/Nm mg/Nm mg/Nm mg/Nm mg/Nm vol %
EAF 50 FH 265000 Bag filter 2 n.d n.d 50 n.d 20
EAF 2 x 50 FH 380000 Bag filter 4 n.d n.d n.d n.d 20
EAF 10 PH 160000 Bag filter 1 1 200 5 0.1 20
FH: Full hooding; PH: Partial hooding
Table 3.10: Typical emission values for EAF furnaces
[29, Batz, 1986]

During melting and refining, CO is generated from the oxidation of the graphite electrodes and
the carbon from the metal bath. Estimated quantities are 6 to 20 Nm
3
CO/tonne (or 7.5 to 25 kg
CO/tonne) depending on the initial carbon content of the charged material and the required
carbon level after treatment. The oxygen injection causes quantities of iron oxide (red smoke) to
be emitted from the metal bath. No other significant emissions have been reported. [32, CAEF,
1997]

Chapter 3
Smitheries and Foundries Industry 105
3.2.3.5 Slags

The chemical composition of EAF slags are given in Table 3.11. The reported values were
based on the analyses of 3 samples.

Compound Average
(%)
Range
(%)
SiO
2
36.2 28.6 41.8
CaO 12.4 7.2 17.7
MgO 22.1 18.3 27.0
Al
2
O
3
8.4 7.4 0.1
FeO 0.7 0.5 1.0
MnO 14.8 4.0 29.6
TiO
2
1.2 0.39 2.7
Na
2
O 0.3 0.11 0.57
K
2
O 0.1 0.1 0.23
Table 3.11: Chemical composition of EAF slag
[171, The Castings Development Centre, 1999]

3.2.4 Induction furnace

Input Output
ferrous material (iron pigs, steel scrap, swarf,
foundry returns )
alloying metal (ferro-alloys )
carburising agents, flux
energy (electric)
cooling water
metal alloy (cast iron, cast steel)
dust
organic and metallic fumes
CO
slag
refractory waste

3.2.4.1 Coreless induction furnace

3.2.4.1.1 Energy input

A coreless induction furnace can melt a tonne of iron and raise the temperature of the liquid
metal to 1450 C using under 600 kWh of electricity. However, in practice, only a few foundries
can achieve this level of specific consumption on a week-by-week basis. The actual energy
consumption varies according to the size and working regime of the furnace. Large furnaces,
working 24 h a day and using a molten heel can achieve a 600 kWh/tonne value. Surveys of
foundries show that consumptions of 520 800 kWh/tonne metal charge are common, the
variation being due to individual melting practice, such as the rate at which the pouring line will
accept the molten metal and whether furnace lids are used effectively. Attention to energy
saving measures should allow figures of 550 650 kWh/tonne metal charge to be achieved. The
typical heat losses from a coreless induction furnace are shown in Figure 3.2.
[47, ETSU, 1992], [174, Brown, 2000], [202, TWG, 2002]

Chapter 3
106 Smitheries and Foundries Industry
Figure 3.2: Typical energy losses from a mains frequency coreless induction furnace
[47, ETSU, 1992]

3.2.4.1.2 Particulate matter

Emission rates of 0.06 to 1 kg/tonne metal charge have been reported in literature, but at
present, emission rates of 0.04 to 3 kg/tonne are normal. The highest emission rates occur
during charging and at the beginning of the melting cycle. Particle sizes range from 1 100 am,
with more than 50 % being smaller than 10 20 am. [32, CAEF, 1997], [202, TWG, 2002]

Dust emission levels and PM distribution from a German IF foundry are given in Table 3.12.

Flue-gas cleaning Volume
(m/h)
Total dust
(mg/m)
PM
10

(%)
PM
2.5

(%)
Hood and bag filter 10400 c. 0.4 78 50
Table 3.12: IF dust emission level and PM size distribution
[202, TWG, 2002]

As for the chemical composition of the particulate matter, there is no published data regarding
the melting of steel, but there is reason to believe that it is close to that of the composition of the
dust emitted during the melting of cast iron as shown in Table 3.13. The presence of Zn, Pb or
Cd will lead to metallic fumes during the meltdown of the charge.

Chapter 3
Smitheries and Foundries Industry 107
Substance Weight per cent
(%)
FeO + Fe
2
O
3
30 70
SiO
2
(depending on the lining material)
5 25
MnO <5
Al
2
O
3
(depending on the lining material)
3 10
CaO <1
ZnO
(depending on the charge material)
<5
Metallic oxides
(depending on the charge material)
<0.1
Loss on ignition 0 10
Table 3.13: Chemical composition of induction furnace dust in cast iron foundries
[32, CAEF, 1997]

3.2.4.1.3 Waste gases

The melting of iron and steel in induction furnaces results in low emissions compared to the
cupola furnace. Emissions due to the combustion of fossil fuel are especially prevented. An
exhaust capture efficiency of up to 95 % is possible using special capture systems, such as side-
draughts, movable hoods and partial covering of the furnace. Filtration of the off-gases is
mainly performed using dry systems. Dust emission levels below 5 mg/Nm can be obtained
[225, TWG, 2003]. Typical emission data are given in Table 3.14.

Capacity Off-gas Flow FGC Dust SO
2
CO NO
x
HF O
2
No of furnaces x
tonne/charge
collection m/h equipment mg/m mg/m mg/m mg/m mg/m vol %
IF (2 x 10) + (3 x 3) Side-
draught
54000 Bag filter 5 n.d n.d n.d n.d 21
Table 3.14: Typical emission data for induction furnace melting in a ferrous foundry
[29, Batz, 1986]

Charging oily scrap or borings in a cold furnace will lead to the presence of organic vapours in
the exhaust gases, which will not be burned since they are created at the beginning of the
melting cycle. Adding this kind of scrap to a molten bath (i.e. the molten heel practice) can be
very dangerous since small explosions in the metal bath may occur, which can then lead to
splashing metal drops and sparks out of the furnace.

Since the scope of the work is limited to the foundry boundaries, no emission data for the
generation of electrical energy have been taken into account.

3.2.4.1.4 Slags

The typical properties of induction furnace slags are given in Table 3.15. Induction furnaces
produce 10 20 kg slag per tonne metal charge. The amount of slag produced depends on the
quality of charge material. The lower limit of the given range applies if internal scrap is cleaned
(blasted) before re-melting.

Chapter 3
108 Smitheries and Foundries Industry
Compound %
SiO
2
40 70
FeO 10 30
Al
2
O
3
2 15
MnO 2 10
CaO 0 3
MgO 0 3
Table 3.15: Typical composition of induction furnace slag
[172, Neumann, 1994]

3.2.4.2 Channel induction furnace

In ferrous metal foundries, the channel induction furnace is mainly used as a holding furnace. It
is the furnace of choice for duplex operation with the hot blast cupola. In this case, its function
is either to hold or to homogenise the chemical composition of the metal, or to serve as a
reservoir of the melted metal for the casting. The furnaces role is not to increase the metal
temperature, but rather to prevent unwanted cooling.

Figure 3.3 presents the energy consumptions of some representative channel induction furnaces.
The consumption depends on process related parameters, such as the holding time. The graph
shows that the consumption decreases with increasing annual tonnage. The extreme values are
80 and 20 kWh per tonne transferred.

Figure 3.3: Energy consumption (kWh/tonne) as a function of transferred tonnage and loss of
temperature for channel induction furnaces
[140, EU Thematic Network Foundry Wastes, 2001]

3.2.5 Rotary furnace

Input Output
ferrous material (iron pigs, steel scrap, swarf,
foundry returns )
alloying metal (ferro-alloys )
carburising agents, flux
energy (electric, oil, gas)
cooling water
metal alloy (cast iron)
dust
organic and metallic fumes
slag
refractory waste

Chapter 3
Smitheries and Foundries Industry 109
3.2.5.1 Input

As is the case with electric melting, the energy source used in the rotary furnace allows a clean
process, especially if natural gas or propane is used. The degree of contamination of the charged
scrap is again very important to the nature and the quantities of the emissions that arise. This
fact explains why no uniform emission pattern exists, and why the measurement data show wide
variations.

3.2.5.2 Particulate Matter

Particulate matter originates from dirt adhering to the charged material, lining wear during
charging and melting, and from the burn out of alloying elements and the different additions to
the charge. The reported data show overall emission rates of 0.3 to 2.9 kg/tonne metal charge.
The grain size is relatively small, from less than 1 am to 100 am, with 20 % <1am, 60 % less
than 10 amand 95 % smaller than 50 am. [32, CAEF, 1997], [204, Carnicer Alfonso, 2001]

The chemical composition of rotary furnace dust is given in Table 3.16.

Compound %
Iron oxides 50 75
MnO <1
SiO
2
<1
MgO 1 - 2
Cr
2
O
3
<0.5
ZnO <1
Pb <0.5
Sn 0.2
Loss on ignition 5 10
Table 3.16: Chemical composition of rotary furnace dust in ferrous melting
[204, Carnicer Alfonso, 2001]

3.2.5.3 Waste gases

CO
2
production is estimated to be 120 kg/tonne metal charge. The thermal efficiency of the
furnace is very high at 50 to 65 %
4
, depending on the capacity. This high yield is achieved by
using pure oxygen instead of air as a combustion medium. During the reducing flame conditions
the combustion of the fuel or gas may be incomplete. CO may be produced under these
conditions. However, at this point the exhaust gases leave the furnace at a temperature of
1500 C and CO burns spontaneously when in contact with the ambient air, which is used to
cool the gases.

If sulphur bearing fuel is used, SO
2
emissions can be significant. However, natural gas or
propane does not show any significant SO
2
emission.

NO
x
emissions of 50 to 250 ppm in the off-gases have been measured in a 5 tonne furnace. The
NO
x
originates from the oxidation of atmospheric nitrogen at high flame temperatures (2800 C)
if ambient air is leaking into the furnace through the burner door. The production of NO
x
is
significant during the oxidising flame regime but only low during the reducing regime. No
reports have been made concerning carbonaceous emissions, probably due to the high flame
temperature in the furnace and the relatively high cleanliness of the charged material. [32,
CAEF, 1997]

4
This is if the energy for oxygen production is not taken into account. With oxygen production, the efficiency should be 10 to
15 % less. Furthermore if the energy for the replacement of the burnt out charge elements (C, Si) is discounted the resulting
primary efficiency is only about 30 to 35 %.
Chapter 3
110 Smitheries and Foundries Industry
For ferrous rotary furnaces operating with simple air/fuel burners and an afterburner, dust
emissions peak at about 250 mg/m
3
for short periods (between 3 seconds continuously but also
intermittently over a period of a minute) during the solid phases of the melt cycle. Thereafter,
once the charge starts to become liquid, the dust emissions reduce to less than 30 mg/m
3
during
normal running. Emissions may remain as high as 150 200 mg/m
3
continually during the solid
phase of the melt. 40 % combustible emissions are common. Peak emissions from rotary
furnaces consist of at least 80 % unburned fuel and occur during charging operations when the
main furnace burner is extinguished and then re-lit. The mentioned emissions only apply if at all
times, including during charging, the fumes are ducted via the afterburner, which is kept fully
operational [163, UK Environment Agency, 2002]. The following raw gas dust levels have been
reported for an oxygas rotary furnace without an afterburner: an average dust load of 400 450
mg/Nm over the two distinct melting phases. These phases are a solid phase with dust emission
levels of 150 mg/Nm, and a liquid metal phase with peak levels up to 1500 mg/Nm at the start
of the rotation and 600 700 mg/Nm during normal operation. [204, Carnicer Alfonso, 2001]

Typical emission data are given in Table 3.17. These were collected from a 1.4 tonne/h cast iron
melting furnace, without any flue-gas cleaning equipment. Inventory data from Italian foundries
showed a dust level below 15 mg/Nm (ranging from 0.6 14.6 mg/Nm) if a bag filter is
applied [180, Assofond, 2002].

Parameter Measured value Parameter Measured value
Gas flow (Nm/h) 9000 Chlorine (mg/Nm) 0.01
Capacity (tonnes/h) 1.4 Dioxins (ngTEQ/Nm) 0.018
SO
2
(mg/Nm) 70 60 PAHs (ng/Nm)
NOx (mg/Nm) 200 200 Naphthalene 548
CO (mg/Nm) 20 10 Phenanthrene 269
Hydrocarbons (mg/Nm) <1 Anthracene 9
HCl (mg/Nm) 1.64 Fluoranthene 102
HF (mg/Nm) 0.91 Pyrene 55
Dust (mg/Nm) 220 Benzo(a)anthracene 10
Mercury (mg/Nm) 0.35 Chrysene 73
Cadmium (mg/Nm) 0.001 Benzo(a)fluoranthene 3
Thallium (mg/Nm) <0.0015 Benzo(b)+(k)fluoranthene 39
Arsenic (mg/Nm) 0.0002 Benzo(a)pyrene 12
Nickel (mg/Nm) 0.015 Benzo(e)pyrene 20
Cobalt (mg/Nm) 0.0001 Indeno(1,2,3-cd)pyrene 10
Lead (mg/Nm) 0.38 Dibenzo(a,h)+(a,c)anthracene 3
Chromium (mg/Nm) 0.022 Benzo(g,h,i)perylene 11
Copper (mg/Nm) 0.196 Perylene 4
Manganese (mg/Nm) 0.38 Anthanthrene 3
Antimony (mg/Nm) <0.0001
Vanadium (mg/Nm) 0.011
Tin (mg/Nm) 0.0187
Selenium (mg/Nm) <0.0001
Platinum (mg/Nm) <0.0006
Palladium (mg/Nm) <0.0029
Rhodium (mg/Nm) <0.0016
Zinc (mg/Nm) 1.768
Iron (mg/Nm) 64.63
- Data for installation without flue-gas cleaning
- Average values for 3 measurements, when the standard deviation is >30 % the value is given
Table 3.17: Measured emissions for rotary furnace, melting cast iron, without flue-gas cleaning
equipment
[110, Vito, 2001]

Chapter 3
Smitheries and Foundries Industry 111
3.2.6 Argon Oxygen Decarburisation (AOD) converter

Input
- molten steel
- alloying metals (ferro-alloys)
- Al, FeSi, lime
- O
2
, N
2
, Ar
Output
- decarburised steel
- metal oxide dust and fumes (Fe,Mn,Cr,Ni)
- gases (CO
2
, CO, inert gases)
- slag (CaO,SiO
2
, AlO
3
)
3.2.6.1 Input

The AOD converter is used mainly in the production of low carbon stainless steels and, in
special cases, carbon and low alloy steels: to remove impurities such as carbon, sulphur and
oxides and also to decarburise high chromium stainless steels. The processed material is liquid
steel that has been melted in EAF furnaces or in induction furnaces and afterwards poured off
into the vessel through ladles. The technique is most common in the steel industry, but it is also
applied in foundries, although to only a limited extent. For decarburisation, oxygen is injected in
the converter and oxidises carbon into CO gas, which escapes from the liquid metal. No energy
source is used except for the energy from the combustion reaction between oxygen and carbon
during oxidation and, afterwards, between oxygen and aluminium (and/or silicon) during
reduction.

The consumption of carbon ranges from 0.4 to 1 %, being about 50 120 m of CO per tonne of
steel and 25 60 m oxygen. To eliminate sulphur it is necessary to use a high basicity slag in a
completely reduced liquid steel. Reduction materials, such as aluminium or silicon and lime are
charged in the AOD converter. The consumption levels depend on the quality of the steel grade
and are given in Table 3.18.

Compound Consumption (per tonne of steel)
CO 50 120 m
O
2
25 60 m
Al 1 2.5 kg
Lime 10 20 kg
Si 1 2 kg
Ar 1 5 m
Table 3.18: Consumption levels for the AOD treatment of steel
[202, TWG, 2002]

3.2.6.2 Output

Dust emissions are comparable to EAF furnaces dusts, both in quantity and quality. AOD dust
emissions have lower levels of residuals (organic) from the scrap charge, but on the other hand
have a higher level of metal oxide (Cr, Ni), as it is mainly stainless steel which is processed in
AOD vessels.

3.2.6.3 Waste gases

The effluent from the mouth of an AOD consists of carbon monoxide and inert gas. The rate of
carbon monoxide evolution depends on the tuyres oxygen injection rate and the oxygen
efficiency, or per cent of oxygen which reacts with carbon. This oxygen efficiency, or carbon
removal efficiency, as it is traditionally labelled in AOD operation, varies during the course of
an AOD blow, in response to combined variables of the bath carbon level, temperature, bath
chemistry, and the mixture of injected gases.

Chapter 3
112 Smitheries and Foundries Industry
The CO and inert gas mixture leaves the vessel approximately at the bath temperature. The CO
mixture exiting the vessel is mixed with excess air in order to fully burn the CO to CO
2
very
early in the exhaust duct. This is done to prevent the presence of combustible or explosive
mixtures persisting downstream into the duct to the filter equipment.

3.2.6.4 Slags

Slag is particularly clean from metal oxides, since, due to the peculiarity of the process, they
are usually consumed. The slag composition is given in Table 3.19.

Compound %
CaO 50 70
Al
2
O
3
5 25
SiO
2
10 25
MgO 5 15
Table 3.19: AOD slag composition
[202, TWG, 2002]

3.2.7 Vacuum Oxygen Decarburisation Converter (VODC)

The VODC process involves decarburisation under reduced pressure. It is much less used than
the AOD process. Vacuum processes (EAF/vacuum) accounted for 5.8 % of the Western
Worlds stainless steel production in 1991. The technique consumes 1 Nm Ar per tonne of
steel, shows low chromium oxidation and has a silicon consumption of 3 5 kg/tonne. The
VODC allows the production of ultra-low-carbon and nitrogen grades. [202, TWG, 2002]

3.2.8 Steel refining and treatment

For deoxidation, aluminium is commonly used in the form of a stick. Additions vary from 0.1 %
for medium carbon steel to 0.2 % for low carbon steel. Recovery is between 35 % and 80 %.
Mechanical feeding of aluminium wire is also used. [174, Brown, 2000].

3.2.9 Cast iron treatment

3.2.9.1 Nodularisation

The efficiency, the qualitative assessment of flue-gas production and the complexity of the
various nodularisation methods are compared in Table 3.20.

Sandwich Tundish
cover
Plunging Flow
through
Wire
injection
Inmold Ductilator
Mg- take-up
efficiency (%)
35 50 45 60 40 60 40 50 20 50 70 90 60 75
Flue-gas
production
High Low Low Low Low No High
Comment Easy
operation
Optimised
sandwich but
more mainte-
nance
High
mainte-
nance
High
mainte-
nance
Expensive
installation
Different
design of
pouring
system
High
mainte-
nance
Table 3.20: Comparison of various nodularisation procedures
[110, Vito, 2001], [225, TWG, 2003]

Chapter 3
Smitheries and Foundries Industry 113
3.3 Melting and the metal treatment of aluminium

3.3.1 Survey of melting furnaces for aluminium

Table 3.21 shows typical furnace properties, consumption and emission data for various types of
aluminium melting furnaces. Due to the wide range of furnace capacities and set-ups, the
literature data does not always provide consistent consumption ranges. Consumption levels are
highly dependent on the furnace capacity and exploitation conditions, such as metal temperature
and charge density.

The use of a pure starting material and mainly electric and gas-fired heating, results in relatively
low emission levels from the melting. Due to the limited concern about off-gas quality,
information on the flue-gas composition is limited.

In an aluminium melting operation there is no generation of metal fume and the metal is only
lost when dross forms. This type of loss is often calle the loss by burning and is an oxidation of
the molten metal. It depends on the amount of covering slag and whether combustion occurs. It
is brought about by a leakage of air into the furnace or by the malfunction of the burner. The
costs for this loss can be high, sometimes even being higher than the energy costs. [148,
Eurofine, 2002]

Data for each of the specific furnace types will be discussed and detailed in the subsequent
sections.

Chapter 3
114 Smitheries and Foundries Industry
Units Rotary furnace Hearth type furnace Shaft furnace Crucible furnace
Subtype One chamber Fuel heated Resistance heated Induction
Energy sources Fuels (liquid, gaseous) Fuels (liquid, gaseous) Fuels (liquid, gaseous) Fuels (liquid, gaseous) Electricity Electricity
Thermal efficiency
1
% 15 40 <30 57 35 60 15 40 65 65 70
Primary thermal efficiency
2
% 15 40 <30 57 35 60 15 40 22 22 25
Spec. Energy demand
3
kWh/t Al
4
600 1250 975 1150 580 900
610 720
900 1200
610 680
750
470 590
475 640
440 470
Batch/continuous Batch Batch Continuous Batch Batch Batch
Melting capacity t 3 10 0.5 30 0.5 4 (-15) 0.1 1.2 0.1 0.4 0.2 25
Holding capacity t n.a n.a 1.5 10 0.1 1.5 0.1 1.5 0.15 6
Meltdown time 2 4 3 4 0.5 1 0.5 1 4 5 0.2 0.5
Refining ability Low Low Low Good Very good Low
Loss by burning % n.d n.d 1 3 1 2 1 2 1 2
Dust generation kg/t Al
4
n.d <1 <1 <1 Minor Minor
NO
x
5
kg/t Al
4
n.d <1 6 <1 6 <1 6 n.a n.a
Investment costs
6
EUR 000 n.d n.d 190 370 20 50 12 100 190 500
Running costs EUR 000 n.d n.d 20 100 3 20 15 45 35 150
Abatement techniques Bag house filter
for bigger installations
Bag house filter
for bigger installations
Bag house filter
for bigger installations
Typically not necessary
due to small furnace size
Not necessary Not necessary
1
Definition: the relationship between the heat of the molten bath and the fuel heat supplied; indicated values give an order of magnitude but largely depend on the exploitation
conditions, such as the metal temperature
2
Efficiency of electrical power generation (fuels) assumed to be 35 %
3
Depends on heat recovery measures; only valid for melting; values given by [148, Eurofine, 2002] in italics
4
Units per tonne Al refer to tonne of molten aluminium alloy
5
Depends on burner design and operational performance
6
Depends on heat recovery measures; only valid for melting
Sources of information: VDG internal survey; Aluminium Taschenbuch, Band 2, 15. Auflage, Aluminium Verlag GmbH, Dsseldorf, 1996; Aluminium recycling, Aluminium
Verlag GmbH, Dsseldorf, 2000; [148, Eurofine, 2002]; comments CTIF
Table 3.21: Typical furnace properties and emission data for aluminium melting
[148, Eurofine, 2002] and comments fromCTIF and VDG
Chapter 3
Smitheries and Foundries Industry 115
3.3.2 Shaft furnace

Input Output
aluminium ingots, foundry returns
energy
deoxidation and gas removal products
molten aluminium
dust
NO
x
, CO
used refractory
Emission factors have been provided for emissions per tonne of good casting, based on the
following assumptions:

- average molten metal yield: 70 % (casting/molten metal)
- average scrap value: 5 % (scrap from finishing/casting)
- global metal yield: 0.7 x 0.95 = 66.5 % (good casting/molten metal)

Consumption and emission data per tonne of good casting for a 3 t/h shaft furnace, melting
aluminium are given in Table 3.22. This also gives the flue-gas emission levels of the raw off-
gas without cleaning.

Input Value
per tonne
Value
per Nm
Aluminium ingots 1503 kg/t
Natural gas 717 kWh/t
Electric energy 172 kWh/t
Total energy use 889 kWh/t
Output
Slag (with 35 40 % Al) 40.3 kg/t
Used refractories 0.3 kg/t
Emissions (without flue-gas cleaning)
Particles 0.12 kg/t 112 mg/Nm
NO
x
0.18 kg/t 113 mg/Nm
VOC 0.12 kg/t
SO
2
0.04 kg/t
CO 150 mg/Nm
Pb + Cr + Cu 0.98 mg/Nm
Cd + Hg 0.01 mg/Nm
As + Ni 0.03 mg/Nm
Data are calculated values per tonne of good casting
Table 3.22: Input and output for aluminium melting in a shaft furnace
[177, Silva Ribeiro, 2002]

Table 3.23 gives the consumption levels for a two tonne melting capacity shaft furnace melting
aluminium.

Total measured consumption Specific consumption per
tonne of charged metal
Metal charges 115332 kg (54 % ingot, 46 % scrap) 1 tonne
Metal tapped 113983 kg 0.988 kg/tonne
Metal loss 1349 kg 0.012 kg/tonne 1.2 %
Dross removed 1412 kg 0.012 kg/tonne 1.2 %
Gas consumption 92786 kWh 804 kWh/tonne
Table 3.23: Results from a 6-day trial melt in a 2 tonne melting capacity shaft furnace for Al
[48, ETSU, 1994]

Chapter 3
116 Smitheries and Foundries Industry
3.3.3 Induction furnace

Input Output
aluminium ingots, foundry returns
electrical energy
cooling water
molten aluminium
dust
used refractory

Aluminium induction furnaces usually range from 500 kg to 2 tonnes capacity and operate at
frequencies of 250 1000 Hz.

For example, in one installation, two 1.5 tonne aluminium capacity steel shell tilters are
powered by a 1250 kW, 250 Hz power supply with a change-over switch which allows alternate
furnaces to be melted. With this system 1.5 tonnes can be melted in 40 minutes.

Induction furnaces are energy efficient melters. The energy consumption for melting is affected
by the density of the charge and the melting practice used. Batch melting is less efficient than
using a molten heel. A 50 % molten heel is the most efficient. The energy consumption varies
from 540 kWh/tonne for a high bulk density charge (small scrap and ingot) to 600 kWh/tonne if
a lower density scrap (such as pressure die-casting runners and ingot) is melted. While the
energy consumption is low, the costs for melting may be higher than for gas-fired furnaces
because of the generally higher cost of electricity as a source of heat. [175, Brown, 1999] [148,
Eurofine, 2002]

The induction furnace coil needs a cooling system. Cooling water may run in a closed circuit or
in an open evaporative system.

3.3.4 Radiant roof furnace (resistance heated)

These resistant heated furnaces are holding furnaces, mainly used for aluminium alloys. The
emission depends on the used metal, the holding-temperature, the surface area and the "metal
out" and "metal in" and especially of the holding-time. The emissions are so low that no
measurements are available. [202, TWG, 2002]

3.3.5 Hearth type furnace

Hearth type (or reverberatory) furnaces exist in various sizes and shapes. Large hearth type
furnaces allow rapid melting and can handle bulky charge material, but the direct contact
between the flame and charge material may lead to high metal losses, gas pick-up and to
considerable oxide contamination. Temperature control can also be difficult. This type of
furnace is used less because of its relatively low thermal efficiency (around 1100 kWh/tonne).
The hearth type furnace is also used for copper alloy melting (see Section 3.5.1). [175, Brown,
1999]

Table 3.24 gives the emission data for a 450 kg/h oil-fired hearth type furnace melting
aluminium, and working without a flue-gas treatment system.

Chapter 3
Smitheries and Foundries Industry 117
Compound Emission level
(mg/Nm)
Mass flow
(g/h)
Yearly mass flow
(tonne/yr)
O
2
17.6 %
CO
2
2.2 %
CO <4 <24 <0.0438
NO
x
45 270 0.4928
SO
2
13 78 0.1424
dust 1 6 0.011
TOC 5 30 0.0548
Al 0.092 0.552 0.001
Flue-gas flow: 6000 Nm/h (dry); operational hours: 1825 h/yr
Table 3.24: Emission data for an oil-fired hearth type furnace melting Al
[183, Goovaerts, 2002]

3.3.6 Crucible furnace (fuel and resistance heated)

Input Output
aluminium ingots, foundry returns or liquid
aluminium if used as a holding furnace
electrical energy or fuel
molten aluminium
dust

Crucible furnaces are indirectly heated using fuel-fired burners or electrical resistors. For fuel-
fired crucibles, the thermal efficiency is not as high as for other melting furnaces, since it is
difficult to make use of the heat of the combustion products. They are relatively inexpensive and
since the flames are not in contact with the molten metal, metal losses are low and the melt
quality is high, and also alloy changes can be readily carried out. [175, Brown, 1999]

For aluminium alloys the amount of particles can roughly be assumed to be around 0.3 kg/tonne
of molten metal. [126, Teknologisk, 2000]

Consumption and emission data per tonne of good casting for a 3 t/h crucible furnace, melting
aluminium are given in Table 3.25.

Input
Natural gas 538 kWh/tonne
Electricity 414 kWh/tonne
Total energy input 952 kWh/tonne
Output
Slag 61 kg/tonne
Refractories 6.87 kg/tonne
Emissions (after bag house)
NO
x
0.18 kg/tonne
SO
2
0.04 kg/tonne
VOC 0.12 kg/tonne
Dust 0.12 kg/tonne
All data are calculated values per tonne of good casting
Table 3.25: Consumption and emission data for the crucible melting of aluminium
[177, Silva Ribeiro, 2002]

3.3.7 Aluminium melt treatment

For the combination of degassing and cleaning in an impeller station, a mixture of argon or
nitrogen with 3 % Cl
2
is generally used. For degassing alone, the gases Ar or N
2
are generally
used without Cl
2.
The applied flow and degassing time depend on the type of alloy used and the
size of the treatment vessel.
Chapter 3
118 Smitheries and Foundries Industry
The consumption of modification agents, grain refiners and fluxes depends on the type of alloy,
but is generally in the order of 100 g 1 kg per 50 kg of molten metal.

3.4 Melting and casting of magnesium and magnesium
alloys

3.4.1 Magnesium melt protection

SF
6
and SO
2
are used as a covering gas to prevent the oxidation (or burning) of molten
magnesium. Since SF
6
is more easy to handle than the toxic SO
2
, it became the preferred cover
gas after its introduction in the mid 1970s. SF
6
has a global warming potential (GWP) of 22200
(for a 100 years time horizon) and an atmospheric lifetime of 3200 years [194, UNEP IPCC,
2002].

The quantities of SF
6
used in various operating conditions are given in Table 3.26 (pressure die-
casting) and Table 3.27 (gravity die-casting). These concentrations need to be maintained close
to the melt surface [191, IMA, et al.]. The tabled values are the recommended practice for the
use of SF
6
from the International Magnesium Association (IMA). In practice, higher
concentrations are sometimes used (e.g. 99.4 % CO
2
, 0.6 % SF
6
) [202, TWG, 2002].

Melt
temp. (C)
Atmosphere over the
melt (vol %)
Surface
agitation
Residual
flux
1)

Melt protection
650 705 Air/0.04 SF
6
2)
No No Excellent
650 705 Air/0.2 SF
6
Yes No Excellent
650 705 75 air/25 CO
2
/0.2 SF
6
Yes Yes Excellent
705 760 50 air/50 CO
2
/0.3 SF
6
Yes No Excellent
705 760 50 air/50 CO
2
/0.3 SF
6
Yes Yes Very good
1) May be present from prior operations
2) Minimum concentration under controlled conditions
Note: Dry air (<0.1 % H
2
O) should be used in the mixing
Table 3.26: The use of SF
6
in pressure die-casting operations under various operating conditions
[191, IMA, et al.]

Quiescent (melting/holding)
low gas flowrate
Agitated (alloying/pouring)
high gas flowrate
Crucible
diameter (cm)
SF
6
(ml/min) CO
2
(l/min) SF
6
(ml/min) CO
2
(l/min)
30 60 3.5 200 10
50 60 3.5 550 30
75 90 5 900 50
Note: The suggested flowrates are 1.7 % 2 % SF
6
by volume
Table 3.27: Use of SF
6
in gravity die-casting operations
[191, IMA, et al.]

To date the only flux-free alternative to SF
6
is SO
2
. In an undisturbed melt, the use would be
typically 1 - 2 % in air or nitrogen at a flowrate of 5 to 10 l/min. [182, Closset, 2002], [218,
Harnisch and Schwarz, 2003]

In 2001, the overall EU production of magnesium parts totalled 39100 tonnes, 37 % of which
were produced with SO
2
and 63 %, or 24500 tonnes, with SF
6
(see Figure 3.4). This issue is
discussed further in Section 4.2.7.1.

Chapter 3
Smitheries and Foundries Industry 119
0
5000
10000
15000
20000
25000
D E I A F S UK DK P FIN
M
g
d
i
e
-
c
a
s
t
i
n
g
p
r
o
d
u
c
t
i
o
n
(
M
g
t
o
n
n
e
s
p
e
r
y
e
a
r
)
SF
6
SO
2
EU Member States
Figure 3.4: Mg die-casting production in tonnes/yr by EU states and by cover gas (SO
2
or SF
6
)
(data 2001)
[218, Harnisch and Schwarz, 2003]

3.4.2 Magnesium melt treatment

Magnesium alloys containing aluminium are treated with hexachloroethane (HCE) for grain
refinement. This treatment applies for alloys for sand and gravity die-casting, which makes up
some 10 % of the total magnesium alloys. The required amount of HCE depends on the type of
component to be manufactured and existing data based on the experience of the individual
foundries. An approximate value of 2 kg of HCE per tonne of aluminium containing magnesium
alloy has been reported. [178, Wenk, 1995]

3.4.3 Magnesium scrap

In a Mg die-casting foundry, the amount of return materials (Mg- scrap) is nearly equivalent to
the amount of good die-castings. About 80 % of the return material complies with scrap grade
class 1 (see Table 3.28).

Class Description
Class 1 Clean, compact scrap with known composition
Feeders: clean
Press residues: clean
Bad castings: clean, uncoated
Class 2 Scrap castings, painted
(partly inserts of Fe, Al, no copper and nickel contamination)
Class 3 Unclean compact metal scrap (oily, wet, contaminated with sand,
copper, nickel, ferrosilicon), i.e. mostly post consumer scrap
Class 4 Turnings: clean, dry
Turnings: oily, wet
Scales, flashings, runners: oily, wet
Class 5 Dross (from metal surface)
Class 6 Crucible slag
Class 7 Flux containing scrap, used salt (black dross)
Class 8 Nonmetallic residue
Class 9 intermetallics
Table 3.28: Magnesium scrap grade classes
[206, Ditze and Scharf, 2000]
Chapter 3
120 Smitheries and Foundries Industry
3.5 Melting and casting of copper and copper alloys

3.5.1 Melting and casting units

Copper and copper alloys are generally melted in crucible furnaces. These furnaces generally
operate within the following ranges:

- charge: 30 1800 kg
- capacity: 30 400 kg
- energy consumption (oil): 0.4 l/kg
- energy consumption (butane): 0.3 Nm/kg

Medium frequency induction furnaces are also used with silica or alumina linings.

Table 3.29 gives a typical mass balance for a brass foundry manufacturing taps. Data are given
for combined melting and low-pressure die-casting operations (but without finishing and core-
making). All data refer to tonnes of good castings sold. The data apply to melting in a gas-fired
crucible furnace, with 1 tonne/h capacity.

Input
Electric energy 1360 kWh
Propane gas 14
Compressed air 48 kWh
Water 85.7 l
Gas removal tablets 0.6 units
Deoxidant 0.9 units
Copper alloy 0.44
Cover flux 0.31
Cores 286
Graphite paint 3.4
Output
Non-poured sand 101
Poured sand 6.7
Graphite paint 61.4 l
Zinc powder 0.075
Slag 36.3
Brass scrap 57.1
Brass swarf 18.4
Emissions (without flue-gas cleaning)
Dust 3.9
VOC 3.3
SO
2
0.1
NO
x
0.03
Copper 0.081
Zinc 26.3
All data per tonne of casting sold, data in kg unless
stated differently
Table 3.29: Mass balance data for a brass low-pressure die-casting foundry (melting + die-casting
operations)
[177, Silva Ribeiro, 2002]

For copper alloys, the amount of particles emitted greatly depends on the zinc content of the
alloys. As a guide the values in Table 3.30 show the relationship between the quantity of
particles emitted and the zinc content.

Chapter 3
Smitheries and Foundries Industry 121
Zn content
(%)
Particle emissions
(kg/tonne molten metal)
Bronze 0 7 0.3 1.5
Brass 20 40 0.5 16
Table 3.30: Particulate emissions from the crucible melting of copper alloys
[126, Teknologisk, 2000]

The production of slags and dross is around 60 kg/tonne of molten metal. This value is similar
for melting in induction or hearth type furnaces and for various copper alloys. The composition
of the slag depends on the type of alloy melted. In general the copper content is in the 45 55 %
range. A constant slag quality may be attained by specific mixing of the various slag types from
the foundry. This constant quality is needed for external recycling.

The production of refractory waste is 8 9 kg/tonne of molten metal for induction melting and
7.5 kg/tonne of molten metal for oil-fired hearth type furnaces.
[34, Binninger, 1994]

3.5.2 Copper and copper alloy melt treatment

When the bubbling of inert gas is used for the degassing of copper, 50 70 litres of gas are used
for each 100 kg of copper. An alternative is the use of specific tablets or briquettes. Degassing
treatment usually takes 3 10 minutes depending on the size of the melt. The precise amount of
deoxidant needed depends on the melting practice used.
[175, Brown, 1999]

3.6 Melting and casting of zinc and zinc alloys

Table 3.31 gives a typical mass balance for a zinc foundry using pressure die-casting. It applies
to the casting of zamac (ZnAl4Cu1 or ZnAl4). Zinc alloy ingots are melted together with rejects
and feeders in a gas-fired crucible furnace of the casting machine. A release agent is sprayed on
the die before casting. The casted piece is cooled in a water bath and the feeder system is
removed. The casting is subsequently degreased and ground. The melting off-gas is collected
and dedusted in a dry system. Data on core production and finishing are given in Section 3.9
and 3.11.

Input
Zinc alloy 1040
Casting rejects 50
Feeders 450
Release agent 20 l
Water 1 m
Electrical energy 700 kWh
Natural gas 70 Nm
Output
Good castings 1000
Swarf 3
Dross 30
Sludge from washing and grinding 2
Off-gas 10000 Nm
Particulates 1
All data per tonne of good casting, data in kg unless stated
different
Table 3.31: Typical mass and energy data for zinc pressure die-casting

Chapter 3
122 Smitheries and Foundries Industry
3.7 Melting and casting of lead

Casting processes for lead may be divided into:
- pressure die-casting of lead
- casting for storage batteries
- sand casting for protective screens.

3.8 Waste gas cleaning

3.8.1 Abatement systems

Various waste and exhaust gas cleaning techniques are used in the foundry industry. Their
principles are discussed in the BREF document for non-ferrous metal industries. The properties
and emission levels of dust abatement systems are given in Table 3.32. A full discussion on the
selection of the abatement technique, its applicability in the various foundry processes and the
achievable emission levels is given as part of the techniques to be considered in the selection of
BAT in Section 4.5.

Technique Particle
Size

(Nm)
Collection
Efficiency
at 1 m
(%)
Maximum
Operating
Temperature
(C)
Range of
Achievable
Emissions
(mg/Nm
3
)
Comments
Hot EP <0.1 >99
Depending
on design
450 5 15
(Pre abatement
>50)
4 or 5 zones. Usual
application is pre-
abatement
Wet EP 0.01 <99 80 1 5
(Visually clear)
EP with 2 zones in
series. Mainly mist
precipitation
Cyclone 10 40 1100 100 300 Coarse particles.
Used to assist other
methods
Fabric Filter 0.01 >99.5 220 1 5 Good performance
with suitable dust
type
Wet Scrubber 1 3 >80 99 Inlet 1000
Outlet 80
4 50 Good performance
with suitable dusts.
Acid gas reduction
EP: Electrostatic precipitator
Table 3.32: Example of current emisions from some dust abatement applications
[155, European IPPC Bureau, 2001]

3.8.2 Dioxins

In melting processes, dioxins may be produced if the conditions that give rise to such pollutants
are present at the same location and time in the process. These conditions are:
- the presence of chloride ions these can arise from contaminated scrap, from the use of
coal, coke, fuel oil or from certain fluxes
- the presence of organic carbon this may arise from contaminated scrap and from coal,
coke or oil used as a fuel
- temperature conditions between 250 C and 450 C, with a sufficient gas residence time in
this temperature interval
- the presence of a catalyst such as copper
- the presence of oxygen.

In evaluating the risk of dioxin formation, a distinction can be made between non-ferrous and
ferrous foundries:
Chapter 3
Smitheries and Foundries Industry 123
- Non-ferrous foundries: In as far as when only ingots and internal scrap are melted, the risk
of dioxin formation in the melting stage is very low. The melting of pure non-ferrous metals
lacks the presence of both the chlorine and carbon required for dioxin (re)formation.
However, the re-melting of external non-ferrous scrap materials for metal production may
involve a risk of dioxin formation. This action, however, falls beyond the scope of this
document and is considered in [155, European IPPC Bureau, 2001]
- Ferrous foundries: Depending on the furnace type and metal load the conditions for dioxin
formation could occur. Considering the high temperatures in the melting furnace, dioxin
emission (if occurring at all) will mainly generate from de-novo synthesis. The above-
mentioned conditions can be used to evaluate the risk of dioxin formation.

Dioxin emission data have been gathered from a broad range of sources. Table 3.33 presents
data for various metals and furnace types. Blank spaces indicate that the indicated number was
not specified. All systems in the survey do not provide specific dioxin capture techniques. A
distinction can be made between two groups of techniques where dioxin levels are very low
(<0.05 ng/Nm) or cover a wide range (<0.01 3 ng/Nm). The first group covers aluminium
melting, the induction melting of iron and the EAF melting of steel. The second group covers
cupola melting and the rotary melting of iron. An analysis of the literature data does not provide
a full contribution of the reasons for the differences within the second group. However, it can be
seen that the presented HBCs with a wet scrubber system show noticeably lower PCDD/F
emission values.

Product type Furnace Melting Flue-gas Abatement O
2
PCDD/F
(tonne/h) (m/h) (%) (ngTEQ/Nm)
Aluminium Hearth type n.d n.d n.d 0.002
Aluminium Hearth type 0.45 9300 None 18.8 0.002
Aluminium Shaft 1.5 8400 None 18.4 0.01
Cast iron CBC 3.4 15900 Bag filter n.d 0.04
Cast iron CBC 3.7 14300 Bag filter 16 0.09
Cast iron CBC 4.5 14300 Bag filter n.d 0.09
Cast iron CBC 3.4 n.d n.d 0.33
Cast iron CBC 5.5 17400 Bag filter 15.9 0.51
Cast iron CBC 6.5 17500 Bag filter n.d 0.51
Cast iron CBC 6 27600 Bag + PC n.d 3.14
Cast iron HBC 45.5 55000 Disintegrator 6 0.003
Cast iron HBC 60 n.d Disintegrator n.d 0.003
Cast iron HBC 40.6 75000 Bag filter 12.5 0.05
Cast iron HBC 50 75000 Bag filter n.d 0.07
Cast iron HBC 15 36400 Bag + PC n.d 0.05
Cast iron HBC 13 n.d n.d 0.10
Cast iron HBC 18.2 29100 Bag filter 8.6 0.20
Cast iron HBC 17.1 22500 Bag filter 7.5 0.29
Cast iron HBC 27 n.d n.d 1.00
Cast iron HBC 28 37000 Bag filter n.d 2.08
Cast iron HBC 21 32000 Bag + PC n.d 3.09
Cast iron IF 19.5 208000 Bag filter 20.2 0.003
Cast iron IF n.d n.d Bag filter n.d 0.01
Cast iron RF 8 n.d n.d 0.004
Cast iron RF 1.4 9000 None n.d 0.02
Cast iron RF 2.1 18600 Bag filter 19.9 0.45
Cast iron RF 3.5 n.d Bag filter n.d 0.61
Steel EAF 5.4 54150 Bag filter 20.9 0.003
Steel EAF 9 5000 Wet scrubber n.d 0.02
CBC: Cold blast cupola; HBC: hot blast cupola; RF: rotary furnace; IF: induction furnace; EAF: electric arc furnace;
Table 3.33: Dioxin emission data for various foundry types
[21, Strau, 1994], [23, Brettschneider and Vennebusch, 1992], [27, Kran, et al., 1995], [43, Batz,
1996], [110, Vito, 2001], [202, TWG, 2002], [225, TWG, 2003], personal communications
Chapter 3
124 Smitheries and Foundries Industry
The dioxin emission data for the various ferrous metal melting furnaces are discussed below.

- Cupola furnaces: In cupola furnaces, a massive surplus of chlorine is always present from
coke. Enough carbon is present from coke too, but an additional input of carbon may be needed
the event of caused by poor scrap qualities. Under specific operational conditions, the
conditions for dioxin formation could occur. Since de novo synthesis mainly occurs during
cooling of the flue-gas, this applies to both hot blast and cold blast cupolas. In Table 3.34, the
result of a statistical analysis of all the measurement data from Table 3.33 for CBC and HBC is
given. Whereas Table 3.33 presents average values per plant, for Table 3.34 individual
measurements were used to perform an overall analysis.

Units Cold blast Hot blast
Average ngTEQ/Nm 0.54 0.75
St. dev. ngTEQ/Nm 1.08 1.3
Median ngTEQ/Nm 0.18 0.09
Minimum ngTEQ/Nm 0.001 0.001
Maximum ngTEQ/Nm 5.1 4.4
Number of measurements 35 18
Number of furnaces 11 11
Table 3.34: Cupola furnace dioxin emissions
[21, Strau, 1994], [23, Brettschneider and Vennebusch, 1992], [27, Kran, et al., 1995], [43, Batz,
1996], [110, Vito, 2001], [202, TWG, 2002], [225, TWG, 2003], personal communications

Data from a German survey (6 installations, 18 measurements) showed a range of 0.006 - 0.22
ng TEQ/Nm, with an average of 0.0724 ng TEQ/Nm (0.0608 i-TEQ/Nm). The data set of the
survey partially overlaps with the data set presented above, but it did not give sufficient
information to expand it. [224, Helber, et al., 2000]

The data show that the average, the standard deviation and the range of values are similar for
both cold blast and hot blast furnaces. The median value for hot blast is lower than for cold blast
furnaces. This confirms the statement from [224, Helber, et al., 2000] that there is no statistical
difference between dioxin emissions for hot and cold blast cupola furnaces. The high standard
deviation shows that the data should be interpreted on a plant-by-plant basis rather than on an
averaged basis.

For wet dedusting with a disintegrator, the off-gas passes through the critical temperature
window (250 450 C) after dedusting, i.e. with a low dust content. Additionally, chlorides
have been washed out. This strongly reduces the risk of de novo synthesis.

- Rotary furnaces: Due to the limited possibility of alloying, the rotary furnace charge
generally consists of clean material. Due to the high flame temperature, the hot gases leave the
furnace at temperatures of between 1000 and 1300 C. Post combustion occurs inside the
furnace volume. De novo synthesis is possible if flue-gases cool slowly. Reported data range
from 0.004 0.61 ngTEQ/Nm.

- Induction furnaces: Due to the limited possibility of alloying, the induction furnace charge
generally consists of clean material. Furthermore the furnace does not produce a ducted high
temperature flue-gas stream that cools down slowly.

- Electric arc furnaces: EAF furnaces allow melt treatment and alloying if operated in basic
lining practice. This enables the use of dirty scrap as a raw material for melting. If scrap
containing organic and/or chlorinated compounds is charged, dioxins may be formed during
cooling of the flue-gas stream. This may be the case when using scrap from e.g. electronic
equipment, transformers, and shredded vehicles. The EAF in acid lining practice does not allow
metal treatment and therefore the possibility for feeding contaminated scrap is reduced. This
also reduces the risk of dioxin formation.

[161, UK Environment Agency, 2002]
Chapter 3
Smitheries and Foundries Industry 125
3.9 Mould and core production

3.9.1 Introduction

Mould and core production involves mixing sand with various chemicals, in some cases
followed by a heating step. Gaseous or volatile reaction-products and excess reagents are
emitted. Table 3.35 presents a survey of the emissions and other environmental impacts during
the mixing, setting and storage of the moulds and cores. A further discussion of the emission
and consumption levels for each of the binder types is given in the subsequent sections.

System name and binder
constituents
Setting method and
relative energy
requirement
Emissions to air
during mixing and
setting
Other environmental impacts
GREEN SAND
Clay
Coal dust or substitute
Water

Pressure low

Particulate matter
no significant
emission to the
environment
Sand spillage around conveyors needs
to be avoided to reduce the likelihood
of fugitive emissions.
Abatement from the mixing process is
not essential (the process is usually
self-contained with displaced air
vented to the foundry)
SHELL SAND
Phenol - formaldehyde
(Novalak) Resin

Heat - high

Formaldehyde*
Ammonia*
Phenol*
Aromatics
HCN

Odour can be an issue as the shell
machines are normally extracted to air

ALKALINE PHENOLIC
Resol - Alkaline phenol-
formaldehyde resin

1. Gas hardened
(alkaline phenolic cold-
box)

2. Self-setting
(alkaline phenolic no bake)
Gas hardened with methyl
formate vapour low

Cold-set with esters - low

Formaldehyde*
Phenol*
Methyl formate

Formaldehyde*
Phenol*
Esters

PHENOLIC
URETHANE
1. Gas hardened: Cold-box
2. Self-setting
(phenolic urethane no
bake)

Amine vapour low

Self-set with substituted
pyridine low

Isocyanate (MDI)
Amine*
Formaldehyde*
Phenol

Isocyanates (MDI)
Formaldehyde*
Phenol

Odour is frequently a problem where
DMEA is used odours arise and
abatement is essential. This may be
incineration or gas scrubbing (using
sulphuric or phosphoric acids) the
latter gives rise to liquors, which are a
special waste.
Where TEA is used scrubbing is only
required if odour problems arise
FURAN
Combination resins of:
Phenol
Urea
Furfuryl alcohol
Formaldehyde

Cold-set with acids low
Formaldehyde*
Phenol*
Furfuryl alcohol*
Hydrogen sulphide
Acid mists

Resins and acids must be kept apart
(unless sand is present) as they are
vigorously exothermic when in contact
HOT-BOX
Combination resins of:
Phenol
Urea
Furfuryl alcohol
Formaldehyde

Heat high

Formaldehyde*
Acids
Furfuryl alcohol*
Phenol*
Ammonia
Isocyanic acid
Methyl isocyanate

Odour can be an issue as the shell
machines are normally extracted to air

Chapter 3
126 Smitheries and Foundries Industry
System name and binder
constituents
Setting method and
relative energy
requirement
Emissions to air
during mixing and
setting
Other environmental impacts
OIL SAND
Linseed oil and starch

Heat high
Acrolein*
Complex organics
Odour can be an issue as the core
ovens are often extracted to air
CO
2
PROCESS
Sodium silicate

Gas hardened with CO
2
gas low

None

SILICATE ESTER
Sodium silicate

Cold-set with esters low Esters

Note 1: All the above processes give rise to spent sand (including broken cores, spillage and mixer residues) that may go to landfill
Note 2: Substances marked with * are those most likely to give rise to odour from the process referred to
Note 3: Any resin binder component would be regarded as special waste for disposal purposes and if spilled could give rise to a risk
of contamination of water systems
Note 4: Amines and methyl formates used for gas hardening are highly flammable and odorous. The avoidance of leaks in storage is
essential
Table 3.35: Environmental impact from binder systems
[126, Teknologisk, 2000], [160, UK Environment Agency, 2002], [225, TWG, 2003], [229, Lilja, et
al., 2000]

Quantitative data for dust emissions from moulding shops have been gathered in an Italian
foundry survey. The results are given in Table 3.36. All exhaust cleaning systems attain a level
<15 mg/Nm.

Concentration (mg/Nm) Emission factor (g/tonne)*
Average Minimum Maximum Average Minimum Maximum
Bag filter 3.2 0.4 12.1 24.4 0.5 108.3
Wet scrubber 5.2 3.6 6.7 6.2 4.0 8.0
Wet Venturi system 9.6 8.5 10.9 34.2 30.1 39.9
Units: g/tonne of good casting
Number of data points: bag filter: 33; wet scrubber: 4; wet venturi system: 3

Table 3.36: Emission values and emission factors for dust emissions from moulding shops after
exhaust cleaning
[180, Assofond, 2002]

The emissions given in Table 3.35 refer to the stages of mixing, setting and storage of the moulds
and cores. Most emissions persist during the metal pouring and casting cooling stages, but now
with the addition of pyrolysis products. These will be discussed in Section 3.10.1.

Table 3.37 gives dust emission data from a German measurement campaign. The dust was
analysed into PM
10
, PM
2.5
and PM
1
fractions.

Activity Off-gas
volume
(Nm/h)
Gas cleaning
equipment
Total dust
(mg/m)
PM
10

(%)
PM
2.5

(%)
PM
1
(%)
Sand
preparation
25600 Bag filter 0.3
Green sand
preparation
24400 Fume hood,
bag filter
0.7 88 38
Sand
preparation
70400 ESP 22 28.3 79 18
Core shop 4670 (dry) Amine
washer
0.7 98 47 62
Moulding
shop
52300 (dry) Bag filter 0.7 95 97 50 60 2 5
Finishing 22000 (dry) Bag filter 5.3 100 45 48 9
Table 3.37: Example dust emissions and particle size for mould and core-making
[202, TWG, 2002]
Chapter 3
Smitheries and Foundries Industry 127
3.9.2 Moulding with clay-bonded sand (green sand moulding)

Input Output
sand
bonding clay (e.g. bentonite)
coal dust, dextrin
water (to prepare moulding mix)
green sand moulds
dust (silica fines, partially burnt-out
clay, unburned coal dust and ash)
The composition of the bentonite added to the sand depends on the specific properties of both
the sand and the bentonite, as well as on the desired strength and gas permeability of the
finished mould. Table 3.38 gives some parameters of clay-bonded sand, as obtained from a
selection of 105 sand samples taken from iron foundries.

Parameter Units Average % of samples in range of
average 10 %
Maximum
value
Minimum
value
Range %
Water content % 4.1 3.4 4.5 48 6.9 2.4
Bentonite content % 8.3 7.5 9.1 45 11.9 5.6
Density g/cm 0.940 1.00 0.85 75 1.06 0.73
Loss on ignition % 5.0 4.5 5.5 20 15.0 1.0
Compressive
strength
N/cm 18.6 16.7 20.5 54 24.5 13.2
Table 3.38: Typical properties of green sand, measured on 105 sand samples from 105 iron
foundries
[36, Winterhalter, et al., 1992]

Further additives are mixed in the green sand mixture:
- Coal dust: Used mainly in iron foundries, but also to a lesser degree in some non-ferrous
foundries. Coal dust levels in green sand vary from 2 or 3 % for small castings, to 7 or 8 %
for heavy section castings. Too much coal dust can give rise to holes in the castings or in
misruns. Various coal dust replacement products exist. They are generally added in a
slightly altered concentration compared to the coal dust
- Cereal binders: Starch and dextrin are used mainly in steel foundries to avoid defects
caused by silica expansion and to control moisture excess in the sand mix. Effective
additions of cereal binders to new green sand facing mixes are between 0.5 and 0.75 %. In
unit type green sands, part of the cereal is destroyed during the casting process. The
recommended addition on each recycle is 0.1 to 0.25 %, depending on the amount of cereal
burn-out and on the dilution by cores and new sand.
[174, Brown, 2000], [225, TWG, 2003]

3.9.3 Moulding with unbonded sand (V-process)

Input Output
dry sand
PE sheets
energy (produce and maintain vacuum)
moulds
dust

The thermal shock induced by the contact of the molten metal breaks up the sand grains,
generating dust that is mostly emitted during shake-out. The remaining dust in the sand has to
be removed in order to allow the re-use of the sand for moulding. Using this method, only a
small amount of dust should occur during the filling of the moulding boxes.

Chapter 3
128 Smitheries and Foundries Industry
This process is of interest because of its very low environmental impact. However, as its process
cycle involves many steps, the moulding rate is low and the gap between starting times for each
new different casting is lengthy. Moreover investment costs are very high. As a consequence,
this process has found only limited application.

3.9.4 Moulding and core-making with chemically-bonded sand

Input Output
sand
resin
catalyst, hardener, additives
sand moulds and cores
excess reagents
reaction products
dust
3.9.4.1 Chemicals consumption levels

The consumption levels of the various types of binders, hardeners, catalysts and additives are
given in Table 3.39.

Hardening
type
Resin type

Resin
addition
%on sand
weight
Catalyst/
hardener
type
Catalyst/
hardener
addition
%on resin
weight
Additive
type
Additive
Addition
%on sand
weight
Cold-setting Furan 0.8 1.5 Sulphonic acid 25 60 Silane 0.1 0.2
Phenolic 1 2 Sulphonic acid 25 50 n.a n.a
Polyurethane 0.8 1.5 Pyridine derivate 2 6 n.a n.a
Resol 1.0 1.5 Ester 22 25 n.a n.a
Alkyd oil 1.0 2.0 Isocyanate 18 20 Catalyst 0.002 0.2
Ester silicate 2.0 4.5 Ester 10 15 n.a n.a
Hardening
type
Resin type

Resin
addition
%on sand
weight
Catalyst type Catalyst
addition
%on sand
weight
Additive
type
Additive
Addition
%on sand
weight
Gas hardening Phenolic/Furan 0.8 1.2 SO
2
0.3 3 Peroxide 0.1 0.5
Polyurethane
Cold-box
1.0 1.8 Amine 0.05 0.15 n.a n.a
Resol 1.2 1.8 Methyl fomate 0.3 0.5 n.a n.a
Resol 2 2.5 CO
2
0.7 1.25 n.a n.a
Acryl/Epoxy 1.2 1.6 SO
2
0.02 0.05 Peroxide 0.05 0.06
Silicate 2 4 CO
2
1 2 n.a n.a
Hardening
type
Resin type

Resin
addition
%on sand
weight
Catalyst type Catalyst
addition
%on resin
weight
Additive
type
Additive
Addition
%on sand
weight
Thermosetting Oil 0.8 4 n.a n.a
Warm-box
furfuryl
alcohol-based
1.0 1.5 Cu salts of
sulphonic acids
10 30 n.a n.a
Hot-box
phenolic or
furan based
1.2 3.0 Ammonium salts 10 25 n.a n.a
Croning
phenolic
1.5 5 Hexamethylene-
tetramine
15 n.a n.a
Table 3.39: Consumption levels of various resins, catalysts, hardeners and additives for chemically-
bonded sand preparation
[174, Brown, 2000], [225, TWG, 2003]
Chapter 3
Smitheries and Foundries Industry 129
3.9.4.2 Emission factors

The use of chemical binders results in the production of various compounds during mixing,
mould- and core-making, storage, pouring and cooling, as indicated in Table 3.35. Some
indicative emission factors for several processes are given in Table 3.40 and Table 3.41.

Furan sand Phenolic sand
Organic solvents 1.4 1.25
Phenol 0.02 0.18
Formaldehyde 0.08 0.15
All data in kg/tonne molten metal
Table 3.40: Emission factors for mould making
[110, Vito, 2001]

Cold-box Hot-box Shell
Dust n.a 0.003 0.003
Furfuryl alcohol 0.1 n.a n.a
Formaldehyde 0.01 n.a 0.003
Organic solvents n.a 0.03 n.a
Aromatic solvents 0.12 n.a n.a
Amines 0.13 n.a n.a
All data in kg/tonne molten metal
n.a: not applicabble
Table 3.41: Emission factors for core-making
[110, Vito, 2001]

3.9.4.3 Cold-setting processes emissions

Emissions for the cold-setting processes can be described as follows:

- Phenolic: Vapours of formaldehyde and phenol may be emitted because of the vapour
pressure of these constituents. But, as the polymerisation occurs at ambient temperature,
these vapour pressures are low and given the consumption rates, the emissions are
insignificant
- Furan: Vapours of formaldehyde, phenol, furfuryl alcohol and alcohols may be emitted
because of the vapour pressure of these constituents. But, as the polymerisation occurs at
ambient temperature, these vapour pressures are low and given the consumption rates, the
emissions are insignificant
- Polyurethane: Vapours of formaldehyde, phenol, isocyanate and aromatic solvents may be
emitted because of the vapour pressure of these constituents. As the polymerisation occurs
at ambient temperature, the vapour pressures of the components are low and given the
consumption rates, the emissions are insignificant. Obnoxious odours may be generated in
the moulding area but are not likely to cause external nuisance
- Resol-ester: The resin contains unreacted phenol and formaldehyde, but their emissions are
extremely low and environmentally insignificant
- Alkyd-oil: During production of the moulds and cores, there are no emission problems,
unless they are cured by heat, in which case odour problems may occur
- Ester silicate: This process does not generate any emission problems.
[32, CAEF, 1997]

Chapter 3
130 Smitheries and Foundries Industry
3.9.4.4 Gas-hardened processes emissions

Emissions for the gas-hardened processes can be described as follows:

- Cold-box: Vapours of formaldehyde, phenol, isocyanate and aromatic solvents are emitted
in low quantities, in spite of their low vapour pressures. Aromatic solvent emissions are
highest during purging. The most significant emission is that of amines, which have low
odour detection thresholds and can create an external nuisance. Amines are hazardous, with
a relatively low exposure value limit, and possess a strong characteristic smell at very low
concentrations
- Resol-ester: The resin contains unreacted phenol and formaldehyde, but their emission
levels, even during the gassing and purging periods, are very low. Methyl formate is not
toxic, does not have an unpleasant smell and its exposure value limit is relatively high
- Resol-CO
2
: The resin has low contents of unreacted phenol and formaldehyde, and their
emission levels, even during the gassing and purging periods, are very low
- SO
2
hardened phenolic and furan resin: The resins and the prepared sands generate
formaldehyde emissions. The major emission problem is caused by sulphur dioxide, which
is very hazardous
- SO
2
hardened epoxy/acrylic: Minimal emissions are generated during the core-making
process
- CO
2
silicate: This process does not generate any emission problems during moulding or
core-making.
[32, CAEF, 1997]

3.9.4.5 Hot curing processes emissions

Emissions for the hot curing processes can be described as follows:

- Hot-box emissions: The patterns are generally heated by open air gas burners, producing
combustion gas emissions. The off-gas may contain phenol, ammonia, formaldehyde and
monoisocyanates (if the resin contains nitrogen)
- Warm-box emissions: Compared to the hot-box process, the emissions are significantly
lower. The emissions do not contain phenol or ammonia, and also formaldehyde emissions
are diminished by a factor of 4. As the pattern temperature is also lower compared to the
one used for the hot-box process, the working conditions are also improved. The
environmental impact is considered to be relatively low
- Shell emissions: Compared to the hot-box process, cured sand consumption is very low.
However, the pre-coated sand contains 2 to 3 times more resin, but as the pattern
temperatures are approximately the same, the resulting impact on working conditions is
similar.
- Linseed oil emissions: As the oils are often in solution in organic solvents such as naphtha
or kerosene, a large amount of VOC is generated by curing, which may result in an external
odour nuisance. This problem is even more persistent during pouring. These problems, as
well as poor productivity, are some of the reasons for the reduction in interest in the use of
this process.
[32, CAEF, 1997]

A mass balance for a hot-box core shop is given in Table 3.42. Data were collected from a core
shop of a low-pressure die-casting brass foundry [177, Silva Ribeiro, 2002].

Chapter 3
Smitheries and Foundries Industry 131
Input
New sand 371
Electric energy 352.9 kWh
Propane gas 7
Compressed air 187.1 kWh
Resin 8.3
Conservative 0.51
Catalyst 0.76
Coating 1.7
Output
Cores 287.7
Sand not poured 84.3
Resin 1.4 l
Metallic and plastic containers
Emissions (after bag filter)
Particulate emissions 0.11
VOCs 0.83
All data are per tonne of castings sold, values in kg unless stated
differently
Table 3.42: Mass balance data for brass foundry core using a hot-box process
[177, Silva Ribeiro, 2002]

3.9.5 Coating of chemically-bonded sand moulds and cores

Input Output
coatings ready to use or mass to be diluted
with water or alcohol
heat for solvent evaporation
VOC (alcohol-based)

3.9.6 Expendable pattern casting (Lost Foam/full mould casting)

3.9.6.1 Unbonded sand Lost Foam

Input Output
EPS granulate
EPS glue
refractory coating
sand
sand moulds
EPS residues
products of pyrolysis and combustion
dust
In producing aluminium castings, around 15 kg/tonne good casting of EPS granulate is
consumed. Refractory coating consumption is around 20 kg/tonne good casting.

Table 3.43 shows a comparative analysis of a green sand and a Lost Foam aluminium foundry,
both producing 5490 tonnes of good castings per year. It should be noted that the values do not
take into account the internal regeneration. In practice, both types of foundry will apply
regeneration, thus reducing the amount of residue effectively left for disposal.

Residue type Lost Foam foundry Green sand foundry
Foundry sand (without internal regeneration) 1.04 2.95
Filter dust (without internal regeneration) 0.056 0.22
Filter dust (with internal regeneration) 0.056 0.61
EPS residue 0.0027 -
All values in kg/tonne good casting
Table 3.43: Comparative data for residue production from Lost Foam and green sand aluminium
casting
[96, Spillner, 1997]
Chapter 3
132 Smitheries and Foundries Industry
Emission data for organic compounds are given in Table 3.44. These apply to an aluminium
Lost Foam foundry producing 1.5 tonne/h compared to a green sand foundry producing
1.2 tonne/h and using 1.9 tonne/h cold-box cores. The Lost Foam foundry was equipped with a
bag house and a post combustion unit. The green sand data apply to a unit with bag house
filtration. The data show that the Lost Foam foundry, with the application of more intense flue-
gas cleaning, emits higher levels of BTEX and formaldehyde, but a clearly lower level of
organic carbon. Post combustion of the casting off-gases is needed to decompose the products
from EPS pyrolysis upon pouring.

Compound Lost Foam foundry Green sand foundry
Benzene 35.33 8.5
Toluene, xylene, ethylbenzene, styrene 354.67 18.58
Unknown organic compounds 96.67 655.0
Total organic carbon 857.33 1283.33
PAH 1.45 1.43
Formaldehyde 18.00 2.08
Phenol 18.00 69.83
All values in g/tonne molten aluminium
Table 3.44: Stack emission data for Lost Foam and green sand aluminium foundries
[96, Spillner, 1997]

3.9.6.2 Chemically-bonded sand full mould

Input Output
EPS granulate
EPS glue
refractory coating
sand
binders
sand moulds
EPS residues
products of pyrolysis and combustion
dust

The mass flow of emissions from full mould and hollow mould casting processes are to be
considered as similar. This can be seen from Figure 3.5 which shows the mass flow of organic
carbon as a function of time upon pouring. Time 0:00 corresponds to the starting of the
pouring. The data were collected using furan-bonded sands, both for full and hollow mould
making.

The full mould process shows a high emission upon pouring and an initial peak emission
immediately after pouring which levels off within 1 hour. The hollow mould process shows a
lower initial emission, but the levelling off goes slower and takes up to 2 hours. Therefore, the
total organic carbon emission over the whole cooling step is equal for both processes.
Additional measurements showed that the same profile applies for benzene, ethyl benzene,
phenol and furfuryl alcohol. The maximum emission for styrene and toluene occurs after 15
30 minutes. This is due to an initial condensation of the molecules on the colder parts of the
mould. [215, Mller, 1996]

Chapter 3
Smitheries and Foundries Industry 133
Figure 3.5: Waste gas load upon pouring in full and hollow moulds
cast alloy: lamellar iron; mould: furan bonded sand; 1.9 tonne sand: 1 tonne Fe
[215, Mller, 1996]

3.10 Casting

3.10.1 Casting, cooling and shake-out, using lost moulds

3.10.1.1 Emission levels

Input Output
finished moulds
molten metal

castings
used sand
combustion products (from preheating of pouring
ladles)
organic pollutants from pyrolysis and thermal
degradation of binder, blackening moulds, etc.
(phenol, formaldehyde, amine, hydrogen cyanide,
PAH, benzene, VOC)
odour
waste from exhaust air cleaning (dry/sludge)
dust from shake-out
Whilst preheating ladles, combustion gases are generated as a result of using natural gas as the
common energy source.

The following emission types can occur during pouring:
- thermally degradable components, such as exothermic sleeves, react releasing fumes and/or
vapours
- chemical compounds from the binder and some blacking systems which can be released as a
result of thermal degradation and/or volatilisation, e.g. combustion gases, water vapour, and
volatile organic compounds. Some of the degradation products can be odorous.

Experience shows that the maximum emissions (related to total-C) only occur 10 minutes or
more after pouring. CO is the main component, with the level of CO being indicative of the
release of other compounds. [110, Vito, 2001]
Chapter 3
134 Smitheries and Foundries Industry
During cooling and shake-out, the thermal decomposition processes proceed and volatile
compounds are generated, mainly controlled by diffusion and evaporation rates.
The removal (shake-out) of the sand moulds and cores from castings is a dust-intensive
procedure, as the moulds, at least partially, have to be crushed.

The second cooling step releases nothing except water vapour from those plants, where water
spray is used as a cooling aid. [32, CAEF, 1997]

The type of emissions during casting are dependent on the binder type used. Emissions are
comparable to those during the mixing stage, with the addition of pyrolysis products occurring
upon contact with the hot metal. Table 3.45 gives the results from a qualitative survey of the
emissions associated with the various binder types.

System name and binder
constituents
Emissions to air during casting Comments
GREEN SAND
Clay
Coal dust or substitute
Water
Particulate matter- soot from coal
burning
Carbon monoxide and carbon
dioxide
Benzene
Toluene
Xylene
Potential odour (may be
associated with the sulphur
content of the coal)
SHELL SAND
Phenol - Formaldehyde
(Novalak) Resin
Particulate matter soot from the
incomplete combustion of carbon
based resins
Carbon oxides
Phenol*, cresols* and xylenols*
Ammonia
Aldehydes
Benzene
PAH
Odour problems more prevalent
treatment may be necessary
although dispersion may suffice
ALKALINE PHENOLIC
Resol - Alkaline phenol
Formaldehyde resin

1. Gas hardened

2. Self-setting
Particulate matter soot from the
incomplete combustion of carbon
based resins
Carbon oxides
Formaldehyde
Phenol, cresols and xylenols
Aromatics
Odour may be a problem
PHENOLIC URETHANE
1. Gas hardened e.g. Cold-box

2. Self-setting
(phenolic urethane no bake)
Particulate matter soot from the
incomplete combustion of the
carbon based resins
Carbon oxides
Nitrogen oxides
Monoisocyanates
Formaldehyde
Phenol, cresols and xylenols
Aromatics (inc. polycyclics)
Anilines
Naphthalenes
Ammonia
Odour may be a problem
Chapter 3
Smitheries and Foundries Industry 135
System name and binder
constituents
Emissions to air during casting Comments
FURAN
Combination resins of:
Phenol
Urea
Furfuryl alcohol
Formaldehyde
Particulate matter soot from the
incomplete combustion of the
carbon based resins
Carbon oxides
Phenol, cresols and xylenols
Formaldehyde
Aromatics (inc. polycyclics)
Sulphur dioxide
Ammonia
Aniline
Isocyanic acid*
Methyl isocyanate*
Odour may occasionally be a
problem
HOT-BOX
Combination resins of:
Phenol
Urea
Furfuryl alcohol
Formaldehyde
Particulate matter soot from the
incomplete combustion of carbon
based resins
Carbon oxides
Nitrogen oxides
Formaldehyde
Phenol, cresols and xylenols
Aromatics (inc. polycyclics)
Aniline
Ammonia
Isocyanic acid*
Methyl isocyanate*

OIL SAND
Linseed oil and starch
Particulate matter soot from the
incomplete combustion of carbon
based resins
Carbon oxides
Butadiene
Ketones
Acrolein
Odour may be a problem
CO
2
PROCESS
Sodium silicate

Carbon oxides

SILICATE ESTER
Sodium silicate

Carbon oxides
Alkanes
Acetone
Acetic acid
Acrolein

* For resin containing nitrogen (urea)

Table 3.45: Environmental impact from binder systems upon pouring, shake-out and cooling
[160, UK Environment Agency, 2002]

Emission factors for iron casting in cold-setting sand moulds as used in Belgium are given in
Table 3.46.

Compound Average kg/tonne melted
CO 1.1
Aliphatic heterocyclic hydrocarbons 0.22
Aromatic hydrocarbons 0.05
HCN 0.03
Formaldehyde 0.02
Sulphur compounds (when using paratoluene sulphonic acid) 0.10
Volatile organophosphorous compounds (when using phosphoric acid) 0.11
Table 3.46: Emission factors for iron casting in cold-setting resin-bonded sand moulds
[110, Vito, 2001]
Chapter 3
136 Smitheries and Foundries Industry
The emission factors during pouring, cooling and shake-out for a mixed sand system, have been
determined through an intensive measurement programme in two automotive foundries in
Mexico. The foundries under study produced iron castings, using green sand moulds and
chemically-bonded sand cores. The emission factors are very process-specific and vary
according to changes in the mould or core composition, or changes in the process parameters
such as cooling time, or in the techniques applied, such as the type of shake-out device applied.
Nevertheless, the emission factors obtained provide useful information on the type of emissions
and on the relative importance of the various process steps. The emission factors for the most
important detected air pollutants are presented in Figure 3.6. The data show that the highest
emissions occur during shake-out and only minor emission occur during pouring. Naphthalene
and monomethylated naphthalenes mainly contribute to the PAH emission. Further analysis of
this category revealed that many of the carcinogenic PAHs, such as benzo(a)pyrene were not
detected.

The major metal emissions include manganese, lead, nickel, copper and chromium, with lead
and manganese being greatest. [141, CERP, 1999]

Figure 3.6: Emission levels of pollutants upon pouring, cooling and shake-out for a green sand iron
foundry
[141, CERP, 1999]

Emission factors for particulate matter are displayed in Figure 3.7. These are based on
measurements of total dust, PM
10
(particles less than or equal to 10am) and PM
2.5
(particles less
than or equal to 2.5am). Note that by definition PM
10
includes PM
2.5
. [141, CERP, 1999]

Chapter 3
Smitheries and Foundries Industry 137
Figure 3.7: Emission levels for particulate matter upon pouring, cooling and shake-out for a green
sand iron foundry
(Total = PM
10
+ PM
>10
)
[141, CERP, 1999]

Data in the above figures apply to raw gas. In an Italian sector survey, emission data for cleaned
exhaust flows were collected and are given in Table 3.47 [180, Assofond, 2002]. All exhaust
cleaning types reach a dust emission level of <20 mg/Nm. Wet scrubbers show the best
performance, though it should be noted that the data are based on only three measurements.

Flue-gas cleaning Concentration (mg/Nm) Emission factor (g/tonne)*
equipment Average Minimum Maximum Average Minimum Maximum
Bag filter 5.8 1.0 16.8 68.5 7.8 206.9
Wet scrubber 2.3 1.8 2.8 16.3 13.3 18.0
Wet multicyclone 18.6 14.6 21.9 202.2 163.6 224.4
Wet Venturi system 11.7 6.2 16.9 116.3 38.2 187.7
* Emission factor in g/tonne of good casting
Table 3.47: Emission values and emission factors for dust emission from shake-out after flue-gas
cleaning
[180, Assofond, 2002]

3.10.1.2 Sand-to-liquid metal ratios

Excessive amounts of sand in the moulding system results in unnecessary capital and operating
costs. A lower sand-to-liquid metal ratio will reduce the total volume of sand in the system and,
therefore, reduce the consumption of new materials.

The distribution of sand-to-liquid metal ratios for green sand in the iron sector is shown in
Figure 3.8. Data were collected from a survey of the UK foundry sector. While the average
sand-to-liquid metal ratio in this sector is 9:1, some foundries are operating either considerably
above or below this figure. The lower ratios are generally associated with certain types of metal
moulds or single-product foundries where box size/casting combinations are more easily
optimised. The higher ratios are generally caused either by jobbing or short-run situations. Here,
many widely varying casting configurations (and thus pattern configurations) will have been
Chapter 3
138 Smitheries and Foundries Industry
involved or the original product/product mix will have changed significantly since the plant was
designed.

Figure 3.8: Green sand-to-liquid metal ratios in iron foundries
[73, ETSU, 1995]

Figure 3.9 shows the distribution of total mixed sand-to-liquid metal ratios for green sand in the
copper sector, where the average sand-to-liquid metal ratio was about 4:1. The reason this value
is lower than that for the iron sector is largely because most copper foundries are product
orientated with an optimised box size.

Figure 3.9: Green sand-to-liquid metal ratios in copper foundries
[73, ETSU, 1995]

Data for chemically-bonded sand-to-liquid metal ratios for various metal types are given in
Figure 3.10.

Chapter 3
Smitheries and Foundries Industry 139
Figure 3.10: Total mixed sand-to-liquid metal ratios in UK foundries
[72, ETSU, 1995]

3.10.1.3 Metal yield

The metal yield is the ratio of the amount of metal melted to the weight of the finished good
castings. Five main factors affect metal yield, i.e.:
- quality requirement
- choice of mould-box size
- the extent of runner and feeder systems
- metal shrinkage
- scrap casting rate.

Metal yield does not have a direct effect on sand use. However, an increase in yield may result
in fewer moulds being produced, which means that less sand is consumed overall. Lower metal
yields are generally associated with higher integrity products, where superior quality standards
may be required, necessitating a more extensive feeding system. Lower yields, however, may
also be indicative of higher scrap rates and excessive feeding systems. In these circumstances,
foundries need to review their process control and mould production methods. The average
metal yields for the main alloy sectors are given in Table 3.48.

Chapter 3
140 Smitheries and Foundries Industry
Sector Average metal yield
(%)
Range of yields reported
(%)
Lamellar iron 68 40 90
Nodular iron 63 40 90
Aluminium 57 40 80
Copper 58 30 90
Steel 45 n.d
Table 3.48: Metal yield in the main alloy sectors
[73, ETSU, 1995], [225, TWG, 2003]

Data were collected through a survey in the UK foundry sector and from the Portuguese foundry
association (the latter for steel). A significant number of survey respondents declined to supply
yield data, particularly in the aluminium and copper sectors. However, some general comments
on metal yield can be made:
- the range of yields reported for lamellar iron and nodular iron was from 40 % to over 90 %
- metal yields reported by aluminium foundries were fairly evenly distributed between 40 %
and 80 %. Over 50 % of aluminium foundries, however, chose not to supply data
- half the small number of copper foundries that supplied data are achieving yields of
between 50 % and 60 %, though the reported range is 30 % to over 90 %.
[73, ETSU, 1995]

3.10.1.4 Used foundry sand

A review and study of foundry waste sand quality and composition was performed in Finland.
Data were collected from literature and from own experiments [169, Orkas, 2001].

In several studies, chemical analyses on used foundry sands from various sources were
performed. A summary of the most significant results is presented in Table 3.49. The data show
that the contents of metallic and organic harmful compounds in used foundry sand are normally
low. Generally, the contents of organic and metallic harmful compounds are higher in green
sands compared to chemically-bonded sands. Inorganic sands are basically very clean.

Compound Green
sand
(mg/kg)
Organic
sand
(mg/kg)
Inorganic
sand
(mg/kg)
Ba 35 118 2.4 5.5 n.d
Cr 1.7 13.5 1.2 7.2 <5
Fe 2950 21000 640 16300 530 1700
Zn 1.5 1450 1.6 49 <10 30.0
Cd 0.03 6.7 0.01 0.03 0.02
Pb 1.6 390 0.4 2.1 1.3
Cu 4.7 5.0 2.7 4.4 <1.5 6.0
Ni <2.5 20.0 0.3 8.5 2.5 8.3
Mn 76 78 22 79 25 34
As 0.2 2.1 0.2 1.8 <0.5 0.51
Phenol 1.1 29.6 0.1 14 0.03
Total PAH 1.0 206.6 0.1 8.8 <1.75
Table 3.49: Results from analyses of used foundry sands from several sources
[169, Orkas, 2001]

The Castings Development Centre (CDC) analysed the PAH-values (total PAH, naphthalene
and carcinogenic PAH) and phenol contents from different used foundry sands. The results are
shown in Table 3.50. For a given sand production line, the variations in the PAH and phenol
contents in used sand are relatively low.

Chapter 3
Smitheries and Foundries Industry 141
Polycyclic aromatic hydrocarbons, (mg/kg) Phenol
Sand system Naphthalene Carcinogenic
PAH
Total PAH Total
(mg/kg)
Leachable
(Ng/l)
Green sand <1.0 7.4 0.03 <1.0 <10 11 1.4 63 26 1600
Alkaline phenolic 1.1 4.8 0.026 0.096 2.3 8.1 1.4 210 0.025 4400
Furan 0.87 <1.0 0.014 1.5 1.0 <10 0.18 15 1.2 19
Resin shell 0.11 7.1 0.01 0.67 0.75 9.3 3.7 3300 0.025 3200
Table 3.50: PAH and phenol contents of used foundry sands
[169, Orkas, 2001]

Leaching tests showed that the leaching of metals is generally low and that the leaching of
chromium caused by the presence of chromite sand is negligible [169, Orkas, 2001].

3.10.2 Casting in permanent moulds

Input Output
- water-based release agent
- cooling water
- molten metal
- cores

casting
oily mist from release agent spraying
organic pollutants from thermal degradation
of core binder
metal oxide containing dusts from holding
furnace
dust from shake-out of cores
waste water
Mass balance data for three different foundries using pressure die-casting of molten aluminium
are given in Table 3.51.

Foundry A Foundry B Foundry C
Input
Water 802 l/tonne 935 l/tonne 1709 l/tonne
Release agent 16 l/tonne 8.26 l/tonne 1.12 l/tonne
Electricity 1103 kWh/tonne 1380 kWh/tonne 652 kWh/tonne
Output
Waste water
1
COD
pH
SS
Oil and grease
BOD5
Al

18000 mg/l
7.5
1300 mg/l
3000 mg/l
2000 mg/l
5 mg/l
122 l/tonne
126 mg/l
7.5
1 mg/l
1 mg/l (total hydrocarbons)
78 mg/l
0.6 mg/l
Exhaust gas
NOx
VOC
Dust

0.006 kg/tonne
0.28 kg/tonne
1.8 kg/tonne
0.14 0.27 kg/tonne
0.030 0.16 kg/tonne
All values per tonne of good casting or in mg/l effluent
1
For foundry C, waste water quality after treatment consisting in a distilling plant and cooling-
evaporation towers
Table 3.51: Mass balance data, exhaust and effluent composition for aluminium presure die-casting
units
[177, Silva Ribeiro, 2002], [202, TWG, 2002]

Chapter 3
142 Smitheries and Foundries Industry
In current die-casting practice, water-based release agents are sprayed on the open die, after
dilution to a 1:50 1:200 ratio. Besides die coating, this spraying is used for die cooling. This
explains the difference in water use displayed in the table. This practice causes a loss and
spillage of water, and up to 40 % of water/agent leakage into the foundry waste water system.

Emission and consumption levels for permanent moulding processes depend on the type of alloy
used, the surface area of the melt, the quantity of the sand cores which are inserted in the mould,
and the surface/volume rate of the casting. The sand cores are the main sources of emissions.
There is no important difference between the different casting techniques. Due to the wide
variety of processes and applications, it is not possible to give average data. [202, TWG, 2002]

Electricity consumption depends on process aspects, such as the closing force of the die, which
are chosen on technical grounds.

Emission factors for the casting of bronze, brass and zamac as used in Belgium are given in
Table 3.52.

Alloy CuO SnO PbO ZnO Al
2
O
3
MnO
Bronze 0.06 0.04 2 20 625 6250 n.a n.a
Brass 0.01 n.a 0.007 1 125 21500 n.a n.a
Zamac n.a n.a n.a 0.004 0.2 0.01
All data in g/tonne molten metal
Table 3.52: Indicative emission factors for casting of bronze, brass and zamac
[110, Vito, 2001]

3.11 Finishing/post-casting operations

Input Output
- unfinished castings
- abrasives
- water, soap
- electricity

finished castings
dust (sand, metal particles) from mechanical
operations
volatised metals, combustion products from
thermal operations
waste water
3.11.1 Slide grinding

Table 3.53 shows input-output data for the slide grinding of aluminium castings. The effluent
water goes to a water treatment system and is re-circulated into the slide grinding equipment.
Abrasive rocks are used and come out unaltered, besides for some minimal wearing.

Input
Electricity 20.7 kWh
Cleansing agent 0.5 l
Water 490 l
Output
Liquid effluent 400 l
All values per tonne of good casting
Table 3.53: Input-output balance for the slide grinding of aluminium castings
[177, Silva Ribeiro, 2002]

Chapter 3
Smitheries and Foundries Industry 143
3.11.2 Shot blasting

Emission levels from the exhaust gas of shot blasting have been collected in an Italian foundry
survey. The results are displayed in Table 3.54. All techniques reach maximum dust emission
levels below 30 mg/Nm. The average value is below 15 mg/Nm. The wet Venturi system is the
least effective system, leaving an emission factor well above those for the other techniques. On
the other hand, it also acts as is a safety device to prevent an aluminium-dust-explosion [225,
TWG, 2003].

Concentration (mg/Nm) Emission factor (g/tonne)
Average Minimum Maximum Average Minimum Maximum
Bag filter 5.3 0.4 19.3 53.1 0.3 327.3
Wet filtration 12.0 4.2 16.4 21.9 7.9 30.1
Wet multi-cyclone 8.2 4.2 14.8 54.9 18.5 135.4
Wet Venturi system 12.8 1.1 22.3 149.0 2.3 523.5
Emission factor data in g/tonne of good casting and for cleaned gas
Table 3.54: Emission data for the exhaust gas from shot blasting, using various dedusting
techniques
[180, Assofond, 2002]

3.11.3 Fettling

Emission levels from the exhaust gas of fettling have been collected in an Italian foundry
survey. The results are displayed in Table 3.55. All techniques achieve dust emission levels
below 30 mg/Nm. The bag filter and Venturi allow maximum levels below 10 mg/Nm. The
wet multi-cyclone is the least effective system, leaving an emission factor well above those for
other techniques. The emission factor is largely dependent on the amount of finishing needed
and thus the type of castings made.

Concentration (mg/Nm) Emission factor (g/tonne)
Average Minimum Maximum Average Minimum Maximum
Bag filter 3.0 0.4 7.5 17.0 2.3 85.6
Wet multi-cyclone 14.8 7.6 23.3 275.6 96.2 497.2
Wet Venturi system 2.5 1.1 5.9 45.5 35.5 63.4
Emission factor data in g/tonne of good casting and for cleaned gas
Table 3.55: Emission data for the exhaust gas from fettling, using various dedusting techniques,
[180, Assofond, 2002]

3.11.4 Finishing operations in steel foundries

Concerning steel foundries, specific cutting, scarfing and welding processes are used. Average
dust emission values are given in Table 3.56. These apply for the use of the following
techniques:
- Cutting (of runner system): Tetrene-acetylene-iron oxide for stainless cutting
- Scarfing: Carbon electrode covered with copper
- Welding: Using electrodes or wire.

Finishing technique Dust emission
(kg/tonne)
Tetrene cutting
Scarfing
Electrode/Wire
11 - 13
9 - 11
3 - 3.5
All values in kg per tonne of good casting
Table 3.56: Average dust emission from steel finishing
[202, TWG, 2002]
Chapter 3
144 Smitheries and Foundries Industry
3.12 Heat treatment

Input Output
finished castings
energy (electric, gas or fuel oil)
quenching: water, oil, air
finished casting (heat treated)
combustion products from firing
water vapour, fume, oil mist
NO
x
, SO
2
The emissions from heat treatment furnaces mainly comprise combustion gases, particularly
from gas- and oil-fired furnaces. The composition of the combustion gases depends on the fuel-
type used. Oil-fired furnaces will generate SO
2
-emission, which are not present for natural gas
burners. Where quenching processes are carried out, emissions of fume, water vapour, or oil
mist will also occur, depending on the quenching media.

Heat treatment furnaces are known to be continuous sources of well known non-air components.
Emission levels are closely related to energy consumption, burner design and maintenance.
Emission collection is trivial in annealing furnaces. The capture of the emissions in the different
furnaces does not differ considerably, and emissions are expelled via the waste gas pipe. In
general, no further waste gas treatment is applied.

Generally, quenching is a batch process. For this reason emission levels alternate substantially.
Emissions from the quenching baths are regarded mostly to be irrelevant for the environment,
although they may become more relevant with large scale production. Data about source
strength (emission rates) and chemical compounds have not been found.
[32, CAEF, 1997]

3.13 Waste water

3.13.1 Sources of waste water

The amount of process water used in foundries is small and mainly depends on the selection of
techniques for dedusting. Waste water is generated mainly in dust removal and waste gas
treatment systems, which are applied in the melting shop, moulding material preparation and
reclamation and in the cleaning shop. Waste water is also generated in core production, if wet
scrubbers are used (depending of the core-making technique and bonding agents used). In a
minority of plants (2 plants reported in Europe), water is used for wet sand regeneration.
Furthermore, water is generally used for cupola shell cooling and for the cooling baths for (die)
castings.

Possible sources of waste water are:
- scrap storage area and site-drainage
- sand preparation
- wet scrubbers used for dedusting in various areas of the foundry
- cooling of pressure die casting machines and tools
- vibratory finishing (slide grinding)
- cooling baths used for heat treatment.

The production-specific waste water quantity averages 0.5 m
3
/t good castings.
[160, UK Environment Agency, 2002], [195, UBA, 2003]

Chapter 3
Smitheries and Foundries Industry 145
3.13.2 Waste water from scrap storage

Scrap may have adhering soil, and external scrap typically exhibits production-specific
impurities. The adhering impurities may be washed off when it rains and may then enter the soil
and groundwater. An overview of the possible impurities is given in Table 3.57:

Scrap type Adhering impurities
Swarf Cutting oils, emulsions (may contain chlorine)
Punching scrap Punching oils (may contain chlorine)
Broken machine castings,
not conforming to specifications
Hydraulic oils, gear oils
Scrap from cold forming and smitheries Phosphates, zinc soaps, graphite, forming oils
Table 3.57: Possible impurities for specified scrap types
[195, UBA, 2003]

If the adhering impurities are substances constituting a hazard to water (oils, emulsions), then
requirements on the storage of substances constituting a hazard to water must be complied with.

3.13.3 Waste water from wet scrubbers used in cupola melting

Wet scrubbers are in use in melting shops to clean the cupola off-gas. Wet scrubbers use water
to remove dust particles (average loading: 10 15 g/Nm
3
) and gases such as sulphur dioxide
from cupola off-gas. In the settling tank, a large part of the solid particles are captured and
thereby almost completely removed from the scrubbing water.

The partly acid gases washed out of the cupola off-gas, such as sulphur dioxide, accumulate in
the scrubbing water, causing its salt concentration to increase (e.g. the formation of e.g. sodium
sulphate) and its pH to decrease. This salt accumulation is supported by evaporation losses, so it
is necessary to draw the water off from time to time.

Waste water from wet dedusting systems in melting shops mainly contains:
- solids such as oxides of silicon, iron and aluminium, calcium carbonates and cyanides
- heavy metals, although with the exception of zinc, these are present only in very low
concentrations
- organic pollutants, which can enter waste water via soiled scrap.

AOX (absorbable organic halides) levels in water used for the scrubbing of cupola off-gas can
amount to several milligrams. Possible causes of AOX are scrap coatings and adhering
chlorinated organic compounds entering the water during the gas scrubbing process.
Chlorinated organic compounds can also reach the water via auxiliaries, such as commercial-
grade hydrochloric acid or iron and aluminium chlorides, which are used as flocculants.

Table 3.58 gives an overview of the pollutant concentrations in filter effluent from the
dewatering of sludge from the wet scrubbers of cupola furnaces.

Chapter 3
146 Smitheries and Foundries Industry
Analyte Units Value
pH value 7.2 - 9.9
El. conductivity S/cm 1400 - 18400
Lead mg/l <0.01 - 2.5
Cadmium mg/l <0.01 - 0.03
Chromium mg/l <0.01 - 0.13
Copper mg/l 0.02 - 0.89
Nickel mg/l 0.04 - 0.23
Mercury mg/l < 0.001
Zinc mg/l 1.8 - 27.9
Sulphate mg/l 430 - 1550
Chloride mg/l 1330 - 3947
COD mg/l 154 - 7580
Table 3.58: Pollutant concentrations in filter effluent from dewatering the sludge from the wet
scrubbers of cupola furnaces
[195, UBA, 2003]

3.13.4 Waste water from the casting, cooling and shake-out area and
from mould production/sand preparation

In the casting, cooling and shake-out area and in mould production/sand preparation, insoluble
fines from the moulding material are captured during dedusting, along with small organic
proportions from the bonding agent used. If wet dedusting is applied, these compounds are
found in the waste water flow. The inorganic solids are iron oxides and clays, which are in part
finely dispersed and difficult to remove.

3.13.5 Waste water from core-making

In core-making shops, chemical scrubbers are in use. They use either acid (cold-box process) or
basic (SO
2
-process and Croning). The quantity of waste water discharged is dependent upon the
accumulation of pollutants and salts in the circulating water. If pollution levels are too high, part
of the polluted water must be drawn off.

Scrubbing solutions from cold-box and hot-box core-making contain readily biodegradable
amines and phenols.

The treatment of amine-containing scrubbing solutions requires a nitrification/denitrification
stage. These treatment steps can also be carried out in an external biological waste water
treatment plant. An alternative to discharge is the recovery of the amines.

Following oxidation, scrubbing solutions from the SO
2
process mainly contain sodium sulphate.
Since high sulphate concentrations (>600 mg/l) can lead to damage to sewerage systems, a limit
value is set by the respective competent authority according to local conditions (e.g. the
presence of sulphate-resistant piping and dilution with other waste water).
[195, UBA, 2003]

Chapter 4
Smitheries and Foundries Industry 147
4 TECHNIQUES TO CONSIDER IN THE DETERMINATION OF
BAT FOR FOUNDRIES

In this section, techniques for environmental protection and energy saving measures are given
for individual processing steps. The various techniques are structured into generic topics, which
allows a thematic approach to be taken for assessing the various different foundry processes and
steps.

The techniques all aim to either optimise the process or to reduce the environmental effects by,
e.g.:
- careful selection and optimisation of each unit operation i.e. in: storage, melting and metal
treatment, mould and core production, casting
- the reduction of emissions to air and water
- the increase in the efficient use of energy
- the minimisation and re-use of residues.

In line with the thematic approach, information has been organised in the following sections:
4.1. Raw material storage and handling
4.2. Metal melting and molten metal treatment
4.3. Mould- and core-making, including sand preparation
4.4. Metal casting
4.5. Fume, flue-gas and exhaust air capture and treatment
4.6. Waste water treatment
4.7. Energy efficiency
4.8. Sand: regeneration, recycling, re-use, disposal
4.9. Dust and solid residues: treatment and re-use
4.10. Noise reduction
4.11. Decommissioning
4.12. Environmental management tools.

In order to allow the selection of best available techniques, all techniques to be considered in the
BAT decision are presented according to a standard structure. The items of this structure are
explained in Table 4.1.

Name of the type of information Type of information included
Description Technical description of the technique
Achieved environmental benefits The main environmental impacts addressed by the technique.
Operational data Data on consumption and emission levels from operational plants
using the technique. Any other information on how to operate,
maintain and control the technique
Cross-media effects Any side effects and disadvantages to other media, caused by
implementation. Environmental effects of the technique in
comparison with others
Applicability Indication of the type of plants in which the technique may be
applied, considering e.g. plant age (new or existing), plant size
(large or small), techniques already installed (i.e.furnace type,
moulding technique, ) and type of product (ferrous, non-ferrous)
Economics Information on costs (both investment and operational) and
possible savings, including details on how these costs have been
calculated
Driving force for implementation Local conditions or requirements which lead to or may stimulate
implementation. Information on reasons other than environmental
ones for implementation (e.g. increase in productivity, safety)
Example plants Reference to plants in which the technique is applied and from
which information has been collected
Reference literature Literature which was used in writing the section and which contains
more details
Table 4.1: Information contained in the discussion of each technique included in Chapter 4
Chapter 4
148 Smitheries and Foundries Industry
This chapter presents the information for each technique. The later balancing and selection of
techniques will be dealt with in Chapter 5, but based on the information of Chapter 4.

4.1 Raw material storage and handling

4.1.1 Introduction

The storage and handling of solid materials, gases and liquids is discussed in the Storage BREF
[205, European IPPC Bureau, 2003]. The document discusses available techniques for e.g.
heaps, silos, tanks and packaged materials (e.g. dangerous and flammable substances). These
are applicable for foundry raw materials such as sand, scrap, coke, fuels (gas, oil), chemicals,
additives and for residues such as used foundry sand, filter dust and slags. In this section, we
will only discuss foundry-specific issues which are not covered by the storage BREF.

4.1.2 Covered storage and impermeable scrap yard

Description
The scrap storage area can be structured and managed in such a way that the following factors
are taken into account:
- the composition of the furnace charge necessitates knowledge of the feed material.
Applying separate storage for various metal types or grades allows control of the charge
composition. This can be achieved using compartments or boxes in the storage area or
bunker
- the introduction of mineral and oxide materials such as rust, soil or dirt causes increased
refractory wear. The use of a cemented scrap yard prevents the entrainment of soil, dirt or
water
- a roof constructed over the storage area can help keep rainwater out and can help contain
dust emissions.
- a water collection and treatment system can be used to prevent pollution to soil or water.

Achieved environmental benefits
The use of a roof and a reinforced yard allows the collection of run-off water and therefore
prevents uncontrolled emissions to water and soil. Direct emissions to soil are also minimised
since mixing of the material with soil is prevented.

Cross-media effects
No cross-media effects apply.

Applicability
The use of an impermeable scrap yard can be applied in the storage of scrap in all new and
existing foundries. Scrap is stored under a roof or with the provision of a water collection and
treatment system to prevent water and soil pollution.

Driving force for implementation
Control of the charge composition improves control of the furnace operation and hence the
metal composition. Also allows the minimisation of refractory wear.

Example plants
This technique is applied in the majority of the foundries.

Reference literature
[202, TWG, 2002]

Chapter 4
Smitheries and Foundries Industry 149
4.1.3 Storage measures for binder chemicals

Description
Binder chemicals carry specific manufacturers recommendations for storage. Failure to comply
with these recommendations will result in unusable or sub-standard products that either require
disposal as special waste or lead to poor casting quality due to sub-standard moulds/cores. Table
4.2 summarises the quality problems arising from the incorrect storage of liquid binder
chemicals. Correct storage takes account of these problems.

Problem Effect on product Implications Effect on castings
Exposure to excessive
cold
Aqueous-based
products may freeze
Product segregation
may occur, causing
mould failure. Affected
products require
disposal
Casting production may
not be possible
Exposure to excessive
heat and sunlight
Premature ageing,
cross-linking of resin
and increased viscosity
Poor mixing properties,
low mould strengths
and poor resistance to
high temperatures
Fining defects
Prolonged storage or
cold storage
Increased viscosity Dispersion onto sand
grains difficult.
Sand mixture flows less
well and is difficult to
compact. Low-strength
moulds
Erosion, exogenous
(sandy) inclusions
finning and dimensional
faults
Contamination by
moisture
Products containing
isocyanates will
deteriorate due to
reaction with water
Binder performance is
poor and product may
need disposal
Increased risk of gas
defects (pinholes)
Sediment disposal in
bulk silicate storage
tanks
Pumped liquid may
become inconsistent
Low-strength moulds Oversize castings and
shrinkage effects
Table 4.2: Problems arising from an incorrect storage of liquid binder chemicals
[71, ETSU, 1998]

Some basic measures include:
- roofed and ventilated area
- collection of spilled liquid
- locked storage area.

Depending on the climate, exposure to excessive cold or heat, and sunlight can be accounted
for. Additional precautions are necessary for the storage of highly flammable liquids such as
methyl formate, triethylamine (TEA), dimethylethylamine (DMEA) and mould coatings
containing isopropyl alcohol.

Achieved environmental benefits
Reduction of the amount of waste chemicals arising, which are unsuitable for use.

Most binder chemicals are hazardous, and are characterised by one or more of the following
properties: toxic, corrosive, flammable. These properties mean that even small spills can pose a
danger to workers health and safety, while larger releases can result in a serious incident. A
major spill that is allowed to enter a surface water drain can cause serious pollution of
watercourses.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to all new and existing installations.
Chapter 4
150 Smitheries and Foundries Industry
Driving force for implementation
Safety measures and optimisation of foundry operations.

Example plants
This technique is applied in the majority of the foundries.

Reference literature
[75, ETSU, 1996]

4.1.4 Use of clean scrap for melting and removal of sand from the
return material

Description
Melting clean scrap prevents the risk of non-metal compounds being taken up by the slag and/or
attacking the furnace lining. These compounds are typically lime, iron oxides, manganese
oxides, and basic oxides (e.g. MgO from nodular iron returns) in combination with silicon
refractory (acid). If the amount of contamination is limited, this will reduce the amount of slag
formed and will allow a longer lifetime of the furnace and ladle lining. For nodular iron
production using a furnace with acid refractory, the adhering silica sand may have a beneficial
effect, since it neutralises the MgO coming from foundry returns.

The presence of contaminants and oxides in the furnace load, will consume part of the melting
energy. Furthermore, the removal of slags necessitates a higher bath temperature to keep the
slags in a liquid state.

Internally recycled scrap consists of feeders systems and rejected casting. Feeder systems are
knocked off after shake-out. In general they do not retain adhering sand, due to their geometry.
Rejected castings are excluded upon quality control, which is carried out after the castings are
cleaned using sand blasting or shot blasting to remove all adhering sand. The use of sand-free
return material therefore, in general, necessitates no extra treatment.

Achieved environmental benefits
This technique reduces the amount of slags and dust that need disposal and limits VOC-
emissions. The energy consumption is reduced (by 10 to 15 %) due to the reduced amount of
slag. Additionally the extracted airflow may be reduced.

Shot blasting of the scrap can be recommended for scrap where its contamination would lead to
the production of excessive amounts of slag. The use of only shot-blasted scrap would actually
require slag-forming additives to create the slag.

Cross-media effects
If foundries only accept clean scrap, the overall recycling rate for dirty scrap will reduce. This
will generate a need for additional cleaning operations and increase the disposal of scrap.

The use of cleaning techniques to remove sand from return material consumes energy. This is
balanced however by the energy gained in the melting stage.

Operational data
An energy consumption to clean 30 - 40 kg of sand from 1 tonne of casting of 12 - 15 kWh has
been reported for an operational foundry. The consumption level is dependent on the size and
type of casting.

For a steel foundry using EAF melting, an increase in molten metal recovery (poured
metal/charged metal) of 2 to 3 % has been reported.

A typical metal charge for EAF is: 55 % foundry steel scrap; 40 % internal returns; 5 % metal
correction additions.
Chapter 4
Smitheries and Foundries Industry 151
Applicability
The removal of sand from internal scrap can be applied in all new and existing foundries. The
use of clean scrap for melting can be applied in all foundry types, but the choice of the scrap
grade should be related to the techniques installed for melting and gas cleaning. If the whole
sector shifts to clean scrap, this will generate a problem for the recycling of dirty scrap.

The use of clean ferrous scrap (both steel and pig iron) will lead to increased purchase costs and
will require investment in new cleaning equipment. There are no difficulties caused by scrap
surface oxidation nor by using pig iron ingots during melting in cupola furnaces (reduction
process). Electric furnaces will have difficulties only with contaminated scrap which does not
comply with the parameters for steel scrap used in foundries.

Cupola furnaces can easily melt scrap which is not clean. There is little coke overconsumption
and little environmental consequence, if the gases are correctly treated.

Limiting the slag is very important for good operation of the coreless induction furnace. The
operation is more affected by the scrap cleanliness than the other furnace types.

The use of dirty scrap has no adverse effects on rotary furnace operation but is the main
determining factor for dust emissions. There is little environmental consequence if the gases are
correctly treated.

Economics
The price for clean scrap (i.e. class 1) is 20 to 30 % higher than that for contaminated (i.e.
class 2) scrap. Furthermore, disposal costs for slags and dust are reduced by applying this
technique.

Driving force for implementation
High disposal fees for residues and high prices for the refractory.

Example plants
This technique is commonly applied in foundries.

Reference literature
[103, Vereniging van Nederlandse Gemeenten, 1998], [110, Vito, 2001], [202, TWG, 2002]

4.1.5 Internal recycling of scrap ferrous metal

Description
Internal scrap is produced from the knock-off of feeders and runners, in quality control and in
finishing operations. The relative amount of internal scrap produced can be calculated from the
metal yield as defined and described in Section 3.10.1.3. In order to minimise residue
production, internal scrap is returned to the scrap metal feed of the furnace.

For steel foundries, a 100 % use of return metal is uncommon, due to the amount of dissolved
gas. Operators consider 60 % to be the maximum amount of return metal (feeders, scrap
castings,...) in the charge metal. Swarf and turnings may cause excessive molten metal
oxidation.

Internal scrap recycling may also be limited in the case of nodular base iron melting, as a high
amount of silicon (about 1 % in most cases) is added in the liquid iron during the metal
treatment. In some cases re-melting the whole internal scrap is not possible because the final
silicon would be too high (typically 2.5 3 % in the castings).

Chapter 4
152 Smitheries and Foundries Industry
In the case of grey or nodular iron, lead or bismuth are very noxious for the properties of the
metal after solidification. If there has been a pollution incident the returns (internal scrap) must
not be recycled.

Achieved environmental benefits
Minimisation of residues through recycling of raw materials.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to all new and existing installations.

Economics
The technique does not involve any additional costs.

Driving force for implementation
Minimisation of residues, optimal use of metal.

Example plants
The technique is applied in all European foundries.

Reference literature
[202, TWG, 2002]

4.1.6 Internal recycling of magnesium scrap

Description
In cell recycling:
Clean feeders can be directly melted in the melting furnaces. Because of oxides and other
inclusions, the possible amount of recycling is limited. Specialised equipment for exact
analytical and metallographical control is necessary.

In house recycling:
Here the scrap is treated and recycled at a separate recycling-plant on-site at the foundry. Two
techniques may be applied:
- Fluxless re-melting: Re-melting of Mg scrap under a covering gas in a hearth type furnace;
applicable only to class 1 scrap. Advantages are the low additional investment and low
energy consumption
- Salt re-melting: Re-melting of Mg scrap under a salt cover. Is applicable to the processing
of all scrap classes except used fluxes and sludge (not treated). If only class 1 2 scrap is
used, HP (high purity) magnesium alloys may be produced.

Both techniques can be used to produce Mg-ingots or molten metal for liquid charging.
Both in fluxless and salt-covered re-melting, dross (the floatings on the melt) and sludge (the
sinkings in the melt) are generated, as well as residues retaining a residual metal content
(70 - 80 % Mg for sludge, 60 90 % Mg for dross). For recovery of the metal content, three
options exist:
- salt re-melting (in the same or a separate furnace)
- re-melting in the aluminium industry
- utilisation in the desulphurisation of iron or steel.

If the scrap is re-melted in a fluxless melting furnace, the salt re-melting of dross and sludge
involves the operation of a separate furnace.

Chapter 4
Smitheries and Foundries Industry 153
The re-melting of mixed class scraps is performed using salt re-melting. The mixed scrap needs
pretreatment. Table 4.3 gives a survey of the input-outputs of a pretreatment installation.

Inputs Equipment Outputs End-of-pipe
technology
- oily or wet Mg-turnings (>2 %
oil/water)
- electricity
Centrifuge

- Mg-turnings (<2 %
oil/water)
- oil/water-emulsion

- centrifuged and all other
turnings
- electricity
Turningpress
(T = 400 C)
- pressed turning
- exhaust air
Particle separator
for the vaporised oil
- feeders, bad castings
- dross (without salt, from the
foundry process)
- electricity
Shredder
(if necessary)
- shredded material
- dust

Table 4.3: Inputs and outputs for a magnesium scrap treatment plant
[202, TWG, 2002]

The recovery of metal from the re-melting salt residue is done using dry crushing and screening
with magnetic separation or by using a wet washing system. The wet system produces a
magnesium fraction and a sludge, which, after dewatering, may be used in fertiliser industry.

Achieved environmental benefits
The main benefit is the optimised recycling of magnesium. Internal recycling has the added
benefit of eliminating transport to an external recycling plant. The metal efficiency of the
recycling is increased by type-specific melting.

Cross-media effects
Fluxless re-melting involves the use of sulphur containing protective gases, which contribute to
global warming (SF
6
) or are poisonous (SO
2
). This issue is discussed in 4.2.7.1.

Operational data
Experimental mass flow diagrams for the re-melting of class 1 scrap are given in Figure 4.1 and
Figure 4.2. These include the internal re-melting of sludge and dross.

Figure 4.1: Mass flow diagram for the fluxless re-melting of magnesium class 1 scrap
[206, Ditze and Scharf, 2000]
Chapter 4
154 Smitheries and Foundries Industry
Figure 4.2: Mass flow diagram for the salt re-melting of class 1 magnesium scrap
[206, Ditze and Scharf, 2000]

Operational data from a 2000 tonne/yr salt re-melting plant for mixed treated scrap (3 furnaces
with a capacity of 600 kg each) give consumption levels of salt: of 3.4 kg/tonne scrap and
natural gas: of 26.5 Nm/tonne scrap.

The wet treatment of the salt fraction produces a magnesium fraction (43 %) and a sludge
(57 %). The sludge is dewatered using a membrane filter press and the water is re-circulated
into the process.

Applicability
This technique is applicable for new and existing magnesium foundry installations.

Economics
Costs for external recycling are estimated to be approximately EUR 1200/tonne, depending on
the type of transport used, the transport distance and the local market. The costs for in-house
recycling are EUR 500/tonne, a difference of EUR 700/tonne. For a single plant with 1500
tonnes/yr recycling material this means a potential saving of more than one million euros per
year.

The given data are general and have to be adapted individually for each enterprise. Depending
on the products and the processes used, the expenses needed for separation of the waste have to
be taken into account. The waste has to be collected specifically for every alloy and every
procedure.

Further factors of major influence are:

- depreciation (approximately 10 % in a depreciation period of 5 years)
- personnel costs (35 40%)
- cost for new material (approximately 30%) to substitute the material loss (estimated 7%)
- costs for energy, maintenance, spare parts, salt, disposal (20 25%).

The amortisation of a flux-based facility with two furnaces is in the range of 8 to 11 months.
The amortisation of a continuous facility with 500 kg magnesium per hour will likely take a few
months more.

Driving force for implementation
Optimisation of magnesium use, with a reduction of the amount of magnesium sent for disposal.

Chapter 4
Smitheries and Foundries Industry 155
Example plants
TCG Unitech, Kirchdorf/Krems (A): uses an in-house recycling facility.
No fluxless re-melting plants have been reported in Europe, although the melting technology is
available on the market.

Reference literature
[202, TWG, 2002], [206, Ditze and Scharf, 2000], [223, Rauch, et al., 2003]

4.1.7 Recycling of used containers

Description
Suppliers of chemicals and additives may take back their empty containers (plastic, wood,
metal) for recycling. The use of the largest size container possible may be considered.

Achieved environmental benefits
Prevents waste and further stimulates recycling.

Cross-media effects
If containers are returned without further cleaning, no cross-media effects apply.

Applicability
This technique is applicable in all new and existing installations.

Economics
Economic data differ from site to site and depend on the conditions agreed with the supplier.

Driving force for implementation
Prevention and recycling schemes for waste and packaging waste regulations.

Example plants
This technique is commonly applied in European foundries.

Reference literature
[110, Vito, 2001]

4.2 Metal melting and molten metal treatment

4.2.1 Cupola furnaces

In this section, techniques concerning melting practices and furnace operation will be discussed.
These techniques may apply either to cold or hot blast operation, or to both. Flue-gas related
techniques, such as post combustion and flue-gas cleaning, are discussed in Section 4.5.2. That
section also discusses the conversion of cold blast to hot blast cupola furnaces.

4.2.1.1 Optimisation of furnace operation

Description
The amount of dust and exhaust gases resulting are directly related to the amount of coke
charged per tonne of iron. Therefore, all measures that improve the thermal efficiency of the
cupola will also reduce the emissions from the furnace.

Chapter 4
156 Smitheries and Foundries Industry
Possible good melting practice measures include:

- Operating the furnace in its optimum regime as much as possible: For any given cupola, the
influence of the blast rate and the coke charge on the tapping temperature of the metal and
the melting rate, can be expressed in the form of a net diagram or Jungblth diagram. A net
diagram is only valid quantitatively for the cupola for which it was obtained. It displays
how the metal temperature and melting rate behave upon changes in the blast rate and coke
charge, and allows determination of the point (or line) of optimal thermal efficiency
- Avoiding excess temperatures of the melt and reducing the superheating temperatures by
taking measures during the handling of tapped metal
- Uniform charging: taking care during charging to get a uniform distribution of metal and
coke
- Improving the control of the charge weight, the blast airflow and the metal temperature
- Minimising air losses: The correct supply of air is essential for efficient cupola operation.
This is often hampered by air losses. Therefore, it is important to pay close attention to the
prevention of all leaks to ensure efficient operation. Slag holes on intermittently tapped
cupolas are often left open and are frequently oversized, resulting in considerable air losses.
Air losses are particularly common for hot blast cupolas. They take place in the heat-
exchanger. The consequence is that oxygen must be injected in the tuyres instead of the air
which is lost
- Avoiding bridging in the cupola: Bridging and scaffolding are the hanging or non-descent
of cupola charges in the cupola shaft. This causes a loss of melting efficiency, and in severe
instances melting can be completely stopped
- Utilising good lining practice: As melting proceeds, the diameter and the area of the melting
zone increase due to lining erosion and wear. This affects the operation, moving it away
from the optimum. Minimising the lining attack is therefore an energy saving measure. For
satisfactory and economic cupola operation the melting zone needs to be efficiently repaired
after each melt.

Achieved environmental benefits
Increased energy efficiency, reduction of cokes consumption, reduction of residues.

Cross-media effects
No cross-media effects are reported

Applicability
This technique applies to all new and exisiting cupola furnaces.

Driving force for implementation
Optimisation of furnace operation.

Example plants
Good melting practice measures are applied in all cupola foundries.

Reference literature
[32, CAEF, 1997], [44, ETSU, 1993], [202, TWG, 2002]

4.2.1.2 Quality control of coke input

Description
The quality of the coke used has a direct bearing on the efficiency of cupola operations. It
particularly affects the initial temperature carbon pick-up and the sulphur content of the iron.
Specifying the foundry coke involves testing the following contents:

Chapter 4
Smitheries and Foundries Industry 157
- Fixed carbon: The higher the carbon content, the higher the calorific value
- Ash content: A high ash content is undesirable since it lowers the calorific value of the coke
and generates a greater volume of slag in the cupola
- Volatile matter: Volatile matter is undesirable since it reduces the fixed-carbon content, and
thus the calorific value of the coke
- Sulphur: Sulphur is well known as an unwanted element in any type of cast iron and leads
to SO
2
emissions. The lower the sulphur content of the feedstock coke, the better. The
sulphur content of the coke depends on the sulphur content of the feedstock coal.
Unfortunately there is no known method to remove sulphur from coal
- Moisture: Moisture in the coke when dispatched from the coke oven is undesirable, since it
reduces the amount of carbon available by weight. However, it is necessary for the coke to
contain some moisture, to avoid fires on conveyer belts and in lorries and wagons
- Size: The size of the foundry coke directly affects the coke consumption per tonne of iron
melted and also the melting rate. To achieve optimum performance the coke size dispatched
from the coke oven generally needs to be sized such that the furnace coke has a mean size
diameter greater than 90 mm, with no more than 4 % smaller than 50 mm. The content of
fines will influence emissions of dust during (un)loading and handling.

Achieved environmental benefits
Optimising the coke input results in a higher process efficiency.

Cross-media effects
No cross-media effects apply.

Operational data
Typical properties of foundry coke are given in Table 4.4:

Property Typical value (%) Limit value (%)
Ash 5.8 5.9 6.5 max.
Volatile matter 0.3 0.4 0.8 max.
Sulphur 0.68 0.70 0.75 max.
Moisture 1.5 2.5 3.0 max.
Strength
M80 Micus index

81 82

78.0 max.
M10 Micus index 8 8.5 9.0 max.
Fixed carbon 93.7 93.8 93.0 min.
Higher calorific value 32200 kJ/kg 31800 kJ/kg
Table 4.4: Typical properties of foundry cokes
[197, Nalonchem, 2002]

The data in Table 4.4 are taken from supplier specifications. Local standards may use higher
limit values.

Applicability
The technique is applicable to all new and existing cupola installations.

Economics
In general, low sulphur feedstocks are more expensive.

Driving force for implementation
To improve the efficiency of the foundry process.

Example plants
This technique is commonly applied in European foundries using cupola furnaces.

Reference literature
[44, ETSU, 1993], [225, TWG, 2003]
Chapter 4
158 Smitheries and Foundries Industry
4.2.1.3 Operation with acidic or basic slag

Description
A flux is used to render the slag fluid enough to separate from the iron and to allow it to flow
freely from the cupola. The most commonly used flux is limestone (calcium carbonate), which
calcines in the cupola shaft to form lime, a basic oxide, which then combines with the other
slag-forming constituents (mainly acidic in character) to provide a fluid slag.

The basicity of a slag is given by the following ratio: (CaO % + MgO %)/SiO
2
%
Most cupolas operate with an acidic or slightly basic slag (basicity <1.2). Basic cupolas
(basicity up to 2) offer 3 advantages:
- higher carbon
- lower sulphur
- possibility of charging lower quality scrap.

But they present the following disadvantages:
- silicon losses are high
- refractory costs are high, unless operating with a liningless cupola
- fluxing material costs are higher
- metal analysis is more difficult to control than acid cupola melting.

Achieved environmental benefits
For acidic cupolas, dry filters can be used. The slag of basic cupolas has a higher melting point.
So a flux based on CaF
2
is generally used to make it fluid. In this type of cupola, there are
emissions of fluor-containing components. This necessitates the application of wet scrubbers for
efficient capture of these components.

Applicability
This technique applies to all new and existing cupola installations. If basic slag is used, a wet
dedusting is necessary.

Example plants
This technique is commonly applied in European foundries using cupola furnaces.

Reference literature
[44, ETSU, 1993], [202, TWG, 2002], [225, TWG, 2003]

4.2.1.4 Increasing shaft height of CBC

Description
The required shaft height for cupolas at various melting rates is given in Table 4.5. These shaft
heights optimise the preheat for the descending burden. However, if the gas is to be burned at
the charge-hole, a shorter shaft height may be considered. In general, the shorter the shaft, the
hotter the top gas and the greater the ease of combustion either spontaneous or assisted by an
afterburner.

Melting rate of Cupola
(tonne/h)
Height from tuyres
to charging door sill
(m)
Up to 5 4.9
5 8 5.8
>8 6.7
Table 4.5: Shaft height requirements

Chapter 4
Smitheries and Foundries Industry 159
The thermal efficiency of the cold blast cupola furnace can be improved by increasing the shaft
height. In general, the higher the furnace shaft, the longer the combustion gases remain in
contact with the charge, and the more heat is transferred to the charge.

Achieved environmental benefits
Depending on the installation type, the height must be optimised to allow burn-out of the gases
and efficient heating of the charge.

Cross-media effects
No cross-media effects apply.

Operational data
Operational data are given in Table 4.6. The modification involves a doubling of the upper zone
volume. This results in a reduction of coke consumption from 140 to 115 kg/tonne, which is a
relative reduction of 18 %. It should be noted that in general an optimal shaft height follows the
rule-of-thumb: height = 5 x diameter at tuyres. The initial situation in the example furnace
was therefore sub-optimal.

Cupola
before modification
Cupola
after modification
Diameter
- melting zone (m) 1.4 1.4
- upper zone (m) 1.4 1.7
Heigth over tuyres (m) 5 6.5
Coke ratio (kg/tonne) 140 115
Table 4.6: Example data for the change in coke consumption upon shaft height increases
[202, TWG, 2002]

Applicability
Optimisation of the furnace height needs to be performed at the design stage of the furnace,
otherwise height increases will usually only be performed during major re-building works of the
furnace.

Driving force for implementation
To increase the efficiency of the furnace operation.

Example plants
Operational data are taken from an example plant in France.

Reference literature
[32, CAEF, 1997], [44, ETSU, 1993], [202, TWG, 2002]

4.2.1.5 Installation of secondary row of tuyres for a cold blast cupola
(CBC)

Description
The thermal efficiency of the cold blast cupola furnace can be improved by the installation of a
secondary row of tuyres. These provide extra oxygen above the combustion zone, which
induces the oxidation of the CO in the combustion gases, the CO being formed by the
endothermic reduction of CO
2
by C (coke). This technique liberates the latent heat of the
combustion gases, thus improving the thermal efficiency of the cupola.

The divided blast cupola is equipped with two rows of tuyres, each supplied with a measured
and controlled quantity of blast air. Compared with a normal cupola, fitted with a single row of
tuyres, the divided blast cupola enables:
Chapter 4
160 Smitheries and Foundries Industry
- a higher metal tipping temperature and higher carbon pick-up to be obtained for a given
charge-coke consumption
- the charge coke consumption to be reduced and, if required, the melting rate to be increased,
whilst maintaining a given metal tapping temperature.

To obtain the maximum benefit from divided blast operation, the blast should be divided
25 - 50 % top and 75 50 % bottom. The 2 rows should be spaced about 1m apart (cold blast)
to 0.5 m (hot blast). Each row of tuyres needs to be provided with its own blast supply system.

Achieved environmental benefits
A reduction in coke consumption and an increase in thermal efficiency.

Cross-media effects
No cross-media effects apply.

Operational data
By the use of two rows of correctly-spaced tuyres with the blast equally divided between them,
the metal tapping temperature can be increased by approximately 45 50 C for a given coke
consumption. Alternatively, the coke consumption can be reduced by 20 32 % and the melting
rate increased by 11 23 %. When operating with a divided blast but without reducing the coke
charge, so that a higher melting temperature is obtained, the carbon pick-up tends to increase
slightly (by approx. 0.06 %), and the melting loss of silicon tends to also increase, by
approximately 0.18 %.

With divided blast operation, the lining burn-out extends further into the furnace shaft. At the
start of a melt it is therefore necessary to measure and adjust the coke bed height. On melts of
short duration, i.e. those less than about 2 3 hours, the saving in coke charge does not usually
compensate for the additional requirements of bed coke. Nevertheless, even on short melts, the
higher tapping temperature and the higher carbon pick-up obtainable with divided blast
operation can be an advantage to many foundries.

Other reported advantages are:
- the exit gas temperature is only 250 C, compared to the conventional cupola, where the
temperature is 450 C
- it can take 100 % bigger chunks of re-melting scrap
- it is possible to increase the steel scrap ratio in the metal charge.

Additional operational data are given in Annex 1.

Applicability
The second row of tuyres is applied as a standard technique for new cold blast cupola
installations and may be applied to existing installations during reconstruction. This technique is
not normally applied for hot blast operation

Economics
The divided blast cupola has been well proven in practice as a means of obtaining economic
operation for a modest investment cost.

The capital cost of converting existing cupolas to divided-blast operation is low compared to the
savings obtained. BCIRA quoted one British foundry where the payback period was only
fourteen weeks. A large Canadian foundry saved CAD 170000 in one year for a conversion cost
of only CAD 18000. An additional benefit from a lower coke use per tonne of iron is the lower
sulphur content, this saves on the desulphurisation plant and gives a higher quality iron.

Additional economic data are given in Annex 1.

Chapter 4
Smitheries and Foundries Industry 161
Driving force for implementation
To increase the efficiency of the melting operation.

Example plants
In France, all the recent cold blast cupolas use 2 rows of tuyres: STAUB (Merville), FONTE
ARDENNAISE (Vivier au court), BERNARD HUET (Vivier au court). There is also 1 hot blast
with 2 rows: FIDAY GESTION (Chassey les Scey)

Reference literature
[32, CAEF, 1997], [196, Unido, 2002], [202, TWG, 2002], [225, TWG, 2003]

4.2.1.6 Oxygen enrichment of blast supply

Description
The thermal efficiency of the cold blast cupola furnace can be improved by oxygen enrichment
of the combustion air. This gives rise to a higher combustion temperature of the coke. In this
way coke consumption can be decreased or higher temperatures of the tapped metal can be
obtained.

Compared with normal operation, the continuous use of oxygen enables one of the following
improvements to be made:
- a higher metal temperature, higher carbon pick-up and lower melting loss of silicon for the
same coke consumption, or
- a lower coke consumption for a given metal temperature, with no increase in carbon pick-up
or reduction in the melting loss of silicon, or
- an increased output from an existing cupola, as a result of the increased melting rate.
Oxygen injection provides the possibility of reacting quickly to process drifts, and then
compensating for small changes in process conditions. Thus, oxygen injection is often used
intermittently, usually whenever the process requires adjustment.

The effectiveness of the oxygen depends on the method by which it is introduced into the
cupola. Three processes have been developed:
- Direct enrichment of the blast supply: Oxygen is fed into the main blast, this technique is
applied in the majority of the cold blast cupolas
- Injection into the well: Oxygen is injected into the coke bed from a ring-main which
supplies water-cooled injectors, whose number depends on the cupola size. Oxygen used in
this way is at least twice as effective as when it is used to enrich the blast supply directly.
However, this type of injection is confined to continuously-tapped cupolas, since with
intermittent tapping there is a risk that slag and or metal may rise to the level of the
injectors. The technique was developed in the UK but it has not found widespread
application
- Injection at the tuyres: Oxygen is introduced into the cupola through injectors inserted into
each tuyre or into alternate tuyres. The effectiveness of this method lies between direct
blast injection and well injection methods. This technique is used in 20 30 % of cases, but
more in hot blast operation.

The principles of oxygen injection are depicted in Figure 4.3.

Chapter 4
162 Smitheries and Foundries Industry
Figure 4.3: Various methods of oxygen injection
[44, ETSU, 1993]

Achieved environmental benefits
The application of oxygen results in a reduced coke consumption and a better process control.
Additionally, a reduction in the emission of dioxins and furans from cold blast cupolas have
been reported (see Section 4.5.1.4).

Cross-media effects
The technique requires oxygen which is produced off-site and involves a consumption of
electric energy.

Operational data
Compared to conventional cold blast operation with one row of tuyres, the increase in tapping
temperature obtained for a given charge of coke, can be expected to be as follows:
- blast enrichment + 15 C
- well injection + 85 C
- tuyre injection + 40 C

Divided blast operation with a direct enrichment of blast air, results in an increase in the tapping
temperature of 85 C compared to cold blast operation. In divided blast operation, neither well
injection nor tuyre injection give any greater benefit than the simple method of direct injection.

An important application of oxygen has been to increase the melting rate of an existing cupola,
well beyond its usual optimum melting capacity. In divided blast cupolas, the melting rate
increases by around 6.8 % and the tapping temperature by about 20 C for each 1 % of direct
oxygen enrichment in the blast air. If an increase in temperature is not required and the coke
charge is reduced, an even greater increase in the melting rate can be obtained for the same blast
rate and oxygen enrichment.

Additional operational data are given in Annex 1.

Applicability
Today, almost all European hot blast cupolas inject oxygen through the tuyres. For cold blast
furnaces, the use of oxygen enrichment can be considered as the standard technique. In this
case, enrichment of the blast supply is usually applied. The oxygen level of the oxidising air
mixture is usually between 22 and 25 % (i.e. an enrichment of 1 % to 4 %).
Chapter 4
Smitheries and Foundries Industry 163
Economics
The effect that using oxygen has on melting costs greatly depends on the price of oxygen, which
in turn is related to the amount used. Foundries with larger outputs can generally buy oxygen
more cheaply. The economic case for using oxygen has to be established on a case-by-case
basis.

The substantial increase in outputs obtainable has enabled foundries to increase outputs without
high capital investments in new plant, and to reduce costs and overtime payments, so much so
that the overall reduction in costs amply justify the cost of oxygen. Such improvements must
take into account the related core and mould making capacity.

Driving force for implementation
To optimise the efficiency and control of the melting process.

Example plants
This technique is commonly applied in European foundries using cupola furnaces.

Reference literature
[32, CAEF, 1997], [156, Godinot, 2001]

4.2.1.7 Superheating of HBC blast air

Description
An alternative technique for increasing the flame temperature is to superheat the blast air to
800 900 C. For this, injection of an air plasma, or heating in tubular resistance heaters is
applied. Experience has shown that increasing the blast temperature by 200 C, from 550 C to
750 C, which takes 60 kWh per tonne of iron, saves 10 kg of coke per tonne melted. The main
benefit, more important even than saving coke, is the flexibility: the hourly output can be
increased by 30 % without modifying the melting bed. Additionally, the application of (plasma)
superheating allows a change in raw material from clean cast iron to steel, with a subsequent
positive economic effect.

Achieved environmental benefits
Reduced coke consumption and increased efficiency of the process.

Cross-media effects
Electrical heating causes an increase in electricity consumption (58 kWh/tonne).

Operational data
Operational data are given in Annex 1. The heating of the blast air produces a lower flue-gas
volume as compared to hot blast long campaign operation. As compared to oxygen injection,
the flue-gas volume and electricity consumption is larger.

Applicability
The technique applies to new and existing HBC furnaces.

The technique (both with electrical and plasma heating) achieves a similar effect as injection of
oxygen through the lances, but can be applied in a more complex installation and produces a
larger flue-gas volume. Additionally, oxygen injection allows make-up for leaks in the air
circuit.

Economics
Economic data are given in Annex 1.

Driving force for implementation
To optimise the efficiency and to control the melting process.
Chapter 4
164 Smitheries and Foundries Industry
Example plants
PSA, Sept-Fons, France
Three foundries in France apply tubular resistance heating.

Reference literature
[156, Godinot, 2001]

4.2.1.8 Minimal blast shut-off periods for HBC

Description
A cupola blown just intermittently will not operate efficiently, and will result in a reduced metal
tapping temperature, as shown in Figure 4.4.

Figure 4.4: Effect of blast shut-off periods on tapping temperature in a post-blast cupola
[44, ETSU, 1993]

Frequently shutting off the blast due to only intermittent demands for metal:
- reduces the average tapping and pouring temperatures and increases the extent of their
variations, with a consequent risk of producing defective castings
- increases the variation in metal composition, particularly the carbon and silicon content,
with a danger of then producing off-grade metal
- increases the coke consumption, through the operators attempt to improve the tapping
temperature
- affects the degree of nucleation in the iron and increases its shrinkage tendencies.

Moulding and casting schedules are therefore programmed to produce a reasonably constant
demand for metal, thereby minimising or even eliminating the need for blast shut-off periods or
for large variations in the blast rate.

Where large fluctuations in demand are unavoidable, the installation of an electric holding
furnace might be considered. This can provide a large buffer reservoir for metal, to take up
variations in demand, so that the cupola can be operated continuously at a reasonably consistent
blast rate. It can also be used to help even out variations in temperatures and composition.

Chapter 4
Smitheries and Foundries Industry 165
Achieved environmental benefits
Reduction of cokes consumption. Higher process energy efficiency.

Cross-media effects
Operating the electric holding furnace incurs a higher energy consumption.

Applicability
This technique is applicable to all new and existing cupola furnaces.

Economics
The economics of installing a holding furnace must be very carefully considered, particularly in
foundries with relatively low production rates.

Driving force for implementation
To increase the efficiency of the foundry process.

Example plants
The applicable blasting regime is part of the operational considerations in all European
foundries using cupola furnaces.

Reference literature
[44, ETSU, 1993]

4.2.1.9 Cokeless cupola furnace

Description
In the cokeless cupola, the metal charge is heated by the combustion of natural gas. Instead of
the traditional coke bed, a bed of refractory spheres on a water-cooled grid supports the metal
charge. The molten metal droplets run through this bed and are collected in the well at the
bottom of the furnace. The life of the spheres exposed to the superheating temperatures of the
metal is limited. A cokeless cupola is therefore run at a reduced temperature (1400 C instead of
1500 C) and superheating of the liquid metal is done in an attached gas-fired or induction
furnace (duplex operation).

Chapter 4
166 Smitheries and Foundries Industry
Figure 4.5: Cokeless cupola furnace in duplex operation
[110, Vito, 2001]

An important operational factor is that the cokeless cupola has to be run continuously. There is a
high risk of clogging and blocking on the cooled grate bars. In case there is a need to stop the
metal flow (e.g. due to problems in the moulding shop), the power of the burners can be reduced
to 35 40 %, in order to compensate for wall heat-losses. The need for continuity needs to be
balanced with the usage of refractory in the melting zone. Therefore, melting campaigns of one
week are typically used.

The oxidising atmosphere and relatively low flame temperature cause increased oxidation
losses. This limits the possibility to feed in steel. A maximum amount of 35 % of steel is used in
the production of nodular iron, though 20 % can be considered general practice. The quality of
the feed needs to be well controlled since the cokeless cupola is more susceptible to bridging
than the coke-fired cupola.

In the production of nodular iron, an important advantage of the cokeless cupola is that there is
no resulphurisation, so the melt may be used immediately after recarburisation.

Achieved environmental benefits
Besides a higher thermal efficiency, this furnace has other environmental advantages. The
combustion of natural gas instead of cokes has the following consequences for the flue-gas:

Chapter 4
Smitheries and Foundries Industry 167
- less dust (0.8 kg/t metal charge instead of 10 to 15 kg/t metal charge for cold blast cupola)
- no CO or SO
2
and less CO
2
(120 kg/t metal charge instead of 450 kg CO
2
/t metal charge for
cold blast cupola)
- the flue-gas rate is smaller (495 m/t metal charge instead of 770 m/t metal charge for cold
blast cupola without water-cooled refractory), and hence the dedusting installation can be
designed much smaller.

Cross-media effects
The use of cokeless cupola melting necessitates duplex operation, in order to allow superheating
of the iron. For superheating in an induction furnace, there is an increased need for electrical
power compared to the hot blast operation.

Operational data
Operational data are given in Table 4.7. The furnace is generally operated at an air factor of
= 1.15. The capacity of the furnace may be high (often in the range 10 12 t/m.h). The
amount of spheres is 1 1.4 % of the metal charge. The energy efficiency of the cupola, without
taking the coupled induction furnace into account, is in the range of 70 %.

Units Consumption
(per tonne of molten
metal)
Natural gas consumption
Preheating of the furnace m 600
Melting m 48.0
Electrical energy consumption
a) Superheater
Superheating kWh 64.0
Holding kWh 15.0
b) Auxiliary equipment kWh 25.0
Consumption of lining material
Furnace shaft kg 5 8
Syphon kg 0.7
Superheater kg 0.5
Water consumption m 0.4
Oxygen consumption m 19.7
Metallic charge Components Part (%/t)
steel 25 35
pig iron 20 30
returns/cast iron scrap 35 55
Alloying and additives Components Part (%/t)
Carburiser 1.1
Briquettes Si 0.8
Ceramic balls 0.95
Slag forming
constituent
0.3
Table 4.7: Typical operational data for a cokeless cupola furnace
[202, TWG, 2002]

Due to the absence of cokes (and CO), no latent heat is lost from the cokeless furnace system.
Full heat recuperation from the flue-gas occurs in the shaft. In duplex configurations (for
example in conjunction with an induction furnace), efficiencies in the range of 40 to 60 % may
be obtained. Thermal efficiencies for coke fired cupolas vary between 25 % (cold blast) and
45 % (hot blast, long campaign).

Chapter 4
168 Smitheries and Foundries Industry
Emission data for cokeless and hot blast cupola furnaces are compared in Table 4.8. These
apply for the following configurations:
- cokeless cupola: below-charge exhaust capture; oxidising atmosphere ( = 1.15); no post
combustion; dry filtration
- hot blast cupola: below-charge exhaust capture; post combustion in separate chamber,
followed by recuperation and dry filtration.

Cokeless Hot blast cupola
Working
conditions
Nm/t kg/t Working
conditions
Nm/t kg/t
Gas 50 Nm/tonne
Energy 500 kWh/Nm

1.15
Coke 12 %
Steel 50 %
Recarburi-
sation
1.9 %
Total
combustible
carbon
9 %
Flue-gas 550 2930
CO
2
9.1 % 50 98 6 % 176 346
H
2
O 18.2 % 100 -
O
2
2.7 % 15 15 %
CO <1 % <5.5 <6.9 10 mg/Nm 29 g/t
NO
x
155 375
mg/Nm
85 210
g/t
205 mg/Nm 150 g/t
SO
2
- 100 mg/Nm 300 g/t
Table 4.8: Emission data for cokeless cupola and hot blast cupola
[170, Godinot and Carlier, 1999]

The following observations can be made:
- the cokeless cupola emits five times less flue-gases than the hot blast cupola. This is mainly
due to the post combustion, which brings excess air into the combustion chamber of the hot
blast cupola. As a consequence, the cokeless cupola can be equipped with a smaller flue-gas
treatment system
- the cokeless cupola emits 3 to 4 times less CO
2
than the hot blast cupola
- the cokeless cupola emits more CO, which is combusted in the case of the hot blast cupola
- NO
x
and SO
2
emission levels are low compared to current emission limit values (e.g. for
France, 500 mg/Nm and 300 mg/Nm respectively)
- if dry filtration is applied, both techniques have low dust emission values.

Applicability
The technique applies to new installations, producing medium to large series. The cokeless
cupola needs a constant and continuous working regime. Due to high oxidation losses and a
high risk of bridging, the technique requires a clean feed with a steel content of max. 35 %.
Since no sulphurisation occurs, the technique is of special interest for the production of nodular
iron.

Economics
Data on operational costs (for 1999) are presented relative to operational costs of hot blast
cupolas (set at 100%) in Table 4.9. The values apply for a 12 tonnes/h equipment and have been
based on a study of 3 cokeless cupolas operating in Europe.

Chapter 4
Smitheries and Foundries Industry 169
Raw
materials
(%)
Auxiliary
materials
(%)
Melting energy
+ Fluids
(%)
Comparative
index
Grey cast iron Cokeless gas cupola 83 6 11 116
Hot Blast coke cupola 69 8 23 100
Nodular cast iron Cokeless gas cupola 81 6 13 104
Hot Blast coke cupola 69 8 23 100
Table 4.9: Operational costs for cokeless cupola, relative to hot blast cupola (set at 100%)
[202, TWG, 2002]

The tabled values were calculated in 1999, however since then, there has been an increase of the
price of coke. From this table it may be concluded that, in Europe a 12 t/h cokeless cupola:
- melts the more expensive grey cast iron
- melts nodular base iron with a price very similar to hot blast cupola.

The comparison is very dependent on the local price of energy and materials.

Driving force for implementation
To reduce cupola emissions.

Example plants
- Dker, Laufach (D): 15 tonne/h nodular iron
- Dker-Kuttner, Lingotes Especiales, Valladolid (E): 16 tonne/h grey an nodular iron
- Hayes Hydraulic Castings (GB), 5 - 6 tonne/h lammellar and nodular iron.

Reference literature
[32, CAEF, 1997], [110, Vito, 2001], [170, Godinot and Carlier, 1999]

4.2.1.10 Gas-fired coke-cupola

Description
The principle of the coke-gas cupola is to replace part of the coke with gas. There are two
techniques for burning natural gas in the cupola:
- using air-gas burners, located above the tuyres
- using oxygas burners, located in the tuyres.

At present, the coke/air-gas cupola furnace finds limited implementation in Europe. This may be
attributed to the difficulty in controlling the process and the increased complication of the
furnace shell.

The oxygas burner was introduced in 1994. Besides oxygas firing, it allows the introduction of
dust into the tuyres for recirculation, although in practice this option is not applied. The oxygas
burners are placed in 1/3 1/2 of the tuyres. About 10 % of the coke energy is replaced by
natural gas, leading to a gas consumption of 8 to 16 Nm/tonne. This is associated with a total
oxygen consumption (burners + lances) of 40 to 60 Nm/tonne. The application of this
technique allows a greater flexibility in the production and/or metallurgy.

The effect and use of the oxygas burner depends on the cupola concerned. In cold blast cupola
furnaces, the technique is used to ensure easy restarts or to reduce the proportion of coke. In hot
blast operation, the technique is used to increase the furnace capacity without modifying the
melting bed. The replacement of part of the coke with CH
4
results in a reduction in flue-gas
volume. This is used as a means of increasing the furnace capacity, without over-saturating the
installed flue-gas cleaning system.

Chapter 4
170 Smitheries and Foundries Industry
The technique results in an increase in the carbon content of the melt and allows an increase in
the amount of steel in the charge. The process provides a means of injecting FeSi, which is less
expensive in powder form than as ore. These properties result in a beneficial economic effect.

Achieved environmental benefits
The replacement of part of the coke with CH
4
results in a reduction in the flue-gas volume. In
addition, the flue-gas is made more combustible due to the higher CO and H
2
contents. If post
combustion is present, the resulting flue-gases will show lower levels of organic compounds
and CO.

The replacement of coke by natural gas reduces the emission levels of SO
2
.
The technique allows the possibility of recirculating cupola dust into the melt. Although, after
some initial trials, this application has not been fully developed.

Reducing the level of coke in the cupola increases the risk of bridging.

Cross-media effects
The production, storage and use of oxygen increases the safety risks. Oxygen production is done
through cryogenic distillation or Vacuum (Pressure) Swing Adsorption, which both involve
elctricity consumption. The consumption of the latter technique is 0.35 0.38 kWh/Nm O
2
.
Oxygen production is often done by an external supplier, who delivers the oxygen to a storage
tank or directly through a pipeline.

Operational data
The AGA-Rayburn Foundry (Coalbrookdale, UK) equipped 3 out of 8 tuyres of their cold blast
cupola with oxygas burners. This allowed a reduction in the overall proportion of coke from
15 % to 10 %. There was a resultant drop in the sulphur content of the produced iron, this then
allowing a reduction to be made in the amount of pig iron used, and thus yielding an economic
gain.

Fritzwinter foundry (D) equipped 3 out of 6 tuyres of their 20 25 tonnes/h hot blast cupola
with oxygas burners. This allowed an increase in production capacity to 28 tonnes/h, without
the need to change the melting bed, nor the flue-gas cleaning system. The effect on the flue-gas
composition before burning is given in Table 4.10.

With oxygen
without oxygas burners
With oxygen
with oxygas burners
H
2
0.8 1.2 2.2 2.4
CO 14 15 19
O
2
2 2
Table 4.10: Effect of oxygas burners on the flue-gas composition for a hot blast cupola
values in %
[184, Godinot and Ressent, 2002]

Applicability
The technique may be applied on both cold blast and hot blast cupolas in both new and existing
installations. The advantages drawn from the application (increased flexibility, economical
benefit, reduced flue-gas volume, increased capacity) will depend on the specific melting
conditions of the installation under consideration. The technique has been reported to cause
difficulties for controlling the process and also increases the complication of the furnace shell
required.

Chapter 4
Smitheries and Foundries Industry 171
Economics
Operational costs before and after switching a cold blast cupola to oxygas operation for the
above-mentioned Aga-Rayburn foundry (UK) are given in Table 4.11. Prices are forecasts as
calculated by CTIF (F).

Unit price Without burners With burners Input Units

EUR
Consumption
per tonne
Cost
EUR/tonne
Consumption
per tonne
Cost
EUR/tonne
Coke Tonne 198 0.15 29.7 0.1 19.8
Gas Nm 0.15 0 - 16.5 2.5
Oxygen Nm 0.38 14 5.4
0.23 40 9.2
Pig iron Tonne 164.6 0.2 32.9 0 -
Scrap iron Tonne 125.8 0.2 37.7 0.5 62.9
Total EUR/tonne 105.7 94.4
All values per tonne molten metal
Table 4.11: Operational costs for cold blast cupola with and without oxygas burners
[184, Godinot and Ressent, 2002]

Operational costs decrease from EUR 105/tonne to EUR 94/tonne molten metal. The economic
gain can mainly be attributed to the reduced proportion of pig iron. This also stresses the fact
that the balance will differ for each specific foundry.

Driving force for implementation
To increase flexibility of the foundry and/or to increase the production capacity of existing
installations without changing the installation layout.

Example plants
- AGA-Rayburn Foundry (Coalbrookdale, UK)
- Fritzwinter foundry (D)

Reference literature
[156, Godinot, 2001], [184, Godinot and Ressent, 2002]

4.2.2 Electric arc furnace

4.2.2.1 Shortening melting and treatment time

Description
Improved control methods can lead to shorter melting or treatment times. Some examples are:
- closer control of the composition (e.g. C, S, P content) and the weight of the charged
materials and slag forming materials
- reliable temperature control of the melt, this can improve the yield of the refining reactions
and avoid overheating
- more efficient methods for sampling and deslagging, these can reduce furnace downtime.

Secondary metallurgy, using AOD/VODC treatment, shortens EAF time and has a positive
impact on energy conservation. The technique is discussed further in Section 4.5.7.1.

Achieved environmental benefits
Increased furnace efficiency through shorter melting times and reduced downtime.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to all new and existing EAF furnaces.
Chapter 4
172 Smitheries and Foundries Industry
Driving force for implementation
Increase the efficiency of the melting and treatment operation

Example plants
This technique finds widespread application in European foundries using EAF melting.

Reference literature
[32, CAEF, 1997], [202, TWG, 2002]

4.2.2.2 Foamy slag practice

Description
The foamy slag practice, currently in use in the steel production industry, consists in
simultaneously injecting oxygen and carbon (in the form of coal dust) into the slag at the end of
the melting. The foam of slag is produced by the action of CO bubbles. The CO gas comes from
the oxidation of carbon in the metal by the injected oxygen and also from the reduction of the
iron oxides (FeO) by the injected carbon.

Creating a foamy slag improves the heat transfer to the charged units and protects the refractory
material inside the furnace. Because of the better arc stability and less radiation effects, slag
foaming leads to a reduction in energy consumption, electrode consumption, noise levels and
increased productivity.

Achieved environmental benefits
Reduction of energy consumption and electrode consumption, lower noise levels and reduction
of flue-gas volume.

Cross-media effects
As the volume of the slag rises so larger slag buckets may be needed. After tapping, the slag
degasses again. There is no reported adverse impact on the possibilities for re-use of the slag.

Operational data
Table 4.12 gives the operational data for a 60 tonne arc furnace and indicates the mains savings
in energy, time, refractory and electrodes.

Units Normal slag Foamy slag
Total power MW 25 30
Radiation loss arc to flue-gas MW 6 0
Flue-gas flow m/h 41000 28000
Final temperature
- metal
- slag
- flue-gas
- refractory
C
1630
1603
1463
1528

1630
1753
1607
1674
Energy-input
- electricity
- fossile (coal)
kWh/tonne
50.8
37.1

37.7
22.6
Energy-output
- metal (H)
- slag (H)
- furnace losses
- flue-gas losses
- other losses
kWh/tonne
10.4
1.1
20.7
53.6
2.5

10.4
9.4
14.1
24.8
1.6
Heating time 1584 - 1630 C
Heating rate
min
C/min
11min 45sec
3.9
7min 30sec
6.1
Table 4.12: Energy and temperature data for EAF furnace melting with normal slag and foamy
slag
[202, TWG, 2002]
Chapter 4
Smitheries and Foundries Industry 173
Slag density is reduced from 2.3 tonne/m to 1.15 1.5 tonne/m.

Applicability
This technique applies to new and existing EAF foundries that practice oxygen injection.

Driving force for implementation
Increasing the efficiency of furnace operation.

Example plants
No specific example plants were reported, but the technique finds application in several
European foundries.

Reference literature
[32, CAEF, 1997], [202, TWG, 2002], [211, European IPPC Bureau, 2000]

4.2.3 Induction furnace

4.2.3.1 Process optimisation: optimising charge materials, charging and
operation

Description
Process optimisation options for coreless induction furnaces include:
- Optimise feedstock condition: This involves avoiding rusty and dirty inputs, utilising
optimum size and density input/scrap, and using cleaner carburisers. These measures
shorten the melting time, reduce the specific energy needed for melting and/or reduce the
amount of slag formed
- Close furnace lid: Oxidation is reduced by avoiding badly fitting lids and
unnecessary/prolonged openings, by quick charging, or by using a protective atmosphere
over the melt (N
2
). Opening times need to be minimised to prevent energy losses. The
necessary opening times for charging, removing slag, temperature measuring, sampling and
pouring vary between 50 % and 25 % of the shift time. The latter figure applies for new
furnaces, working in optimised conditions. A well-fitted closed lid limits the surface heat
loss to about 1 % of the input power. When open, the heat loss can amount to up to 130
kWh/tonne for a 10 tonne capacity unit. When melting under a closed lid, care should be
taken not to overheat the furnace
- Restrict holding to a minimum: Since melting is the first step in the foundry process chain, a
shortening of the holding time can only be obtained by an integrated process optimisation
for the whole foundry process and by minimising delays, problems and irregularities in any
of the foundry departments. During holding, composition adjustment is performed based on
the results of an analysis made on a chill cast sample. Optimisation of the sampling, testing
and adjusting procedures is another measure for reducing the holding time
- Operate at maximum power input level: Furnaces are more efficient in their use of energy
when they are being operated at maximum power input levels, and best results are obtained
when the available power can be fully utilised for the largest proportion of the melting
cycle. This also includes reducing cold start melts (optimise production programme), and
allowing follow-up using monitoring and computer control
- Avoid excessive temperature and unnecessary superheating: It is important that the metal
reaches the required temperature just at the time that the moulding department is in a
position to receive it. Good co-operation between the melting and moulding shop is
essential in order to minimise the electricity usage
- Optimise high temperature melts for slag removal (good balance): Low melting point slag
build-up can be reduced by heating the furnace up to elevated temperatures (1580 C vs.
1450 C normal). This results in a higher energy consumption, and can affect the
metallurgical aspects of the melt. If the slag is allowed to build up on the furnace lining, this
can affect the electrical efficiency of the furnace. Slag removal requires the opening of the
furnace lid, and thus causes a thermal loss. A good balance needs to be found between
increasing the melt temperature and the slag removal practice
Chapter 4
174 Smitheries and Foundries Industry
- Prevent slag build up: More common and more troublesome are instances where high
melting point build-up is involved. This mainly results from charging sand and, in the case
of iron melting, metallic aluminium into the melt. Some furnace operators have attempted
flux additions and cleaning routines but prevention is better than cure in this respect. This
involves minimising the presence of sand and Al in the raw materials
- Trickle the oxygen injection: Instead of using conventional decarburisation
- Minimise and control the refractory wall wearing: The refractory life depends on the choice
of materials as a function of the slag chemistry (acidic or basic), the operational temperature
(steel, cast iron, non-ferrous), and the care taken upon relining (sintering). The lifetime may
vary from 50 (steel, cast iron) to 200 300 (cast iron) melts. Operational control measures
are taken to follow the refractory wear. These include visual inspection, physical
measurement and instrumental monitoring programmes. Good charging practice measures
prevent the cumulative effects of physical chokes and mechanical stresses. These include
the use of automatic charging systems, hot charging, avoiding high drops and the use of
compact and dry scrap.

Achieved environmental benefits
Increased furnace efficiency through shorter melting times and reduced downtime.

Cross-media effects
No cross-media effects apply.

Operational data
A typical coreless furnace can melt a tonne of iron and raise the temperature of the liquid metal
to 1450 C using under 600 kWh of electricity. However, in practice, only a few foundries can
achieve this level of specific consumption on a week-by-week basis. Some foundries consume
in the region of 1000 kWh for every tonne of iron produced by their coreless furnaces.
Prevailing circumstances in many foundries can restrict the scope for good energy management,
but in fact almost all coreless melting operations could be improved in some way, in order to
achieve a worthwhile saving in the quantity of electricity used per tonne of iron processed.

Applicability
This technique applies to all new and existing induction furnaces.

Driving force for implementation
Increasing the efficiency of furnace operation.

Example plants
Process optimisation measures are commonly applied in European foundries using induction
furnaces.

Reference literature
[32, CAEF, 1997], [47, ETSU, 1992], [110, Vito, 2001], [145, Inductotherm], [225, TWG,
2003]

4.2.3.2 Change from mains frequency to medium frequency furnaces

Description
Medium frequency (250 Hz) furnaces have a higher power density (up to 1000 kW/tonne) than
mains frequency (50 Hz) furnaces (300 kW/tonne). This allows the use of a smaller crucible (up
to a factor of three smaller) which results in a smaller total heat loss. The thermal efficiency of
medium frequency furnaces is 10 % higher than for the mains frequency types. Additionally,
mains frequency units need to be operated with a molten heel of up to 2/3 of the crucible
capacity to optimise specific energy consumption and also require specific starter-blocks for
cold start-up. Medium frequency furnaces can readily be started with a cold charge and can be
emptied at the end of each working shift or melting batch.
Chapter 4
Smitheries and Foundries Industry 175
When a foundry converts from mains frequency melting to operating a medium frequency
installation, it is important that the furnace personnel undertake some retraining. Operating
techniques formally used to date will have to be abandoned, and new specific procedures
designed to give good specific energy consumption will have to be adapted. If retraining is
overlooked, the available improvements in energy usage may not be fully realised.

Achieved environmental benefits
Increased energy efficiency of the melting operation.

Cross-media effects
No cross-media effects apply.

Operational data
Operational data are given in Section 3.2.4.

Applicability
This technique applies to new installations and for major refurbishments of existing installations

Driving force for implementation
To increase the efficiency of the foundry operation.

Example plants
The technique is commonly applied in foundries installing a new furnace.

Reference literature
[47, ETSU, 1992]

4.2.4 Rotary furnace

4.2.4.1 Increasing the furnace efficiency

Description
All measures that increase the thermal efficiency of the furnace will in turn lead to a lower CO
2
output. A major improvement has been achieved by introducing oxygen instead of air as the
combustion medium. This is discussed in Section 4.2.4.2.

Further improvements of furnace yield can be obtained through tight control and optimisation
of:
- burner regime
- burner position
- charging
- metal composition
- temperature.

Achieved environmental benefits
The optimisation results in lower amounts of dust and residues and in higher energetic
efficiency.

Cross-media effects
No cross-media effect apply.

Operational data
An optimisation programme on a 3 tonnes/h cast iron melting furnace suggested the following
as good operational practice:
Chapter 4
176 Smitheries and Foundries Industry
- use of clean scrap and loading in the following order: (1) ingots and materials with low Si-
content; (2) internal return material and foundry scrap; (3) alloying elements and melt
protection; (4) steel scrap
- melt protection: use of anthracite for melt protection (2 % of metal charge) and silicon
(2 %)
- rotation: in discontinuous phase, 1/3 of a rotation every minute with a total of 7.5 turns until
the phase change. During continuous rotation: 1.7 rotations/minute
- power and angle of burner: use a parallel burner-head position for the lower injectors. Start
at maximum power during 20 minutes, reducing 10 % every 20 minutes until change of
phase (60 minutes after start).

Using these measures, a metal efficiency (molten metal/charged metal) at >95 % could be
maintained.

Applicability
The principles of the optimisation are generally valid for iron melting in rotary furnaces using
oxygen burners. The operational data have been established for a 3 tonnes/h furnace. For
furnaces of other sizes, a site-specific optimisation is necessary.

Driving force for implementation
Optimisation of furnace operation and to increase the melting efficiency.

Example plants
Process optimisation measures are commonly applied in European foundries using rotary
furnaces.

Reference literature
[204, Carnicer Alfonso, 2001]

4.2.4.2 Use of an oxyburner

Description
Flame temperatures are increased by the application of pure oxygen instead of air in the burners
used for melting or preheating the pouring ladles. This enables a more efficient heat transfer to
the melt and reduces the energy use.

If the air supply is blocked by a tight closure of the recipient, no NO
x
can be formed through the
oxidation of atmospheric nitrogen. Additionally, the total flow of flue-gases from an oxyburner
is smaller due to the absence of nitrogen ballast. This allows the application of a smaller
dedusting installation.

Achieved environmental benefits
The techniques reduce energy consumption and lower the emissions of NO
x
and CO
2
, through
higher combustion temperatures.

Cross-media effects
The production, storage and use of oxygen increases the safety risks. Oxygen production is done
through cryogenic distillation or Vacuum (Pressure) Swing Adsorption, which both involve
electricity consumption. The consumption of the latter technique is 0.35 0.38 kWh/Nm O
2
.
Oxygen production is often done by an external supplier, who deliver the oxygen to a storage
tank or directly through a pipeline.

Fuel or heavy oil gives rise to SO
2
or NO
x
emissions, depending on their S or N content. The use
of cleaner carburants such as natural gas and propane will not cause any additional pollution,
except for CO
2
, as is the case for all combustion processes.

Chapter 4
Smitheries and Foundries Industry 177
Operational data
Table 4.13 gives, for the "oxygas" melting of cast iron and various furnace capacities, the
theoretical consumption of several fuels and of oxygen per tonne melted:

Energy source Units Furnace capacity (tonnes)
3 5 8 12 20
Light fuel oil kg/tonne 33 38 33 38 33 38 32 37 32 37
Natural gas Nm
3
/tonne 38 43 38 43 38 43 38 42 38 42
Propane Nm
3
/tonne 15 17 15 17 15 17 14 16 14 16
Oxygen Nm
3
/tonne 130 150 130 150 130 150 130 145 130 145
Table 4.13: Table of energy consumptions (minimum melt)
[148, Eurofine, 2002]

Oxygen enrichment used in conjunction with a recuperator generally achieves a 30 % energy
saving. Additionally, the higher combustion temperature assists in reducing the overall
emissions. The exhaust gas volume is also reduced. Full oxy/fuel firing may offer energy
savings of up to 50 %, and can reduce the exhaust gas volume by up to 72 %.

A German installations change from oil-air firing to gas-oxygen firing caused a reduction in
noise emissions of 15 18 dB(A) in the immediate vicinity of the furnace, dependent on the
point of analysis. At the installation border, a value of 48 dB(A) was measured. With the
application of energy recuperation for charge preheating, a total energy saving of 53 % was
reported.

Applicability
This technique can be applied on any rotary furnace and in the preheating of pouring ladles.
Oxyburners do not find implementation in non-ferrous foundries, although they are used e.g. in
secondary copper smelting.

Economics
Investment costs: EUR 3400 4500.
Operational costs: dependent on process operation.

Driving force for implementation
Optimisation of furnace operation and to increase melting efficiency.

Example plants
This technique finds wide application in ferrous foundries using rotary furnaces.

Reference literature
[23, Brettschneider and Vennebusch, 1992], [32, CAEF, 1997], [110, Vito, 2001], [103,
Vereniging van Nederlandse Gemeenten, 1998], [148, Eurofine, 2002], [163, UK Environment
Agency, 2002], [232, Novem, 2000]

4.2.5 Selection of cupola vs. induction or rotary furnace for cast iron
melting

Description
Given the fact that various melting techniques show overlapping fields of application, a basic
technique is the selection of the melting technology. In this selection the decisive criteria
include:

Chapter 4
178 Smitheries and Foundries Industry
- metal type
- continuous or batch production
- size of the series
- metal throughput or capacity
- flexibility towards input material type and cleanliness
- flexibility towards alloy change
- type of product made
- emissions and other environmental considerations
- availability of raw materials
- availability of fuels/electricity.

The application of the different melting techniques is very dependent on the indicated criteria.
From the discussions mentioned above, the following generalities can be deduced:

- Metal type: Electrical furnace better placed for nodular or alloyed cast iron
- Continuous production: Cupola is better placed
- Batch production: Electrical or rotary furnace is better placed
- Flexibility towards input material type and cleanliness: Cupola is better placed
- Flexibility towards alloy change: Induction and rotary furnace is better placed
- Environmental considerations:
cupola is better placed only if well de-dusted, cokeless show better environmental
performance than HBC or CBC
induction has lower CO, SO
2
, NO
x
, dioxin, slag, but this it needs to be borne in mind
there will also be emissions from the required electricity production.
- Availability ofraw materials: When cheap scrap is available, the cupola is better placed

In fact, all these criteria must be considered together. Table 4.14 gives a summary, based on the
technical considerations. Economic data for the various cupola types and the induction furnace
are given in Annex 1.

On the basis of the stated criteria, the replacement of the cupola by induction or rotary furnaces
may be considered. The selection of induction or rotary furnaces is given priority over cold blast
cupolas for small foundries casting a variety of products in several European countries (e.g.
Austria, Belgium (Flanders)).

Chapter 4
Smitheries and Foundries Industry 179
CRITERION ONLY GREY CAST IRON GREY +
NODULAR
ONLY NODULAR CAST IRON ONLY
MALLEABLE
CAST IRON
Type of product
made
Counter
weights
Sewer castings,
pipes, urban
furniture
Enameled
castings,
heating
appliances
Mechanical parts Mechanical
parts
Sewer
castings,
pipes, urban
furniture
Mechanical parts Mechanical
parts
Size of series Mediumto
large
All Mediumto
large
Batch to
small
Medium
to large
All All Batch to
small
Mediumto
large
Mediumto
large
Cupola Cold
blast
++ ++
(<10 - 15 t/h)
++
(<10 - 15 t/h)
++
(<10 - 15 t/h)
(+)
Hot
blast
+ ++
(>10 - 15 t/h)
++
(>10 - 15 t/h)
0 ++
(>10 - 15 t/h)
++
(>10 - 15 t/h)
0 +
(>10 - 15 t/h)
Cokeless 0 + + 0 +
Induction + + ++ ++ ++ + ++ ++ ++
Rotary furnace (+) batch ++ + ++ (+) batch or
small series
++ ++
++: technically more adapted;
+: technically adapted
(+): adapted in some cases
0: not adapted
Table 4.14: Technical choice of the melting equipment for cast iron melting
[202, TWG, 2002]
Chapter 4
180 Smitheries and Foundries Industry
Achieved environmental benefits
Replacement of a cupola by an induction or rotary furnace results in a reduction in the direct
emissions of CO and SO
2
and a reduced amount of slag. Induction furnaces show a reduced
emission of NO
x
and a lower risk of dioxin formation. The indirect emissions generated by the
use of induction furnaces, depend on the local electricity generation infrastructure.

Cross-media effects
Replacement of a cupola by an induction furnace generates a strong increase in electricity
consumption.

Applicability
The replacement of a cold blast cupola by an induction or rotary furnace is applicable under the
criteria stated above and upon major refurbishment of the installation.

For new installations, the criteria of Table 4.14, apply besides local and site-specific
considerations. If cupola and induction or rotary furnaces are indicated as most adapted, the
induction and rotary types may be preferred on the basis of their environmental benefits.

Economics
A survey of economic data for the various types of cupola furnaces and for the induction
furnace is given in Annex 1.

In Table 4.15, a summary is given of cost data collected by an example Spanish foundry that
wanted to replace an existing cupola by a rotary or electric furnace. No emission collection or
filtration equipment costs are included. Data are from 1997 - 1998. The comparison involves the
following three systems:
- cold blast cupola: 3.5 tonne/h capacity
- oxygas rotary furnace: 3 tonne capacity
- 2 medium frequency induction furnaces: 2 + 1 tonne capacity.

The cost calculation was performed for melting costs for a melting production of 1 tonne/h,
working 8 hours/day. For the induction furnace a distinction is made between peak-time
electricity costs and off-peak-time. Since electricity is the major component of the costs, this
distinction makes a big difference to the final cost. Operation of the induction furnace during
off-peak time (night-time) results in costs comparable to rotary furnace costs. The data show
that the investment has a 5 9 years payback period.

Costs in EUR/tonne of
molten metal
Cupola Rotary Induction
Energy 30.7 29.8 40.8 (peak)
27.2 (valley)
Refractory 3.6 1.3 1.4
Man power 52.9 22.7 22.7
Raw materials 205 205 205
Total cost 292 259 270
Yearly costs in EUR
Cost for melting
1320 tonne/year
385000 341000 356000 (peak)
338000 (valley)
Savings none 44000 29000 (peak)
47000 (valley)
Investment none 250000 260000
Building Not needed Needed
Table 4.15: Example plant cost data for replacement of an existing cupola with a rotary or
induction furnace
Data from 1997 - 1998, Spain
[202, TWG, 2002]
Chapter 4
Smitheries and Foundries Industry 181
Driving force for implementation
Reduction of direct emissions from cast iron melting.

Example plants
Most Austrian foundries, which are small and medium-sized enterprises, have changed from
cupola melting to induction furnaces. As a consequence there are only three Austrian foundries
left which still use cupola melting.

Reference literature
[202, TWG, 2002], [110, Vito, 2001], [225, TWG, 2003]

4.2.6 Hearth type furnace

4.2.6.1 Use of an oxyburner

See Section 4.2.4.2.

4.2.7 Other furnace types

4.2.7.1 Alternatives to the use of SF
6
as a cover gas for Mg-melting

Description
Alternative covering gases for magnesium melting are of big interest due to the high global
warming potential of the generally used SF
6
. The substitution of SF
6
has been the topic of
research for several years. A worldwide research project to develop and propose alternatives is
currently being co-ordinated by the International Magnesium Association and will be finalised
in 2004. Gas mixtures containing HFC-134a (CF
3
CH
2
F) and Novec 612 (perfluoroketone,
C
3
F
7
C(O)C
2
F
5
) have been successfully tested. Research is on-going on the applicability of both
compounds. HFC-134a falls under the restrictions of the Kyoto protocol. Both compounds show
a lower global warning potential than SF
6
: SF
6
GWP = 23900; HFC-134a GWP = 1300;
Novec 612 GWP = 1.

At present, SO
2
is a viable alternative. Some magnesium foundries have long experience with
this gas, and never turned to SF
6
when this gas was first introduced at the beginning of the
1970s. SO
2
reacts with liquid magnesium and builds a protective film on the surface. A mixing
unit for blending a low concentration SO
2
-air mixture is available. Furthermore, handling
procedures for SO
2
have been established to reduce workers exposure and risk.

The protective gas is supplied through a manifold with several outlets. The positions of the
outlets are chosen to give protection to all areas of the metal surface. Areas closer to hatches
that will be opened frequently during operation need a higher flow of gas than areas where
interaction with the surrounding atmosphere is small.

In order to optimise safety and minimise gas consumption, totally encapsulated melting and
casting processes are applied.

If replacement is not feasible, the following technical measures allow a reduction of SF
6
consumption and emissions:
- improved sealing of furnaces
- fully automatic cover gas dosage
- electronic control of both gas mix and flowrate
- reduction of overdosing.

Achieved environmental benefits
The replacement of SF
6
avoids using this greenhouse gas, which has a GWP of 22200 over a
100 years time horizon.

Chapter 4
182 Smitheries and Foundries Industry
Cross-media effects
SO
2
is a toxic gas and exposure limit values for workers should be taken into account. The
occupational exposure limit in most countries is 2 ppm (5 mg/m) over 8 hours. Sulphur- and
oxygen-containing deposits may form on the furnace wall. Under unfavourable conditions these
deposits can be immersed into the molten metal where they cause reactions leading to metal
eruptions from the surface. Frequent removal of scaling can prevent this from happening.

The technique results in emissions of small amounts of SO
2
, the amount depending on the
foundry capacity and the number of furnaces. One estimate could be 50 500 kg per year for
melt protection.

Operational data
The consumption of SF
6
depends on the melting technique used, and varies between less than
0.1 and 11 kg/tonne of finished castings. Technical improvements and measures have resulted in
a reduction in the average specific SF
6
consumption, from over 3 kg/tonne of casting to 0.9
kg/tonne. Application data are given in Section 3.4.1.

In an undisturbed melt, the use of SO
2
would be typically 0.7 % in air at a flowrate of 5 to
10 litre/minute. In closed (but not airtight) furnaces, such as the melting and dosing furnaces of
die-casting machines, pure SO
2
is applied. Small amounts of air entering through the lid are
needed to provide the covering action. If a carrier gas is used, nitrogen is the preferred gas
because of its inertness. The SO
2
is then mixed in concentrations of 1 2 %.

Mass balance data for the fluxless re-melting of magnesium scrap are given in Section 4.1.6.

Applicability
There is a great potential to reduce the SF
6
consumption, but at the moment it does not seem
possible to fully replace it during non-standard situations. SO
2
may be applied in general during
normal melting operations. When repairing the furnace, SF
6
remains needed as a protection gas,
due to the health risks correlated with SO
2
exposure. Additionally, SO
2
cannot be used to
extinguish magnesium fires. In this case also, SF
6
remains in use.

Economics
The price of SF
6
has increased dramatically over the last few years. Conversion from SF
6
to SO
2
will therefore have an associated economic cost. Table 4.16 gives estimates for the running
expenses for the use of SF
6
and SO
2
based on information from a die-caster running three die-
casting machines and the results from industrial scale trials. Even though the figures are only
rough estimates, they clearly show that there is a large cost saving potential when substituting
SO
2
for SF
6
.
Property Units SF
6
SO
2
Concentration of gas % 0.4 0.7
Price EUR/kg 36.5 4.5
Inverted density (at 0 C and 1 atm) l/kg 153 350
Yearly consumption of gas kg/yr 339 259
Cost/year EUR 12373 1165
Data for 3 die-casting machines, run at 300 days/yr, 24 h/day with flowrate of gas to each
machine of 10 l/min
Table 4.16: Cost comparison between SO
2
and SF
6
used for protective gas atmospheres
[190, Schubert and Gjestland]

If a new investment is considered, the SO
2
gas system, meeting all modern safety and dosage
requirements, costs EUR 70000, for a 1000 tonnes/yr plant. A new SF
6
system costs
EUR 23300. Operational data, accounting for this investment and for the use of 1.5 % of SO
2
and 0.2 % of SF
6
, both in nitrogen, are given in Table 4.17. The total annual cost for SO
2
is
some EUR 2500 more than for SF
6
, on the other hand 12300 tonnes of CO
2
equivalent
emissions are avoided. The bigger the plant, the less the annual costs differ between the two
systems. At an annual output of 1500 tonne/yr, the turning-point is passed and the CO
2
reduction cost turns negative.
Chapter 4
Smitheries and Foundries Industry 183
General Casting Data
Capacity of the ovens 2000 kg
Maximum smelting performance 1000 kg/h
Casting time per day (Cold chamber) 13.2 h
Working days 250/yr
Utilisation 76 %
Gross melting weight (shot weight) 2000 t/yr
Scrap factor 50 %
Net weight of the Mg parts 1000 t/yr
Sales per year (EUR 15.00/kg) EUR 15 million
Surface of the Mg baths 6 m
3
Gas (carrier + cover gas) per m
2
surface 300 litres per hour, all year long
Extra gas dosage while charging + 25 %
Gas Data SF
6
SO
2
Carrier gas nitrogen nitrogen
Cover gas concentration in carrier gas 0.2 vol. % 1.5 vol. %
Cover gas dosage per hour 46.9 g 154.2 g
Cover gas dosage per year 514 kg 1688 kg
Coefficient cover gas/net Mg output 0.51 kg/t 1.69 kg/t
Emissions in tonnes CO
2
equivalent 12278 -
Cost Data SF
6
SO
2
1 kg cover gas EUR 20.00 EUR 3.00
1 m
3
carrier gas (nitrogen) EUR 0.20 EUR 0.20
Invest-cost of new gas equipment EUR 23333 EUR 70000
Discount rate per year 10 % 10 %
Depreciation period 10 yr 10 yr
Annualised invest cost of equipment EUR 3797 EUR 11392
Operating cost of cover gas EUR 10274 EUR 5065
Annual running cost (without nitrogen) EUR 14072 EUR 16457
Additional total cost of using SO
2
- EUR 2385
Specific emissions reduction cost - EUR 0.19/tonne CO
2
eq.
Share of additional cost in sales 0.02 %
Note: Most data are valid only for this 1000 tonne/yr plant and will change with plant size
Table 4.17: Operational and cost data for a new die-casting plant of 1000 tonne/yr Mg output
[218, Harnisch and Schwarz, 2003]

For existing installations, the reduced cost for SO
2
consumption needs to pay back the
investment cost for the conversion of the cover gas installation. The payback time is dependent
on the actual cover gas consumption. Based on an SF
6
consumption of 0.9 kg/tonne, the turning
point at which the total costs of using SO
2
get lower than the comparable cost of SF
6
is below
900 tonne/yr output. At 500 tonnes/yr the SO
2
is still more expensive, but the specific reduction
cost of 1 tonne CO
2
eq. amounts to EUR 0.31. The share of additional costs in the plants sales
amounts to 0.04 %. Therefore at this capacity, the technique can also be considered.

As a conclusion, it is found to be economic to use SO
2
instead of SF
6
as a cover gas in
magnesium die-casting, at least in plants with an annual metal output of 500 tonnes and more,
regardless of whether the plant is to be newly built or still runs with SF
6
.
Driving force for implementation
Reduction measures for greenhouse gas emissions, in particular, the Kyoto climate protocol
controlling the use of fluorinated gases: HFCs, PFCs and SF
6
. These gases have a high global
warming potential but are not ozone depletors. This will form the basis for the EU Directive on
the use of fluorinated gases. Furthermore, some regions have already issued proposals to ban its
use, such as e.g. Denmark, Germany and Austria.

Example plants
LM Leichtmetall Systemtechnik, Felbach (D)
In Austria, Denmark and Sweden SF
6
is no longer used as a cover gas.
Chapter 4
184 Smitheries and Foundries Industry
Reference literature
[182, Closset, 2002], [190, Schubert and Gjestland], [191, IMA, et al.], [192, Gjestland and
Westengen, 1996], [194, UNEP IPCC, 2002], [218, Harnisch and Schwarz, 2003]

4.2.8 Non-ferrous metal treatment

4.2.8.1 Aluminium degassing and cleaning using an impeller station

Description
The degassing of aluminium is needed to remove hydrogen from the melt. The removal or
reduction of small amounts of undesirable elements and impurities necessitates treatment of the
melt with halogens, such as chlorine, fluorine or bromine. This treatment is referred to as
cleaning, and is usually combined with the degassing treatment.

Adequate degassing with physically operating treatment processes does not achieve sufficient
cleaning in most aluminium castings. Vacuum degassing achieves a very low hydrogen content
in the melt but, at the same time, the lower number of nuclei means that crystallisation is less
effective.

In a combined degassing and cleaning process, gas mixtures of argon and chlorine or nitrogen
and chlorine are introduced into the melt. The chlorine concentration in the carrier gas must be
optimised in order to ensure good cleaning as well as low emissions. In most cases, only a
degassing treatment is necessary. This treatment is done without Cl
2
-gas. Depending on the
treatment vessel, the efficiency of the degassing can be improved with porous plugs and
impellers.

In some isolated cases, an Ar/SF
6
-mixture has been used. The latter is a greenhouse gas with a
high global warming potential, which falls under the Kyoto protocol. (see also Section 4.2.7.1)

Achieved environmental benefits
The use of an impeller with Ar/Cl
2
or N
2
/Cl
2
avoids the use of hexachloroethane in the cleaning
treatment of aluminium. The use of hexachloroethane was banned from 30 June 2003 in the EU.

Cross-media effects
The application of this type of degassing unit has allowed an end to using hexachloroethane for
degassing and cleaning purposes.

Operational data
A mixture of nitrogen with 3 % Cl
2
is generally used for combined degassing and cleaning. For
degassing only, Cl
2
is not necessary.

A mobile impeller station can be used for holding furnaces and ladles of 50 250 kg of Al. The
treatment takes 3 to 5 minutes. A metal treatment unit is used for capacities from 400 to 1000 kg
Al. The melt is treated for 1.5 to 5 minutes with a gas flow between 8 and 20 litre/minute. The
graphite rotor has a life of 100 150 treatments, depending on the temperature of the melt.

Applicability
Impeller stations have been developed for holding furnaces and ladles from 50 to 1000 kg of
molten aluminium.

Economics
Investment costs are given Table 4.18.

Description Cost (EUR)
Impeller plant including accessories 15000
Facilities for mixing of argon gas with chlorine 40000
Total 55000
Table 4.18: Investment costs for an impeller station
[178, Wenk, 1995]
Chapter 4
Smitheries and Foundries Industry 185
Driving force for implementation
European Directive 97/16/EC prohibiting the use of hexachloroethane in the manufacturing or
processing of non-ferrous metals.

Example plants
The technique is commonly applied in aluminium foundries.

Reference literature
[175, Brown, 1999], [178, Wenk, 1995],

4.3 Mould- and core-making, including sand preparation

4.3.1 Selection of the mould type

The selection of the mould type is mainly based on technical criteria. The applicability of the
various moulding types is summarised in Table 4.19. Additionally, Table 4.20 gives the general
properties of various systems.

Lost moulds Permanent moulds
Moulding methods Casting technique
G
r
e
e
n
s
a
n
d
m
o
u
l
d
i
n
g
S
h
e
l
l
s
a
n
d
P
h
e
n
o
l
/
F
u
r
a
n
P
e
p
s
e
t
/
w
a
t
e
r
g
l
a
s
s
L
o
w
-
p
r
e
s
s
u
r
e
d
i
e
-
c
a
s
t
i
n
g
P
r
e
s
s
u
r
e
d
i
e
-
c
a
s
t
i
n
g
(
h
o
t
)
P
r
e
s
s
u
r
e
d
i
e
-
c
a
s
t
i
n
g
(
c
o
l
d
)
C
e
n
t
r
i
f
u
g
a
l
c
a
s
t
i
n
g
C
o
n
t
i
n
u
o
u
s
c
a
s
t
i
n
g
Cast Iron
Grey cast iron X X X X O X X
Nodular cast iron X X X X X X
Malleable cast iron X O X O
White cast iron X X X
Steel
Unalloyed X X X X X
Low alloy steel X X X X X
Highly alloyed
manganese steel
X X X X X
Stainless and heat
resistant
X X X X O X
Heavy Metals
Brass X X X X X
Bronze X X X X
Copper X X X X X X X X
Zinc alloys O X X
Light Alloys
Aluminium alloys X X X X X X X
Magnesium alloys X O X X X X X
Titanium O O X X O
X: The method can be used
O: The method is possible but not often employed
Table 4.19: Applicability of the various moulding types
[126, Teknologisk, 2000]

Chapter 4
186 Smitheries and Foundries Industry
Green sand
casting
Chemically-
bonded sand
Low-pressure and
gravity die-casting
High-pressure
die-casting
Relative cost in quantity

Low Medium high Low Lowest
Relative cost for small
number
Lowest Medium high High Highest
Permissible weight of
casting
Up to about 1
tonne
Up to a few
hundred tonnes
50 kg 30 kg
Thinnest section castable,
centimetres
0.25 0.25 0.3 0.08
Typical dimensional
tolerance, centimetres
(not including parting
lines)
0.03 0.02 0.07 0.02
Relative surface finish

Fair to good Fair to good Good Best
Relative mechanical
properties
Good* Good* Good* Best
Relative ease of casting
complex design
Fair to good Good Fair Good
Relative ease of changing
design in production
Best Fair Low Lowest
Range of alloys that can
be cast
Unlimited Unlimited Copper base and
lower melting point
metals preferable
Aluminium base
and lower melting
preferable
(*) Mechanical properties may be enhanced by heat treatment
Table 4.20: Technical properties of the various moulding types
[42, US EPA, 1998]

4.3.2 Moulding with clay-bonded sand (green sand moulding)

4.3.2.1 Preparation of clay-bonded sand by vacuum mixing and cooling

Description
The mixing and cooling process are combined into a single process step. This is achieved by
operating the sand mixer under reduced pressure, which results in cooling by the controlled
vaporisation of the water. The special mixer needs to be hermetically closed. It has a pressure
casing and is connected to a vacuum system. The layout of the system is given in Figure 4.6.

Chapter 4
Smitheries and Foundries Industry 187
Figure 4.6: Moulding-sand preparation plant with vacuum mixer-cooler
[207, Drews, 1996]

As compared to a normal mixing plant (Figure 2.28), this technique does not require a separate
cooler and associated equipment. The vacuum plant therefore is a more compact size. Since the
total airflow is reduced, the vacuum installation has a smaller central dust extraction filter. The
size and energy consumption of the filter are 50 % of that of a conventional plant. The reduction
of the airflow, results in a reduced removal of active bentonite and thus in an overall reduction
of additive consumption. Due to the enclosed operation, the technique is not affected by outside
climatic conditions.

The use of steam in the absence of air leads to a quick activation of the bentonite, resulting in a
very quick attainment of the optimal compressive strength for the bonded sand.

Water is added in two separate stages:
- water to moisten the return sand to the prepared sand moisture content
- water to cool the sand by vaporisation to the required temperature of 38 40 C.

The disadvantage of the system is the operation and management of the vacuum system. The
process also affects the sand technical properties (e.g. compressive and tensile strength,
permeability), which diminish after 72h (the 72h effect). This can be prevented by re-mixing
for about 90s.

Achieved environmental benefits
Reduction of the total airflow results in a reduced off-gas volume and amount of dust for
disposal, as well as a reduced consumption of binder as compared to conventional cooling and
mixing plants (although this also depends on the efficiency of the regeneration plant).
Additionally, the system uses less energy.

Cross-media effects
No cross-media effects apply.

Chapter 4
188 Smitheries and Foundries Industry
Applicability
The technique is applicable in new green sand installations.

Economics
The commercial limits of application appear to be processes where sand cooling is required and
which have a capacity of >60 tonnes of sand per hour.

Driving force for implementation
Reduction of dust for disposal and minimisation of additives consumption. Limitation of space.

Example plants
In Japan, a vacuum mixing and cooling plant has been in operation since 1993, with a prepared
sand throughput of 100 m/h.
Various installations are also currently running in Italy, France, Germany and England.

Reference literature
[143, Inasmet and CTIF, 2002], [207, Drews, 1996], [214, Gerl, 2003]

4.3.3 Moulding and core-making with chemically-bonded sand

4.3.3.1 Minimisation of binder and resin consumption

Description
The minimisation of chemicals consumption can be achieved through optimisation of the
process controls and material handling. The latter category is dealt with in Section 4.1. In this
section process measures are discussed.

Overuse to compensate for poor process control is the most common way in which binder
chemicals are wasted. For instance, key parameters that relate to good binder management
include:
- Sand consistency: Use of a sand quality which is consistent with the binder system. Good
management of sand storage and sand testing (purity, grain size, shape, moisture) are of the
main importance. A low content of fines and maximum amount of re-used sand will reduce
the amount of resin needed
- Temperature control: The sand temperature must be maintained in a narrow range, with
regular checks and adjustment of the amount of hardener addition. Positioning the sand
heater just before the mixer allows close temperature control
- Mixer maintenance and cleaning
- Mould quality: Checking, solving and preventing moulding defects
- Addition rates: The appropriate binder addition depends on the binder type, the surface area
of the sand and the size of the casting
- Mixer operation: Optimisation of the mixer performance involves monitoring and control of
its operation.

Table 4.21 lists the mixing process variables that are easy to measure using readily available
instrumentation. Interfacing the controls with an alarm system allows the operator to be alerted
to any out-of-specification condition. This needs to be corrected by cleaning, maintenance and
recalibration before a major problem develops.

Chapter 4
Smitheries and Foundries Industry 189
Process variables Comments Instrumentation
Sand flow Can vary as a result of obstructions
at the discharge gate that
reduce/stop the sand flow or
worn/loose stop nuts that produce a
faster flow
Impact plate flowmeters
Sand temperature Determines the speed of hardening
and the amount/type of hardener
required
A process control instrument can
be used to adjust hardener
additions to compensate for
changes in sand temperature
Power consumption Provides an indicator of mixer
cleanliness
Motor load monitor
Resin and hardener flowrate Can vary as a result of worn pumps,
viscosity changes with temperature,
blockages or leaks in pipework and
nozzles, sticking of non-return
valves, etc.
Positive displacement,
electromagnetic or coreolis
flowmeters

Operating hours Useful for assessing performance
against capital and maintenance
costs

Table 4.21: Process variables and how they indicate mixer performance
[71, ETSU, 1998]

For foundries that require greater sophistication, a range of fully automated mixer control
systems are available. These systems use microprocessor technology to provide automatic
corrective control of mixed sand production and need only limited operator involvement. The
additional benefits achievable with automated control systems are mainly due to the much
reduced dependence on the operator and the more rapid correction of variations in conditions.
The latter is also performed without having to stop production.

Achieved environmental benefits
Optimisation of the binder and resin use results in a minimisation of the consumption of
chemical additives.

Volatile organic compounds make up to 50 - 60 % by weight of the binder components. The
amount depends on the binder system type. Most are emitted during sand mixing and metal
pouring. A reduction in the use of binder results in a corresponding reduction of VOC
emissions.

Cross-media effects
No cross-media effects apply.

Operational data
In most cases, reductions of 5 % in binder levels and 1 % in scrap moulds are easily achieved
using modern mixer control systems. Many foundries have realised savings that have been
considerably higher. Reductions in binder addition of 5 to 25 % have been reported for various
foundries, as well as almost eliminating scrap moulds production.

The installation of an automated mixer control system in an iron foundry using cold-setting
resins allowed the foundry to reduce the resin addition rate from 10 kg/minute (1.22 % resin to
sand weight) to 8.89 kg/minute (1.09 %), which correspondingly resulted in a 10 % saving in
the amount of catalyst used. The number of defective moulds fell by over 60 % and cost savings
of over GBP 37000/year (1995 UK prices) were achieved.

Applicability
This technique applies to all new and existing foundries using chemically-bonded sands. Mixing
monitoring control equipment can be retrofitted to existing plants.
Chapter 4
190 Smitheries and Foundries Industry
Economics
For a tonne of mixed sand, the binder addition is typically only 1 3 % by weight, but in terms
of cost, binders account for 30 60 % of the total raw material cost. It is estimated that cost
savings of 5 10 % could be achieved through better management of binder materials.

Incorporating all the equipment suggested in Table 4.21 in a mixer system costs around
GBP 10000 (1997 UK prices). However, due to the significant improvement in process control,
this capital investment will have a relatively short payback time. Although the cost of installing
an automated control system is approximately twice that for simple on-line monitoring,
substantial cost savings are possible.

Some example economic data are given in Table 4.22. These refer to the iron foundry described
under the operational data heading of this section.

Cost type EUR
Saving due to 60 % reduction in defective moulds 6416
Saving due to 10 % reduction in resin use 29966
Saving due to 10 % reduction in catalyst use 9050
Total saving in material cost 45433
Annual running cost 320
Estimated annual maintenance cost 24
Net cost saving 45088
Purchase cost of unit 24166
Payback period 7 months
Table 4.22: Annual savings, cost and payback for example plant mixer control
(1995 prices)
[75, ETSU, 1996]

Driving force for implementation
Optimisation of operational costs and minimisation of VOC emissions.

Example plants
Good binder management measures are used in the vast majority of the foundries using bonded
sand.

Reference literature
[71, ETSU, 1998], [75, ETSU, 1996], [202, TWG, 2002], [225, TWG, 2003]

4.3.3.2 Minimisation of mould and core sand losses

Description
Modern mould and core-making machines allow the production parameters of the various
product types to be saved in an electronic database. This allows easy change-over to new
products, without the loss of time and materials from finding the right parameters by simply
using trial and error. For new products, the settings for similar products may be used to shorten
the optimisation time.

Achieved environmental benefits
Reduction of the amount of waste sand and energy, through minimisation of the test run period.

Cross-media effects
No cross-media effects apply.

Chapter 4
Smitheries and Foundries Industry 191
Applicability
Series production of small series with high production throughput. An errorless change-over
requires good control and continuity of the sand quality.

Economics
The investment cost for a core shooting machine depends on the core volume, and range from
EUR 150000 (5 l) to EUR 400000 (100 l). Operational costs are 5 10 % of investment costs
per year.

Driving force for implementation
Optimisation of processes that require frequent change of core or mould type.

Example plants
This technique is applied in several foundries around Europe.

Reference literature
[110, Vito, 2001]

4.3.3.3 Best practice for cold-setting processes

- Phenolic: The temperature of the sand is kept as constant as possible, i.e. at around
15 C - 25 C, which is low enough in order to prevent emissions caused by evaporation.
Direct contact between the resin and catalyst has to be avoided as the reaction is exothermic
and can be violent

- Furan: The temperature of the sand is very important for this process and is kept as constant
as possible, around 15 25 C; in order to control the setting time of the binder and to
minimise catalytic additions. Direct contact between the resin and catalyst needs to be
avoided as the reaction is exothermic and can be violent

- Polyurethane (phenolic isocyanate): The sand temperature is kept between 15 and 25 C, to
maintain process control and to minimise emissions. Best practice is to work with three
pumps and to mix the catalyst and the phenolic resin with isocyanate and sand directly in
the mixer [202, TWG, 2002]

- Resol ester (Alkaline phenolic ester hardened): The temperature of the sand is controlled
and kept at an optimal level of 15 35 C. The process is less sensitive to temperature
fluctuations than other self-hardened systems. The velocity of hardening is managed by the
choice of the hardener type [225, TWG, 2003]

- Ester silicate: The sand temperature is controlled within the range 15 25 C. As moulds
and cores pick-up moisture, they need to be used as quickly as possible after achieving full
strength. Prolonged storage can only be done in dry conditions.

4.3.3.4 Best practice for gas-hardened processes

- Cold-box: Amine vapours need to be captured at the core-making machine. Any excess
amine is captured through the core-box. Additionally, it may also be necessary to ventilate
the core storage area. When possible, hoods are placed over the moulding or coring
machines as well as over the temporary core storage area.
Amine consumption can be minimised to the extent that its consumption matches the
production needs of the sand cores, since the cost of the raw material and its treatment is
high. Minimisation of amine consumption is helped by optimising the amine distribution
process within the core, typically through a simulation and optimisation of the gas flow
[202, TWG, 2002].
Chapter 4
192 Smitheries and Foundries Industry
The sand temperature has to be maintained as constant as possible, between 20 and 25 C,
otherwise a too low temperature will involve a longer gassing time, thus resulting in a larger
amine consumption. A too high temperature drastically reduces the lifetime of the prepared
sand.
Water is detrimental to this process. The moisture of the sand has to be maintained below
0.1 %, and the gassing and purging air must be dried.
Note: Amines are flammable and explosive in some proportions with air. Considerable care
is required in their storage and handling, and the supplier's instructions need to be observed
at all times.

- Resol (Alkaline phenolics methy formate hardened): It is advisable to ventilate the working
area, at least in order to prevent any danger of fire. Methylformate consumption needs to be
minimised to such an extent that its consumption matches the production needs of the sand
cores. The sand temperature has to be maintained higher than 20 C, in order to prevent the
condensation of methylformate. Methylformate vapours are heavier than air; this should be
kept in mind when designing the exhaust system.
Note: Methylformate is flammable when its concentration in air reaches 6 to 20 % and
explosive in some proportions with air. Considerable care is required in storage and
handling, and the supplier's instructions need to be observed at all times.

- Resol-CO
2
: The resin has low contents of unreacted phenol and formaldehyde, and their
emission levels, even during the gassing and purging periods, are very low. It is advisable to
ventilate the working area. [225, TWG, 2003]

- SO
2
hardened phenolic or furan: The working area has to be ventilated, and the emissions
collected at the emission point by totally hooding the moulding or coring machines and by
applying venting.
These collected gases must be treated before emission. This is easily completed by a
scrubber containing a sodium hydroxide solution. The pH and the concentration of the
scrubbing solution have to be continuously supervised. The scrubbing solution needs to be
occasionally replaced to remove concentrated salts, it then requires disposal as a hazardous
waste.
Sulphur dioxide consumption needs to be minimised to the extent that its consumption
matches the production needs of the sand cores.
Normally, the purging cycle lasts 10 times longer than the gassing cycle.

- SO
2
hardened epoxy/acrylic (free radical curing). As for gassing and purging, collection
and treatment are described in the SO
2
hardened furan resins section. The best gassing
conditions for achieving cores with good characteristics require:
the sand to be dry before use
the blowing or shooting air to be dried too
the concentration of sulphur dioxide in CO
2
or nitrogen to be from 5 % to 100 %,
depending on the resin used (minimal for acrylic, maximal for epoxy/acrylic)
the purging cycle to last 10 times longer than the gassing cycle.

4.3.3.5 Replacement of alcohol-based coatings with water-based coatings

Description
Coatings are applied to the surface of moulds and cores to create a refractory barrier at the
mould-metal interface and to ensure a good surface appearance. Coatings serve to reduce
veining, erosion and metal penetration defects and also help to reduce finishing operations.
Alcohol-based coatings are mainly based on isopropylalcohol. The coating is dried by
evaporation or by a burning-off of the solvent. This produces VOC emissions. Water-based
coatings have been developed as an alternative.

Chapter 4
Smitheries and Foundries Industry 193
The advantages of water-based coatings are:
- safety (no fire risk)
- workers health (less exposure to organic vapours)
- reduced costs of reagents (water vs. alcohol)
- mostly better surface finish of castings.

The difficulties for implementation are:
- the need for greater (process) time and space. Also cores need a longer drying time, which
results in the need for a drying line, passing through a drying furnace
- changing the process requires a case-by-case optimisation period
- growth of bacteria, which cause short shelf-life of coatings (1 - 2 weeks) and causes odour
emissions
- cost of re-approval by specific customers in aerospace and defence related applications.

Water-based coatings can be dried in ambient air, or by using a drying furnace or a microwave
or an infra-red furnace. They generally need a longer drying time compared to alcohol-based
coatings. The drying does not generate noxious emissions, but may cause odour emissions.
Drying is done using a drying line, which transports the cores from the core-making to the
mould assembly, thus allowing for the necessary drying time. The line may also pass through a
drying furnace. Microwave and infra-red drying is applied for small, medium and large series.

The different rheological properties of water-based coatings compared to solvent-based coatings
has necessitated the development of new application techniques. These allow a constant coating
quality.

Achieved environmental benefits
A life cycle assessment (LCA) study evaluated the impacts of water- and alcohol-based
coatings. The use of water-based coatings with drying in ambient air shows clear environmental
benefits, mainly due to the reduced emission of VOCs and the low energy use. When a stove is
used for drying, the reduced emissions are partially counterbalanced at the LCA-level by an
increased energy use. The overall environmental index using a drying stove shows a small
benefit over isopropyl alcohol (IPA) use and burning-off.

Cross-media effects
Aqueous coatings contain a variety of (organic) chemicals to enhance their properties. This may
affect the emission profile upon shake-out.

Drying the aqueous coating results in an increased volatilisation of binder solvents. This may
cause increased odour emissions, due to the evaporation of BTX-containing core solvents.

Water-based coatings require a higher use of energy due to transport (during air drying) and
drying in a furnace.

Drying in ambient air may result in a reduction of the air temperature in the foundry. In a Dutch
foundry, an overall reduction of the ambient temperature of 2 C was experienced. This results
in a higher consumption of fuel for heating during wintertime.

Operational data
Changing from IPA-based to water-based coatings is becoming more widespread in various
foundries nowadays. Automotive foundries have changed the majority of their production to
water-based, leaving solvent-based coatings being used only for specific applications (see
applicability).

Operational data were obtained from a Dutch foundry. When drying in ambient air, the air
velocity is the main factor (rather than temperature or humidity). In order to allow sufficient air
velocity over the drying trajectory, additional fans and ventilators were installed in the foundry.
In order to reach a good and constant quality, the fines content of the (regenerated) core-making
sand was reduced. Water-based coatings can cause a swelling of the fine residual (bentonite and
coal dust) particles, which then causes core defects.
Chapter 4
194 Smitheries and Foundries Industry
Operational data were obtained from a foundry in France, operating a stove with a capacity of
540 cold-box cores, which is heated with a natural gas burner. Two ventilators are used to
evacuate the fumes and to mix the fresh air. The design temperature of the hot air is 165 C. A
heat balance was set-up on the basis of measurements using 0.4 kg cores with a humidity of
5.5 %. The balance, as given in Figure 4.7, applies for a throughput of 390 cores, which is
equivalent to 72 % of full capacity.

Losses by
air renewal:
13%
Losses to
exhaust:
25%
Total losses:
50%
Wall losses:
12%
Heating of
cores:
6 %
Heating of
trays:
36%
Drying of
coating:
8 %
Total energy
supply to charge:
50%
Energy burner
Energy ventilators
Figure 4.7: Thermal balance of a stove for core drying, working at 72 % capacity
[143, Inasmet and CTIF, 2002]

The balance shows that 50 % of the energy is lost, with indeed effectively less than 10 % being
used for core drying. The balance gives a consumption of 8.5 kWh per kg of water evaporated,
which corresponds to 465 kWh per tonne of cores at a throughput of 156 kg/h. When the stove
works at a lower capacity rate, the losses increase further. Similar measurements in another
foundry confirmed the high (>50 %) losses and low (<10 %) effective energy use. Energy
efficiency can be increased using microwave drying. This will be discussed in Section 4.3.3.6.

Applicability
Water-based solvents can be used in most situations. However, alcohol-based coatings cannot
be replaced in some specific applications and will still need to be used:
- for big or complicated moulds/cores, where drying problems may occur due to difficult
penetration of drying air
- for water glass bonded sands
- in Mg casting: water creates Mg(OH)
2
and thus causes technical problems
- in the production of manganese steel with MgO-coating.

The implementation of water-based coatings is applicable for new and existing large-scale, large
series foundries. For new and existing small-scale foundries, the implementation may be
hindered by technical or economic factors.
Economics
The investment cost depends on several factors, such as the available space for a drying line in
the foundry, the choice of the drying technique, and the need to adapt the sand mix.

The complete replacement of alcohol-based coatings by water-based coatings in a Dutch iron
foundry, using ambient air drying involved a total investment of EUR 71000. This included the
installation of ventilators and optimisation of the sand quality. A reduction of the IPA-emission
of 161 tonne/yr corresponded to EUR 62.5/tonne IPA emission reduction.

Chapter 4
Smitheries and Foundries Industry 195
The operational costs involve increased controls of the coating and cores (viscosity, layer
thickness, water content, product quality) and amount to EUR 9000/yr. Drying 1 tonne of
coating consumes 2100 kWh of energy.

On the benefits side of the economics balance is the reduction in expenses for the purchasing of
solvent. For bigger foundries, examples exist where investment in a water-based system was
paid back in 2 years, on the basis of the reduced purchasing costs of the alcohol only.

The operational cost of drying is EUR 0.01/kg (non-ferrous foundry in France).

Driving force for implementation
Pressure from authorities increasing attention to the emissions of organics.

Example plants
- PSA Sept-Fons (F)
- Fonderie Bra, Montluon (F)
- De Globe b.v., Weert (NL).

Reference literature
[129, infoMil, 1999], [143, Inasmet and CTIF, 2002], [149, Beauvais and Choplin, 2001]

4.3.3.6 Microwave drying for water-based coatings

Description
Water-based coatings require a drying treatment. The application of microwave furnaces for this
purpose gives operational advantages over the use of ambient or heated air drying. Microwaves
of 2450 MHz frequency are used, and have the following properties:
- good selectivity of energy transfer: The microwave energy is transferred preferentially to
the water molecules, rather than the sand material. This implies a preferential and
homogeneous drying of the surface, even for complex cores
- self adaptive coupling: The drying mechanism provides a self-controlling homogenisation
of the humidity of the material
- quick drying: The speed of drying depends on the applied power. The high drying velocity
results in a limitation of the water-core contact time. This results in a good mechanical
stability of the cores.

The application of microwave drying presents difficulties in the following cases:
- heterogeneous loading of the furnace (with a range of core masses and volumes)
- use of cores or inserts with iron
- inhomogeneous drying of thicker parts of the coating
- risk of deformation of complex cores.

Achieved environmental benefits
Increased energy efficiency of the drying process.

Cross-media effects
Microwave drying also brings polymerisation reactions to a further completion. This reduces the
gaseous emissions in the proceeding stages of the foundry process (i.e. pouring, cooling, shake-
out).

Operational data
The use of microwave drying has not been implemented on an industrial scale, although large-
scale industrial tests have been performed. The results of these tests are given below. One test
campaign has been run using a 50 kW microwave furnace. The stove was loaded to 50 % of its
capacity with a charge of 800 kg of cores. Cores are placed on plastic trays, which are not
heated by the microwaves. The thermal balance is given in Figure 4.8. It shows that more than
Chapter 4
196 Smitheries and Foundries Industry
30 % of the energy input is used for drying. Additionally, the cores are only slightly heated up
(temperature at exit = 40 C), which allows immediate handling. The data from two
measurement campaigns are given in Table 4.23, and compared to analogous campaigns on hot
air furnaces (as discussed in 4.3.3.5).

Losses by air renewal:
52%
Total losses:
55%
Wall losses:
3%
Heating of
cores:
13 %
Drying of
coating:
32 %
Total energy
supply to charge:
45%
Energy microwaves
Energy ventilators
Figure 4.8: Thermal balance for the microwave drying of cores
for a 50 kW furnace working at 50 % capacity
[143, Inasmet and CTIF, 2002]

Foundry A Foundry B Foundry B Foundry C
Furnace type Hot air Hot air Microwave Microwave
Core load 156 kg/h 270 kg/h 800 kg/h 500 900 kg/h
Humidity 5.5 % 2.7 % 2.7 % 2 3.2 %
Thermal energy supply 73 kW 70 kW 50 kW 40 kW
Duration 48 min 113 min 8 min 8 min
Specific consumption per kg
evaporated water
8.5 kWh 9.3 kWh 2.3 kWh 2 kWh
Specific consumption per
tonne wet cores
465 kWh 280 kWh 63 kWh -
Table 4.23: Results of test campaigns on core drying, using hot air and microwave furnaces
[143, Inasmet and CTIF, 2002]

The data show that the drying time is considerably reduced from 1 2 h to less than 10 minutes.
Additionally, the energy consumption is reduced by a factor of 4 5, and the throughput is
increased.

Applicability
This technique is applicable for the drying of all water-based coatings. Furthermore, it allows
the greater applicability of water-based coatings to more complex core shapes and smaller
series.

Economics
Microwave drying represents high investment costs. The actual investment is proportional to the
power requirement, which corresponds to the amount of water to be evaporated. The high
investment is partially counterbalanced by:

Chapter 4
Smitheries and Foundries Industry 197
- higher energy efficiency, though the technique uses electricity rather than gas in the hot air
furnace
- reduced solvent cost for water-based vs. solvent-based
- reduced maintenance costs due to a more compact installation compared to the use of (hot)
air drying or a cooling line.

Driving force for implementation
Pressure from authorities, with increasing attention being paid to the emissions of organics,
combined with the need for higher energy efficiency.

Example plants
The technique is reported to be applied in foundries in France and Spain.

Reference literature
[143, Inasmet and CTIF, 2002], [225, TWG, 2003]

4.3.3.7 Use of non-aromatic solvents for cold-box core production

Description
Classical cold-box systems use organic solvents. These cause releases of noxious and odorous
VOCs during core production and storage. Furthermore VOC emissions (benzene, toluene,
xylene) occur during pouring, cooling and shake-out. Alternative solvents for cold-box core
production are based on protein or animal fat (e.g. vegetable-oil methyl esters) or on silicate
esters. These solvents are not noxious to workers health, nor inflammable and therefore
transport and storage are easier.

The low volatility of the vegetable-oil methyl esters enhances the storage capabilities in humid
atmospheres and their persistence when applying water-based coatings.

The resulting cores have a higher strength, lower sticking of the sand and good shake-out
properties.

Achieved environmental benefits
There is no vaporisation due to the high boiling point (about 300 C) and thus there is no solvent
odour emission during storage.

VOC emissions are reduced during core manufacturing and storage, and (more importantly)
during pouring, cooling and shake-out.

Cross-media effects
Vegetable-based solvents cause increased fume production during pouring and (in the case of
die-casting) upon opening of the die. This is due to the low evaporation and consequently high
residual solvent content upon pouring. This effect does not occur with silicate ester solvents.

Protein and animal fat based solvents produce a distinctive odour, and have been reported to
cause odour problems inside the foundry.

Operational data
Table 4.24 shows operational emission values for benzene, toluene, xylene and phenol, as
measured in one aluminium green sand foundry. Measurements were performed both at the
shake-out grate and in the exhaust-air stack. Data are given for the traditional cold-box system
and for one using vegetable-oil based solvents. A clear reduction in BTX and C-emissions is
observed (25 50 % of stack emissions).

Chapter 4
198 Smitheries and Foundries Industry
At shake-out At exhaust-gas stack
Compound Aromatic
based
(mg/m)
Vegetable-
based
(mg/m)
Difference
(%)
Aromatic
based
(mg/m)
Vegetable-
based
(mg/m)
Difference
(%)
Benzene 0.08 0.05 - 44 0.1 0.07 - 30
Toluene 0.12 0.05 - 58 0.08 0.06 - 25
Xylene 0.09 0.04 - 56 0.09 0.05 - 41
Phenol 14.8 14.6 - 2 7.2 6.57 - 9
Total C 61.0 29.5 - 52 37.0 18.5 - 50
Table 4.24: Emission data for cold-box core systems in an aluminium foundry, measured at the
shake-out grate and in the exhaust-air stack
[189, Httenes-Albertus, 2002]

Results of emission measurements in a German iron foundry are given in Table 4.25. For this
table, the measured emission of each compound is expressed relative to the emission of the
same compound for the aromatic-based system. For example: the emission of toluene at the
second cooling is only 58 % of the toluene emission for the aromatic system. The table shows a
clear reduction of VOC emissions in the second cooling and shake-out phase.

Benzene Toluene Xylene Phenol
Pouring and cooling 1 83 100 100 100
Cooling 2 78.5 58 46 74
Shake-out 78 78 78 12
Emission data expressed as %, relative to emission of the aromatic based system
(set at 100 % for each measurement)
Table 4.25: Emissions (%) of selected compounds from vegetable-based cold-box core systems
[216, Hobelsberger, et al., 1997]

The total mass flow reduction of the specified compounds was:
- benzene: - 21 %
- toluene: - 26 %
- xylene: - 30 %
- phenol: - 62 %

A survey measurement of the total carbon emission over the full foundry process is given in
Figure 4.9. The data show that the vegetable-based solvents partially result in a shift of
emissions from the core-making area to the finishing area. Nevertheless an overall reduction of
17 % carbon emission reduction is reached.

Chapter 4
Smitheries and Foundries Industry 199
100 100 100 100 100 100
44
134
98
212
83
138
0
50
100
150
200
250
Core drying
Pouring and
cooling 1
Cooling line 2 Shake-out Product handling Total
T
o
t
a
l
C
e
m
i
s
s
i
o
n
(
%
)
Aromatic based
Vegetable based
Figure 4.9: Total carbon emissions from various process steps, using aromatic- and vegetable-based
solvents
(values in % expressed relative to emission of the aromatic based system)
[216, Hobelsberger, et al., 1997]

Applicability
Non-aromatic solvents can be applied in all processes using cold-box sand. The technique is
relatively new and operational results as stated above need to be confirmed through repeated
application. The technique has been implemented in foundries for the full range of automotive
castings, as well as in castings for machinery, and in the construction and electronics industry.
Both ferrous and non-ferrous applications are reported.

Economics
Data from the Italian foundry association indicate that non-aromatic solvents have a cost that is
double that of the aromatic equivalent, e.g. EUR 0.82/kg against EUR 0.36 0.48/kg.

Operational data from a large series ferrous foundry showed that binder prices for aromatic cold
box were around EUR 2/kg (2001) and went up with some 30 %, upon changing to the non-
aromatic system. After one year (2003), the price lowered again however, to a level 10 % above
the price of the traditional system.

Driving force for implementation
To reduce VOC and odour emissions.

Example plants
The technique is reported to be used in large-scale foundries in Germany.

Reference literature
[143, Inasmet and CTIF, 2002], [189, Httenes-Albertus, 2002], [202, TWG, 2002], [216,
Hobelsberger, et al., 1997], [225, TWG, 2003]

Chapter 4
200 Smitheries and Foundries Industry
4.3.4 Alternative moulding/core-making methods

4.3.4.1 Lost Foam casting

Description
The general principles of Lost Foam casting have been described in Section 2.5.7.1. Due to the
absence of binders, the technique produces a reduced amount of solid waste and emissions
compared to sand casting methods.

The technique allows the production of pieces with tighter tolerances, less feeder and riser
channels and better as-cast features. This all results in less machining and clean-up time.

Lost foam foundries include a foam workshop, a melting shop and a casting shop. They do not
have an extensive core-making shop and finishing shop. The foam shop involves foaming
machines, steam generators and coolers and ambient air dryers.

The Lost Foam models are made of polystyrene (EPS) or PMMA, with small amounts of
pentane, glue and a mineral coating. Since both EPS and pentane are pure hydrocarbons,
organic carbons are formed upon pyrolysis of the model. In order to minimise emissions of the
organic decomposition products of EPS, post combustion of the off-gases is performed.

The technique uses unbonded sand, which results in no binderrelated emissions upon pouring
and shake-out and which may be internally re-used without major treatment. EPS pyrolysis
results in a slow build-up of organic material in the sand. This may be prevented by de-dusting
and partial renewal of the circulating sand. A 5 % renewal is sufficient to maintain sand quality.
The evacuated sand may be thermally regenerated for internal re-use.

Achieved environmental benefits
Since unbonded sand is used, the technique shows no binder-related emissions upon pouring.
However pyrolysis of the EPS or PMMA results in organic decomposition products, which
necessitate post combustion.

The use of energy in the Lost Foam process is significantly lower than in conventional casting
methods. This is mainly due to the reduced energy uses for post casting operations, melting and
sand preparation. The production of cores for conventional moulding methods is relatively more
energy intensive than the production of the foamed model.

LCA analysis has confirmed that the environmental benefit as stated above, can be extended to
an overall better environmental performance of the Lost Foam process over sand casting, in as
far as the casting of core-intensive castings are concerned. For simple castings (e.g. lids), the
life cycle approach shows that there is no overall better performing process.
Cross-media effects
The technique represents a reduction in energy consumption and a decrease in mould making
and casting finishing operations. The off-gases show increased values of BTEX and
formaldehyde, but emissions only occur during pouring and they can be more easily captured
and treated. The technique produces a reduced amount of residual sand and dust and the sand
may be more easily regenerated (shortened thermal treatment).

An LCA-based assessment of the Lost Foam process (also taking into account processes outside
the foundry borders) shows a lower overall environmental impact for the production of core-
intensive castings.

Operational data
Emission data and data on residue production have been given in Section 3.9.6.1. The technique
results in lower residue production compared to a sand foundry of equal capacity. Flue-gases
need more intensive cleaning (post combustion) and (after cleaning) result in higher organic
compound (BTEX, formaldehyde) stack emissions.
Chapter 4
Smitheries and Foundries Industry 201
Comparative consumption levels for the production of a comparable compressor casing using
sand casting and Lost Foam casting are given in Table 4.26. This shows the higher amount of
return material in the melt and the lower weight of the casting. For the Lost Foam method, the
total amount of moulding-sand is greatly increased, while no cores are used.

Input material Green sand Chemically-bonded
sand
Lost Foam
Cast iron 98.0 98.0 96.1
Foundry returns 30.0 30.0 38.1
Weight casting 68.0 68.0 58.0
Moulding-sand 256.8 233.0 1101.4
Core sand 122.0 150.7 n.a.
Weight foam model + feeder n.a. n.a. 0.212
All data in kg
n.a: not applicable
Table 4.26: Operational data for the production of a similar cast iron compressor casing, using
various methods
[130, de Wilde and ten Houten, 1999]

Applicability
This technique applies to the serial production of small and medium sized products in ferrous
and non-ferrous metal. Maximum casting sizes are 1000 x 1000 x 550 mm. The technique finds
most of its implementation in aluminium foundries and has a 2 % share of aluminium casting.

Since the technique implies a basic change in the mould making procedure and infrastructure, it
applies primarily to new installations. Application in existing foundries requires a conversion of
the foundry process with regard to the moulding, pouring and finishing operations and a review
of these stages for each of the castings produced. To introduce Lost Foam for a given casting
requires funds, manpower and the necessary time and flexibility, as well as the co-operation of
the customer.

Economics
Economic data for an aluminium Lost Foam installation are given in Table 4.27. Investment
costs include equipment, installation, start-up and training.

Investment costs Description EUR
Model making equipment Model making machine
Steam generator
Coating installation
1300000
Casting equipment Automatic casting machine 2540000
Off-gas cleaning Dry dedusting
Thermal regenerative treatment
608000
Sand preparation Sieving, cooling, transport 160000
Total Investment, installation, start-up, training 4608000
Operational cost EUR/tonne good
casting
Consumption EPS-granules, glue, coating, steam, combustibles 202
Foundry details: aluminium castings, 5500 tonne/yr production good castings, 6864 tonne/yr production molten metal
Table 4.27: Economic data for an aluminium Lost Foam foundry
[96, Spillner, 1997]

Driving force for implementation
To reduce wastes from sand casting and to reduce energy consumption.

Example plants
Several example plants are reported in Germany and France.
Chapter 4
202 Smitheries and Foundries Industry
Reference literature
cast iron: [130, de Wilde and ten Houten, 1999]
aluminium: [96, Spillner, 1997], [143, Inasmet and CTIF, 2002]

4.3.4.2 Ceramic shell moulding

Description
Ceramic shell moulding is a patented process (Replicast) in which a polystyrene model is
covered with a ceramic shell (2 3 mm thickness), based on ethyl silicate and refractory sand.
The shell is hardened using ammonia and sintered at 1000 C. The sintering cures the shell and
burns away the polystyrene model. The metal is then poured into the shell.

The sintering furnace is equipped with post combustion to reduce emissions.

The technique allows a casting design with no parting lines, cores and draft angles and reduced
feed metal. This reduces the need for finishing and machining operations.

Achieved environmental benefits
The technique results in a minimisation of dust emissions from moulding and finishing, as
compared to sand moulding. The emissions of VOC are eliminated as no gas is emitted for the
inert ceramic mould. Additionally the amount of waste (dust, metal) is reduced. The reduction
of feeder systems results in a higher yield of castings per melt.

Cross-media effects
No cross-media effects apply.

Operational data
Operational data from the owner of the technique are reported in Table 4.28, for the casting of
valves, as compared to sand casting.

Property Relative weight reduction
(%)
Finished weight of castings 20 26
Feed metal 24 60
Poured metal 24 50
Table 4.28: Relative weight reduction for ceramic shell casting of valves as compared to sand
casting
[219, Castings Technology International, 2003]

Applicability
The technique applies for the fabrication of castings which need a high surface finish, in low-
carbon alloys such as ultra low carbon stainless steel and nickel-based alloys with a finished
weight of up to 550 kg. The ceramic shell process is a patented process and may only be used in
accordance with the terms and conditions of the licence granted to the licensee.

Driving force for implementation
Regulation of VOC emissions and high disposal costs for residual wastes.

Example plants
- CMS srl, Urbisaglia (I): 2 automated shelling lines, producing steel valves up to 150 kg
finished weight
- Saint-Gobain SEVA, Chalons-sur-Saone (F): steel and super alloy components for glass
industry.

Reference literature
[110, Vito, 2001], [219, Castings Technology International, 2003]
Chapter 4
Smitheries and Foundries Industry 203
4.3.5 Permanent (metal) moulds and pressure die preparation

4.3.5.1 Minimisation of release agent and water consumption

Description
A water-based solution of release agent is sprayed on the open HPDC die before closing it. This
allows cooling and coating of the die. Some simple process measures allow the minimisation of
the release agent and water consumption. These will also prevent the formation of a mist. These
measures are:
- Automated spraying process: Robotisation of the spraying process allows a careful control
of the amount of release agent used and adaptation of the amount used to the local needs of
the cast piece
- Optimise dilution factor: The release agent dilution factor should be chosen such that the
spraying action has the required balance between coating and cooling of the die
- Apply in-die cooling: The cooling action may be partially taken over by internal cooling
using an integrated water circuit.

Achieved environmental benefits
Minimisation of water and chemicals consumption. Prevention and/or reduction of (diffuse)
emissions.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to HPDC foundries. The extent to which the minimisation can be applied
depends on the type of castings and the machines used.

Driving force for implementation
Minimisation of consumptions and emissions.

Example plants
This technique is applied in major HPDC plants, e.g. in the manufacture of car parts (D, F).

Reference literature
[202, TWG, 2002]

4.3.5.2 Closed-mould application of release agent

Description
A release agent is applied to the closed die-casting mould in a vaporised form. The high
temperature results in a condensation and deposition of a release agent film. This technique is an
alternative to the spraying of a water-based solution of the release agent onto the open die. It
allows the water usage to be reduced and also reduces the consumption of release agents. The
active substances for the process, i.e. getting the casting out of the die, are very similar to the
open-mould technique: paraffin wax, aliphatic carbons, polysiloxanes, polyglycol.

Water spraying, however, also causes the necessary cooling of the die surface. The alternative
technique therefore requires enhanced temperature management (mainly by a cooling effect), by
means of an internal die cooling system. This may be obtained by reducing the thermal oil
temperature. Combined water spraying and closed-mould release agent application is possible in
places where the moulds cannot achieve complete thermal equilibrium by means of internal
cooling systems alone.

Chapter 4
204 Smitheries and Foundries Industry
Achieved environmental benefits
The technique allows a reduction in the usage of both water and the release agent. This results in
a lower amount of waste water and reduced emissions of steam containing oily particles. The
active compounds of the release agent are comparable to the traditional ones. Emissions due to
decomposition of the release agent (and upon opening the die to get the casting out) therefore
are comparable in nature to those with the open-mould technique.

Cross-media effects
The technique requires an increased effort to cool the dies. In simple cases, this may involve a
lowering of the thermal oil temperature, but for more complex dies, it may involve the
application of an additional cooling circuit.

Operational data
Some operational examples in aluminium foundries have been reported. These result in a 50
80 % reduction in the spraying time, and a 80 % reduction in sleeve-parting agent consumption.

Applicability
The technique is applicable in existing installations, but may require the production of specific
dies with integrated or adapted cooling circuit. The applicability is limited to specific types of
castings, dies and casting machines and is mainly based on technical considerations. This
technique is not a general alternative for the application of release agents.

Driving force for implementation
Foundry strategy to reduce waste water production and to reduce the consumption of water and
chemicals.

Example plants
Some operational examples in aluminium foundries have been reported.

Reference literature
[189, Httenes-Albertus, 2002], [202, TWG, 2002]

4.4 Metal casting

4.4.1 Improvement of metal yield

Description
Metal yield is defined as the ratio of metal melted to the weight of finished castings. The
difference between the two values is due to metal losses (e.g. melting losses, spilt metal,
grinding losses) and return metal (e.g. pigged metal, runners, scrap castings). An improvement
of the metal yield involves reducing the metal loss and amount of return metal.

Improving metal yield is possible using one or more of the following measures:
- applying on efficient method: proper design of runners, risers, ingates, pouring basin and
optimised box yield (castings/metal poured in a mould). A valuable tool in designing on
efficient method is the use of computer aided simulation of pouring and solidification
- applying good procedures in melting and pouring operations: in order to reduce melting
losses, excessive pigging, scrap rates, etc.
- applying good practice in the moulding and core-making departments: in order to reduce
scrap due to deficiencies in the mould- and core-making operations.

Achieved environmental benefits
Increasing the metal yield results in a lower consumption of energy, sand and additives per unit
of good casting. The overall efficiency of the process is increased.

Cross-media effects
No cross-media aspects apply to this technique.
Chapter 4
Smitheries and Foundries Industry 205
Operational data
Table 4.29 provides a guide to the typical yields obtained for different types of iron casting
production. It is not possible to recommend a target yield figure which can usefully be adopted
by a particular foundry, as so much depends on the type of metal, type of casting, the production
facilities and the market which has to be served. It is therefore necessary that a foundry sets
individual targets, based on an analysis of its own current performance.

Casting type Yield (%)
Heavy grey iron, simple shape 85 95
Medium sized grey iron jobbing or small batch 65 75
Mechanised repetition, general quality small to medium sized grey iron engineering and
municipal castings
65 75
Mechanised repetition, high quality small to medium sized grey iron engineering castings,
relatively simple design
60 65
Mechanised repetition, high quality small to medium sized grey iron engineering castings,
complicated heavy cored design
55 60
Medium sized nodular iron jobbing or small batch 50 60
Small or very small grey iron repetition 45 55
Mechanised repetition of malleable iron and small nodular iron castings 40 50
Table 4.29: Typical metal yields for different iron casting types
[45, ETSU, 1990]

In response to a questionnaire, 82 UK foundries reported their metal yield performances for the
period 1981 - 1987. The improvement in yield per metal type (weighted average) are given in
Table 4.30.

Yield in 1981
(%)
Yield in 1987
(%)
Improvement in yield
(%)
Grey iron 60.5 63.0 2.5
Nodular iron 51.8 55.7 3.9
Malleable iron 36.4 39.2 2.8
Table 4.30: Weighted average of yield performances in 82 UK ferrous foundries, 1981 - 1987
[45, ETSU, 1990]

Applicability
The technique is applicable in all existing ferrous and non-ferrous foundries.

Economics
Improving the metal yield is possible through simple low-cost practical measures and controls.
The benefits may be high since every percentage improvement results in a corresponding
reduction of melting energy consumption and a reduction in the consumption of sand and
chemicals.

The introduction of a computer-aided simulation for pouring and solidification requires
investment and training costs.

Driving force for implementation
Optimisation of process efficiency.

Example plants
Metal yield considerations are part of good operational practice in the majority of European
foundries.

Reference literature
[45, ETSU, 1990], [140, EU Thematic Network Foundry Wastes, 2001]
Chapter 4
206 Smitheries and Foundries Industry
4.5 Fume, flue-gas and exhaust air capture and treatment

4.5.1 General principles

Various process steps in the foundry have the potential to produce dust, fume and other gases,
e.g. material storage, handling and processing. Techniques to reduce emissions to air involve
prevention, minimisation and fume collection.

Furnace sealing (or the use of sealed furnaces) combined with process control may be applied to
prevent or contain emissions from a process plant. Sections 4.5.2 4.5.6 covering furnaces
indicate where furnace sealing is possible and where other collection techniques may be used to
provide integral gas collection.

Other techniques are available to collect the emissions that cannot be prevented or contained.
Gases and fumes that escape from the processes are released into the working area and then
escape into the surrounding environment. They may affect operator health and safety and
contribute to the environmental impact of the process. Process gas collection techniques are
used to prevent and minimise these fugitive emissions. Hoods are designed to be as close as
possible to the source emission while leaving room for process operations. Movable hoods are
used in some applications. Some processes use hoods to collect primary and secondary fumes.

Fugitive emissions may be very important, but are hard to measure and quantify. Methods of
estimating ventilation volumes or deposition rates can be used to estimate them. One reliable
method, which has been applied to primary copper smelting, shows that the magnitude of
fugitive emissions can be much more significant than collected and abated emissions. Fugitive
emissions can be more than two to three times the quantity of controlled emissions. [155,
European IPPC Bureau, 2001]

The melting shop, core-making shop, sand plant and post-casting shop are considerable sources
of emissions. The emitted air pollutants are mainly dust (possibly with heavy metal particles),
sulphur dioxide, carbon monoxide and odorous organic compounds. Table 4.31 gives the results
of a survey of the pollutants generated in the different parts of the ferrous foundry process. Both
inorganic and organic compounds are listed as individual and group compounds. Dust emissions
are of specific importance, since thermal processes can generate considerable amounts of heavy
metals.

Chapter 4
Smitheries and Foundries Industry 207
SOURCE
RELEASES
R
a
w
m
a
t
e
r
i
a
l
s
t
o
r
a
g
e
a
n
d
h
a
n
d
l
i
n
g
F
u
r
n
a
c
e
o
p
e
r
a
t
i
o
n
s
D
e
s
u
l
p
h
u
r
i
s
a
t
i
o
n
o
f
m
o
l
t
e
n
i
r
o
n
N
o
d
u
l
a
r
i
s
a
t
i
o
n
P
r
e
p
a
r
a
t
i
o
n
o
f
c
o
r
e
s
a
n
d
m
o
u
l
d
s
C
a
s
t
i
n
g
S
h
a
k
e
-
o
u
t
,
r
e
c
l
a
m
a
t
i
o
n
F
e
t
t
l
i
n
g
,
d
r
e
s
s
i
n
g
a
n
d
f
i
n
i
s
h
i
n
g
o
f
c
a
s
t
i
n
g
s
Oxides of sulphur X X X X X
Oxides of nitrogen X X X X
Carbon dioxide X X X X X X
Carbon monoxide X X X X X X
Hydrogen sulphide X X X
Ammonia X X X
Oxides of iron X X X X X X
Alkali metal compounds X X
Alkaline-earth metal compounds X X X X
Metal oxide particulates X X X X X X
Non-metallic particulates X X X X X X X
Metallic iron X X
Hydrogen cyanide X
Sulphur X
Amines/amides X X
Dioxins X
Volatile organic compounds X X X X
Acid vapours X X X
Noise X X X X
Substances include their compounds, except where separate reference to the compound is made.
Releases to air may also be released to land or water, depending upon the abatement technology
employed, e.g. via collected dusts, sludges or liquors.
Some releases are specific to a particular binder system
Table 4.31: Survey of air emissions from different ferrous foundry stages
[160, UK Environment Agency, 2002]

The principles and techniques for waste gas collection and treatment are the same for foundries
and (primary) non-ferrous metal industries, therefore for a full discussion of the principles the
reader is referred to the BREF document on non-ferrous metal industries. Section 4.5.1.3 details
a short survey of the techniques that apply.

4.5.1.1 Reduction of fugitive emissions

Description
Fugitive emissions occur when emissions from specific process sources are not collected.
Besides the process-related emission sources mentioned in Section 4.5.1, potential sources of
fugitive emissions to air include:

Chapter 4
208 Smitheries and Foundries Industry
- storage areas (e.g. bays, stockpiles, heaps)
- the loading and unloading of transport containers
- transferences of material from one vessel to another (e.g. furnace, ladle, silos)
- the mixing and curing of chemical binders (inorganic and organic chemical emissions)
- mould coatings (solvents)
- conveyor systems for moving material around
- pipework and ductwork systems (e.g. pumps, valves, flanges, catchpots, drains, inspection
hatches, etc.); these type of emissions are discussed in detail in the LVOC-BREF
- poor building containment and extraction
- bypass of abatement equipment (to air or water)
- an accidental loss of containment from a plant or equipment failure, including leakages, e.g.
from the sand reclamation plant
- spills.

In order to minimise fugitive dust emissions, the following techniques may be employed:
- the covering of skips and vessels
- the avoidance of outdoor or uncovered stockpiles
- where outdoor stockpiles are unavoidable, using sprays, binders, stockpile management
techniques, windbreaks, etc.
- cleaning wheels and roads (i.e. avoiding the transfer of pollution to water and wind)
- using closed conveyors, pneumatic conveying (although note the higher energy needs), and
minimising drops
- vacuum cleaning of the moulding and casting shop in sand moulding foundries, with the
exception of areas where the sand has a technical or safety-related function, e.g. the pouring
area, and with the exception of hand-moulding jobbing foundries
- keeping outside doors shut, e.g using an automatic shutter system or flaps
- carrying out good housekeeping, i.e. ensuring that regular inspections are carried out by
responsible and delegated staff as a practice of good housekeeping and keeping up-to-date
records.

Fugitive emissions to water may arise from subsurface structures or defective surfacing. These
types of fugitive emissions can be minimised by the following actions:
- establishing and recording the sources, direction, and destination of all installation drains
- identifying and recording all subsurface sumps and storage vessels
- carrying out routine programmed inspections
- having an inspection and maintenance programme of impervious surfaces and containment
kerbs
- being able to justify where operational areas have not been equipped with:
an impervious surface
spill containment kerbs
sealed construction joints
connection to a sealed drainage system.

Achieved environmental benefits
A reduction of uncontrolled emissions.

Cross-media effects
No cross-media effects apply.

Applicability
These techniques are applicable to all new and existing installations.

Driving force for implementation
Limiting fugitive emissions will limit the overall odour and dust emissions to the close vicinity
of the plant. For installations in crowded areas, maintaining good relations with neighbours
could therefore be a driving force for implementation.

Chapter 4
Smitheries and Foundries Industry 209
Example plants
Vacuum cleaning of the moulding and casting shop: applied in various visited green sand
foundries.

Reference literature
[160, UK Environment Agency, 2002], [163, UK Environment Agency, 2002], [155, European
IPPC Bureau, 2001]

4.5.1.2 Use of a multi-flue stack

Description
In order to obtain maximum advantage from thermal buoyancy, hot emissions may be combined
into the minimum practicable number of chimneys. A multiplicity of discharge points may thus
be avoided. This is particularly important when new plants are being designed or when changes
are being made to existing processes.

Achieved environmental benefits
Gathering multiple off-gas streams into one stack allows control of the emissions and increases
the total volume (and load) going to treatment and thus reduces the total emission level.

Cross-media effects
No cross-media effects apply.

Applicability
This technique can only be implemented in new plants or when existing installations require
substantial changes.

Driving force for implementation
To increase the possibility to measure and control the foundry operation.

Example plants
This technique finds multiple implementation around Europe.

Reference literature
[160, UK Environment Agency, 2002]

4.5.1.3 Abatement techniques

Description
For cleaning collected off-gases, various wet and dry systems may be used. Selection of the
appropriate technique depends on the composition, flow and conditions of the off-gas stream.
The design of the abatement process is critical. Factors such as efficiency, suitability of the
method and the input and output loading of the material to be collected are used. A full
discussion of the working principles is given in the BREF document for non-ferrous metal
industries. The following techniques are used in the foundry industry.
- Dust and particle removal:
cyclones
fabric or bag filters
wet scrubbers
- Gas scrubbing systems (SO
2
, Cl, amine removal):
wet scrubbing using scrubber towers, venturi scrubbers and disintegrators.
- Oil-mist separators:
wet electrostatic precipitators
- CO and organics removal:
post combustion
biofilter.
Chapter 4
210 Smitheries and Foundries Industry
For dust and particle removal, both wet and dry systems are used. The main advantage with
using dry systems is that the dust is captured dry, which allows its subsequent re-use.
Furthermore, no pollution is transferred to another medium, as is the case for wet systems. For
gaseous compounds, such as SO
2
and chlorides, bag filters are inefficient, since they are not
adsorbed onto the filter surface. Experiments with lime injection have not provided good results.
The following describes the system used for dust and particle removal:

- Cyclones: If the right measures are taken (i.e. heat resistant steel, refractory lining) a
cyclone filter may be used for dedusting hot flue-gases (500 600 C). The efficiency is too
low to achieve emission levels in the 20 mg/Nm range. The cyclone is mainly used as a
spark separator in front of a bag filter. Cyclones are used as a pretreatment step to other
abatement systems
- Multicyclones: The separation efficiency of a cyclone increases with decreasing diameter.
The use of a parallel series of small cyclones allows the separation of small dust particles,
without a sharp increase in the pressure drop over the cleaning equipment
- Bag filter: This type of separator is widely applied in various parts of the foundry process,
due to its good efficiency and low emission values. It can achieve good efficiencies in
controlling the fine particulate matter found in melting operations. Sub-micron particles
such as metallic oxides may be separated as well. For proper operation, the following
measures need to be taken: flue-gas cooling (T = 130 - 160 C) and separation of sparks
(using a cyclone). The post combustion of organic material (to reduce the fire risk) may be
applied for an off-gas with a high VOC content. Post combustion as a "firewall" for bag
filter protection has been reported in some cases for off-gas from the unpacking of drums.
This technique is not applied in general; instead waste gas streams with high dust loads are
mixed with gas streams with increased VOC contents to prevent the filter and the exhaust
ducts from becoming sticky. A view of the inside and outside of a bag filter unit is given in
Figure 4.10.

Figure 4.10: Bag filter units; inside (l.) and outside (m., r.) views
[237, HUT, 2003]

- High temperature filter systems (using a ceramic filter medium): These are available on the
market but not currently applied in the foundry industry

- Electrostatic precipitators (ESP): These are not widely applied for foundry flue-gas
dedusting. Due to their sensitivity to gas flow, gas temperature and humidity, they are only
suitable for continuous melting regimes. They also incur a considerable explosion risk, due
to the high gas volume they contain. The removal of dust to reduce this risk requires
frequent cleaning, and therefore might incur economic problems. The main field of
application for electrostatic precipitators in foundries is oil/mist removal from pressure die-
casting exhaust gases.

Wet dedusting systems, such as venturi and disintegrators, are applied in melting flue-gas
treatment. Scrubber towers are used for dedusting of non-melting off-gas. Compared to dry
systems the wet systems have the following disadvantages: higher energy consumption, higher
maintenance (corrosion, bacteria), and they result in waste water and a sludge for disposal.
Chapter 4
Smitheries and Foundries Industry 211
Advantages are the capture of water-soluble compounds (such as SO
2
, chlorides), quick cooling
- which prevents dioxin reformation, low investment costs, and less restriction on input
temperature.

- Venturi scrubbers: Water is sprayed into the gases as they pass through a venturi. The
acceleration of the gas flow in the venturi throat causes an intensive mixing of both media.
The dust particles are damped, making them heavier, so that they can be separated in a
cyclone or other system placed downstream. If the gas flow drops, the venturi throat is
adjusted to maintain the collection efficiency
- Disintegrators: These so-called dynamic scrubbers consist of concentric rotor and stator
mounted pins through which the gas stream is driven by means of a fan placed downstream
or by fan blades at the outer end of the rotor. Water injected into the centre of the rotor, is
broken into fine droplets by the pins and dispersed in the gas stream. The wet particles
impinge on the stator walls and are collected at the bottom of the disintegrator. The system
works efficiently when the gas flow is reduced.

Post combustion systems and biofilters will be discussed in Sections 4.5.2 and 4.5.9,
respectively.

Achieved environmental benefits
Reduction of emissions to the atmosphere.

Cross-media effects
Wet scrubber systems result in the production of a waste water flow, which then needs
treatment. The cleaned water may be recirculated into the process. Sludge cakes, produced from
the waste water treatment need to be disposed of.

If the foundry is located in a cold climate (heating season more than 6 months), all increases in
air ventilation will also cause a growth in the energy consumption, since the workplace has to be
heated more.

The use of all abatement systems require the use of energy to induce the draught of the gases
through the abatement system.

Operational data
A comparison of the properties of wet and dry systems is given in Table 4.32. In the subsequent
sections, techniques that are specific for the various melting furnaces and for the various process
stages will be discussed.

Chapter 4
212 Smitheries and Foundries Industry
Dry systems

Wet systems Abatement
techniques
Multi cyclone Bag filter Venturi Disintegrator
Dust emission
level*
100 200 mg/Nm <5 20 mg/Nm <20 150 mg/Nm <20 150 mg/Nm
Investment cost Low High Low Medium
Energy
consumption
Low Low-medium High High
Advantages/reason
for choice
Applicable for pre-
cleaning of gases
prior to other
methods
Good performance
for suitable dusts if
well monitored.
The potential to
recycle dust to the
process
Partial SO
2
-
capture
Low risk of de
novo synthesis
Compact
installation
Low risk of de
novo synthesis
Disadvantages Low efficiency
when there is a
disturbed flow
pattern (dust
blocking of
distributor).
limited efficiency
for fine particles
Fire risk, large
volume, blocking
upon condensation
Wet sludge, waste
water treatment, a
loss of efficiency
with wearing
Higher energy use,
wearing, wet
sludge, waste
water treatment
* Values from operational practice, that can be maintained throughout the service life of the installation
Table 4.32: Properties of wet and dry abatement systems for foundries
[110, Vito, 2001], [155, European IPPC Bureau, 2001], [202, TWG, 2002]

Figure 4.11 gives a comparison of the operational conditions for hot blast cupola furnaces with
a wet and dry dedusting system. The main differences are the temperature profile of the flue-gas
and the energy consumption.

CUP Combust. Sec. heat Scrubber Recup.
Cyclone
Cyclone
Combust. Recup. Sec. heat Filter
35 C
20 mg/m
10 mg/m
3
10 mg/m
3
- 10 kWh/t
Sludge + 0.23 MW/t
- 25 kWh/t
600 C
Dust
> 0.1mm
200 mgCO/m
Dust + 0.23 MW/t 600 C 200 mg CO/m
Dust
> 0.1mm
950/800 C 650 C
CUP
950/800 C
170 C
Wet dedusting
Dry dedusting
Figure 4.11: Operational conditions for wet and dry dedusting of hot blast cupola off-gas
[230, CAEF, 2003]

Chapter 4
Smitheries and Foundries Industry 213
Applicability
The applicability of the various abatement systems will be discussed in the following sections of
this chapter.

Economics
Investment costs and energy usage for bag filters and wet separators are compared in Table
4.33.

Abatement technique Investment cost*
(EUR/Nm)
Energy consumption
(kW/1000 Nm)
Bag filter 2.5 5 1 3
Wet separator 1.5 5 1 3
Biofilter 7.5 10
* Excluding pipes and ducts, but including assembly
Table 4.33: Investment and energy consumption data for various abatement systems
[32, CAEF, 1997], [110, Vito, 2001], [202, TWG, 2002]

Driving force for implementation
Air emission regulations.

Example plants
This technique finds general application in European foundries.

Reference literature
[32, CAEF, 1997], [110, Vito, 2001], [155, European IPPC Bureau, 2001], [202, TWG, 2002]

4.5.1.4 Dioxin prevention and abatement

Description
Dioxins are relevant for thermal processes which have metals present. Dioxins or their
precursors may be present in some raw materials and there is a possibility of de-novo synthesis
in furnaces or abatement systems. Dioxins are easily adsorbed onto solid matter and may be
collected by all environmental media as dust, scrubber solids and filter dust. Field tests have
shown that the formation of dioxins in cupola furnaces cannot be correlated to one (or a few
single) operational parameter(s). A combination of measures is needed to minimise the risk of
dioxin formation.

In-process or primary measures to prevent dioxin emissions include:
- post combustion of the furnace off-gas in the CBC shaft or in a HBC combustion chamber.
Combustion of cupola off-gas is fully described in Sections 4.5.2.2 and 4.5.2.3
- continuous temperature monitoring and control in the HBC combustion chamber
(T >850 C) and maximising of the residence time (preferably >2s)
- maintaining the particulate matter concentration in the recuperator at a level <20 mg/m, this
is possible for HBC when using wet dedusting
- providing quick quenching of the dust laden off-gases, through the de-novo synthesis
temperature range of 250 450 C
- preventing or minimising the build-up of dust along the cooling trajectory of the flue-gas,
especially in the heat-exchanger, e.g. using vertical exchanger tubes, efficient internal
cleaning, high temperature de-dusting
- melting clean scrap. This technique is described in Section 4.1.4
- using oxygen injection to ensure complete combustion. This technique is described in
Section 4.2.1.6.

Although dioxins are destroyed at high temperature (i.e. above 850 C) in the presence of
oxygen, the process of de-novo synthesis is still possible as the gases are cooled through the
Chapter 4
214 Smitheries and Foundries Industry
reformation window (250 450 C). This window can be present in heat-exchangers or
abatement systems and in cooler parts of the furnace, e.g. the feed area. Care must be taken in
the design of cooling systems to minimise the residence time in the window and to avoid dust
build-up, in order to prevent de-novo synthesis. An alternative is to dedust the off-gas by quick
quenching using a wet system. Sufficient oxygen also needs to be present in the hot gases and
for this oxygen injection can be used to ensure complete combustion. Nevertheless, excess
oxygen should be prevented since this may support de-novo synthesis.

Sulphur has an inhibiting effect on the formation of dioxins, through depletion of molecular
chlorine. The use of coal with a higher sulphur content in large combustion plants has been
shown to provide lower PCDD/F concentrations. The reducing inhibiting effect is related to the
S/Cl ratio, with a critical ratio of 0.64. A further increase does not result in less dioxins and
furans. This effect has not been demonstrated in foundries, but may be studied. [231, UNEP,
2003]

The great spreading and big variability in the dioxin emission levels (even for the same
installation) show that primary measures alone may not allow a stable and low dioxin emission
value. Therefore, besides primary measures, the following abatement measures may be
considered:
- injection of additive powders into the gas stream, such as activated carbon, open-hearth
furnace coke or zeolite, so that dioxins are absorbed onto the surface. High efficiency dust
filtration is then used to remove the dust and dioxins. The additive is injected into the off-
gas stream before filtration. The adsorption process mainly takes place while the absorbents
adhere to the filter bag. The filter dust may be recirculated back to the flue-gas to attain a
higher efficiency. When using a carbon-based additive, special measures should be taken to
prevent fire and explosion risk. The collected dusts may have high dioxin concentrations
and will need to be disposed of or treated carefully
- catalytic oxidation systems are available for the destruction of dioxins. Fabric filters that
incorporate a catalytic layer are used for the destruction of dioxins. In other sectors (e.g.
steel, municipal waste incineration) this technique has been implemented successfully and
implementation in the foundry industry is considered feasible. However, in order to prevent
deactivation of the catalyst layer, a prior removal of coarse dust particles may be needed.

These are techniques to be considered depending on the application. They can all be
incorporated into existing processes. The choice of the most effective and economically viable
technique will depend on the specific site, safety aspects, and operational stability, as well as on
economic factors.

Although the absence of one of the five dioxin building conditions mentioned above hinders
dioxin synthesis, it is currently not possible to precisely foresee dioxin emissions by considering
known operational parameters. The building of a new furnace therefore needs the careful
consideration of primary measures as well as the option to add secondary measures in case of
unexpectedly high values.

Achieved environmental benefits
Reduction of the emission of dioxins and furans to air.

Cross-media effects
The adsorption of dioxins and furans on activated carbon generates a PCDD/F-loaded dust
stream. In order to prevent the explosion risk in the bag filter, mixing of the activated carbon
with lime may be needed. This will increase the total amount of residue for disposal and will
limit the possibility to re-use the filter dust.

Operational data
In-process measurements of dioxins in a hot blast cupola with dry dedusting have shown that
high PCDD/F-levels (5 ngTEQ/Nm) occur in the heat-exchanger. Other parts of the flue-gas
system show much lower values. Reduction measures should therefore aim at minimising the
Chapter 4
Smitheries and Foundries Industry 215
contact between dust and flue-gas in this zone, by minimsing dust or reducing the dust residence
time.

A PCDD/F-emission level of 0.5 ng TEQ/Nm
3
can be achieved by using primary measures; and
better than 0.5 ng TEQ/Nm can be achieved by using one or more of these techniques. A
German survey concluded that without secondary measures the level of 0.1 ng TEQ/Nm is
passed only in a limited number of installations and then only by a limited extent. The
operational data as given in 3.8.2, however show that the level should be evaluated on a plant-
by-plant basis.

Secondary measures in other sectors have been proven to allow a reduction to below
0.1 ng TEQ/Nm.

For a flue-gas rate of 8000 m/h at a blast rate of 3000m/h, quick quenching of the cupola off-
gas from 800 C to 150 C requires a water consumption of 4 m/h.

Applicability
These techniques are applied in other industrial sectors, such as steel and non-ferrous metal
production and waste incineration. Judging on a technical basis, they may be transposed to
foundry furnace types that show a risk of dioxin formation: cupola, rotary and electric arc
furnaces melting iron and steel (Section 3.8.2). For new and existing installations primary
dioxin reduction measures, such as efficient combustion, furnace design modifications and scrap
quality control have to be taken into consideration on a case-by-case basis, before turning to
secondary measures.

The application of additive injection involves the installation of an additive silo, an injection
system and, in the case of carbon injection control measures to prevent additive build-up. In
order to prevent fire risks, activated carbon may be mixed with lime and may be injected, after
the first filtration but before a dedicated secondary filter unit.

The application of catalytic filtration involves the least technical modification for existing
plants, since only the filter bags need to be replaced by the catalytic type.

Economics
Primary measures do not involve additional investment costs. Operational costs are restricted to
the use of oxygen or a higher price for cleaner scrap.

A cost estimation has been made for a hot blast cupola, as given in Table 4.34. Operational
costs depend on the coke injection dose used and are given in Table 4.35. Costs for staff,
maintenance, spare parts as well as possibly required early bag exchange or the installation of
another type of bag quality have not been taken into consideration.

Parameter Units Value
Volume flow Nm/h 50000
Operating hours h/yr 6250
Additive powder Open-heart furnace coke
Additive price EUR/t 400
Landfill price EUR/t 300
Energy costs EUR/kWh 0.09
Particle content 2 g/Nm - 100 kg/h
Investment cost:
incl. silo, safety equipment, recirculation, integration
- total cost
- yearly cost

EUR
EUR/yr

350000
52500
Energy consumption
Cost of energy consumption
kW
EUR/yr
10
6000
Table 4.34: Investment cost estimation for carbon injection installation, added to HBC
[230, CAEF, 2003]
Chapter 4
216 Smitheries and Foundries Industry
Parameter Units Values
Coke content g/Nm humid 0.2 0.3 0.4 0.5
Coke consumption t/yr 63 94 126 156
Purchase cost EUR/yr 25200 37600 50400 62400
Landfill cost EUR/yr 18900 28200 37800 46800
Abated investment cost EUR/yr 52500 52500 52500 52500
Energy cost EUR/yr 6000 6000 6000 6000
Total cost EUR/yr 102600 124300 146700 167700
Table 4.35: Total costs for carbon injection installation, added to HBC
[230, CAEF, 2003]

Catalytic bag filter systems are used in the waste incinerator sector. Cost data from this
application are transposed on basis of the flue-gas volume to be treated, but without taking into
account operational parameters such as air-to-cloth ratio. For a depreciation time of 5 years, the
investment and operational costs can be estimated at EUR 0.4 0.5/tonne molten metal for
HBC and EUR 0.9 1.3/tonne molten metal for CBC. This price estimation was nor confirmed
nor discarded by a catalytic bag filter supplier.

Driving force for implementation
Regulation limiting dioxin and furan emissions from the melting of metals.

Example plants
Reduction by installing oxygen injection: 3 example CBC plants in the Netherlands
Activated carbon injection: 1 example plant in Germany

Reference literature
[155, European IPPC Bureau, 2001], [161, UK Environment Agency, 2002], [202, TWG, 2002],
[224, Helber, et al., 2000], [230, CAEF, 2003]

4.5.1.5 Odour abatement

Odour emissions are mainly associated with processes involving sand binders. The actual
odorous products of pyrolysis may vary according to the type of system used, but generally the
phenolic breakdown products, i.e. cresols and xylenols, are the most common source of odour
complaint, due to their very low odour detection thresholds. The dispersion of odours during
casting, cooling and shake-out involves mixing with large volumes of air, which makes
collection and treatment difficult. The use of inorganic binders, such as sodium silicate, can
substantially reduce emissions. No totally effective proven method of eliminating foundry
generated smells is known to be currently available. A general approach is to ensure good
ventilation and a rate of air change that ensures that emissions are quickly and efficiently
dispersed to the atmosphere. Process-integrated measures include the substitution of binders or
binder solvents (see Section 4.3.3.7). End-of-pipe techniques aim to abate pyrolysis products
(see Section 4.5.8.5) and amines (see Section 4.5.8.4). These include adsorption techniques,
post combustion, wet scrubbers and biofilters (see Section 4.5.8.6).

4.5.2 Cupola furnace

4.5.2.1 Gas collection, cleaning and cooling

Description
The design of the collection and gas cleaning system is based on the conditions occuring during
blow-down, as these are often the most severe conditions experienced during the system
operation. At the end of a melting campaign, the furnace is no longer filled with charge
materials. Gas temperatures will increase gradually since they are no longer cooled by a cold
charge in the stack. In contact with oxygen, CO will burn automatically. Temperatures can
Chapter 4
Smitheries and Foundries Industry 217
therefore reach up to 1200 C, or even higher. The off-gas collection and treatment system has
to be able to cope with these severe conditions.

Collection
Two systems are in use for top gas collection:
- Above charge-hole offtake: The exhaust gases are withdrawn at the end of the cupola stack
by means of ductwork and a fan placed downstream. The opening above the charge-hole
allows an important inflow of air, necessary to prevent cupola gases being emitted from the
area. This volume of extra air may be many times the cupola gas flow. This increases the
size and cost of the collection and cleaning system. Reducing charge-hole sizes may have
some merit but this option may be limited because of the explosion hazard present when too
little oxygen is mixed with the CO containing cupola gases (pulsating combustion)
- Below charge-hole offtake: The cupola top gases are collected through an annular ring
below the charge-hole. No air inflow is required since the gases cannot be emitted out of the
hole, provided the control system is sensitive enough to function properly during the
variation of the blowing rate of the cupola. Too little offtake rate can result in the emission
of uncleaned gases through the cupola stack, too great an offtake flow can result in air
ingress, leading to combustion and overheating of the gases (i.e. an explosion hazard).

Cooling
Following collection, the gases may need cooling depending on the dust abatement system used.
In the hot blast operation, the heat recovered from cooling may be used for preheating the blast
air.

Several options are possible for cooling the collected gases, including:
- Using tube coolers: Running the collected gases through long ducts, decreases the
temperature by natural convection and radiation. This system is simple but takes up a lot of
space and does not offer controlled cooling (therefore there is a risk of condensation)
- Using a forced air/gas heat-exchanger: Cold ambient air is forced through an arrangement
of tubes or plates to cool down the gases. Dust collection and the subsequent need for
cleaning the heat-exchanging surfaces may lead to a complex and expensive design of the
system. One advantage of this system is the possible use of the heated air for external
heating purposes. Recuperative hot blast cupolas are equipped with a post combustion unit
and a heat-exchanger (recuperator) to heat the blast air
- Using an oil/gas heat-exchanger: This is similar to the above system but more expensive
because of the need for a secondary cooling system. The heat-exchanger is generally cooled
with a circulation of mineral oil. Cooling with a water/gas heat-exchanger is not (or only
very rarely) practised
- Saturation with water: Here the gases are cooled by the evaporation of the water sprayed
into the gas stream. Wet scrubbers perform better if the gases are cooled in a saturation
chamber prior to cleaning. When using fabric filters only, partial saturation is possible to
prevent clogging of the fabric due to the condensation of water. A good control system is
necessary to guarantee correct functioning of the system. Quenching the gases has the
advantage that rapid cooling reduces the risk of dioxin formation.

Dedusting
Dust capture equipment of various types can be used to remove particulate matter from the
waste gases. Generally wet scrubbers have low capital costs and maintenance, but require a high
energy input to achieve acceptable collection efficiencies. The removal of the sludge is difficult
and the scrubber water has to be treated prior to discharge. Dry collection systems have more
expensive capital costs and need better control of the inlet gas conditions (temperature,
condensation of water or organic vapours, CO:O
2
ratio, sparks) but usually use less energy than
that needed for wet scrubbing. Furthermore, dry cupola dust can be recycled into the cupola (see
4.9.4.2).

Chapter 4
218 Smitheries and Foundries Industry
Both venturi scrubbers and disintegrators are used with cupola systems. Descriptions of these
systems are given in Section 4.5.1.3. A separator to remove small particles entrained in water
droplets, is located after the wet scrubber.

With regard to dry systems, the following remarks can be made:
- Multicyclones: These are often used in conjunction with a fabric filter, acting as coarse dust
arrestors. They help to prevent incandescent coke particles reaching the filter cloth.
Provided refractory lining and high grade steel are used in the design of the cyclone, they
can operate at high temperatures. Note, the collection efficiency from cyclones alone is not
sufficient to meet todays regulations, hence they are usually used in combination with other
dedustung systems
- Bag filters: These are ideal when the gases are burned prior to the dedusting. This avoids
problems of the deposition of carbonaceous material or fire hazards. Bag filters can be
designed to provide good efficiency for collecting metallurgical fume particles such as ZnO
- Electrostatic precipitators: These systems are less common in the European foundry
industry. This system is best suited to more or less constant working conditions, such as in
long campaign cupolas, because of its sensitivity to variations in gas temperature, flow and
humidity. There is an explosion hazard when dedusting unburned gases mixed with air, due
to the relatively large volume of the precipitator. The precipitator therefore needs to be
flushed before applying electrical power.

A schematic representation of a cold blast cupola with heat recovery and a bag filter is given in
Figure 4.12.

Figure 4.12: Flow sheet of cold blast cupola with heat recovery, cooling and a bag filter
[29, Batz, 1986]

Post combustion
Post combustion of the waste gases is used to optimise heat recovery (chemically bonded as
CO) and to provide cleaner exhaust gases. In burning CO, any residual carbonaceous material is
simultaneously oxidised to CO
2
and H
2
O. The generated heat can be recovered using a heat-
exchanger and then transported to an internal user (e.g. blast air preheating).

Typical arrangements are:
- a post combustion chamber placed before (bag filter) or after (wet-filter) the dedusting unit
(for below charge-hole offtake), this is discussed in Section 4.5.2.2
- (natural gas) burners or controlled air injection in the cupola shaft (for above charge-hole
offtake), this is discussed in Section 4.5.2.3.

Chapter 4
Smitheries and Foundries Industry 219
The design of the system needs to ensure that the waste gases remain at a temperature above
800 C and with a suitable residence time, i.e. 2 seconds, to guarantee the complete oxidation of
the waste gases. The different systems are fully described and discussed below.

Achieved environmental benefits
Exhaust capture and cleaning is a necessary measure to reduce the emission products from coke
combustion such as NO
x
, SO
2
, HF, PCDD/F, and dust.

The post combustion of CO allows (additional) heat recovery from the cupola off-gas.
Additionally, it allows the melting of scrap contaminated with oil and grease without additional
environmental effects and thus stimulates the recycling of metals. Post combustion without heat
recovery shows a negative environmental balance (See cross-media effects).

Cross-media effects
Dedusting systems produce a residual dust for disposal. 5 13 kg of dust are produced per
tonne of liquid iron. The dust may be re-circulated into the cupola. This is discussed in Section
4.9.4.2. The characteristics of the dust have already been given in Section 3.2.2.2.

The post combustion of CO without heat recovery has negative impacts on the environment,
since powerful burners with a power of tens of kW need to be installed. The burners generate
emissions of combustion gases and consume additional oxygen.

Operational data
Fabric filters have an efficiency of over 99 %. Daily average dust emission values stay well
below 10 mg/Nm. The sum of Pb, Zn, Cr, Cu, Mn,V, Sn, Cr, Ni, As, and Cd reaches about
20 % of the total dust content. The data represented in Table 4.36 were gathered within the
framework of investigations of the Federal Environmental Agency in Germany of operational
plants. Data for 3 example plants are given.

Units Plant D Plant E Plant F
Initial After rebuild
Melting capacity
- design value
- actual value

t/h
t/h

7.5 8
7.0

12.0
11.0

6 7
5.5

4 5
3.7
Exhaust gas stream
- design value
- actual value

Nm/h
Nm/h

25000
19800

30000
22300

20000
17400

n.d
14300
Year of construction
of the filter
1981 1995 1988 1985
Last filter cloth
change
n.d 1995 1988 1993
Date of the
measurements
07/1981 11/1997 03/1993 03/1993
Emissions
- dust
- raw gas
- clean gas
average
1
min.
max.
- gaseous
2
SO
2
NO
x
C
total

CO
CO
2
mg/Nm
3
%
1623 - 2674

21.5
18.0
25.4

288
43
n.d
700
7
<1
<1
<1

n.d
n.d
n.d
n.d
n.d

n.d

3
1
5
174
24
22
11890
4.9

n.d

3
2
4
227
31
7
18980
3.9
Chapter 4
220 Smitheries and Foundries Industry
Units Plant D Plant E Plant F
Initial After rebuild
Heavy metal
emissions

- Cd
- Cr

- Pb
- Zn
- Ni
- As
- Mo
mg/Nm
3
Raw Clean

0.0184 0.0019
0.7287 0.0384

29.895 0.2952
16.464 0.2862
0.2024 0.0077
0.7665 0.0149
0.2672 0.0420
Clean gas

0.00313 (Cu,
Mn, Cr, V)

0.00057
n.d

n.d
PCDD/F-emissions
- clean gas
3
- emission factor

ngTEQ/Nm
3
ag TEQ/t Fe
n.d
0.512
1.620

0.085
0.330
Filter dust
- collected amount
- PCDD/F-Gehalt

kg/t Fe
ag TEQ/kg

6.5 disposal

8.2 re-use

4.850

0.960
Filter material Polyester cloth
with PA coating
Synthetic fibre Polyester Needle
cloth
Filter cleaning Pulse jet Medium
pressure
counter flow
Pneumatical
with pressure
drop control

Capital cost EUR 000 385 (1981) 370 (1995/96) n.d 350
Operating cost EUR/tonne
good casting
9.8 (1982) = 3 %
of the casting
costs
9.04 (1998) =
2.8 % of the
casting costs
n.d n.d
1
Dust average is calculated on basis of 3 - 5 half-hour measurement data
2
The concentrations of sulphur dioxide, nitrogen oxide, total carbon, carbon monoxide and carbon dioxide are
calculated averages from continuous measurements over several hours during sampling for dust or PCDD/F
3
Sampling time of 6 hours
n.d: no data
Table 4.36: Operational data of cold blast cupola furnaces with a bag filter for dust abatement
data taken from [43, Batz, 1996] and [202, TWG, 2002], [225, TWG, 2003]

Data for hot blast cupolas are given in Section 4.5.2.2.

Applicability
This technique applies to all new and existing cupola furnaces.

Economics
Table 4.36 shows emission values and economic data. The most important operational cost
factors are the electrical energy requirements for overcoming the pressure drop and for the filter
material.

For a melting installation incorporating two cold blast cupolas with a melting rate of 4.5 t/h, a
sales quotation was made with the following details:
- abatement equipment to treat up to 12400 m
3
/h blast, with combustion chamber at a
temperature of 820 C accepting up to:
VOC 1 g/Nm
3
CO 59100 g/Nm
3
- two burners to be used with methane Power/burner 390 kW

Installation and start-up price: EUR 350000.
Foreseen running cost, considering a 4500 tonne casting production, five year payment period
plus 10 % for energy and maintenance per year: EUR 23.3/t.

Chapter 4
Smitheries and Foundries Industry 221
Driving force for implementation
Regulation of air emissions from foundries.

Example plants
The mentioned techniques are generally applied in European foundries with cupola furnaces.

Reference literature
[29, Batz, 1986], [32, CAEF, 1997], [110, Vito, 2001], [202, TWG, 2002], [225, TWG, 2003]

4.5.2.2 Post combustion in an HBC combustion chamber

Description
A post combustion chamber with a burner is installed after the cupola. Normally a separate post
combustion unit has to be preheated with a natural gas burner. Once the cupola is up and
running either a smaller burner sustains the combustion of the waste gases or the gases self-
ignite.

The type and position of the chamber can vary according to the process composition. Both
horizontal and vertical combustion chamber types exist.

- Hot blast cupola with a recuperator and wet scrubber (Figure 4.13): In this arrangement,
the gases are dedusted prior to combustion. This reduces dust build-up in the recuperator,
which improves the rate of heat transfer. One disadvantage is the higher energy
consumption in the post combustion unit, caused because the gases are cooled down in the
wet scrubber. Early cooling of the offtake gases is continuously carried out to reduce the
size of the dedusting unit.

Figure 4.13: Hot blast cupola with a recuperator and wet scrubber
[32, CAEF, 1997]

Chapter 4
222 Smitheries and Foundries Industry
- Hot blast cupola with a recuperator and bag filter (Figure 4.14): The hot, dust laden, top
gases are fed directly into the post combustion unit. Close process control is necessary to
prevent sintering of the dust particles to the walls of the recuperator, which need to be
cleaned regularly. The gases need further cooling before entering the bag filter since they
leave the recuperator at temperatures of 500 to 600 C.

Figure 4.14: Hot blast cupola with a recuperator and bag filter
[32, CAEF, 1997]

Achieved environmental benefits
Post combustion limits the emissions of CO and eliminates the majority of organic compounds.
If not combusted, these would be captured in the dust or emitted through the chimney.
Furthermore, post combustion reduces the risk of fire in the filter.

Cross-media effects
Dry dedusting techniques produce dust for disposal (4 12 kg/t liquid iron). The dust may be
re-circulated into the cupola. This is discussed in Section 4.9.4.2. Wet systems produce a sludge
fraction.

Operational data
Operational data for hot blast cupolas are given in Table 4.37. Data for two example plants are
given:

1. Foundry G is a 3 shifts/days and 5 days/week operated foundry with a melting capacity
of 50 t/h. The off-gas is collected below the charge-hole and combusted in a
recuperator. The waste gas is then separated: one part goes to hot blast production (T =
600 C), another part goes to a steam boiler. The steam is fed into a turbine, which
powers a generator or compressor. The residual heat is used for preheating the
combustion air of the recuperator. Waste gas is then cleaned in a bag filter. A concept
drawing and further discussion of the installation is given in Section 4.7.3. Dust is
recycled into the cupola, after mixing with petcokes. This is discussed in Section
4.9.4.2.

Chapter 4
Smitheries and Foundries Industry 223
2. Foundry H is a 3 shifts/day and 5 days/week operated foundry with a melting capacity
of 70 t/h. The off-gas is collected below the charge-hole and washed in a disintegrator,
before post combustion in a recuperator. Heat is used for blast air preheating and goes
to a further heat recovery, before leaving through the stack at a temperature of 220 C.
Further discussion of the heat recovery installation is given in Section 4.7.3. Waste
water is re-circulated after settling. The circulating water volume is 440 m/h. The
sludge from the settling tank is dried in a filter press to a 50 % DS content, before
disposal. Some 80 m/day of waste water are disposed to the municipal waste water
treatment.

Units Plant G Plant H
Melting capacity tonne/h 50 50
Waste gas flow Nm/h 75000 55000
Hot blast temperature
o
C 600 570
Exhaust gas temperature
o
C 127 220
Year of construction of the
flue-gas treatment
1989 1983
Date of the measurements 10/1990 Control measurement
09/1993
Emissions
- Dust
Raw gas
Clean gas: average
1
maximum
- Gaseous
2
SO
2
NO
x
C
total

CO
O
2
mg/Nm
3
%
1300 4300
1.1
1.8

33
44
<5
32
12.2

8000 20000
6.1
7.3

15.6
52.5
28.6 (FID)
<100
6.4
Heavy metal emission
- Cr
- Cr
total

- Pb
- Zn
- Ni
mg/Nm
3
Clean gas
3
<0.001
<0.001
<0.001
<0.011
n.d
Clean gas
<0.0022
n.d
0.11
0.36
<0.004
PCDD/F-emissions
1
- Clean gas
4
- Emission factor

ng TEQ/Nm
3
ag TEQ/tonne
Fe

0.048
0.089

0.003
0.004
Filter dust/sludge
- solitary quantity
- PCDD/F-content

kg/tonne Fe
ag TEQ/kg
Dust recycling
4.5
0.176
Filter cakes to disposal
5.5
1.4
Waste heat utilisation Hot blast air, steam for
electricity production up
to 3 MW electric
Hot blast air, thermal oil for
the conversion of the waste
heat for heating and drying
purposes up to 21 MW
Investment cost DEM 000 26400 22700 in 1980/81
Operational cost DEM/tonne
good casting
See data in the text for
foundry G
See data in the text for
foundry H
1
Dust average is calculated on basis of 5 - 6 half-hour measurement data and 2x2 hour values for PCDD/F
2
The concentrations of sulphur dioxide, nitrogen oxide, total carbon, carbon monoxide and carbon dioxide are
calculated averages from continuous measurements over several hours during sampling for dust or PCDD/F
3
At the time of measurement, the dust recycling was not operational
4
Sampling time of 2 hours
n.d: no data
note: Raw gas sampling was performed before the bag filter in G and before the disintegrator for H; clean gas
sampling was performed after the bag filter for G and after the recuperator for H
Table 4.37: Operational data of hot blast cupola furnaces using a bag filter and a disintegrator for
dust capture
[17, Strau, 1983], [27, Kran, et al., 1995], [202, TWG, 2002]
Chapter 4
224 Smitheries and Foundries Industry
Burning the fumes in the post combustion chamber does not consume much energy, providing
there is sufficient carbon monoxide in the fumes, which is generally the case. But the whole
system for treating the fumes (combustion chamber + heat-exchanger + filter or wet scrubber +
fans) also needs electrical energy and regular maintenance. Table 4.38 gives some examples of
energy consumption for hot blast cupolas.

Hourly rate
of the cupola
Type of dedusting Gas consumption for
the combustion
chamber
(kWh/t charged)
Electricity consumption
for the fume treatment
system
(kWh/t charged)
12 Filter 59 46
12 Filter 124 72
26 Filter 42 n.d
17 Electro filter (wet) 16 38
Table 4.38: Energy consumption of hot blast cupolas

Applicability
During the design stage, considerable care has to be taken to minimise the total flow of the
gases to be treated. The combustion air intake has to be kept to a strict minimum. This is the
reason why a separate combustion unit is always installed in conjunction with a below charge-
hole offtake.
For economic reasons (see below) the post combustion chamber is only used on hot blast
cupolas.

Economics
The high energy consumption of the post combustion is only economically reasonable if the
released heat of the waste gases can be re-used, as it is in the recuperative hot blast cupola.
However, switching from a cold blast to a hot blast cupola for the sole reason of gas combustion
might encounter economical limitations in certain situations. Hot blast cupolas, preferably in a
long campaign configuration, involve higher investment costs and are only used with production
rates of 10 tonne Fe/h or more. In smaller foundries this production method might not be the
right choice.

Table 4.37 gives economic data for the example plants. For foundry G, the investment costs for
a hot blast cupola with bag filter and extensive heat recovery are given. The operational costs
for 1994 (after refurbishment of the melting shop) were 25 % lower than those for 1985, i.e.
with the old melting furnace. For foundry H, investment costs for 1980 - 1981 are given. After
refurbishment, the operational costs went down by 2 %/tonne of liquid iron.

Driving force for implementation
Emission limit values and continuous monitoring of CO and emission limit values for organic
compounds.

Example plants
The technique finds general application in HBCs around Europe.

Reference literature
[17, Strau, 1983], [27, Kran, et al., 1995], [32, CAEF, 1997], [157, Godinot, et al., 1999], [202,
TWG, 2002]

4.5.2.3 Post combustion in the cupola shaft

Description
The gases are combusted by an injection of air into the upper part of the charge or at a position
above the charge top level. The airflow is adjusted so that the off-gases ignite spontaneously,
due to their CO content and temperature. The injection nozzles can be placed on one or two
levels. The partitioning of the airflow over the various levels, the choice of diameter and the
Chapter 4
Smitheries and Foundries Industry 225
position of the nozzles is based on experience. The goal of the optimisation is to burn the CO
without ignition of the coke. The draught will also suck in air from the charge door. This air
excess allows a more complete burn-out of the CO.

A supporting burner may be provided to maintain the flame. When using very low coke charges
(i.e. <6 8 %) the precautionary measure is reasonable.

Post combustion of the off-gas must be combined with a gas cooling, if a bag filter is used. For
hot blast cupolas, the cooling is combined with a blast air preheating. In cold blast operation, a
rapid cooling may be applied using water injection in the furnace shaft. Alternatively, an (off-
gas air) heat-exchanger may be used. This is illustrated in Figure 4.15.

Figure 4.15: Principle of post combustion and off-gas cooling in cold blast cupolas
[157, Godinot, et al., 1999]

Installing post combustion on cold blast cupolas can be combined with a full retrofit to hot blast
operation. In general, this choice is based on operational considerations. The characteristics of
hot blast and long campaign furnaces are discussed in Section 2.4.1.

Achieved environmental benefits
Post combustion limits the emissions of CO and eliminates the majority of organic compounds.
If not combusted, these would be captured in the dust or emitted through the chimney.
Furthermore, post combustion reduces the risk of fire in the filter.

This technique is known to avoid explosion risks under certain circumstances. The positive
environmental effect is limited to those cases when the off-gas is burning autothermally most of
the time. Otherwise, the energy consumption will counterbalance the CO reduction.

Cross-media effects
If the cupola off-gas does not ignite spontaneously, the installation of ignition or support
burners is necessary. These incur a significant power use and increase the overall flue-gas
volume.

Operational data
Two configurations of (in-shaft post combustion and) cooling on cold blast cupolas were
studied on an industrial scale and compared by CTIF; cupola I was equipped with a water
injection in the cupola shaft, a cyclone and a bag filter, cupola J was equipped with a gas-air
heat-exchanger, a cyclone and a bag filter. Operational data are given in Table 4.39. Results of
the measurement campaigns and reference to the applicable legislation are given in Table 4.40.
Chapter 4
226 Smitheries and Foundries Industry
Parameter Units Foundry I Foundry J
Melting capacity tonne/h 10.3 9.2
Primary + secondary airflow Nm/h 7389 6484
Post combustion airflow Nm/h 2372 1549
Water flow l/h 2678
Flue-gas flow (chimney) Nm/h 26780 39179
Residence time combustion to bag filter s 10.5 12.3
Residence time cooling stage s <1.7 6.01
Overall CO combustion efficiency % 66 96.5
Table 4.39: Operational data of two example configurations using post combustion with water
cooling (I), and air cooling (J)
[157, Godinot, et al., 1999]

Foundry I Foundry J Arrt du 02/02/98
Compound analysis flux analysis flux limit value limit for
consideration*
Dust 1.7 2.8 mg/Nm
3 5 g/t
0.04 0.07 kg/h 1 mg/Nm
34 g/h
100 mg/Nm
200 g/t
<1 kg/h
melt = 8 tonne/h
CO 450 Nm/h 560 kg/h 35 Nm/h 44 kg/h 50 kg/h
HCl 22 mg/Nm 0.4 kg/h 9.2 mg/Nm 0.35 kg/h 50 mg/Nm 1 kg/h
HF 13 mg/Nm 0.24 kg/h 4.8 mg/Nm 0.18 kg/h 5 mg/Nm 0.5 kg/h
HCN <0.007 mg/Nm 0.13 g/h <0.01 mg/Nm 0.38 kg/h 5 mg/Nm 50 g/h
NH
3
0.61 mg/Nm 12 g/h 0.10 mg/Nm 3.8 g/h 50 mg/Nm 100 g/h
NO
x
9 mg/Nm 0.17 kg/h 15 mg/Nm 0.57 kg/h 500 mg/Nm 25 kg/h
VOC non-
methane
13 mg/Nm 0.25 kg/h 6 mg/Nm 0.23 kg/h 110 mg/Nm 2 kg/h
*: If the flux is below the limit of consideration, the measured value may exceed the limit value
Table 4.40: Analytical results and applicable (French) legislation, for post combustion in a cold
blast cupola with water cooling (I), and the air cooling (J) of flue-gases
[157, Godinot, et al., 1999]

The water spray cooling achieves a CO-destruction efficiency of 66 % and allows compliance
with the applicable legislation. Flue-gas cooling over a heat-exchanger provides better
efficiency (98 %). For discussion on cooling performance, see also 4.5.2.1.

Applicability
To change the offtake system of an existing cupola from above to below the charge-hole system
is in most cases impossible. Post combustion of the waste gases in the furnace shaft therefore is
a more viable solution for cupolas with an above charge-hole offtake.

For economic reasons, the application of post combustion has mainly been related to hot blast
cupolas. However, recently, a post combustion system for cold blast furnaces, without the
complexity of a hot blast installation, has also been developed. This system is currently in
operation in France. In-shaft post combustion therefore applies to both hot blast and cold blast
cupola operation.

In the Czech Republic, the whole range of cold blast cupola installations was studied for
applicability of in-shaft post combustion. The combustion was not spontaneous in any of the
set-ups over the whole melting period. In each case, it was necessary to install ignition burners
of significant power. The cupola emission volume also increased.

Economics
The high energy consumption of the post combustion needs to be evaluated against its benefits.
In hot blast operation, the released heat of the waste gases can be re-used, which makes the
investment more viable.

The high energy consumption of the post combustion is only economically reasonable if the
released heat of the waste gases can be re-used, as it is in the recuperative hot blast cupola.
However, switching from a cold blast to a hot blast cupola for the sole reason of gas combustion
Chapter 4
Smitheries and Foundries Industry 227
is not possible. Hot blast cupolas, preferably in a long campaign configuration, involve higher
investment costs and are only used with production rates of 10 tonne Fe/h or more. In smaller
foundries, this production method might not be the right choice.

The investment cost for a twin cold blast cupola of 850 mm internal diameter, which produces
4.5 tonnes an hour, 10 hours a day, 5 days a week is in the order of EUR 300000.

Driving force for implementation
Emission limit values and continuous monitoring of CO and emission limit values for organic
compounds.

Example plants
The technique is reported in 2 example foundries in France.

Reference literature
hot blast: [150, ETSU, 1998], [202, TWG, 2002]
cold blast: [157, Godinot, et al., 1999], [202, TWG, 2002]

4.5.3 EAF

4.5.3.1 Off-gas collection

Description
Particulate matter from an EAF is very fine and difficult to capture. The capture of smoke and
dust is the most difficult problem to solve when installing a dust arrestment system on an
electric arc furnace. There are several methods that offer different advantages and
disadvantages. Figure 4.16 depicts the principle of roof mounted hoods, side-draught hoods and
fourth-hole evacuation.

Figure 4.16: Principle drawings of (a) a roof mounted hood, (b) a side-draught hood, and (c) direct
evacuation through a fourth-hole
[173, Huelsen, 1985]

- Roof mounted hoods: A hood is fixed on the furnace roof and collects fumes through a gap
between the furnace shell and the roof. In addition, fumes arising from the working door
and the tap-hole are also captured. Roof mounted hoods are heavy and subject to distortion
from heat. If fitted to an existing EAF, problems may arise due to increased structural load
of the roof lifting and rotating mechanism. Typical extraction rates for this system are 7500
to 15000 Nm
3
/tonne
- Side-draught hoods: A hood is mounted on the furnace roof and collects the fumes emitted
through the electrode ports. Vanes direct the airflow towards the hood. Additional hoods
above the working door and the tapping-hole are used to capture emissions at these
locations. Higher extraction rates are necessary here, compared to roof mounted hoods, as
high air draught velocities are needed to capture the fumes. Some leakage of pollutants to
the plant atmosphere can be expected, unless a tight seal is maintained. It has been reported
that side-draught hoods increase electrode consumption
Chapter 4
228 Smitheries and Foundries Industry
- Canopy hoods: A large exhaust hood is placed over the complete furnace, and above the
charging system. This creates a large gap between the furnace and the extraction system,
making it difficult to control the rising smoke and dust, even when using high extraction
rates. Cross-draughts can seriously distort the collection efficiency. These disadvantages
make the use of these collection systems unattractive
- Direct furnace or fourth-hole evacuation: The fumes are collected through a water-
cooled or refractory-lined duct connected to a fourth-hole in the furnace roof. Ambient air is
aspirated through a joint in the ductwork, thus providing oxygen to burn the CO gas in the
undiluted and hot offtake gas. This prevents explosions in the dust collecting system. The
dimensions of the combustion chamber are very critical in order to guarantee sufficient
combustion at all stages of the melting cycle. The burnt gas (900 C) is cooled downstream
by dilution with ambient air, water injection, heat-exchangers (water jackets) or by the use
of long ducts. This cooling process is needed to protect the dust filter equipment. The use of
a pressure control system in the furnace allows a relatively small extraction rate at the
furnace at 2000 to 4000 Nm
3
/tonne. Electrode consumption is generally higher due to the
effect of oxidation
- Partial furnace enclosure: Mobile or fixed enclosures are mounted around the furnace and
tapping area. The former are retracted in order to accommodate charging and tapping, while
the latter are equipped with a mobile roof and side-walls for the same purpose. Exhaust
gases are collected through a main exhaust duct at the top of the enclosure, assisted by fans
inside the duct to guide the fumes to the exhaust. With these types of system collection,
efficiencies of up to 98 % are possible
- Total furnace enclosure: This system consists of a large fixed evacuated room completely
surrounding the melting furnace and ladle pit. The enclosure is equipped with a movable
roof and/or with side panels to accommodate both charging and tapping operations. These
may be equipped with air curtains to reduce gas losses. Disadvantages of the technique are
the high temperatures and noise levels inside the enclosure. However, the advantages
include low outside noise levels and low energy consumption. The energy consumption is
estimated to be 30 to 50 % less than that required for a canopy hood.

Achieved environmental benefits
Flue-gas capture allows the controlled evacuation and treatment of the flue-gas stream, and
results in a minimisation of both fugitive and guided emissions.

Cross-media effects
Capturing the exhaust increases the energy consumption. As it allows the cleaning of the
exhaust gas, it will also generate dust for disposal or re-use.

Operational data
Table 4.41 provides typical exhaust flowrates and removal efficiencies for various capture
systems.

Exhaust flowrate
for given furnace capacity
Particulate removal
efficiency (%)
3.9 t/h 9.1 t/h 22.7 t/h Range Typical max.
Side-draught hood 12.9 19.8 50.00 90 100 99
Roof hood 7.7 11.9 30.00 95 100 99
Direct evacuation 3.2 5.0 12.5 90 100 99
Total enclosure 35 42
Table 4.41: Typical exhaust flowrates (in m/s) and particulate removal efficiency of EAF exhaust
capture systems
[173, Huelsen, 1985]

In a Polish foundry, the installation of a total enclosure over two EAFs with a capacity of
8.5 tonnes each, resulted in a reduction of the noise level from melting from 91 dBA to less than
85 dBA.

Chapter 4
Smitheries and Foundries Industry 229
Applicability
The above-mentioned techniques are applicable to all new and existing EAF furnaces, with the
following limitations:
- when installing a roof mounted hood on an existing furnace, problems may arise due to the
increased structural load from the roof lifting and rotating mechanism
- direct furnace evacuation using a fourth-hole is limited to big electric arc furnaces, as the
roof must be big enough to accommodate a fourth-hole without any structural weakening.
The technique is not applicable for iron melting, because the in-flow of fresh air generates
an excessive oxidation of carbon.

Economics
The construction and assembly of a doghouse over two EAFs with a capacity of 8.5 tonnes
each, involved an investment of EUR 275000 (1996 prices), excluding costs for further
refurbishment of the ducting and filtration system.

Driving force for implementation
Workers health-related prevention measures. It allows application of a gas cleaning technique
by guiding the exhaust gas flow.

Example plants
doghouse: Metalodlew Foundry, Krakow (PL)
off-gas collection: Metso Lokomo Steels and Sulzer Pumps Karhula Foundry (FIN)

Reference literature
[29, Batz, 1986], [32, CAEF, 1997], [173, Huelsen, 1985], [199, Metalodlew s.a., 2002]

4.5.3.2 Exhaust gas cleaning

Description
Gas cleaning systems for EAF flue-gas need to be highly efficient since the particle size is
small. Fabric filters are widely used to dedust the exhaust gases. Venturi scrubbers or
disintegrators are applied less due to their higher energy consumption, the necessity for water
treatment prior to discharge, and due to the need to dispose or recycle wet filter sludge. Fabric
filters are widely used in favour of electrostatic filters, due to the wide fluctuations in gas
temperature and the particulate concentration of the exhaust gases.

If a fabric filter is used, several precautions need to be taken in order to protect the filter
medium as discussed in Section 4.5.1.3. These include:
- using a high temperature cyclone or multicyclone
- using gas cooling equipment. Here the exhaust gases are cooled by dilution with ambient
air, water injection (quenching), heat-exchangers or by the use of long ducts. Water
injection, in particular, has an additional advantage, in that the quick cooling down of the
exhaust gases prevents the recombination of any partially burned organic substances,
thereby leading to a cleaner emission. The cooling system needs an effective control system
to prevent gases which are too hot from entering the fabric filters.

Achieved environmental benefits
Reduction of particulate emissions from EAF.

Cross-media effects
The application of wet dedusting techniques causes a higher energy consumption, the necessity
for water treatment prior to discharge, and the need to dispose or recycle wet filter sludge.

Operational data
Operational data are given in Section 3.2.3.4. Current bag filters have air to cloth ratios between
2:1 and 3:1. The air to cloth ratio is the ratio of volumetric airflow through the filter to the
Chapter 4
230 Smitheries and Foundries Industry
media area and is expressed in (m/s)/m, which yields the same result as the air speed through
the filter media.

Applicability
These techniques can be applied to all new and existing installations.

Economics
The costs for installing a new dust capture and filtration unit in an existing Polish EAF foundry
using two 8.5 t/h melting furnaces are summarised in Table 4.42. Dust emissions have been
reduced from 10 13 kg/h and 145 150 mg/Nm to 0.2 0.25 kg/h and 2.8 2.9 mg/Nm.
There has also been a major reduction of diffuse emissions from the melting shop.

Investment costs EUR*
Foundation and remodelling of old collector 115000
Construction and assembly of doghouse housings over 2 furnaces 275000
Purchase and assembly of filtration unit and electrical devices 560000
Research, implementation and supervision 80000
Operational costs EUR/year
Depreciation cost 100000
Electric power 88000
Repairs and service 1500
* Prices converted from zlotys 1999
Table 4.42: Cost data for EAF flue-gas capture and cleaning equipment
[199, Metalodlew s.a., 2002]

Driving force for implementation
Regulation limiting the dust emissions from the melting of metals.

Example plants
doghouse with bag filter unit: Metalodlew Foundry, Krakow (PL)
Metso Lokomo Steels and Sulzer Pumps Karhula Foundry (FIN)

Reference literature
[173, Huelsen, 1985], [32, CAEF, 1997], [29, Batz, 1986], [199, Metalodlew s.a., 2002]

4.5.4 Induction furnace

4.5.4.1 Off-gas collection

Description
The capture of smoke and dust is the most difficult problem to solve when installing an off-gas
collection system on a coreless induction furnace, since there is no exhaust shaft. Several
methods have been developed in the past decade, each with advantages and disadvantages.
- General ventilation of the workplace: A combination of wall mounted louvres and roof
mounted ventilators situated over the furnace platform are used to increase the natural
convection of smoke and fumes and to direct them outside. Even with baffles suspended
from the roof and using high extraction rates the efficiency is often poor and easily
disturbed by draughts.
- Canopy hood extraction: Since lower placed hoods will interfere with crane charging
systems, larger hoods have to be installed above the charger. This creates a large gap
between the furnace and the extraction system, making it difficult to control the rising
smoke and dust, even when using high extraction rates. Cross-draughts can seriously distort
the collection efficiency of the system. These disadvantages make the use of these
collection systems unattractive.
Chapter 4
Smitheries and Foundries Industry 231
- Swing aside hoods: These hoods are more efficient when used in conjunction with vibrating
feeders. Cut-outs in the hood can facilitate charging. During tapping, the hood is swung
over the ladle, allowing efficient fume extraction.
- Side-draught hoods: Placing the extraction hood beside the furnace offers the advantage of
good furnace accessibility and no interference with the charging systems. Due to the high
buoyancy of the exhaust gases, large extraction rates are achieved, therefore giving good
efficiency, especially when the hood is placed outside the furnace platform. In this case
extraction control is poor during tapping. Attachment to the furnace platform overcomes
this problem but may interfere with charging operations. The efficiency can be improved by
installing airjets at the opposite side of the hood to blow the dust and fumes into the hood.
Unfortunately, this facility does not work if there are any interferences in the airstream,
which is the case during charging.
- Lip extraction: A suction ring is placed on top of the furnace and arranged so that it moves
with the furnace during deslagging or tapping operations. This system does not interfere
with the charging operations. With the lid closed, lip extraction offers very good control,
since it is as near to the emission source as possible and involves the lowest extraction rates.
The fumes do not pass through the breathing zone of the furnace operators. However, the
extraction control decreases significantly when the furnace lid is opened, for instance during
charging.
The design of this extraction equipment has been subjected to many studies. Suppliers offer
solutions to overcome some of the disadvantages.
- Cover extraction: The gas is exhausted through the furnace cover. This method is very
effective. It is used by the majority of furnace producers. Exhaustion is managed according
to furnace regime: melting, charging, pouring.

Attention has to be paid to the material used for hoods and ducts since the gases may be at high
temperatures when the intake of the collection system is positioned close to the furnace. The
heating required by radiation or convection from the molten metal bath needs to be taken into
account in the design stage. Proper maintenance in combination with heat-sensors reduces the
risk of fire.

Again scrap cleanliness plays an important role. When the scrap contains organic matter,
collected gas temperatures may rise due to the combustion of the material, thus requiring the use
of heat resistant steel or even refractory linings. Oily deposits, formed by condensation of oil
vapour in the ductwork, accumulate dust and can present a fire hazard if not removed regularly.
When using clean scrap, a mild steel construction is adequate and does not need accessibility for
cleaning.

Achieved environmental benefits
Flue-gas capture allows controlled evacuation and treatment of the flue-gas stream, and results
in a minimisation of both fugitive and guided emissions.

Cross-media effects
Applying exhaust capture increases the energy consumption. In addition, as it allows the
cleaning of the exhaust gas, it will also generate dust for disposal or re-use.

Operational data
Through the use of specific capture systems, such as side-draught hoods, movable extraction
hoods and partial housing of the furnace, a capture efficiency of more than 95 % is possible.

Operational experience in a German cast iron foundry showed that the furnace lid is open on
average during 25 % of the working time of the furnace. During the opening periods, dust-
generating process steps such as the feeding of additives, deslagging and pouring are performed.
The lip extraction system installed on the furnace lid, does not allow extraction of the generated
fumes. Installation of a telescopic canopy hood allowed an efficient exhaust capture during the
opening of the furnace lid.

Chapter 4
232 Smitheries and Foundries Industry
Applicability
The installation of exhaust capture equipment is applicable to all new and existing induction
furnace installations, both in ferrous and non-ferrous foundries.

Driving force for implementation
Regulations concerning emissions to air.

Example plants
These techniques are commonly used on induction furnaces in Europe. A specific example is
Walter Hundhausen GmbH & co KG (D)

Reference literature
[29, Batz, 1986], [18, Rademacher, 1993], [32, CAEF, 1997]

4.5.4.2 Exhaust gas cleaning

Description
Gas cleaning systems for induction furnace flue-gas need to be highly efficient as the particle
sizes need to trap are small. Fabric filters are widely used to dedust the exhaust gases. Fabric
filters are used in favour of electrostatic filters, as they are more suited to the wide fluctuations
in gas temperature and particulate concentration of the exhaust gases.

When using fabric filters care has to be taken with respect to the presence of oil in the scrap, as
oily vapour could condense on the filter cloth, where it could block the pores and make the
removal of adhering dust impossible. There is also a possible fire hazard. The rapid increase in
the pressure loss of the system when pores get blocked quickly reduces the extraction rate of the
system. Therefore, in order to prevent workplace pollution the filter media have to be changed
or regenerated (cleaned) much quicker than when using clean scrap. The use of coated fabrics or
an injection of lime in the ductwork may be a solution to this problem. Furthermore, if the oil
vapour is likely to burn in the ductwork, the process must allow enough time for combustion to
be completed before the vapour enters the filter housing. The gas temperature may not exceed
the design temperature of the fabric, in which case cooling of the gases may therefore be
needed.

The wet scrubbers used generally need to be of the high energy type (venturi), because
carbonaceous and metallurgical smoke consists of very small particles. These require
considerable fan power to create sufficient turbulence in the scrubber to collect the particles.
Exhaust flowrates are therefore kept to a minimum by using exhaust capture systems with the
least entrainment of ambient air. Under normal conditions, corrosion is not a problem. If borings
containing cutting fluids are charged into the furnace, it should be noted that some of these
fluids might contain sulphur, which could lead to the formation of SO
2
. This could cause
problems in the equipment, as the absorption of SO
2
in the scrubber leads to an acidification of
the water and eventually to corrosion of the equipment if water treatment is not applied.

Some aluminium foundries melt material recovered from the electrolysis of aluminium. In this
instance, inorganic pollutants such as hydrogen fluoride may be formed. These can be cleaned
from the flue-gas using a chemisorption step, which can be added onto the dust abatement
system. In this treatment calcium hydrate or aluminium oxide can be used as the adsorbant.

Achieved environmental benefits
To reduce particulate and acidifying emissions from induction melting furnaces.

Cross-media effects
Exhaust gas cleaning increases the energy consumption. The dedusting of the exhaust gas,
generates dust for disposal or re-use. The application of wet dedusting techniques lead to a
Chapter 4
Smitheries and Foundries Industry 233
higher energy consumption, the necessity of water treatment prior to discharge, and the need to
dispose of or recycle wet filter sludge.

Operational data
Cleaning of the captured gases is generally performed using filters. Daily average dust emission
values stay well below 10 mg/Nm. Operational data are given in Section 3.2.4.1.

Table 4.43 gives the operational data for a German cast iron foundry in which a centralised flue-
gas capture system is installed, with dedusting carried out using a bag filter. The system collects
exhaust gas from the various parts of the foundry, including: the four induction furnaces (each
with lip extraction and canopy hood), the scrap storage and preheating, the metal treatment, the
sand regeneration, and the casting areas. Data are given for the raw melting of the furnace off-
gas, the raw combined gas flow and the cleaned gas flow.

Compound Melting off-gas Combined exhaust gas Cleaned gas*
Dust 89.3 237 <1
NO
x
1.6 8.3 7.9
CO 2.2 4.2 3.8
SO
2
3.5 3.9 3.7
Total C 21.8 34.7 34.9
PCDD/-F 0.036 x 10
-6
0.0027 x 10
-6

* NO
x
, CO, SO
2
and total C are not captured in the bag filter. The difference between raw and
cleaned gases are due to the slightly changed gas mix after the ID fan
Melting capacity 14 tonne/h, total flue-gas flow 240000 m/h
All data in mg/Nm
Table 4.43: Emission data for a cast iron foundry, using induction melting and a centralised
exhaust system with a bag filter
[18, Rademacher, 1993]

Applicability
This technique is applicable to all new and existing induction furnace installations, both in
ferrous and non-ferrous foundries.

Economics
The cost and consumption data for a bag filter installation on a melting unit of crucible
induction furnaces, melting 15 t/h, able to treat 120000 Nm
3
/h are given in Table 4.44.

Dust emission level
(mg/Nm)
Investment cost
(EUR)
Power consumption
(kW)
<5 350000 250
<20 200000 150
Table 4.44: Investment cost and power consumption for a bag filter unit on induction furnaces,
with varying final dust emission levels
data for Portugal 2003
[225, TWG, 2003]

Driving force for implementation
Regulation limiting the dust emissions from the melting of metals.

Example plants
Flue-gas cleaning is applied in most ferrous foundries using an induction furnace and in a
limited number of non-ferrous foundries.

Reference literature
[18, Rademacher, 1993], [32, CAEF, 1997], [29, Batz, 1986]
Chapter 4
234 Smitheries and Foundries Industry
4.5.5 Rotary furnace

4.5.5.1 Off-gas collection and waste gas cleaning

Description
In most cases, dedusting equipment is needed to meet the applicable regulations. Generally, bag
filters are installed for this purpose, but it is also technically possible to use wet dedusting
techniques.
In order to lower the temperature of the exhaust gases, they are diluted with ambient air. This is
achieved by the intake of air through a gap between the furnace exhaust and the elbow shaped
exhaust pipe. This gap is always present as it allows the furnace body to be able to rotate and
tilt. Often the exhaust pipe is retractable. In general, dilution for emission reduction is
unacceptable. If dilution is used (and is necessary) for cooling, the end-of-pipe technology
should have the right dimension for the larger gas flow. Sometimes the diluted waste gases are
then fed through an air-gas heat-exchanger for further cooling. This way the gas temperature is
reduced from the initial 1500 C to 200 C, or lower. At this temperature, the gases can be
introduced into a bag filter for dedusting.

The application of an afterburner allows the reduction of organic carbon emissions and
combustible particles. This technique may also be effective in reducing the risk of dioxin
formation upon cooling of the gases. The afterburner is installed after the furnace and before the
heat-exchanger. One of the following afterburner types may be used:
- thermal incinerator: combustion in an open flame
- catalytic incinerator: combustion at lower temperatures through the use of a catalyst,
resulting in a higher efficiency and lower NO
x
emissions
- recuperative incinerator: combustion with a heat recovery for preheating the combustion air,
resulting in a higher thermal efficiency and lower fuel consumption
- catalytic recuperative incinerator: a combination of the previous two types.

Achieved environmental benefits
Reduction of particulate emissions from rotary furnaces.

Cross-media effects
Exhaust gas cleaning increases the energy consumption. The dedusting of the exhaust gas
generates a dust for disposal or re-use.

Operational data
Operational emission data are given in Section 3.2.5.3.

An afterburner can be expected to achieve between 80 % and 98 % efficiency for burning the
combustible particulates emitted from the rotary furnace. Hot gases from the afterburner can be
ducted through a recuperator and can assist in preheating the combustion air to the main furnace
burner. Recuperators offer an energy saving of up to 15 %.

Applicability
This technique is applicable to all new and existing rotary furnace installations, both in ferrous
and non-ferrous foundries.

Driving force for implementation
Regulations limiting the dust emissions from the melting of metals.

Example plants
This technique is commonly applied in foundries using a rotary furnace.

Reference literature
[23, Brettschneider and Vennebusch, 1992], [32, CAEF, 1997], [163, UK Environment Agency,
2002], [202, TWG, 2002]
Chapter 4
Smitheries and Foundries Industry 235
4.5.6 Crucible, hearth type and radiant roof furnaces

4.5.6.1 Prevention of visible and fugitive emissions during melting and
metal treatment

Description
Under normal operating conditions, melting processes using pure metal melts do not emit
visible smoke. However, during charging of the furnace it is possible that a visible smoke may
be produced. This can be either due to the burning of contaminants in the charge, such as oil or
paint, or due to the burner flame being extinguished and unburned fuel being emitted, as may be
the case with solid or liquid fuels. In these circumstances, an afterburner may be installed.

Hoods may also be installed to capture visible smoke and fugitive emissions. The melting of
clean scrap prevents or minimises these types of emissions.

Achieved environmental benefits
A reduction of the emission to air of the products from incomplete combustion.

Cross-media effects
Smoke capture will involve a usage of electrical energy.

Applicability
This technique applies to new and existing crucible and hearth type furnaces.

For implementation, any constraints due to the current furnace design and operation need to be
considered. These include physical constraints, for example due to crane tracks, the existing
structural elements of the building, or due to the furnace itself. To be effective the extraction
need to be fitted in such a way that it moves with the furnace body when it is tilted to pour, as
tapping-out is one of the peak emission operations. This is not always easy to achieve with
existing furnace bodies. There can also be physical problems related to the location of
associated ducting. In some cases, major modifications to the furnace platform and surrounding
area may be necessary to enable the ducting to be installed.

Economics
A cost estimate for an extreme (worst case) scenario, i.e. needing to capture all the dusts from
charging and tapping, was made by CTI. Table 4.45 summarises this financial data for an
average foundry with the costs depreciated over a presumed 10 year lifespan of the extraction
and abatement equipment. This estimation however, can broadly vary, depending on the
foundry layout and the techniques used.

Costs
EUR
Costs
EUR/tonne of iron melted
Extraction
rate
m/minute
Production
rate
tonnes/hr Capital Civil
Power
kW
Running Financial Total
142 0.5 117573 15676 40 15.24 18.81 34.05
(1150 tonnes
per year)

Note:
- Civil costs are taken as 12 % of capital costs
- Running costs are based on maintenance and waste disposal costs and are equivalent to 10 % of the capital
costs and power at EUR 0.06/kWh
- Financial costs are based on the plant being amortised over ten years with interest charges of 10 %
Table 4.45: A cost estimate for installation of an abatement system for visible fumes during
charging and tapping
[161, UK Environment Agency, 2002]

In this example, under the pre-conditions used, the costs for abatement of these visible
emissions would be about EUR 34.05 per tonne of metal melted.
Chapter 4
236 Smitheries and Foundries Industry
Driving force for implementation
Regulations limiting the occurrence of visible emissions.

Example plants
This technique is commonly applied in foundries around Europe.

Reference literature
[161, UK Environment Agency, 2002]

4.5.7 Metal treatment

4.5.7.1 AOD Converter: exhaust gas capture and treatment

Description
Pollution control equipment need to be designed to be able to handle the peak effluent
condition, even though the peak may only persist for a relatively short period.

Fume control hoods for AOD vessels come in a great variety of shapes and sizes. The
alternative to the direct extraction hood is the roof canopy. Roof canopies are in use in several
AOD plants and efficiently capture both AOD process fumes and smoke from other operations
such as charging and tapping. These roof canopies may be used in conjunction with accelerator
stacks over the AOD mouth. The accelerator stack performs several useful functions, including
forming a more coherent plume out of the AOD, reducing the size of canopy required, and
shielding the surrounding equipment and personnel from the sometimes intense radiation of the
AOD flame. The main advantage of the roof canopy is the capture of both process and fugitive
effluents in the same hood.

Achieved environmental benefits
Flue-gas capture allows controlled evacuation and treatment of the flue-gas stream, and results
in a minimisation of both fugitive and guided emissions.

Cross-media effects
Exhaust gas cleaning increases the energy consumption. The dedusting of the exhaust gas
generates dust for disposal or re-use.

Applicability
This technique is applicable to all new and existing AOD installations.

Driving force for implementation
Regulations limiting the dust emissions from the melting of metals.

Example plants
Example plants are reported in Italy and Finland.

Reference literature
[202, TWG, 2002]

4.5.7.2 Nodularisation: off-gas collection and dedusting

Description
The type of exhaust collection applied depends on the technique used for nodularisation (see
2.4.12.4). Depending on the technique, significant amounts of MgO can be released as a white
smoke. Properties of the various nodularisation techniques are given in Table 3.20. In general
techniques with a higher take-up efficiency result in lower emissions. This is the case for the
in-mould process, where the nodularisation takes place during pouring.
Chapter 4
Smitheries and Foundries Industry 237
For in-crucible nodularisation, techniques using a lid or cover result in a significant reduction
of the emission.

If the nodularisation off-gas is not captured, this could result in the melting shop filling-up with
a white smoke, and MgO dust drop out. Exhausting without filtration can result in a visible
emission from the foundry. A large volume of visible particulate can be generated in a relatively
short period of time, but usually only lasts for a short duration (5 to 10 minutes for each batch
treated).

In-crucible nodularisation may be performed at a specific stand or location in the melting shop.
The crucible with the molten metal is brought to this point after pouring, but before taking it to
the casting furnace or station. This allows the installation of a fixed hood for exhaust capture.

The collection of the MgO fume is hampered by the fact that the gases are very hot and that the
intensive Mg reaction causes sparks. Due to the high upward velocity and temperature large
quantities of surrounding air need to be extracted as well. This requires a large sized, and high
cost, installation.

Dry filtration (using bag filters) of the exhaust gas, results in a MgO powder that may be re-
used in pigments or for refractory material production.

Achieved environmental benefits
Mg has no harmful effect on the environment and is an essential nutrient for animals and plants
in small concentrations. UK occupational exposure limits give a long-term exposure limit (8-
hour TWA) of 4 mg/m for MgO dust and respirable fume (expressed as Mg).

Cross-media effects
Capture of the magnesium oxide fume requires an increased energy use and therefore results in
increased emissions from energy production. External re-use of the MgO powder leads to a
reduced need for primary material.

Operational data
A typical addition rate for magnesium to the melt is around 0.1 % of the melt weight, measured
as magnesium (the actual addition of magnesium containing alloy, such as magnesium
ferrosilicon, could be up to 2 % of the melt weight according to the alloy used).

This addition provides about 0.05 % of the magnesium in the melt, most of the remainder
oxidises and escapes to atmosphere as MgO, where it will agglomerate fairly rapidly in the air.
Where there is no fume capture, the fumes can spread through the foundry and a proportion will
drop out in the foundry as dust. There is no accurate information available on this amount but a
reasonable figure may be 50 % of the fume released. Therefore, for each tonne of metal treated
there would be around 500 g of magnesium released to air at the ladle as MgO (i.e. 833 g of
MgO released per tonne of metal treated) and about 400 g of MgO released to the external
atmosphere.

Applicability
The technique of exhaust capture and filtration applies for foundries using in-crucible
nodularisation techniques. No information on applicability to the in-mould nor ductilator was
provided.

Economics
In applying this technique, the extraction required to successfully capture most of the MgO
emissions from the treatment ladle would be around 280 m/min. The combined capital and
installation costs per foundry would be around EUR 180000.

The table below summarises the financial data for an average foundry, with the costs
amortised over the presumed 10 year lifespan of the extraction and abatement equipment.
Chapter 4
238 Smitheries and Foundries Industry
Costs
EUR
Costs
EUR/tonne of iron treated
Extraction
rate
m/min
Production
rate
tonne/h Capital Civil
Power
kW
Running Financial Total
280 0.5 180280 23514 50 21.95 29.35 51.30
(1000 tonnes
per year)

Note:
Civil costs are taken as 15 % of capital costs. Running costs are based on maintenance and waste disposal
costs and are equivalent to 10 % of capital costs and power at EUR 0.06/kWh. Financial costs are based on
the plant being amortised over ten years with interest charges of 10 %
Table 4.46: Cost calculation for MgO fume abatement
[161, UK Environment Agency, 2002]

The costs for abatement of the magnesium oxide fume emissions are estimated at about
EUR 51.30 per tonne of metal treated.

Driving force for implementation
Regulations limiting visible emissions and workplace atmosphere quality measures.

Example plants
- Rmheld & Moelle Mainz (D)
- Fuchosa Atxondo (E).

Reference literature
[161, UK Environment Agency, 2002], [110, Vito, 2001]

4.5.8 Mould and core-making

4.5.8.1 Exhaust capture of dust from green sand preparation (sand plant)
and dedusting

Description
Only a few of the different green sand preparation stages cause significant dust generation,
namely the vibrating screen, dedusting and cooling process stages. These stages are equipped
with dust capture and treatment. The units are enclosed and connected to a centralised dedusting
unit. The set-up of the dedusting system has to take account of the dew point of the exhaust air
and the type of dust.

The exhaust air from the green sand preparation is saturated with water. Therefore, wet systems
(often of the low pressure type) seem to be most suitable for exhaust capture. However, the wet
system is widely being replaced by dry dust abatement. The latter has the advantage that part of
the dust may be recirculated and that no waste water stream is generated. Additionally, the wet
systems are prone to internal corrosion and the build-up of dust and oxidation products. The wet
separation of exhausted air impurities can result in problems for the waste water cleaning. Dust
contains bentonite which is difficult to dispose of due to its anti-sedimentable effects. [225,
TWG, 2003]

When applying fabric filters, condensation problems need to be prevented. Condensation may
cause a deposition of dust, clogging and/or a tearing of the filter. The addition of cooling water
to the sand needs to be done in a controlled way in order to minimise steam formation. The dew
point of the airflow may be increased through heating with gas burners.

One benefit is that the collected dust is dry and so can be easily transported. After size
separation, the coarse fraction and part of the fine fraction may be recirculated into the sand
mixing.
Chapter 4
Smitheries and Foundries Industry 239
Achieved environmental benefits
Reduction of dust emissions to air.

Cross-media effects
The use of wet systems cause the production of a waste water flow. Which needs to be treated, a
sludge fraction, which needs to be a disposed of. Some systems work without producing waste
water. The sludge is thickened and, if it contains a sufficient amount of bentonite, is re-used in
the moulding-sand cycle.

Operational data
Using fabric filter equipment, a residual concentration well below 10 mg dust/Nm
3
in the
exhaust air can be achieved. For wet scrubbers, the efficiency is slightly less, normally resulting
in emission levels of 50 100 mg dust/Nm
3
, though lower values have also been reported (see
also Table 4.47)

Equipment Exhaust gas flow
(m/h)
Abatement technique Emission level
(mg/m)
Sand preparation 160 tonne/hour 64000 Venturi 4
Sand preparation 11 tonne/hour 48080 Wet scrubber 7
Sand preparation 57400 Bag filter 6
Shake-out unit 51070 Wet scrubber 7
Shake-out unit 50000 Bag filter 2
Grinding 54000 Bag filter 5
Shot blasting 17000 Bag filter 4
Table 4.47: Dust emission values from moulding and post casting shops
[29, Batz, 1986]

Continuous monitoring data for a large-scale automotive foundry are given in Annex 2. These
data show dust monitoring for sand preparation, sand transport and shot blasting.

Applicability
This technique is applicable to new and existing green sand preparation plants.

Driving force for implementation
Regulation of dust emissions.

Example plants
Generally applied throughout Europe.

Reference literature
[29, Batz, 1986], [32, CAEF, 1997], [110, Vito, 2001], [225, TWG, 2003]

4.5.8.2 Exhaust capture of dust from the green sand moulding shop work
area and from dedusting

4.5.8.3 Exhaust capture from core-making general considerations

The type of binder and hardener used will determine the type of emissions that occur. The
exhaust gas from core-making with chemical binders consists mainly of a mixture of organic
solvents. Emissions of phenol, formaldehyde, or ammoniac etc. are much smaller and depend on
the binder type. Mixing, core shooting and drying can be enclosed to varying degrees,
depending on the emissions occurring. Various binder-specific techniques apply, and these are
discussed in the subsequent sections. Volatile organic carbon emissions occur with all
chemically-bonded sand types, except for water glass sands. For self-setting and cold-setting
Chapter 4
240 Smitheries and Foundries Industry
binders, the emissions during mixing are low, given the low resin usage (1 2 % resin:sand
ratio). The emissions during hardening, coating and drying can be considered for capture and
treatment.

Upon regeneration of the chemically-bonded sand, dust generation occurs in the shake-out
drum. Sand cooling and dedusting is generally performed in a dry unit, since there are no
problems with water condensation. It is not possible to recirculate the dust. [110, Vito, 2001]

4.5.8.4 Cold-box: capture of amine vapour and exhaust treatment

Description
The gassing of cold-box cores produces off-gases that need cleaning before emission into the
atmosphere. In order to safeguard the working atmosphere, the core shooting machines are
enclosed and cores are flushed with air to remove all the gas residues before the machine is
opened up to take out the freshly made core.

Exhaust capture systems may be installed to secure good working conditions during the
checking, handling and storage of the freshly made cores. For this, hoods or evacuation systems
are installed, for example at the checking table, above the handling area and above the
temporary storage area (the waiting area for the tray of ready cores).

The exhausted amine vapour requires treatment to prevent an odour nuisance. One of the
following methods may be used:
- Adsorption on activated carbon: This process is very efficient, but the operational costs are
excessive, so it may only be feasible if the exhausted volume is small
- Combustion: For this process to be efficient, the afterburner chamber must be correctly
designed, to ensure a temperature of at least 800 C, with a minimum dwell time of
2 seconds. Energy consumption is high and the process is therefore expensive to operate. In
foundries operating a cupola furnace, the exhaust gas may be fed to the cupola for
combustion
- Chemical scrubbing: Amines are strong bases, and react with sulphuric or phosphoric acid.
The pH of the solution is normally controlled to below 3. As a consequence, the scrubbing
solution is replaced from time to time to remove concentrated salts, which subsequently
require disposal as a hazardous waste. It is technically possible to process the scrubbing
solution to reclaim the amine for re-use. This is discussed in Section 4.6.5
- Biofilter: See Section 4.5.8.6.

In winter, it may be necessary to heat the exhaust air in order to prevent amine condensation
inside the ductwork. This may be done using waste heat from compressors or from other nearby
devices.

Achieved environmental benefits
Exhaust capture allowing a reduction in amine emissions and related odour emissions.

Cross-media effects
When using acid scrubbers, the amine is transferred to the scrubbing solution, which then needs
further treatment. However, recovery of the amine from the solution is possible. This is
discussed in Section 4.6.5.

Operational data
Whatever the treatment method applied, amine emissions can be kept lower than 5 mg/Nm
3
.
Emission data for a cold-box core-making shop using an acid scrubber are given in Table 4.48.
The scrubber uses a 75 % phospohoric acid solution. In normal operation, 15 kg amine is
disposed per day through the scrubber solution. The solution is saturated and stored in a tank for
external treatment or disposal. The installation operates at 35.5 kW power and for 2000 h/yr.
Chapter 4
Smitheries and Foundries Industry 241
Compound Emission level
mg/m
Dust 0.42
Amine 2.4
Phenol 0.53
Oxygen 21 %
Exhaust gas volume 25100 m/h
Table 4.48: Emission data for a cold-box core-making shop, using an acid scrubber
[29, Batz, 1986]

Applicability
The technique applies for all existing and new cold-box core-making shops.

Economics
Operational costs for the installation mentioned in the operational data section are
EUR 6.3/tonne good casting.

The investment cost for an example installation, using a wet scrubber is given in Table 4.49.

Property Value
Exhaust airflow 30000 m
3
/h
Crude gas amine concentration 150 mg amine/Nm
3
Cleaned gas amine concentration <1 mg amine/Nm
3
Power supply 45 kW
Investment cost EUR 187000
Table 4.49: Specification and investment cost for an amine scrubber on cold-box core-shop exhaust
data for Portugal, 2003

Driving force for implementation
To ensure workers health and to prevent odour emissions.

Example plants
The technique is generally applied for foundries using cold-box binders.

Reference literature
[29, Batz, 1986], [32, CAEF, 1997], [15, Gwiasda, 1984]

4.5.8.5 VOC-collection and removal

Description
VOC emissions (mainly solvents, BTEX, and to a lesser extent phenol, formaldehyde, etc.)
result from chemically-bonded sand preparation and also, further along the process, from
pouring, cooling and shake-out (see Table 3.35 and Table 3.45). These compounds are noxious
and generate odour emissions.

The abatement of VOCs is hindered by the fact that large volumes of ambient air are entrained
by the collection systems (e.g. canopy hoods) used.

The following techniques are applicable for the removal of VOCs:
- adsorption to activated carbon
- post combustion
- biofilter.

Chapter 4
242 Smitheries and Foundries Industry
For adsorption to activated carbon, the exhaust gas flows through a carbon bed. Upon
saturation, the carbon is thermally regenerated. Activated carbon has a very high adsorption
(and abatement) efficiency. For benzene, the efficiency is >99 %. Nevertheless, it has the
following disadvantages:
- the high flue-gas volumes require large amounts of activated carbon
- dust and aerosols need to be filtered out of the off-gas before the carbon adsorption. Since
very fine dust particles show a tendency to stick, the filtering is only possible using wet
dedusting techniques and this thus generates a waste water flow.

To successfully exploit post combustion to eliminate VOCs from off-gases, specific minimum
concentrations are necessary. These limit values are compound specific and depend on the
chosen technique. Post combustion is possible for the exhaust gas from shell moulding.
Generally, the exhaust gas from the casting shop does not have high enough VOC levels to
operate post combustion. One alternative to post combustion is to use exhaust air from the core
blowers as combustion air for the cupola.

The use of biofilters is discussed in detail in Section 4.5.8.6.

Achieved environmental benefits
Reduction of VOC-emissions to air.

Cross-media effects
Energy is consumed in the collection of the off-gas flow, which may also involve entraining
large volumes of ambient air.

Driving force for implementation
Legislation on the emissions of VOC.

Reference literature
[20, Gapp, 1998], [110, Vito, 2001]

4.5.8.6 Exhaust gas cleaning using biofilter

Description
Biofiltration is based on the ability of microbes living in a fibre-peat filter bed to oxygenate
malodorous gases and convert them into odourless compounds. The gas to be purified is blown
by a fan through a bed of humid filter material, exiting odourless from the upper side. The
odorous compounds are adsorbed in the water phase and are decomposed by micro-organisms
living on the filter material. The good performance of the filter depends on the balance between
the supply of nutrient (compounds for abatement/decomposition) and the number of micro-
organisms.

In foundries, biofilters are applied for the removal of odorous gases, which mainly consist of
amines from the off-gas from cold-box core-making processes and for the removal of VOCs
(e.g. benzene) from casting shop exhaust gases.

Critical parameters for good functioning of the biofilter are:
- Permeability of the filter layer: Gases need to flow through the bed in good contact with the
filter bed material. An excess of fine material in the filter bed may cause clogging, pressure
build-up and blowout of bed material
- Wetting of the filter medium: A water dispersion system may be installed for continuous or
intermittent wetting of the bed material
- Conditioning of the exhaust gas: The exhaust gases need to be moistened before entering
the filter. For optimum operation, a constant temperature (30 C) has to be maintained. This
may require preheating of the exhaust gas
Chapter 4
Smitheries and Foundries Industry 243
- Adequate water treatment: The biofilter produces residual water, containing phenol and
cresol. This is treated in a waste water system before disposal or re-circulation. Re-
circulation requires an additional salt removal step.
Introduction of the biofilter technique into a German foundry required a long optimisation
period, until it was able to function satisfactory. It was found that the following characteristics
appear to be important:
- closure of the biofilter beds with a lid, in order to prevent CO- and condensation problems
throughout the installation. The filtered air is evacuated through a centralised chimney
- close control of water quality (e.g. salt content), especially when applying a system not
producing waste water
- the addition of nutrient salts to the sprinkling water, to maintain bed activity, and
- control of the pH-value of the bed material and the addition of lime to maintain a neutral
pH-value.

Achieved environmental benefits
Reduction of odourous, amine and/or VOC emissions from the core shop and casting shop. The
elimination of benzene and other VOCs takes place in biofilters but may not be regarded as
reliable. The efficiency varies remarkably according to a wide spectrum of factors, such as
weather conditions, age of substrates, etc. The use of bioflters alone is therefore not valid as a
stand-alone technique for the reduction of VOCs. Their main purpose consists in the reduction
of odour.

Cross-media effects
The biofilter produces a waste water flow. This may be re-circulated internally after treatment,
yielding a sludge cake for disposal.

Energy is consumed during the transfer of the off-gas to and through the filter, as well as for the
surrounding equipment.

It has been reported that emissions of N
2
O, a greenhouse gas, may be a problem.

Operational data
Pilot scale data for amine removal have been provided by Finland. A biofilter unit (diameter
0.8 m, height 1 m) was installed in a core-making off-gas line, after an acid scrubber. The
equipment included a fan and a dewing unit, with which the dampness of the biofiltration unit
was adjusted to an optimum level. The odour removal was measured using an olfactometer.
Results for the odour and total hydrocarbon levels are given in Table 4.50. Variations in the
input analysis are due to process variations. The results show that filtration results in a
significant odour and hydrocarbon reduction.

Compound Units Before biofilter After biofilter
Odour OU/m 410 3000 150 310
Total hydrocarbons mg C/m 20 35 5 10
OU = Odour units
Table 4.50: Operational data for the biofiltration of a cold-box core-making off-gas
[112, Salminen and Salmi, 1999]

Industrial scale data for VOC removal have been provided by Germany. A biofilter unit
(specification: surface 300 m, height 1 m, in 5 rectangular beds, with a total throughput of
32330 Nm/h dry gas) was installed for the treatment of exhaust gases from pouring and cooling
in a green sand cast-iron foundry, using croning and cold-box cores. The emission data are
given in Table 4.51. Olfactometric measurements showed an average reduction of odour
emissions by a factor of 94.5 %.

Chapter 4
244 Smitheries and Foundries Industry
Parameter Units Before
biofilter
After
biofilter
Reduction
%
O
2
% 20.74
CO
2
% 0.17
NO
x
mg/m 3.65
CO mg/m 382 299 21.7
Benzene mg/m 15.80 1.44 90.9
Toluene mg/m 9.37 0.92 90.1
Ethylbenzene mg/m 3.00 0.46 84.7
Xylene mg/m 4.90 1.54 68.6
Formaldehyde mg/m 0.37 0.01 68.6
Dust mg/m 13.63 2.03 85.1
Ammonia mg/m 8.97 0.16 98.2
Phenol mg/m 4.67 <0.02 >99.6
Cresol mg/m 3.73 <0.02 >99.5
PCDD/F ng/m 0.0056 0.0041 26.8
Benzo(a)pyrene) mg/m 0.0001 0.00003 70.0
Table 4.51: Emission data for the biofiltration of exhaust from green sand pouring and a cooling
line
[20, Gapp, 1998]

Applicability
This technique finds its application in green sand foundries and cold-box core-making shops.
The applicability is limited to 3-shift large series casting foundries, as biofilters need very stable
conditions throughout the year. It should be noted in each known case of application great
individual efforts of research and development have been required to achieve an acceptable
performance.

Economics
Investment costs for the biofilter unit in the German example (specification: surface 300 m,
height 1 m, in 5 rectangular beds, with a total throughput of 32330 Nm/h dry gas), including a
water treatment unit to allow operation without the production of waste water, have been
estimated at EUR 3247000. This includes EUR 2333000 for the filter and surrounding
equipment and EUR 914000 for the exhaust capture and ductwork. With a design load of
40000 Nm/h, operational costs are estimated at EUR 170000/yr (costs based on DEM for
1998).

These estimated operational costs are much higher than generally stated investment and
operational costs for biofilter units. This is probably due to the large throughput and the extent
of the surrounding equipment. In a 1998 paper the cost for an open-bed biofilter, treating
17000 Nm/h was estimated at USD 70000 USD 100000 for investment and a yearly operating
cost for water, labour and overheads of USD 15000 USD 25000.

Driving force for implementation
Reduction of nuisance effects (mainly odour) on neighbouring locations.

Example plants
Halberg Guss, Saarbrcken-Brebach (D)

Reference literature
[20, Gapp, 1998], [112, Salminen and Salmi, 1999], [202, TWG, 2002], [208, Devinny, 1998]

Chapter 4
Smitheries and Foundries Industry 245
4.5.8.7 Permanent moulding: collection of release agent emissions

Description
The spraying of a water-based release agent on HPDC dies generates a mist with low
concentrations of decomposition products. The mist is collected using a suction hood and an EP.

Achieved environmental benefits
Reduction in the emission of release agent containing organic compounds. Prevention of diffuse
emissions.

Cross-media effects
Capture and collection of the mist requires the consumption of energy.

Operational data
Operational data from an example plant are given in Table 4.52. The data refer to a water
miscible release agent (1:40 dilution factor), based on synthetic polymer and polysiloxane. Each
die-casting machine is equipped with a separate suction hood and an EP. The hot and cleaned
air is blown back into the hall (i.e. for heat recovery).

Compound Half hour measured values
(mg/Nm
3
)
Mass flow
(g/h)
Dust 1.5 4.3 5.5
Organic substances
(values in total carbon)
(mg/m
3
)
4 -6 9
Table 4.52: Emission data from EP-cleaned HPDC off-gas
[202, TWG, 2002]

Applicability
This technique applies to existing installations. It is an end-of-pipe measure, applied in cases
where prevention or minimisation of the mist has failed. Prevention and minimisation measures
are discussed in Section 4.3.5.1.

Driving force for implementation
Minimisation of diffuse emissions and legislation on the emission of oil mist (e.g. in Sweden).

Example plants
Reference plant: TCG Unitech, Kirchdorf/Krems (A)
General practice for older die-casting machines (D, F, B,...) and for all die-casting machines in
Italy.

Reference literature
[202, TWG, 2002], [225, TWG, 2003]

4.5.9 Casting/Cooling/Shake-out

4.5.9.1 Introduction

The emissions during pouring, cooling and shake-out are diverse and differ broadly in quality
and quantity from one foundry to another. Basically, dust emissions are produced, as well as
inorganic and organic gaseous compounds. These are mainly reaction products, caused by the
high temperature and the reducing atmosphere upon pouring and cooling. The composition of
pouring fumes is complex. They consist mainly of CO, CO
2
, H
2
, and methane as the main
representatives of the organic decomposition products. Polycyclic aromatic hydrocarbons and
benzene occur, depending on the composition of the mould- and core-system.
Chapter 4
246 Smitheries and Foundries Industry
The capture and treatment of the off-gases of these process stages is generally possible for
automated moulding and pouring installations. [29, Batz, 1986]

Relative emissions (as kg/t of molten metal) increase over the pouring-cooling-shake-out
sequence. Exhaust capture efforts will need to take this variation into consideration.

4.5.9.2 Pouring and cooling lines encapsulation

Description
The collection of emissions during pouring, cooling, and shake-out is as a rule more extensive
as the plant becomes bigger.

The further the pouring process can be restricted to a fixed area or position, the smaller the
amount of extracted air will be, and the easier it will be to capture emissions with ventilators
and housings, and therefore air-treatment can be carried out more effectively.

In serial pouring, the emissions increase with increasing production capacity. Without
extraction, the air at the pouring plant could at some point reach unacceptable concentration
levels of noxious substances. To lessen the pollution in the work area, extractor ventilators or
extractor surfaces are fitted as close to the moulds as possible, but without hindering the pouring
process. The extractor elements are arranged in such a way that all emissions occurring during
pouring, are moved away from the work area to the extracting equipment. The air-speed at the
free extraction cross-section is kept between 0.5 and 1 m/s.

Achieved environmental benefits
Reduction of diffuse emissions of CO, PAHs and other organic decomposition products.

Cross-media effects
Extraction of exhaust air consumes electrical energy.

Applicability
This technique is applicable to new and existing installations utilising serial pouring and
cooling.

Driving force for implementation
Reduction of diffuse emissions. Collection of emissions in order to allow treatment.

Example plants
The technique is commonly applied in large-series foundries.

Reference literature
[32, CAEF, 1997]

4.5.9.3 Exhaust capture and treatment from shake-out

Description
The technique used for the emission collection from shake-out depends on the degree of
mechanisation, the emissions to be extracted and the size range of the castings, in particular the
smallest and largest item cast on the same machine.

The system of reducing emissions most commonly used is a ventilator panel on the side of the
shaker. Usually, panels are installed on both sides of the shaker, regardless of the shaker size. If
possible the rear of the shaker is enclosed in this way too. Frequently an extraction point is
provided under the shaker in the sand-box, operating through a projecting pipe. This presents
the risk of sand being extracted if the box is filled with sand up to the pipe. Moreover the pipe's
inlet is often obstructed by sand lumps or adhering steam and dust.
Chapter 4
Smitheries and Foundries Industry 247
The best way of achieving good emission levels with relatively small ventilation rates is when
shake-out is performed in enclosed units. Roof openings, possibly with a moving screen, doors
and entry/exit chutes make it possible to use a crane or other means of transport. In addition, the
closed cabins reduce the noise level.

In automatic moulding plants, the casting-mould package is often pressed out of the moulding
box by a hydraulic jack with a mounted pressing plate. Afterwards the casting is separated from
the sand firstly on a jig-shaker and then to a rotating shake-out drum or tube or a sand cooler.
Extraction drums or tube drums, which are now increasingly being used, are much more suited
to the collection of emissions than the usual shakers, but they have the disadvantage of possible
odour emissions.

The quantity of extracted air is not only determined by the workplace limit values required, but
also by processing requirements. Large air volumes may be required for cooling and for
protection of the filter bags. This needs to be balanced against the demands of any pollution
abatement techniques, that usually need highly charged waste gases with small air volume
streams in order to be effective.

Appropriate techniques for dedusting are cyclones combined with wet scrubbers or dry filters.
Biofilters are also used and are discussed in Section 4.5.8.6.

For gravity die-casting and centrifugal casting, post combustion, chemical washing and
adsorption to activated carbon are used in order to reduce organic emissions and odour. These
systems have already been described above.

Achieved environmental benefits
Reduction of the emissions of VOC and dust to air.

Cross-media effects
Dust collection produces a residue for disposal.

Operational data
With free-standing extraction shakers up to 2 m in width, for every m
2
of shaker surface 7000 to
9000 Nm
3
/h of air is extracted in the case of a one-sided ventilation screen. With a two-sided
screen, 5000 Nm
3
/h per m
2
of shaker surface is installed.

If the extraction point is under the shaker in the sand-box, the amount of air extracted per m
2
of
shaker surface is designed to be about 700 Nm
3
/h.

In a Polish example foundry for large-size cast steel castings, a removable doghouse type cover
was constructed over the shake-out grate. The doghouse consists of two sideways moving doors,
which are opened and closed hydraulically, as shown in Figure 4.17. The cover reduces dust and
noise emissions into the foundry hall. The exhaust is cleaned using dry de-dusting. The noise
level for shake-out was reduced from 100 dBA to below 85 dBA.

Chapter 4
248 Smitheries and Foundries Industry
Figure 4.17: Doghouse cover for a shake-out grate, for large-size batch castings
[209, Metalodlew s.a., 2003]

Applicability
In plants where clay-bonded sand moulds are shaken out, the emission of steam is significant.
The combination of dust and steam can lead to huge obstructions in the ductwork if the
temperature drops below the dew point or worse, below freezing point. To overcome this
problem, hot air must be injected into the ducts and the piping and filter-housing must be
insulated, or water (free of dust) must be sprayed into the ventilator piping to prevent these
obstructions.

Economics
For the doghouse in the Polish example foundry, as mentioned above, the investment cost was
EUR 220000, including some additional equipment.

Driving force for implementation
Regulations on the emission of VOC and dust.

Example plants
Metalodlew s.a., Krakow (PL)

Reference literature
[16, Gerber and Gwiasda, 1981], [32, CAEF, 1997], [209, Metalodlew s.a., 2003]

4.5.9.4 Off-gas cleaning using biofilter

Description
Biofilters are used to treat emissions of VOC and benzene and are therefore used for the
treatment of off-gas from pouring, cooling and shake-out. The full description and discussion of
the technique is given in Section 4.5.8.6.

Example plants
Several examples in Germany.

Reference literature
[225, TWG, 2003]

Chapter 4
Smitheries and Foundries Industry 249
4.5.10 Finishing/post-casting operations: off-gas collection and
treatment

4.5.10.1 Finishing off-gas collection

Description
Considerable amounts of fume have to be captured and extracted when using thermal separation
techniques. The work area may be as enclosed as much as possible, but ensuring the freedom of
movement and that the inhaled air supply of the worker can be assisted. Dust collection
techniques during abrasive cutting and grinding are differentiated between stationary machines
and hand machines. Rigid hoods are customary at stationary machines, partly in connection with
partial housings.

For stationary grinding machines, the tightly limited abrasive blast goes into an extracting
funnel and is extracted. In the case of abrasive cutting-off stations, it is sensible to additionally
extract emissions from the cabin in the interest of workers health and safety. Handheld machine
emissions are occasionally extracted via the splint protection hood. This is an effective but
unpopular method, as it increases the weight of the machine and impairs its manageability.
Normally hand grinding and hand cutting-off machines are housed in cabins (i.e. partial
housings).

Collection of the dust takes place via extracting walls, roof domes, moveable hoods or
extraction work benches. An effective aid when collecting emissions are hot-air curtains which
direct additional air into the cabin. The additional air used may be cleaned recycled extraction
air, in order to economise on heating energy. However, it should be noted, that a portion of fresh
air will still always need to be provided from the outside.

Dust or pollutant collection is in most cases not needed for sawing, rapping, pressing, chiselling,
stamping and milling. Screens may sometimes be used to protect the operator against rough
particles. Welding is divided into processes with and without additional materials. Work
protection regulations requiring extraction exist for most processes.

Achieved environmental benefits
Reduction of diffuse emissions of metallic particles and dust.

Cross-media effects
The collection and extraction of dust-laden off-gas consumes energy.

Operational data
Operational data have been given in Section 3.11 and show that emission from cutting, scarfing
and welding (without off-gas cleaning) are low compared to those for fettling and shot blasting.
Applicability
As indicated in Table 4.53, some finishing techniques do not generate dust and consequently do
not necessitate an off-gas collection system.

Chapter 4
250 Smitheries and Foundries Industry
Roof
ventilation
Dome-shaped
roof
Rigid hood Adjustable
hood
Housing
Abrasive cutting x x x
Burning x x x x x
Chiselling, needling Off-gas collection seldom necessary
Fettling x x x x
Milling No off-gas collection necessary
Rapping, pressing No off-gas collection necessary
Sawing No off-gas collection necessary
Slide grinding x x x x x
Shot blasting x
Stamping No off-gas collection necessary
Welding x x x x
Table 4.53: Applicability of dust capture techniques for various finishing operations
[32, CAEF, 1997], [225, TWG, 2003]

Driving force for implementation
Regulations on dust emissions and occupational health and safety.

Example plants
The technique is applied in the vast majority of the European foundries.

Reference literature
[32, CAEF, 1997]

4.5.10.2 Exhaust air cleaning techniques

Description
- Blasting: Blasting generates a lot of dust. Off-gas capture is unproblematic due to the total
enclosure of the blasting process in a closed cabin. Customary exhaust air decontamination
processes are wet washing and dry filters, usually with a cyclone as a pre-filter
- Slide grinding, drumming: Both processes do not require emission reduction in normal
conditions. Aerosols that can form in quickly moving drums can be neglected
- Cutting: All thermal separation processes generate emissions. Exhaust capture is achieved
through enclosing the work place and carrying out an intensive extraction of the arising
fume. It is important to place the capture elements as close as possible to the emission
source, but without hindering the operating process. In some cases it has proven beneficial
to combine a mobile arm for direct extraction and additional cabin extraction. By carefully
arranging the extraction elements, the dust-filled air can be guided away from the breathing
area of the worker. The usual exhaust air decontamination processes are wet washing and
dry filters. It should be noted that the systems are set up for the smaller grain sizes of the
emitted fumes. Inertia force separators are applied as pre-separators
- Abrasive cutting: Stationary abrasive cutting installations are extracted. Customary exhaust
air decontamination processes are cyclones, wet washing and dry filters
- Sawing, rapping, pressing: These handling processes cause few emissions and do not
require reduction measures in normal conditions
- Chiselling, needling: In these processes mainly coarse particles are produced which are
difficult to remove through extraction. The work is mainly carried out in cabins for safety
reasons. In special cases, e.g. when removing burnt-on sand, the generated dust may be
extracted by an extraction arm. The exhaust air is decontaminated in cyclones, wet washing
and dry filters

Chapter 4
Smitheries and Foundries Industry 251
- Grinding: Emission capture when grinding on stationary machines takes place in a similar
way as with abrasive cutting, i.e. via fixed funnels into which the abrasive blast is directed.
An enclosed work place is used for manual grinding and abrasive cutting. Extraction walls
can then be applied in the housing. Air decontamination processes are cyclones, wet
washing and dry filters
- Stamping, milling: These handling processes cause few emissions and do not require
reduction measures in normal conditions
- Welding: More or less emissions occur according to the type of the selected welding
process, which is usually be best collected by extraction arms. For exhaust air cleaning, wet
washing and dry filters and occasionally electrostatic filters are used.

Achieved environmental benefits
Reduction of dust emissions.

Cross-media effects
The extraction and cleaning of dust-laden off-gas consumes energy.

Operational data
Wet scrubbers and fabric filters have been installed and operated successfully for many years.
Fettling cabins with high-efficient dust arrestment units can be run stackless, as their output is
cleaner than the usual fettling shop air. They also help to improve workplace conditions.

Operational data is given in Section 3.11 and show that emission factors for cutting, scarfing
and welding (without off-gas cleaning) are low, as compared to those for fettling and shot
blasting.

Using fabric filter equipment, emission levels below 10 mg dust/Nm
3
can be achieved.
Using wet dust removal systems, emission levels below 20 mg dust/Nm
3
can be achieved.

Applicability
These techniques apply to new and existing foundries. In general, non-ferrous foundries do not
apply intensive (dust generating) finishing operations.

Driving force for implementation
Regulations on dust emissions and occupational health and safety.

Example plants
The technique is applied in the vast majority of the European foundries.

Reference literature
[32, CAEF, 1997], [180, Assofond, 2002]

4.5.11 Heat treatment waste gas collection and cleaning

4.5.11.1 Use of clean fuels in burner-heated heat treatment furnaces

Description
A basic process-integrated measure for emission reduction in burner-heated heat treatment
furnaces is the use of clean fuels i.e. natural gas or low-level sulphur content fuel.

Additionally, automated furnace operation allows close control of the working regime and
temperature and minimises excessive energy consumption.

Achieved environmental benefits
Reduction of combustion-based pollutants, such as CO, SO
2
, NO
x
.
Cross-media effects
No cross-media effects apply.
Chapter 4
252 Smitheries and Foundries Industry
Operational data
In an example foundry in Poland, up to 1998 the heat treatment was performed in 3 coal gas
furnaces with manually controlled burners. The coal gas came direct from the coke oven and
was of a variable quality (e.g. CO-content up to 15%). This resulted in high emissions and the
risk of intoxication, together with a poor control of the temperature profile.

In 1998, 2 of the 3 furnaces were newly equipped with computer controlled natural gas burners.
Additionally the furnace lining was renewed. The results of these changes on the operation
were:
- automated control of the temperature profile
- 40 %reduction of gas consumption (volume)
- reduction in cost due to the use of cheaper natural gas
- reduction of emissions of SO
2
, NO
x
, CO and aromatics (see Table 4.54).

Coke oven gas burners Natural gas burners Compound
Concentration
(g/Nm)
Emission
(kg/h)
Concentration
(g/Nm)
Emission
(kg/h)
CO 0.006 0.074 0.000 0.000
SO
2
0.011 0.136 0.003 0.064
NO
x
0.016 0.197 0.004 0.085
Aromatics 0.001 0.012 0.00025 0.0054
Table 4.54: Emission levels from a heat treatment furnace before and after conversion to natural
gas burners system

Applicability
This technique applies to burner-fired heat treatment furnaces.

Driving force for implementation
Regulation of emissions of SO
2
, NO
x
, CO.

Example plants
Metalodlew, Krakow (PL)

Reference literature
[32, CAEF, 1997]

4.5.11.2 Quenching baths

Description
Here, emission capture and reduction essentially involves the capture of fume at the quenching
baths, especially at the oil quenching baths. Roof ventilation, extraction domes and edge
extractors are used. Here similar difficulties arise as with shake-out grids. The baths must be
frequently loaded by a crane, so the extraction hoods need to be installed high above the hall
floor.

Abatement systems are not widely applied at this pointin general, but if any abatement is
performed, electrostatic filters are mostly used.

Achieved environmental benefits
Reduction of diffuse emissions of oil-mists.

Cross-media effects
Exhaust extraction consumes energy.

Chapter 4
Smitheries and Foundries Industry 253
Applicability
The applicable techniques for different types of baths are characterised in Table 4.55.

Roof
ventilation
Dome-shaped
roof
Dome, rigid Dome,
adjustable
Edge
extraction
Water x
Emulsion x x x x x
Hardening oil x x x x x
Note: x : applicable
Table 4.55: Applicability of fume collection techniques for quenching baths
[32, CAEF, 1997]

Edge extractors are only effective up to a certain size and lose their effectiveness almost
completely at the time of the highest emissions, i.e. when a hot workpiece is submerged in the
bath. In spite of this, edge extraction is the most sensible capture measure, utilising big baths
combined with a blast veil and a roof dome.

Driving force for implementation
Regulations on emissions and occupational health and safety.

Example plants
The technique is used in several foundries around Europe.

Reference literature
[32, CAEF, 1997]

4.6 Waste water prevention and treatment

4.6.1 Measures to prevent the generation of waste water

Description
The following measures result in a marked reduction in waste water:

- Use of dry dedusting systems: Dry dedusting systems can be used for most foundry exhaust
gas flows. However, in exceptional cases, it may be necessary to use a wet scrubber for
removal of iron-containing ultra-fine dust fractions. The applicabilities of dry and wet
exhaust cleaning techniques are discussed in Section 4.5
- Biological waste gas scrubbers or compost filters: The use of a biological waste gas
scrubber can be considered if the gases to be treated contain readily biodegradable
substances such as phenols, etc. Biological scrubbers generate less waste water than
conventional wet scrubbers because biologically treated waste water can be recirculated
more often.
Note: Compost filters (biofilters) do not generate any waste water, although the compost
material must be sufficiently moist. The technique of biofiltration is discussed in Section
4.5.8.6
- Internal recycling of process water: An extensive recycling of scrubbing water requires the
use of treatment facilities. These include sedimentation tanks, possibly with an integrated
flocculation stage, and filtration equipment. Depending on the process, cooling may be
necessary. Evaporation losses and substance inputs from gas scrubbing lead to an accumu-
lation of salt in the circulating water. Depending on the salt concentration and the applicable
discharge limit values (in the case of indirect discharge), circulating water may have to be
drawn off. It may be economically favourable to evaporate this waste water and to use the
condensate to make up evaporation losses
Chapter 4
254 Smitheries and Foundries Industry
- Multiple use of treated waste water: Multiple use of treated waste water needs to be
considered, e.g. using process water from sand preparation in the moulding sand cycle or in
slag granulation. The use of cooling water in wet scrubbing may also be considered
- Use of waste heat for the evaporation of waste water: This technique can only be applied
when waste heat is available on a continuous basis. A case-by-case evaluation is needed to
evaluate its applicability and economic viability
- Prevention of waste water from scrap storage: The generation of waste water from the
storage of contaminated scrap can be prevented by roofing the storage area. The run-off
water is collected as uncontaminated rainwater. All techniques for storage are discussed in
Section 4.1
- Avoidance of AOX formation in waste water: Careful selection of scrap types can prevent
the uptake of chlorinated compounds in the production cycle. Analyses for AOX
compounds prior to scrap purchase are a helpful measure. AOX may also be contained in
auxiliaries used in waste water treatment, e.g. commercial-grade hydrochloric acid, iron
chloride or aluminium chloride
- Keep water types separate:Various water flows with distinct pollutant levels are kept
separate, in order to minimise the need for waste water treatment and to optimise the water
use.

Achieved environmental benefits
Prevention of the production of waste water.

Cross-media effects
Cross-media effects may apply, depending on the chosen technique. Information is given in the
specified sections.

Operational data
Operational data are given in the description of the example plant in Section 4.6.3.

The waste water system of an example aluminium foundry consists of 4 separate flows:
- surface run-off: this joins to the central rainwater channel, through an oil separator
(2xNG80), and onto a nearby river
- used water = process water (open cooling baths) and sanitary water: this goes through an
open air channel on site, into a public channel and then onto the public sewage works of the
town
- cooling water: this is initially taken from the nearby river, over a sand filter, and goes
through the installation in a closed percolation cooling circuit, and then back to the river,
with a temperature limited to 28 C. Roof water also enters the cooling water circuit
- water from off-gas scrubbers (green sand, post casting): this is partially evaporated, so that
only the remaining sludge leaves the installation.

Applicability
This technique applies to all new and existing installations.

Economics
Economic data are given in the specified sections for alternative techniques and in the Example
plants section of Section 4.6.3.

Driving force for implementation
Minimisation of waste water production.

Example plants
Honsel, Meschede (D)

Reference literature
[195, UBA, 2003]

Chapter 4
Smitheries and Foundries Industry 255
4.6.2 Treatment of scrubbing water and other waste water flows

Description
For furnace gas cleaning, dry filtration systems eliminate potential waste water streams,
however with a wet cleaning system there is no need for an aqueous discharge if appropriate
measures are taken to clean the water and recycle it. As long as solids are removed to a level
acceptable to the scrubbing device, soluble compounds can normally be allowed to reach
saturation without any adverse effects. There are a variety of techniques, or combinations
thereof, which could adequately separate the solids.

The following treatment techniques are applied:
- sedimentation
- hydroxide precipitation
- multi-stage precipitation
- wet oxidation
- filtration procedures.

The waste water can contain undissolved and dissolved heavy metals, phenols and cyanides.
The treatment needs to be adapted to the type of pollutants.

Undissolved heavy metals have to be removed from waste water by physical methods
(sedimentation, filtration, possibly flotation). Concentrations well below 0.5 mg/l can be
achieved by the use of these methods.

Dissolved heavy metals must first be transformed into poorly soluble compounds using suitable
precipitants. Especially strong bases (lime milk, caustic soda solution, soda) are used as
precipitants for hydroxide precipitation. If this is not sufficient, sulphide precipitation with
organic sulphides or alkali sulphides must be carried out.

Phenols and cyanides can be biologically degraded or removed by physical-chemical treatment.
Normally they are present in such low concentrations that there is no need for targeted
pretreatment. Combined treatment in a sufficiently large biological waste water treatment plant
is sufficient.

Achieved environmental benefits
Reduction of the pollutant load in discharged waste water.

Cross-media effects
Waste water treatment generates a sludge fraction for disposal.

Operational data
Waste water from foundries mainly contains zinc, in addition to iron. Hydroxide precipitation in
a pH range of 8.5 to 11 can reduce the zinc concentration to values below 2 mg/l. This pH range
must be respected, because at a pH in excess of this range amphoteric zinc is re-dissolved, in the
form of zincates.

By way of example, Figure 4.18 shows the combined conditioning and precipitation reaction of
non-dewatered sludge from wet scrubbers of cupola furnaces. Following removal from the wet
scrubbers water treatment system, the sludge is conditioned in a separate treatment stage using
caustic lime. This results in an increase in the pH value and the precipitation of heavy metals.
Organic pollutants are also adsorbed. The sludge is subsequently dewatered in a chamber filter
press. Following turbidity measurements, the filtrate is passed to a storage tank incorporating a
monitoring of pH and conductivity. Depending on its condition and actual needs, the filtrate is
returned to the sludge silo, sent to slag granulation or sent for dry dust humidification via
equalising tanks, or discharged to drain as surplus (following the requirements for discharge to a
municipal waste water treatment plant).

Chapter 4
256 Smitheries and Foundries Industry
Sludge/waste water from wet
process dust separator and
Coagulation
aid
Lime
Storage tank
Air
Sludge silo
Conductivity and
pH-value measuring
Turbidimetry
Filter press
Waste water
Storage tank
for filtrate
Equalising tank for filtrate
To dross granulation
To the humidification
of dry dust
Figure 4.18: Treatment of waste water and sludge from the wet de-dusting system of a cupola
furnace
[195, UBA, 2003]

Additional operational data are given in the description of the example plant in Section 4.6.3.

Applicability
The type of waste water treatment to be applied needs to be selected on the basis of the waste
water composition and the local conditions for disposal of water from the plant into the
receiving water medium.

Economics
Economic data are given in the description of the example plant in Section 4.6.3.
Driving force for implementation
Regulations for pollutants discharge into water.

Example plants
See Section 4.6.3

Reference literature
[195, UBA, 2003], [160, UK Environment Agency, 2002]

4.6.3 Example plant for waste water prevention and reduction

Description
A German example plant produces parts from cast iron (grey and nodular cast iron) and die-cast
aluminium.
The following operations are carried out:
- Melting: Gas-fired melting furnaces and flue-gas treatment plant (dry filter)
- Manufacture of diecast parts: Automated casting-punching cells for aluminium diecasting
- Sand casting: Manufacture of special alloys
- Iron casting (core production): For core production, the cold-box and Croning processes
are used. Scrubbing concentrates are generated in amine scrubbing
- Iron casting (melting): The melting shop consists of a mixing plant, a hot blast cupola
furnace, several electric furnaces, a converter and a holding furnace
- Iron casting (moulding): Use of lost moulds made of bentonite-bonded sand.
Chapter 4
Smitheries and Foundries Industry 257
Production data and water consumption are given in Table 4.56.

Production
Good iron castings, total 46710 t
Good non-ferrous metal castings, total 2355 t
Water consumption
Drinking water (town water) 59630 m
3
Surface water/groundwater 64998 m
3
Table 4.56: Production and water consumption data (on yearly basis) for example foundry
[195, UBA, 2003]

Wet scrubbers are applied where large volumes of waste gas containing relatively high
concentrations of organic substances are generated (waste gas from core-making) or where high
flue-gas temperatures occur (flue-gas from hot-blast cupola furnaces).

A schematic representation of the water circuit of the wet dedusting system is given in Figure
4.19. The waste gas from under-charge cupola flue-gas extraction is fed to a cyclone separator
and a venturi scrubber for dedusting. The subsequent removal of gaseous pollutants is achieved
in a packed bed scrubber. The waste water from the venturi scrubber is pretreated in a float and
sink separator and then passed to a settling tank for secondary treatment. The scrubbing liquid
from the packed tower is fed directly into the settling tank. The settled sludge from the settling
tank can be pumped discontinuously to the float and sink separator. The sludge from the float
and sink separator is pumped to a scraper conveyor, agglomerated by the addition of flocculants,
and passed into a storage tank. From there, the sludge is fed to the sludge silo and is
subsequently solidified in a mixing plant by the addition of dry dust.
Chapter 4
258 Smitheries and Foundries Industry
Packed bed
scrubber
Srubber
DSSS
FA
Air Scraper conveyor
Setting
tanks
Dry dust Sludge silo
Mixer
Figure 4.19: Water circuit for the wet dedusting system of a cupola furnace
DSSS: double shell sinking separator; FA: flocculant, M: motor
[195, UBA, 2003]

In this basic layout, the plant experienced some operational problems, due to salt accumulation
and the low final water quality. An extension was made to the plant, with the aims of:
- increasing the settling velocity, by the use of other flocculants
- reducing the pollutant content in the scrubbing medium
- improving sludge dewatering
- re-using the scrubbing medium from the gas scrubbing cycle.

For this purpose, a chamber filter press with a volume of 2 m
3
and a sludge collection tank
(30 m
3
in volume) with an agitator were installed. The sludge was passed from the scraper
conveyor directly into the storage tank and preconditioned with lime.

During the trial phase, which lasted several months, it was found that approximately 25 kg of
lime per batch were needed to achieve the best result in terms of dewatering and heavy metal
removal. The average dry matter content was 70%. Analysis of the filtrate for heavy metals
showed values in the range of detection limits. However, the relatively high sulphate
concentrations of about 1 g/l posed a problem for re-use of the filtrate in the scrubbing cycle.
Chapter 4
Smitheries and Foundries Industry 259
As a practicable alternative, the sludge treatment was further extended with other compounds as
shown in Figure 4.20. The water/solids mixtures from wet scrubbing were pre-conditioned by
the addition of flocculant and the resulting sludge was fed to a mixing tank by a scraper. Lime is
added to the mixing tank in a quantity which needs to be determined previously through tests,
and the sludge is fed to the sludge silo by a pump (pump P1 in Figure 4.20). Pump P2 pumps it
to the filter press.

The filtrate is passed via a turbidity meter into a storage tank in which pH and conductivity are
monitored. There are three possibilities for where it goes from there:
- via an equalising tank to slag granulation or dry dust humidification
- back to the sludge silo (if the turbidity limit is exceeded)
- to drain (emergency measure only).

Sludge silo
Sludge/waste water from wet
dedusting and cleaning
Lime
Flocculant
Air
Filter-press
Turbidity
P 1
P 2
P 4
P 5 Pumping chamber for filtrate
Channel
V 3
V 1
V 2
Buffer for filtrate
To dry dust humidification
To slag granulation
pH, conductivity
Figure 4.20: Treatment system for sludge from the wet dedusting of cupola flue-gas
[195, UBA, 2003]

Economics
Economic data concerning the extension of the system with the sludge treatment are given in
Table 4.57.

Before extension After extension
Generated sludge with a dry
matter content of 30%
2000 t/yr Generated sludge with a dry
matter content of 70%
850 t/yr
Dry dust for sludge
solidication
1100 t/yr Dry dust for sludge
solidication
0 t/yr
Landfill price EUR 100/t Landfill price EUR 100/t
Disposal costs per year EUR 310000 EUR 85000
Table 4.57: Disposal costs for the extension of a waste water system with sludge treatment
[195, UBA, 2003]

The sludge with 70 % dry matter content does not need further stabilisation before disposal. For
the wet sludge, a solidification with dry dust is needed. The saving in disposal costs is
EUR 310000 EUR 85000 = EUR 225000. This compares to an investment cost for the
extension totalling EUR 175000 and an additional annual operating costs of about EUR 50000.
The total cost saving therefore amounts to EUR 175000/yr, giving a 1-year payback period. The
calculation does not include the cost savings from the re-use of the filtrate.
Chapter 4
260 Smitheries and Foundries Industry
Example plants
Sach Giesserei GmbH, Kitzingen (D).

Reference literature
[195, UBA, 2003]

4.6.4 Oil interceptors

Description
Run-off from all open areas, but particularly from raw materials stocking areas, contain
suspended solids, which can be removed by settlement or other techniques. Oil interceptors are
used in draining scrap handling areas. Drainage sumps are designed with a sufficient size to
handle storm-water and to accommodate any storm surge, in order to prevent carry-over of
unsettled material.

Oil interceptors are used on waste water from permanent mould casting foundries. The
hydraulic systems of die casting automates can potentially leak oil. The water collection system
is thus designed in such a way that any oil from leakage is collected and the resulting waste
water flow is treated using an oil interceptor.

Achieved environmental benefits
Prevention of water pollution.

Cross-media effects
Oil interception generates a residue for disposal.

Applicability
This technique applies to new and existing die-casting (permanent mould) foundries and to all
other foundries with a waste water discharge directly to surface water.

Driving force for implementation
Regulations for pollutants discharge into water.

Example plants
Honsel, Meschede (D).

Reference literature
[160, UK Environment Agency, 2002]

4.6.5 Recovery of amine from scrubbing water

Description
When acid washing the cold-box exhaust gas, amine sulphate is formed (see Section 4.5.8.4).
This amine may be recovered through neutralisation with sodium hydroxide, followed by
distillation.

As shown in Figure 4.21, the salt that is formed from tertiary amine (e.g. DMEA, DMIA and
triethanolamine (TEA)) in the acid scrubber, and the scrubbing acid (e.g. sulphuric acid) are
retransformed by reaction with a strong base (e.g. caustic soda solution), thereby forming free
amine and e.g. sodium sulphate. The amine is expelled with water vapour and is then cleaned
and concentrated in a column, to an extent that allows it to be re-used. The sodium sulphate
solution (together with polluted scrubbing solution) can be recovered or may be disposed of.
If the amine is not recycled, the scrubbing concentrates can be treated in a biological waste
water treatment plant with nitrogen elimination.

Chapter 4
Smitheries and Foundries Industry 261
Wet scrubber liquid
Storage tank
Storage tank
Column
Condensator
NaOH
Emission condensator
Waste air to the
amine scrubber
Brine
Tank Tank
Tank
Brine Brine
Steam
Water Discharge
of amine
Discharge
of amine
Na
2
SO
4
Figure 4.21: Amine recycling plant
[195, UBA, 2003]

Achieved environmental benefits
Recovery of a chemical compound, which may be re-used.

Cross-media effects
If the operation is performed in a centralised plant, it involves transport of the scrubber liquid to
the treatment unit.

Applicability
This technique is applicable to all amine-containing scrubber liquors from core-making exhaust.
The implementation is limited by economic factors (e.g. transport costs), since sufficient
quantities of amine scrubber liquid need to be available. In general, the treatment is applied in a
centralised location, treating the effluents of several foundries. Since the operation is subject to
an explosion hazard, any necessary measures to prevent explosion risks should be taken.

Driving force for implementation
Recovery of chemical compounds. Disposal costs for industrial waste.

Example plants
- Centralised recycling plants Voerde (D)
- Centralised plant for group of automotive foundries: Poitou (F).

Reference literature
[195, UBA, 2003]

Chapter 4
262 Smitheries and Foundries Industry
4.6.6 Reduction of glycols in die-casting waste water streams

Description
Hydraulic systems for die-casting use water-glycol mixtures, such as hydraulic liquid. Leaks in
the hydraulic system and consequent transport of the leaked liquid into the water system could
lead to the presence of glycols in the foundry waste water. Removal of the glycol is not possible
using filtration or flotation techniques.

The applicable treatment techniques are:
- distillation or vacuum evaporation
- biological degradation.

The cleaned waste water may be re-used as a solvent for the release agent.

Achieved environmental benefits
Reduction of water pollution.

Cross-media effects
In the waste water treatment, a sludge cake is produced and needs disposal.

Operational data
In an example Belgian foundry, process water is treated in a vacuum evaporation unit (capacity
3 m/day). Three sources feed the unit: excess release agent (collected under the HPDC units),
leakage hydraulic fluid from casting automates (water + glycol), and water from E-filter
cleaning. The vacuum evaporation produces a sludge cake for disposal and an effluent. The
effluent has a high COD and a low hardness (alkalinity). This makes it well suited for re-use as
a solvent for the release agent.

Applicability
This technique applies to all new and existing die-casting foundries.

Driving force for implementation
Regulation for pollutant discharge into water.

Example plants
Metaalgieterij Giessen, Hoboken (B): aluminium foundry using HPDC and gravity die-casting.

Reference literature
[202, TWG, 2002]

4.7 Energy efficiency

4.7.1 Introduction

Metal melting and the holding of molten metal in the molten state generally take up a significant
part of a foundrys energy consumption. In many foundries more energy is used in holding the
metal in the molten state than in the actual melting process. Furthermore, considerable energy is
expended in areas other than that of metal supply. For example, foundries are generally large
users of compressed air. Other large users of energy are likely to be items such as high-pressure
die-casting machines, whose mainly hydraulic power units are also driven by electricity. Sand
and gravity die foundries may employ hot core-making processes, such as shell or hot-box,
using gas or electricity to heat the boxes. The heating of dies and ladle and furnace linings can
amount to an appreciable fraction of the total energy used. The typical relative distribution of
energy use in two foundry types is given in Table 4.58.
[64, ETSU, 1997]

Chapter 4
Smitheries and Foundries Industry 263
Activity Non-ferrous foundry
(%)
EAF steel foundry
(10000 t/yr good casting)
(%)
Melting 30 44
Holding 30
Plant actuation 15
Air compression 14 10
Tool heating 3
Heat treatment (gas) 7
Others * 8 39
(*) Includes all activities for which no values are given in the same column and activities
which are not mentioned
Table 4.58: Typical energy use in a non-ferrous and EAF steel foundry
[64, ETSU, 1997], [202, TWG, 2002]

These data show that foundry services (i.e. all but melting and holding) are responsible for up to
half of foundry energy consumption. This includes activities using items such as motors and
drives, compressed air, lighting, space heating and boiler plant. Energy efficiency measures
should have regards to both the melting and the services field. [46, ETSU, 1995]

Energy efficient operation is brought about by applying good practice measures to reduce
energy use in all of the mentioned steps. Improved energy efficiency is one of the main
environmental benefits mentioned throughout the techniques discussions this chapter. The
following sections focus on specific techniques aimed at heat recovery and heat transfer to other
parts of the foundry.

4.7.2 Induction furnace: waste heat utilisation

Description
A significant proportion of the electrical energy which is supplied to an induction-melting
furnace is converted into waste heat. About 20 to 30 % of the total energy input to the plant is
dissipated through the cooling system. The furnace cooling circuit not only deals with the
electrical losses in the induction coil, but also protects the coil from heat conducted through the
furnace lining from the hot metal in the crucible. The heat in the furnace cooling system is used
in some installations for space-heating, the heating of shower water and for drying raw
materials.

- Drying of raw materials: Where metallic charge materials are added to a molten heel in an
induction-melting furnace, the presence of water in the scrap can potentially be very
dangerous. Although scrap may be stored under cover at the foundry, it is possible that it
may be wet when delivered by the scrap dealer. The heat in the furnace cooling water can
be extracted in an air/water heat-exchanger and a fan can be used to convey the warmed air
to the bases of the stockyard bunkers. A diagrammatic layout of such an installation is
shown in Figure 4.22.

Chapter 4
264 Smitheries and Foundries Industry
Figure 4.22: Use of waste heat for scrap drying
[47, ETSU, 1992]

- Space heating and hot water supply: A similar system to the one discussed above may be
used to blow hot air into the foundry hall for space heating. Alternatively, a water-water
heat-exchange is used to heat a water circuit for radiators or for hot water supply.

Since the temperature of the cooling water in an unpressurised system is normally unlikely
to exceed 60 70 C, radiators will need extra large surface areas are necessary in order to
transfer heat efficiently. Alternatively, the temperature of the water can be boosted by some
other heat source, such as supplementary gas or oil burners or an electrical heater. Some
alternative heating system may be necessary to take over when the furnaces are not
operating; for example, early in the morning, to raise the temperature of the work areas to a
comfortable level.

The normal furnace cooling facilities must be kept operating during the summer period and
at other times when the space-heating facilities are not extracting an appropriate quantity of
heat from the furnace cooling system. It is essential to ensure the integrity of the furnace
cooling system. The whole installation must be designed to be fail-safe, to protect the
integrity of the furnaces at all times. Also water returning to the furnaces must not be too
cold (i.e. not below 30 C). Temperature operated alarms should be provided and
emergency bypass pipework with easily accessible manual control valves, should be
installed, to ensure that the heat recovery facilities can be isolated quickly from the primary
cooling circuit in the event of any problems.

Achieved environmental benefits
Increased energy efficiency.

Cross-media effects
No cross-media effects have been reported.

Operational data
A heat recovery system using the cooling oil of induction furnaces was installed in a Belgian
foundry. The foundry operates two induction holding furnaces in duplex with a cupola furnace.
Chapter 4
Smitheries and Foundries Industry 265
The inductors of the electrical furnaces are cooled using thermal oil. The thermal oil gets heated
up to 200 300 C and looses its heat over an outside oil-air heat-exchanger. Before installation
of the heat recovery system, 1 MW of heat was dissipated into the air. An alternative system
was installed to use the waste heat for space heating. The heated air is introduced into the core
shop. This allows the recuperation of 1/3 of the dissipated heat and replaces the original gas-
fired heating system. The implementation was possible at low cost because the oil-air heat-
exchanger is installed next to the core shop. Space heating in other parts of the foundry may be
considered later, but will require more piping (and will subsequent thus involve further losses).

Applicability
Before heat recovery can be applied, a number of criteria must be met:

- a worthwhile application for the waste heat must be reasonably nearby and the times at
which this recovered heat can be utilised must match the times at which the furnace is
operating. However, typically the heat available is fairly low grade. The temperature for the
cooling water must not exceed 70 C
- the relatively low temperatures involved mean that heat-exchangers need to be much larger
than those which are normally encountered
- the furnace water must not be returned to the furnaces at a temperature lower than about
30 C, otherwise this may give rise to condensation problems
- maintaining the integrity of the cooling circuits is absolutely essential. The cooling circuit is
provided to protect the coil - if it fails in its task the results can be disastrous.

The above aspects, particularly the question of furnace integrity, discourage most furnace
operators from even contemplating the utilisation of heat from the cooling circuit.

Economics
A foundry attempting to make use of the heat from the cooling circuit needs to fully evaluate the
benefits and then compare them with the cost of the additional equipment and the safety of the
furnace and operators.

Driving force for implementation
Increasing energy efficiency at the foundry.

Example plants
Space heating using hot air: Proferro, Oudenaarde (B).
Metso Paper Jyvskyl Foundry (F).

Reference literature
[47, ETSU, 1992]

4.7.3 Cupola furnace: waste heat utilisation

Description
The need to cool cupola off-gases before they enter the bag filter leads to the possibility of
attaching to a secondary user and applying waste heat utilisation. The secondary user may be
e.g.:
- a steam boiler
- a thermal oil circuit
- a heating circuit
- a hot water circuit.

Achieved environmental benefits
Recovery of waste heat, which otherwise would be lost to the outside, allowing a reduction in
fuel (or other sources of energy) consumption.

Chapter 4
266 Smitheries and Foundries Industry
Cross-media effects
No cross-media effects apply.

Operational data
The two example plants described in Section 4.5.2.2, are equipped with a waste heat utilisation
system. Plant data have been given in Table 4.37.

Plant G uses cupola waste heat for electricity production. Part of the off-gas flow is fed to a
steam boiler, which drives a turbine, coupled to a generator or compressor. A schematic
representation of the installation is given in Figure 4.23. In total 29 % of the introduced cokes
heat is converted into additional utilisation. Around 2.9 MW of electrical energy are generated.
This means that the plant generates an electrical power of 75 kWh/tonne liquid iron.

Figure 4.23: Schematic representation of a hot blast cupola with a steam boiler, turbine and
generator
[27, Kran, et al., 1995]

Plant H uses cupola waste heat in a thermal oil circuit. After heat exchange for blast preheating,
the off-gas passes a gasoil heat-exchanger. The heated oil is used for core drying. A maximum
heat recovery of 21 MW can be attained. A process flow diagram is given in Figure 4.24.

Chapter 4
Smitheries and Foundries Industry 267
Figure 4.24: Process flow diagram of a hot blast cupola with heat recovery
[17, Strau, 1983]

Applicability
This technique applies to new installations and should be taken into account when designing the
process. For existing plants, the technique can be applied during major refurbishment of the
plant, however, small add-on units can generally be accomodated in existing plants.

Economics
The stated examples were installed as part of a major rebuilding of the considered installation. It
is therefore not possible to extract specific cost data.

Driving force for implementation
Increasing energy efficiency of industrial processes.

Example plants
The two mentioned example plants are located in Germany.

Reference literature
[17, Strau, 1983], [27, Kran, et al., 1995], [202, TWG, 2002]

4.7.4 Reduce energy loss/improve preheating practice of ladles

Description
Energy is wasted if the molten metal transfer system allows an excessive loss of metal
temperature between furnace tapping and mould pouring. Losses can be prevented by using
good practice measures. These imply the following:
- the utilisation of clean ladles, preheated to bright red heat
- the utilisation of distribution and pouring ladles, which are as large as is practicable and are
fitted with heat-retaining covers
- keeping the covers on ladles which are standing empty or putting ladles upside down when
not in use
- minimising the need to transfer metal from one ladle to another
- always conveying the metal as quickly as possible, while still complying with safety
requirements.

Chapter 4
268 Smitheries and Foundries Industry
Achieved environmental benefits
To reduce energy losses.

Cross-media effects
No cross-media effects occur.

Applicability
Since this technique involves measures related to good practice, it is applicable to all new and
existing foundries.

Economics
No economic data can be given.

Driving force for implementation
Energy efficient foundry management.

Example plants
These measures are used to a varying extent in European foundries.

Reference literature
[44, ETSU, 1993]

4.8 Sand: regeneration, recycling, re-use and disposal

4.8.1 Introduction

Since foundries make intensive use of sand as an inert primary material, the regeneration of this
sand is a major point of consideration as part of its environmental performance. A clear
distinction must be made between green sand and chemically-bonded sand. Green sand can be
easily reconditioned after use. Indeed, recirculated green sand shows a better technical quality
than new sand. Most green sand foundries perform primary regeneration.

Primary regeneration, also known as attrition or particulation, involves breaking down the sand
from moulds or cores back to its original grain size. This includes screening the sand, removing
tramp metal, and separating and removing fines and over-sized agglomerates. The sand is then
cooled before being sent for storage, returned to the sand system or blended with new sand. At
this stage, the sand grains are likely to retain a partial coating of spent binder. This affects the
amount of reclaimed sand that can be used to make moulds and, more particularly, cores. New
sand therefore has to be added to ensure that the sand mix produces adequate mould and core
strength and subsequenty aids good casting quality. Primary reclaimed sand is not generally of
sufficient quality to be used for core-making, without further processing to remove residual
binder materials, and is therefore used principally for moulds. The main primary regeneration
techniques are vibration, rotating drum or shot blasting.

Secondary regeneration involves the further processing of the previously particulated sand to
remove residual binder. The sand is returned to a quality similar to, or better than, that of new
sand. Foundries using secondary regeneration have, in some cases, virtually eliminated the need
for new sand. To remove residual binder, more aggressive techniques are needed than for
primary regeneration. The main secondary regeneration techniques are:
- cold mechanical treatment:
low energy attrition: friction, impact (for cold-setting resins)
high energy attrition: pneumatic chafing, grinding, centrifugal friction
- thermal treatment (usually in a fluidised bed)
- wet scrubbing.

Chapter 4
Smitheries and Foundries Industry 269
Sands bonded with cold-setting resins may be regenerated using simple treatment techniques,
due to the fragility of the binder layer. Mechanical regeneration systems (e.g. fluidised bed
systems) are based on interparticle friction or impact.

Sands bonded with gas-hardened and thermosetting resins need more intensive treatment to
remove the binder layer. These include grinding, pneumatical chafing and centrifugal friction.
Silicate sands can only be regenerated mechanically using pneumatical treatment.

Thermal treatment involves the burning of the organic binder. Bentonite is inactivated by the
high treatment temperature. For sand flows containing green sand, any thermal treatment should
therefore be combined with a mechanical treatment.

Wet regeneration involves binder removal through interparticle grinding. This technique applies
only for green sand and silicate or CO
2
-bonded sands and is not widely applied.

Secondary regeneration of green sand as a monosand flow finds limited implementation. For
chemically-bonded sands, mechanical regeneration is most widely applied (>200 units in
Germany in 1999) [80, ERM Lahmeyer International GmbH, 1999]. The applicability of the
various regeneration techniques and of the regenerated sand are summarised in Table 4.59
(monosands) and Table 4.60 (mixed sand). Each of the presented techniques will be discussed
in more detail in the sections below.

The most important monosand flows for secondary regeneration are the core sands in non-
ferrous foundries. Due to the low thermal load they are easily separated from the green sand.
Furthermore monosands are produced from mould and core-making with purely organic
systems such as croning, furan resin and urethane cold-box. A smaller monosand stream is non-
cured core sand, arising from broken or rejected cores in the core-making shop and the residual
sand of core-making machines.

Mixed sands generally contain bentonite-bonded sand as well as chemically-bonded sand. They
are mainly generated in iron foundries and represent some 75 % of the total waste sand
production.

Chapter 4
270 Smitheries and Foundries Industry
Type of sand Regeneration
technique
Regeneration
equipment
Utilisation Borderline conditions Minimal
quantity
(tonne/h)
Organic
monosystems

Cold-setting
resins
Mechanical or
thermal
Mechanical:
friction, impact,
pneumatic chafing
Thermal: turbulent
bed, fluidised bed
or rotary furnace
- for mould-
making using
cold-setting
resins
- 20 25 %
substitution of
new sand for
cold-setting
core-making
- mechanical: only if
binder shells have
become fragile enough
upon pouring
- the target values for
regenerate quality
must be met
1.5
Cold-Box,
SO
2
, Hot-Box
and Croning
sand
Mechanical or
thermal
Mechanical:
pneumatic chafing,
centrifugal friction,
fluidised bed
chafing
Thermal: turbulent
bed, fluidised bed
or rotary furnace
In core-making,
as new sand
substitute
- mechanical: only if
binder shells have
become fragile enough
upon pouring
- the orientative values
for regenerate quality
must be met
- re-use of fines
0.75
Resol- ester
methyl
formate
hardened
sand
Mechanical Mechanical:
friction, impact,
pneumatic chafing
With limitations,
in mould-
making with
methyl formate
sands
- regeneration as resol-
ester cold setting but
with lower yield
- embrittlement of
binder components

Inorganic
monosystems

Green sand Mechanical pneumatic chafing,
grinding
Renewal sand
for green sand
circuit
- requires predrying
- re-use of fines
0.75
Sodium
silicate sand
Mechanical Only for making
moulds and
cores using
sodium silicate
sand
Embrittlement of binder
components at 200 C
0.5
Table 4.59: Fields of application of different regeneration systems for monosands
[128, IHOBE, 1998], [225, TWG, 2003]

Type of sand Regeneration
technique
Regeneration
equipment
Utilisation Borderline conditions Minimal
quantity
(tonne/h)
Mixed
organic
systems
Mechanical or
thermal
Mechanical:
pneumatic chafing,
centrifugal friction,
fluidised bed
chafing
Thermal: turbulent
bed, fluidised bed or
rotary furnace
- core-making
substituting
new sand
- mechanical: only if
binder shells have
become fragile
enough upon pouring
- the orientative values
for regenerate quality
must be met
- re-use of fines
0.75
Mixed sands
containing
bentonite
Mechanical or
mechanical-
thermal-
mechanical
Mechanical:
grinding, pneumatic
chafing, fluidised
bed chafing
Thermal: turbulent
bed, fluidised bed or
rotary furnace
- core-making
substituting
new sand
- renewal sand
for the green
sand circuit
- require pre-drying
- combined thermal
regeneration requires
efficient mechanical
regeneration to
remove active
bentonite
- re-use of fines
0.75
Table 4.60: Fields of application of different regeneration systems for mixed sands
[128, IHOBE, 1998]

Chapter 4
Smitheries and Foundries Industry 271
The applicability of the various treatment systems is summarised in Table 4.61 and will be
discussed further in the specific applicability sections.

Cold mechanical
S
i
m
p
l
e
m
e
c
h
a
n
i
c
a
l
s
y
s
t
e
m
s
G
r
i
n
d
i
n
g
I
m
p
a
c
t
d
r
u
m
P
n
e
u
m
a
t
i
c
c
h
a
f
i
n
g
W
e
t
r
e
g
e
n
e
r
a
t
i
o
n
T
h
e
r
m
a
l
M
e
c
h
a
n
i
c
a
l

t
h
e
r
m
a
l

m
e
c
h
a
n
i
c
a
l
Monosands
Cold setting x x x x x x 0
Cold-box, SO
2
, Hot-box, Croning 0 x x x 0 x 0
Silicate (CO
2
or ester) 0 0 0 x x 0 0
Green sand (primary) x 0 0 0 0 0 0
Green sand (secondary) 0 x 0 x x 0 0
Mixed sands
Mixed organic 0 x x x 0 x 0
Mixed green + organic 0 x 0 x x 0 x
x: Applicable; 0: Not applicable
Table 4.61: Applicability of various sand regeneration techniques to various sand types

The application of both primary and secondary regeneration in a mixed sand foundry in order to
achieve a total regeneration of 92 % is shown in Figure 4.25. This simplified scheme does not
take into account the various losses in the processing steps. The addition of new sand can be
reduced to a minimum by integrating (the coarse fraction of) the filter dust sands (from the
exhaust of hand forming lines, de-coring lines, storage silos, etc).

Figure 4.25: Sand balance diagram for a thermal/mechanical regeneration system
[70, ETSU, 1998]
Chapter 4
272 Smitheries and Foundries Industry
An overall reclamation ratio of 92 %, as given above, is a normal value for mixed green sand
chemically bonded sand systems. Regeneration ratios of up to 98 % have been reported. The
actual ratio depends on the volume and chemical composition of the used cores. For furan cold
setting monosands, values around 78 % are reported.

Generally the mixing of different types of sands has a negative effect on the strength of the
cores and subsequently the moulds made with the regenerated sand, although there are a few
exceptions to this general principle. In order to produce a good quality regenerated sand it is
therefore of great importance to keep non-compatible sand types separate. Optimisation of the
regeneration potential may therefore imply changing to compatible binder systems, if mixed
sands are used, or the application of (shake-out) techniques that allow the separation of various
sand types. Table 4.62 provides a cross-compatibility table.

Goal system Cold-box
S
o
u
r
c
e
s
y
s
t
e
m
Bentonite Silicate
Cold-
setting
Hot-
box
Croning
Methyl-
formate
Amine SO
2
Bentonite + 0 0 - 0 0 0 0
Silicate 0 + - - - - - -
Cold-setting + - + - 0 - 0 +
Hot-box 0 - + + + - 0 +
Croning + + + + + + + +
Methyl-
formate
0 + 0/- - - 0 0 -
Amine + + + + + 0 + +
C
o
l
d
-
b
o
x
SO
2
+ - + - 0 0 + +
+: Compatible, 0: Limited compatibility, -: Incompatible
Table 4.62: Compatibility of regenerated sources sands with various binders
[37, Winterhalter, et al., 1992], [225, TWG, 2003]

[37, Winterhalter, et al., 1992], [70, ETSU, 1998], [128, IHOBE, 1998], [138, Metaalgieterij
Giesen B.V., 1996], [153, Umweltbundesamt, 2002], [202, TWG, 2002], [225, TWG, 2003]

4.8.2 Regeneration of green sand under optimised conditions
(primary regeneration)

Description
One of the major advantages in using green sand moulding is that the sand from the moulds can
be reconditioned after pouring for multiple re-use. The addition of a minimum percentage of
new sand is performed in order to maintain the quality of the moulding-sand. The amount of
new sand added is determined by the input of core sand and the losses in the process. For
coreless moulding, the average sand renewal ratio is 2 5 %. In casting processes using cores,
sand renewal occurs through the introduction of the core sand into the loop. In any case, surplus
sand is removed from the sand circuit after the shake-out screen or from the storage silos. The
general treatment is depicted in Figure 2.28. This internal recirculation of green sand with
minimal treatment is referred to as primary regeneration. This regeneration basically has three
aims: (1) to break the sand into its original grain size or small particles, (2) to remove the fines,
and (3) to cool the sand before blending with new sand.

Various techniques are applied for the breaking and separation:
- Vibration: The vibratory grid, or screen equipment, is the most widely used technique for
primary reclamation purposes. The sieved sand is removed for subsequent treatment, e.g.
for cooling, size classification, and thermal reclamation, and the residual material is
collected for disposal
Chapter 4
Smitheries and Foundries Industry 273
- Drum: Sand from the knock-out operation is loaded into a rotating drum fitted with both
lifting and travelling bars. As the sand travels along the drum, the rotating and lifting action
causes the sand particles to grind against each other and break down to individual grains.
The sand grains fall through screens at the end of the drum, while oversize and tramp
materials are removed for landfill disposal
- Shot blast: The mould and casting is loaded directly into the shot blast machine. The action
of the shot blast results in the complete disintegration of the mould and also cleans the
surface of the casting. The sand and shot are subsequently separated. However, this
technique is not very common.

During cooling of the castings, the sand gets heated-up. In order to reach good sand mixing
conditions, the sand needs to be cooled down to 40 45 C. Evaporative coolers are used with
turbulent and fluidised beds, as well as heat-exchangers. If shake-out is done using the vibrating
transporters of revolving drums, the cooling can occur at the same time.

If a fluidised bed cooler is used, the drying air may be heated using gas or electricity. The
average treatment time in the bed is 10 minutes. In general, the sand enters the fluidised bed
with a humidity of 2 3 % and a temperature of 250 300 C. Temperature and humidity
control may be performed before the fluidised bed cooler with a subsequent addition of
controlled amounts of water. This allows minimisation of the amount of fines that are removed
during the fluidised bed drying. The fines contain bentonite that may be re-activated. The level
of humidity of the return sand needs to be kept between 2 and 2.2 % at 35 C. Consequently,
during storage, the bentonite contained in the cooled and wet sand starts to activate and, during
the mixing, the water and bentonite that need to be added are reduced, as is the cycle time.

A good homogenisation of the sand allows a constant sand quality and better and easier sand
preparation. Homogenisation systems consist of the use of a few small silos instead of one big
one, or in the re-circulation of the sand in the silo.

In systems using chemically-bonded cores, the mixing of the core sand may have a negative
effect on the sand quality, depending on the binder type and the amount of core sand mixed. The
negative effects are more significant with acidic and alkaline cores, than compared to neutral
systems (SO
2
-epoxy, PUR-cold-box). During shake-out, core sand and moulding sand will
inevitably get mixed. Uncured cores and non-broken cores, however, may be left out or taken
out of the sand before feeding back to the primary regeneration.

Achieved environmental benefits
Reduced usage of primary materials (sand and bentonite), reduced amount of material for
disposal.

Cross-media effects
Sand regeneration uses electricity, therefore it increases the overall electric energy consumption
of the installation. Since mechanical techniques are mainly applied, this increase is low.
Sand cooling and the separation of fines leads to dust-laden exhaust gas. The exhaust needs to
be filtered to prevent dust emissions. The collected dust is sent to disposal or re-used (see
Section 4.8.12).

Operational data
While the level of new sand addition is governed by a number of factors, the usual range is
10 - 20 % of the poured metal weight. However, it is more convenient to consider new sand
additions as a percentage of sand throughput. For most foundry processes, a 5 % addition is
considered sufficient, but many foundries work at lower rates.

For a green sand monosystem, regeneration ratios of 98 % may be achieved. Systems with a
high degree of incompatible cores, may achieve a regeneration ratio of 90 94 %.

Chapter 4
274 Smitheries and Foundries Industry
Applicability
This technique is applicable to all green sand foundries, in new and existing installations.

Economics
The potential cost benefit of reclamation for a foundry that does not currently reclaim its sand is
summarised in Table 4.63.

Description Cost
(EUR/tonne of sand)
Average price of silica sand 32.64
Average cost of waste sand disposal 14.56
Total cost of sand purchase and disposal 47.2
Estimated depreciation cost to recoup
equipment expenditure in one year
18.24
Average operating cost 7.76
Anticipated cost saving in first year 21.2
Anticipated cost saving in subsequent years 39.44
Table 4.63: Cost benefits from primary reclamation
(UK estimation 1995)

The investment for a mixer, dosing unit and process control unit is in the range of
EUR 0.05 - 1 million. The investment for the size reduction of core-breaks is EUR 0.1 million,
although this is only applied if the size reduction cannot be done in an available mechanical or
pneumatic regeneration unit. Operational costs (on a yearly basis) can be estimated at 5 10 %
of the investment cost.

Driving force for implementation
Many foundry operators state the reduction in costs as the sole reason for starting various sand
reclamation programmes. Regulations encourage a reduction in the amount of dumped material,
through an increase in disposal costs.

Example plants
Primary reclamation is used by virtually all green sand foundries, though the degree of
sophistication of the reclamation plant varies widely; from a simple manual operation to one
fully automated with computer-controlled equipment.

Reference literature
[72, ETSU, 1995], [73, ETSU, 1995], [108, FEAF, 1999], [110, Vito, 2001], [128, IHOBE,
1998], [140, EU Thematic Network Foundry Wastes, 2001], [143, Inasmet and CTIF, 2002]

4.8.3 Simple mechanical regeneration of cold-setting sand

Description
Simple mechanical techniques are used for the regeneration of cold setting monosands (e.g.
furan sand) and uncured core sand. These techniques include the breaking of lumps, segregation
of the sand grains and cleaning by intergranular friction, with consequent dedusting and cooling
down to operational temperature. Various types of crushers and mills are used, e.g. impact
crusher, jaw crusher, ball mill.

Achieved environmental benefits
A reduction of the amount of sand for disposal and of the consumption of new primary sand.

Cross-media effects
The regeneration of sand requires additional energy and causes additional dust emissions and
residual dust for disposal.
Chapter 4
Smitheries and Foundries Industry 275
Operational data
For furan cold setting monosands regeneration ratios around 78 % are reported.

Applicability
The technique can be used for all cold-setting sand, excluding silicate sand. The regenerated
sand can be re-used in the same moulding cycle, with small additions of new sand to level-off
quality losses.

The technique can be used for uncured coresand with organic binders. The regenerated sand
may be re-used for core-making using the same binder type, after mixing with new sand. It may
also be used, within certain limits, for the renewal of moulding-sand.

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
Simple mechanical regeneration of furan sand is applied in many foundries around Europe, and
is specifically widespread in Germany and Finland.

Reference literature
[153, Umweltbundesamt, 2002], [202, TWG, 2002]

4.8.4 Cold mechanical regeneration using a grinding wheel unit

Description
This is a widely applied commercial grinding system (see Figure 4.26). The system was
originally developed to regenerate mixed bentonite-organic sands. In this system, a horizontally
rotating grinding wheel is used to remove the hard oolitic bentonite layer from the sand. The
grinding can also remove chemical binders from sand grains. Around the grinding wheel is a
slowly rotating paddle wheel, which continuously moves the sand onto the grinding wheel.
Above this, a dedusting unit extracts the dusts and the fines. To be fit for processing, the sand
must be dry. A pre-drying step, using a fluid bed or other dryer, is required to bring the moisture
content down to below 0.2 %.

Figure 4.26: Cold mechanical regeneration using grinding
[128, IHOBE, 1998]

Chapter 4
276 Smitheries and Foundries Industry
The off-gas stream of the regenerator is dedusted using a cyclone and a bag filter. The filter dust
contains residual active bentonite and coal dust. It may be recirculated into the core-making,
thereby allowing a reduction in the use of lustrous carbon. Furthermore, the moulds demonstrate
better technical quality (wet tensile strength, flowability), due to the remaining bentonite
content, which furthermore leads to a reduction in the amount of scrap moulds and a reduction
in finning on the castings.

Achieved environmental benefits
A reduction of the amount of sand for disposal and of the consumption of new primary sand.
A reduction in the consumption of lustrous carbon, in the case of green sand.
An increase in moulding sand properties, resulting in a reduction in the number of scrap moulds
and in the number of rejects in finished castings.

Cross-media effects
The regeneration of sand requires additional energy and causes additional dust emissions and
residual dust for disposal. However, the use of regenerated sand in the moulding process allows
a reduction in the amounts of lustrous carbon used.

Operational data
For clay-bonded sand, the maximum regeneration rate is 65 - 75 %. This corresponds to the
quartz grain content of the material. For a chemically-bonded sand, the amount of recyclable
quartz sand is 90 95 %.

The total sand circuit flow and the need for new sand depends on the amount of cores (and core
sand) used. Therefore, data on the sand circuit are very process-specific. Figure 4.27 gives data
for a plant in the Netherlands.

Core shop Foundry
Scrap core
breaker
Sand cleaner
New sand
70 kg
SiO
2
424 kg
Liquid iron
1000 kg
Losses
28 kg
396 kg
Losses
42 kg 354 kg
57 kg
Internal
48 kg
External
9 kg
Figure 4.27: Operational data on the sand circuit of a Dutch green sand foundry
[140, EU Thematic Network Foundry Wastes, 2001]

In general, waste green sand consists of about 80 % quartz sand grains and about 20 % fines
(bentonite, coal dust, etc,). Out of 100 % waste sand, about 70 % is transferred back to the core-
making as reclaimed sand. The degree of efficiency with regard to quartz is about 88 %.

The filter dust contains approximately 40 % active bentonite, which has a loss of ignition (LOI)
of 17 % and contains 43 % fines. Re-use of the dust in the sand plant results in a reduced
requirement for lustrous carbon by up to 30 %. Results of the improved sand quality at the
example foundry show the amount of scrap moulds were practically halved.
Chapter 4
Smitheries and Foundries Industry 277
The cyclone dust (20 % of regenerated green sand weight) contains 25 30 % bentonite and
high levels of organics, but its chemical composition falls outside the required specification to
allow its re-use as a secondary construction material. In the Netherlands, a specific application
in cover layers on disposal sites is allowed. In other regions, this fraction will need disposal of.
However, legislation limiting the organic content of material for disposal, may limit the disposal
options.

The grinding of the sand causes accelerated wearing of the sand grains. This and the
recirculation of the filter dust cause shifts in the grain size distribution of the sand. The overall
sand composition therefore needs good control and follow-up.

Applicability
Cold mechanical regeneration is mainly applied for removing bentonite layers from green sand
and for removing chemical binders in no-bake systems. The grinding technique is the most
widely applied cold mechanical treatment. Vibration- and impact systems are also used for
chemically-bonded sands but produce sand of only low or average quality. These techniques are
more appropriate for the conventional sand loop (primary regeneration). The secondary
regeneration of green sand has limited implementation.

Economics
The investment cost for a 50 tonne/day unit, including drying, cooling, regeneration unit,
cyclone and bag filter is around EUR 1135000. Data from Germany, state an investment cost of
EUR 600000 for a 1.5 tonne/h unit.

The consumption levels are as follows (based on yearly average figures):

Electricity: 39 kWh/tonne
Natural gas: 3 Nm/tonne (depending on the moist input of the sand)
Compressed air: 36 Nm/tonne
Wear parts: 1.18 EUR/tonne
Manning level: 0 %
Maintenance: routine.

Cost curves for the mechanical regeneration of organic bounded sand are given in Figure 4.28.
Fixed costs vary according to the total capacity and investment. Variable costs can be up to
around EUR 11/tonne, although the actual costs depend on the specific installation and local
conditions. The total costs (fixed plus variables) range from EUR 12 - 40/tonne of regenerated
sand, depending on the size and type of equipment.

0
5
10
15
20
25
30
35
40
0 5000 10000 15000 20000 25000 30000 35000 40000
Regeneration capacity (tonne/yr)
Investment cost EUR 270000 Investment cost EUR 540000
Depreciation: 8 years
Discount rate: 8 %
F
i
x
e
d
c
o
s
t
s
(
E
U
R
/
t
o
n
n
e
)
Figure 4.28: Fixed costs in euros per tonne of regenerated sand, for the mechanical regeneration of
sand with cold-setting binders
[82, IfG - Institut fr Gieereitechnik, 1996]
Chapter 4
278 Smitheries and Foundries Industry
Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
- De Globe, Weert, the Netherlands, 60 tonnes per day plant using two lines, in operation
since 1995
- Eisenwerk Brhl, Brhl; D (1.5 tonne/h), 1999
- Giesserei Fritz Winter GmbH & Co KG, Stadtallendorf, D (9.5 tonne/h)
- Mercedez-Benz AG, Mannheim, D (9 tonne/h).

Reference literature
[82, IfG - Institut fr Gieereitechnik, 1996], [110, Vito, 2001], [128, IHOBE, 1998], [133, De
Globe B.V., 1999], [140, EU Thematic Network Foundry Wastes, 2001], [151, Gemco, 1999]

4.8.5 Cold mechanical regeneration using an impact drum

Description
This mechanical regeneration technique is based on intergranular grinding of the sand and gives
the best results for chemically bonded monosand. The sand is fed into a drum with a spinning
internal axis, equipped with small blades. Sand grains are impacted against the drum wall and
against each other. This impact produces a mechanical abrasive cleaning action. Fines are
removed with the exhaust gas. The impact drum operates in a batch wise regime. The
installation of two units allows continuous operation.

When applied on a mixed bentonite-organic sand, the regeneration is preceded by a magnetic
separator, to remove green sand. Due to the presence of unreacted bentonite, the green sand
shows a very weak magnetism, which allows a magnetic separation to be carried out. The
regeneration system allows the introduction of a limited amount (15 %) of uncured core sand
(core breaks from production). The combination of magnetic separation and impact drum
cleaning allows an optimised chemically bonded sand regeneration from a mixed sand flow,
with re-use of the regenerated sand in core-making.

Achieved environmental benefits
Internal re-use of core sand, therefore limiting the amount of material for disposal and the need
for raw materials.

Cross-media effects
The regeneration of sand requires additional energy, and causes additional dust emissions and
residual dust for disposal.

Operational data
The regeneration drum has a batch wise regime, with a 20 minute treatment time for each
1.5 tonne sand charge. The energy consumption of the full installation (including magnetic
separation, exhaust treatment, sand transport) is 55 kWh/tonne of treated sand, of which 35 %
may be attributed to sand transport and feeding. The installation uses compressed air at
48 Nm/tonne sand.

The regenerated sand is of the following quality:
- average grain size: 0.30 0.33 mm
- share of fines: 0.4 1 %
- pH: 8.7
- loss on ignition: 0.25 0.5 %.

Cores made with 100 % regenerated sand have acceptable properties. In practice, 10 70 % of
regenerated sand is used for new cores, the actual amount depending on the core type.

Chapter 4
Smitheries and Foundries Industry 279
Dust generation in the installation is 10 % of the regenerator input. This dust is collected using a
cyclone (90 %) and a bag filter (10 %).

Applicability
The combined separation-regeneration technique can be applied for the treatment of mixed
green sand and chemically bonded sand. Regeneration is performed on the chemically-bonded
sand flow and may include uncured core sand. The technique may be applied in new and
existing installations.

Economics
Investment costs for a 3 tonne/h unit with 1 impact drum, magnetic separation, exhaust gas
dedusting and transport and a feeding system are EUR 1.3 million. Operational costs for this
unit are EUR 10/tonne. These include energy consumption, wear and tear of parts, maintenance
and disposal of residues. This represents a net benefit of EUR 37/tonne, compared to the costs
of new sand purchase and used sand disposal (both incl. transport).

For a 380 tonne/day unit in France, an operational cost of EUR 15/tonne was quoted. For this
plant this represents a net benefit of EUR 18/tonne, compared to the costs of new sand purchase
and used sand disposal (both incl. transport).

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
This technique is used in the following plants:
- PSA, Sept-Fons (F): 6 magnetic separators and 2 grinders, working alternatively
- GF, Leipzig (D): 2 magnetic separators and 1 grinder
- Dktas Turkey: 5 magnetic separators and 2 grinders.

Reference literature
[185, Spitz, 2002], [122, Kirst, 1999], [153, Umweltbundesamt, 2002]

4.8.6 Cold regeneration using a pneumatic system

Description
In a pneumatic system, binders are removed from the sand grains using abrasion and impact.
The kinetic energy is provided by a compressed air stream. This results in a simultaneous
dedusting. The advantage of this type of system is that the direction and velocity of the sand can
be controlled. Due to the low energetic efficiency of air compression, the energy consumption is
higher compared to the purely mechanical treatment. The principle of the reactor is depicted in
Figure 4.29.

Chapter 4
280 Smitheries and Foundries Industry
Figure 4.29: Cold mechanical regeneration using a pneumatic system
[122, Kirst, 1999]

The reactor uses a fluidised bed with a central raiser tube. The sand is blown into the tube using
compressed air and impacts on either a rubber deflector plate or a conical target plate. From
there, the sand falls down again and the cycle starts over again. The cleaning happens through
intergranular abrasion, abrasion against the impact plate and through pure impact. The shape
of the plate determines the main cleaning mechanism: impact (bell-shape) or abrasion (conus).
The generated dust is removed with the airflow over a bag filter. Treatment is performed as a
batch system or through a series of coupled pneumatic units. In this case, the cleaned sand is
sucked out of the reactor through a deflector below the impact plate.

Achieved environmental benefits
Reduction of the amount of sand for disposal and of the consumption of new primary sand.

Cross-media effects
The regeneration of sand requires additional energy and causes additional dust emissions and
residual dust for disposal.
The general evaluation of internal regeneration versus external re-use and the corresponding
cross-media effects are discussed separately below.

Operational data
A waste sand mix of green sand with chemically-bound core sand having the following
characteristics: 8 12 % fines content, 3 5 % LOI, <2 % humidity; produces a regenerate with
the following characteristics:

- regenerate yield, based on SiO
2
content of waste sand: 70 80 %
- ultra fines content (<0.063 mm): max. 2 %
- fines content: <2 %
- loss on ignition: <0.5 %
- average grain size: unaltered.
Chapter 4
Smitheries and Foundries Industry 281
The technique uses a modular installation build-up, using 0.8 1.2 t/h basic units. The supplier
mentions an electric energy consumption of 15 20 kWh/t (excluding dedusting equipment).
German operational data give an electric energy consumption of 41 kWh/t for a 8 t/h unit and
62 kWh/t for a 0.75 t/h unit. Reported sand regeneration efficiencies based on sand input range
from 65 % to 85 %. Spanish data give a total energy consumption of 120 kWh/t of regenerated
sand for a 1.2 t/h unit.

Applicability
The pneumatic system can be used for the regeneration of organic mixed and monosands and
mixed sand containing bentonite. It also finds application as a pre- or post-treatment in
combined mechanical-thermal-mechanical treatment Here, the main use is the removal of
residual dust from the sand grains and cooling. The regenerated sand from simple mechanical
regeneration may be used in mould making (using 100 % regenerated sand) or mixed with new
sand in core-making (using 40 - 60 % regenerated sand).

Furthermore, the technique may be applied for the regeneration of core sand of the CO
2
-water
glass type from aluminium foundries. This is discussed separately below.

Economics
Investment costs for a single 0.8 1.2 tonne/h unit are EUR 330000. Operational costs are
estimated at EUR 22/tonne, which gives a total regeneration cost of EUR 36.5/tonne of sand.

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
This type of equipment is built by various suppliers: e.g. KGT Jet Reclaimer, Knkel-Wagner
GmbH Turbo Dry, Kernfest Webac AB. The technique is used in several plants in Western
Europe and China.

Reference literature
[32, CAEF, 1997], [82, IfG - Institut fr Gieereitechnik, 1996], [108, FEAF, 1999], [110, Vito,
2001], [122, Kirst, 1999], [128, IHOBE, 1998]

4.8.7 Thermal regeneration

Description
Thermal regeneration uses heat to combust binders and contaminants. All thermal processes
need an initial mechanical step in order to bring the sand to the correct grain size and to screen
out any metallic contaminants. This pretreatment may also involve a (partial) abrasion of
bentonite and dust removal. The heating of the sand is usually achieved by means of a fluidised
bed furnace, operating at temperatures mostly between 700 to 800 C. Rotary kilns or multiple-
hearth furnaces are also used. Heat may be provided by gas combustion, electric heaters or by
short wave infra-red emitters. The throughput of these systems ranges from 250 kg/h to more
than 5 t/h.

The emerging gases are combusted in order to eliminate carbon monoxide and any VOCs that
may be present. This may be done in the freeboard of the furnace, if it is large enough, by
adding complementary air or by gas-fired afterburners. If the temperature of the combustion
gases is not high enough or if the time during which the gases are at a high temperature is not
sufficient, a separate afterburning device is added. In all cases, emissions can be considered as
insignificant.

Chapter 4
282 Smitheries and Foundries Industry
The exhaust gases are filtered, mostly by means of fabric filters. Therefore cooling is needed,
which may be done using water injection, heat exchange/recovery or by mixing with air
naturally entering through openings/slits etc. The cooling may incorporate a preheating of the
fluidisation air.

Achieved environmental benefits
Reduction of the amount of sand for disposal and a reduction in the consumption of new
primary sand.

Cross-media effects
Thermal regeneration requires fuel and generates emissions of dust and combustion related
compounds (NO
x
, CO; and in the case of oil:SO
2
).
The high energy consumption and complexity of the installation are balanced by a low wearing
of the sand grains and the recovery of the dust in a thermally inert form.

Thermal regeneration of furan bonded sand with paratoluenesulphonic acid as a hardener,
requires flue-gas cleaning: i.e. a, post-combustion of CO and the adsorption of SO
2
.
Operational data
The sand is usually heated to temperatures of 700 850 C. Theoretically, this requires
200 kWh/t. In practice, energy uses of 150 350 kWh/t are reported, depending on the energy
recuperation and the calorific value of the sand.

Operational data from 3 thermal regeneration plants, treating various types of sand and of
various capacities are given in Table 4.64.

Chapter 4
Smitheries and Foundries Industry 283
Property Units Plant
Reference plants Foundry K Foundry L Foundry M
Type of sand Cold-box core units Green sand
moulding, Cold-box
cores
Handforming in
furan bonded sand
Regeneration unit
supplier
CEC/VAW Richards Siempelkamp
Technique Multiple hearth
furnace (500 C)
with sieving and
classification
Fluidised bed
furnace (780 C),
mechanical size
reduction and
dedusting
Fluidising
combustion chamber
with magnetic pre-
separation,
simultaneous
processing of sand
and dust from
mechanical furan
sand regeneration
Sand throughput t/h 12 16 1.0 2.0 used sand, 0.5
dust
Capacity t/yr 28600 3840 5000
Regeneration ratio,
based on used sand
throughput
% 95 95 (99
(1)
) 95
Application of
regenerated sand
Core-making Core-making Mould and core-
making
Quality criteria for
regenerated sand
Dust <1 %
AFS 60 70
LOI <0.1 %
Identical to new
sand F33
LOI <0.5 %
pH 6.0 6.5
Mould strength upon
addition of 1 %
binder and 0.3 %
harder: 350 N/cm
after 24 h
Consumption of
electrical energy
kWh/t 119 65
Consumption of
natural gas
m/t 24.5 38
Residue production t/year 660 Dust: 10
Bentonite sand: 1700
200
Final destination of
residues
Mining cavities Cement production Construction
industry/disposal site
Waste gas flow Nm/h 7000 16509 6560
Waste gas
temperature
C 75 60 35
Type of filter Post combustion and
fabric filter
Fabric filter Lime injection +
fabric filter
Emission levels
(2)

- dust
- C
total

- CO
- O
2
- PCDD/F

mg/Nm

%
ng TEQ/Nm

0.1
2.2
2.3
14
0.006

4.9
5.7
1.6
n.d
n.d

0.4
10
0
20.95
n.d.
(1) The higher recuperation rate (99%) is reached if only in-house core sand is regenerated. A rate of 95 % applies for
mixed internal and external core sand
(2) Emission levels are calculated average values from continuous monitoring data over a period of several hours, during
sampling for dust or PCDD/F
Table 4.64: Operational data of 3 German reference foundries applying thermal sand regeneration
[153, Umweltbundesamt, 2002]

The performance of the regeneration technique and the quality requirements for the regenerated
sand are evaluated on a location-specific basis. Specific data may be found in the references:
[186, Stephan, 1996], [187, Stephan, 1997], [188, Stephan, 2000].

Applicability
Thermal systems are normally used for chemically bonded sand systems and mixed sand
systems, providing the share of chemically bonded sands (cores) is high enough. Nevertheless, a
distinction can be made:
Chapter 4
284 Smitheries and Foundries Industry
- some binders may leave inorganic salt residues at the grain surface. When the temperature
increases above the melting point, these spread out over the sand surface and may cause the
sintering of the sand grains upon cooling. Known examples are resol resin and silicate resin-
bonded sands
- the thermal recycling of furan resins causes the emission of SO
2
if sulphonic acids are used
as a hardening agent. The temperature of the exhaust system must be sufficiently high
enough to prevent sulphuric acid condensation. If the emission is high, a wet flue-gas
cleaning may be needed. Phosphor (from the phosphoric acid used as a hardener) does not
evaporate but remains on the surface in salt form. This may lead to P accumulation upon
recycling. If the P-concentration rises above 0.5 0.7 %, a metallurgical reaction may
occur, causing the rejection of the casting.

The degree of implementation of thermal regeneration is low compared to mechanical
regeneration. In Germany, only 10 installations were reported to apply this technique in 1999
(compared to 200 mechanical units).

The Siempelkamp thermal regeneration unit in Krefeld has the ability to treat foundry dust
along with the waste sand. Tests have been run on other mineral waste types as well.

Economics
Investment costs vary according to the size and type of equipment and are given in Figure 4.30.
The data relate to thermal systems with a mechanical pre- or post-treatment step, as well as for
intensive mechanical regeneration (e.g. grinding units). Small scale installations have a capacity
of 0.75 t/h or 1500 t/yr. A medium size foundry will be between 2500 and 5000 t/yr. For a small
scale unit, the capital cost (with 8 years amortisation and 8 % interest rate) would be EUR 55/t.
These costs do not include expenses for energy, personnel nor for the disposal of residues.

0
10
20
30
40
50
60
70
80
90
100
110
120
0 5000 10000 15000 20000 25000 30000
Regeneration capacity (tonne/yr)
Investment cost: EUR 500000 Investment cost EUR 3000000
Investment cost EUR 6000000
F
i
x
e
d
c
o
s
t
s
(
E
U
R
/
t
o
n
n
e
)
Depreciation: 8 years
Discount rate: 8 %
Figure 4.30: Fixed costs in euros per tonne of regenerated sand for thermal regeneration and high
level mechanical regeneration of mixed sands
[82, IfG - Institut fr Gieereitechnik, 1996]

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Chapter 4
Smitheries and Foundries Industry 285
Example plants
- Fluidised bed, Metaalgieterij Giesen, the Netherlands
- Fluidised bed centralised regeneration plant, Bielefeld, Germany
- VAW Alucast GmbH, Dilingen, Germany, system CEC/VAW
- Werner Rietschle GmbH, Schopfheim, Germany, system Richards
- Siempelkamp Giesserei gmbH & Co, Krefeld, Germany, system Siempelkamp
- Sofogir, Ronchamp (F) system Regetherm 500
- Duranton Sicfond, Vnissieux (F), system Richards
- FOAM, Leumann (I), system Eco-rec
- Fonderie Bra, Montluon (F).

Reference literature
[9, Schneider, 1993], [32, CAEF, 1997], [80, ERM Lahmeyer International GmbH, 1999], [128,
IHOBE, 1998], [138, Metaalgieterij Giesen B.V., 1996], [186, Stephan, 1996], [187, Stephan,
1997], [188, Stephan, 2000]

4.8.8 Combined regeneration (mechanicalthermal-mechanical) for
mixed organic-bentonite sands

Description
In mixed organic-bentonite sands, cured bentonite and organic binders are present on the sand
grains. The dust is composed of active and cured bentonite, coal dust (only for iron foundries),
quartz fines and organic binder residues. Mixed sands occur mainly in iron foundries and
represent some 75 % of the total used sand production. The regeneration can be performed using
mechanical, pneumatic, thermal or combined systems.

The sand is pretreated (sieving, magnetic separation) and dried, in order to reduce the water
content to <1 %. After this, the sand is mechanically or pneumatically cleaned in order to
remove part of the binder. In the thermal step, organic constituents are burned, and inorganic
constituents are transferred to the dust or burned onto the grains. In a final mechanical
treatment, these layers are removed mechanically or pneumatically and discarded as dust. A
typical layout for a system using pneumatic treatment and fluidised bed thermal treatment is
depicted in Figure 4.31.

Figure 4.31: Mechanical-thermal-mechanical sand regeneration unit
[108, FEAF, 1999]
Chapter 4
286 Smitheries and Foundries Industry
Achieved environmental benefits
Reduction of the amount of sand for disposal and of the consumption of new primary sand.

Cross-media effects
The regeneration of sand requires additional energy and causes additional dust emissions and
residual dust for disposal.
The general evaluation of internal regeneration versus external re-use, and the corresponding
cross-media effects, are discussed separately below.

Operational data
Operational data for a mechanical-thermal-mechanical treatment are given in Table 4.65.

Parameter Units Foundry
N
Foundry
O
Foundry
P
Sand flow (tonne/h) 1 2 3
Maximum humidity of the sand input (%) 1 1 1
Total installed electric power (kW) 150 215 400
Natural gas
(Nm
3
/h) 31 62 155
Thermal treatment
- average airflow
(fluidisation + combustion)
(Nm
3
/h) 700 1200 3000
Treatment temperature (C) 800 800 800
Average treatment time (min) 20 20 20
Flue-gas flow (Nm
3
/h) 200 300 500
Cooling
- type
- water flow in the circuit

(m
3
/h)

Fluidised bed with water
circulation
15 30 75
Sand temperature at exit (C) 30 40
Total energy consumption:
- electric
- gas

100
260

90
260

70
260
Compressed airflow (Nm
3
/h) 150 210 330
Efficiency of the installation 70 80 %
Regenerate sand quality:
- granulometry
- loss on ignition

Unchanged
0.1 %
Re-use of the regenerated sand:
- mould making
- core-making

100 %
70 80 %
Table 4.65: Operational data of a mechanical-thermal-mechanical treatment system (pneumatic -
fluid bed - pneumatic)
[108, FEAF, 1999]

German data for a 85000 tonne/yr (13 15 tonne/h) installation provide the following off-gas
composition:

- SO
2
: 118 mg/Nm
- NO
x
: 150 mg/Nm
- total C: 10 mg/Nm
- CO: 30 mg/Nm
Chapter 4
Smitheries and Foundries Industry 287
Applicability
This regeneration technique is used for mixed sands containing bentonite. The economical and
technical success of the regeneration depends on the selection of the sand to be regenerated. It is
not suitable to regenerate the whole sand volume. Sand which is not thermally damaged can be
re-used directly for new sand preparation. Its regeneration is not suitable, because active binders
and additives would be removed. A selection and separation has to be made during shake-out,
before homogenisation. The technique cannot be applied for core sands which disrupt bentonite
characteristics (binders of acid nature) or which change green sand characteristics (e.g. water
glass sand).

The regenerated sand can be used for core-making in the original process, for cores with low or
medium geometrical demands. The applicability for core-making is related to the initial amount
of chemically-bonded sand. The applicability in other binder systems must be tested in each
case. Furthermore, these sands may be applied without restrictions for the replacement of losses
in green sand moulding cycles. Restrictions may apply in processes that use water glass or
methyl formate bonded sands.

In recent years, various types of installations have been developed for the treatment of mixed
sands, but they have not yet found broad implementation in the sector. Some only work on a
pilot scale. Others have been developed for a specific foundry and produce a regenerated sand
suitable for re-use only in the same process, and results cannot be transposed to other foundries.

Economics
Costs for a 3-step installation (mechanical-thermal-mechanical) using pneumatic cleaning as the
mechanical step (3 tubes in 1 chamber, KGT type Jet Reclaimer) with a capacity of 2.5 tonne/h
are as follows: operational costs (consumption, personnel, maintenance) - EUR 21/t, investment
costs (8 year amortisation) - EUR 30/t, thus yielding a total regeneration cost of EUR 51/tonne.

The consumption levels are as follows:

- Electricity: 50 kWh/t
- Natural gas: 18 Nm/t (depending on the moisture content of the sand)
- Compressed air: 60 Nm/t
- Wear parts: 5 EUR/t

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
Halberg Guss GmbH, Saarbrucken (D), thermal-mechanical system. This plant treats 13 15 t/h
of mixed 30 % organic, 70 % inorganic bonded sand. The organic bounded sand is a 50/50
mixture of Croning and cold-box sand. The regeneration yield on the basis of the amount of
used sand treated is 78 %. The regenerated sand is applied for core-making (100 % Croning;
70 100 % cold-box).

The mechanical primary regeneration + separation of chromite (from cores) and quartz sand
(both chemically-bonded) + thermal/mechanical treatment of quartz sand was demonstrated on a
pilot scale in Germany (1993). Separation of both sand types was needed because sintering and
eutectics formation occurred in the mixed sand. The plant later closed down due to other
economic reasons.

Reference literature
[11, Schachtner and Mller-Spth, 1993], [37, Winterhalter, et al., 1992], [108, FEAF, 1999],
[122, Kirst, 1999], [128, IHOBE, 1998], [153, Umweltbundesamt, 2002], [225, TWG, 2003]

Chapter 4
288 Smitheries and Foundries Industry
4.8.9 Wet sand regeneration

Description
After metal removal, the sand is mixed with water to produce a sludge for easily separating the
binder and to allow subsequent wet screening (1.6 mm). The removal of grain-bound binder
residues is performed in the wet regeneration unit, through intensive interparticle rubbing of the
sand grains. The binders are released into the wash-water. The washed sand is dried to a final
moisture content of max. 0.3 %, dry screened (at 1.2 mm), and then cooled. This may be
followed by an additional magnetic iron extraction and a final dust collection. The binder
residue is separated from the suspension and treated to allow its safe, disposal.

The main advantage of the process over mechanical and thermal processes is that it allows real
time process monitoring. Through pH control, continuous follow-up of the process is possible,
allowing corrective real-time action if needed and producing a continuous quality regenerated
sand. The technique allows removal of the binder layer through wet mechanical action,
combined with chemical action when needed, without wearing the sand grains.

The process mainly aims to lower both the oolitisation degree of used sand (LOI max. 1 %) and
the acid content.

Achieved environmental benefits
Reduction of the amount of sand for disposal and of the consumption of new primary sand.

Cross-media effects
The wet process generates a sludge for disposal and a waste water stream. A serious problem of
this waste water flow is the bad sedimentation of bentonite, and its difficult removal out of
waste water. Waste water from water glass sands regeneration shows the presence of alkaline
ions. These ions are difficult to separate from the waste water.

Applicability
The wet regeneration system can only be applied to green sand and silicate- or CO
2
-bonded
sands. Regeneration from these types of processes allows full re-use of the regenerated sand in
both moulds and cores. Tests on regenerated green sand showed the possibility of producing
good quality cold-box cores with an acceptable binder quantity (1.8 % in total).

Economics
Within the Italian foundry market, a centralised wet regeneration plant can offer regenerated
sand at a price lower than that of new sand.

Driving force for implementation
This system is reported to be able to cope with changes in used sand quality better than
mechanical or thermal processes. This makes it suitable when considering a centralised sand
regeneration plant.

Example plants
Safond centralised green sand wet reclamation plant, which since 1981 has been treating
230000 tonnes/yr of used sand.

One Polish foundry (Odlewnia Zeliwa Srem s.a., Srem (PL)) operates a wet regeneration of
furan bonded sand. The operation is governed by site-specific factors and the technique cannot
be considered for transposing to other sites.

Reference literature
[37, Winterhalter, et al., 1992], [140, EU Thematic Network Foundry Wastes, 2001], [225,
TWG, 2003]

Chapter 4
Smitheries and Foundries Industry 289
4.8.10 Regeneration of water glass sand using pneumatic systems

Description
Water glass sand has been traditionally difficult to regenerate. The use of a pneumatic system
has allowed the setting up of regeneration plants operating of a 60 % regeneration ratio. The
system works along the same principle as mentioned above (see Section 4.8.6). For this type of
binder however, the sand needs to be heated to 220 C before the regeneration, in order to make
the silicate layer brittle. The sand should have a humidity below 0.3 % before regeneration. The
regenerated sand may be re-used in the same system. In order for the ester to work properly, the
regenerated sand must be cooled to below 20 C, before feeding back into the moulding cycle.
Water glass sand regeneration systems comprise the following process steps: breaking
drying/heating (pneumatical) reclamation cooling - filtration.

This technique has a lower performance than the thermal regeneration of organically bound
sands. The following limitations occur:
- lower immediate compressive strength
- shorter applicability times of the core sand mixtures
- stability loss upon storage of cores
- loss of retained compressive strength, especially with the use of binders with high
collapsibility.

In order to compensate for these problems, changes to the processing scheme or the binder
mixes are needed.

Achieved environmental benefits
The application of water glass sand as a binder has a low environmental impact, compared to
the application of organic binders. This technique allows the (partial) regeneration of the water
glass bonded sand and thus reduces the need for used sand disposal and primary raw material
use.

Cross-media effects
In order to heat up the sand, natural gas is combusted. The corresponding CO
2
-emission is
estimated to be 18 kg/tonne of used sand. NO
x
emissions will also occur.

Operational data
A German example plant operates a pneumatical unit in a step-wise cycle. The sand is first dried
by the introduction of heated air (5 min/220 C). After this, the pneumatic cleaning is started by
injecting of shots of compressed air (70 min.). This is followed by a final dedusting phase,
during which only fluidising air is introduced (2 min). There is no need for further cooling,
since the sand cools down to a workable temperature.

The yield of regenerated sand for one cycle operation is reported to be 85 % of the initial weight
(on the basis of SiO
2
). In order to produce stable cores, and taking into account the further
reduction of sand quality upon a second regeneration cycle, a maximum regeneration ratio of
62 % may be achieved (leaving 38 % new sand addition).

Specific consumption levels are as follows (for a 1500 tonne/yr; 0.5 tonne/h unit):
- natural gas consumption: 104.4 kWh/t used sand
- electricity consumption: 74.5 kWh/t used sand.

A Polish example plant (in a cast steel foundry) consists of the followig units: vibratory crusher
drier impact plate regenerator cascade dedustor electromagnetic separator. The technical
details of the installation are given in Table 4.66. The regenerated sand is re-used in the
preparation of silicate-bonded moulding-sand. The silicate-sand is used as backing sand in
mould making, with alkaline sand used as contact sand. The total sand balance (on a yearly
basis) shows a use of 45 - 47 % regenerated sand.

Chapter 4
290 Smitheries and Foundries Industry
Characteristic Value
Capacity 10 tonne/h
Recovery ratio 90 %
Content of regenerated sand in the moulding-sand 50 60 %
Working power 76 kW
De-dusting system efficiency (2 bag filters) 99.4 %
Installation area for regeneration site 220 m
Table 4.66: Operational data of a silicate sand regeneration unit
[200, Metalodlew s.a., 2002]

A Spanish example plant (cast steel foundry) reported a re-use of 80 - 88 % (with a yearly
average figure of 12.5 % new sand addition) of regenerated silicate-ester sand in a manual
moulding line, used for the casting of big pieces. The regeneration system consists of: vibratory
crusher heating in an indirectly heated reactor (quemador with gas burner in a side chamber)
cooling in a water-cooled heat-exchange tower impact cleaning in a rotary mixing chamber.

Italian example plants also use special equipment for heating sand to temperatures of 140 C to
150 C, to enable the removal of defects and the embritlement of the water glass coating. This is
followed by mechanical treatment in a rotary wiper, dedusting and finally cooling. A
schematical drawing of the rotary regeneration unit is given in Figure 4.32.

Figure 4.32: Rotary mechanical regeneration unit
[225, TWG, 2003]

The equipment has an energy consumption of 35 kW/tonne of regenerated sand. The operational
loss of sand is about 5 %. For moulding and core making mixtures, over 90 % of regenerated
sand can be used. The achievable regeneration ratio depends on the intensity and time of
regeneration and the operational sand loss. Sand losses occur during each part of the whole
cycle of mixing, preparation, transportation, mould production and regeneration.

Applicability
The applicability of regeneration for water glass sands depends on the catalyst used. With
classic polyacetate glycerol esters, regeneration is no longer possible after a number of cycles.
With carbonaceous esters, regeneration is possible and even easier than for silicate-CO
2
sand. If
the sand cannot be cooled below 20 C before returning to the mixer, esters with a slow reaction
need to be used. These cannot be regenerated. This situation may occur during summer in hot
climates.

Chapter 4
Smitheries and Foundries Industry 291
The re-use of the regenerated water glass sand for the preparation of organically bound moulds
or cores is problematic. The elevated content of electrolytes (binders) prohibits their use in other
core binder systems. The highly basic water glass residues have a negative effect on bonding
and core stability, in both neutral as well as alkaline binder systems. Applications in green sand
cycles and acidic core binder systems have not yet been successfully implemented. This may be
due to a reduced interest in the development of this application.

Because of the low quality of secondary regenerated sand, it is necessary during de-coring and
sand collection to identify the amount of regenerated sand and its origin (e.g. number of cycles),
and to then discard any unusable sand.

Due to the high costs and relatively low regeneration ratio, depreciation of the installation in a
reasonable time can only be guaranteed for plants with a capacity >2500 tonnes/yr.

Economics
The processing cost of this technique for the German example plant is around EUR 60/tonne.
This high cost was the reason one operator closed his plant down and went back to the disposal
of the water glass sand, which he could do for the lower cost of EUR 30/tonne.
The reported operational cost by the Spanish example plant is EUR 10/tonne.

Driving force for implementation
To enable the regeneration of water glass sand, which has a good environmental performance in
certain applications.

Example plants
- KGT Jet Reclaimer centralised regeneration plant for 4 foundries, constructed 1998 at Fa.
Brer, Schwetzingen (D) but stopped operation in 2001
- Metalodlew s.a., Krakow (PL)
- Lur Sue s.l., Lora Del Rio (E)
- Fonderia Arno Metallurgia, S. Vittore Olona (I)
- Faser Spa, Rogeno (I)
- Talleres De Amurrio S.A. Amurrio (E)
- Daros Piston Rings AB, Partilla (S).

Reference literature
[128, IHOBE, 1998], [152, Notzon and Heil, 1998], [154, Baum, 2002], [200, Metalodlew s.a.,
2002], [210, Martnez de Morentin Ronda, 2002], [225, TWG, 2003], [228, Galante, et al.,
1997]

4.8.11 Internal re-use of uncured core sand

Description
Core production generates sand residues in the form of broken cores, cores with small mistakes
and excess sand from the core-making machines. The excess sand can be hardened in a specific
unit. Subsequently the various unused core sand flows are fed to a breaking unit. The resulting
sand may be mixed with new sand for the production of new cores.

Achieved environmental benefits
Internal re-circulation of 5 - 10 % of the core sand, which otherwise would be disposed of.

Cross-media effects
No cross-media effects occur.

Applicability
This technique applies for polyurethane (cold-box) and furan resin-bonded sands. Other binders
do not allow this technique.
Chapter 4
292 Smitheries and Foundries Industry
The implementation of the technique necessitates changes in the process control. The optimum
ratio of re-circulated and new sand will need to be determined. Additionally changes in binder
addition or composition may be needed.

Economics
The technique requires an investment cost of EUR 250000 500000 for the combined treatment
and breaking unit. Operational costs are about EUR 12/t. These costs are balanced by a
reduction in disposal costs and purchasing costs for new sand.
The technique is only viable for core intensive production.

Driving force for implementation
Optimisation of the use of primary materials and reduction of the amount of waste for disposal.

Example plants
The technique was reported by several large-scale foundries.

Reference literature
[82, IfG - Institut fr Gieereitechnik, 1996], [110, Vito, 2001]

4.8.12 Re-use of dusts from the green sand circuit in mould making

Description
Dust is collected through the exhaust filtration from the shake-out installation and from the
dosing and handling stations for dry green sand. The collected dust contains active binder
compounds and may be recycled into the green sand circuit.

Achieved environmental benefits
Reduction in the use of binders (bentonite), and additives (carbon), through internal re-
circulation.

Cross-media effects
No cross-media effects exist.

Operational data
An iron automotive foundry, using an automated moulding line with a production of 8000 t/yr
castings produces 480 t/yr of dust. This sand is collected, cooled and recycled into the sand
mixer. The collected dust contains 23 % active bentonite and 10 % carbon. 50 % of the dust can
be re-circulated without any risk of loss of quality due to fine dust.

Applicability
The technique may be applied in new and existing green sand installations.

Economics
The technique requires an investment of EUR 25000 for storage and transport equipment. The
amortisation period is 8 years, with a discount rate of 8 %, for a 240 tonne/yr installation,
resulting in a capital cost of EUR 17/tonne of treated sand. There are no additional operational
costs since the operation of the overall installation does not change.

Driving force for implementation
To optimise the use of primary materials and to reduce the amount of waste for disposal.

Example plants
Iron automotive foundry in Germany.

Reference literature
[82, IfG - Institut fr Gieereitechnik, 1996]
Chapter 4
Smitheries and Foundries Industry 293
4.8.13 External re-use of used sand and the undersize from the sand
circuit and regeneration processes

Description
Used sand and undersize sand from the sand circuit or sand regeneration may find some external
applications, with the main areas of application being:
- the construction industry (road construction, highway construction)
- the building materials industry (cement, bricks, limestone manufacture)
- in the filling of mining cavities
- in landfill construction (roads on landfills, permanent covers).

The limits of these applications are given by either technical criteria for construction materials
and/or environmental criteria for the given application. Environmental criteria are generally
based on leaching properties and the content of organic compounds. These differ between
various European regions. Used sands generally show a low metal leaching potential. An
exceedence of the limit values may occur for materials with a high content of organic binder or
with specific additives, such as lustrous carbon.

Due to its high quartz content and appropriate granulometry, used sand may be applied as a
virgin sand substitute in road construction. Application in the production of building materials
(concrete, bricks, tiles, glass wool, ) are technically feasible but require a higher level of
composition control and logistics. Industrial scale trials have been successfully applied in:
- road bases
- filling material
- drainage material
- concrete elements
- cement production (depending on the silicon need of the process. In Germany this
application is of major importance)
- filling of mining cavities (due to its good supporting capacities. Used sand is used in this
application in Germany)
- final cover on landfills (mixed with water glass and other wastes, examples in the
Netherlands)
- construction material for the reinforcement of dikes
- vitrification of hazardous waste.

A survey table which gives more detailed information for the different sand types is given in
Section 4.9.

Other fields of application are the fabrication of bricks, secondary copper smelting and zinc
recuperation.

It should be stressed that discussions on the (type of) external utilisation fall beyond the scope
of the current reference document. However, in general, it can be stated that the sand usually
needs no pretreatment and is transported from the foundry after collection and intermediate
storage. Usually in order to guarantee a continuous quality of the material, a separate collection
and storage system will be developed. Furthermore, the quality of the material needs to be
controlled through regular analyses.

Achieved environmental benefits
In external applications, the used sand substitutes, and therefore saves, virgin materials. In
general, the external application does not require any treatment of the sand and therefore does
not generate any energy expense for the foundry.

Cross-media effects
Residues of organic coatings are possible sources of aromatic hydrocarbons.

Chapter 4
294 Smitheries and Foundries Industry
Applicability
This technique is applicable for new and existing installations, as long as they can find a buyer
for the waste sand.

The use of surplus foundry sand does not reduce the wear resistance of asphalt concrete. The
use of surplus foundry sand together with the dust from the same foundry process is possible in
Portland cement concrete production. The use of surplus foundry sand together with fly ash and
steel slag is possible in mineral wool production. The use of surplus foundry sand in the
composting process of biowastes does not negatively affect the environmental or technical
features of the final product, i.e. the topsoil. The use of surplus green sand in mineral liners is
technically and environmentally possible in the top layers of the landfill.

Economics
Costs for external utilisation depend on the local market and the transport and storage costs
needed.

Investment costs for separate collection and storage are minimal. Often these can be realised
through organisational measures. Operational costs for analyses and administration may be up
to EUR 5000/yr. On the benefit side, there is a reduction in the costs for disposal (up to
EUR 125/tonne).

Driving force for implementation
Legislation using high disposal fees to reduce the amount of residues for disposal.

Example plants
Multiple examples have been reported throughout Europe, e.g. by Finland, the Netherlands,
Belgium, Germany, UK.

Reference literature
[82, IfG - Institut fr Gieereitechnik, 1996], [110, Vito, 2001], [128, IHOBE, 1998], [169,
Orkas, 2001], [171, The Castings Development Centre, 1999]

4.9 Dust and solid residues: treatment and re-use

4.9.1 Introduction

A good segregation of materials is essential to facilitate opportunities for recovery, recycling
and re-use. Waste should always be recycled or recovered unless a satisfactory justification has
been accepted by the regulator that recovery is technically and economically impossible.

In general, the waste streams comprise:
- raw material dusts collected in bag or cartridge filters
- slag from desulphurisation
- slag/dross from melting
- melting dust and fume collected in a filter plant
- casting dust and fume collected in a filter plant
- fettling dusts collected in an abatement plant
- chips and turnings from fettling
- used abrasives from shot blasting
- scrubber liquors and sludges and output from the effluent treatment plant
- refractory waste from launders and ladles
- sand
- chemical and oil containers
- general inert industrial waste.

Chapter 4
Smitheries and Foundries Industry 295
Waste should be recovered wherever practicable. Other than sand, the most significant wastes
are:
- slag from the melting and metal treatment processes
- dust collected from abatement plants
- collected sludge
- refractory waste.
[160, UK Environment Agency, 2002]

4.9.2 (Pretreatment for) external re-use of solid residues

Description
In order to allow an external re-use of solid residues, the material may need treatment. Table
4.67 gives the required processing for the various (non-sand) solid residues and possible
limitations for their re-use.

For cupola furnace slag, the physical form, and therefore to some extent its options for re-use,
depend on the type of de-slagging used. Dry slagging, i.e. pouring of the slag in pots for cooling
and solidification, produces a crystalline non-porous mineral. Concerning technological aspects,
this slag is comparable to blast furnace slag. Wet slagging generates a granular slag by cooling
the slag with a water jet.

Residue type Treatment Limitations
Air-cooled cupola slag Crushing - generation of glassy dust
- handling requirements due to glassy
nature
Water quenched cupola slag None - generation of glassy dust
- handling requirements due to glassy
nature
Induction melting slag Crushing - generation of glassy dust
- handling requirements due to glassy
nature
- little data available on leachates
EAF slag Crushing - generation of glassy dust
- handling requirements due to glassy
nature
- little data available on leachates
Desulphurisation slag Extraction of metal and
other coarse particles
- handling requirements, CaC
2
needs
careful handling to avoid injury
- run-off
- may be a special waste
Dusts and sludges Sludge pressing, drying
and granulation needed for
most applications
- health and safety requirements for some
dusts
- problems with transport of dusty
materials
- high leaching potential due to nature of
product and large surface area
Table 4.67: Necessary treatment and possible limitations for external re-use of solid residues
[171, The Castings Development Centre, 1999]

Achieved environmental benefits
In external applications, the re-used material will serve as a secondary construction material and
can substitute for virgin materials. The recycling results in a reduction in the amount of material
for disposal.

Cross-media effects
Pretreatment requires energy and may generate dust emissions.

Chapter 4
296 Smitheries and Foundries Industry
Applicability
The technique is applicable for new and existing installations, as long as there is a suitable a
local market for the residue.

A survey of external re-use applications for various solid wastes is given in Table 4.68.

SAND SLAGS OTHER
SUMMARY OF RE-USE
APPLICATIONS

Construction type uses
G
r
e
e
n
s
a
n
d
A
l
k
a
l
i
n
e
p
h
e
n
o
l
i
c
P
h
e
n
o
l
i
c
u
r
e
t
h
a
n
e
F
u
r
a
n
R
e
s
i
n
s
h
e
l
l
S
o
d
i
u
m
s
i
l
i
c
a
t
e
A
i
r
-
c
o
o
l
e
d
c
u
p
o
l
a
W
a
t
e
r
q
u
e
n
c
h
e
d
c
u
p
o
l
a
I
n
d
u
c
t
i
o
n
E
l
e
c
t
r
i
c
a
r
c
f
u
r
n
a
c
e
D
e
s
u
l
p
h
u
r
i
s
a
t
i
o
n
I
n
v
e
s
t
m
e
n
t
c
a
s
t
i
n
g
s
h
e
l
l
D
u
s
t
s
a
n
d
s
l
u
d
g
e
s
Asphalt x x + + + O x
Ballast + +
Block making + x + + x + x
Brick manufacture x x + + + + +
Cement x x + x x
Coarse aggregate substitute x x x x
Concrete x + + +
Fine aggregate substitute x x + + + +
Foamed concrete, etc. x x +
Insulating/mineral/glass
wools
+ + + + + + + + +
Lightweight aggregate
production
+
Mortar production +
Road base construction x + + x x x x x
Roofing felt x
Re-use in another foundry
As new sand addition to
greensand
x
Soil type uses
Artificial topsoils x + + x + +
Decorative ground cover + +
Ground cover in riding
stables
+
Fertiliser filler x
Landfill - capping x x + +
Landfill - daily cover x x + x x x
Landfill - liner + +
Soil modifier/improver + + + + + + + + +
Miscellaneous
Abrasives/blasting media x + x +
Absorbent media +
Blast furnace slag cement
manufacture
+ +
Chemical/industrial
applications
x
Slaked lime replacement + +
Smelting fluxes + + + +
Waste vitrification + + + + +
x Proven re-use application, with successful projects running in the UK
+ Re-use application that has been proven in theory but no re-use project is currently running in the UK
O Unsuitable for re-use in untreated from
Table 4.68: Summary of external re-use applications for solid foundry residues (status 1999)
[171, The Castings Development Centre, 1999], [202, TWG, 2002]
Chapter 4
Smitheries and Foundries Industry 297
Economics
The costs associated with re-use are mainly the processing and transport costs, but these can
possibly be reduced by taking into account the income generated from selling the material.
However, the latter in general may be low or close to zero, whereas typically mechanical
treatment costs are in the range of EUR 8/tonne.

The further the materials have to be transported, the greater the increase in costs. In order to
assess the economic viability of the re-use option, the applicable disposal costs should be taken
into account. These differ between various regions and depend on the residue type.

Driving force for implementation
Legislation stimulating the recycling of mineral wastes, partly by setting high disposal fees to
reduce the amount of residues for disposal.

Example plants
Multiple examples of sand and slag re-use have been reported throughout Europe, e.g. in
Finland, the Netherlands, Belgium, Germany, UK

Reference literature
[171, The Castings Development Centre, 1999]

4.9.3 Minimisation of slag forming

Description
The production of slag can be minimised using in-process measures, such as:
- using clean scrap
- using a lower metal temperature
- preventing overshoots (temporary high temperatures)
- preventing long standing times of molten metal in the melting furnace
- making adequate use of fluxes
- making adequate use/choice of refractory lining
- applying water cooling of the furnace wall to avoid usage of refractory lining.

Achieved environmental benefits
Minimisation of residue production and reduction of emissions to air.

Cross-media effects
No cross-media effects occur from these minimisation measures. Concerning the melting of
clean scrap, the considerations raised in Section 4.1.4 should be taken into account.

Applicability
This technique applies to all new and existing installations. The applicability of the use of clean
scrap has been fully discussed in Section 4.1.4.

Economics
This technique does not involve any investement, since it concerns operational measures.

Driving force for implementation
High disposal costs for residues.

Example plants
This technique is part of existing operational procedures in the majority of European foundries.

Reference literature
[103, Vereniging van Nederlandse Gemeenten, 1998]

Chapter 4
298 Smitheries and Foundries Industry
4.9.4 Cupola furnace

4.9.4.1 Collection and recycling of coke breeze

Description
The handling, transport and charging of coke results in the production of coke breeze. Specific
measures may be adopted for the collection and recycling of this material e.g. collection systems
below conveyor belts or, at charging points.

The collected material may be recycled into the process by injection into the cupola or in its
utilisation for recarburisation.

Achieved environmental benefits
Minimisation of the production of residues.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to new and existing installations.

Driving force for implementation
High disposal fees for solid residues.

Example plants
The technique is reported for several foundries in Europe.

4.9.4.2 Recirculation of filter dust into cupola furnace

Description
Cupola filter dust is re-injected into the cupola furnace. The aim of this technique is to achieve
an accumulation of zinc in the dust, up to a level that allows reprocessing with Zn recovery
(>18 %). Zn recovery is performed using the Waelz process.

Recirculation of the dust can be done through re-injection at the tuyres or by charging dust
pellets through the charging door. Both techniques are applied on an industrial scale. A mass
balance analysis can be used to show that the zinc is sublimated and reappears in the
recuperated dust. After a number of cycles the dust is rich enough in Zn to allow recovery.
Recovery is technically possible above an 18 % Zn level.

There are two limits to recycling dust:
- the zinc essentially appears in the form of Zn
2
SO
4
, a compound containing 64 % Zn. This
therefore sets the maximum level of enrichment
- the dust is also enriched in alkali, which causes flame extinction in the combustion
chamber. This effect may be solved by injecting water vapour into the combustion air.

Injection at the tuyres can be based on several principles. There are fifteen (status May 2001)
European foundries recycling dust by injection at the tuyres. Eleven use the positive pressure
technique, and four use the venturi suction technique. The advantage of the venturi technique is
that it consumes less injection air and so cools the melting zone less. With both techniques, the
dust becomes sticky after successive cycles. In the case of the venturi technique, the dust must
therefore be mixed with petcoke to allow its transport in the pipes.

The recirculation of dust through agglomeration and charging is much less common than tuyre
injection. This is because the process is much more difficult to automate. Furthermore, good
control and knowledge of the pelletisation process is needed to give the pellets the necessary
Chapter 4
Smitheries and Foundries Industry 299
consistency. Ferrosilicon can be added to produce pellets with good consistency. In practical
tests a mix of 2/3 FeSi with 1/3 of dust was needed to allow good operation. On the other hand
the pellet technique can be used to treat large quantities. For cupolas with larger diameters,
tuyre injection of the dust into the centre of the fire is difficult, this therefore favours the use of
pellets.

The properties of both techniques are compared in Table 4.69.

Re-use of the fines collected in the
purification of the cupola off-gas
Injection through
nozzles
Introduction
as pellets
Observations
Influence of the waste
Limitation in the introduced amount
of fines
7 8 kg/tonne metal

Problem of
extinction of flame
Reduction in the generated amount of
fines
40 % 50 % Approx.
(1)

Increase in the content of zinc in the
fines
30 % 20 % Approx.
(1)

Increase in slag production ? ?
Influence on the operation of the cupolas
Amount of coke necessary to
counteract the temperature loss
0.5 0.6 % 0.2 0.3 % Approx.
(1)

Variation in the analysis of off-gases No No n.a
Influence in melted metal
Increase in the metal content (Zn, Pb)
in the iron
Limited No (in the order of
thousandths)
Influence in the losses to fire 10 30 % No Approx.
(1)(2)

Effectiveness to add other products
(C, FeSi) with the same method
Yes No n.a
Costs
Reduction in the disposal cost of the
fines
60 % 50 % Approx.
(1)

Operations
Possibility of automating the process Yes No n.a
Additional Problem No Yes Crash-resistance of
the pellets
Implementation of technology
Existing Facilities Yes Yes
New facilities Yes Yes
Preferably cupolas
of large size
(1)
Approximation, real value depending on the initial zinc oxide content the fines, of the degree of
necessary enrichment for its re-use, and on the characteristics of the cupola
(2)
Function of the amount of steel in the load of the cupola
n.a: not applicable
Table 4.69: Properties of injection techniques for recirculation of cupola furnace dust
[108, FEAF, 1999]

Achieved environmental benefits
The main benefit of the technique is a decrease in the net amount of dust sent out by the
foundry, for disposal or recovery. The foundry typically sends out a smaller amount of dust, but
with a higher load of metals. This allows the recovery of Zn from the cupola dusts. The
technique results in an enrichment of the Zn in the cupola dust, while for Pb a larger share ends
up with the iron. If the Zn is effectively recuperated from the dust, the technique decreases the
heavy metal load of the final dust for disposal, as well as its amount. On the other hand, the
amount of slag produced increases.

Cross-media effects
Both techniques consume energy, i.e. coke, to melt the product. An increase in coke use of 0.15
to 0.25 kg cokes/kg dust at a thermal efficiency of 40 % of the furnace is reported.

Chapter 4
300 Smitheries and Foundries Industry
In addition, the injection of cold dust in the hot melting zone will lower the flame temperature,
which accordingly will have to be restored by an injection of oxygen. The total amount of slag
produced will also increase.

The loading of sticky dust to several of the internal parts of the system will cause medium-term
loss of efficiency and possibly operational problems. These have not been quantified.

Operational data
An analysis of the input and output flows of French cupolas resulted in the mass balance shown
in Figure 4.33 and Figure 4.34. The contribution of each flow to the Zn-in/output is given
together with the zinc-level (as mass and %). The percentages given are average values over the
indicated measuring period.

Figure 4.33: Zinc distribution for tuyre injection, upon injection during 20 days
[156, Godinot, 2001]

Chapter 4
Smitheries and Foundries Industry 301
Figure 4.34: Zinc distribution for recirculation through the charge, upon charging during 11 days
[156, Godinot, 2001]

The analysis of the input and output flows do not allow a full closure of the zinc balance. This
indicates that a considerable part of the zinc (20 % for tuyre injection, 10 % for pellet
charging) remains and accumulates in the system. Upon recirculation, the dust gets sticky and
deposits in the combustion chamber, heat-exchanger and filters. The recirculation will therefore
require increased cleaning efforts and will result in a material for disposal at this cleaning stage.

The recirculation of cupola dust (yearly iron production: 30000 tonnes) in a Dutch iron foundy
reduced the net amount of dust by 156 t on a yearly basis, without any significant effect on the
environmental and construction qualities of the slag.

Applicability
Recovery is technically possible above an 18 % Zn level. Cupolas with a Zn-rich charge have
Zn-levels in the dust >20 %, without recirculation. A minimal content of 40 % Zinc is needed
for cost neutral recovery.

It is not possible to re-inject the full dust production of a furnace. Injection of more than
8 kg/tonne metal results in an extinguishing of the flame.

New and existing cupola furnaces with continuous operation can be equipped with these dust
recirculation installations.

Economics
Recovery is technically possible above an 18 % Zn level. Operators report profitability of the
recirculation and Zn recovery occuring for a Zn-level >40 %.

Chapter 4
302 Smitheries and Foundries Industry
From the practical data from two foundries in France, additional operational costs were
calculated. Tuyre injection showed an additional operational cost of EUR 0.35/tonne cast iron;
pellet injection EUR 0.90/tonne cast iron. This calculation does not take into account any extra
costs for slag disposal or maintenance works.

In the case of Venturi injection, the following data apply. In a German reference plant over a
5 year period an average of 7.5 kg zinc dust/tonne melted iron could be delivered to the
processing industry. Disposal costs were thereby reduced by EUR 1.97/tonne liquid iron.
Furthermore, 20 % of the cokes could be replaced by petcokes. This allowed a reduction in the
fuel cost of 15 %, which corresponded to a cost reduction of EUR 2.56/tonne liquid iron. This
calculation does not take into account any extra costs for slag disposal or maintenance works.
Driving force for implementation
High disposal costs for furnace dusts.

Example plants
Venturi (NPT):
Pont-a-Mousson, Brebach (F)
Positive Pressure (VELCO):
John Deere, Mannheim (D)

Reference literature
[108, FEAF, 1999], [110, Vito, 2001], [140, EU Thematic Network Foundry Wastes, 2001],
[156, Godinot, 2001], [181, Ruffin and Godinot, 1998], [198, Georg Fischer, ], [202, TWG,
2002]

4.9.5 EAF

4.9.5.1 Recycling of filter dust and sludge (from EAF melting)

Description
Dry collected furnace dust can, under certain circumstances, be recycled in the furnace. The
dust will preferably be pretreated, e.g. by pelletising or briquetting, as this will reduce the
amount of dust that is just blown through the furnace. The idea is to recover the metallic part of
the dust and to melt the inorganic part to slag. In general, the dust is added at the beginning of
each melting cycle.

As discussed for cupola furnaces, and for the EAF, in some cases where significant quantities of
Zn containing scrap are used, recycling can lead to a Zn enrichment of the furnace dust up to a
level where it can be used in the zinc industry for the recovery of Zn if economically viable (up
to 30 - 40 %).

Achieved environmental benefits
The main benefit of the technique is a decrease in the net amount of dust sent out by the
foundry, for disposal or recovery. The produced dust contains a higher load of metals. This
allows the recovery of Zn from the EAF dusts. The technique results in an enrichment of the Zn
in the cupola dust, while for Pb a larger share ends up in the iron. If the Zn is effectively
recuperated from the dust, the technique decreases the heavy metal load of the final dust for
disposal, as well as its amount. On the other hand, the amount of slag produced, increases.

Cross-media effects
Dust recycling possibly reduces furnace efficiency and rises the consumption of electrical
energy (approx. 20 - 30 kWh/tonne). The technique results in an increased production of slag.

Operational data
Operational data from EAF steelmaking show a plant which recycles 75 % EAF dust from an
original yield of 20 22 kg/t, and so only has to take care of about 50 % of the dust, with an
average zinc content of 35 %.
Chapter 4
Smitheries and Foundries Industry 303
Applicability
The technique generally applies for dusts with a high metallic load. The actual applicability
though depends on many factors, that may be dissimilar for different plants.

Economics
The technique does not involve any additional investment.

Driving force for implementation
High disposal costs for furnace dusts.

Example plants
This technique is performed in several European foundries.

Reference literature
[32, CAEF, 1997], [211, European IPPC Bureau, 2000]

4.9.5.2 Recycling of Al slags and residues

Description
The possibilities of recycling depend on the product and the generated wastes. Swarfs are
mostly recovered for internal remelting while other scrap (containing too much other metal) will
usually be sold to secondary aluminium production.

Achieved environmental benefits
Minimisation of the production of residues.

Cross-media effects
No cross-media effects apply.

Applicability
This technique applies to new and existing installations.

Driving force for implementation
High disposal fees for solid residues.

Example plants
The technique is commonly applied in aluminium foundries.

Reference literature
[225, TWG, 2003]

4.10 Noise reduction

Description
The foundry process contains various point sources of noise. These include:
- scrap handling
- furnace charging
- burners
- HPDC automates
- shake-out
- grit-blasting
- core (and mould) shooting
- finishing
- all motors and hydraulic systems
- transport (tipping, loading, etc.).
Chapter 4
304 Smitheries and Foundries Industry
The overall reduction of noise levels involves developing a noise reduction plan. Here each of
the sources needs to be checked and evaluated. Alternative techniques with lower noise levels
may be applied and/or point sources may be enclosed. Examples for alternative techniques have
been given in Section 4.2.4.2 (oxyburner in RF) and 4.2.2.2 (foamy slag for EAF). Some
examples of the enclosure of sources have been given for furnaces (Section 4.5.3.1) and shake-
out (4.5.9.3).

General measures include:
- using noise reducing flaps on all outside doors and closing all doors (and keeping them shut
as much as possible), especially during the night
- actively blowing air into the foundry hall. This causes a small increase in the indoor
pressure and keeps the noise inside
- enclosing fans, insulating of ventilation pipes and using of dampers
- minimising the number of transport activities during the night.

Full enclosure of the foundry building may be considered as well. This will also require setting-
up a climate control system, to limit the temperature inside the building.

Achieved environmental benefits
Reduction of noise pollution.

Cross-media effects
Enclosing (parts of) the foundry building may necessitate the installation of a climate control
system, the operation of which requires energy.

Operational data
A Belgian aluminium foundry is currently developing a noise reduction plan. The aim is for the
general noise level to be reduced from 50 dBA to 40 dBA. This involves the study of 170
sources. Specific measures are taken to reduce the overall noise level between 22.00 h and
6.00 h. Additionally air is blown into the foundry hall, creating a small pressure difference to
keep the noise inside. The total inside air volume is refreshed 36 times each hour.

Another Belgian foundry is running a project to reduce noise emissions by fully closing the
foundry building. This requires the setting-up of a climate control system, in order to achieve a
maximum inside-outside temperature difference of 8 C. The system is currently going through
subsequent revisions, so tests are still ongoing.

Applicability
This technique applies to all new and existing installations. The level to which measures should
be taken depends on the foundry location.

Driving force for implementation
Regulation on noise levels from industrial installations.

Example plants
- MGG, Hoboken (B): noise reduction plan
- Hayes-Lemmertz, Hoboken (B): total building enclosure.

Reference literature
[225, TWG, 2003]

Chapter 4
Smitheries and Foundries Industry 305
4.11 Decommissioning

Description
Some techniques to consider are:
- considering later de-commissioning at the design stage, thereby minimising risks and
excessive costs during later de-commissioning
- for existing installations, where potential problems have been identified, putting in place a
programme of improvements. These improvements designs need to ensure that:
underground tanks and pipework are avoided where possible (unless protected by
secondary containment or a suitable monitoring programme)
there is provision for the draining and cleaning-out of vessels and pipework prior to
dismantling
lagoons and landfills are designed with a view to their eventual clean-up or surrender
insulation is used which can be readily dismantled without dust or hazards arising
any materials used are recyclable (although bearing in mind they still meet operational
or other environmental objectives)

- developing and maintaining a site closure plan, to demonstrate that, in its current state, the
installation can be decommissioned to avoid any pollution risk and to return the site of
operation to a satisfactory state. The plan should be kept updated as material changes occur.
However, even at an early stage, the closure plan can include:
either the removal or the flushing out of pipelines and vessels where appropriate and
their complete emptying for any potentially harmful contents
plans of all underground pipes and vessels
the method and resource necessary for the clearing of lagoons
the method of ensuring that any on-site landfills can meet the equivalent of surrender
conditions
the removal of asbestos or other potentially harmful materials unless agreed that it is
reasonable to leave such liabilities to future owners
methods of dismantling buildings and other structures, for the protection of surface and
groundwater at construction and demolition-sites
testing of the soil to ascertain the degree of any pollution caused by the activities and
the need for any remediation to return the site to a satisfactory state as defined by the
initial site report

Achieved environmental benefits
Prevents environmental issues during de-commissioning.

Applicability
Techniques mentioned here are applicable throughout the installations operational lifetime,
during the design and building stage of the site and activities and immediately after the site
closure.

Reference literature
[236, UK Environment Agency, 2001]

4.12 Environmental management tools

Description
The best environmental performance is usually achieved by the installation of the best
technology and its operation in the most effective and efficient manner. This is recognised by
the IPPC Directive definition of techniques as both the technology used and the way in which
the installation is designed, built, maintained, operated and decommissioned.
Chapter 4
306 Smitheries and Foundries Industry
For IPPC installations an Environmental Management System (EMS) is a tool that operators can
use to address these design, construction, maintenance, operation and decommissioning issues
in a systematic, demonstrable way. An EMS includes the organisational structure,
responsibilities, practices, procedures, processes and resources for developing, implementing,
maintaining, reviewing and monitoring the environmental policy. Environmental Management
Systems are most effective and efficient where they form an inherent part of the overall
management and operation of an installation.

Within the European Union, many organisations have decided on a voluntary basis to
implement environmental management systems based on EN ISO 14001:1996 or the EU Eco-
management and audit scheme EMAS. EMAS includes the management system requirements of
EN ISO 14001, but places additional emphasis on legal compliance, environmental performance
and employee involvement; it also requires external verification of the management system and
validation of a public environmental statement (in EN ISO 14001 self-declaration is an
alternative to external verification). There are also many organisations that have decided to put
in place non-standardised EMSs.

While both standardised systems (EN ISO 14001:1996 and EMAS) and non-standardised
(customised) systems in principle take the organisation as the entity, this document takes a
more narrow approach, not including all activities of the organisation e.g. with regard to their
products and services, due to the fact that the regulated entity under the IPPC Directive is the
installation (as defined in Article 2).

An environmental management system (EMS) for an IPPC installation can contain the
following components:

(a) definition of an environmental policy
(b) planning and establishing objectives and targets
(c) implementation and operation of procedures
(d) checking and corrective action
(e) management review
(f) preparation of a regular environmental statement
(g) validation by certification body or external EMS verifier
(h) design considerations for end-of-life plant decommissioning
(i) development of cleaner technologies
(j) benchmarking.

These features are explained in somewhat greater detail below. For detailed information on
components (a) to (g), which are all included in EMAS, the reader is referred to the reference
literature indicated below.

(a) Definition of an environmental policy

Top management are responsible for defining an environmental policy for an installation
and ensuring that it:

is appropriate to the nature, scale and environmental impacts of the activities
includes a commitment to pollution prevention and control
includes a commitment to comply with all relevant applicable environmental legislation
and regulations, and with other requirements to which the organisation subscribes
provides the framework for setting and reviewing environmental objectives and targets
is documented and communicated to all employees
is available to the public and all interested parties.

Chapter 4
Smitheries and Foundries Industry 307
(b) Planning, i.e.:

procedures to identify the environmental aspects of the installation, in order to
determine those activities which have or can have significant impacts on the
environment, and to keep this information up-to-date
procedures to identify and have access to legal and other requirements to which the
organisation subscribes and that are applicable to the environmental aspects of its
activities
establishing and reviewing documented environmental objectives and targets, taking
into consideration the legal and other requirements and the views of interested parties
establishing and regularly updating an environmental management programme,
including designation of responsibility for achieving objectives and targets at each
relevant function and level as well as the means and timeframe by which they are to be
achieved.

(c) Implementation and operation of procedures

It is important to have systems in place to ensure that procedures are known, understood and
complied with, therefore effective environmental management includes:

(i) Structure and responsibility
defining, documenting and communicating roles, responsibilities and authorities,
which includes appointing one specific management representative
providing resources essential to the implementation and control of the environmental
management system, including human resources and specialised skills, technology
and financial resources.

(ii) Training, awareness and competence

identifying training needs to ensure that all personnel whose work may significantly
affect the environmental impacts of the activity have received appropriate training.

(iii) Communication
establishing and maintaining procedures for internal communication between the
various levels and functions of the installation, as well as procedures that foster a
dialogue with external interested parties and procedures for receiving, documenting
and, where reasonable, responding to relevant communication from external
interested parties.

(iv) Employee involvement
involving employees in the process aimed at achieving a high level of environmental
performance by applying appropriate forms of participation such as the suggestion-
book system or project-based group works or environmental committees.

(v) Documentation
establishing and maintaining up-to-date information, in paper or electronic form, to
describe the core elements of the management system and their interaction and to
provide direction to related documentation.

(vi) Efficient process control
adequate control of processes under all modes of operation, i.e. preparation, start-up,
routine operation, shutdown and abnormal conditions
identifying the key performance indicators and methods for measuring and
controlling these parameters (e.g. flow, pressure, temperature, composition and
quantity)
documenting and analysing abnormal operating conditions to identify the root causes
and then addressing these to ensure that events do not recur (this can be facilitated
by a no-blame culture where the identification of causes is more important than
apportioning blame to individuals).
Chapter 4
308 Smitheries and Foundries Industry
(vii) Maintenance programme
establishing a structured programme for maintenance based on technical descriptions
of the equipment, norms etc. as well as any equipment failures and consequences
supporting the maintenance programme by appropriate record keeping systems and
diagnostic testing
clearly allocating responsibility for the planning and execution of maintenance.

(viii) Emergency preparedness and response
establishing and maintaining procedures to identify the potential for and response to
accidents and emergency situations, and for preventing and mitigating the
environmental impacts that may be associated with them.

(d) Checking and corrective action, i.e:

(i) Monitoring and measurement
establishing and maintaining documented procedures to monitor and measure, on a
regular basis, the key characteristics of operations and activities that can have a
significant impact on the environment, including the recording of information for
tracking performance, relevant operational controls and conformance with the
installation's environmental objectives and targets (see also the Reference document
on Monitoring of Emissions)
establishing and maintaining a documented procedure for periodically evaluating
compliance with relevant environmental legislation and regulations.

(ii) Corrective and preventive action
establishing and maintaining procedures for defining responsibility and authority for
handling and investigating non-conformance with permit conditions, other legal
requirements as well as objectives and targets, taking action to mitigate any impacts
caused and for initiating and completing corrective and preventive action that are
appropriate to the magnitude of the problem and commensurate with the
environmental impact encountered.

(iii) Records
establishing and maintaining procedures for the identification, maintenance and
disposition of legible, identifiable and traceable environmental records, including
training records and the results of audits and reviews.

(iv) Audit
establishing and maintaining (a) programme(s) and procedures for periodic
environmental management system audits that include discussions with personnel,
inspection of operating conditions and equipment and reviewing of records and
documentation and that results in a written report, to be carried out impartially and
objectively by employees (internal audits) or external parties (external audits),
covering the audit scope, frequency and methodologies, as well as the
responsibilities and requirements for conducting audits and reporting results, in order
to determine whether or not the environmental management system conforms to
planned arrangements and has been properly implemented and maintained
completing the audit or audit cycle, as appropriate, at intervals of no longer than
three years, depending on the nature, scale and complexity of the activities, the
significance of associated environmental impacts, the importance and urgency of the
problems detected by previous audits and the history of environmental problems
more complex activities with a more significant environmental impact are audited
more frequently
having appropriate mechanisms in place to ensure that the audit results are followed
up.

Chapter 4
Smitheries and Foundries Industry 309
(v) Periodic evaluation of legal compliance
reviewing compliance with the applicable environmental legislation and the
conditions of the environmental permit(s) held by the installation
documentation of the evaluation.

(e) Management review, i.e.:

reviewing, by top management, at intervals that it determines, the environmental
management system, to ensure its continuing suitability, adequacy and effectiveness
ensuring that the necessary information is collected to allow management to carry out this
evaluation
documentation of the review.

(f) Preparation of a regular environmental statement

- preparing an environmental statement that pays particular attention to the results
achieved by the installation against its environmental objectives and targets. It is
regularly produced from once a year to less frequently depending on the significance
of emissions, waste generation etc. It considers the information needs of relevant
interested parties and it is publicly available (e.g. in electronic publications, libraries
etc.).

When producing a statement, the operator may use relevant existing environmental
performance indicators, making sure that the indicators chosen:

i. give an accurate appraisal of the installations performance
ii. are understandable and unambiguous
iii. allow for year on year comparison to assess the development of the
environmental performance of the installation
iv. allow for comparison with sector, national or regional benchmarks as
appropriate
v. allow for comparison with regulatory requirements as appropriate.

(g) Validation by certification body or external EMS verifier

- having the management system, audit procedure and environmental statement examined
and validated by an accredited certification body or an external EMS verifier can, if
carried out properly, enhance the credibility of the system.

(h) Design considerations for end-of-life plant decommissioning

- giving consideration to the environmental impact from the eventual decommissioning of
the unit at the stage of designing a new plant, as forethought makes decommissioning
easier, cleaner and cheaper
- decommissioning poses environmental risks for the contamination of land (and
groundwater) and generates large quantities of solid waste. Preventive techniques are
process-specific but general considerations may include:
i. avoiding underground structures
ii. incorporating features that facilitate dismantling
iii. choosing surface finishes that are easily decontaminated
iv. using an equipment configuration that minimises trapped chemicals and
facilitates drain-down or washing
v. designing flexible, self-contained units that enable phased closure
vi. using biodegradable and recyclable materials where possible.

Chapter 4
310 Smitheries and Foundries Industry
(i) Development of cleaner technologies

- environmental protection should be an inherent feature of any process design activities
carried out by the operator, since techniques incorporated at the earliest possible design
stage are both more effective and cheaper. Giving consideration to the development of
cleaner technologies can for instance occur through R&D activities or studies. As an
alternative to internal activities, arrangements can be made to keep abreast with and
where appropriate commission work by other operators or research institutes active in
the relevant field.

(j) Benchmarking, i.e.:

carrying out systematic and regular comparisons with sector, national or regional
benchmarks, including for energy efficiency and energy conservation activities, choice
of input materials, emissions to air and discharges to water (using for example the
European Pollutant Emission Register, EPER), consumption of water and generation of
waste.

Standardised and non-standardised EMSs
An EMS can take the form of a standardised or non-standardised (customised) system.
Implementation and adherence to an internationally accepted standardised system such as
EN ISO 14001:1996 can give higher credibility to the EMS, especially when subject to a
properly performed external verification. EMAS provides additional credibility due to the
interaction with the public through the environmental statement and the mechanism to ensure
compliance with the applicable environmental legislation. However, non-standardised systems
can in principle be equally effective provided that they are properly designed and implemented.

Achieved environmental benefits
Implementation of and adherence to an EMS focuses the attention of the operator on the
environmental performance of the installation. In particular, the maintenance of and compliance
with clear operating procedures for both normal and abnormal situations and the associated lines
of responsibility should ensure that the installations permit conditions and other environmental
targets and objectives are met at all times.

Environmental management systems typically ensure the continuous improvement of the
environmental performance of the installation. The poorer the starting point is, the more
significant short-term improvements can be expected. If the installation already has a good
overall environmental performance, the system helps the operator to maintain the high
performance level.

Cross-media effects
Environmental management techniques are designed to address the overall environmental
impact, which is consistent with the integrated approach of the IPPC Directive.

Operational data
No specific information reported.

Applicability
The components described above can typically be applied to all IPPC installations. The scope
(e.g. level of detail) and nature of the EMS (e.g. standardised or non-standardised) will
generally be related to the nature, scale and complexity of the installation, and the range of
environmental impacts it may have.

Economics
It is difficult to accurately determine the costs and economic benefits of introducing and
maintaining a good EMS. A number of studies are presented below. However, these are just
examples and their results are not entirely coherent. They might not be representative for all
sectors across the EU and should thus be treated with caution.
Chapter 4
Smitheries and Foundries Industry 311
A Swedish study carried out in 1999 surveyed all 360 ISO-certified and EMAS-registered
companies in Sweden. With a response rate of 50%, it concluded among other things that:

- the expenses for introducing and operating EMS are high but not unreasonably so, save in
the case of very small companies. Expenses are expected to decrease in the future
- a higher degree of co-ordination and integration of EMS with other management systems is
seen as a possible way to decrease costs
- half of all the environmental objectives and targets give payback within one year through
cost savings and/or increased revenue
- the largest cost savings were made through decreased expenditure on energy, waste
treatment and raw materials
- most of the companies think that their position on the market has been strengthened through
the EMS. One-third of the companies report increasing revenue due to EMS.

In some Member States reduced supervision fees are charged if the installation has a
certification.

A number of studies
5
show that there is an inverse relationship between company size and the
cost of implementing an EMS. A similar inverse relationship exists for the payback period of
invested capital. Both elements imply a less favourable cost-benefit relationship for
implementing an EMS in SMEs compared to larger companies.

According to a Swiss study, the average cost for building and operating ISO 14001 can vary:

- for a company with between 1 and 49 employees: CHF 64000 (EUR 44000) for building the
EMS and CHF 16000 (EUR 11000) per year for operating it
- for an industrial site with more than 250 employees: CHF 367000 (EUR 252000) for
building the EMS and CHF 155000 (EUR 106000) per year for operating it.

These average figures do not necessarily represent the actual cost for a given industrial site
because this cost is also highly dependent on the number of significant items (pollutants, energy
consumption,...) and on the complexity of the problems to be studied.

A recent German study (Schaltegger, Stefan and Wagner, Marcus, Umweltmanagement in
deutschen Unternehmen - der aktuelle Stand der Praxis, February 2002, p. 106) shows the
following costs for EMAS for different branches. It can be noted that these figures are much
lower than those of the Swiss study quoted above. This is a confirmation of the difficulty to
determine the costs of an EMS.

Costs for building (EUR):
- Minimum: 18750
- Maximum: 75000
- Average: 50000

Costs for validation (EUR):
- Minimum: 5000
- Maximum: 12500
- Average: 6000
5
E.g. Dyllick and Hamschmidt (2000, 73) quoted in Klemisch H. and R. Holger, Umweltmanagementsysteme in kleinen und
mittleren Unternehmen Befunde bisheriger Umsetzung, KNI Papers 01/02, January 2002, p 15; Clausen J., M. Keil and M.
Jungwirth, The State of EMAS in the EU.Eco-Management as a Tool for Sustainable Development Literature Study,
Institute for Ecological Economy Research (Berlin) and Ecologic Institute for International and European Environmental
Policy (Berlin), 2002, p 15.
Chapter 4
312 Smitheries and Foundries Industry
A study by the German Institute of Entrepreneurs (Unternehmerinstitut/Arbeitsgemeinschaft
Selbstndiger Unternehmer UNI/ASU, 1997, Umweltmanagementbefragung - ko-Audit in der
mittelstndischen Praxis - Evaluierung und Anstze fr eine Effizienzsteigerung von
Umweltmanagementsystemen in der Praxis, Bonn.) gives information about the average savings
achieved for EMAS per year and the average payback time. For example, for implementation
costs of EUR 80000 they found average savings of EUR 50000 per year, corresponding to a
payback time of about one and a half years.

External costs relating to verification of the system can be estimated from guidance issued by
the International Accreditation Forum (http://www.iaf.nu).

Driving force for implementation
Environmental management systems can provide a number of advantages, for example:

- improved insight into the environmental aspects of the company
- improved basis for decision-making
- improved motivation of personnel
- additional opportunities for operational cost reduction and product quality improvement
- improved environmental performance
- improved company image
- reduced liability, insurance and non-compliance costs
- increased attractiveness for employees, customers and investors
- increased trust of regulators, which could lead to reduced regulatory oversight
- improved relationship with environmental groups.

Example plants
The features described under (a) to (e) above are elements of EN ISO 14001:1996 and the
European Community Eco-Management and Audit Scheme (EMAS), whereas the features (f)
and (g) are specific to EMAS. These two standardised systems are applied in a number of IPPC
installations. As an example, 357 organisations within the EU chemical and chemical products
industry (NACE code 24) were EMAS registered in July 2002, most of which operate IPPC
installations.

In the UK, the Environment Agency of England and Wales carried out a survey among IPC (the
precursor to IPPC) regulated installations in 2001. It showed that 32 % of respondents were
certified to ISO 14001 (corresponding to 21 % of all IPC installations) and 7 % were EMAS
registered. All cement works in the UK (around 20) are certified to ISO 14001 and the majority
are EMAS registered. In Ireland, where the establishment of an EMS (not necessarily of a
standardised nature) is required in IPC licenses, an estimated 100 out of approximately 500
licensed installations have established an EMS according to ISO 14001, with the other 400
installations having opted for a non-standardised EMS.

Reference literature
[78, ETSU, 1996]

(Regulation (EC) No 761/2001 of the European parliament and of the council allowing
voluntary participation by organisations in a Community eco-management and audit scheme
(EMAS), OJ L 114, 24/4/2001, http://europa.eu.int/comm/environment/emas/index_en.htm)
(EN ISO 14001:1996, http://www.iso.ch/iso/en/iso9000-14000/iso14000/iso14000index.html;
http://www.tc207.org)
Chapter 5
Smitheries and Foundries Industry 313
5 BEST AVAILABLE TECHNIQUES FOR FOUNDRIES

In understanding this chapter and its contents, the attention of the reader is drawn back to the
preface of this document and in particular the fifth section of the preface: How to understand
and use this document. The techniques and associated emission and/or consumption levels, or
ranges of levels, presented in this chapter have been assessed through an iterative process
involving the following steps:

- identification of the key environmental issues for the foundries sector
- examination of the techniques most relevant to address these key issues
- identification of the best environmental performance levels, on the basis of the available
data in the European Union and worldwide
- examination of the conditions under which these performance levels were achieved; such as
costs, cross-media effects, main driving forces involved in implementation of the techniques
- selection of the best available techniques (BAT) and the associated emission and/or
consumption levels for this sector in a general sense all according to Article 2(11) and
Annex IV of the Directive.

Expert judgement by the European IPPC Bureau and the relevant Technical Working Group
(TWG) has played a key role in each of these steps and in the way in which the information is
presented here.

On the basis of this assessment, techniques, and as far as possible emission and consumption
levels associated with the use of BAT, are presented in this chapter that are considered to be
appropriate to the sector as a whole and in many cases reflect current performance of some
installations within the sector. Where emission or consumption levels associated with best
available techniques are presented, this is to be understood as meaning that these levels
represent the environmental performance that could be anticipated as a result of the application,
in this sector, of the techniques described, bearing in mind the balance of costs and advantages
inherent within the definition of BAT. However, they are neither emission nor consumption
limit values and should not be understood as such. In some cases it may be technically possible
to achieve better emission or consumption levels but due to the costs involved or cross-media
considerations, they are not considered to be appropriate as BAT for the sector as a whole.
However, such levels may be considered to be justified in more specific cases where there are
special driving forces.

The emission and consumption levels associated with the use of BAT have to be seen together
with any specified reference conditions (e.g. averaging periods).

The concept of levels associated with BAT described above is to be distinguished from the
term achievable level used elsewhere in this document. Where a level is described as
achievable using a particular technique or combination of techniques, this should be
understood to mean that the level may be expected to be achieved over a substantial period of
time in a well maintained and operated installation or process using those techniques.
Where available, data concerning costs have been given together with the description of the
techniques presented in the previous chapter. These give a rough indication about the magnitude
of costs involved. However, the actual cost of applying a technique will depend strongly on the
specific situation regarding, for example, taxes, fees, and the technical characteristics of the
installation concerned. It is not possible to evaluate such site-specific factors fully in this
document. In the absence of data concerning costs, conclusions on economic viability of
techniques are drawn from observations on existing installations.

It is intended that the general BAT in this chapter are a reference point against which to judge
the current performance of an existing installation or to judge a proposal for a new installation.
In this way they will assist in the determination of appropriate "BAT-based" conditions for the
installation or in the establishment of general binding rules under Article 9(8). It is foreseen that
Chapter 5
314 Smitheries and Foundries Industry
new installations can be designed to perform at or even better than the general BAT levels
presented here. It is also considered that existing installations could move towards the general
BAT levels or do better, subject to the technical and economic applicability of the techniques in
each case.

While the BAT reference documents do not set legally binding standards, they are meant to give
information for the guidance of industry, Member States and the public on achievable emission
and consumption levels when using specified techniques. The appropriate limit values for any
specific case will need to be determined taking into account the objectives of the IPPC Directive
and the local considerations.

The foundry industry is a differentiated and diverse industry. The elements of BAT applicable
to a specific foundry need to be selected according to the type of activity. A foundry basically
consists of a melting shop and a casting shop, both with their supply chain. For lost mould
casting this supply chain includes all activities related to moulding and core-making. In this
chapter, distinction will be made on the following basis: melting of either ferrous or non-ferrous
metal and casting in either lost or permanent moulds. Each foundry may be classified as a
combination of a melting with a moulding class. BAT is presented for each class as well as
generic BAT, common to all foundries.

5.1 Generic BAT (for the foundry industry)

Some BAT elements are generic and apply for all foundries, regardless of the processes they
apply and the type of products they produce. These concern material flows, finishing of
castings, noise, waste water, environmental management and decommissioning.

Material flows management
The foundry process involves the use, consumption, combination and mixing of various
material types. BAT requires the minimisation of raw materials consumption and the furthering
of residue recovery and recycling. Therefore, BAT is to optimise the management and control of
internal flows.

BAT therefore is to:

- apply storage and handling methods for solids, liquids and gases as discussed in the Storage
BREF
- apply the separate storage of various incoming materials and material grades (Section
4.1.2), preventing deterioration and hazards (Section 4.1.3)
- carry out storage in such a way that the scrap in the storage area is of an appropriate quality
for feeding into the melting furnace and that soil pollution is prevented as described in
Section 4.1.2. BAT is to have an impermeable surface for scrap storage with a drainage
collection and treatment system. A roof can reduce or eliminate the need for such a system
- apply internal recycling of scrap metal, under the conditions discussed in Section 4.1.4,
4.1.5 and 4.1.6
- apply the separate storage of various residue and waste types to allow re-use, recycling or
disposal
- use bulk or recyclable containers (Section 4.1.7)
- use simulation models, management and operational procedures to improve metal yield
(Section 4.4.1) and to optimise material flows
- implement good practice measures for molten metal transfer and ladle handling (Section
4.7.4).

Finishing of castings
For abrasive cutting, shot blasting and fettling, BAT is to collect and treat the finishing off-gas
using a wet or dry system. The BAT associated emission level for dust is 5 - 20 mg/Nm. The
techniques for off-gas collection and exhaust air cleaning are discussed in Section 4.5.10.1 and
4.5.10.2.
Chapter 5
Smitheries and Foundries Industry 315
For heat treatment, BAT is all of the following, to:
- use clean fuels (i.e. natural gas or low-level sulphur content fuel) in heat treatment furnaces
(Section 4.5.11.1)
- use automated furnace operation and burner/heater control (Section 4.5.11.1)
- capture and evacuate the exhaust gas from heat treatment furnaces.

Noise reduction
BAT is all of the following, to:
- develop and implement a noise reduction strategy, with general and source-specific
measures
- use enclosure systems for high-noise unit operations such as shake-out (see Section 4.5.9.3)
- use additional measures as described in Section 4.10, according to local conditions.

Waste water
BAT is all of the following, to:
- keep waste water types separate according to their composition and pollutant load
- collect surface run-off water and ue oil interceptors on the collection system before
discharge to surface water, as discussed in Section 4.6.4
- maximise the internal recycling of process water and the multiple use of treated waste water
(Section 4.6.1)
- apply waste water treatment for scrubbing water and other waste water flows, using one or
more of the techniques mentioned in Section 4.6.2 and 4.6.3.

Reduction of fugitive emissions
BAT is to minimise fugitive emissions arising from various non-contained sources in the
process chain, by using a combination of the following measures. The emissions mainly involve
losses from transfer and storage operations and spills, and are discussed in Section 4.5.1.1.
- avoid outdoor or uncovered stockpiles, but where outdoor stockpiles are unavoidable, to use
sprays, binders, stockpile management techniques, windbreaks, etc.
- cover skip and vessels
- vacuum clean the moulding and casting shop in sand moulding foundries according to the
criteria given in 4.5.1.1
- clean wheels and roads
- keep outside doors shut
- carry out regular housekeeping
- manage and control possible sources of fugitive emissions to water.
These techniques are further specified in Section 4.5.1.1. Additional storage techniques are
detailed in the Storage BREF.

Additionally, fugitive emissions may arise from the incomplete evacuation of exhaust gas from
contained sources, e.g. emissions from furnaces during opening or tapping. BAT is to minimise
these fugitive emissions by optimising capture and cleaning, taking into account the associated
emission levels as given in Sections 5.2 and 5.3. For this optimisation one or more of the
following measures are used, giving preference to the collection of fume nearest to the source:
- hooding and ducting design to capture fume arising from hot metal, furnace charging, slag
transfer and tapping
- applying furnace enclosures to prevent the release of fume losses into the atmosphere
- applying roofline collection, although this is very energy consuming and should only be
applied as a last resort.

Environmental management
A number of environmental management techniques are determined as BAT. The scope (e.g.
level of detail) and nature of the EMS (e.g. standardised or non-standardised) will generally be
related to the nature, scale and complexity of the installation, and the range of environmental
impacts it may have.
Chapter 5
316 Smitheries and Foundries Industry
BAT is to implement and adhere to an Environmental Management System (EMS) that
incorporates, as appropriate to individual circumstances, the following features: (see Section
4.12)

- definition of an environmental policy for the installation by top management (commitment
of the top management is regarded as a precondition for a successful application of other
features of the EMS)
- planning and establishing the necessary procedures
- implementation of the procedures, paying particular attention to
structure and responsibility
training, awareness and competence
communication
employee involvement
documentation
efficient process control
maintenance programme
emergency preparedness and response
safeguarding compliance with environmental legislation.
- checking performance and taking corrective action, paying particular attention to
monitoring and measurement (see also the Reference document on Monitoring of
Emissions)
corrective and preventive action
maintenance of records
independent (where practicable) internal auditing in order to determine whether or not
the environmental management system conforms to planned arrangements and has been
properly implemented and maintained.
- review by top management.

Three further features, which can complement the above stepwise, are considered as supporting
measures. However, their absence is generally not inconsistent with BAT. These three
additional steps are:

- having the management system and audit procedure examined and validated by an
accredited certification body or an external EMS verifier
- preparation and publication (and possibly external validation) of a regular environmental
statement describing all the significant environmental aspects of the installation, allowing
for year-by-year comparison against environmental objectives and targets as well as with
sector benchmarks as appropriate
- implementation and adherence to an internationally accepted voluntary system such as
EMAS and EN ISO 14001:1996. This voluntary step could give higher credibility to the
EMS. In particular EMAS, which embodies all the above-mentioned features, gives higher
credibility. However, non-standardised systems can in principle be equally effective
provided that they are properly designed and implemented.

Specifically for the foundry sector, it is also important to consider the following potential
features of the EMS:

- the environmental impact from the eventual decommissioning of the unit at the stage of
designing a new plant
- the development of cleaner technologies
- where practicable, the application of sectoral benchmarking on a regular basis, including
energy efficiency and energy conservation activities, choice of input materials, emissions to
air, discharges to water, consumption of water and generation of waste.

Chapter 5
Smitheries and Foundries Industry 317
Decommissioning
BAT is to apply all necessary measures to prevent pollution upon decommissioning. These are
described in Section 4.11 and include:
- minimising later risks and costs by careful design at the initial design stage
- developing and implementing an improvement programme for existing installations
- developing and maintaining a site closure plan for new and existing installations.
In these measures, at least the following process parts are considered: tanks, vessels, pipework,
insulation, lagoons and landfills.

5.2 Ferrous metal melting

Furnace selection
Steel is melted in both electric arc furnaces (EAF) and induction furnaces (IF). The choice
between furnace types is based on technical criteria (e.g. capacity, steel grade). Due to its
refining ability, the EAF allows the melting of lower grade scrap. This is an advantage in terms
of the recycling of metals, but requires an appropriate flue-gas capture and cleaning system, as
will be presented below.

For cast iron melting: cupola, electric arc, induction and rotary furnaces are applicable. The
selection will be based on technical and economic criteria.
For the operation of any selected furnaces type, elements of BAT are given below.

Cupola furnace melting of cast iron
For the operation of cupola furnaces, BAT is all of the following, to:
- use divided blast operation (2 rows of tuyres) for cold blast cupolas (Section 4.2.1.5)
- use oxygen enrichment of the blast air, in a continuous or intermittent way, with oxygen
levels between 22 and 25 % (i.e. 1 % - 4 % enrichment) (Section 4.2.1.6)
- minimise the blast-off periods for hot blast cupolas by applying continuous blowing or long
campaign operation (Section 4.2.1.8). Depending on the requirements of the moulding and
casting line, duplex operations must be considered
- apply good melting practice measures for the furnace operation as listed in Section 4.2.1.1
- use coke with known properties and of a controlled quality (Section 4.2.1.2)
- clean furnace off-gas by subsequent collection, cooling and dedusting using a combination
of the techniques described in Section 4.5.2.1. BAT for dedusting is to use a bag filter or
wet scrubber. BAT associated emission levels are given below (Table 5.1, Table 5.2)
- apply post combustion in the cupola shaft of CBC, if the off-gases can burn autothermally
and then to recover the heat for internal use (Section 4.5.2.3). For HBC, use a separate
combustion chamber (Section 4.5.2.2), and recover the heat for blast air preheating and
other internal uses (Section 4.7.3)
- evaluate the possibility of waste heat utilisation from holding furnaces in duplex
configuration and to implement heat recovery if applicable (Section 4.7.2)
- prevent and minimise dioxins and furan emissions to a level below 0.1ngTEQ/Nm, using a
combination of measures as given in Section 4.5.1.4. In some cases this may result in a
preference for wet scrubbing. Industry has expressed doubts on the implementation of
secondary measures that have only been proven in other sectors and in particular questions
the applicability for smaller foundries
- use a wet scrubber system when melting with basic slag (basicity up to 2) (Section 4.2.1.3).

Residues produced by cupola melting include dust, slags and coke breeze.
BAT for residue management is all of the following, to:
- minimise slag forming using one or more of the in-process measures listed in Section 4.9.3
- pretreat the slags in order to allow external re-use (Section 4.9.2)
- collect and recycle coke breeze (Section 4.9.4.1).

Chapter 5
318 Smitheries and Foundries Industry
Electric arc furnace melting of steel and cast iron
For the operation of electric arc furnaces, BAT is all of the following, to:
- apply reliable and efficient process controls to shorten the melting and treatment time
(Section 4.2.2.1)
- use the foamy slag practice (Section 4.2.2.2)
- capture furnace off-gas using one of the techniques discussed in Section 4.5.3.1
- cool the furnace off-gas and dedust using a bag filter (Section 4.5.3.2).

Residues produced by EAF melting include dust and slags.
BAT for residue management is to:
- recycle filter dust into the EAF furnace (Section 4.9.5.1).

BAT associated emission levels are given below (Table 5.1, Table 5.3)

Induction furnace melting of cast iron and steel
For the operation of induction furnaces, BAT is all of the following, to:
- melt clean scrap, avoiding rusty and dirty inputs and adhering sand
- use good practice measures for the charging and operation as discussed in Section 4.2.3.1
- use medium frequency power, and when installing a new furnace, to change any mains
frequency furnace to medium frequency (Section 4.2.3.2)
- evaluate the possibility of waste heat recuperation and to implement a heat recovery system
if applicable (Section 4.7.2)
- use a hood, lip extraction or cover extraction on each induction furnace to capture the
furnace off-gas (Section 4.5.4.1) and to maximise off-gas collection during the full working
cycle
- use dry flue-gas cleaning (Section 4.5.4.2), taking into account the BAT associated emission
levels as given in Table 5.1
- keep dust emissions below 0.2 kg/tonne molten iron.

Rotary furnace melting of cast iron
For the operation of rotary furnaces, BAT is all of the following, to:
- implement measures to optimise furnace yield as discussed in Section 4.2.4.1
- use an oxyburner (Section 4.2.4.2)
- collect the off-gas close to the furnace exit, apply post combustion, cool it using a heat-
exchanger and then to apply dry dedusting (Section 4.5.5.1), taking into account the BAT
associated emission levels as given in Table 5.1 and Table 5.4
- prevent and minimise dioxins and furan emissions to a level below 0.1ngTEQ/Nm, using a
combination of measures as given in Section 4.5.1.4. In some cases this may result in a
preference for wet scrubbing. Industry has expressed doubts on the implementation of
secondary measures that have only been proven in other sectors and in particular questions
the applicability for smaller foundries.

Ferrous metal treatment
If an AOD converter is used for steel refining, BAT is to
- extract and collect the exhaust gas using a roof canopy.

For the production of nodular iron, nodularisation is performed. BAT for nodularisation is to:
- select a nodularisation technique with no off-gas production or to capture the produced
MgO smoke, using a lid or cover equipped with extraction equipment or by using a fixed or
movable hood, and to
- dedust the exhaust gas using a bag filter and to make the dust available for recycling.

BAT associated emission levels are given below (Table 5.1)

Chapter 5
Smitheries and Foundries Industry 319
BAT associated emission levels
The following emission levels are associated to the BAT measures stated above. All associated
emission levels are quoted as an average over the practicable measuring period. Whenever
continuous monitoring is practicable, a daily average value is used. Emissions to air are based
on standard conditions, i.e. 273 K, 101.3 kPa and dry gas.

Parameter Emission level
Dust
(1)
5 20 mg/Nm
PCDD/PCDF 0.1 ng TEQ/Nm
3
(1) The emission level of dust depends on the dust
components, such as heavy metals, dioxins, and its mass flow.
Table 5.1: Emission to air associated with the use of BAT for ferrous metal melting and treatment

Type Parameter Emission level
(mg/Nm)
Hot blast Carbon monoxide 20 1000
SO
2
20 100
NO
x
10 200
Cold Blast SO
2
100 400
NO
x
20 70
NM - VOC 10 20
Cokeless NO
x
160 400
Table 5.2: Emissions to air associated with the use of BAT for the cupola melting of ferrous metals

Parameter Emission level
(mg/Nm)
NO
x
10 50
CO 200
Table 5.3: Emissions to air associated with the use of BAT for the EAF melting of ferrous metals

Parameter Emission level
(mg/Nm)
SO
2
70 130
NO
x
50 250
CO 20 30
Table 5.4: Emissions to air associated with the use of BAT for the rotary melting of ferrous metals

5.3 Non-ferrous metal melting

For non-ferrous metals, this document (only) considers the melting of ingots and internal scrap,
since this is standard practice in non-ferrous foundries.

For aluminium melting, multiple furnace types apply. The selection of the furnace type is based
on technical criteria (e.g. regime, capacity, type of casting line). These are given in Section 3.3
and Table 3.21. One foundry can use several furnace types. Operational practice and logical
reasoning shows that centralised melting in higher capacity furnaces has a more favourable
energy efficiency than small crucible melting for large-scale plants. No data are available
however to select it as BAT.

For the melting of copper, lead and zinc and their alloys, induction or crucible furnaces are
used. For copper alloys, hearth type furnaces are used as well. The selection depends on
technical criteria.

For magnesium melting, only crucible furnaces are used. A cover gas is used to prevent
oxidation.
Chapter 5
320 Smitheries and Foundries Industry
Induction furnace melting of aluminium, copper, lead and zinc
For the operation of induction furnaces, BAT is all of the following, to:
- use good practice measures for the charging and operation as discussed in Section 4.2.3.1
- use medium frequency power, and when installing a new furnace, to change any mains
frequency furnace to medium frequency (Section 4.2.3.2)
- evaluate the possibility of waste heat recuperation and to implement a heat recovery system
if applicable (Section 4.7.2)
- minimise emissions in accordance to the associated emission levels stated below, and if
needed to collect the furnace off-gas (Section 4.5.4.1) maximising off-gas collection during
the full working cycle, and to apply dry dedusting.

Rotary furnace melting of aluminium
For the operation of rotary furnaces, BAT is to:
- implement measures to optimise the furnace yield as discussed in Section 4.2.4.1
- collect the off-gas close to the furnace exit and to evacuate it through a stack, taking into
account the BAT associated emission levels given below.

Hearth type furnace melting of aluminium and copper
For the operation of hearth type furnaces, BAT is all of the following, to:
- collect the furnace off-gas and to evacuate it through a stack, taking into account the BAT
associated emission levels given below
- capture fugitive and visible emissions according to the BAT elements for fugitive emissions
as mentioned above (Section 5.1) and to apply hooding as discussed in Section 4.5.6.1.

Shaft furnace melting of aluminium
For the operation of shaft furnaces, BAT is all of the following, to:
- allow efficient exhaust collection upon tilting of the furnace, and to evacuate the exhaust
gas through a stack, taking into account the BAT associated emission levels given below.

Radiant roof furnace holding of aluminium
For the operation of radiant roof furnaces, BAT is all of the following, to:
- follow the BAT elements for fugitive emissions as mentioned above (Section 5.1) and to
apply hooding under the conditions discussed in Section 4.5.6.1.

Crucible melting and holding of aluminium, copper, lead and zinc
For the operation of crucible furnaces, BAT is to:
- follow the BAT elements for fugitive emissions as mentioned above (Section 5.1) and to
apply hooding under the conditions discussed in Section 4.5.6.1.

Degassing and cleaning of aluminium
BAT for the degassing and cleaning of aluminium is to:
- use a mobile or fixed impeller unit with Ar/Cl
2
or N
2
/Cl
2
gas (Section 4.2.8.1).

Melting of magnesium
For the melting of magnesium, BAT is to:
- use SO
2
as a cover gas or to replace SF
6
by SO
2
as a cover gas. This applies for installations
with an annual output of 500 tonnes and more (Section 4.2.7.1)
- for smaller plants, use SO
2
as a cover gas or take measures to minimise SF
6
consumption
and emissions, as discussed in Section 4.2.7.1. In the case where SF
6
is used, the BAT
associated consumption level is <0.9 kg/tonne casting for sand casting and <1.5 kg/tonne
casting for pressure die-casting.

Note: At the time of reaching this conclusion, other alternatives to SF
6
were just being brought
into use (see 4.2.7.1). SO
2
can be of concern on health and safety grounds and can be corrosive
to equipment.

Chapter 5
Smitheries and Foundries Industry 321
BAT associated emission levels
The following emission levels are associated to the BAT measures stated above. All associated
emission levels are quoted as an average over the practicable measuring period. Whenever
continuous monitoring is practicable, a daily average value is used. Emissions to air are based
on standard conditions, i.e. 273 K, 101.3 kPa and dry gas.
The BAT AEL for dust for non-ferrous metal melting and treatment is 1 20 mg/Nm.
Additional values for the melting of aluminium are given in Table 5.5.

The emission factor associated with BAT for dust emissions from aluminium melting is 0.1
1 kg/t of molten aluminium.

In order to comply with these BAT associated emission levels it may be necessary to install a
flue-gas cleaning installation; in this case BAT is to use dry dedusting.

Furnace
Type
Parameter Emission level
(mg/Nm)
General Chlorine 3
Shaft SO
2
30 50
NO
x
120
CO 150
VOC 100 150
Hearth SO
2
15
NO
x
50
CO 5
TOC 5
Table 5.5: Emissions to air associated with the use of BAT in the melting of aluminium

5.4 Lost mould casting

Lost mould casting involves moulding, core-making, pouring, cooling and shake-out. This
includes the production of green sand or chemically-bonded sand moulds and chemically-
bonded sand cores. Mould and cores are assembled and the molten metal is poured into the
assembled mould. The casting is allowed to solidify and cool before shake-out. The available
techniques and their environmental effects for moulding and casting are closely interlinked.
BAT elements will be presented in three categories: green sand moulding, chemically-bonded
sand moulding and pouring/cooling/shake-out.

Green sand moulding
Green sand preparation starts with the mixing of the sand, clay binder and necessary additives.
This may be done in atmospheric- (the most common situation) or vacuum mixers (see Section
4.3.2.1). Both methods are considered to be BAT. For vacuum mixing, an additional condition
is that the sand capacity needs to be higher than 60 t/h.

Furthermore, BAT for green sand preparation is to:
- enclose all the unit operations of the sand plant (vibrating screen, sand dedusting, cooling,
mixing operations) and to dedust the exhaust gas (Section 4.5.8.1), taking into account the
BAT associated emission levels as given in Table 5.6. If the local market allows, the
captured dust is made available for external re-use (Section 4.8.13). Dust collected from
shake-out, dosing and handling is recycled into the green sand circuit, at up to 50 % of the
collected mass (Section 4.8.12)
- apply primary regeneration as discussed in Section 4.8.2. The new sand addition depends on
the amount of cores used and their compatibility. For green sand monosystems, regeneration
ratios (mass of regenerated sand/total sand mass) of 98 % is associated with the use of BAT.
For systems with a high degree of incompatible cores the BAT-associated regeneration ratio
is 90 94 %.

Chapter 5
322 Smitheries and Foundries Industry
Chemically-bonded sand mould and core-making
Various binder types are in use, all with their own specific properties and applicability. All
binders are determined as BAT if they are applied according to the discussed good practice
measures, which mainly involve process control and exhaust capture measures to minimise
emissions (Section 4.3.3.3, 4.3.3.4). BAT associated emission levels are given below (Table
5.6).
For chemically-bonded sand preparation, BAT is all of the following, to:
- minimise the binder and resin consumption and sand losses, using process control measures,
i.e. (manual or automated) mixer control, as discussed in Section 4.3.3.1. For series
production involving frequent changes of production parameters and with high production
throughputs, BAT is to apply the electronic storage of production parameters (see Section
4.3.3.2)
- capture exhaust gas from the area where cores are prepared, handled and held prior to
dispatching
- use water-based coatings and to replace alcohol-based coatings for the refractory coating of
moulds and cores, in foundries producing medium and large series. The use of alcohol-
based coatings is BAT
for big or complex moulds and cores
for water glass bonded sands
in magnesium casting
in the production of manganese steel with MgO-coating.
Both water-based and alcohol-based coating techniques are BAT in small-scale foundries
and large-scale jobbing foundries (see Section 4.3.3.5). Implementation of the water-based
technique in these two foundry types is supported by the availability of microwave drying
(Section 4.3.3.6) and other novel drying techniques for which no information was
submitted. When alcohol-based coatings are used, BAT is to provide evacuation at the
coating stand, using movable or fixed hoods, noting that in jobbing foundries that apply
floor moulding this is not feasible.

Additionallly, for amine-hardened urethane-bonded (cold-box) core preparation, BAT is all of
the following, to:
- treat the evacuated exhaust gas for cold-box core preparation, using one of the methods
mentioned in Section 4.5.8.4. The amine emission can be maintained below 5 mg/Nm
- recover amine from cold-box exhaust scrubbing liquor, provided the total volume allows
economic operation (Section 4.6.5)
- use either aromatic-based or vegetable-based (i.e. non-aromatic) solvents. Both methods are
BAT (Section 4.3.3.7).

BAT is to minimise the amount of sand going to disposal, by adopting a strategy of regeneration
and/or re-use (see section 4.8.13) of chemically-bonded sand (as mixed or monosand). In the
case of regeneration, the following conditions apply:
- cold setting monosands (e.g. furan sand) are regenerated using simple mechanical
techniques, as presented in Section 4.8.3. This applies to all cold setting monosands, except
silicate sand. A regeneration ratio of 75 80 % may be achieved
- uncured cold-box and cold setting core sand are hardened and broken in a specific unit,
allowing a minimum internal re-circulation of 5 10 % of the core sand (Section 4.8.11)
- silicate monosand is regenerated using heating and pneumatic treatment. A regeneration
ratio of 45 85 % (as yearly average) may be achieved (Section 4.8.10). The use of slow-
reacting esters should be minimised
- cold-box, SO
2
, hot-box and Croning monosands and mixed organic sands are regenerated
using one of the following techniques: cold mechanical regeneration (e.g. grinding, impact
drum, pneumatic chafing) or thermal regeneration (Sections 4.8.4, 4.8.5, 4.8.6, 4.8.7). The
overall regeneration ratio depends on the amount of cores used. For core-making 40
100 % of regenerated sand can be used; in mould making 90 100 % of regenerated sand
can be used
- mixed green and organic sand are regenerated using mechanical-thermal-mechanical
regeneration (Section 4.8.8), grinding (Section 4.8.4) or pneumatic chafing (Section 4.8.6).
Chapter 5
Smitheries and Foundries Industry 323
For core-making 40 100 % of regenerated sand can be used; in mould making 90 100 %
of regenerated sand can be used
- the quality and composition of the regenerated sand is monitored
- regenerated sand is re-used only in compatible sand systems. Non-compatible sand types are
kept apart (see Section 4.8.1).

Alternative moulding methods (Section 4.3.4) and inorganic binders (Section 6.5) are
considered to have a promising potential for the minimisation of the environmental impact of
moulding and casting processes.
Pouring, cooling and shake-out
Pouring, cooling and shake-out generate emissions of dust, VOCs and other organic products.
BAT is to:
- enclose pouring and cooling lines and to provide exhaust extraction, for serial pouring lines
(Section 4.5.9.2), and
- enclose the shake-out equipment, and to treat the exhaust gas using wet or dry dedusting as
discussed in Section 4.5.9.3. The BAT associated emission level for dust is specified in
Table 5.6.

BAT associated emission levels
The following emission levels are associated to the BAT measures stated above. All associated
emission levels are quoted as an average over the practicable measuring period. Whenever
continuous monitoring is practicable, a daily average value is used. Emissions to air are based
on standard conditions, i.e. 273 K, 101.3 kPa and dry gas.

Emission Source Parameter Emission level
(mg/Nm)
General Dust 5 20
Core shop Amine 5
Regeneration units SO
2
120
NO
x
150
Table 5.6: Emissions to air associated with the use of BAT for moulding and casting using lost
moulds

5.5 Permanent mould casting

Permanent mould casting involves the injection of molten metal into a metal mould. The mould
is opened after solidification and the casting is taken out for finishing. Chemically-bonded sand
cores are used to a limited extent in gravity and low-pressure die-casting.

HPDC dies need coating and cooling in order to provide good solidification and release
properties. For these purposes, a release agent and cooling water are sprayed onto the die.
BAT for permanent mould preparation is all of the following, to:
- minimise the consumption of the release agent and water for HPDC dies, using one or more
of the process measures discussed in Section 4.3.5.1. This prevents the formation of a mist.
If prevention measures do not allow the BAT associated emission level for organic
substances as defined in Table 5.7 to be reached, use hooding and EP, as discussed in
Section 4.5.8.7
- collect run-off water into a waste water circuit for further treatment
- collect water leakage liquid from hydraulic systems into a waste water circuit for further
treatment, using oil interceptors (Section 4.6.4) and distillation, vacuum evaporation or
biological degradation as discussed in 4.6.6.

Chapter 5
324 Smitheries and Foundries Industry
BAT for chemically-bonded sand preparation is analogous to the elements mentioned in Section
5.4. Since lower amounts of waste sand are generated, BAT for shake-out and used sand
management are different. BAT for used sand management in permanent mould foundries is to:
- enclose the de-coring unit, and to treat the exhaust gas using wet or dry dedusting as
discussed in Section 4.5.9.3, taking into account the BAT associated emission level given in
Table 5.7, and
- if a local market exists, make sand from de-coring available for recycling (Section 4.8.13).

BAT associated emission levels
The following emission levels are associated to the BAT measures stated above. All associated
emission levels are quoted as an average over the practicable measuring period. Whenever
continuous monitoring is practicable, a daily average value is used. Emissions to air are based
on standard conditions, i.e. 273 K, 101.3 kPa and dry gas.

Parameter Emission level
(mg/Nm)
Dust 5 - 20
Oil mist, measured as total C 5 - 10
Table 5.7: Emissions to air associated with the use of BAT for permanent mould casting (incl.
HPDC)

Chapter 6
Smitheries and Foundries Industry 325
6 EMERGING TECHNIQUES FOR FOUNDRIES

6.1 Use of low cost combustible materials in cupola melting

Description
In order to reduce the consumption of (high quality) coke, techniques have been developed to
allow the use of high calorific value solid waste and lower grade coke as a fuel.

A particular type of cupola, named the FAR furnace, has been developed in Brazil, that allows
the melting of a conventional charge using tyres and plastic pieces (automotive shredder residue
- ASR, fluff, etc.), mixed with second rate coke. The cupola structure is totally different from
the conventional one. Only the metallic charge enters from the top. The solid fuel is fed in from
the side, so that it reaches the hot zone very quickly. Here hydrocarbons are cracked and
combusted.

Alternative injection systems allow the addition of fluff or dust directly through the tuyres or
by using specific injectors.

Achieved environmental benefits
A reduction in the consumption of cokes, as they are replaced by solid waste (ASR, fluff),
which is difficult to treat in conventional combustion systems.

Operational data
For the FAR furnace, the best results have been obtained using 35 % fluff and 65 % coke. This
technique has been developed up to pilot scale.

Cross-media effects
The application of alternative fuels will cause a change in the flue-gas composition; leading to
higher amounts of dust for disposal, possibly with a higher content of pollutants and an
increased risk of dioxins, PAHs and heavy metals. The thermal efficiency will be reduced.

Applicability
The application of this specific cupola type is only possible for new installations, since it has a
specific furnace layout. Injection systems for solid or fluffy fuels may be applied on existing
cupola furnaces.

Economics
Since the process has only been developed up to pilot scale, there are no economic data
available for industrial scale application.

Driving force for implementation
To reduce the costs for fuels and to integrate the foundry into a waste management policy.

Reference literature
[140, EU Thematic Network Foundry Wastes, 2001]

6.2 Recycling of metal-bearing filter dust (ferrous metals)

Description
Metal-bearing dust can be agglomerated either using a binder (most preferably cement) or by
mixing it with chips from machining, when the foundry has a machining shop. In the latter case
the agglomerates may also include metal-bearing sludge. These agglomerates will be mixed in
the induction furnace charge and melted. The metal yield can be over 90 %. The rest of the dust
goes into the slag. Suitable agglomeration machines exist on the market.

Chapter 6
326 Smitheries and Foundries Industry
Achieved environmental benefits
The amount of waste to be disposed of in landfills will be reduced, as will the need to buy
metallic material outside the foundry. The mineral part of the metallic dust will not melt and is
taken up by the slag. Thus it is converted into a less hazardous form of waste.

Operational data
Operational data are currently not available.

Cross-media effects
It may be possible that the slag formed will increase lining wear. Also the energy used in
melting may increase slightly. In steel foundries, there is a risk of carbon pick-up.

Applicability
The technique applies to new and existing iron and steel foundries using induction melting
furnaces.

Economics
In its current stage of development, no ecomic data are yet available. However, in initial
estimates for one iron foundry with a machining shop, the payback time was calculated to be
between two and four years.

Driving force for implementation
The main driving force for implementation may be that melting dust could, at least in some
cases, be classified as hazardous waste. This would incur very high disposal costs for foundries.

Example plants
No example plants are available.

Reference literature
The technique is under development in the EU in an FP5 Growth project BRICETS "Metal By-
Product Recovery in Induction Furnaces - Commercial, Environmental and Technical
Solutions", contract number G1RD-CT-2001-00482. The project is due to finish in spring 2004.
[140, EU Thematic Network Foundry Wastes, 2001], [202, TWG, 2002]

6.3 Amine recovery from the core-making waste gas by gas
permeation

In a pilot plant, amine (DMIA) was recycled from the exhaust air of a core shop by means of
gas permeation. Its composition was as follows:
- 80 90 % amine (DMIA)
- 10 20 % solvent
- 0 % water.

After several test series, the recycled amine was used to gas brake-disk cores. It was found that
there was no difference between these cores and the cores which were gased with standard
commercial DMIA. The test cores were poured under normal production conditions and the
finished brake disks underwent the usual technical tests.

All the results turned out to be within permissible tolerance. This showed that recycled DMIA
can be used under normal production conditions. Economic efficiency may be achieved after
further optimisation of the plant.
[217, Paul, et al., 1994]

Chapter 6
Smitheries and Foundries Industry 327
6.4 Separate spraying of release agent and water in
aluminium die-casting

Description
In aluminium high-pressure die-casting, the permanent moulds are cooled and sprayed with
release agent prior to casting to prevent the casting adhering to the mould. The conventional
technique for release agent use is that a mixture of release agent and water is applied to the hot
mould in one go, by means of several linearly arranged nozzles. Some of the water evaporates,
cooling the mould and leaving the release agent in place, whilst another part of the mixture runs
off the mould, leading to a loss of release agent and water. The evaporation may also cause the
formation of a mist. (see Section 4.5.8.7)

In an alternative process, water and release agent are applied separately. For this purpose, a row
of nozzles was added to the spray head for the separate application of release agent. These
nozzles can be actuated individually via a computer (see Figure 6.1).

Figure 6.1: Spray head with separate nozzles for water and release agent

Water is sprayed on first, cooling the mould by approximately 20 C, followed by spray
application of the release agent. The preliminary cooling causes less release agent to evaporate
and improves its adhesion to the mould. This reduces the consumption of release agent by some
25 %. Thermographic measurements showed that the technique cools moulds more strongly and
more evenly. Tests proved that it involves no leakage of release agent which otherwise would
have had to be collected and disposed of.

Release agent is applied only to those parts of the mould which come into contact with the
molten aluminium. This can reduce the consumption of release agent by another 30 %.

The use of computer-assisted actuation of the nozzles for the release agent has made the
previously used existing array of nozzles for compressed air redundant. This function is now
taken over by the nozzles for the release agent.

An advantage with small series and frequent mould changes is that the spraying programme can
be saved on the computer and can thus be reproduced quickly after a mould change.

Achieved environmental benefits
The consumption of release agent can be reduced by 25 % by the separate application of water
and release agent. Separate actuation of the nozzles makes it possible to achieve a further saving
of up to 30 %, depending on the geometry of the casting, as the release agent is applied only to
those parts of the mould which come into contact with the casting.

Water consumption can be reduced by about 15 %. There are no losses of water or release agent
to be disposed of. No gaseous emissions due to evaporation of release agent are generated.

Cross-media effects
No cross-media effects apply. No increase in energy costs is expected.
Chapter 6
328 Smitheries and Foundries Industry
Applicability
Previous tests were carried out in aluminium die-casting. The technique can also be applied in
zinc die-casting and in the injection moulding of plastics. Modifications are necessary when the
spray head is used at higher temperatures. These have not yet been tested.

Economics
Economic advantages result from the considerable savings in release agent and water, and from
the elimination of disposal costs.

Further advantages result from the major time savings achieved in the mould change, due to the
fact that spraying programmes can be saved and re-run.

The investment costs for this technique are higher than those of the conventional process. On
the plus side, different moulds can be sprayed individually with the same spray head.

Driving force for implementation
Considerable saving of material and water.
The loss of release agent-water mixture, which needs to be collected, is avoided.
Short time-scales required for machine set up and adaptation.
High reproducibility of the results (consistent quality).

Example plants
The spray heads are produced and sold by a German producer from the end of 2003. Enquiries
from the industry have already been received. Industrial use is expected to start by the end of
2003.

Reference literature
[234, Anders, 2003]

6.5 Inorganic binder material for core-making

Description
In order to reduce the consumption of organic binding material, which is responsible for
emissions and odour in foundries (which can lead to complaints from neighbours), different
compositions of inorganic binding materials have been developed for use in core-making in
aluminium die-casting and have already been used in the series production of intake pipes for
car engines.

The cores are produced using a binder that contains magnesium sulphate and/or polyphosphate.
The inorganic binders consist essentially of salt-water mixtures, as well as small amounts of
additives (e.g. ceramic) to prevent sand adhering or sticking to the casting. The proportion of
binder relative to the quantity of sand by weight is 3 to 8%. About half of this amount is solvent
and crystal water.

In order to achieve short drying times of 10 to 20 seconds, preheated sand (60 80 C) is blown
into heated core-shooting tools (120 140 C), in which the solvent water and the crystal water
evaporate and are flushed out by air. The inorganic ceramic core will then have a temperature
stability of more than 1000 C, and will maintain high strength. When coming directly into
contact with water, the core disintegrates within a few seconds. The mixing and shooting of
warm sand is a patented core-shooting method.

After casting, core removal can be performed either wet or dry. In dry core removal, the lumps
of sand are ground to grain size, which leaves the binder layer on the quartz sand largely intact.
Therefore, once core sand is in the core sand cycle, binder must be added to it at only 5 % of the
binder quantity originally used. In wet core removal, the castings are cooled rapidly by
quenching. This can cause a desired improvement of the microstructure and causes the binder to
dissolve completely.
Chapter 6
Smitheries and Foundries Industry 329
The different sand flowing properties will require revision of the core models and venting upon
implementation of this technique in an existing plant. In the drying step, all the moisture needs
to leave the core, a requirement which may limit the core dimensions.

Achieved environmental benefits
Unlike the cold-box process, which gives rise to emissions of organic compounds such as
amines, pyrolysis products and smoke, this technique involves no emissions from the binder,
neither during core-shooting nor during casting. No waste is generated through the disposal of
sand residues, and there is no need for costly sand regeneration. The quantity of sand to be
replaced is very small.

The ceramic additives contained in the binder eliminate the need to coat the cores.

When a dry technique is used for core removal, all the binder remaining in the sand can be re-
used.

Operational data
The fluidity of the binder-sand mixture is sufficient to produce complex shapes such as water
jacket cores (cylinder head). The stability of test cores is 250 - 300 N/cm
3
. The stability can be
adapted to core requirements by adding more binder.

As no gases form during casting, no problems with gas permeability arise. The binder therefore
allows the use of very fine sands, which significantly improve surface quality (e.g. depth of
roughness). As the regeneration rate is high, it is economically viable to use even expensive
alternative synthetic ceramic sands or e.g. zircon sand.

Cross-media effects
As there are no emissions from the binder, there is no need for waste gas collection and
treatment. The heating of the binder-sand mixture and the preheating of the core-shooting tools
give rise to higher energy costs compared to other core-making techniques.

Applicability
This technique is applicable for core-making in aluminium die-casting in new plants and in
existing plants after adaptation of the core-shooting automates. Further developments are aimed
at adapting the binder for use in iron casting and as a moulding material.

Economics
Despite the need to heat core-shooting tools, this technique leads to overall cost savings for
core-making, due among other things to the lower consumption of new sand and the fact that
costly thermal regeneration or the disposal of sand is no longer necessary. Costs savings can
also be generated by recirculating the binder. Considerable cost reductions are achieved by the
fact that there is no need for waste gas treatment systems to remove organic compounds, such as
amines and pyrolysis products, which can arise during core-shooting and casting when organic
binders are used.

According to rough estimates, the use of this technique reduces the costs for core-making by 30
to 50 % compared to the cold-box process.

Driving force for implementation
Considerable cost reduction.
Reduction of odour problems and organic emissions.

Example plants
Foundry of VW AG, Hanover, Germany

Reference literature
[235, Bischoff, 2003]

Chapter 7
Smitheries and Foundries Industry 331
7 CONCLUDING REMARKS

7.1 Timing of the work

The first plenary meeting of the TWG was in April 1999. However, shortly after this, the project
was put on hold from January 2000 due to a change of lead expert in the EIPPCB. The work
recommenced in November 2001. A second plenary meeting of the TWG was organised in
January 2002 to review the status of the work and the conclusions of the first meeting, taking
into account possible new developments in the sector. The first draft of the BREF was then sent
out to the TWG for consultation in November 2002. The comments were assessed and
integrated into the document and a second draft, including proposals for BAT conclusions was
sent out in May 2003. The final plenary meeting of the TWG was organised in October 2003.
After the final meeting there was a short consultation on the concluding remarks and executive
summary chapters before the final version of the document was produced.

7.2 Sources of information

Many documents were sent in by research centres, authorities and by industry as a basis of
information for this document. The BAT documents from Belgium [110, Vito, 2001] and from
the industry [32, CAEF, 1997] may be considered as the cornerstone documents. These were
used as a starting point in the writing of the chapter on techniques to be considered in the
determination of BAT (Chapter 4). These were supplemented by BAT documents and good
environmental practice guides from the UK, Spain, Denmark and the Netherlands. Valuable
information on specific techniques and their implementation in example plants was provided by
Germany. Throughout the project there was a good and open communication with foundry
experts from research centres and associations in France, Germany, Spain, UK, Finland,
Belgium, Italy, Sweden and Portugal. Site visits were carried out in Germany, France, Spain,
Belgium and Poland. The consultation rounds provided specific feedback from foundry
operators, remarks on applicability and implementation of some techniques and additional
operational data. Throughout the project, special attention was given to the involvement of the
new Accession Countries that have important foundry industries. This resulted in the active
participation of Poland, the Czech Republic and Hungary in particular in the information
exchange.

In order to get a better view on what information was available on dioxin emission and
abatement, the industry delegation organised a special workshop, in September 2003. This
workshop was attended by over 30 participants, both TWG members and industry experts. It
provided additional operational data and useful technical information.

The majority of the documents arising from the information exchange related to ferrous
foundries. Non-ferrous foundry processes have been underrepresented throughout the exchange
process. Nevertheless, there was a good communication with a few industry experts, although
they did not attend the plenary meetings of the TWG. Valuable information was gathered during
site visits at non-ferrous foundries in Germany, France and Belgium. The International
Magnesium Association provided useful input on cover gases for magnesium melting.

The data quality of this document shows an imbalance, in that, for instance, consumption and
emission levels mentioned in Chapter 3 and Chapter 4 vary from survey data for the sector in a
whole country, to data covering several example plants, to specific data on single plants.

Emission data for non-ferrous metals are mainly based on single installations. For the melting of
copper and zinc, only emission factors (given in kg/tonne of metal or kg/tonne of casting) were
supplied. For the melting of magnesium and lead, no consumption and emission levels were
provided. For the melting of aluminium, no emission data were provided for induction, rotary or
Chapter 7
332 Smitheries and Foundries Industry
crucible furnaces. Where available, the emission factors are mentioned in Chapter 3. However,
there was no clear correlation between the mentioned emission factors and applied techniques.
Therefore BAT conclusions with an associated emission level were not possible for the melting
of non-ferrous metals other than aluminium.

The document contains information on waste water prevention and treatment techniques.
However, no data were supplied on the emission levels associated with these techniques. No
operational data on waste water quality in relation to the applied treatment were provided. It was
therefore not possible to determine a BAT associated emission level for waste water.

No information was provided concerning techniques for the abatement of VOC-emissions. The
lack of information on abatement techniques and associated emission levels made it impossible
to formulate a BAT conclusion. Techniques for the reduction of VOC-emissions related to
coating processes (as applied to cores and moulds) may be found in the BREF on Surface
Treatment Using Solvents.

In the discussion on the re-use of dust and residues, no information was provided on treatment
or re-use techniques for dust from induction furnace melting, for dross from aluminium melting,
fettling dusts, abrasives or for blasting grit, although a re-use of most of these materials into
secondary metal production is possible.

Legislation surveys were sent in by two member states, but these are not necessarily
representative for the whole of the European Union. Information on the applicable legislation in
these two member states can be found at the following internet addresses:
- Austria: http://www.ris.bka.gv.at/
- Germany: http://www.bmu.de/de/txt/download/b_taluft/
7.3 Degree of consensus reached

The conclusions of the work were agreed at the final plenary meeting in October 2003 and a
high level of consensus was reached. The industry expressed one concern on the dioxin issue, as
expressed in detail in Section 5.2. Besides this concern, there was full agreement on all the BAT
conclusions as presented in this document.

The main issues of discussion at the final meeting concerned the techniques for the collection
and treatment of off-gas, the minimisation of fugitive emissions, noise reduction, the role of
environmental considerations in technical choices and the BAT associated emission levels. For
some processes, mainly in the field of non-ferrous metal melting, the information was
incomplete and therefore no BAT AEL could be agreed. During the meeting, there was little
discussion on the techniques that are specific for non-ferrous foundries, partially due to the
absence of industry experts in this field.

The issue about whether environmental considerations play a detrmining role in the basic choice
of melting furnaces and moulding methods, as formulated in Chapter 4 of the document, was
carefully considered. There was a general agreement that these choices are complex and should
be based on technical reasons and market considerations, on a case-by-case basis. Therefore the
issue was not selected as BAT.

The discussions on the treatment of off-gases concluded with an agreement that for most
operations in ferrous foundries both dry systems and well-performing wet systems are BAT.
The associated emission levels were based on the information available in the document.

Concerning the reduction of fugitive emissions and noise, there was some discussion on the
presentation of the topics and the level of detail to be given in Chapter 4 and Chapter 5. There
was a general agreement not to conclude on a specific level of noise which could be associated
with BAT.
Chapter 7
Smitheries and Foundries Industry 333
7.4 Recommendations for future work

The information exchange and the result of this exchange, i.e. this document, present an
important step forward in achieving the integrated prevention and control of pollution from the
foundry industry. On a few topics, however, the information is incomplete and did not allow
BAT conclusions to be reached. The main issues have been presented in Section 7.2. Future
work could usefully focus on collecting the following information:

- Techniques for VOC-abatement: VOC-emissions are generated by the thermal
decomposition of the binders upon pouring, in coating the cores and moulds and in the
melting of dirty scrap, and are one of the key environmental issues for the sector. VOC-
emissions in foundries typically involve large-volume, low-level flows, which represent a
significant load to the environment and may result in an odour emission. Information is
needed on techniques for the efficient capture and treatment of these flows, together with
associated emission and economic data. The prevention of VOC emissions through the use
of alternative binder and coating materials, also deserves further study. Data should be
gathered and reported from the implementation of these techniques in operational practice
- Waste water treatment: Waste water treatment is presented in a general descriptive way in
this document, and indeed operational data are given for only one example plant. The basis
for discussion should be broadened, through an exchange of information on the performance
of a broader range of treatment plants and providing further operational data. In view of the
selection of wet scrubbing as part of BAT, information should be gained on the associated
treatments of waste water and on emission levels to surface and sewage water flows
- Melting of non-ferrous metals: The discussion on non-ferrous metal melting is based on the
melting of clean ingots and clean internal return material. On this basis, low emission levels
are presumed. There is a need for more complete information on both guided and fugitive
emissions (e.g. of metals, organics) from non-ferrous metal melting in foundries, based on
operational practice and expressed both as an emission level (mass/exhaust volume) and as
a mass flow (mass/output of metal or castings)
- Economic data: There is a lack of economic information for many of the techniques
presented in Chapter 4. This information needs to be collected by industry and Member
States from ongoing projects dealing with the implementation of the presented techniques
- Energy Efficiency: Examples of techniques for heat recovery for induction and cupola
furnaces are given in this document. There is a need for additional examples of heat
recovery systems, with performance data. Information should be gathered concerning enrgy
consumption and energy losses for all types of furnaces, related to specific techniques to
increase the energy efficiency. This may then allow the determination of BAT-associated
energy consumption levels for metal melting.

7.5 Suggested topics for future R&D projects

The information exchange has also exposed some areas where additional useful knowledge
could be gained from research and development projects. These relate to the following subjects:

- Dioxin monitoring and abatement: In order to allow the full implementation of primary
measures for dioxin prevention, there is a need for a better understanding of the contribution
of the process parameters to the formation of dioxins. This requires the monitoring of dioxin
emissions for various installations and under varying conditions. Additionally, there is a
need for research on the use and effectiveness of secondary measures in the foundry
industry. Secondary measures (e.g. additive injection, catalytic filter bags) have been
implemented in other sectors (e.g. steel, waste incineration, non-ferrous metals production),
but there is very little experience of their performance and of possible operational problems
which may occur in applying then in foundries

Chapter 7
334 Smitheries and Foundries Industry
- Mercury emissions: Stack emissions of metals are largely related to dust. The minimisation
of dust emissions, as presented in the BREF will therefore also minimise possible metal
emissions. This holds true for all metals of concern for the sector, besides mercury.
Uniquely for metals the high volatility of mercury may cause gaseous emissions which are
not related to dust. In view of the implementation of a European policy on mercury
emissions, there is a need for research on the emissions of mercury from melting processes
in general and more specifically from (non-ferrous) foundries in particular
- Oxygas burners and their use in cupola furnaces: The application of oxygas burners in the
cupola furnace is described in Section 4.2.1.10. The technique shows some environmental
benefits. The difficulties encountered in the implementation of this technique, however limit
its widespread take-up. The TWG reported that new applications have been set up as a
result of ongoing research. There is a scope for further research and development here, to
bring this technique up to a development level that allows its further implementation.
- Alternative replacement gases for SF
6
in magnesium melting: Alternative covers gases to
replace SF
6
such as HFC-134a and Novec 612 have been developed and successfully tested
with some recent industrial implementation. These gases present an alternative for the
substitution by SO
2
. Further research and demonstration projects would provide more
detailed knowledge on the protective properties decomposition behaviour and emission
pattern of these compounds. This should allow a clearer indication of the applicability and
may support industrial implementation.

The EC is launching and supporting, through its RTD programmes, a series of projects dealing
with clean technologies, emerging effluent treatment and recycling technologies and
management strategies. Potentially these projects could provide a useful contribution to future
BREF reviews. Readers are therefore invited to inform the EIPPCB of any research results
which are relevant to the scope of this document (see also the preface of this document).

References
Smitheries and Foundries Industry 335
8 REFERENCES

2 Hoffmeister, J.; Reichardt, N. and Alwast, H. (1997). "NE-Metallindustrie -
Betreiberleitfaden fr Anlagen zum Gieen in verlorenen Formen - Sandguverfahren",
Ministerium Umwelt, Raumordnung u. Landwirtschaft NRW.

9 Schneider, B. (1993). "Verwertung von Gieereialtsanden durch thermisch-
mechanische Aufbereitung in einer zentralen Anlage", Umweltbundesamt, 30 441-5/9.

11 Schachtner, J. and Mller-Spth, G. (1993). "Demonstrationsanlage zur thermischen
Altsandregenerierung mit Komponententrennung fr Quarz- und Chromerzsand
(Demonstration Plant for Thermical Regeneration of Used Sand with Components
Separation of Silica and Chromite Sand - Final Report 9010)", Umweltbundesamt,
UBA3 0 441-2/4.

15 Gwiasda, F. (1984). "Erfassung und Reinigung von Abgasen bei der Kernfertigung
einer Gieerei", Umweltbundesamt.

16 Gerber, E. and Gwiasda, F. (1981). "Erfassung und Reinigung von Abgasen an
Ausschlagrosten einer Gieerei", Umweltbundesamt.

17 Strau, D. (1983). "Emissionsminderung druch Umbau einer Kupolofenanlage
(Reduction of Emissions by Reconstruction of Cupola Furnace Melting Plant - Final
Report 1018)", Umweltbundesamt.

18 Rademacher, H. (1993). "Errichtung und Betrieb einer Absaugung der Induktionsofen-
Schmelzerei und Nebenbetriebe (Establishment and Operation of Exhausting the
Induction Furnaces Melting Equipment and its Sub-Plants - Final Report 1107",
Umweltbundesamt, 50 441-5/168.

20 Gapp, H. (1998). "Reduzierung des Gehaltes an Benzol in den Abgasen von Gie- und
Khlstrecken durch den Einsatz eines Bio Filters (Reduction of the Benzene Content in
the Exhaust Gas from Casting and Cooling Lines by Using of a Biological Filter - Final
Report 2041)", Umweltbundesamt, 50 441 - 7/10.

21 Strau, D. (1994). "Emissionsminderung durch Anwendung moderner Prozesstechniken
in einer Gieerei (Reduction of Dust and Gaseous Emissions in a Foundry by Applying
Modern Process Techniques - Final Report 1095)", Umweltbundesamt, 0 441-5/145.

23 Brettschneider, D. and Vennebusch, B. (1992). "Luftreinhaltung - Verbesserung der
Emissionssituation eines Gieereischmelzbetriebes mit zwei Drehtrommelfen
(Improvement of Conditions for Emissions of a Melting Range with Two Rotary
Furnaces) - Final Report 1119", Umweltbundesamt, 50 441 - 1/56.

27 Kran, H.-P.; Rudolph, A. and Reinhold, M. (1995). "Reduction of Emissions of a Hot
Blast Copula Furnace by Application of a Fabric Filter and by Using of Waste Heat and
Residue - Final Report", Umweltbundesamt, UBA 0 441-/ZII 1.1-50 441-5/178.

29 Batz, R. (1986). "Stand der Technik bei der Emissionsminderung in Eisen-, Stahl- und
Tempergieereien", Umweltbundesamt.

32 CAEF (1997). "BAT for the Abatement of Atmospheric Pollution in the Ferrous
Foundry Industry", European Commission - DG XI.E.1.

34 Binninger, W. (1994). "Vermeidung von Abfllen durch abfallarme
Produktionsverfahren - Reststoffe aus Schelzanlagen fr Nichteisenmetalle (NE-
Schwermetallgu)", Abfallberatungsagentur (ABAG).
References
336 Smitheries and Foundries Industry
36 Winterhalter, J.; Mauersberger, G.; Bars, P. and Toussaint, D. (1992). "Vermeidung von
Abfllen durch abfallarme Produktionsverfahren - Gieereialtsande aus
Nichteisenmetallgieereien", Abfallberatungsagentur (ABAG).

37 Winterhalter, J.; Mauersberger, G.; Bars, P. and Toussaint, D. (1992). "Vermeidung von
Abfllen durch abfallarme Produktionsverfahren - Gieereialtsande aus Eisen-, Stahl-
und Tempergieereien", Abfallberatungsagentur (ABAG).

42 US EPA (1998). "EPA Office of Compliance Sector Notebook Project: Profile of the
Metal Casting Industry", U.S. EPA, EPA/310-R-97-004.

43 Batz, R. (1996). "Dioxin- und Furanemissionen aus Schmelzanlagen in Eisen-, Temper-
und Stahlgieereien", Umweltbundesamt, 0722 186X.

44 ETSU (1993). "Cupola Melting of Cast Iron in Iron Foundries (Good Practice - Guide
58)", Energy Efficiency Office - Department of the Environment.

45 ETSU (1990). "Achieving High Yields in Iron Foundries (Good Practice - Guide 17)",
Energy Efficiency Office - Department of the Environment.

46 ETSU (1995). "Energy Saving in Foundry Services (Good Practice - Guide 166)",
Energy Efficiency Office - Department of the Environment.

47 ETSU (1992). "Guidance Notes for the Efficient Operation of Coreless Induction
Furnaces (Good Practice - Guide 50)", Energy Efficiency Office - Department of the
Environment.

48 ETSU (1994). "Improving Metal Utilisation in Aluminium Foundries (Good Practice -
Guide 142)", Energy Efficiency Office - Department of the Environment.

55 ETSU (1993). "Electric Holding of Hot Metal in Iron Foundries", Energy Efficiency
Office - Department of the Environment.

64 ETSU (1997). "Non-Ferrous Foundries (second edition)", Energy Efficiency Office -
Department of the Environment.

70 ETSU (1998). "Optimising Sand Use in Foundries", Energy Efficiency Office -
Department of the Environment.

71 ETSU (1998). "Cost-Effective Management of Chemical Binders in Foundries
(GG104)", Energy Efficiency Office - Department of the Environment.

72 ETSU (1995). "Chemically Bonded Sand: Use and Reclamation (EG4)", Energy
Efficiency Office - Department of the Environment.

73 ETSU (1995). "Foundry Greensand: Use and Reclamation (EG 5 Guide)", Energy
Efficiency Office - Department of the Environment.

75 ETSU (1996). "Improved Process Control Reduces Mould Losses", Energy Efficiency
Office - Department of the Environment.

78 ETSU (1996). "Environmental Management Systems in Foundries (GG 43 Guide)",
Energy Efficiency Office - Department of the Environment.

80 ERM Lahmeyer International GmbH (1999). "Aufkommen und Entsorgung von
Gieereialtsanden (Amount of Waste Foundry Sand and Disposal)", Umweltbundesamt.

References
Smitheries and Foundries Industry 337
82 IfG - Institut fr Gieereitechnik (1996). "Untersuchung von Eisen- und NE-
Metallgieereien - Branchengutachten", Ministerium Baden-Wrttemberg, 0941-780 X.

96 Spillner, A. (1997). "Vermeidung von Kernsanden und Aminabfllen durch den Einsatz
des Lost-foam-Verfahrens im Leichtmetall-Seriengu", Abfallberatungsagentur
(ABAG).

103 Vereniging van Nederlandse Gemeenten (1998). "Werkboek milieumaatregelen Metaal
- en Elektrotechnische Industre", VNG Uitgeverij.

108 FEAF (1999). "Fundicin de Metales Ferrosos", Fundacin Entorno Empresa y Medio
Ambiente.

110 Vito (2001). "Beste Beschikbare Technieken voor de Gieterijen", Vito, ISBN 90 382
0312 8.

112 Salminen, S. and Salmi, J. (1999). "The Biological Air Purification System in Purifying
Odours Produced in a Foundry Process", Vapo Oy Biotech.

122 Kirst, J. (1999). "Regenerierung von Kernformstoffen und Formstoffen",
Umweltbundesamt, UBA III 2.2-50441-14/7.

126 Teknologisk (2000). "A Survey of the Pollution and Environmental Conditions in the
Foundry Industry", Danish Technological Institute.

128 IHOBE (1998). "Arenas de Moldeo en Fundiciones Ferreas", Gobierno Vasco.

129 infoMil (1999). "Reductie van de uitstoot van isopropylalcohol door ijzergieterijen",
infoMil.

130 de Wilde, J. M. and ten Houten, M. R. (1999). "LCA van drie gietsystemen voor
gietijzer op basis van de LCA van een compressor casing ten behoeve van een rotorhuis
Z3", TNO Industrie.

133 De Globe B.V. (1999). "Aanvraag revisievergunning Wet milieubeheer", Provincie
Limburg.

138 Metaalgieterij Giesen B.V. (1996). "Emissiemeetprogramma Thermische
Zandregeneratie", Provincie Limburg.

140 EU Thematic Network Foundry Wastes (2001). "Foundry Waste Possibilities in the
Future", Inasmet.

141 CERP (1999). "Foundry Process Emission Factors: Baseline Emissions from
Automotive Foundries in Mexico", Casting Emission Reduction Program.

143 Inasmet and CTIF (2002). "Eurofond 2002: Sand Moulding: Key to Competitiveness"
Eurofond 2002, San Sebastian (E).
144 US Dept. of Energy (1998). "Advanced Lost Foam Casting", Office of Industrial
Technologies, http://www.oit.doe.gov/factsheets/metalcast/pdfs/lostfoam.pdf.

145 Inductotherm (2002). "Inductotherm Safety Manual On-line", Inductotherm,
http://www.inductotherm.com/safety/safety.htm.

148 Eurofine (2002). "European Foundry Internet Network",
http://eurofine.reflexe.fr/BDdebut_an.html.
References
338 Smitheries and Foundries Industry
149 Beauvais, P. and Choplin, L. (2001). New technologies for the application of water-
based coatings", Fonderie Fondeur Aujourd'hui, 201, pp. 9-29.

150 ETSU (1998). "Long campaign hot-blast cupolas in iron foundries", GPCS366.

151 Gemco (1999). "Sand Cleaner", Gemco Engineers Foundy Projects.

152 Notzon, I. and Heil, D. (1998). "Verminderung von Kernsanden aus dem CO2-
Wasserglas-Verfahren durch mechanische Regenerierung", Abfallberatungsagentur
(ABAG).

153 Umweltbundesamt (2002). "Techniken zur Regenerierung von Mono- und Mischsanden
aus Giessereien", Umweltbundesamt, personal communication.

154 Baum (2002). "personal communication", firma Brer, personal communication.

155 European IPPC Bureau (2001). "Reference Document on Best Available Techniques in
the Non-Ferous Metal Industries", European Commission.

156 Godinot, P. (2001). Evolution of the Cupola in Europe", Fonderie Fondeurs
Aujourd'hui, 205, pp. 28-40.

157 Godinot, P.; Charbonnier, M.; Dervin, B. and Morin, A. (1999). Le traitement des
fumes de cubilot", Fonderie Fondeurs d'Aujourd'hui, 188, pp. 37-43.

158 Charbonnier, M.; Godinot, P. and Stephan, J. (1998). Devenir des poussires de
cubilots dpoussiers sec", Fonderie Fondeur Aujourd'hui, 174, pp. 44-52.

160 UK Environment Agency (2002). "Interim Guidance for the A2 Ferrous Foundries
Sector - External Consultation Draft", IPPC S(A2)2.01 (2002).

161 UK Environment Agency (2002). "Process Guidance Note Draft Electrical, Crucible
and Reverberatory Furnaces", UK Environment Agency, PG 2/3 (2002).

163 UK Environment Agency (2002). "DRAFT Process Guidance Note: Hot and cold blast
cupolas and rotary furnaces", UK Environment Agency, 163.

164 UK Environment Agency (2002). "Draft Process Guidance Note Processes for melting
and producing aluminium and its alloys", UK Environment Agency, PG 2/6a(2002).

165 UK Environment Agency (2002). "Draft Process Guidance Note Processes for melting
and producing magnesium and its alloys", UK Environment Agency, PG 2/6b(2002).

168 CAEF (2002). "Data relating to the European Foundry Industry", Committee of
European Foundry Associations, http://www.caef-eurofoundry.org/industry.htm.

169 Orkas, J. (2001). "Technical and environmental requirements for surplus foundry sand
utilisation", Helsinki University of Technology.

170 Godinot, P. and Carlier, P. (1999). Le Cubilot sans Coke", Fonderie - Fondeurs
d'Aujourd'hui, 185, pp. 32-38.

171 The Castings Development Centre (1999). "Beneficial Re-use for Managers", The
Castings Development Center.

172 Neumann, F. (1994). "Gusseisen", expert verlag.

References
Smitheries and Foundries Industry 339
173 Huelsen, W. B. (1985). Arc furnace emission control", The British Foundryman, july
1985, pp. 302-309.

174 Brown, J. R. (2000). "Foseco Ferrous Foundryman's Handbook", Butterworth-
Heinemann, ISBN0 7506 4284 X.

175 Brown, J. R. (1999). "Foseco Non-ferrous Foundryman's Handbook 11th Ed.",
Butterworth Heineman, ISBN 0 7506 4286 6.

176 ETSU (1998). "Induction Melting of Aluminium", Good practice guide 229.

177 Silva Ribeiro, C. A. (2002). "Mass balance data for non-ferrous foundries", Associao
Portuguesa de Fundio.

178 Wenk, L. (1995). "Use of hexachloroethane for refinement of aluminium alloys and
grain refinement of aluminium containing magnesium alloys", Verein Deutscher
Giessereifachleute (VDG), personal communication.

179 Hoppenstedt (2002). "Guss Produkte Jahreshandbuch", Giessereien - Zulieferer -
Ausstatter.

180 Assofond (2002). "Italian Foundry dust emission data", Assofond.

181 Ruffin, A. and Godinot, P. (1998). Recyclage des poussires du cubilot", Fonderie
Fondeurs Aujourd'hui, 175, pp. 18-41.

182 Closset, B. (2002). "SF6 replacement", International Magnesium Association, personal
communication.

183 Goovaerts, L. (2002). "Emission data from site visits", Vito, personal communication.

184 Godinot, P. and Ressent, S. (2002). Bibliographical study and survey on the coke-gas
cupola", Fonderie fondeur d'aujourd'hui, 216, pp. 32-41.

185 Spitz, G. (2002). "Operational data PSA foundry", PSA Sept-Fons, personal
communication.

186 Stephan, J. (1996). Rgnration thermique des sables de fonderie - caractrisation
d'une installation Regetherm 500", Fonderie Fondeurs d'Aujourd'hui, 158, pp. 37-47.

187 Stephan, J. (1997). Rgnration thermique des sables de fonderie - caractrisation
d'une installation Richards PX800", Fonderie Fondeurs d'Aujourd'hui, 170, pp. 21-53.

188 Stephan, J. (2000). Rgnration thermique des sables de fonderie - caractrisation
d'une installation Fataluminium Eco-Rec", Fonderie Fondeurs d'Aujourd'hui, 198, pp.
54-69.

189 Httenes-Albertus (2002). "Die neue Cold-box Generation".

190 Schubert, W. and Gjestland, H. Use of SO2 as protection gas in magnesium die
casting".

191 IMA; Erickson, S. E.; King, J. F. and Mellerud, T. "Recommended practices for the
conservation of sulphur hexafluoride in magnesium melting operations", International
Magnesium Association, technical committee report.

References
340 Smitheries and Foundries Industry
192 Gjestland, H. and Westengen, H. (1996). "Use of SF6 in the megnesium industry, an
environmental challenge" 3rd International Magnesium Conference, Manchester.
194 UNEP IPCC (2002). "Climate change 2001: the Scientific basis", Intergovernmental
Panel on Climate Change, http://www.grida.no/climate/ipcc_tar/wg1/001.htm.

195 UBA (2003). "Waste Water from Iron, Steel and Malleable Iron Foundry",
Umweltbundesamt.

196 Unido (2002). "Cleaner Production in Metalworking (foundry)", Unido Information
Sources on Cleaner Production,
http://www.unido.org/ssites/env/sectors/sectors303.html#1.

197 Nalonchem (2002). "Foundry coke", http://www.nalonchem.com/productos.htm.

198 Georg Fischer "Einsaugsystem fr Kupolfen", GF.

199 Metalodlew s.a. (2002). "De-dusting system for Electric Arc Furnaces in the cast steel
foundry Metalodlew s.a.", Metalodlew s.a.,.

200 Metalodlew s.a. (2002). "A site for regenerating of self-hardening moulding core sands
in the cast steel foundry Metalodlew s.a.", Metalodlew.

201 CEN (2000). "Definition and classification of grades of steel (EN10020)", European
Committee for Standardisation, EN10020.

202 TWG (2002). "Comments to first draft".

203 Linxe, D. (2002). "Le prototypage rapide et ses applications en fonderie", CTIF.

204 Carnicer Alfonso, P. L. (2001). "Emisiones de un Horno Rotativo de Oxigas para la
fabricacin de fundicin frrea. Diagnstico y medidas correctoras.", Universidad del
Pais Vasco.

205 European IPPC Bureau (2003). "Reference Document on Best Available Techniques on
Emissions from Storage", European Commission.

206 Ditze, A. and Scharf, C. (2000). "Remelting of Magnesium type 1 scrap with or without
flux?", Magnesium Industry Website, http://www.magnesium-
industry.com/Library/Highlights/Paper%20of%20the%20Month/papers/Remelting%20
of%20Magnesium%20Type%201%20Scrap%20with%20or%20without%20Flux.pdf.

207 Drews, B. (1996). Simultaneous mixing and cooling of moulding sand under vacuum",
Casting Plant+Technology International, 2, pp. 4-10.

208 Devinny, J. S. (1998). Cleaning the Air, Biologically", Civil Engineering, september,
pp. 46-49.

209 Metalodlew s.a. (2003). "Environment-frienly conditions and improvement of work
conditions", Metalodlew, http://metalodlew.nss.pl/eng/produkcja/srodowisko/index.php.

210 Martnez de Morentin Ronda, J. (2002). "Operational data Lur-Sue Foundry", Lur-Sue,
personal communication.

211 European IPPC Bureau (2000). "Best Available Techniques Reference Document on the
production of iron and steel", European Commission.

References
Smitheries and Foundries Industry 341
212 Zalensas, D. L. (1993). "Aluminium Castings Technology, 2nd ed.", American
Foundrymen Society.

213 CTIF and CQRDA (2002). "Cast Aluminium Alloys - The reference guide", Artecomm
document engineering.

214 Gerl, S. (2003). Aufbereitung von Feststoffmischungen mit Evactherm-
Vakuummischern", Giesserei-Praxis, 1-2003, pp. 17-22.

215 Mller, G. (1996). "Direkterfassung von Emissionen aus Formksten im Gro- und
Einzelgu (Hohl- und Vollform)", BIA, ISBN 3-88383-425-4.

216 Hobelsberger, H.; Krger, H. and Nisi, D. (1997). Erste Erfahrungen met einem
weiterentwickelten umweltvertrglichen Cold-Box-System bei Daimler-Benz",
Giesserei, 84/21, pp. 48-49.

217 Paul, H.; Nisi, D. and Timm, J. (1994). "Solvent (amine) recycling from the exhaust air
of a core shop by means of gas permeation" GIFA-Kongress Giessereitechnik '94,
Dsseldorf (D).
218 Harnisch, J. and Schwarz, W. (2003). "Cost and impact on emissions of potential
regulatory framework for reducing emissions of hydrofluorocarbons, perfluorocarbons
and sulphur hexafluoride", Ecofys GmbH, ko-Recherche, B4-
3040/2002/336380/MAR/E1.

219 Castings Technology International (2003). "Replicast (R)", CTI,
http://www.castingsdev.com/tech_services/replicast/replicast.htm.

223 Rauch, E.; Sigmund, A. and Galovsky, U. (2003). "Technische und
betriebswitrschaftliche Aspekte des Inhouse-Recycling von Mg-Legierungen",
http://www.lkr.at/german/news/news1/pdf/technische_betriebswirtschaftliche.pdf.

224 Helber, J.; Bautz, H. and Wolf, G. (2000). "PCDD-/PCDF-Emissionnen bei Kupolfen"
Conference Internationale sur le Cubilot, Strasbourg.
225 TWG (2003). "Comments to the Second draft".

226 CTIF (1997). "Evolution des moyens d'laboration de la fonte liquide", Ademe,
4.02.0025.

227 Godinot, P. (2004). "The economical future of cupola" International cupola conference,
Trier.
228 Galante, G.; Michilli, O. and Maspeto, R. (1997). "No-bake as we see it", Impianti
Machine Fonderia.

229 Lilja, B. G.; Westberg, H. and Naystrm, P. (2000). "Survey of isocyanates in
foundries, parts 1-5", Swedish Foundry Association - Svenska Gjuterifreningen,
000317.

230 CAEF (2003). "Presentations of the Dioxin Workshop" Dioxin Workshop 24/09/2003,
Brussels.
231 UNEP (2003). "Formation of PCDD and PCDF - an overview" Stockholm convention
on persistent organic pollutants, expert group on BAT and BEP, Research Triangle
Park (USA).
References
342 Smitheries and Foundries Industry
232 Novem (2000). Vervangen cokes-gestookte kroesoven door aardgas/zuurstof-gestookte
trommeloven", Tenders Industrile Energiebesparing, 39520/0047.

233 IARC (1997). "Silica, some silicates, coal dust and para-aramid fibrils", International
Agency for Research on Cancer.

234 Anders, U. (2003). kologisch und konomisch optimierter Trennstoffeinsatz beim
Aluminium-Druckguss", Integrierter Umweltschutz in Gieereien, Verein Deutscher
Gieereifachleute.
235 Bischoff, U. (2003). "Untersuchungen zum Einsatz eines wasserlosen, anorganischen
Kernbinders auf Basis von Magnesiumsulfat in einer Aluminiumleichtmetallgieerei",
Universitt Bergakademie.

236 UK Environment Agency (2001). "Guidance for the recovery and Disposal of
hazardous and non-hazardous waste (other than by incineration and landfill)", SEPA,
IPPC S5.06.

237 HUT (2003). "Foundry pictures database", Helsinki University of Technology.

Note: The numbers do not follow sequentially. They refer to the document numbers in the
EIPPCB database (Technical material on Smitheries and Foundries). Only the references
utilised in this document are quoted here.

Glossary
Smitheries and Foundries Industry 343
9 GLOSSARY

9.1 Abbreviations

AEL associated emission level
ADI austempered ductile iron
AFS American Foundrymen Society
AFS-number measure of particle size distribution of foundry sand
AOD converter argon oxygen decarburisation converter
AOX absorbable organic halides
BAT best available techniques
BREF BAT reference document
BTEX benzene, toluene, ethylbenzene,xylene
BTX benzene, toluene,xylene
CAD (before number): Canadian Dollar - ISO currency unit
CAD (general): Computer aided design
CAEF Committee of Associations of European Foundries
CAM computer-aided manufacturing
CBC cold blast cupola
CNC computer numerical controlled
DEM Deutsche Mark (German Mark - ISO currency unit)
DMEA dimethyl ethyl amine
DMIA dimethyl isopropyl amine
DMPA dimethylpropylamine
DS dry solid
EAF electric arc furnace
EEA European Environment Agency
EFR Emission Factor Rating
EIPPCB European IPPC Bureau
EMS environmental management system
EP electrostatic precipitator
EPER European Pollutant Emission Register
EPS expanded polystyrene
EUR euro
FA furfuryl alcohol
GWP global warming potential
GBP Great Britain Pound ISO currency unit
HBC hot blast cupola
HCE hexachloroethane
HFC hydrofluoro carbon
HIP hot isostatic pressing
HP high purity (for Magnesium alloys)
HPDC high pressure die-casting
ID-fan induced draught fan
IEF Information Exchange Forum (informal consultation
body in the framework of the IPPC Directive)
IF induction furnace
IMA International Magnesium Association
IPA isopropyl alcohol
IPPC integrated pollution prevention and control
IPTS Institute for Prospective Technological Studies
JRC Joint Research Centre
l. left
LCA life cycle assessment
LOI loss on ignition
Glossary
344 Smitheries and Foundries Industry
m. middle
MDI methyl di-isocyanate
n.a not applicable
n.d no data
NM-VOC non-methane volatile organic carbons
OU odour units
PAH polycyclic aromatic hydrocarbon
PCDD/F polychlorinated dibenzodioxins and -furans
PCB polychlorinated biphenyl
PE polyethylene
PEVA polyethylene vinyl acetate
PF phenol - formaldehyde
PFC perfluorinated carbon
PM particulate matter
PMMA polymethylmethacrylate
ppm parts per million
r. right
RF rotary furnace
RP rapid prototyping
RPM respirable particulate matter
SG spheroidal graphite
SME small and medium enterprise(s)
TEA triethyl amine
TEQ toxicity equivalent factor
TWA time-weighted average
TWG technical working group
VOC volatile organic compound
VODC vacuum oxygen decarburisation converter
UF urea-formaldehyde
USD United States Dollar ISO currency unit
Glossary
Smitheries and Foundries Industry 345
9.2 Terms

Annealing a heat treating process that softens the alloy by slow cooling
the metal through its critical temperature range
Austenitising a heat treating process that holds the alloy above the critical
temperature for a sufficient period of time to ensure that the
matrix is fully transformed to austenite
Air factor , the ratio of the air added to a combustion process over the
stoechiometric quantitiy of air needed for complete oxidation
Cast iron a ferrous alloy which solidifies with an eutectic solidification
reaction, whatever the chemical composition, but normally
with a carbon content above 2 %
Casting (noun) general term used for products at or near their finished shape,
formed by solidification of a metal or alloy in a mould (ISO
3134-4: 1985)
Diffuse emission emissions arising from direct contact of volatile or light dusty
substances with the environment (atmosphere, under normal
operating circumstances). These can result from:
- inherent design of the equipment (e.g. filters, dryers)
- operating conditions (e.g. during transfer of material
between containers)
- type of operation (e.g. maintenance activities)
- or from a gradual release to other media (e.g. to cooling
water or waste water).
Fugitive emissions are a subset of diffuse emissions
Emission the direct or indirect release of substances, vibrations, heat or
noise from individual or diffuse sources in the installation
into the air, water or land
End-of-pipe technique a technique that reduces final emissions or consumptions by
some additional process but does not change the fundamental
operation of the core process. Synonyms: "secondary
technique", "abatement technique". Antonyms: "process-
integrated technique", "primary technique" (a technique that
in some way changes the way in which the core process
operates thereby reducing raw emissions or consumptions)
Existing installation an installation in operation or, in accordance with legislation
existing before the date on which this Directive is brought
into effect, an installation authorised or in the view of the
competent authority the subject of a full request for
authorisation, provided that that installation is put into
operation no later than one year after the date on which this
Directive is brought into effect
Ferrous materials those materials of which iron is the major constituent, i.e. the
content of Fe (%w/w) is higher than that of any other element
(following EN10020:2000; 2.1)
Full mould process moulding technique using a foam model in chemically
bonded sand, in which the model is lost upon pouring,
generally used only for large castings
Grain refinement a liquid metal treatment made late in the melting operation to
produce a finer and equiaxed grain during solidification.
Green sand mix of sand, clay and additives as used for mould making
Grey iron cast iron with grey-coloured fracture; may be found with
lamellar, nodular and compact graphite but the term is most
often used for lamellar iron
Hearth type furnace static furnace with direct heating, also called reverberatory
furnace and bale-out furnace
Glossary
346 Smitheries and Foundries Industry
Inclusion (specific term for this sector) impurity in the metal structure,
producing structural defects; mostly used for non-ferrous
metals (e.g. oxides in magnesium, hydrogen in aluminium)
Jungbth or net diagram diagram representing the influence of the blast rate and the
coke charge on the tapping temperature of the metal and the
melting rate; used for cupola furnaces
Lamellar iron cast iron with carbon in the form of flakes
Lost Foam process moulding technique using foam models in unbonded sand, in
which the model is lost upon pouring of the metal, used for
series production
Lost mould or single use moulds, a mould that is specially made for each
casting and is destroyed after pouring. The moulds are
generally made of sand, chemically-bonded or clay-bonded,
or even unbonded. Investment casting can also be included in
this family
Lost mould casting all processes of the moulding and casting shop in a foundry
using lost moulds, such as sand preparation, moulding, core-
making, pouring, cooling and shake-out
Malleable iron iron that is capable of extension or of being shaped under the
hammer; produced by decarburising (whiteheart malleable
iron) or not decarburising (blackheart malleable iron) heat
treatment of graphite-free iron with white fracture surface
(white cast iron)
Melting capacity the aggregate plated capacity of the furnaces at an
installation; where the hourly rate multiplied by 24 should be
used if appropriate
Modulus silica-to-soda ratio SiO
2
/Na
2
O, used to characterise (sodium)
silicates
Nodular iron cast iron with carbon in nodular/spheroidal shape, commonly
refered to as ductile iron
Nodularisation magnesium treatment of cast iron to change the carbon
particles into nodular/spheroidal shape
Non-ferrous materials all materials that are not ferrous
Normalising a heat treating process that cools the alloy in air from a
temperature above its critical range
Off-site sand re-use utilisation of sands in various ways
Permanent mould or multi use a mould, mould that is used for gravity and low-
pressure casting, pressure die-casting, or centrifugal casting;
typically the moulds are metallic
Permanent mould casting all processes of the coreshop and casting shop in a foundry
using permanent moulds, such as sand preparation, core-
making, die-casting, take-out and decoring
Pollutant individual substance or group of substances which can harm
or affect the environment
Post combustion ignition and burning of exhaust gas through the injection of
air or the use of a burner; used in order to reduce the amount
of CO and (volatile) organic compounds
Primary regeneration the processing of used green sand for internal re-use in the
sand circuit, using screening, sieving, cooling and mixing
with new sand, fresh bentonite and other additives
Production capacity production of good castings and capacity based on the
theoretical capacity of the foundry if it was operated for 24
hours, provided that it is technically able to operate that way
Regeneration ratio the ratio between the mass of regenerated sand and the total
mass of sand used in mould and core-making, expressed as a
percentage
Glossary
Smitheries and Foundries Industry 347
Sand circuit the various parts of the installation for storage, transport,
preparation and treatment of sand
Sand recycling on-site used-sand preparation, for external re-use
Secondary (sand)
regeneration
measures (mechanical, pneumatic, thermal and wet) aimed at
removing spent binder layers from the sand corn (=
reclamation); returning the sand to a quality similar to, or
better than, new sand
Steel a ferrous alloy which solidifies according a preritetic reaction.
The (mass) content of iron is bigger than that of any other
element, and the carbon content is generally lower than 2%;
the alloy also usually contains other elements
Tempering a heat treatment that after the hardening of the alloy consists
of heating to a temperature below Ac
1
and cooling in air; after
austenitising and liquid quenching, this provides to the alloy
the highest possible hardness and strength
Glossary
348 Smitheries and Foundries Industry
9.3 Lexicon
English Deutsch Franais Nederlands Espaol Italiano Magyar
bentonite Bentonit bentonite bentoniet bentonita bentonite bentonit
bridging Brckenbildung accrochage brugvorming colgadura formazione di corona adagfennakads
burr Grat barbe braam rebarba bava fnc
cast iron Gusseisen fonte gietijzer hierro ghisa ntttvas
cast steel Stahlguss acier moul gietstaal acero moldeado getti d'acciaio nttt acl
casting Giessen couler het gieten fundicin colata nts
chamfer Abschrgung chanfrein gietsysteem chafln cianfrinatura formzsi ferdesg
channel induction furnace Rinnenofen four induction canal kanaal inductieoven horno de induccin de
canal
forno a canale csatorns indukcis kemence
coating Schlichte enduit, poteyage coating/deklaag pintura vernice, verniciatura formabevon anyag
cold-box kalter Kernkaste, Cold-
box
bote froide cold-box caja fria cassa fredda cold-box
core Kern noyau kern macho anima mag
corebox Kernkasten bote noyaux kerndoos caja de macho cassa d'anima magszekrny
core-making Kernfertigung noyautage kernmakerij macheria formatura di anime magkszts
crucible furnace Tiegelofen four creuset kroesoven horno de crisol forno a crogiolo tgelykemence
decoring Entkernen debourrage ontkernen extraccin de macho sterratura delle anime kimagozs
die Druckgiessform moule vorm mole metlico stampo kokilla, nyomsos ntvforma
die lubricant Druckgiessschlichte poteyage koelsmeermiddel lubricante distaccante kokilla-, nyomsos ntvforma
kenvanyag
drying oven Backofen tuve droogstoof estufa essiccatore szrt kemence
fettling putzen barbage ontbramen desmazarotado sbavatura tisztts
grinding schleifen meulage slijpen esmerilada molatura csiszols, kszrls
induction furnace Induktionsofen four induction inductieoven horno de induccin forno ad induzione indukcis kemence
lamellar iron or grey cast
iron
Grauguss, GGL fonte lamellaire,
fonte grise
lamellair gietijzer hierro gris ghisa a grafite
lamellare,
ghisa grigia
lemezgrafitos vagy szrke ntttvas
lustrous carbon Glanzkohlenstoff carbone brillant glanskoolvormer carbono brillante carbonio brillante fnyes karbon
malleable iron temperguss (GT) fonte mallable smeedbaar gietijzer fundicin maleable ghisa malleabile tempervas
mixer Menger malaxeur menger mezclador mescolatore keverv
mould Form mote vorm molde forme forma
nodular iron, ductile iron,
SG iron
GGG,
Gusseisen mit
Kugelgraphit
fonte graphite
sphroidal,
fonte ductile
nodulair gietijzer hierro nodular ghisa sferoidale gmbgrafitos ntttvas
pouring Giessen couler gieten colada colata nts
sand Sand sable zand arena sabbia homok
Glossary
Smitheries and Foundries Industry 349
English Deutsch Franais Nederlands Espaol Italiano Magyar
shaft furnace Schachtofen four cuve schachtoven horno de cuba forno a tino akns kemence
shake-out Ausleeren dcochage uitschudden desmoldeo distaffatura rts, kirzs
shot blasting Schleuderstrahlen grenaillage gritstralen granallado granigliatura szemcseszrs
sprue Einguss canal de coule voedingskanaal canal de colada canale di colata ll bemlv
turnings Spne copeaux spanen viruta trucioli forgcs
veins rissen gerces vinnen venas crestine repedsek
Annexes
Smitheries and Foundries Industry 351
10 ANNEXES

10.1 Annex 1 Overview of data for cast iron melting in
various cupola types and in the induction furnace

A survey of the consumption and emission levels of the various technical modifications of
cupola melting was set up by Neumann in 1994, as given below. All data refer to a system for
the production of 10 tonne/h nodular iron with 3.6 % C and 1.6 % Si and at a pouring
temperature of 1530 C. The balances in Figure 10.1 show inputs, outputs and process
temperatures. The latter will be higher compared to operational practice. Comparison of the
various balances allows an assessment of the effect of all the modifications.

Additional operational data for cold and hot blast cupolas, with various flue-gas cleaning
equipment types, are given and discussed in Section 4.5.2. Those data are taken from operating
plants.

The pictures show the following techniques:
(A): cold blast, secondary row of tuyres
(B): cold blast, oxygen injection
(C): cold blast with gas burner, oxygen injection
(D): hot blast (500 C), long campaign
(E): hot blast with electrical support heating (750 C)
(F): hot blast, oxygen injection (6 %)
(G): hot blast, with plasma support heating (1450 C)
(H): cokeless cupola in duplex operation (electrical superheating).

A B
Annexes
352 Smitheries and Foundries Industry
C D
E F
Annexes
Smitheries and Foundries Industry 353
G H
Figure 10.1: Input-output balance for various technical modifications of cupola melting
[172, Neumann, 1994]

Operational data are given in Table 10.1, Table 10.2 and Table 10.3. These include economic
data based on the stated unit costs. Operational data are based on the production of a cast iron
with 3.3 % C and 2 % Si. In the cold blast operation 35 % steel scrap is used, while the hot blast
operation allows 60 %. This is taken into account when considering the necessary carburisation.
For this, C is also supplied through the addition of the SiC briquettes.

Calculation of the heat balance results in the stated heat transfer efficiencies. The cold blast
cupola shows an efficiency of <30 %. The application of oxygen or secondary air increases the
efficiency to 37 40 %. The hot blast cupola shows a further increased efficiency, providing the
furnace wall is refractory lined. In liningless operation, the efficiency drops below 40 %, which
may be somewhat compensated for by adding oxygen. The cokeless cupola with inductive
superheating results in a very high efficiency, close to 60 %.

The calculated operational costs include depreciation of the investment cost, but exclude
personnel costs. The cost difference between the various techniques is rather small. The
calculation takes account of higher cokes consumption and more expensive scrap/raw material
costs for the cold blast cupola. In the case of air preheating using a plasma torch, the increased
electricity use leads to increased costs. In practice, this increase may be balanced by an
increased usage of swarf as raw material. The cokeless cupola shows the lowest cost. In this
evaluation, however, local cost effects and uncertainties, such as the effect of hourly rate, have
not been taken into account, but need to be considered. The quoted data are all for installations
with a significant hourly rate. This disfavours the performance of cold blast cupolas.

Annexes
354 Smitheries and Foundries Industry
Units or
Cost per unit
Normal O
2
Second-
ary row
Second-
ary + O
2
Specific quantity
Melting cokes kg/tonne 150 125 120 105
Cokes for carburisation kg/tonne 10 10 10 10
Natural gas m
3
/tonne
Electricity kWh/tonne 20 20 20 20
Oxygen % - 3 - 2
Blast air m
3
/tonne 711 562 604 517
Cupola off-gas m
3
/tonne 902 724 731 628
Filtered off-gas m
3
/tonne 3022 2150 1765 1646
Slags kg/tonne 36 35 46 45
Refractory kg/tonne 5.2 5.2 4.8 4.8
SiC- briquettes(45 % SiC) kg/tonne 28.5 28.5 31.3 30.3
Total heat input MJ/tonne 5003 3643 3588 3416
Heat transfer to iron % 26.9 37.2 37.8 39.8
Costs
Investment EUR million 2.81 2.91 3.06 3.16
Alloy EUR 332/tonne 9.44 9.44 10.36 10.36
Additives EUR 10/tonne 0.20 0.20 0.20 0.20
Cokes EUR 204/tonne 32.65 27.55 26.53 23.47
Steel scrap EUR 128/tonne 44.64 44.64 44.64 44.64
Scrap iron EUR 128/tonne 12.76 12.76 12.76 12.76
Pig iron EUR 194/tonne 38.78 38.78 38.78 38.78
Natural gas EUR 153/1000 m
3
0.31 0.31 0.31 0.31
Electricity EUR 89/MWh 1.79 1.79 1.79 1.79
Oxygen EUR 179/1000 m
3
- 3.01 - 1.84
Graphite EUR 561/tonne - - - -
Refractory EUR 765/tonne 3.98 3.98 3.67 3.67
Capital 10 years/8 % 11.22 11.63 12.24 12.65
Total melting cost EUR/tonne 155.77 154.08 151.28 150.46
Iron composition: 3.3 % C; 2.0 % Si; 0.3 % Mn; 0.1 S; 1500 C
Load composition: 35 % steel scrap; 35 % returns; 20 % pig iron; 10 % scrap iron
Table 10.1: Operational data of a cold blast cupola (10 tonne/h)
[172, Neumann, 1994]

Annexes
Smitheries and Foundries Industry 355
Units or
Cost per unit
Type A* Type B Type C Type D Type E
Characteristics
Daily furnace change
Oxygen enrichment
Lining
Yes
No
Yes

No
No
Yes

No
Yes
Yes

No
No
No

No
Yes
No
Specific quantity
Melting cokes kg/tonne 95 100 85 115 100
Cokes for carburisation kg/tonne 8 7 7 8 8
Oxygen % - - 3 - 3
Air m
3
/tonne 571 566 425 626 473
Cupola off-gas m
3
/tonne 680 667 519 746 586
Filtered gas m
3
/tonne 1798 1471 1134 1819 1473
Slag kg/tonne 60 60 60 60 60
Refractory kg/tonne 15 6 6 1.5 1.5
SiC- briquettes(45 % SiC) kg/tonne 50 53 53 50 50
Total heat input MJ/tonne 3756 3630 3076 4098 3534
Heat transfer to iron % 39.8 40.2 44.9 36.5 38.7
Costs
Investment EUR million 3.42 3.57 3.65 3.49 3.55
Alloy EUR 332/tonne 16.58 17.60 17.60 16.58 16.58
Additives EUR 10/tonne 0.20 0.20 0.20 0.20 0.20
Cokes EUR 204/tonne 21.02 21.84 18.78 25.10 22.04
Scrap steel EUR 128/tonne 76.53 76.53 76.53 76.53 76.53
Scrap iron EUR 128/tonne 12.76 12.76 12.76 12.76 12.76
Oxygen EUR 179/1000 m
3
- - 3.06 - 3.42
Refractory EUR 765/tonne 11.48 4.59 4.59 1.12 1.12
Capital 10 years/8 % 13.67 14.29 14.59 13.98 14.18
Total melting cost EUR/tonne 152.24 147.81 148.11 146.28 146.84
* Is no longer built as new plant
Iron Composition: 3.3 % C; 2.0 % Si; 0.3 % Mn; 0.1 S; 1530 C
Load composition: 60 % steel scrap; 30 % returns; 10 % scrap iron
Table 10.2: Operational data of a hot blast long campaign cupola (10 tonne/h)
[172, Neumann, 1994]

Annexes
356 Smitheries and Foundries Industry
Hot blast cupola
Units or
Cost per unit
Cokeless
Cupola
with O
2
+
induct.
superh.
air preh.
using
plasma
air preh.
using
electric
resistance
Specific quantity
Melting coke kg/tonne - 81 90
Coke for carburisation kg/tonne - 10 12
Natural gas m
3
/tonne 55 2 2
Electricity kWh/t 80 175 82
Oxygen % 1.5 - -
Air m
3
/tonne 5.30 419 474
Cupola off-gas m
3
/tonne 692 451 531
Cleaned off-gas m
3
/tonne 815 1313 1264
Slags kg/tonne 58 59 60
Refractory kg/tonne 16 9.3 4.3
SiC- briquettes (45 % SiC) kg/tonne - 13.7 15.4
FeSi- briquettes
(1 kg Si per briquette)
kg/tonne 8 - -
Total heat input MJ/tonne 2562 3952 3320
Heat transfer to iron % 58.4 41.4 41.5
Cost
Investment EUR million 2.81 5.10 3.83
Alloy EUR 332/tonne 4.54 5.10
Fe-Si-briquettes EUR 454/tonne Si 3.62
Additives EUR 10/tonne 0.56 0.56 0.56
Cokes EUR 204/tonne 18.57 23.06
Electricity EUR 89/MWh 7.14 15.61 7.30
Gas EUR 128/1000 m
3
6.99 0.31 0.31
Scrap steel EUR 128/tonne 44.64 44.64 44.64
Scrap iron EUR 128/tonne 44.64 44.64 44.64
Graphite EUR 561/tonne 6.73
Oxygen EUR 179/1000 m
3
1.43
Refractory EUR 102/tonne 0.61 0.31 0.20
Refractory EUR 765/tonne 7.65 4.80 1.79
Capital 10 years/8 % 11.22 20.41 15.31
Total melting cost EUR/tonne 135.26 154.39 142.91
Iron composition: 3.3 % C; 2.0 % Si; 0.3 % Mn; 0.1 S; 1530 C
Load composition: 35 % steel scrap; 30 % returns; 35 % scrap iron
Table 10.3: Operational data of a cokeless cupola and hot blast cupola with air preheating
[172, Neumann, 1994]

CTIF also calculated the costs of liquid cast iron in a comparative study about the different
melting ways [226, CTIF, 1997]. This study, made in 1997, was updated in 2003 [227, Godinot,
2004].

Annexes
Smitheries and Foundries Industry 357
GRADE
Lamellar cost iron (ENGJL 250) Nodular cost iron (ENGJS 400 15)
Units
Melting device
Cold blast
cupola
Hot blast
cupola
Cokeless
cupola
Induction
furnace
Hot blast
cupola
Cokeless
cupola
Induction
furnace
Metallic charge 95.80 59.52 96.68 72.97 49.75 73.52 59.80
Ferro alloys + additives 7.62 13.09 27.22 25.72 5.27 24.17 24.85
Energy and fluids 33.14 35.26 20.29 27.28 40.12 21.07 27.28
Linings 3.73 4.29 8.79 1.47 4.55 8.79 1.47
Personnel costs 10.05 10 11 9.10 10.42 11 9.10
Maintenance 1.55 3.15 0.16 0.93 3.15 0.16 0.93
Environment 4.50 4.30 3.15 0.55 4.30 3.15 0.55
Desulphurisation 10.33
Depreciation of investment 5.92 14.81 8.89 8.89 14.81 8.89 8.89
EUR/t
Total per tonne charged 162.31 144.42 176.78 146.91 142.70 150.75 132.97
%
Fire loss 5 5 5 3 5 5 3
EUR/t
Total per tonne liquid 170.85 152.02 185.45 151.45 150.21 158.68 136.98
%
Returns percentage 35 40 35 40 50 50 50
EUR/t
Total per tonne of casting 262.85 253.36 285.31 252.42 300.42 317.37 273.96
Table 10.4: Comparison of liquid cast iron costs -June 2003. Mass production 10 t/h
Annexes
358 Smitheries and Foundries Industry
Cold blast cupola Induction Coreless furnace Hot blast cupola Cokeless cupola
Characteristic Unit Price
EUR/unit
Ratio Price
EUR/t
Price
EUR/unit
Ratio Price
EUR/t
Price
EUR/unit
Ratio Price
EUR/t
Price
EUR/unit
Ratio Price
EUR/t
Price
EUR/unit
Ratio Price
EUR/t
Coke
Electricity
Oxygen
Gas
tonne
kWh
Nm
3
kWh
(NCV)
210
0.0513
0.535
0.0296
0.145
20
2
20
30.45
1.027
1.07
0.59
0.0448 609 27.28 0.0448 609 27.28
210
0.0513
0.535
0.228
0.125
128
2
60
26.25
6.57
1.07
1.37
0.0513
0.23
0.0186
165
11
500
8.46
2.53
9.30
Sub total 33.14 27.28 27.28 35.26 20.29
Returns
Pig iron
Scrap iron
Scrap steel (E3)
Scrap steel (E8)
Scrap steel (E1 C)
tonne
tonne
tonne
tonne
tonne
0
185
167
101.5
0.35
0.20
0.20
0.25
0
37
33.4
25.4
0
185
101.5
99.5
0.40
0.15
0.225
0.225
0
27.75
22.83
22.39
0
200
99.5
0.50
0.10
0.40
0
20
39.80
0
101.5
96.9
0.40
0.30
0.30
0
30.45
29.07
0
185
167
101.5
0.35
0.25
0.15
0.25
0
46.25
25.05
25.38
Sub total 95.8 72.97 59.80 59.52 96,68
Graphite
FeSi
FeMn
CaCO
3
+ div
Balls
kg
kg Si
kg Mn
tonne
tonne
0.68
0.595
12.2
8
2.7
0.048
5.44
1.60
0.58
0.70
1.13
0.79
16.7
10.6
2.6
11.69
11.98
2.05
0.80
1.13
17.92
9.39
14.34
10.61 0.68
0.59
12.2
15.8
3.2
0.0375
10.74
1.89
0.46
0.7
0.68
0.595
8.12
11. 71
1.45
5.68
7.96
0.86
12.72
Sub total 7.62 25.72 24.95 13.09 27.22
Table 10.5: Price of lamellar cast iron: Metallic charge + energy
Annexes
Smitheries and Foundries Industry 359
Hot blast cupola Cokeless cupola Units
Price
EUR/unit
Ratio Price
EUR/t
Price
EUR/unit
Ratio Price
EUR/t
Coke
Electricity
Oxygen
Gas
tonne
kWh
Nm
3
kWh
(NCV)
210
0.0513
0.535
0.0228
0.14
140.5
4
60
29.40
7.21
2.14
1.37

0.0513
0.23
0.0186

180
11
500

9.23
2.53
9.30
Sub total 40.12 21.07
Returns
Pig iron
Scrap iron
Scrap steel (E8)
Scrap steel (E3)
tonne
tonne
tonne
tonne
tonne
0
99.55

0.40

0.50
0
49.75

0
199
167

101.5
0.40
0.20
0.05

0.25
0
39.8
8.35

25.38
Sub total 49.75 73.52
Graphite
FeSi
FeMn
CaCO
3
+ div
Balls
kg
kg Si
kg Mn
tonne
tonne

0.68

12.2

7
0.042

4.76

0.51
0.7
0.68

12.37
4.11

8.66
2.79

12.72
Sub total 5.27 24.17
Table 10.6: Price of nodular cast iron: Metallic charge + energy

Table 10.4 summarises the costs of cast iron calculated for the 3 main cupola types compared
with coreless induction furnace. Some items greatly depend on the melting device and the
grade: energies and fluids, metallic charge and ferro alloys. They are detailed in Table 10.5 and
Table 10.6.

The compared melting devices melt at 10 - 12 t/h, in 2 shifts for a mass production typically,
e.g. the automotive industry. The investments are paid in 10 years and correspond to the
industrial practice in each case, i.e.:

Cold blast cupola
- 2 daily drop cupolas (one per day)
- treatment of fumes: combustion, cooling, dry filtration.

Hot blast cupola
- long campaign cupola
- treatment of fumes: combustion, blast heating, cooling, dry filtration
- holding in a channel induction furnace.

Cokeless cupola
- long campaign cupola
- treatment of fumes: cooling, dry filtration
- heating and recarburisation in a superheater
- holding in a channel induction furnace.

Introduction furnace
- 2 furnaces 12 tonne
- 1 power supply 10MW 250 Hz.

The cost of cast iron must take into account its downstream recarburisation (cokeless cupola),
its holding and also the importance of the fume treatment system.

The price of cast iron was compared for 2 different alloys in the case of mass automotive
industry:

- lamellar cast iron (ENGJL 250)
- nodular graphite base cast iron (base for ENGJS 400-15).
Annexes
360 Smitheries and Foundries Industry
The cold blast cupola is increasingly less used for nodular cast iron and was not considered in
this case.

Comparing the calculations of CTIF and Neumann:

The position of the melting devices, from the least to the most expensive, is not the same
depending on the type of alloy and the authors:

- Lamellar graphite cast iron (LG cast iron)
CTIF: induction furnace, hot blast, cold blast, cokeless cupola
Neumann: cokeless, hot blast, cold blast cupola
- Nodular base cast iron
CTIF: induction furnace, hot blast, cokeless (little different).

The comparison of the different items for Neumann and Godinot leads to the following
conclusions:

1. Energies and fluids
- the cokeless cupola is the least expensive in all cases
- it is noticeable that the energy costs strongly depend on the country, particularly with regard
to the price and availability of electricity and gas. In the CTIF study, energy costs
correspond to the French case (prices published in the review Energy Plus).

2. Metallic charges and ferro alloys
The classification technique applied of disadvantages the cokeless cupola for CTIF and not for
Neumann. The principal differences are the following ones:

- Neumann indicates 0 % pig iron for cokeless cupola, when Godinot indicates a minimum of
25 % according to the industrial practice. Pig iron is the most expensive metallic charge.
- Neumann gives the same cost for steel scrap and cast iron scrap. In the French case, the
ratio between these 2 costs has been between 2 and 1.3 for 10 years. This ratio was 1.6 in
June 2003, the date considered for the calculation, as indicated in the attached Figure 10.2.

It is noticeable that the cost of raw materials, including scrap, are fixed today at an international
level.

0.0
50.0
100.0
150.0
200.0
250.0
J
a
n
-
9
3
J
u
l
-
9
3
J
a
n
-
9
4
J
u
l
-
9
4
J
a
n
-
9
5
J
u
l
-
9
5
J
a
n
-
9
6
J
u
l
-
9
6
J
a
n
-
9
7
J
u
l
-
9
7
J
a
n
-
9
8
J
u
l-
9
8
J
a
n
-
9
9
J
u
l
-
9
9
J
a
n
-
0
0
J
u
l-
0
0
J
a
n
-
0
1
J
u
l
-
0
1
J
a
n
-
0
2
J
u
l
-
0
2
J
a
n
-
0
3
C
o
s
t
(
E
U
R
/
t
o
n
n
e
)
SG pig iron
Lamellar
pig iron
Scarp
iron (VF2)
Scrap
steel (E1C)
Figure 10.2: Cost of raw materials (France: 1993 to 2003)
Annexes
Smitheries and Foundries Industry 361
3. Other items
The investment is less important than for cold blast, cokeless cupola and coreless induction
furnace.
- in the case of the cold blast cupola liquid, cast iron is produced at a good temperature and
analysis is possible at the spout of the cupola; moreover the metallurgical results depend
little on the hourly production. It is not necessary to hold the cast iron in a channel induction
furnace
- in the case of the hot blast cupola, or of the cokeless cupola, it is difficult to modify the
hourly production. A reservoir of liquid metal is necessary between the cupola and the
moulding shop, i.e. generally a channel induction furnace
- in the case of the cokeless cupola, the specific volume of fumes to be treated is low and it is
not necessary to burn CO
- finally, in the case of the coke cupola (hot or cold blast), a desulphurisation is generally
practised before holding the base nodular liquid cast iron. The cost of this operation is high,
which is favours the cokeless cupola, whose cast iron does not contain any sulphur.

4. Conclusion
Taking industrial practice into account, according to Godinot, the cokeless cupola has the
advantage of a low energy consumption but it incurs extra cost for the raw charging materials.

In the case of base nodular cast iron, the cokeless cupola produces a metal without sulphur and
with a cost level similar to the hot blast. In the case of lamellar cast iron, in France, this melting
device produces a more expensive metal than coke cupolas.

Today, for this type of production in France the limit between cupola and induction is 10 t/h
(lamellar cast iron) and probably more in the case of nodular cast iron (the cupola is less
expensive for 20 t/h).

These calculations have been made for an automotive type mass production. For other
applications, they need to be slightly modified. At least it is possible to say the relative order
depends on the local conditions, especially energy costs which depend strongly on the country
of application.

10.2 Annex 2 Dust control equipment at a large-scale
automotive foundry

Dust emission monitoring data were provided by a Swedish large-scale automotive foundry.
The foundry uses a continuous particulate emission monitoring system on 17 bag filter exhausts.
Data were plotted on a graph for various time ranges and for 4 measurement points as shown in
Figure 10.3, Figure 10.4 and Figure 10.5. The figures show a continuous dust level well below
1 mg/Nm for sand preparation and shot blasting. A 24h-average of 1.3 1.6 mg/Nm was
obtained for the sand transport.

Annexes
362 Smitheries and Foundries Industry
Figure 10.3: Monitoring data for dust emissions from 4 locations in the sand plant (24 hours
continuous monitoring)

Figure 10.4: Monitoring data for dust emissions from 4 locations in the sand plant (2 weeks
continuous monitoring)

Annexes
Smitheries and Foundries Industry 363
Figure 10.5: Monitoring data for dust emissions from 4 locations in the sand plant (30 days
continuous monitoring)

This example plant has a permitted emission limit value for dust of 10 mg/Nm. The monitoring
system has an alarm limit of 5 mg/Nm. If this limit is passed, the filter is checked and tested
with fluorescence material. If the measured value passes 10 mg/Nm, the production and filter
are stopped. Calibration of the monitoring equipment is performed twice a year by an external
contractor.

The service and maintenance instructions are as follows:
- daily
pressure drop check
visual check
magnetic valve check
condensation water drain
- monthly
check tightness
- quarterly (every 3 months)
control with fluorescence material.