Research Online Research Online

University of Wollongong
Research Online
University of Wollongong Thesis Collection University of Wollongong Thesis Collections

1954-2016
1992
The effects of bed depth, basicity and suction on sintering technology

Liborio Münive
Follow this and additional works at: https://ro.uow.edu.au/theses
Copyright Warning
You may print or download ONE copy of this document for the purpose of your own research or study. The University
does not authorise you to copy, communicate or otherwise make available electronically to any other person any
copyright material contained on this site.
You are reminded of the following: This work is copyright. Apart from any use permitted under the Copyright Act
1968, no part of this work may be reproduced by any process, nor may any other exclusive right be exercised,
without the permission of the author. Copyright owners are entitled to take legal action against persons who infringe
their copyright. A reproduction of material that is protected by copyright may be a copyright infringement. A court
may impose penalties and award damages in relation to offences and infringements relating to copyright material.
Higher penalties may apply, and higher damages may be awarded, for offences and infringements involving the
conversion of material into digital or electronic form.
Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily
represent the views of the University of Wollongong.
Recommended Citation
Münive, Liborio, The effects of bed depth, basicity and suction on sintering technology, Master of
Engineering (Hons.) thesis, Department of Materials Engineering, University of Wollongong, 1992.
https://ro.uow.edu.au/theses/2503
Research Online is the open access institutional repository for the University of Wollongong. For further information
contact the UOW Library: research-pubs@uow.edu.au
T H E E F F E C T S OF BED D EPTH , B A S IC IT Y AND
SU CTIO N ON SIN TERIN G TECH N O LO G Y
By
LIBORIO MÜNIVE, B.E. (Chem. Eng.)

Dip. Proj. Eng., & Chemistry.
A Thesis submitted in fulfilm ent of

requirements for the Degree of
M ASTER OF ENGINEERING (HONOURS)
THE UNIVERSITY OF WOLLONGONG

DEPARTMENT OF MATERIALS ENGINEERING
January 1992
UNIVERSITY OF
WOLLONGOMG
LIBRARY
O í 8072
ACKN O W LED G EM EN TS
I would like to acknowledge the support, encouragement and guidance of my supervisor

Dr. N. Standish, Associate Professor of the Department of Materials Engineering.
Appreciation is expressed to the Head of Department of Materials Engineering,

Professor N.F. Kennon, W.J. Plumbridge, Mrs Sharon Nightingale, Dr G.W. Delamore,
and Mr P.G. Wood - Manager Academic Services, for their support and encouragement.
The author is indebted to Messrs J. Greenlees, W.R. Gadsden, A.S. Kristensen and R.J.
Brown from BHP Steel - Slab & Plate Products Division, (formerly Australian Iron &
Steel Pty Ltd) Port Kembla ironmaking and quality control (research) management,
for their permission to investigate the effects of bed depth, suction and basicity on
sintering.
I would also like especially to thank the following:
. Messrs A.K. Farey, B. Peden, A. Russell, G. Hamilton and J.D. Brace well from Port
Kembla Sintering Plant, who carried out the plant trials and exploited the research
test results mentioned herein.
. My colleagues I. McAuley, W.S. Felton, M. Dal Molin and B. Johnston from Ores &
Sinter Research, for their assistance during the development of research tests and as
technical observers and hourly communicators during the first stages of plant trials.
. Messrs E.J. Bagnall from CRL., R.J. O’Brien and R. Dieckmann from Port Kembla
Refractories Research, for their assistance relating to the microstructural observations
of the product sinters.
. Messrs P. Hamblin, B. Clift, D. Troth and K.J. Preuhs from the Central Laboratory.
. Messrs R. Perkuhn and F.W. Kolodziej, from the Instrument Room, for their help
during the experimental period.
. Mr M. Burns - Assistant General Manager Ironmaking; Messrs D. Lathlean, A. Talbot,

J. Drinkwater, A. Blakey and Mrs A. McCarthy from Ironmaking Technology &
Development - Port Kembla Steelworks; J. Campbell and J. Woolfrey from Newcastle
Sintering Plant, for their assistance and supply of fresh information.
. Messrs K. Price, R.T. Williams, W.R. Gadsden, J. Greenlees and L.A. Jenkins, who
assisted me by sharing their insights and criticizing my initial writing efforts. Mr
D.S. Pierce, Mrs Nelly F. Miinive-Pierce, Mrs Rosa M. Münive de Jenkins and my wife
Filoména E. Münive, who helped with the word processing of this thesis.
. The last but not least, I wish to thank Messrs E.C. Herkenhoff, S.S. Alderman, Jr and
P.R. Porath for their encouragements and support during my technical training in
America.
. Finally this work is gratefully dedicated with the deepest love to my parents, par
ticularly my mother, for their support during my undergraduate studies, and to my
wife and my children for their support and encouragement.
Table of Contents
1.0 INTRODUCTION ............................................................................................ 1
1.1 Overview ...................................................................................................... 1
1.2 Scope ............................................................................................................ 3
1.3 Sintering Plant Trials................................................................................. 4
1.4 Summary of Research Contributions ...................................................... 6
2.0 DEVELOPMENTS IN AGGLOMERATION PROCESSES ....................... 9
2.1 Overview ..................................................................................................... 9
2.2 Methods of Agglomeration ................................................................ 9
2.3 The Sintering Process .............................................................................. 10
2.3.1 Overview .............................................................................................. 10
2.4 DEVELOPMENT OF THESINTERING PROCESS ............................... 13
2.4.1 Overview .............................................................................................. 13
2.4.2 The Huntington-Heberlien Process.................................................. 14
2.4.3 The Greenawalt Process................................................................... 14
2.4.4 The Down-Draught Dwight-LloydSintering Process ..................... 16
2.5 DEVELOPMENT OF SINTERING OF FERROUS ORES IN
AUSTRALIA ..................................................................................................... 20
2.5.1 Overview .............................................................................................. 20
2.6 B.H.P STEEL - SLAB & PLATE PRODUCTS DIVISION,PORT
KEMBLA ........................................................................................................... 21
2.6.1 Overview ............................................................................................. 21
2.6.2 No 1 Sintering M achine.................................................................... 21
2.6.3 No 2 Sintering Machine ................................................................... 22
2.6.4 No 3 Sintering M achine.................................................................... 23
2.6.5 Port Kembla Blast Furnaces ............................................................ 23
3.0 RAW MATERIALS FOR SINTERING ........................................................ 29
3.1 IRON O R E S................................................................................................ 30
3.1.1 Overview ............................................................................................. 30
3.1.2 Origin of Iron Ores .......................................................................... 30
3.2 IRON ORES IN AUSTRALIA ................................................................ 32
3.2.1 Overview .............................................................................................. 32
3.2.2 Iron Ore Types and their Genetic Categories................................. 32
3.2.3 Iron Ore Deposits, their Characteristics and Modes of
Occurrence .................................................................................................... 34
3.2.4 Mineralogy ......................................................................................... 41
3.2.5 Clay Minerals ..................................................................................... 43
3.2.6 Silica M inerals.................................................................................... 43
3.2.7 Carbonate Minerals ............................................................................ 44
3.2.8 Iron Silicate M inerals........................................................................ 44
3.2.9 Manganese Oxide M inerals............................................................... 45
3.3 FLUX .......................................................................................................... 46
3.3.1 Overview .............................................................................................. 46
3.3.2 O rig in ................................................................................................... 47
3.3.3 M ineralogy.......................................................................................... 48
3.3.4 Im purities............................................................................................ 48
3.3.5 Classification of Flux ........................................................................ 49
3.3.6 Serpentine ........................................................................................... 49
3.3.7 Manganese Ore ................................................................................... 50
3.3.8 Dune Sand .......................................................................................... 50
3.4 FUEL ............................................................................................................ 51
3.4.1 Overview .............................................................................................. 51
3.4.2 Coke Breeze Characteristics ............................................................ 51
3.4.2.1 Size Distribution ....................................................... 52
3.4.2.2 Coke M oisture............................................................................. 53
3.4.2.3 Ash C ontent................................................................................ 53
3.4.2.4 R eactivity.................................................................................... 54
3.5 RETURN FINES ........................................................................................ 55
3.5.1 Overview ............................................................................................ 55
3.5.2 Return Fines Characteristics ............................................................ 55
3.5.3 Return Fines Effects onSintering .................................................. 56
3.5.3.1 Permeability andoperating stability ....................................... 56
3.5.3.2 Degree of Sinter Oxidation ....................................................... 56
3.5.3.3 Productivity - Hot Return Fines .............................................. 57
l
3.6 MISCELLANEOUS MATERIALS ............................................................ 58
3.6.1 Overview ............................................................................................. 58
3.6.2 Mill Scale ............................................................................................. 58
3.6.3 Brecketts S c ra p ................................................................................... 58
3.6.4 Flue Dust ............................................................................................ 59
4.0 EXPERIMENTAL .............................................................................................. 60
4.1 DESCRIPTION OF APPARATUS FOR SINTERING ........................ 60
4.1.1 Mixer ...................................................................................................... 60
4.1.2 Sintering Reactor(Pot) U n it............................................................... 60
4.1.3 B urner.................................................................................................... 61
4.1.4 Anemometer.......................................................................................... 61
4.1.5 Sintering Suction Fan ........................................................................ 61
4.2 EXPERIMENTAL SINTERING PROCEDURE ....................................... 65
4.2.1 Sintering Feed Mix ............................................................................. 65
4.2.2 Sintering Testwork ................................................................ 65
4.3 SINTER QUALITY CONTROL ................................................................. 67
4.3.1 Physical Properties .............................................................................. 67
4.3.1.1 Sinter Size Distribution ............................................................. 67
4.3.1.2 Cold Sinter Strength ................................................................... 68
4.3.1.3 Hot Sinter Strength .................................................................... 70
5.0 PRELIMINARY EXPERIMENTAL WORK .................................................. 73
5.1 EFFECTS OF IRON ORE GANGUE AND ULTRAFINES CONTENT
ON SINTERING ............................................................................................... 73
5.1.1 Overview ............................................................................................... 73
5.1.2 Sample Preparation ............................................................................ 73
5.1.3 Cyclone Classification ........................................................................ 73
5.1.4 M ineralogy............................................................................................ 74
5.1.5 Sintering Feed Materials Data ......................................................... 77
5.1.6 Sintering of the Crude Fines and Classified P ro d u cts.................. 77
5.1.6.1 Test Results and Data Analysis ................................................ 78
5.1.6.2 Test Results and Discussion ..................................................... 79
5.1.7 Effects of the Koolan Overflow Material Addition to the
Cockatoo Classified Product, on Sintering................................................. 81
5.1.7.1 Sintering Test R esults................................................................. 81
5.1.7.2 Data Analysis on test R esults.................................................... 81
5.1.8 Effects of the Cockatoo Overflow Material Addition to the
Koolan Classified Product, on S intering................................................... 82
5.1.8.1 Sintering Test R esults................................................................. 82
5.1.8.2 Data Analysis on Test Results ................................................. 82
5.2 CONCLUSIONS ......................................................................................... 87
5.3 CHARACTERIZATION OF THE RAW MATERIALS ......................... 88
5.3.1 Overview ............................................................................................... 88
6.0 BASIC EXPERIMENTALWORK .................................................................. 89
6.1 Overview........................................................................................................ 89
6.1.1 Correlation between the Port Kembla Experimental Reac
tor Unit and the Kwinana Sintering Plant .............................................. 90
6.2 Some Fundamental Facts ............................................................................ 94
6.2.1 Effects of Ore Feed Size Distribution on Sintering ....................... 94
6.2.2 Effects of Limestone Flux and Coke Breeze Size
Distribution on Sintering.............................................................................. 95
6.3 Development of Sintering Process Models .............................................. 104
6.3.1 Overview ......................................... 104
6.3.2 Sintering Process Model I ................................................................... 105
6.3.3 Sintering Process Model II ................................................................. 105
6.3.4 Sintering Process Model III ............................................................... 105
7.0 EFFECTS OF BED DEPTH ON SINTERING ............................................. 107
7.1 Overview......................................................................................................... 107
7.1.1 Presentation of Data and R esu lts.................................................... 110
7.1.2 Interpretation of Data and Results .................................................. 117
7.2 CONCLUSIONS ....................................................................................... 125
8.0 EFFECTS OF BASICITY ON SINTERING ............................................... 127
8.1 Overview........................................................................................................ 127
8.1.1 Presentation of Data and R esults.................................................... 131
8.1.2 Interpretation ofData and Results .................................................... 131
ii
8.2 CONCLUSIONS 140
9.0 EFFECTS OF SUCTION ON SINTERING .................................................. 142
9.1 Overview ....................................................................................................... 142
9.1.1 Presentation of Data and Results .................................................. 142
9.1.2 Interpretation of Data and R esults................................................ 142
9.2 CONCLUSIONS ....................................................................................... 151
10.0 SINTERING PLANT TRIALS ..................................................................... 153
10.1 Overview ..................................................................................................... ¡53
10.1.1 Pre-Trial Sintering Plant Operating Conditions and R esu lts... 154
10.1.2 Port Kembla Sintering Plant Trial and R esults......................... 155
10.2 CONCLUSIONS ....................................................................................... 167
REFERENCES ...................................................................................................... 168
APPENDICES
A. PRESSURE AGGLOMERATION PROCESS ............................................... 175
B. NODULIZING PROCESS ................................................................................ 177
C. DEVELOPMENT OF THE PELLETIZING PROCESS ................ 178
C.l Overview ...................................................................................................... 178
C. 2 Pelletising Plants in Australia ............................................................... 182
C.2.1 Savage R iv e r...................................................................................... 182
C.2.2 Hamersley............................................................................................ 182
C.2.3 Whyalla .............................................................................................. 183
C.2.4 Robe River_.......................................................................................... 183
D. DEVELOPMENT IN SINTERING OF NON-FERROUS ORES IN
AUSTRALIA ........................................................................................................... 185
D. l Overview ...................................................................................................... 185
D.2 Fundamental Studies and Developments inSintering ........................... 186
D.3 Development of the Era of Large Sintering Machines.......................... 187
D.4 Development of the Up-dranght Sintering Technology........................... 188
E. DEVELOPMENT OF SINTERING OF FERROUS ORES IN AUSTRA
LIA B.H.P STEEL-SLAB & PLATE PRODUCTS DIVISION, PORT
KEMBLA...................................................................................................................... 192
F. B.H.P-ROD & BAR PRODUCTS DIVISION, NEWCASTLE SINTER
PLANT ................................ 199
F.l Overview................ 199
F.2 Sintering Process ....................................................................................... 199
F.3 Basic O perations........................................................................................ 200
F.4 Major Sintering Plant Equipment .......................................................... 201
F.5 Typical Sintering Machine Feedand Operating Parameters ............... 201
F .6 Quality Control .......................................................................................... 202
F. 7 Blast Furnace Ferruginous Borden ......................................................... 204
G. AUSTRALIAN IRON & STEEL PTY. LTD. - KWINANA STEELWORKS
...................................................................................................................................... 205
G. l Overview.......................................... 205
H. FLUX ................................................................................................................. 206
H. l Origin ( Continued ) ................................................................................. 206
H.2 Geochemical Evolution .............................................................................. 206
H.3 Chemical Properties .................................................................................. 207
H.4 Physical Properties .................................................................................... 208
H.5 Classification of Lim estone..................................................................... 208
H .6 Limestone U ses........................................................................................... 209
111
ABSTRACT
Experimental studies on the effects of bed depth, basicity and suction on sintering at
Australian Iron & Steel Pty Ltd were successfully carried out and introduced through
Sintering Plant trials.
Two new Sintering Process Models were developed during the experiment. The
effectiveness of the second model was verified by its producing an orderly positioning
of each combination of parameters, particularly, in the test series to study the effects
of suction on sintering. This was a significant contribution to sintering technology.
An improvement in productivity and sinter quality, and a marked changes in the

sinter microstructural composition and texture were achieved.
As a result:
. In terms of sinter microstructural composition, Port Kembla sinter achieved the
same level as Japanese sinter.
. The sinter mineralogy changed from a high magnetite-glass bond to high calcium
ferrite bond. The texture changed from a homogeneous, highly fused sinter to
heterogeneous, less fused and more porous sinter.
. Fuel consumption was significantly reduced, and sintering temperatures

decreased, with a resultant lowering of costs.
. The primary aim of producing an improved blast furnace burden material was
accomplished.
1.0 INTRODUCTION
1.1 Overview
Sintering as applied to ores, is the formation of a coherent bonded mass of fine ore
particles by applying thermal energy generated through the combustion of coke. This
heat energy brings fine particles into a "solidus and liquidus reaction" at a temperature
below the melting point of the nucleus material.
Fine ore particles exhibit characteristics which are to some extent undesirable for a
direct application in smelting, and the sintering process is a means of agglomerating
iron ore fines which would otherwise be unsuitable for blast furnace consumption.
Fig 1-1 shows the introductory concept of raw materials proportioning, sintering
process and the product sinter texture; and Fig 1-2 illustrates the sintering plant
process schematic.
A systematic literature study of why, when and how agglomeration technology was
developed is presented in the first chapters of this thesis, for the following reasons:
a) Sintering is a highly complex process, so a broad introductory knowledge is

necessary.
b) The properties of the sinter product play a substantial role in the high-temperature
reduction process that takes place in the continuous smelting operation in blast
furnaces.
c) Sinter has an important influence on the costs of blast furnace production, which
is still the most economic process in ironmaking.
The properties of sinter are determined by two sets of factors:

. The properties of raw materials, and
. The sintering plant processing parameters.
Major properties of the raw materials are the chemical composition and physical
properties such as size distribution, mineralogy, structural texture and porosity.
3 0009 02982 2728

Major plant process parameters are the proportioning of raw materials, granulating
and feeding conditions, bed heights, ignition conditions, suction during ignition and
sintering, and sintering temperatures.
Since plant parameters can not usually be varied greatly, the most important factors
affecting sinter quality and productivity are the raw materials and their properties.
For this reason, a review of selected literature about raw material characteristics has
also been included in this thesis. These characteristics are:
. Ore types and their genetic categories.
. Ore deposits, their characteristics and modes of occurrence; e.g.,

friable, hydrous, hard blocky, hard massive; friable granular, soft, porous; and
conditions of association with impurities.
. Geochemistry; the physical and chemical processes which have produced the
observed distribution of the elements in the phases.
. Mineralogical and textural structure of the mineral deposits; e.g., hematite from
soft earthy to rock crystalline, magnetite, limonite, etc.
The systematic interpretation and correlation of these characteristics, added to the

most important parameter, the size distribution of the raw material, has been a key
factor in the better understanding of the specific behaviour of each raw material type
during sintering.
Sinter is a major component of blast furnace burden materials, so it is important that

both the raw material properties and the sintering plant process parameters be adjusted
to blast furnace requirements. These include cold state properties of sinter strength,
size distribution, chemical composition, mineralogy and structural texture, as well as
size degradation of sinter during reduction (RDI. -3mm% index) in the low temperature
zone (400°C-600°C); and the softening under-load (S.U.L) property of sinter in the
high temperature zone (1200°C-1500°C).
2
1.2 Scope
During the engineering of this research work, a deeper approach has been taken, to
characterize each raw material by a qualitative and quantitative description of its
morphological parameters (size of microstuctures), the homogeneous and heterogeneous
textures, and the distribution of metamorphosed gangue minerals (Figs 5.1-4 to 5.1-19).
These observations were of significant assistance in understanding why limestone

fluxes with identical chemical composition behave quite differently during sintering,
or why sinter products with the same chemical composition are different in physical
properties.
For example, the same sintering feed mix may be processed one day at lower or higher
sintering temperature, and the next day at the aim temperature, resulting in sinters
with the same chemical composition but different in physical properties. In addition,
sinter usually contains unreacted and partially reacted particles, which reflect, totally
or partially, their original properties during blast furnace operation.
Preliminary research was carried out to improve the established sintering process
parameters, and included the following topics:
i. Optimization of granulation of the green sintering feed, with the aim of

improving permeability and coke combustibility during sintering. This study
was conducted in the laboratory by comparing the performance of two types
of mixers, looking at mixing time, angle of mixing, number of revolutions per
minute and the moisture content of the mixture etc.
ii. Optimization of the sintering reactor unit (pot) by designing a cylindrical reactor
with temperature and suction sensors and with efficient insulation to minimise
the heat loss; and
iii. Optimization of ignition conditions, which are related to the type of burner,
ignition time and suction level during ignition.
The basic raw materials investigated were:

i. Various types of iron ore fines,
ii. Coke breeze,
3
iii. Various types of fluxes, and
iv. Return fines.
These raw materials were investigated under constant sintering process parameters,
in order to determine the effects on sintering. The following variables were
investigated:
i. Mineralogy and structural texture
ii. Size distribution
iii. Selection of different types of flux by close examination of the chemical and
physical properties and economic considerations.
iv. Coke breeze levels in the sintering feed mix
v. Return fines levels, and
vi. Proportioning of the different types of iron ore fines.
The fundamental experimental work was conducted using a projected sintering plant
feed mix, and the following effects on sintering were investigated:
i. Bed depth, 250 to 550 mm.
ii. Suction, 9 to 15 kPa; and
iii. Basicity, 1.2 to 2.8 CaO/Si0 2 ratio.
1.3 Sintering Plant Trials.
Once optimum parameters were established in the experimental work, the first plant
trial was designed to operate for two days, but was terminated by the operators after
eight hours. The changes in the product sinter quality and productivity were
insignificant.
However, after evaluation of this trial and further discussion, management approved
a revised plan for a one week uninterrupted plant trial with the condition: "If the
experimental mode shows a significant contribution to sinter quality and productivity
during the trial period, the sintering plant will indefinitely continue operating at
these conditions. Otherwise, trials will be suspended and normal operating conditions
resumed".
In the belief that expert visual observations are the simplest and fastest system of
communication, and instrumental tests are the proof of the facts, it was arranged that
4
the trials should be observed at all times by a technical staff member of the Ores &
Sinter Research Section, as hourly communicator and supervisor of raw materials and
product sinter hourly sampling.
Plant trials were conducted during the second stage using the No 2 Sintering Machine
on a 24 hour continuous operation basis.
Plant operating conditions were:

Operating Factors Normal Experimental
Basicity, CaO/Si0 2 1.10 1.55
Bed height, mm 254 457
Suction, kPa 9 9 to 10
The results in summary, included:

a) Visual observations
Hours 0-3, Highly fused sinter, no significant difference compared to normal
sinter.
Hour 4, Sinter predominantly lumps and less fused.
Hour 5, Sinter containing unreacted particles and less fused.
Hour 6, Sinter much less fused and more porous. Remarks for the remainder
of the trial were similar.
b) Chemical analysis:
Determinations Hours 0-3 Hour 4 Hour 5 Hour 6
MP 62 26 24 22
FeO % 15.8 7.1 7.1 6.9
c) Physical properties:
- 6.35 mm % 22 19 16 13
Rumbler index % 68 73 75 74
d) Mineralogy:
Hematite % 13 22 43 42
Magnetite % 62 29 25 19
Ca-Ferrite % 6 35 21 30
Glass % 19 14 11 9
MP = Magnetic permeability.
Encouraged by these results, management approved extensive trials at the Sintering

Plant for three months, involving all blast furnaces at Port Kembla Steelworks.
5
Comments by Port Kembla sintering plant operators at the end of the three months
plant trial were:
"Experimental work in pot testing and the study of microstructures by Ores & Sinter
Research Department, indicated that it was possible to produce a high calcium
ferrite-glass bond type of sinter. Furthermore, the plant trial demostrated that the
same effect could be achieved on the plant, using the fines normally encountered
here" [52,18] .
1.4 Summary of Research Contributions
1. The most significant contribution was the change in the mineralogical composition
and the microstructural texture of the product sinter. The deep bed sinter
mineralogy changed from a high magnetite-glass bond to highly calcium
ferrite-glass bond. The texture changed from homogeneous, highly fused sinter
to heterogeneous, less fused and more porous sinter. This was using Kwinana,
Port Kembla and Newcastle reference sinters [52] [53].
2. The rumbler index improved, from the usual 66-67 up to 75-76, indicating a
significant improvement in sinter strength.
3. A marked decrease in FeO content, hence, a marked increase in the degree of

sinter reducibility, was achieved.
4. The total return fines loading dropped from 38% to 27% during the trial. As a
result, sinter productivity was increased.
5. A significant reduction in fuel consumption was achieved, showing a potential

financial saving.
6. Sintering operating temperatures were reduced, with a resultant lowering of costs

for maintenance of hot screens, coolers and product handling conveyor belts.
6
FIG. 1-1 THE SINTERING PROCESS
RAW MATERIALS FUEL

OLD SINTER
HOMOGENEOUS
TEXTURE
IRON ORE SINTER
FINES RETURN COKE
FINES OVENS
GAS
—6.4m m * 0
PRODUCT
FUEL SINTER
COKE LUMP
FLUX BREEZE
KINETIC ENERGY
A + A
MODERN SINTER
—3m m * 0 —5.6m m *
HETEROGENEOUS
TEXTURE
FIG. 1-2 SINTERING PROCESS SCHEMATIC
2.0 D EVELO PM EN TS IN AGGLOM ERATION P R O C ESSES
2.1 Overview
In general terms, agglomeration is the process of consolidating fine particles into large
lumps to impart suitable properties for subsequent use or for further treatment.
Fine particles exhibit characteristics which are undesirable for many direct appli
cations. Because of this, they are agglomerated using a large variety of techniques.
Agglomeration products are extensively used in medicine, food technology, chemical

engineering, metallurgy, agriculture and other fields.
2.2 Methods of Agglomeration
The technology of the consolidation of fine particles is carried out at conditions

varying from cold, ambient to moderate and high temperatures, through either purely
physical action, chemical reaction or a combination of both.
Four basic methods are outlined below (although another important process, crys
tallization of dissolved materials, may be considered an agglomeration method):
i. Pressure agglomeration by
.Roll presses
.Isostatic presses
.Extrusion
.Tableting
.Briquetting, etc.
ii. Nodulizing by rotary kiln.
9
iii. Pelletizing by
.Shaft furnace
.Straight grate
.Grate-Kiln.
iv. Sintering process by

.Huntington-Heberlein (H&H) batch updraught,
.Greenawalt batch downdraft,and
.Dwight-Lloyd continuous downdraught and updraught.
Details of pressure agglomeration, nodulizing and pelletizing processes are presented

in Appendices A, B and C respectively.
2.3 The Sintering Process
2.3.1 Overview
The practice of sintering initially was developed in the non-ferrous industry in the
1890’s. The exhaustion of suitable oxide ores, and increasing intrusion of mixed lead
and zinc sulphides into the mine orebody brought the need for beneficiation of the
leaner sulphide ores.
The introduction of froth flotation technology required some form of pre-treatment

of the fine concentrates, and there were two things necessary to be done with the
concentrates to prepare them for blast furnace smelting:
i. reduce the sulphur content by roasting;
ii. convert the roasted concentrates to a lump form suitable for smelting.
The alternative briquetting process, in which roasted concentrates were mixed with
some finely divided fluxes, overcame some difficulties in the smelting operation; but
the fragile characteristics of the briquettes led to slow furnace operation and poor
lead recovery until the invention of the Huntington-Heberlein pots or "converters" in
1896.
10
The need to develop more effective types of sintering devices during the first half
of the 20th century led to the Dwight-Lloyd process, invented in 1905. This became
the most successful sintering machine.
By the beginning of the 1920’s the iron and steel industry faced the need to use lower
quality lump ores, as high quality iron ore deposits became exhausted. Finer and
leaner lump iron ores were increasingly used in blast furnaces. The poor performance
of the burden materials generated higher tonnages of flue dust per tonne of iron and
lowered pig iron production.
In addition to this almost world-wide problem, there was already a large accumulation
of flue dust and other ferruginous dusts around the steelworks, for which an economic
use was required.
These factors were the first driving forces for the development of the agglomeration
process. At this stage, the sintering process was mainly a scavenger process geared to
the agglomeration of stocks of flue dusts and pyrite residues. For this reason, the early
sintering plants did not produce high quality ferrous-sinter, and sinter proportions
in blast furnace burden were limited.
However, explorations after World War II resulted in the discovery of rich iron ore
deposits in Australia, Africa and South America, far from the main steel producers.
This radically changed the raw material picture for the world’s iron and steel industries
at the beginning of the 1950’s, particularly in Japan and Italy, countries solely
dependent on iron ore imports.
The export of high quality iron ore lumps with a specific size distribution requires
crushing and screening operations, from which large quantities of low cost iron ore
fines are generated. This influenced not only sintering and pelletizing technology,
but also methods of concentrating low grade iron ore fines, particularly on the
American continent, in Europe and in Japan.
During the second half of the 20th century, there were three important events in
sintering development:
11
1) . Sintering plant construction and production had declined in the U.S.A since the
early 1960’s. Pelletizing production increased, achieving an average of 58% as blast
furnace burden material, for three reasons:
i. Most of iron ore deposits in the U.S.A. are low grade magnetite ore, and they
are located far from steelworks.
ii. Very fine ore concentrates are not suitable for sintering, and sinter is not
suitable for long distance transportation.
iii. The majority of sintering plants in the U.S.A. used a high percentage of iron
bearing waste oxide materials, that are by-products of iron and steelmaking
processes [6].
Furthermore, the general consensus in the U.S.A. was that the older sintering plants
would remain in operation at reduced tonnage or be upgraded for high tonnage without
replacing the strands [7].
2) . Significant technological improvements in the sintering process were achieved in

Europe, Japan and Australia, where an average of 68% sinter in the blast furnace
burden was used at the end of the 1960’s through the 1970’s, increasing to an average
of 80% during the 1980’s.
Modern sintering plants located at steelworks provided :
- Economic gains, as sinter production cost was reduced to one-third to one-fourth

that of pellets [6,7].
- An effective fully beneficiated blast furnace feed, and
- A significant saving in blast furnace fuel cost, especially after the oil price
crisis in 1973 and the rapid increase in the price of coking coal.
It is also interesting to note that 60% of European and 62% of American ironmaking
plants operated using acid to self-fluxed (0.61 to 1.3 CaO/Si02) sinters, but that the
Japanese ironmaking plants have operated since the 1960’s using self-fluxed to
super-fluxed (1.3 to 1.8 CaO/Si02) sinters, as have Australian plants since 1972.
12
3). Attention to blast furnace burden materials commenced in the 1950‘s, increased
through the 1960’s and by the 1970’s was of major interest, in Europe and particularly
in Japan and Australia. Technologies examined were:
i. Careful selection of raw materials and preparation of the sintering feed mix
based on studies of their characteristics during sintering and of the product
sinter performance as blast furnace burden material.
ii. Improved plant design for higher productivity and better environmental
conditions.
iii. Use of local scientific and technical personnel and the exchange of expertise
via overseas technical missions.
iv. Computerised operating techniques and constant reviews and training of plant
operational and technical personnel; and
v. A management system based on fundamentals of economic viability and

profitability.
As a result, the ironmaking process significantly reduced its coke rate (kg/tHM), and
increased its productivity. Hence, the iron and steel industries survived the 1982
world wide economic recession, and currently they are gradually recovering
approximately one-third production capacity loss.
2.4 DEVELO PM ENT OF T H E SINTERING PRO CESS
2.4.1 Overview
The development of modern sintering devices began with the invention of the Hun
tington-Heberlien (H&H) process in 1896, and was followed by the Greenawalt,
Schlipenbach, Dwight-Lloyd, and other sintering machines.
The continuous down-draught Dwight-Lloyd (D&L) process was invented in U.S.A. in

1905. However, the patents were purchased and improved by Metallbank und
Metallurgischen Gesellschaft of Germany (currently Lurgi Chemie und Huttentechnik
G.M.B.H., Frankfurt, Germany). This process has achieved prominence for the sintering
13
of iron ore fines.
The continuous up-draught D&L process has specialised application in the non-ferrous
ores: in the case of sintering lead, zinc, and copper sulphides, it enables better con
centration of sulphur dioxide for sulphuric acid production, with minimum lead loss
by sublimation [8].
2.4.2 The Huntington-Heberlien Process
The H&H process consisted of a line of cast iron pots or "converters”, each pot receiving
approximately 10 tonnes of sintering mix. The charge was poured over an igniting
bed of hot coke and the influence of an up-draught or upward blast of air generated
sufficient heat for sintering the material into a hard semi-fused mass that had to be
broken up with sledge hammers.
This sinter product was a considerable improvement, both physically and chemically
over the weak friable briquette product [9].
2.4.3 The Greenawalt Process
The Greenawalt process was a batch down-draught operation in which the sintering
feed mix was held in a bin, and pans were individually filled from a roll feeder.
The filled pan was placed over a windbox, where suction was applied and the surface
was ignited. On completion of sintering, the pan was inverted to discharge the sinter
cake, and it was then returned for filling [ 10].
Both the Greenawalt and the Dwight-Lloyd processes employ the same principles.
Sintering is practicable as either batch or continuous process. For large tonnages, the
continuous D&L process is unquestionably more economical and technically more
controllable.
Typical Greenawalt and Dwight-Lloyd Sintering Plants
Full-scale Greenawalt and D&L sintering plants at the Appleby-Frodingham Steelworks

in the UK showed that output was closely related to average air flow through the
strand, and the sinter characteristics were similar. The Pennsylvania Steel Co also
had Greenawalt plants at Steelton and Sparrows Point, and a D&L plant at E&G
Brooke Iron Co at Birdboro in the U.S.A [10].
14
It is interesting that a normal Greenawalt design plant built in 1934, at Domnarvet,
Sweden, produced about 700,000 tonnes of sinter per year, using Swedish concentrated
iron ore fines.
The plant characteristics were:
. Six pans, each 7.9m x 3.0m

. Depth of the bed 305mm
. The time between successive charges was 31 to 32min, of which about:

- two minutes were used for charging, and
- four minutes for ignition.
From 1915, when the first sintering plant in Sweden was built, the average of sinter
in the burden of Swedish blast furnaces is illustrated in Table 2.4-1 [11].
Table 2.4-1 Blast Furnace Ferruginous Burden in Sweden, 1920 to 1950.
Year Sinter burden %

1920 10
1930 25
1940 82
1950 89
Sinter Analysis % (1950)
Fe 51.1 MgO 1.8
CaO 11.5 P205 2.6
Si02 7.5 Mn 0.8
CaO/Si02 1.53 S 0.01
Degree of oxidation 98.2
15
2.4.4 The Down-Draught Dwight-Lloyd Sintering Process
The process of sintering is today the most important and widely used method of
agglomerating iron ore fines. Principles and methods of sintering are well established,
but are subject to continuous development and search for improvement.
In the sintering process, the sintering feed, thoroughly mixed, moistened and granu
lated, is loosely fed on to a grate of a continuous sintering machine, in a layer of
440mm to 530mm thick.
Combustion is initiated by igniting the fuel on the surface of the ore mix, and it is
maintained by drawing air through the charge; combustion proceeds in the direction
of the gas stream and causes the sintering of the charge.
The combustion proceeds in a thin horizontal layer, moving vertically through the
charge. This is called the "differential layer", the thickness of which corresponds to
only a fraction of the total thickness of the charge layer. This is characteristic of
down-draft sintering.
A prerequisite for effective sintering is satisfactory gas permeability of the charge.

However, it is not sufficient for the ore mix to have a high gas permeability only in
a moistened state. The sintering feed mix must also retain this condition after drying,
as well as in the state of softening and at the beginning of fusion.
Above the combustion zone, there is a layer of highly heated sinter material (hot
sintering zone), that is cooled by fresh air which, preheated in this way, arrives at
the "combustion zone". Below this zone, the heat content of the waste gases is absorbed
almost completely by the lower layers (cold charge), so that the cold layers are dried
and preheated.
The combustion air and charge meet in a highly preheated state in the "combustion
zone", and the temperatures produced (1000°C-1250°C) result in most cases in an
immediate melting in the material. High thermal efficiency is therefore produced by
a heat accumulation within a partial layer of the charge, the "sintering zone". This
sintering layer travels at 10-30mm per minute towards the grate.
16
The combustion process with down-draught sintering, therefore, has a character
different from a conventional combustion process. In sintering, heat transmission
within the charge takes place largely by the gas passing through the bed owing to the
large particle size of the granulated material. Hence, a large portion of the waste-gas
and combustion heat is utilised directly for the sintering process itself [ 12].
Table 2.4-2 shows typical D&L sintering plants at UK during the first half of the
20th century [13].
Figs 2.4-1 and 2.4-2, show the material flpw for the downdraught D&L sintering
process of iron ore fines, and a schematic diagram of the automatic control system
[15] [16].
Development in sintering of non-ferrous ores in Australia is outlined in substantial

detail in Appendix D.
17
Table 2.4-2 Typical Dwight-Lloyd Sintering Plant at U.K. during
the first half of the 20th century [13].
Dagenham Cleveland
Built in, 1932 1949
Strand, single single
Length, 15.85m 32.92m
divided into 8 windboxes divided into 16 windboxes
Width, 1.83m 1.83m
Bed height, 305mm (max) 330mm (maximum)
Grate area, 28.46m2 60.26m2
Iron ore screened to, n/a -6.4mm
Coke breeze crushed to, n/a 70% <-1.59mm
Strand speed,
variable: 1.52m/min (max) 2.32 m/minute (maximum)
Fan: 16-bladed paddle type a double box-paddle type machine
Driven by 425 h.p. motor 1050 r.p.m. motor
Rated at 2549m3 S.T.P./min 3540 m3 per minute
R.P.M. 1450 750 r.p.m.
Suction, 6.23 kPa 6.23 kPa
at an efficiency of 70% The motor was directly coupled to the fan, and
.No bedding layer was used. air volume was adjusted by vane control.
.The pallets were pushed .The wind legs were connected to a common
along the strand by the main.
action of the head pulley .The sinter was discharged down a chute to a
and along lubricated slides rotary shaft breaker with spaced claws, which
to seal the air. emptied periodically into a ladle, and sprayed
.At the end of the strand, and cooled. It was then screened on a 25.4mm
the sinter fell down a grizzly and through 7.9mm vibrating screens.
chute, over stationary bars
12.7mm apart, directly into Code:
a transfer car. Then the n/a = not available.
sinter was quenched by
water sprays while in the car.
18
Fig. 2.4-1 A typical modern Sintering Plant material flow incorporating
a pre-blended sintering mix and product sizing [7].
ORE BINS riOlOR

GENERATOR REGULATOR
Fig. 2.4-2 Schematic diagram of automatic control system on the

downdraught Dwight-Lloyd type sintering machine for
iron ore [80].
19
2.5 D EVELO PM EN T OF SIN TERIN G OF FERROUS O RES IN
A U STR A LIA
2.5.1 Overview
In 1897, The Broken Hill Proprietary Company Limited (B.H.P.) acquired iron ore
mine leases in the Middleback Ranges for the purpose of using this ore as a flux in
smelting the Broken Hill lead ores at Port Pirie. These orebodies contained large
quantities of high grade hematite.
The discovery of the Middleback Ranges iron ore deposits, about 40 km west of
Whyalla, South Australia, and the abundance of coking coal in N.S.W., led B.H.P to
make the decision to change from the non-ferrous to iron and steelmaking industry.
As the initiator of this new enterprise, B.H.P. established its first steelworks at
Newcastle, N.S.W., in 1915.
By the beginning of the 1950’s, there was a world wide trend to develop usage of iron
ore fines and concentrates. This became imperative, for three main reasons:
i. A marked expansion of the iron and steelmaking industry,
ii. The decrease of high quality lump ore reserves, and
iii. The significant production increase of inexpensive iron ore fines.
This trend has been responsible for a change in the status of the sintering plant, from
a simple, relatively unimportant unit, used predominantly for the reclamation of blast
furnace flue dust along with some ore fines, to a large, fully mechanised installation
integrated with comprehensive ore handling, screening and blending facilities [16].
B.H.P. Steel, as Australia’s only fully integrated iron and steelmaking industry, has a
high dependence on properly prepared blast furnace burdens. Hence, iron ore sinter
forms the highest proportion of the blast furnace ferruginous burden [17].
20
Currently agglomeration of iron ore fines by sintering is carried out at two of the
three B.H.P. Steel ironmaking centres: at Port Kembla, the largest centre, and Newcastle,
both in N.S.W.
2.6 B.H.P S T E E L - SLA B & PLA TE PRO D U CTS DIVI

SIO N ,PO RT KEM BLA
2.6.1 Overview
In 1943, a 1.07m sintering machine discarded by Port Pirie, was installed at the former
Australian Iron and Steel Pty. Ltd. Port Kembla Steelworks, N.S.W., for the prime
purpose of sintering flue dusts. The product sinter was discharged over a fixed grizzly
screen with 38.1mm aperture. The oversize was discharged directly into rail wagons,
where it was cooled by water spray. This plant was closed in 1945, due to a shortage
of labour.
In 1950, all B.H.P. blast furnaces suffered serious operating difficulties with the
introduction of Cockatoo ore, because of its friable nature. In addition, plans were
made to develop the Koolan ore, which was also known to have a high fines content.
Under these circumstances, an agglomeration process became imperative, and a decision

was made to use the Dwight-Lloyd sintering process, after examination of overseas
practice in 1954.
2.6.2 No 1 Sintering Machine
The No 1 Sintering Machine, including all buildings and bins was designed, supplied
and erected by Lurgi Gesellschaft fur Chemie und Hüttenwesen G.m.b.H. of Germany.
The belt conveyors, fine ore and miscellaneous materials storage bins, were designed
and mechanicals supplied by the Fraser and Chalmers division of B.G.E. Co. Ltd. of
England.
The No 1 Sintering Machine was based on the down-draught D&L process. This was,
21
when designed, the largest machine in the world for ferrous sintering [18].
The No 1 Sintering Machine, with a grate area of 100m2, was commissioned on 21

January, 1957, and ceased operation on 8 February, 1975. The total sinter produced
was 13,249,518 tonnes [20].
Tables E-l, E-2 and E-3 (Appendix E.l), respectively show the No 1 Sintering Machine
equipment description; guaranteed production figures; and the physical and chemical
analyses of major raw materials [18] [19].
The No 2 Sintering Machine, including pan conveyors, sinter screening station and
cooler, was designed and built by Lurgi Gesellschaft fur Chemie of Germany.
The structural sections of the plant were designed and fabricated by Improved
Constructions Ltd. The erection of the plant was carried out by the Australian Iron
and Steel Pty. Ltd., with the assistance of Lurgi engineers.
The dedusting system of the dry electrostatic type was designed and supplied by Lurgi,
with the fans, ductwork and certain other parts being supplied by local firms. This
system of dedusting was extended to the No 1 Sintering Machine as well.
The No 2 Machine was at that time the largest in the world for ferrous sintering,
based on the down-draught D&L process.
This Machine, with a grate area of 176m2, was commissioned on 22 April, 1960, and
temporarily stopped in July 1975. It was recommissioned on 17 April, 1981, and ceased
operation on 15 September, 1981, after a total production of 23,461,815 tonnes sinter
[20 ].
Table E-4 (Appendix E) indicates the No 2 Sintering Machine major equipment data;
ore handling and screening plant, and miscellaneous materials handling and coke
crushing plants.
22
The No 3 Sintering Machine with a grate area of 420m2, commissioned on 22 February

1975, incorporated much of the technologies developed over the previous years of
active research, setting high quality and production standards, including the following
record levels [20]:
19 Apr 1984, the lowest reduction degradation (RDI), 24.7 %

29 Sep 1986, the lowest FeO content, 3.92 %
07 Oct 1987, the highest reducibility index (RI), 69.5 %
29 Nov 1989, the highest daily production rate, 15855 tonnes.
A significant factor was also the installation of preheating of the ignition hood air
by recirculatin the sinter cooler hot air on 14 April 1987. Another landmark was the
total production of 60,766,036 tonnes sinter to 31 October 1990.
Table 2.6-1, provides essential details of the plant equipment for Nos 2 and 3 Sintering
Machines (No 2 Sintering Machine with an enlarged grate area of 200m2) [20]. Further
details are presented in Appendix E (Table E-5).
Tables 2.6-2, 2.6-3 and 2.6-4, show the No3 Sintering Machine operation data; sinter
quality, and raw materials chemical and physical analyses respectively.
2.6.5 Port Kembla Blast Furnaces
The sintering capacity indicates the extent of the dependence of ironmaking operations
on prepared burdens. Sinter product as a major burden component not only has a
vital role in a better blast furnace performance, it also lowers the iron production
cost by this high-temperature reduction process through the continuous smelting
operation in the reactor-blast furnace.
Table 2.6-5, indicates the ferruginous burdens used in the current Port Kembla blast
furnaces [21].
Sintering at B.H.P, Rod and Bar Products Division - Newcastle is presented in Appendix
F (F.l to F.7).
23
Table 2.6-1 Sintering Plant Equipment Description [20].
No 2 MACHINE No 3 MACHINE
Raw Material Bins

Number and Size 12 x 350 m3 12x350m3, 1x300 m3,
1 x 50 m3
Mixing/Rolling Drum
Dimension (L x 0): 13.25 x 3.8 m 25 x 4.4 m
Rotational Speed: 5.4 rpm 5.5 rpm
Residence Time: 4.5 min 4.5 min
Inclination: 1.7 degrees 4.5 degrees
Design Capacity: 515 tph 1,185 tph
Sinter Strand
Type: Lurgi Lurgi
Dimensions (W x L): 4 m x 50 m 5 m x 84m
Design Capacity: 27.5 t/m 2/d 40 t/m 2/d
Suction Area: 200 m2 420 m2
Suction: 10 kPa 16 kPa
Bed Height: (typical) 400 - 450 mm 480 - 530 mm
Feed Roll Diameter: 1.3 m 2 m (outside diam)
Ingition Hood
Dimensions (W x L): 4mx5m 5mx 3m
Burner: Urquart, three Direct impingment
per side (Kawasaki line burner)
Primary Crusher
Type: Single Spike Roll Single Spike Roll
Capacity: 600 tph 1140 tph
Cooler
Type: Straight Line Grate Circular Grate
Effective Cooling
Area: 210m2 (3.5m x 60m) 380 m2
24
Table 2.6-2 Port Kembla No 3 Sintering Machine Operating Data.
Materials, 1990 Jan. Feb. Mar. Apr. May.

Blended fines % 59.8 59.6 59.6 59.1 60.1
Limestone % 10.5 11.6 11.5 10.1 10.1
Dolomite % 1.7 1.9 0.9 2.6 2.0
Serpentine % 1.2 0.9 0.5 0.6 1.0
Dune sand % 0.1 0.1 0.2 0.2 0.1
Manganese ore % 0.2 0.2 0.2 0.3 0.3
Coke breeze % 3.8 4.0 4.0 3.9 3.9
Return fines % 22.1 21.2 22.8 23.0 22.3
Quick lime % 0.6 0.5 0.3 0.2 0.2
Total % 100.0 100.0 100.0 100.0 100.0
Operation
Mix moisture % 5.6 5.7 5.7 5.6 5.8
Limestone, kg/t-S 162.0 180.2 180.1 161.6 157.0
Dolomite, kg/t-S 26.1 28.2 12.8 40.9 31.3
Serpentine, kg/t-S 17.1 14.3 20.0 8.8 15.5
Coke rate, kg/t-S 54.0 56.3 56.5 56.5 56.3
C.O.gas, Nm3/t-S 1.4 1.5 1.5 1.5 1.2
Total fuel, kl/t-S 58.8 59.7 59.0 59.2 58.0
Bed height, mm 547.6 534.4 475.4 478.6 542.6
Strand speed, m/min 3.4 3.3 3.3 3.3 3.2
Suction WB1-A, kPa 12.4 12.0 9.4 8.8 11.2
Ign. hood temp. °C 729 715 794 831 850
Production
Production t/month 394084 362573 341807 352296 390494
t/day 12712 12949 11026 11743 12597
Prod, rate t/h r 585.1 576.2 507.4 526.2 593.1
Productivity t/m 2/d 33.4 32.9 29.0 30.1 33.9
Yield % 85.6 85.1 85.7 85.2 85.7
Availability % 90.6 93.4 86.8 92.9 87.7
Reliability % 93.3 94.2 94.4 96.8 92.7
25
Table 2.6-3 Port Kembla No 3 Sintering Machine Operation Data.
1990 Jan. Feb. Mar. Apr. May.

Energy Consumption
Main fan kwh/t-S 21.3 21.0 20.1 17.5 21.8
MJ/t-S 76.8 75.6 72.5 63.1 78.4
Preheat MJ/t-S 49.6 50.9 54.8 55.1 49.0
Annealing MJ/t-S 20.3 18.6 21.0 20.8 18.3
Ignition gas MJ/t-S 28.2 28.7 29.7 28.7 30.2
Ign. air+gas Mj/t-S 30.5 30.9 32.1 31.0 32.2
Sinter Quality
Analysis
Fe % 55.2 56.2 55.7 55.5 55.4
FeO % 6.8 6.6 6.6 6.7 6.6
S i0 2 % 5.9 5.8 5.8 5.7 5.6
Al2Og % 2.2 2.0 2.0 2.1 2.1
CaO % 11.1 11.4 10.8 10.9 11.0
MgO % 1.6 1.6 1.4 1.6 1.6
Mn % 0.4 0.5 0.5 0.5 0.5
CaO/Si0 2 1.9 2.0 1.9 1.9 2.0
Sizing
+ 40 mm % 5.6 5.2 5.2 5.3 4.8
+ 10 mm % 59.2 58.9 59.4 58.1 58.3
+ 5 mm % 93.8 93.8 94.3 93.7 93.9
AMD mm 16.7 16.5 16.7 16.5 16.2
SMD mm 9.9 9.8 10.0 9.7 9.8
Quality Indices
RDI, -3mm % 39.2 36.9 39.0 37.9 35.9
Reducibility % 60.6 61.5 61.3 61.1 61.4
26
Table 2.6-4 Port Kembla No 3 Sintering Machine Raw Materials Data.
Blended Feed
Analysis, 1990 Jan. Feb. Mar. Apr. May.
Fe % 61.2 61.6 61.4 61.6 61.7
S i0 2 % 4.9 4.7 4.9 4.6 4.6
A120 3 % 2.1 1.9 1.9 1.9 1.9
C a0 2 % 1.1 1.2 0.8 1.1 1.2
MgO % 0.3 0.4 0.6 0.3 0.3
Mn % 0.3 0.3 0.3 0.3 0.3
h 2o % 5.1 5.5 5.4 5.6 5.8
Ignition loss % 3.3 2.9 3.0 3.1 2.9
CaO/Si0 2 0.2 0.3 0.2 0.2 0.3
Sizing
+ 8 mm % 7.3 7.5 6.3 6.5 10.4
+ 5 mm % 22.7 25.0 22.3 26.0 29.8
+ 4 mm % 30.3 32.3 30.3 35.0 39.6
+ 2 mm % 45.7 50.0 47.3 49.0 55.4
+ 1 mm % 49.7 53.8 51.0 55.5 60.6
+ 500 micron % 57.3 61.3 59.7 64.3 68.2
+ 250 micron % 64.7 68.3 66.0 70.0 74.8
+ 125 micron % 73.0 75.8 74.3 77.0 81.6
+ 63 micron % 82.0 83.8 83.3 83.8 87.6
Mean size mm 2.7 2.9 2.7 2.9 3.3
SMD mm 0.1 0.2 0.1 0.2 0.2
Code: Sizing = Size distribution, cumulative % retained.

AMD = Arithmetic mean diameter
SMD = Sauter mean diameter.
Tumbler = Tumbler strength index = TI
RDI = Reduction degradation index.
Mineralogy = Mineralogical structure of the sinter.
Hematite, Fe 20 3. Magnetite, Fe30 4. Calcium ferrite, Ca0.2Fe 20 3.
27
Table 2.6-5 Port Kembla-Blast Furnace Ferruginous Burdens [21]
1990 Jan. Feb. Mar. Apr, May.

No 5 Blast Furnace
Production, t/month 95753 160480 195717 196207 182830
(Taphole), t/day 6534 6520 6591 6623 6374
Productivity,IV. t/m 3/d 2.2 2.1 2.2 2.2 2.1
Coke rate, kg/tHM 448.6 445.4 444.9 458.3 446.9
Nat gas, kg/tHM 22.9 25.9 26.3 25.7 24.1
Fuel rate, kg/tHM 471.5 471.3 471.2 484.0 471.0
Ferruginous Burden:
Sinter, % 69.9 68.3 71.7 71.0 68.0
Lump Ore, % 10.1 10.9 8.7 9.5 11.0
Pellets, % 20.1 20.9 19.6 19.5 21.0
No 4 Blast Furnace
(Taphole) t/day 2817 2677 2495 1381 2501
Coke rate, kg/tHM 504.3 508.4 544.9 605.7 519.3
Nat. gas, kg/tHM 22.1 21.7 14.8 36.4 22.8
Fuel rate, kg/tHM 526.5 530.1 559.6 642.1 542.1
Ferruginous Burden:
Sinter, % 80.1 80.0 80.0 80.1 80.0
Lump Ore, % 19.9 20.0 20.0 19.9 20.0
No 2 Blast Furnace
(Taphole) t/day 1989 2027 1883 1942 1916
Coke rate, kg/tHM 451.2 457.4 457.0 461.2 472.8
Nat.gas, kg/tHM 40.3 43.5 43.9 38.4 38.7
Fuel rate, kg/tHM 491.5 500.9 500.9 499.5 511.5
Ferruginous Burden:
Sinter, % 70.0 70.0 70.6 71.9 70.0
Lump Ore, % 30.0 30.0 29.4 28.1 30.0
28
3.0 RAW M ATERIALS FOR SINTERIN G
IRON ORE FINES (-6.35 mm x O size)

Why alia (1) (SA) . Goldsworthy (WA)
Cockatoo (2) (WA) . Marra Mamba (WA)
Mount Newman (3) (WA) . Yandicoogina (WA)
Dowd’s Hill (4) (WA) . Crookwell (NSW)
Koolan (5) (WA) . Romeral (Chile)
New Caledonian (6) (NC) . Kudremukh (India)
Koolyanobbing (WA) . Samarco (Brazil)
Robe River (WA) . Carajas (Brazil)
Hamersley (WA) . Esperanca (Brazil)
Tasmanian magnetite. (TAS) . San Isidro (Venezuela)
MAJOR FLUXES (-3 mm x O size)
. Limesand (1) (SA-Coffin Bay) . Serpentine (3)
. Limestone (2) (NSW-Marulan & Japan) . Dolomite
MINOR FLUXES
. Jaspilite (1) . Manganese ore . Ilmenite sand
. Quicklime (2) . Ferro manganese . Bentonite
. Lime Kiln . Dune sand . Olivine
FUEL
. Coke breeze (-5 mm x O) (1) . Natural gas . Coal
. Coke Ovens gas (2) . Heavy oil
RETURN FINES
. Cold return fines (1) . Hot return fines (2)
MISCELLANEOUS MATERIALS
. Flue dust (1) . Scrap fines
. Mill scale (roll scale) (2) . Plant spillage
. Cyclone cake (3) . Plant sludge
. Brecketts (BOS slag) (4) . Lime sludge
. Scarfing scale (5) . Steel shavings
Note: The numbered materials were used for experimental work and during the plant
trial period. The numbers represent relative proportions (1 = highest). The other raw
materials were gradually introduced between 1973 and 1990.
29
3.1 IRON O R E S
3.1.1 Overview
Iron ore from low grade deposits in NSW Australia, was first smelted in 1849 [27],
and over the following 65 years, a number of attempts to establish a viable iron and
steel industry were made. These attempts succeeded when BHP decided to change
from non-ferrous operation to establish the steelmaking plant at Newcastle in 1915,
based on the large iron ore bodies containing high grade hematite at the Middleback
Ranges in South Australia.
Owing to the demand for high iron content and for uniform chemical and physical
characteristics, almost all iron ores are upgraded by crushing, sizing and beneficiation.
Customer specifications are increasingly stricter, not only for iron content but also
for the marginal limits of Si02, A120 3, P, S, CaO, alkalis and other minor components.
Depleting of sources of lump ore, increased severity of mining practices and crushing,
screening and transport operations result in high proportion of fine ore (-6.3 mm) in
the final product: the fines content can reach 30% to 60% of the total weight. The
price for fine ore is significantly lower than that for lump ore.
3.1.2 Origin of Iron Ores
The earth, with a core of high density iron-nickel alloy and a diameter of 12,800 km,
presents only a thin crust about 16 km deep from which ores may be extracted. The
gradual disintegration of the thin outer shell by the physical, thermal, and chemical
forces of nature produced the millions of tonnes of iron ore extracted annually. They
were formed as early as 2.5 billion years ago [23]. Currently iron ores are being
formed in marshes as oxides precipitate, certain mineral deposits slowly weather and
leach, and other minerals, oxides, and compounds undergo transformation in sand
deposits and on beaches. Geographically, iron deposits are located in nearly all areas
of the earth's crust.
Both the origin and location of all ores were determined by the manner in which the
mineral-bearing deposits evolved from the earliest physical condition of the earth -
gas and molten rock. With time, the original fluid earth separated into a single gas
30
phase - the atmosphere - and three distinct liquid phases: metallic, sulfide, and silicate,
each immiscible in the others. Under the influence of gravity the three liquids assumed
a concentric shell arrangement, during a period when minerals that could exist under
conditions of great heat and pressure crystallized from the molten rock (magma). The
metallic layer was innermost, the sulfide in the middle, and the silicate layer outermost.
Igneous ores were formed by fractional crystallization of molten rock during the
earliest period of the earth's cooling cycle. These ores are usually high in iron content;
magnetite and hematite are igneous ores.
Sedimentary ores were formed by slow erosion, disintegration, transportation, and

settling of elements leached from igneous rocks. In addition to these forms of
mechanical deposition, chemical and biochemical activity are important in the for
mation of sedimentary ores. Oolitic hematite, siderite, silicate, and sometimes limonite,
along with fossils and sand grains with cemented hematite, are found in sedimentary
deposits [23].
Contact ores occur near or at the junction between igneous and sedimentary rocks.
Pyrites and carbonates, formed by reaction between the igneous rock and the sedi
mentary rock (usually limestone) occur widely.
Hydrothermal ores were formed by the solvent action of hot solutions, resulting in
the filling of open activities in existing rock and the chemical replacement of soluble
rock formations with mineral deposits. The ores in such instances are usually siderites
and oxides.
Metamorphic ores are those formed through a series of gradual transformations by

biochemical and concentration processes. Vegetable, animal, and mineral constituents
contributed to these deposits, which are a type of sedimentation ore. Magnetite,
taconite and a type of hematite are examples of metamorphic ore [23].
31
3.2 IRON O R E S IN A U S T R A L IA
3.2.1 Overview
Currently, the most common ways of using the iron ore fines are agglomeration
processes: sintering and pelletizing. Each iron ore type has different sintering
characteristics, and these are related to their place of origin, distribution and mode
of occurrence, and mineralogy. Basic information on ferrous ores, flux and fuel is
therefore included.
The systematic interpretation and correlation of the raw material characteristics with
the sintering factors will assist in the better understanding of the specific behaviour
of each iron ore type, not only during sintering processes but also blast furnace
operation.
3.2.2 Iron Ore Types and their Genetic Categories
The most common Australian iron ore types which are used in the local iron and
steelmaking industry and their genetic categories are summarised in Table 3.2-1.
32
Table 3.2-1 Australian Iron Ore Types and their Genetic Categories [24].
Iron Ore Types
Trade Name Orebody Location Genetic Category

1) Whyalla Middleback Ranges Iron-enriched
Whyalla, S.A. amphibolites.
2) Koolyanobbing Koolyanobbing, Iron-enriched
Dowd’s Hill Bungalbin W.A. Archaean banded
iron formations.
3) Mt. Newman Mt. Whaleback Iron-enriched
Pilbara Region W.A. Proterozoic banded
iron formation.
4) Deepdale Deepdale Pisolitic limonites.
Robe River Robe River,
Pilbara Region W.A.
5) Cockatoo Yampi Sound Metamorphosed

Koolan Cockatoo and Koolan ferruginous
Islands W.A. sediments.
6) Tasmanian Savage River, Granular to massive
Magnetite TAS low grade
magnetite ore.
7) Frances Creek Roper Bar Oolitic hematite
Mt. Bundey N.T. and magnetite
ore formations.
8) Constance Constance Range Oolitic hematite,
QLD. limonite, siderite
and chamosite
sediments.
S.A. = South Australia. W.A. = Western Australia.

TAS. = Tasmania. N.T. = Northern Territory. QLD. = Queensland.
33
3.2.3 Iron Ore Deposits, their Characteristics and Modes
of Occurrence
1) The hematite deposits of the Middleback Ranges, about 40 km west of Whyalla,
South Austalia, form a discontinuous line of ridges and ranges. The principal
orebodies in decreasing order of size are Iron Monarch, Iron Duke, Iron Baron,
Iron Prince, and Iron Knight. These ranges and hills occur in a series of folded
Lower Proterozoic jaspilites which are bands of iron-rich siliceous rocks and
interbedded schists, amphibolites, dolomites and quartzites with local concen
tration of high grade iron ores. This ore is a mixture of hard blocky hematite
and soft granular hematite [24] [25].
. Iron Knob orebody was the first to be developed. The ore is mainly soft, friable
and hydrous.
. Iron Monarch orebody consists of hard massive and friable granular ores, ranging
from high grade hematite (assaying 68% Fe) to manganiferous hematite. The soft
ores are generally composed of high grade hematite with negligible manganese
content. The talus ore consists essentially of hard hematite pebbles and cobbles in
a matrix of sandy clay and limonite. Quarrying of the Iron Monarch deposit
commenced in August 1962.
. Iron Prince deposit is almost wholly hematite, both massive and loosely coherent.
. Iron Baron orebody is essentially a southern extension of Iron Prince.
. Iron Duke hematite ore occurs only in the western limb of the syncline and generally
to depths of the order of 122m, where it passes, in descending succession, into
magnetic jaspilite, magnetite-talc ore, magnetite-dolomite ore and hematite-dolo
mite ore. All biotite schists and amphibolites in contact with metasomatically
altered carbonate and talc rocks have been completely chloritised. All schists and
amphibolites encountered in quarries in close association with iron ore, are heavily
decomposed [25].
2) The hematite ore deposits at Koolyanobbing Range and Koolanooka Hills, belong
to the iron-enriched Archaean banded iron formations. These are typically iron
34
enrichments of small discontinuous and steeply dipping bodies of banded
iron formations and meta-sediments [24]. The Koolyanobbing Range is a line
of low hills about 450 km east of Perth (W.A.).
In the late 1930’s, fears were expressed at the capacity of Australian resources
to meet long term requirements of the local steel industry. After official
estimates gave iron ore resources at 263 million tonnes, the Commonwealth
Government imposed an embargo on exports, taking effect in 1938 [25].
The lifting of this embargo in 1960 was the main reason for the exploration
boom in Western Australia during the 1960’s, which continued during the
seventies and beyond on an expanding and more sophisticated basis [27].
Systematic exploration for iron ore in the Hamersley and Ophthalmia Ranges,
situated 1127 km north of Perth in the north-west division of Western Australia,
commenced in 1961. Two years of extensive effort, by numerous companies,
revealed that the iron ore resources of this region far exceed all other known
resources within Australia, and closely rival of those of the major iron ore
fields of the world, such as Brazil, India, and the USSR [25].
On 19 November, 1960, BHP signed an agreement to develop an integrated Iron

and Steel industry at Kwinana in Western Australia in return for leases covering
iron ore deposits at Koolyanobbing and nearby Bungalbin. Production stated
in April, 1967 [26].
3) The iron ore deposits of the Hamersley and Ophthalmia Ranges constitute an
integral mineral province, named Hamersley Iron Province, which is related to
a succession of Proterozoic rock sediments, with which the iron ores are asso
ciated. These deposits are quite distinct from those related to the Archaean
Jaspilites in the northern part of the Pilbara region, the Mt Goldsworthy iron
ore.
. The first major iron ore development in the Pilbara region was undertaken at
the Mt. Goldsworthy iron ore deposit. Exports began in June, 1966. The known
reserves of high to medium grade ores in the Pilbara region are very large and
35
could be in excess of 20,320 million tonnes. If low grade and beneficiable ores
are taken into account, the reserves, life can be reckoned in hundreds of years
[27].
. The iron ores of Mt. Whaleback and Mt. Tom Price of the Hamersley Iron
Province are banded iron ore formations of Proterozoic age. These are the rocks
known as taconite in North America; itabirite in Brazil; hematite quartzite in
India; quartz-banded ore in Scandinavia and banded jaspilite in Australia. The
origin of these deposits is little understood but they appear to have developed
in shallow seas and lakes.
Production from the hematite ore deposits at Mt. Tom Price commenced in
August 1966, and the first shipment of iron ore from Mt. Whaleback left Port
Hedland on 1st April 1969.
The important hematite ores of this category are the result of localised earth
processes which have converted primary banded iron formation into a generally
layered, hematite-rich ore by leaching out the chert (quartz) layers and by
martitization (the replacement of primary magnetite by hematite). These
processes result in a strong lump hematite ore with variable internal porosity
[24].
Mt. Whaleback, the principal high grade hematite ore deposit in the Mt. Newman
area, is over 451 km by railway from the coast at Port Hedland. Mt Newman
ore is a hard uniform high grade hematite averaging 64.4% Fe, associated with
low impurities (<0.036%P) and has one of the best metallurgical characteristics
in the world. At present, -10mm + 6mm lump ore is the most suitable blast
furnace feed. The -6mm x O (-6mm) primary and secondary fines are
agglomerated by sintering and pelletising.
4) The iron ore deposits of the Robe River including the middle Robe and the Mt.
Enid at Deepdale are pisolitic limonites. These apparently were deposited in
either very mature river valleys or estuaries during the late Mesozoic period.
During the Tertiary period, changes in the drainage pattern resulted in the
partial erosion of the limonite sediments, leaving mesaform deposits [28].
36
While these deposits are similar, being composed of cemented pisoliths of layered
hematite and goethite, individual deposits differ in mineral proportions, clay
and quartz contents and minor element geochemistry [24].
Examination of the terrace conglomerate, particulary in the Robe River system,

shows that there is a transition between jaspilite conglomerate and hematite
conglomerate, which reflects the progressive stages of desilication of jaspilite
detritus.
The most extensive deposits of pisolitic iron ore are in the western section of the
Hamersley Iron Province, in the long established drainage systems of the Robe River
and Duck Creek [25].
5) Cockatoo and Koolan Islands in Yampi Sound are about 1900 km north of Perth.
The existence of the Yampi Sound iron ore deposits was known at the end of
the 19th century. Cockatoo ore was quarried and shipped in 1951, and production
from Koolan Island commenced in 1965. The ore contains an average of 62 to
66 % Fe.
The hematite ore deposits on Koolan and Cockatoo Islands occur in steeply
dipping sandstones, quartzites, schists and conglomerates. Genetically, these
are categorised as metamorphosed ferruginous sediments.
. On Cockatoo Island, the orebody has two distinct parts, a high grade zone,
and a low grade zone. The high grade zone consists of steely blue almost pure
hematite, hard in outcrop but often friable below the surface. The low grade
comprises a larger number of friable, lenticular beds, which range from sche-
matite to rich hematite sandstone and rich hematite schist.
The hematite is considered to be of sedimentary origin. The alteration from

magnetite (martite crystals) must be attributed to an earlier cycle of weathering,
presumably predating deposition. The Yampi iron ores are considered to be
clastic sediments derived by erosion from pre-existing iron rich rocks [25].
. On Koolan Island, porous and friable hematites are graded into hemati
te-containing sandstones, conglomerates and schistose rocks. The presence of
disseminated iron oxide grains in these sediments and metasediments, and the
37
presence in the ores of detrital grains of tourmaline, garnet, hornblende, etc.,
has indicated that the hematite ores are the result of localised iron enrichment
of metamorphosed clastic sediments [29].
The hematite ore varies from dense and compact to soft and powdery. Under
the microscope it commonly shows a definite polygonal structure suggestive of
the textures of metamorphosed silicate minerals [24].
. Apart from the above iron ore deposits, other places in Australia are being
exploited on a lesser scale:
6 ) Savage River, in the north-west corner of Tasmania, the granular and massive
low grade magnetite ore is upgraded from 38 to 67% Fe through a beneficiation
plant and pelletising processes.
7) Frances Creek and Mt. Bundey at Roper Bar, 168 km and 90 km from Darwin,
Northern Territory, respectively. These oolitic hematite and magnetite ore
formations contain about 63% Fe.
8) Constance Range in Queensland, consists of oolitic hematite, limonite, siderite

and chamosite beds of the Wabana (Newfoundland) type. The mineralogy of
these ironstones includes a variable mixture of ochrous red hematite, finely
crystalline blue-black hematite, limonite, quartz grains, quartz cement, shale
and clay minerals, rare relict siderite, and/or chamosite.
Athough the concentration of the major elements, determined by chemical analysis

of iron ore, reflects the quantitative mineralogy of the ore, the mode of occurrence
of minor elements, e.g., zinc, potassium etc., is not easily understood. Studies by J.
Oswald [24] and others on the mineralogy and geochemistry of Australian iron ores
may be summarised:
Iron (Fe) is related to the major ore minerals: hematite, goethite, and magnetite. Iron
may also occur in carbonates and in silicates.
Silica (S i0 2) is generally present as quartz, either as relict chert or as introduced

quartz in opal veinlets and coatings. Clay mineral content of the ore is another
significant source of Si02.
38
Phosphorus (P) occurs as apatite in Western Australian banded iron formations. Iron
ore at Dowd’s Hill, Koolyanobbing is reported to contain a complex
carbonate-apatite-fluoride Ca3(P04 )2CaF 2, leucophosphite layers
KFe 2(P0 4)2(0 H).2H 20 as well as apatite crystallites X 5(Y 04)3Z where X is usually
Ca+2 or Pb+3, Y is P +5 or As+5, and Z is F", Cl- or OH" [30] [31].
Aluminium (Al) is present in Australian iron ores in the form of clay minerals, chiefly
kaolinite; as aluminous goethite; as gibbsite and related aluminium hydroxides; as
aluminous hematite and as other complex aluminous silicates. At Mt. Whaleback,
kaolinite gangue predominates but in the adjacent iron enriched Marra Mamba iron
formation, gibbsite occurs in association with aluminous goethite and kaolinite in iron
enriched shales [25]. In Deepdale "J" and "K" ore, alumina occurs as interstitial
kaolinite and as aluminous goethite [24].
Limestone is composed primarily of CaC0 3 or a combination of calcium and magnesium

with varying amounts of impurities, commonly silica and alumina. Limestone generally
occurs at low levels in iron ores, usually as carbonate minerals which fill solution
cavities in ores.
Fluorine (F) is the most electronegative element and the most chemically energetic of
the nonmetallic elements. It usually occurs as a component of the complex
carbonate-apatite-fluoride [30].
Manganese (Mn) occurs in a variety of oxide minerals such as pyrolusite, cryptomelane,

romanechite, lithiophorite, chalcophanite, birnessite, todorokite.
Cobalt (Co) commonly occurs associated with manganese oxide minerals such as
lithiophorite which develop during lateritization. Cobalt may also enter the magnetite
spinel structure, and Ayres [32] found cobalt in relict magnetite grains in the Mt.
Whaleback ore.
Zinc (Zn) also occurs in manganese oxide minerals in iron ores, as minerals of
low-temperature-weathering origin such as chalcophanite, or as metamorphic spinels.
In Iron Monarch ore, zinc tends to be concentrated in manganese spinel segregations,
allowing selective mining to produce low zinc iron ore. Zinc may also occur in a
variety of clay minerals.
39
Alkali metals potassium and sodium (K, Na) probably occur in relict sheet silicates
in Whaleback ore [33] and as clay minerals in Deepdale "J" and "K" ore.
Magnesium (Mg) occurs as traces of dolomite in Whaleback ore [33], and probably as
magnesium silicates which are widely formed in nature.
Nickel (Ni) and Copper (Cu) may occur as trace disseminated sulphides, but an
association with manganese oxides in lithiophorite is more likely. Lindly [33] suggested
that trace amounts of nickel in Mt. Whaleback ore were associated with the iron oxide
minerals.
40
3.2.4 Mineralogy
The mineralogy of iron ores has a marked effect on agglomeration processes, as well
as on direct reduction and ironmaking. The mineralogy and mode of occurrence of
the most common commercial iron ores and associated minerals are described as follows
[24]:
Hematite (Fe 20 3, ferric oxide) is the most significant ore mineral in currently-mined
iron ores in Australia. Pure hematite contains 69.94% Fe and 39.06% O, and it is an
anhydrous compound. Hematite is found in a variety of forms, from soft earthy to
rock crystalline. The Australian hematite ores may be classified as [24]:
i) Martite, a hematite replacement of octohedral magnetite. Martitization is a

common ore forming process, usually associated with desilicification of banded
iron formations to produce hematite ores, for example at Mt. Whaleback.
ii) Dense polycrystalline hematite, microtypes with low internal porosity, may result
from mobilization of iron and recrystallization (Mt. Whaleback).
iii) Secondary bladed hematite is a common form in iron ores developed from
banded iron formations (Mt. Whaleback).
iv) Specular hematite or specularite, occurs as thin plate-like flattened parallel

crystals which are highly reflecting optically (Koolyanobbing).
v) Polygonal hematite crystals, an unusual morphology, occurs in Koolan hematite

ore where individual grains occur in a polygonal mosaic.
Magnetite, or ferrous oxide, is traditionally considered as Fe30 4 but is currently

presented as Fe+2Fe 2+30 4. Pure magnetite contains 72.4% Fe and 27.6% O, and ranges
in colour from dark grey to black. The mineral commonly occurs as octahedral crystals
in layers alternating with siliceous layers in the banded iron formations. Low
temperature oxidation effects, facilitated by the leaching process and water table
movements, commonly convert the magnetite crystal structure to hematite grains with
the octohedral form of magnetite. This form of hematite is called martite.
41
Limonite is a general name given to hydrous iron oxides, sometimes called brown
hematite. Pure limonite has the formula 2Fe20 3. X H 20 , and contains 60% Fe and 8
to 20% combined (X) water. Limonite occurs in a loose, porous form and in compact
varieties with a fibrous structure:
. Goethite, a FeO.OH is the major mineral of the pisolitic limonites of mesas and
terraces along the low-gradient valleys within and around the Hamersley Ophthalmia
Ranges. It occurs in low but variable proportions in various hematite ores, especially
in the hard capping of these ores.
The goethite observed in iron ores is variable in texture, ranging from dense struc
tureless material to cellular yellow ochre. Cavities in iron ores are often filled with
delicately layered goethite which exhibits mammilated and colloform banding. Such
goethite layers suggest colloidal (gel-type) deposition and this texture type is known
as metacolloidal goethite.
Although much of the goethite in Deepdale type limonites and in banded iron formation
derived ores is probably metacolloidal in origin, all goethite does not necessarily have
a colloidal origin. Deepdale ore commonly contains particles of fossil wood in which
the wood cellular tissue has been replaced by goethite, no doubt the result of solution
replacement [24].
Maghemite, y Fe 20 3 occurs as a minor iron oxide mineral in a range of Australian

iron ore deposits. The mineral is widespread in the pisolitic limonite deposits. Although
the formula is normally written Fe 20 3 due to its chemical analysis, the mineral has
high magnetic susceptibility and remanence [24].
Ferrihydrite, Fe 20 3.nH20 is a little known mineral phase which may occur metastably
in some of the Western Australian iron ores. The ferrihydrite commonly forms as a
precipitate by the action of iron ore oxidising bacteria such as Gallionella, Leptothrix
and Toxothrix. Following precipitation, ferrihydrite either remains metastable or
converts to hematite. Ferrihydrite is just one of a number of poorly-defined hydrated
iron oxide minerals, including protolepidocrocite, feroxyhyte and akageneite which
may occur metastably in ochres [34] [35] [36].
42
3.2.5 Clay Minerals
Since the important Australian hematite iron ores are either of low temperature aqueous
origin or have been enriched by low temperature reactions involving water transport,
complex silicate minerals in the rocks will tend to decay to stable clay minerals of
which kaolinite is by far the most common mineral. Other clay minerals occur but
their distribution and exact determination are often uncertain, and their influence,
on such processes as mining and agglomeration, is largely unknown [24].
In view of the complexity of identification of minor amounts of clay minerals, their

occurrence in Australian iron ores is not well known.
Kaolinite, Al2Si20 5(OH)4 is a common constituent in the gangue of Whaleback ore,

and in the goethite ore from Deepdale deposits.
Montmorillonite, (Na,Ca)0.33(Al,Mg)2Si4O 10(OH)2.nH 2O, is widely distributed in small

amounts in iron ores. It is a common minor constituent of shale bands of the Brockman,
Mt. McRae and Marra Mamba formations [37] [38], and occurs to a lesser extent in
related ores. It is apparently always subordinate to kaolinite.
Illite, (K,H 3,O)(Al,Mg,Fe)2(Al,Si)4O 10[(OH)2,H 2O] occurs in low levels in Australian

iron ores. Some recent studies suggest that trace illite clays in Deepdale ore may the
host phases for trace potassium present in some orebodies [24].
Gibbsite, Al(OH)3, boehmite, yAlO.OH and diaspore, a AlO.OH are the major minerals
in alumina-rich laterites (bauxites) and this suggests that their occurrence in iron ores
is a result of lateritization; in particular, from the desilicification of clays [39]. The
rather variable occurrences of these aluminium oxyhydroxides in Australian iron ores
is probably the result of superficial weathering.
3.2.6 Silica Minerals
Quartz, S i02, and its cryptocrystalline equivalent chalcedony, occur widely in iron
ores. Western Australian iron ores contain minor amounts of relict banded iron
formation, which is essentially finely granular quartz (chert) containing martite.
43
3.2.7 Carbonate Minerals
Calcite, CaC03, Dolomite, CaMg(C03)2 and Ankerite, Ca(Fe,Mg)(C03)2 occur in the

gangue of Australian iron ores.
3.2.8 Iron Silicate Minerals
A number of complex iron silicate minerals, either prismatic or platey in habit, occur
in the Western Australian banded iron ore formations and, generally chemically altered,
in iron ore derived from these formations. The following mineral species have been
found [24]:
Stilpnomelane, [(Al,Fe+3)x(Fe+2,Mg,Mn)48_x(OH)48_x][Si3o-36Al6_o0 6o]2> is a complex

layered silicate occurring in iron formation and low grade metamorphosed volcanics
and sediments. It is common in Marra Mamba iron formation shales.
Minnesotite, (Fe,Mg)3Si4O10(OH)2, is a secondary mineral occurring as sheaves and

acicular crystals, and as spherules. It occurs chiefly in Marra Mamba banded iron
formation.
Riebeckite, Na 2(Fe+2,Mg,Fe+3)5Si80 22(OH)2, is a fibrous amphibole mineral, crocidolite

being the asbestiform variety. Hamersley Province riebeckite is probably a secondary
stress-induced crystallization product [19].
Biotite, K(Mg,Fe)3(Al,Fe)Si3O 10(OH,F)2 and Chlorite M5_6(Al,Si)4O 10(OH)8, where M =

MgFe+2Fe+3Al are typically platey minerals occurring in shale bands of detrital,
authigenic or low grade burial metamorphic origin.
Chamosite, (Fe+2,Fe+3Mg)5(Al(Si3Al)O 10(OH)8). Gole [41] indicated that this mineral

occurs in Archaean banded iron formation.
Greenalite, (Fe,Mg)6Si4O 10(OH)8, occurs in very low-grade Archaen banded iron for
mation [41].
44
3.2.9 Manganese Oxide Minerals
A variety of manganese oxide minerals occur as minor constituents of Western Aus

tralian iron ores. These generally occur as cavity fillings and veinlets resulting from
the oxidation of manganese containing solutions migrating through the ores. Minerals
which develop in this manner include [24]:
. Pyrolusite, MnOz
. Cryptomelane and Manjiroite with the complex range composition from A 0.64B 8O 16
to A0.8B8Oi6>where A = essentially K+ with minor substitution by Ba+2 and Na+,and

B = essentially Mn+4 with substitution by Mn+2, Zn+2, A1+3, Cu+2, Fe+3.
Manjiroite is a form in which A = essentially Na+.
. Lithiophorite, (Al,Li)Mn2(OH)2, often containing small amounts of Ni, Cu, Co and

Zn.
The manganiferous iron ores of the Middleback Range contain a variety of manganese
oxides of metamorphic and low temperature weathering origin.
. Zincian hausmannite, (Mn,Zn)0Mn 20 3, and

. Zincian Jacobsite, (Mn,Mg,Zn)0.Fe20 3, Both have been identified in manganese-rich
of ore from Iron Monarch, together with secondary cryptomelane and psilomelane
[24].
45
3.3 FLU X
3.3.1 Overview
Flux, in sintering, is a substance used to promote fusion of the sintering feed ores,
and to form the phase component glass in the "combustion zone" during sintering. The
major roles of the flux are to lower the melting point of the sintering mix material
and to provide the basicity ratio CaO/SiC>2 required by blast furnaces.
Abundance of this material is evidenced by the fact that an estimated 3.5% to 4% of

the earth's crust is calcium and 2% magnesium [43]. Calcite and dolomite represent
about 20% of all sedimentary rocks, and occur in all continents, in strata of every
age-from Precambrian to Quaternary [44].
Limestone is a sedimentary rock composed dominantly of calcium carbonate and

magnesium carbonate with varying amounts of impurities. The major impurities are
silica, alumina and iron oxides. The impurities may be finely disseminated throughout
the rock (e.g., clay minerals, iron oxide, phosphorus, bituminous matter), present as
individual grains (e.g., pyrite, quartz, feldspar), or aggregated into large nodules (e.g.,
chert or flint).
In mineralogical terms, limestone is composed of four major minerals, two of which

are calcite (CaC03) and dolomite MgCa(C03)2; industrially, calcite is called limestone.
Three types of basic flux are used in sintering: limestone, limesand and dolomite.
However, limestone and limesand are chemically the same, but different in texture
and some minor chemical composition.
Limestones range in texture from lithographic limestone to oolitic limestone. The

first is simply called limestone, e.g., Marulan limestone; and the last is locally called
"limesand", e.g., Coffin Bay limesand. In appearance limesand is similar to beach-sand
[30].
In 1971, Miinive L., et al. [30], made extensive comparative studies of Port Kembla
sintering plant raw materials. Microscopic observations of the flux materials revealed
46
that:
. Limestones from Marulan in N.S.W., Rapid Bay in S.A., and from Japan are
typical lithographic limestones, consisting mainly of medium to fine grained
crystals of massive calcium carbonate, occasionally associated with fossiliferous
fragments.
Marulan limestone is the largest deposit close to Port Kembla Steelworks, and is quarried
by the Southern Portland Cement Co. The Japanese limestone, slightly better quality
and finer grain than Marulan limestone, is imported from Japan.
. Limesands from Coffin Bay in S.A., and Wanneroo in W.A., are classified as skeletal
oolitic limestones, consisting mainly of coarse to fine, well rounded oolites or
flattened bodies of cemented calcium carbonate, associated with minor impurities
of free quartz grains, iron oxides, pink crystals of alkali silicates, fossilised
fragments and other impurities.
Extensive deposits of limesand at Coffin Bay were opened in 1966, to supply BHP
sintering plants at Port Kembla, Newcastle and Kwinana.
Although limestone and limesand are similar in chemical composition, limestone is

preferred to limesand for sintering. During sintering, limesand is not well assimilated
in the sinter matrix; a maximum of 25% limesand can be tolerated, but one hundred
percent limestone is the preferred feed material.
3.3.2 Origin
Limestone, the most important and abundant of all sedimetary rocks that is employed
commercially, is usually of organic origin.
Limestones and dolomites are formed by a wide variety of processes. Limestone may
be produced by the cementation of carbonate sand grains derived from erosion of a
former carbonate land form (e.g., island, hill, plateau); in this case it is a classic rock,
a sandstone, whose grains have been transported and deposited in accordance with
the laws of hydraulics [44].
47
On the other hand, a limestone might result from direct precipitation of the carbonate
through a saturated solution. In this case it is a chemical rock, whose formation has
been governed by either an increase in temperature, which reduces the solubility of
the calcium carbonate, or through evaporation and pressure [45].
The texture and crystalline form of limestone depends on the size and the shape of
the calcareous particles or grains that are deposited. Usually, this deposition is
contaminated with varying amounts of impurities, like silica, that through time become
an intimate part of the limestone, cemented to the carbonate particles [30].
3.3.3 Mineralogy
Limestone is composed of four major minerals, exclusive of impurities:
Calcite Aragonite Dolomite Magnesite

CaC0 3 CaC0 3 CaMg(C03)2 MgC03
Crystal rhombo- ortho rhombo- rhombo-
structure hedral rhombic hedral hedral
Molecular wt., 100.1 100.1 184.4 84.3

Sp. gravity, 2.72 2.94 2.83 3.00
Molecular vol, 36.8 34.0 65.2 28.1
Hardness, 3.0 3.5-4.0 3.5-4.0 3.5-4.5
Colour, Colorless White often Colorless White, tan
often tinted by often tinted or brown.
tinted by impurities. pink or tan.
impurities
Of the above minerals, the only point of disagreement is that dolomite may be:
. a double carbonate of calcium and magnesium, or
. a mechanical mixture of calcite and magnesite.
3.3.4 Impurities
There are two classifications of impurities:
48
i) Homogeneous, in which the impurities are well dispersed throughout the for
mation, e.g., clay, silt and sand or other forms of silica, like quartz.
ii) Heterogeneous, in which the impurities collect only in the crevices or between
the strata or as siliceous pieces or nodules of sand, chert, or flint loosely embedded
in the limestone.
The major impurities are silica, alumina, iron oxide, clay, sand, sulphur,
phosphates, alkali (Na, K), or organic materials; other impurities usually occur
in small quantities.
3.3.5 Classification of Flux
Basic fluxes include limestone (CaC03), dolomite CaMg(C03)2, and lime (CaO). These
fluxes incorporated in the sinter combine with the silica (Si02) and alumina (A120 3)
in the furnace burden to form a fluid slag with a melting temperature near that of
iron. So both slag and iron is cast from the furnace as separate fluids.
Another important function of the lime (CaO) is to combine with the sulfur from the
coke, ore and flux.
Acid fluxes the only important acid flux is silica (Si02), which reacts with basic
material that is present. Si0 2 is added to increase slag fluidity. Silica also produces
a progressive polymerisation process in the continuous silica network. And acid oxide
absorbs oxygen ions when dissolved in a basic melt .
A basic oxide provides oxygen ions when dissolved in a melt. This behaviour can be
summed up by general definitions of acid and basic oxides.
Neutral fluxes fluorspar (CaF2), and borax (Na 2B40 7), are both used to increase fluidity
of slag [42, 43]. Neutral fluxes (CaF2) were used at AIS on trial basis.
3.3.6 Serpentine
Serpentine (Mg,Fe)3 Si20 5(0H )4 is a group of green, greenish-yellow or greenish-gray

ferromagnesian hydrous silicate rock-forming minerals having a greasy or silky luster
and a slighty soapy feel.
49
The magnesia addition to the blast furnace had been made principally via serpentine
addition at the sintering plant, until mid-1985, when a magnesia-bearing Australian
iron ore pellet was introduced to the burden of Port Kembla No 5 blast furnace. As
a result, a progressive replacement of serpentine by dolomite as the sintering plant
magnesia source was carried out [49].
3.3.7 Manganese Ore
Manganese ore occurs in nature as manganite MnO(OH), and the most common com
pound is the dioxide M n02. The manganese ore is transported to Port Kembla Steelworks
from Groote Eylandt (N.T). These ore fines are added to the sintering mix to provide
adequate manganese content in the hot metal, in order to remove sulphur and other
impurities during the refining process of iron and steelmaking.
3.3.8 Dune Sand
Dune sand is a silica (Si02) load material added to the sintering mix to reduce the
Al20 3/S i0 2 ratio, in order to maintain adequate slag fluidity at the blast furnaces.
50
3.4 FUEL
3.4.1 Overview
The most common fuel used in the sintering of ferrous ores is coke breeze, the fine
fraction of the product of coking coal.
The coarse coke (+20mm), is used in the blast furnaces and the undersize material
(-20mm x O), is used in the sintering plant. Blast furnace coke screen undersize is
also used in the sintering plant.
Both coke breeze and nuts (small coke) can be used in the sintering plant, after crushing
in rod mills to pass approximately 95% -4mm.
Sizing data from coke ovens primary screens and blast furnace underbin screens are
indicators of sintering plant feed coke quality on which crushing practice is based.
Coke makes up about one half of the volume of material in the blast furnace. It is
the only material which keeps its solid form until it is gasified in the raceway, and
forms a permeable skeleton from the furnace stockline to the hearth; it is also the
major heat supplier for the process [46].
In contrast, coke utilization in sintering is 3% to 5% of the total sintering feed mix.

Even so, coke breeze plays an important part in the sintering process:
. as the main source of heat, and
. to support a reductive process in the bed.
The time the charge materials spend in the high temperature zone is also important
as it determines the development of processes of heat exchange and fusion, and the
formation of quality agglomerated sinter.
3.4.2 Coke Breeze Characteristics

The main coke breeze characteristics are:
. Size distribution
. Moisture
. Ash content
. Reactivity
51
3.4.2.1 Size Distribution
The influence of the size distribution of the coke breeze particles on the rate of
reaction and the rate of combustion is critical, since the combustion zone, due to its
high temperature and liquid formation, is considered to be the major area of gas flow
resistance. Hence, the effects of fine and coarse size distribution in sintering are:
- Fine size coke (-1mm x O) tends to ignite earlier than large coke particles, then
the flame front profile widens. Ultrafine coke also decreases the permeability
and hence the reduction degradation index (RDI) becomes poor. To overcome
this problem it is necessary to decrease the bed height and regain RDI at the
expense of physical quality [47].
- Fine coke particles have poor thermal efficiency, as they burn rapidly ahead
of the flame front and the ratio C 0 /(C 0 2 + CO) increases in the exhaust gas
during the sintering process.
- Excessive ultrafine coke content (eg., -152 micron) in the sintering feed mix,
as well as all ultrafine materials have a detrimental effect on the sintering gas
permeability, hence sinter quality and productivity decreases. Each 1% of
ultrafines in the raw mix corresponds to a productivity loss of 0.02t/m 2/hr.
- When the mean size diameter of the coke increases, the rate of reaction decreases,
hence the reduction degradation index (RDI) and gas permeability improves,
but care must be taken or tumbler index (TI) reduces if the coke is too large.
- Coarse coke particles are not efficiently utilized in sintering on a grate, as they
tend to burn behind the flame front, resulting in increased fuel consumption.
- Segregation of the feed flowing onto the strand depends on the particle size
range, and this has a marked effect upon the permeability of the bed and the
air flow through it during sintering.
- In general terms, the largest particles segregate, accumulating towards the bottom
of the grate.
52
3.4.2.2 Coke Moisture
Minimum variation of coke moisture is essential for adequate blending control of the
green sintering feed on a dry weight basis. An on-line moisture meter allows this
variability to be decreased, with a consequent lowering of FeO and raising of RDI
[46]. In addition, there are resulting coke savings, with implications for the works
coke balance. However, moisture meters do not have the capability to detect large
coke moisture variations. Most coke moisture probes provide accurate moisture value
when the variability of coke moisture measurements is less than + 0.5% about the set
coke moisture value at the probe instrument indicator [46].
Port Kembla samples show that coke moisture varies between 8% and 14% on samples
representative of 8 hours operation. As a result, sinter FeO and RDI exhibit excessive
variability [46].
Significant coke moisture variations mostly reflect heavy rainfalls or some irregular
wet quenching operations at the coke ovens or uncontrolled water sprays on conveyor
belts because the work environment is dusty [46].
3.4.2.3 Ash Content
The coke breeze ash content is determined by blast furnace requirements and the
proportioning in coal blending-cokemaking operations. Blast furnace coke ash percent
is usually used as an indicator of expected coke breeze ash.
The coke ash variability at Port Kembla Steelworks has been markedly improved since
the washed coal blending yard was commissioned in September 1983, and coke ash
content of 15% - 16% has been lowered to 14% by changes in coal cleaning practices.
53
Coke Ovens operation targets for coke ash in 1983 and 1990 [46] [21] were:
1983 1990
Ash: Apr-Oct. Jan-Oct.
bed average 14.5% 11.3%
between bed difference +0.2 %.-0.2 % +0.2 %,-0.2 %
within bed standard deviation 0.08% 0.03%
Low coke ash content decreases the slag volume and increases porosity and reducibility
index (RI) in the sinter. However, a low content of basic components (e.g., below 10%
Fe 203>CaO, MgO, K 20 , Na 20) in the coke ash, decreases the reactivity and increases
the post-reaction strength of the coke [46].
3.4.2.4 Reactivity
In sintering, the combustion proceeds in a thin horizontal layer, moving vertically
through the charge. Hence, heat is transmitted through the charge largely by the gas
atmosphere (N 2+ 0 2+C0 2+C0) passing through the bed.
The reactivity of the coke breeze, under constant sintering parameters, determines
the time of combustion and therefore the thermal condition in which the sinter
mineralogy is formed by the solidus and liquidus reation in the "combustion zone".
54
3.5 RETURN FIN ES
3.5.1 Overview
The return fines represent the fines which, after discharge and pre-crushing of the
sinter cake, are separated by screens from the lump product sinter; they are recycled
to the sintering process.
The mechanical stress which the product sinter undergoes at the plant, which generates
the sinter fines, is simulated in the laboratory by dropping the sinter cake four times
from two metres height (shatter test), and screening through 6.35 mm. The ratio of
the amount of sinter fines (%) produced through the shatter test to the plant return
fines (%) used in the sintering mix, is called "return fines balance".
The total return fines in the sintering feed mix is between 20 to 30%. For this reason,
the role of return fines in sintering is important.
3.5.2 Return Fines Characteristics
Return fines are a mixture of reacted, partially reacted and unsintered particles,
usually -6.35 mm x 0 size. The return fines are also a mixture of fines produced at
the sintering plant, blast furnace under bin screening fines and plant spillages.
The quality of the return fines reflects the quality of sintering operation. Return
fines containing predominantly sintered particles produce good quality sinter; return
fines containing noticeable amounts of unsintered and ultrafine materials produce
low quality sinter.
The distribution of return fines in the sinter cake is not uniform: the unsintered
materials are encountered mostly in the upper and lower parts of the sinter cake.
Inevitably, there is a tendency for the top of the bed to be weaker because this region
is normally heated to a lower temperature for a shorter time than deeper layers of
the bed, where the combustion is intensified by the heat transferred down from the
55
upper levels. Cooling is also more rapid at the top of the bed. On the other hand,
the ore particles are larger in the lower part of the sinter, so the permeability is higher
and the combustion zone speeds up [48].
A high circulating load is caused partly by shallow bed depths. Approximately 25

mm of weak sinter is produced on the top of the bed, regardless of the total bed depth
[16].
The return fines have a lower melting point than the raw ore mix, and sinter first
when the flame temperature arrives, and so they really start the sintering [50].
3.5.3 Return Fines Effects on Sintering
3.5.3.1 Permeability and operating stability
The permeability of a sintering feed material is defined as the quantity of air which
flows through a unit cube of charge bed under a unit pressure head. The permeability,
for a given sintering feed and bed height, is controlled by the size and number of
voids in the bed. So return fines particles, which are of larger size than the ore
particles and have a more regular porous structure, improve the air utilization
efficiency and plant operating stability.
For example, the deleterious effects of flue dust and concentrates containing high
ultrafine particles is compensated by an increase of return fines which partially
restores the permeability and hence the flame front speed and the sintering time.
Endothermic and exothermic reactions as well as the water content of the sintering
mix, when using sludges, can be regulated by an adequate addition of return fines in
such way that the different reaction speeds match the advance of the flame front
[50].
3.5.3.2 Degree of Sinter Oxidation
The effects of the return fines rate and the fuel content (coke breeze) in the mix on
the FeO content in the sinter are important, so return fines are used as a manipulated
56
variable for the control of the FeO content in the sinter. "The lower Fe++ value is.
given by a higher return fines rate" [50]. However, an increase in sinter productivity
can only be achieved by increasing the percent of new raw material.
3.5.3.3 Productivity - Hot Return Fines
In 1956, Bennett et al found, when sintering ores, that much improved production
rates are obtained when using hot return fines as compared with cold return fines.
This proved the case with Port Kembla ores, and experimental results with a particular
mix showed an increase in output of 30% by using return fines at 204 °C as compared
with cold return fines [16].
In 1961, Jennings and Grieve from Huntington Heberlein Co. Ltd., stated that the hot
return fines added to the sintering feed raise the mix temperature to around 60 - 70
°C before ignition [55]. They also stated that investigations have confirmed that the
increased bed resistance is associated with the distribution of the moisture in the
vertical plane, causing the lower part of the bed to collapse. They observed that when
the mix is initially at a temperature of about 60 °C, the moisture driven off by ignition
does not condense in the lower layers of the mix during sintering, but escapes in the
waste gas instead. A comparative study, of the effects of hot and cold return fines
on permeability, indicated that hot return fines increased the output rate by 65% at
a constant suction [55].
At the Port Kembla sintering plant, addition of hot return fines in the sintering feed
was discontinued on 22 February 1984 and the hot screen was removed on 19 November
1985, mainly due to the high maintenance cost and reduced operations availability.
There was some worry that sinter quality and productivity would change, but the
maintenance cost and availability problems were much more significant [20].
Nevertheless, the sintering feed mix temperature can be increased by adding hot water
instead of cold water in the mixing and granulating drums.
57
3.6 M ISCELLAN EO U S M ATERIALS
3.6.1 Overview
During the course of sintering operations, a number of by-products arise in the form
of spillages, fume and flue dust recovered, and these must be uniformly incorporated
in the charge for retreatment. The main miscellaneous materials are :
3.6.2 Mill Scale
Mill Scale is the scale which has been formed and removed during hot working
operations. The scarfed material that is utilised during sintering has a high iron
content, much of which is in the form of magnetite (Fe30 4) that gives off heat when
oxidised. A 1% increase in mill scale in the raw feed bedding fines decreases coke
consumption by 0.8 kg per tonne sinter.
The use of mill scale is limited by two factors :
i. It contains oil, caused by leakage of bearing oils and greases, and hydraulic
oils during the rolling process. If excessive hydrocarbons are present in the
sinter feed (in the order of 0. 1%) they can enter the waste stream and cause
glow fires in the electrostatic precipitators.
ii. The use of mill scale above approximately 10% in the sintering feed causes a
decrease in the production rate.
3.6.3 Brecketts Scrap
Brecketts Scrap is a fine metallic material reclaimed from iron (BF) and steelmaking
(BOS) slags using a screening and magnetic separation plant.
The magnetic fines (-6 mm) are used in the sintering process, and the non-magnetic
slag fines (-6mm), by-products of the metal recovery, are used in the sintering blend
to control the silica (Si02) level. Useful calcium and manganese is also contained in
these slag fines. However, the high phosphorus content in this material is an undesirable
constituent.
58
3.6.4 Flue Dust
Flue Dust is material collected from the dust catchers at the blast furnace. It is a
mixture of ferrous materials, coke and limestone, which must be agglomerated to be
used as blast furnace feed.
The use of flue dust in the sintering green mix saves on fuel and flux which normally
must be added to the mix. Flue dust decreases coke rate because of its contained
carbon.
Assuming that the carbon in the flue dust is 30%, the flue dust in the raw mix is 0.7%,
and the carbon in the coke is 87%, then the carbon content in the raw mix is equivalent
to 0.7 x 0.30 x 100/87 = 0.24%. As a result, 0.24% less coke need be added to the total
new material.
However, the use of flue dust in the sintering mix is limited by its fineness, which
reduces the permeability during the sintering process, and results in a decrease in
productivity when used above approximately 10% in the green mix blend.
59
4.0 EXPERIM EN TA L
4.1 D ESCR IPTIO N OF A PPA R A TU S FOR SIN TERIN G
Photographs of the set up of the main sintering equipment used during the preliminary
and the fundamental experimental work are shown in Figs 4.1-1 to 4.1-4 and these
are described below.
4.1.1 Mixer
Optimization of granulation of the green sintering feed was conducted in the laboratory
by comparing the performance of two type of mixers, looking at mixing time, angle
of mixing, number of revolutions per minute and the moisture content of the mixture.
The selected unit and mixing conditions were:

. A 50 dm 3 (0.05m3) cement mixing unit
. Dry mixing of raw materials for 2.0 minutes
. Water addition under constant spray for 1.0 minute, and continuation of mixing
for a further 2.0 minutes.
This method of mixing proved consistent, and is comparable to, if not better than,
other methods providing the mixer is set at the horizontal position to minimise
segregation of the coarse particles.
4.1.2 Sintering Reactor (Pot) Unit

Most of the preliminary experiments were conducted using the original rectangular
reactor (Figs 4.1-1 and 4.1-2), with an effective grate area of 1200 cm2 and a bed
height of 254 mm.
The cylindrical sintering reactors Nos 2 and 3 (Figs 4.1-3 and 4.1-4), designed by the
author and locally made mostly using scrap materials from the steelworks, were used
to develop bed depth and high suction sintering technology.
The basic experimental work was carried out using the reactor No 3 with an effective
grate area of 345 cm 2 and a variable bed height of 500 to 650 mm. This unit (Fig
4.1-4) was designed with five temperature and pressure sensors to measure sintering
temperatures and differential pressures as the flame front proceeds through the bed.
60
4.1.3 Burner
The overhead radiant burner for the rectangular reactor was locally redesigned using
the latest technical advance in burners. The new burner for the reactors 2 and 3 was
also locally designed and made. Ignition time and suction levels during ignition were
experimentally improved.
Table 4.1-1, shows the geometry, dimensions, correction factors to calculate produc
tivity, and other sintering parameters for the three reactors. The burner operating
conditions are also included.
4.1.4 Anemometer
A 102mm Vane Anemometer with a velocity range of 18 to 1219 m/min., suitable for
temperatures approximately 5X to 121^C, and with an automatic timing device for
direct reading of air velocities during sintering, was supplied by Esdaile & Sons Pty
Ltd, 31 Bay Street, Glebe, NSW 2037.
The reset button for setting the pointer to zero in this instrument was specially located
outside the propeller housing. This unit including the mounting handle and transport
case was particularly designed for the cylindrical sintering reactors.
4.1.5 Sintering Suction Fan

The original 889mm diameter impeller disc, powered by a 35 b.h.p. motor with a
driving unit speed of 1470 r.p.m, with approximately 20 a.m3/niin capacity (actual
volume measured under fan inlet condition) at a static pressure differential across
the fan of 11 kPa at 500X1, was supplied by D. Richardson & Sons Pty Ltd.
This fan unit maintained a constant suction of 10 kPa during the whole of sintering
time; suctions of 15 kPa were maintained for approximately 90% of the sintering time.
The need to conduct sintering tests with a deeper bed at a constant suction between
12 kPa and 16 kPa, and to produce sufficient product sinter (35kg) to carry out
complete physical tests in accordance with the ISO procedures, required an increase
in the existing fan unit capacity.
This involved the purchase of another unit to deliver 19 a.m3/min at 316X against a
static pressure of 12 kPa, powered by an 18 b.h.p motor, and the installation of a two
61
stage cross-over blower, and a cyclone deduster, together with electrical and mechanical
connections. This equipment was supplied by Davidson of Australia Pty Ltd, Victoria
House, 11 Anderson Street, Chatswood, NSW 2067.
Table 4.1-1 Geometry, dimentions and other sintering parameters

for reactors and burners.
Sintering Reactors Reactor 1 Reactor 2 Reactor 3

Geometry: Rectangular Cylindrical Cylindrical
Dimensions: length,mm 610 diameter 203 210
width,mm 305 radius 102 105
height,mm 330 460 650
area,cm 2 1458 324 345
Effective bed height, mm 254 254 - 500 500 - 650
Effect, grate area, cm2 1200 324 345
Productivity factor: 7.8 44.4 41.9
Capacity, kg sinter 75 13 - 25 27 - 30
Hearth layer, kg 11.3 1.8 1.8
Burner home made home made home made
Ignition time, sec 75 75 90

suction, kPa 6.2 7.5 7.5
fuel gas C.O. C.O. L.P.G.
Std. deviation of
productivity [57], 0.07 0.08 0.08
Note: The productivity factor is composed of:

. Hearth area factor, and
. A time factor to give a production rate in terms of t/m 2/d.
62
Figs 4.1-1 (top) and 4.1-2 (bottom), the original rectangular sintering reactor.
63
Fig 4.1-3 at the top shows the cylind ideal sintering reactor used during the
preliminary testworks, and Fig 4.1-4 at the bottom, illustrates the reactor used for the
fundamental experimental work.
64
4.2 EXPERIM EN TA L SIN TERIN G PROCEDURE
4.2.1 Sintering Feed Mix
A typical mix calculated on a dry basis for the large and cylindrical sintering reactors
was:
% Kg Kg
Ore mix 56.8 43.7 18.2
Limestone 12.2 9.4 3.6
Coke 3.0 2.3 0.8
Return fines (RF) 28.0 21.6 7.4
Total 100.0 77.0 30.0
The weighed raw materials were added in sequence to the mixer, and dry pre-mixed
for 2.0 min. Following pre-mixing, the required amount of water was constantly added
for 1.0 min using a spray gun. The mixing-granulation was continued for a further
2.0 min with the mixer set at horizontal position.
On completion of mixing, all the greenfeed was removed from the mixer, the "Weight
in" was recorded, and samples for moisture and permeability determinations were
taken.
Meantime, the hearth layer consisting of -12.7mm +6.4mm sinter was spread over the
grate bars to a depth of approximately 25mm. Then the mix was loaded into the
sintering reactor unit via a cylindrical feeder with a louvered bottom to minimise
segregation.
4.2.2 Sintering Testwork
Ignition of the sintering reactor greenfeed was provided by the use of a gas fired
burner. The burner was preheated to 1100<C and the reactor content ignited for 90
seconds with an initial suction of 7.5 kPa.
After this time, the suction was rapidly increased to the required suction, and this
level was maintained until the natural fall off in suction at the completion of sintering.
65
Sintering time was calculated from the start of ignition till the point at which the
waste gas reached its maximum temperature.
On completion of sintering, the sinter cake was fan cooled and the reactor unit
discharged when the waste gas temperature was approximately 80e€. Finally, the
sinter cake was weighed and dropped 3 times from a height of 1.8m to simulate plant
handling on relative terms. The -6.4mm fines generated (7.2 kg) in the dropping test
must balance with the input return fines in the mix with a ratio of 1.0 ±0.05 for a
sintering test to be accepted.
Calculation:
Wet mix in 27.5 kg
Dry mix in 26.4 kg
Dry coke in 0.8 kg
Return fines in (C) 7.4 kg
Hearth layer in (B) 1.8 kg
Sinter cake out (A) 24.5 kg
Sintering time 19.5.min.
Return fines in = dry mix x % RF = 26.4 x 28% = 7.4 kg

Coke in = dry mix x % coke = 26.4 x 3% = 0.8 kg
Ore mix in = 26.4 - 7.4 - 0.8 = 18.2 kg
Sinter produced = A - B - C = 24.5 - 1.8 - 7.4 = 15.3 kg
S i n t e r p r o d u c e d x 44 A ( f actor')
Sintering rate =
sint ering t im e ( m i n )
1 5 .3 x 4 4 .4
= 3 4 . 8 t / m z/ d
1975
Yield =
R F o ut _ 7 .2
RF b a l a n c e =
RFi n 7 .4
66
4.3 SIN T ER Q U ALITY CO N TRO L
Sinter qualities such as chemical composition, physical properties and mineralogical
structure are not uniformly reported. They are to certain degree different at each
plant and in each country. Hence, sinter quality at Port Kembla Steelworks is reported
as:
Chemical Composition
Fe, FeO, S i02, A120 3, CaO, MgO, Mn (%)
and CaO/Si0 2 ratio.
Mineralogical Structure
Hematite
Calcium-ferrite
Magnetite
Slag (glass)
Porosity (%)
4.3.1 Physical Properties
4.3.1.1 Sinter Size Distribution

+40, +10, +5mm
. Sauter Mean Diameter, based on an expression of Ergun’s:
Where Mi - proportion of weight of size fraction "i"

di - arithmetic mean diameter of size fraction V
. Arithmetic mean size is defined as:
¿.M.S.-ZMi.di
Where Mi = proportion of weight of size fraction "i"

di = arithmetic mean diameter of size fraction "i"
67
4.3.1.2 Cold Sinter Strength
Cold sinter strength is the resistance of the sinter to fragmentation and fines production
during handling and the operation of charging into the blast furnace.
Measurements of the cold sinter strength indices vary for each plant and in each
country, due to the fact that this property is determined using different standards
(ASTM and JIS) by:
Size index (3 drops ,23 kg sample); % cum. +6.4 mm

Shatter index (4,10,20 drops,23 kg sample); % -6.4mm
Rumbler index (AIS drum, 23 kg and 11.3kg sample); % cum +6.4 mm
Tumbler index (ASTM drum, 12kg and 4.5 kg sample); % cum. +6.4 mm or +10mm.
Abrasion index (ASTM drum); % cum. +8mm and % -5 mm
Three indices are used at Port Kembla Steelworks:
Size index: After weighing the product sinter cake, it is dropped 3 times from a height
of 1.8m (currently 2.0m), the -6.4mm fines being screened out after 2 drops to minimise
the cushioning effect of the fines. After the third drop, the material is mechanically
screened through 50, 25, 10, 8 and 6.4mm screens (450mm dia.) and the percentage of
each fraction is reported as the experimental sinter size distribution. The size index
is the cumulative percentage on 6.4mm, and the % -6.4mm output is used to calculate
the return fines balance (RFB).
The sinter size distribution is reported from this sizing test result, then the required
amount of sinter fractions for rumbler index or tumble index, hot testing and other
physical evaluations are separated.
The remaining +6.4mm sinter is recomposed to its original sizing, crushed to -3.3mm,
mixed and riffled down to approximately 300 grams of sample for chemical analysis,
reserve and mineralogical studies.
Shatter Test
The shatter test differs from the size test mainly in the number of drops and sample
68
weight. The size test is conducted on the whole experimental sinter cake, while the
shatter test uses a sinter sample obtained from a conveyor belt close to the blast furnace
top or ofter the blast blast furnace underbin screen. Hence, the size test simulates
the sintering plant handling conditions from the sinter strand to the hot screen dis
charge, and the shatter test simulates the sinter handling conditions after the plant
hot screen or after the blast furnace underbin screen to the blast furnace top.
A 23 kg sinter sample is dropped 4 times, the -6.3mm fines from each second drop is
separated to avoid the cushioning effect. After the fourth drop, the material is
mechanically screened through 50, 25, 10, 8 and 6.4mm screens, the percentage of each
fraction is reported as the plant sinter size distribution.
The shatter test is expressed as shatter index (SI), the percentage of -6.4mm fines
generated after 4 drops.
The drop shatter test can be extended to 10, 20 drops in special cases, and the %
-6.4mm from each second drop is used to determine the sinter degradation trend.
Rumbler Index: A 23 kg or 11.4.kg of -50 +6.4mm sinter recomposed to the original

size distribution was rumbled for 200 revolutions at 25 rpm. The rumbler index was
the cumulative % +6.4mm after rumbling. This test was conducted from the esta-
blishement of Port Kembla sintering Plant until July 1971. Hence some experimental
work was conducted using this index. The rumbler index was fair to poor compared
to the -6.4mm x O (16-25%) sinter fines content on the blast furnace top.
Tumbler Test: Assesment of the sinter cold strength using the ASTM tumbler drum
(914mm inside diameter (I.D) x 557mm long, 6mm thick and 2 lifters each 50mm), was
introduced at Port Kembla steelworks in August 1971.
A 12.0 kg sinter feed sample is prepared from the original sinter size distribution as
follows:
Size, mm kg
-50 + 25 1.2
-25 + 12.5 4.8
-12.5 + 10 3.0
-10 + 6.4 3.0
Total 12.0
69
This sample is placed in the standard tumbler drum and rotated for 200 revolutions
at a speed of 25 rpm. After tumbling, the sample is removed and screened. Weights
of the various screen fractions are recorded, and the cumulative percentage retained
on 6.4mm is the tumbler index (TI). Some sintering plants report this tumbler index
as abrasion index. This tumbler index sensitivity is good because the sinter strength
is higher and more consistent (65% to 75%) compared to the sinter produced in the
past (45% to 52% using 11.4 kg sample). These represent the sinter strength prior to
the sinter underbin screen, which is comparable to international standards. The -6.4mm
x O sinter fines content is below 12% on the blast furnace top.
The introduction of the cylindrical sintering reactors with much lower amount of
sinter production resulted in an insuffient quantity of material for a standard ASTM
tumbler test, a 4.5 kg tumbler test was devised. After 200 revolutions at 70 rpm, the
tumbled sinter is screened through 12.5, 10 and 6.3mm screens. The tumbler index is
given as the cumulative % on 6.4mm. This tumbler index (84% to 93%) may be
considered fair because of the reduced tumbler feed mass. This represents the sinter
strength after the sinter underbin screen or blast furnace top (approximately 88% +
6.4mm).
J.I.S. Tumbler Test: 23 kg of -50 +10mm sinter recomposed to the original size
distribution is tumbled for 200 revolutions at 24 rpm. The tumbler index is the %
+10mm after tumbling, and the abrasion index is the % -5mm.
Tumbler dimensions are: 914mm I.D. x 557mm, 6mm thick, 2 lifters each 50mm.
4.3.1.3 Hot Sinter Strength
Development of hot sinter testing procedures and apparatus have been markedly
improved at Port Kembla Steelworks during the last two decades. Measurement of
the RDI index was introduced at the beginning of the 1970's, using the Linder apparatus
followed by the Nagoya unit.
Currently, the low temperature reduction degradation on sinter uses the JIS reduction
testing apparatus and the NSC test method. The reducibility of sinter is determined
using the Japanese industrial standard (JIS).
The hot sinter strength properties are measured by the following indices:
70
. Reduction degradation index (RDI), % -28mm at 550 ±5°C.
. Reducibility index (RI), % 0 2 removed from the iron oxide
at 900 ilO 'C
. Softening under load (SUL) between 1100eC- 1500^
Hence, the hot state properties of sinter simulates blast furnace operating conditions
from the lumpy zone (upper part of shaft) to the melting zone (lower part of shaft).
Reduction degradation index: A 500 gram of sample of 20 - 15mm material is heated

to 550^ under a nitrogen gas flow of 3 1/min. When thermal equilibrium has been
reached the gas composition is changed from 100% nitrogen to mixture of 70% nitrogen
und 30% carbon monoxide at 15 1/min.
The sample is reduced for 30 mins, at 550*C and then cooled to room temperature
under nitrogen gas at 3 1/min. The sample is then removed and tumbled for 900
revolutions at 30 rpm. After tumbling the sample, the product is screened over 10, 5
and 2.8mm sieves. The R.D.I. is the % passing 2.8mm, measured to the first decimal
place.
Reducibility index: A 500g sample of -20 +19mm material is held in the desiccator, its
mass (Wi) is measured, and the sample is placed into the reduction tube. The ther
mocouple is inserted into the sample layer close to the centre. The reduction tube is
loaded into the RI test electric furnace, connected to the weighing device, and the
gas circuit is completed.
After replacing the atmosphere in the reduction tube with inert nitrogen gas, the
temperature is raised to 900*C over one hour (60 - 90 min) while passing the inert gas
through the circuit. When the temperature reaches 900T!, this temperature is maintained
for 30 min.
The mass (Wo) of the sample is measured after the 30min time period, and then the
reducing gas (70% N2, 30%CO) is introduced immediately at 900 ±10°C for the reduction
duration of 180 min. The mass (Wf) of the sample is measured at 10 min intervals
for the first 60 min, and at 15 min intervals thereafter. The sample is then cooled
to ambient temperature within the flow of inert gas.
71
Calculation: The final degree of reduction is calculated using the analyzed values of
total iron and ferrous oxide from the chemical analysis sample before reduction (A,B),
and the measurement of the mass of sample during the reduction test. The following
equation is used, the result being rounded down to the first decimal place [56]:
Where, R = the final degree of reduction (%)

Wi =» the mass of the sample as weighed out (g)
Wo = the mass of the sample immediate before starting
the reduction (g)
Wf = the mass of the sample after 180 min reduction (g)
A = the total iron of the sample before reduction (%)
B » the ferrous oxide of the sample before reduction (%)
72
5.0 PRELIM INARY EXPERIM EN TA L WORK
5.1 E F F E C T S OF IRON ORE GANGUE AND U LTRA FIN ES

CO N TEN T ON SINTERIN G
5.1.1 Overview
The gangue in iron ores and sinter product is the sum of % (Si0 2 + A120 3 +CaO +
MgO) content. However, only the major components such as silica and alumina were
considered for this study; and the particles passing 152 microns were classified as
ultrafine material. .
Cockatoo and Koolan iron ore fines were selected for this study. Cockatoo iron ore
has higher sintering rate (22.2t/m2/24hrs) and produces stronger sinter compared to
Koolan Main Orebody type (14.0t/m2/day). However, anomalies occur within Main
Orebody Koolan types and sintering rates in excess of 22.2t/m2/24hrs may occur [58].
To enable a better understanding of the factors causing these differences, much of

the gangue and ultrafine particles contained in the crude fines (from both fine ore
types) were separated by hydraulic classification using cyclones.
Finally, the effects of exchanging the cyclone overflow material containing most of
the gangue and ultrafine particles were investigated by a gradual addition of the
Koolan overflow material to the Cockatoo classified product and vice versa.
5.1.2 Sample Preparation

A representative sample of 907 kg of Koolan iron ore fines was dry screened on 3.2mm
square mesh. The -7.9mm + 3.2mm screen fraction was stored to be blended with the
cyclone underflow material. This recomposed material will be referred to as the
classified product. The -3.2mm x 0 fraction was used as cyclone feed. A similar
amount of Cockatoo iron ore fines was prepared following the same procedure.
5.1.3 Cyclone Classification

A single Warman 76mm rubber-lined cyclone fed by a 45mm diameter pump and fitted
with a motor of 3 hp and 1415 rpm, was used for the hydraulic classification.
73
A cyclone underflow product slightly contaminated with some clay was obtained
during the first stage. Recycloning of this product was carried out in a second stage,
simulating the operation of a helical classifier to separate the clay. This cyclone
underflow product, containing smaller amounts of fine particles than the crude fines,
was dried and weighed, then reconstituted with the -7.9mm + 3.2mm dry screened
fraction and this blend was used for sintering tests.
Both the cyclone overflow products from the first and second stage were decanted,
dewatered, dried and weighed, and used for sintering tests.
. The hydrometallurgical process results are illustrated in Figures 5.1-1, 5.1-2 and
5.1-3.
5.1.4 Mineralogy
Microscopic observations of the classified products (dry screened and cyclone
underflow recomposed mix), and the cyclone overflow material were:
The Koolan Main Orebody classified product consisted of high grade hematite con
glomerate, hematite quartzite and siliceous hematite. These hematite ores varied from
compact and dense to soft, powdery and porous conglomerated particles.
The cyclone overflow fraction, consisting of 90.1% ultrafine (-152 micron) particles
and containing 5.8% gangue (Si0 2 + A120 3), revealed mainly ferruginous and chlorite
schists associated with small amounts of limonite, quartzite and clay minerals.
The Cockatoo classified product consisted of high grade ore, ranging from hard but
often friable steely blue almost pure hematite, to hematite schist and rich hematite
quartzite. These sedimentary hematites occurred in quartzose matrices, and varied
from hard to friable conglomerate particles.
The cyclone overflow material,consisting of 96.8% ultrafine particles and containing

21.3% gangue, consisted mainly of ferruginous schists and clay minerals associated
with minor amounts of quartzites and limonite. Fig 5.1.3 shows degradation in size
results by cyclone classification on Koolan and Cockatoo ores.
74
Fig 5.1-1 Metallurgical Breakdown
Koolan Crude Fines
Mass 100.0 % Dist

Fe 66.7 100.0
S i0 2 1.5 100.0
AI2Ó3 1.6 100.0
I.L. 1.2 100.0
+6.35mm 3.1
-152micron 61.3
Dry Screening
r !
Oversize (-7.9mm + 3.2mm) Undersize (- 3.2mm x 0)
% Mass 12.2 % Dist. % Mass 87.8
% Fe 64.0 11.8 % Fe 66.3.
% SiO , 2.0 19.0 % S i0 2 1.5
% A120 3 4.2 31.4 % A120 3 1.4
% I.L. 3.4 34.8 % I.L. 1.2
% T i0 2 0.33 % Solids 16.0
% +6.4mm 4.1 Feed pressure 241 kPa.
V ortex size 1 16mm dia.
Apex size | 6.4mm dia.
Cyclone - First Stage
1------
y f
Underflow Overflow
% Mass 65.4 % Mass 22.4
D ensity 1.8 Density 1.11
% Solids 1 57.3 % Solids 0.9
Y
Cyclone - Second Stage
Y
Underflow Overflow
% Mass 62.0 % Mass 3.4
D ensity 1.7 Density 1 1.1
%Solids 1 50.2 % Solids l 0.7 ^
t
i
Cyclone Underflow Product Total Overflow
% Mass 62.0 % Dist. % Mass 25.8 % Dist
% Fe 67.3 63.7 % Fe 63.6 24.5
% S i0 2 1.0 24.1 % S i0 2 3.1 56.9
% a i 2o 3 1.0 22.8 % A120 3 2.7 45.8
% I.L. 0.9 25.3 % I.L. 1.9 39.9
% T i0 2 0.25 - % TiOo 0.48 -
% -152micron 51.9 % -152micron 90.1
75
Fig 5.1-2 Metallurgical Breakdown
Cockatoo Crude Fines
% Mass 100.0 % Dist.

% Fe 64.5 100.0
% S i0 2 2.9 100.0
% AI2O3 2.2 100.0
% I.L. 1.2 100.0
% +6.35mm 7.7
% -152micro , , 48.6
!
Dry Screening
Oversize (-7.9mm + 3.2mm) Undersize (- 3.2mm x 0)

% Mass 16.7 °/o Dist. % Mass 83.3
% Fe 66.8 17.3 % Fe 64.6
% S i0 2 1.7 9.2 % S i0 2 3.2
% AI2Ö3 1.4 12.7 % A120 3 2.4
% I.L. 1.00 12.4 % I.L. 1.3
% T i0 2 0.15 % Solids 15.5
% +6.4mm 9.5 Feed pressure 207 kPa.
Vortex size 16mm dia.
Apex size ^r 6.4mm dia.
Cyclone - F irst Stage
i
U nderflow Overflow
% Mass 68.0 % Mass 15.3
D ensity 1.71 D ensity 1.10
% Solids 43.2 % Solids 1.1
Cyclone - Second Stage
Overflow
1
U nderflow
% Mass 64.5 % Mass 3.5
D ensity 1.86 Density 1.04
%Solids „ 62.0 % Solids , f 0-5 ,
I Total Overflow
Cyclone U nderflow Product
% Mass 64.5 % D isi % Mass 18.8 % Dist,
% Fe 67.8 68.7 % Fe 48.5 14.0
% S i0 2 0.7 13.3 % S i0 2 12.4 77.5
% AI2O3 0.5 11.5 % AI2Ö3 8.9 75.8
% I.L. 0.3 16.0 % I.L. 4.7 71.6
% T i0 2 0.19 - % T i0 2 0.29 -
% -152micron 38.5 % -152micron 96.8
76
5.1.5 Sintering Feed Materials Data
% wt
6.4 -152 % Analysis
Ore +mmi Micron Fe S i0 2 a i 2o 3 Gangue
Crude Fines
Koolan 3.1 61.3 66.7 1.5 1.6 3.1

Cockatoo 7.7 48.6 64.5 2.9 2.2 5.1
Classified Product (dry sceeened + cyclone underflow)
Koolan 0.7 43.4 66.8 1.2 1.5 2.7

Cockatoo 2.0 30.6 67.5 0.9 0.7 1.6
Cyclone Overflow Material
Koolan 0.0 90.1 63.6 3.1 2.7 5.8

Cockatoo 0.0 96.8 48.5 12.4 8.9 21.3
5.1.6 Sintering of the Crude Fines and Classified Products
The cyclone underflow fraction (-3.2mm x O) recomposed with the dry screened
fraction (-7.9mm + 3.2mm) for each iron ore type is called classified product. This
blend was used for sintering tests.
77
The effect of Koolan and Cockatoo iron ore gangue and ultrafines content on sintering
productivity and sinter quality were investigated by a gradual addition of the cyclone
overflow material to the classified product.
Sintering parameters:
Reactor Large unit
C a0 /S i0 2 ratio 1.1
Flux 100% limesand (Coffin Bay)
Coke in mix 5.5%
Return fines in mix 50%
Return fines size -7.9mm x O
Sintering feed moisture 5.9 ±0.1%
Return fines balance 1.0 ±0.05
Bed height 229mm
Suction 10 kPa
5.1.6.1 Test Results and Data Analysis

% % % Product Cum %Retained
Sintering +6.4 -152 ivity on 6.4mm
Feed Ore mm Micron Gangue t/m 2/d SI RI
a. Crude Fines
Koolan 3.1 61.3 3.1 14.0 54 54
Cockatoo 7.7 48.6 5.1 22.2 63 59
Difference,% -4.6 +12.7 -2.0 -36.9 -9 -5
b. Classified Product
Koolan 0.7 43.3 2.7 19.2 63 57
Cockatoo 2.0 30.6 1.6 20.6 65 63
Diff., % -1.3 + 12.8 + 1.1 -6.8 -2 -6
c. Koolan Main Orebody
Classified 0.7 43.4 2.7 19.2 63 57
Crude fines 3.1 61.3 3.1 14.0 54 54
Diff., % -2.4 -17.9 -0.4 +37.1 +9 +3
d. Cockatoo Ore
Classified 2.0 30.6 1.6 20.6 65 63
Crude fines 7.7 48.6 5.1 22.2 63 59
Diff., % -5.7 -18.0 -3.5 -7.2 +2 +4
78
5.1.6.2 Test Results and Discussion
a. Sintering of the Koolan crude fines gave a productivity of 14.0 t/m 2/24h with
54% size index and 54% rumbler index compared to a productivity of 22.2 t/m 2/24h
with 63% size index and 59% rumbler index for the Cockatoo crude fines.
A difference of 36.9% lower productivity with 9% lower size index and 5% lower
rumbler index was calculated for the Koolan crude fines compared to the Cockatoo
ore. This significant low productivity and poor sinter strength of the Koolan
ore is basically attributed to the 12.7% higher ultrafine particles content, 2%
lower gangue content and 4.6% lower +6.4mm grains content in the Koolan crude
fines. In addition, the Koolan ore fines mineralogical structure of dense to
powdery rich hematite containing 61.3% ultrafine particles, evidently detri
mental, affected the sinter bed permeability.
b. Sintering of the classified Koolan product gave a productivity of 19.2 t/m 2/24h
with 63% size index and 57% rumbler index compared to a productivity of 20.6
t/m 2/24h with 65% size index and 63% rumbler index for the recomposed Cockatoo
product.
In this case, a difference of only 6.8% lower productivity, with 2% lower size
index and 6 % lower rumbler index, was calculated for the Koolan classified
product compared to the Cockatoo recomposed product. This is mainly attributed
to the marked decrease in the ultrafines particles content from 61.3% in the crude
fines to 43.4 % in the classified product, and to a lesser degree to the 1. 1% higher
gangue content in the Koolan ore compared to the Cockatoo product.
c. As described before, the classified Koolan product gave a productivity of 19.2

t/m 2/24h with 63% size index and 57% rumbler index, compared to its crude
fines with a productivity of 14.0 t/m 2/24h, 54% size index and 54% rumbler
index.
A difference of 37.1% higher productivity, 9% higher size index and 3% higher

rumbler index, was calculated for the classified Koolan product compared to its
79
crude fines. This significant improvement in productivity and sinter strength
is particularly attributed to the 17.9% lower ultrafine particles content in the
recomposed Koolan product.
d. Also as described before, the classified Cockatoo product gave a productivity of

20.6 t/m 2/24h with 65% size index and 63% rumbler index, compared to its crude
fines with a productivity of 22.2 t/m 2/24h with 63% size index and 59% rumbler
index.
A difference of 7.2% lower productivity, but 2% higher size index and 4% higher
rumbler index, was calculated for the classified Cockatoo product compared to
its crude fines. The decrease in productivity is attributed to the 3.5% lower
gangue content and the 5.7% lower +6.4mm grains content. The increase in sinter
strength is due to the 18.0% lower ultrafine particles content in the recomposed
Cockatoo material compared to its crude fines.
80
5.1.7 Effects of the Koolan Overflow Material Addition
to the Cockatoo Classified Product, on Sintering.
5.1.7.1 Sintering Test Results
Koolan % % Produc Cum % Ret’d

O’F -<-6.4 -152 % tivity on 6.4 mm
% wt mm Micron Gangue t/m 2/d SI RI
0.0 2.0 30.6 1.6 20.6 65 63

5 1.9 33.6 1.8 20.7 66 60
10 1.8 36.6 2.0 21.1 67 59
20 1.6 42.5 2.5 21.7 67 58
30 1.4 48.5 2.9 21.1 67 58
5.1.7.2 Data Analysis on test Results
5.0 1.9 33.6 1.8 20.7 66 60

0.0 2.0 30.6 1.6 20.6 65 63
Diff.% -0.1 +3.0 +0.2 +0.5 +1 -3
10.0 1.8 36.6 2.0 21.1 67 59

0.0 2.0 30.6 1.6 20.6 65 63
Diff.% -0.2 +6.0 +0.4 +2.4 +2 -4
20.0 1.6 42.5 2.5 21.7 67 58

0.0 2.0 30.6 1.6 20.6 65 63
Diff.% -0.4 + 11.9 +0.9 +5.1 +2 -5
30.0 1.4 48.5 2.9 21.1 67 58

0.0 2.0 30.6 1.6 20.6 65 63
Diff.% -0.6 +17.9 +1.3 +2.4 +2 -5
81
5.1.8 Effects of the Cockatoo Overflow Material Addition
to the Koolan Classified Product, on Sintering
5.1.8.1 Sintering Test Results
Cockatoo % % Produc Cum % Ret’d

O’F +6.4 -152 % tivity on 6.4 mm
% Wt mm Micron Gangue t/m 2/d SI RI
0.0 0.7 43.4 2.7 19.2 63 57

5 0.7 46.1 3.6 18.5 61 56
10 0.6 48.7 4.6 18.7 63 58
20 0.6 54.1 6.4 18.9 64 . 59
30 0.5 59.4 8.2 19.5 67 62
5.1.8.2 Data Analysis on Test Results
5.0 0.7 46.1 3.6 18.5 61 56

0.0 0.7 43.4 2.7 19.2 63 57
Diff.% 0.0 +2.7 +0.9 -3.8 -2 -1
10.0 0.6 48.7 4.6 18.7 63 58

0.0 0.7 43.4 2.7 19.2 63 57
Diff.% -0.1 +5.3 +1.9 -2.7 0 +1
20.0 0.6 54.1 6.4 18.9 64 59

0.0 0.7 43.4 2.7 19.2 63 67
Diff.% -0.1 +10.7 +3.7 - 1.6 +1 +2
30.0 0.5 59.4 8.2 19.5 67 62

0.0 0.7 43.4 2.7 19.2 63 57
Diff.% -0.2 +16.0 +5.5 +1.5 +4 +5
82
% Wt Degradation %Wt degradation in size = [ f - ( \l - o) ] 100
Fig 5.1-3 Degradation in size by cyclone classification on Koolan

and Cockatoo iron ore fines (-3.2mm x o )
83
Productivity (t/m 2 /2 4 hour») Productivity (t/m 2 /2 4 houro)
22
20 18
30 35 « 45 50
40 45 50 55 60
(a) Ultrafineo (% -152 micron) (c) Uttraflncs (% -152 micron)
22 20
20
t.6 1.8 2 22 2.4 2.6 2.8 2 3 4 5 6 7 8
(b) Gangue (XS102 + %AI203) (d) Gangue (%SI02 + XA1203)
Figs (c) 5.1-7 and (d) 5.1-8
Figs (a) 5.1-4 and (b) 5.1-5
Relationship between sintering Relationship between sintering productivity
productivity and the uHrafine and and tho ultraflnes and gangue content
ganguo content In tho fc-on ore mix In tho iron ore mix (Cockatoo Cyclone
(Koolan Cyclone Overflow Material Overflow material added t o tho
added t o the Cockatoo classified Koolan classified base product).
base product).
22 20
21 - 19 -
20 ü 18
0 8 10 20 30 0 5 10 20 30
Koolan Cyclone Overflow (%Wt) Cockatoo Cyclone Overflow (XWt)
Fig 5.1-6 Fig 5.1-9
Sintering productivity as a function Sintering productivity as a function

of the Koolan Cyclone Overflow material of Cockatoo Cyclone Overflow material
added t o the Cockatoo classified base added t o the Cockatoo classified base
product product
84
Size Index (%} SI (%)
Gangue (%Si02 + XA1203)
Fig 5.1-10 Fig 5.1-12

Relationship between size index and Relationship between size index ant
the gangue content in the iron ore the gangue content of the iron ore
mix (Koolan Cyclone Overflow material mix (Cockatoo Cyclone Overflow
added to the Cockatoo classified material added to the Koolan
base product) . classified base product).
Rutnbler Index (%) RI (X)
62 -
59
60
68 -
57
56 55
1.6 1.8 2 2 .2 2 .4 2 .6 2 .8 i 2 3 4 5 6 7 8 9
Gangue (%Si02 + XA1203) Gangue (%Si02 + XA1203)
Fig 5.1-11 Fig 5.1-13
Relationship between rumbler index Relationship between rumbler Index

and the gangue content of the iron ore and the gangue content in the iron
mix (Koolan Cyclone Overflow material ore mix (Cockatoo Cyclone Overflow
added to the Cockatoo classified base material added to the Koolan
product). classified base product) .
85
Size Index (%)
68 J r 0-2*1 X .+ 48.S>
66
64
o .io x + 6 2 . &
a. 7?
62
60
30 35 40 45 50 40 45 50 55 60
(a) Ultra fines (%—152 microns) (c) Ultrafines (X-152 microns)
Rumbler Index (%) RI (X)

64 62 vi~ r\. 34 DC + A 1.5
30 35 40 45 50 40 45 50 55 60
(b) Ultrafines (X-152 microns) (d) Ultra fines (X-152 microns)
Figs (a) 5.1-14 and (b) 5.1-15 Figs (c) 5.1-17 and (d) 5.1-18
Relationship between size index, rumbler Relationship between size index,

index and the ultrafines content in the rumbler index and the ultrafines
iron ore blend (Koolan Cyclone Overflow content in the iron ore mix
material added to the Cockatoo (Cockatoo Cyclone Overflow material
Classified Base Product). added to the Koolan Classified
Base Product).
68 68
66 ... .SI., ix)) . .. 66
SI (%)^*
*#... . / ****
y
64 64 ......
62
62 'RÏ'ïxi'?"'
60
60
58
h-"~~.^RI (%}
58 56
66 ....•. 54
0 5 Î0 20 30 o 5 10 20 30
(%Wt) (e)
Fig 5.1-16 Fig 5.1-19
Sinter strength as a function of the Sinter strength as a function of the

Koolan Cyclone Overflow material added Cockatoo Cyclone overflow material
to the Cockatoo classified base product. added to the Koolan Classified base
p rod u ct.
86
Figs 5.1-4 to 5.1-9, illustrate the relationship between sintering productivity and the
ultrafine and gangue content in the iron ore mix.
Figs 5.1-10 to 5.1-13, show the relationship between size and rumbler indices and
gangue content in the iron ore mix.
Figs 5.1-14 to 5.1-19, indicate the relationship between size and rumbler indices and
ultrafine content in the iron ore mix.
5.2 CO N CLU SIO N S
1. The mineralogical structure of iron ores and their mode of occurrences determines
the size distribution and the gangue content.
2. The size distribution is the most important factor affecting both sintering pro
ductivity and sinter quality.
3. The gangue content is another prima factor in sintering productivity, but to a

lesser degree in sinter quality. The gangue content is particularly beneficial in
sintering when it is contained in the ultrafine fraction, provided that this material
is effectively pelletized during the mixing-granulation of the greenfeed, and
that the granules are physically hard enough to maintain a consistent high
permeability during the sintering process.
87
5.3 CH A R A CTER IZA TIO N OF T H E RAW M ATERIALS
5.3.1 Overview
. C om parison of the rectangular sintering reactor u n it results versus the cylindrical

reactors was conducted, and then each reacto r was compared to the in d u strial No2
S intering M achine, to ensure the rep ro d u cib ility of the experim ental works and to
assess the degree of variab ility [76].
. E ffe cts of the m oisture level, each raw m aterial level and th eir respective size
d istrib u tio n on sintering were investigated, using a constant base ore mix and constant
sintering param eters [69]:
- R etu rn fines
- Coke breeze
- Flux, including effects of various types of fluxes, and
- V arious types of iron ore fines
. F inally, e ffe cts of the proportioning of the miscellaneous m aterials on sintering were
also investigated:
- Flue dust
- M ill scale
- F ilte r cake
- Jaspilite
- Serpentine, etc.
As a result, the fun d am en tal experim ental work was conducted on optim ised sintering
param eters, using a projected sintering plan t feed mix and selected sintering c h ar
acteristics of raw m aterials.
88
6.0 BA SIC EXPERIM EN TA L WORK
6.1 Overview
The quality of sinter produced at Kwinana (WA) prompted a review of the investi
gations made previously in the Port Kembla Sinter Research Section. This raised two
specific points:
a. Physical strength of the Kwinana sinter was reported to be high, and this
appeared to be anomalous for operation at a basicity of about 1.4 to 1.6, an area
recognised generally to be a trough of physical quality expressed by size and
rumbler indices.
b. Record sinter production levels and iron make occurred subsequently using 80%
sinter burden of basicity 1.8 to 2.0, with crushed Japanese limestone comprising
50 to 100% of the sintering flux.
It was known from Kwinana that the Dowd's Hill ore used during the later period
was of improved quality, with silica in the sinter not exceeding 6 %, and that it was
well blended in the primary bed. It was likely that sinter productivity, exceeding
32.8t/m2/24 hrs, could be explained in terms of size distribution and bed permeability,
particularly influenced by the Japanese limestone which is known to crush to a coarse,
wide distribution.
Experimental studies with a comparable ore mix (i.e. containing 20% Koolan fines)
showed a trough, albeit somewhat shallow, at the basicity level about 1.5. It followed
that Kwinana sinter was relatively stronger in relation to basicity than Port Kembla's.
A mineralogical examination was made to explain this.
The mineralogical structures of Port Kembla sinters were mainly magnetite in glass,
and magnetite and calcium ferrite in glass. The significant point was that the
distribution of phases indicated good assimilation of the gangue and flux, conducive
to effective bonding, This had been the philosophy derived from the basic studies
during the 1960's and from reactor tests utilising gangue (blast furnace slag) additives
during 1970. However, in Kwinana's circumstances, the sinter strength appeared to
have been achieved by thermal conditions in the bed [53, 57, 58].
Snap samples of three sinters of basicity of about 2.0 were microscopically examined.
Apart from some microscopic free lime, the structure was uniform, principally calcium
ferrite, magnetite and hematite. The assimilation again was good (homogeneous sinter
texture). This assimilation may have been due to some of the following factors:
89
i. high temperature during sintering,
ii. high solubility of gangues and lime in the sintering liquids,
iii. greater reduction during the early stages of sintering, resulting in high FeO
content. The final FeO content of the Kwinana sinter was about 13% at
basicity 1.5.
The presence of unusual quantities of hematite in the sample of basicity 2.0 and the
high ferrite content suggested high oxidising conditions during the post-ignition stage.
This could be due to bed permeability favouring rapid and effective passage of air.
High ferrite content, low glass and the presence of hematite would result in high
reducibility [53, 57, 58].
The various explanations put forward as to how Kwinana operators achieved this
favourable result were all relatively subjective in nature. The preliminary observation
led to a request for more extensive and reliable data at Kwinana and further
investigation at Port Kembla [57].
Twenty six 44-gallon drums containing Kwinana plant materials were received at Port
Kembla on 27 January 1971 [30]. All these samples were taken at the Kwinana plant
over a stabilised 5 hours of operation on 1st December 1970 [59].
6.1.1 Correlation between the Port Kembla Experimental

Reactor Unit and the Kwinana Sintering Plant
A direct correlation between the Port Kembla experimental reactor unit and the
Kwinana sintering plant was carried out using the materials taken in the Kwinana
plant. Furthermore, the Port Kembla sintering plant feed mix was also investigated
over the same period using the same sintering reactor. Similar assessment criteria
were used in order to establish the differential effect of the Port Kembla raw materials
on sintering, and the sinter characteristics produced at Port Kembla operating
conditions. Results are shown in Figs 6.1.1-1 to 6.1.1-7 and in Tables 6.1.1-1 to
6.1.1-3.
Reasons for the high sintering rate and the high quality of the Kwinana product
sinter were investigated by L. Miinive and colleagues [30]. The basic hypothesis
attributed the high quality sinter and productivity of the Kwinana sintering plant to
bed depth, basicity and suction, and hence to the thermal conditions prevailing during
the sintering process [30].
Investigation of the hypothesis postulated above was initiated in 1971 and is the basis
of this thesis.
90
Table 6.1.1-1 Sintering process and results on Kwinana and
Port Kembla Sintering Plant raw materials
100% Dowd’s Hill ore Port Kembla feed mix
PK Kwinana PK PK
Sintering Reactor Sinter Sinter Reactor
Parameters Unit Plant Plant Unit
Flux 1:1 JL/W1 1:1 JL/W1 100% CB1 100% CB1
RF in mix, % 40 40 41 41
Coke in mix, % 5.5 3.7 3.9 5.9
Moisture in mix, % 6.0 4.9 5.9 5.8
Bed depth, mm 254 381 254 254
Basicity, CaO /Si0 2 1.6 1.6 1.2 1.2
Suction, kPa 9 15 9 9
Ign time, sec. 75 120 90 75
Ignition fuel LP gas Fuel oil C.O.gas LP gas
Sintering Results
PR, t/m 2/d 28.9 34.5 27.8 26.5
Size index, % 63 - _ 62
Rumbler index, % 59 70 66 56
Sinter Analysis
FeO % 10.7 9.0 11.3 13.9
Total Fe % 59.1 59.1 59.1 59.3
CaO % 8.4 8.0 6.8 6.9
S i0 2 % 5.2 5.1 5.8 6.0
AU03 % 2.1 2.2 2.8 2.8
CaO /Si0 2 ratio 1.62 1.57 1.17 1.15
Mineralogy (vol %)
Calcium ferrite 18 18 11 10
Hematite 16 18 28 18
Unreacted ore 10 18 0 0
Magnetite 39 29 39 51
Glass 17 17 22 21
Total 100 100 100 100
Code:
Reactor Unit = Experimental reactor.
91
Table 6.1.1-2 Size distribution and chemical analyses of Kwinana and
Port Kembla Sintering Plant feed ore, coke and flux.
Size Distribution: Cumulative % Passing
Flux Flux
size Feed ore Coke; breeze Kwinana Port Kembla
mm K PK K PK JL W1 CB1 ML
9.5 98 100 100 100 100
6.4 87 90 97 98 100 98
3.2 65 68 88 88 92 87
1.68 49 53 76 66 60 100 73
0.85 35 45 61 45 32 100 99 56
0.42 27 38 50 31 21 98 79 46
0.30 23 32 41 24 16 85 52 41
0.152 16 27 23 14 11 29 5 34
Chemical Composition (%)

Total Fe 60.2 60.5 - - - - -
CaO - 0.43 1.9 54.4 48.6 51.3 52.0

SiC>2 48 6.3 10.2 0.7 4.9 0.9 2.9
AI2O3 1.9 3.1 4.5 0.3 0.6 0.2 0.9
MgO 0.1 0.5 0.2 0.6 2.7 2.7 0.5
P 0.13 0.07 - - - - -
Ign loss 5.7 2.4 81.3 43.3 42.2 43.6 41.4

Ash % - - 18.3 - - - -
Code:
K = Kwinana, Western Australia
PK= Port Kembla, NSW, Australia
JL= Japanese limestone
Wl= Wanneroo limesand
CB1= Coffin Bay limesand
ML= Marulan limestone
92
Table 6. 1.1-3 Size distribution and chemical analyses of Kwinana
and Port Kembla sintering Plant hearth layer and
return fines
Size Distribution
Cumulative % Passing
Size Hearth layer Return fines
mm K PK K PK
25.4 100
19.1 100 83
12.7 98 53 100
9.5 69 31 99
7.9 60 25 98
6.4 23 14 100 89
3.2 1 1 31 24
Chemical Composition (%)
FeO 9.0 15.0 8.0 14.4
Total Fe 59.2 57.9 59.6 58.4
CaO 8.2 7.2 7.6 7.0
Si0 2 5.4 6.3 5.1 6.3
AI2O3 2.3 3.3 2.0 3.2
MgO 0.3 0.5 0.3 0.5
P 0.13 0.05 0.13 0.05
CaO/Si0 2 1.52 1.14 1.49 1.11
Code:
PR= Productivity or production rate= t/m 2/d= t/m 2/24hrs
SI= Size index, + 6.35 mm %
RI= Rumbler index, + 6.35 mm %
0.152 mm= 152 microns = ultrafine particles
Sintering feed ore = Iron ore fines, ore mix
F= Primary fines
PSF= Secondary fines
93
6.2 Some Fundamental Facts
6.2.1 Effects of Ore Feed Size Distribution on Sintering
On the basis of the broadly known effects of the size distribution of raw materials
and the bed permeability on sintering, and considering the degree of sensitivity and
sizing cost, all particles passing 152 micron screen were considered as ultrafine material.
Hence, the percentage of particles passing 152 microns was defined as the ultrafine
content of the feed materials and the largest (top) size particle defined, particularly
for return fines, coke breeze and limestone flux, was 6.35mm.
The typical ranges of ultrafine particles and gangue content of Port Kembla raw
materials were:
% Ultrafines % Gangue
Iron Ore Fines -152 micron (S i02 + a i 2o 3>
Koolan 53 - 61 2.5 - 3.5
Cockatoo 49 - 58 4.5 - 5.5
Whyalla 21 - 26 5- 15
Dowd's Hill 16 - 29 8- 16
Mt Newman PF 18 - 25 6- 11
Mt Mewman SF 16 - 17 6- 12
Ferruginous Feed
Filter coke 85 - 87 9 - 14
Flue dust 35 - 43 9 - 15
Fuel and Flux % Ash
Coke breeze 15 - 16 15 - 26
Marulan limestone 2 7 -4 7 -
The fines content of the ores is outside the plant o p e ra to r control. However, the
proportioning of the sintering ore mix is under operators' control.
The influence of size distribution on bed permeability, and hence sinter production
and quality was known. Therefore, the ore mix was aimed at a maximum of 24%
ultrafines.
Table 6.2.1-1, illustrates the effect of return fines level on sintering.
The addition of Koolan and Cockatoo fines and some ferruginous materials such as
the filter cake and flue dust were minimised due to the high content of ultrafines in
these materials.
94
As may be seen in Table 5.1.5, the ultrafines particles content in Koolan and Cockatoo
ore are the highest compared to other ores and their effects are detrimental in
productivity and sinter quality (higher RDI). However, their gangue content is the
lowest compared to the other ores. Hence, the use of Koolan and Cockatoo ore in a
limited amount is beneficial, decreasing the silica content of the sinter.
The role of iron ore gangue content in sintering is experimentally demonstrated by

the exchange of Koolan gangue (cyclone overflow material) in the Cockatoo classified
material and vice versa. There is no effective sintering without sufficient gangue
components in the feed mix, as these are the melt-formers, and the main bonding
phase in terms of sinter strength and reducibility. Figs 5.1-4 to 5.1-19, show the
relationship between sintering characteristics and the ultrafine and gangue content
in the iron ore mix.
6.2.2 Effects of Limestone Flux and Coke Breeze

Size Distribution on Sintering
Improvement in the size distribution of these two materials can be achieved by the
operators. The Marulan limestone and coke size distributions were improved by
crushing using different types of crushers. The top size was standardised at 6.35mm
mesh and the ultrafines content was decreased to:
% Ultrafines
- 152 micron
Coke breeze 1 1 -1 4
Marulan limestone 1 7 -2 0
Table 6.2.2-1, illustrates the effects of Marulan limestone flux size distribution on
sintering.
The improvement of the size distributions of the Marulan limestone and the coke
significantly increased productivity and sinter quality. The sintering feed mixes in
which the Marulan limestone contained 12% and 17% ultrafines gave a productivity
higher than those with Japanese limestone crushed in Japan and in Australia, which
during the correlation between Port Kembla experimental reactor and the Kwinana
sintering plant, gave higher productivity than the feed mix with the Marulan limestone
containing 34% ultrafines [60].
95
The use of limesand also produced more magnetite, hence a higher FeO content, which
was due to the required use of a higher coke rate. Table 6.2.2.-2, shows the influence
of different flux types on sintering.
The improvement of the coke size distribution was significant because the effect of
coke level in sintering is sensitive and severe. Consistency of the coke size distribution
and of the coke level in the feed mix are essential, as the coke controls the thermal
conditions in the combustion zone. Minor changes in the thermal condition during
the sintering process vary the sinter mineralogy, hence the sinter properties. This was
the key factor for the development of the Sintering Process Model III.
The relatively high coke rate used in the experimental sintering process, compared to
the plant, was mainly due to the small mass of sinter cake produced, and to some
extent to the inefficient insulation of the reactor unit walls. This insulation was
improved by increasing the refractory thickness and by gradual maintenance and
repair. As a result, the coke level decreased slightly, closer to the plant value.
Table 6.1.1-1 shows that the mineral phases present were similar for the experimental
reactor and the plant, with the exception of the magnetite, followed by the hematite
and the unreacted ore. This is attributed largely to the reducing conditions during
the experiment which were caused by the high coke level and low oxygen potential.
As a result, the rumbler indices for the experimental sinters were lower than those
for Kwinana and Port Kembla plants.
The experimental and plant sinters, produced using the Port Kembla raw materials,
consisted of exsolution magnetite-glass, hematite and calcium ferrite phases with no
unreacted ore, and were homogeneous in texture. This homogeneous texture is
characterised by internal stressing and microcracking, which results in poor physical
characteristics [61].
However, the Kwinana plant sinter was heterogeneous in texture, and consisted
basically of unreacted ore particles embedded in a homogeneous matrix of exsolution
hematite, magnetite, calcium ferrite-glass phases. This type of bond is stronger than
that of the above Port Kembla sinter.
96
Table 6.2.1-1, Effect of return fines level on sintering.
Constant Parameters: Variable Parameters:

Feed mix "M4" RF in mix (%)
Flux 100% ML Coke in mix (%)
Flux ultrafine (%) 17 RF size -6.4 and -4.8mm x 0
Suction (kPa) 10 RFB (ratio)
Bed depth (mm) 250
Nominal basicity 1.0
Sintering Process and Results

RF size in feed mix, -6.35 mm x 0
RF in mix, % 50 45 40 35 30 25
Coke in mix, % 4.5 4.8 5.3 6.2 9.0 10.7
RFB, ratio 0.90 0.93 0.93 1.00 1.03 1.03
PR, t/m 2/24h 28.5 29.9 32.1 32.0 29.0 27.6
SI, +6.4 mm % 55 58 63 65 69 74
TI, +6.4 mm% 51 52 52 51 51 50
Sinter Sizing (cum % retained)
+ 25mm % 1 4 4 2 1 1
+ 10mm % 34 36 39 39 44 37
+ 6.4 mm % 55 58 63 65 69 74
RF Size in feed mix, -4.76 mm x 0
RF in mix, % 50 45 40 35 30 25
Coke in mix, % 2.9 3.0 3.3 3.5 4.1 5.0
RFB, (-4.8) 0.90 0.89 0.85 0.94 0.87 0.92
PR, t/m 2/24h 37.4 37.8 38.7 42.1 40.9 39.3
SI, +4.8 mm % 55 60 66 67 74 77
TI, +6.4 mm % 52 51 51 53 51 48
Sinter Sizing (cum % retained)
+ 25 mm % 2 3 3 3 6 4
+ 10 mm % 20 22 28 30 36 30
+ 6.4 mm % 38 41 50 48 58 52
+ 4.8 mm % 55 60 66 67 74 77
97
Table 6.2.2-1, Effect of Marulan flux size distribution on sintering.

Feed mix "M4" Coke in mix (%)
RF size (mm) -6.35 Flux, top size (%) +6.35 mm
Flux 100% ML Flux, ultrafines (%) -152 microns
RF in mix (%) 30 RFB, (ratio)
Suction (kPa) 10
Bed depth (mm) 460
Normal basicity 2.5
Sintering Process and Results

Flux size, +6.4mm % 36 1 Nil Nil Nil
Flux size -152micron % 12 17 26 37 41
Coke in mix, % 5.6 5.6 4.9 5.0 4.1
RFB, ratio 1.07 0.93 1.00 1.00 1.00
PR, t/m 2/24h 39.7 38.7 33.6 32.8 27.8
PR, tFe/m 2/24h 21.9 21.4 18.7 18.2 15.5
SI, +6.4 mm, % 68 72 70 70 70
TI, +6.4 mm, % 91 93 92 92 91
Sinter Analysis
FeO % 5.1 4.8 5.1 7.7 7.7
Total Fe % 55.1 55.4 55.8 55.4 55.9
CaO % 13.1 12.6 12.5 12.7 12.9
S i0 2 % 5.2 5.3 5.2 4.8 5.1
AI2O3 % 3.2 3.2 2.9 2.9 2.6
CaO /Si0 2 ratio 2.5 2.4 2.4 2.6 2.5
Code:
JL = Japanese limestone CB1 = Coffin Bay limesand
ML = Marulan limestone W1 = Wanneroo limesand
RBL = Rapid Bay limestone
98
Table Ó.2.2-2 Influence of flux types on sintering
Constant Parameters: Variable Parametes:

Feed mix "MX" Flux types
RF size (mm) -6.35 RF in mix (%)
Suction (kPa) 10 Coke in mix (%)
Bed depth (mm) 250 RFB (ratio)
Nominal basicity 1.2, 1.8
Basicity, 1.2 JL ML RBL CB1 W1

RF in mix, % 40 40 40 45 45
Coke in Mix, % 4.8 4.8 5.5 5.5 5.5
RFB, ratio 0.98 0.93 1.00 1.00 1.00
PR, t/m 2/24h 32.0 30.8 29.9 27.4 26.7
PR, tFe/m 2/24h 19.1 18.2 17.6 16.1 15.6
Size index, % 61 63 60 55 55
Tumbler index, % 52 57 53 52 57
FeO % 12.0 13.2 13.3 14.7 13.3
Total Fe % 59.6 59.2 59.0 58.8 58.4
CaO % 7.7 8.6 8.5 8.0 7.8
SiOo % 6.2 6.7 6.7 6.0 6.5
AloU« % 2.5 2.6 2.6 2.4 2.6
C ab /S i0 2 1.2 1.3 1.3 1.3 1.2
Basicity, 1.8
RF in mix % 40 40 40 35 45
Coke in mix % 4.9 4.7 5.0 5.5 5.5
RFB, ratio 0.95 0.98 0.93 1.03 1.00
PR, t/m 2/24h 32.3 31.2 30.6 27.6 23.5
PR, tFe/m 2/24h 18.8 18.1 17.7 15.8 13.3
Size index, % 62 61 63 64 55
Tumbler index, % 57 58 55 52 51
FeO % 8.4 9.6 9.5 10.0 7.8
Total Fe % 58.3 58.0 57.8 57.3 56.7
CaO % 10.5 10.1 11.5 10.3 12.5
Si02 % 5.5 5.4 6.3 5.5 6.5
AloO* % 2.3 2.2 2.0 2.0 2.0
CaO /Si02 1.9 1.9 1.8 1.9 1.9
Flux Quality:
Total CaO % 54.4 52.0 51.2 51.3 48.3
-152 micron % 11 34 21 5 29
99
Fig 6.1.1-1 shows the unreacted hematite (UH) and unreacted magnetite (UM)
which had not been assimilated in the sintering liquid. (Specimen from
Kwinana plant sinter, with basicity of 1.6 [30]). Magnification x 200.
Fig 6.1.1-2 illustrates the magnetite (M), hematite (H), calcium ferrite (CF),
UH and silicate glass (C). (Specimen representing the Port Kembla experimental
reactor sinter using Kwinana raw materials [30]). Incident light 200x.
100
Fig 6.1.1-3 shows the typical hematite glass calcium ferrite structure sinter
containing an additional component: dicalcium silicate (DS). Specimen repre
senting the Kwinana plant sinter. In this case the hematite content is greater
than that of magnetite [30]. Incident light 200x.
Fig 6.1.1-4 shows the typical magnetite glass calcium ferrite structure sinter
with an additional component: pores (P). (Specimen from the PK experimental
reactor sinter using Kwinana raw materials). In this instance the magnetite
content is much higher than the hematite content [30]. Magnification x 200.
101
Fig 6.1.1-5 illustrates the typical magnetite glass structure sinter with magnetite
(M) and silicate glass (G). (Specimen representing the experimental PK reactor
sinter using PK feed mix, at a basicity of 1.2 [30]).Incident light 200x.
Fig 6.1.1 -6 shows the typical magnetite glass phases sinter, illustrating micro
cracks. (Specimen representing the experimental PK reactor sinter using PK
feed mix, with basicity of 1.2 [30]). Incident light 200x.
102
Fig 6.1.1-7 illustrates a typical homogeneous sinter texture (a) and heterogeneous
texture (b) [61].
6.3 Development of Sintering Process Models
6.3.1 Overview
In general terms, the sintering mix components were constant for each test series,
except two control variables: the coke and the return fines input in the feed mix. The
coke rate or the return fines level was adjusted to maintain the fines balance close
to unity.
The return fines balance is the ratio between the return fines produced (output) and
the return fines input. A high precision sintering work requires a return fines balance
ratio of 1 + 0.05. However, a ratio of 1 + 0.1 is acceptable.
The typical sinter property ranges for blast furnace burden material are:
Tumbler index (TI) >65%
Reduction degradation index (RDI) 36 - 39%
Reducibility index (RI) 59 - 66%
FeO content 4.5 - 6.5%
Product sinters with specifications outside of these ranges are considered poor quality
burden for blast furnaces.
In addition, the dissociation and melting temperatures of the sinter microstructural

components are also important, the lower they are being more economical for blast
furnace operation. Such temperture ranges are:
°C
Silica glass 1200 - 1250
Calcium ferrites 1210 - 1436
Hematites 1250 - 1395
Magnetites 1370 - 1590
Dicalcium silicates 2130
The microstructural composition of the sinter presented above is in the simplest form.
In fact, each component crystallises in different forms, mainly related to the thermal
conditions in the combustion zone. Hence, each form for each component has its own
temperature of dissociation and melting point; but they are within the ranges given
above [62 to 68].
Figs 7.1.1-1 to 9.1.2-3 illustrate the quantitative variations or fluctuations of these

microscopic components related to the three variables studied (bed depth, basicity and
suction) and show the structure by which improvement in sinter quality was achieved.
104
6.3.2 Sintering Process Model I
This is the traditional model, where both control variables, the coke rate and the
return fines level, may be varied to adjust the return fines balance close to unity.
As may be seen in Fig 8.1.1-1, some of the observation points are scattered due to the
fact that either the coke or the return fines level were adjusted, e.g., from five sintering
tests for each conbination of parameters, three were done adjusting the coke rate and
two the return fines level. This indicated that three changes in the thermal conditions
in the combustion zone occurred, and a further two changes in the bed permeability
and thermal conditions when the return fines level was adjusted to maintain the return
fines balance close to unity.
6.3.3 Sintering Process Model II
In Model II, one control variable, coke rate in the feed mix, was held constant. The
other control variable, the return fines level, was adjusted to maintain the return
fines balance close to unity for each combination of parameters.
This model resulted in higher correlation than Model I, due to the fact that the bed
permeability changed gradually, producing faster flame front speed, more highly
oxidising conditions and lower sintering time. The thermal conditions changed
indirectly as a result of the bed permeability change.
The sensitivity of the return fines level change started from approximately 1 %, and
the effect on sintering was less severe compared to coke rate.
However, the coke rate was adjusted if necessary after each series of tests or when
the degree of change in the range of parameters is significantly large, but was then
held constant during any one test series to suit each series.
6.3.4 Sintering Process Model III
In this model, one control variable, return fines, was held constant, while the coke
rate was the control variable for the whole test series. Any minor change of the coke
rate, to adjust the return fines balance close to unity, directly affects the thermal
conditions in the combustion zone.
105
The sensitivity of the coke rate change was high, in the order of 0.1%, and the effect
on the bed permeability was much less than the return fines as the mass of coke used
to adjust the return fines balance close to unity was very small. However, the effect
of this variable component is more severe than the effect of the return fines on the
thermal conditions. As a result, the Model III gave an orderly set of experimental
results.
One of the most significant developments was that when using the Sintering Process
Model III, the magnetite content in the microstructural composition of the sinter,
which in the Model II follows a positive gradient, in Model III is reversed to a negative
slope.
This change in the magnetite slope is one of the most important contributions to
sintering technology, since one of the prime aims was to decrease the microstructural
components with high dissociation and melting temperatures and to provide high sinter
strength.
In the same manner as for the magnetite content, the gradients for total hematite,
calcium ferrite and dicalcium silicate microstructural components were reversed in a
favourable way. However, in the RDI relationship, the negative slope was changed to
positive, but the RDI value, to some extent, is not significant providing a critical level
is achieved (Table 7.1.2-1).
106
7.0 E F F E C T S OF BED D EPTH ON SINTERING
7.1 Overview
World wide computer searches through the Dialog System of the Lockheed Corporation
California, USA and the British Library in the UK, were conducted in 1972 for
references to the effects of bed depth (deep bed, bed height) on sintering. No
international articles, books, patents or theses were published on this technology until
1974, when Italy first and subsequently Russia published regularly. The articles
contained very few details and emphasised only the effects on sintering, which verified
the findings of L. Miinive during the experiments conducted in 1971 and plant trials
in 1972 [30, 69].
Japan, one of the most technologically advanced countries in this field used the bed
depth technology only in a limited range (275 to 410mm). This information was
obtained by direct contact in 1972
Both Newcastle and Port Kembla Steelworks investigated this technology in parallel
during 1971 and 1972. The Newcastle Steelworks report No NC/SP/72/002 on deep
bed trials summarised as follows:
"Little benefit was obtained from deep bed operations for the particular bed height,
ore mixes and basicity used, the only significant advantage being a reduction in coke
rate and increased Nagoya index".
However, the Port Kembla report No R. 24/1172, written by MUnive L. [69] reported
the successful result of a comparative study made between the bed depth of 254mm
(routine plant operation) and the proposed bed height of 508mm. The sinter quality
and productivity were significantly improved as follows:
i. An increase in size and tumbler indices, 66% to 76%.

ii. A decrease of 7% to 39% in coke consumption,
iii. An increase of 7% to 18% in iron unit productivity.
107
The fundamental reasons why the bed depth technology in sintering was very slowly
developed are:
1). Although the sintering technology was invented in 1896, the downdraught D
& L process with improved control facilities commenced production in 1926
with a bed depth of 102mm to 203mm [12].
2). In 1950, theories were developed to correlate the effects of bed permeability,
bed depth and other factors controlling the rate of sinter production, using
a bed depth of 267mm before ignition [72].
3). In 1953, Voice, Brooks and Gledhill [73] put forward the formula on sintering
bed permeability:
P = (F/A) (h/s)n, Where n = 0.6
They were careful to point out that only in the case of the permeability of
the cold feed mix before sintering (pre-ignition permeability) could this be
regarded as a precise relationship, and that other permeabilities could only
be regarded as "effective" since a variety of complex reactions were occurring
simultaneously [73,74].
The above formula shows that a relationship exists between bed depth (h)
and suction under bed (s) with the permeability index (P). As a result, a bed
depth of 330mm was considered as a very high level bed.
4). Also in the 1950*5, on the basis of an economic rating of a sinter machine
using the D & L process, the concept of "System Resistance" was introduced
[75]:
SR = (h)°-6/P x V a (power costs per ton)0-6
This new criterion similarly recommended that the bed depth should be as
shallow as possible for minimum system resistance, varying from 178mm to
330mm according to the type of materials to be sintered.
108
5). In 1958, Bogan L.C. and Worner H.K. [19] pointed out that bed depths of
152mm to 229mm are associated with suctions of the order of 7.5 to 8.5 kPa.
They also reported that the values for "n" were 0.458 to 0.606 for a bed depth
of 177mm and a suction of 8.5 kPa, which incidentally was common on the
Port Kembla sintering plant strand using the Australian rich fine hematite
ores.
6). However, it was overlooked that the formula on permeability was developed
using silica chip beds. In fact, the complex phenomena of the kinetic energy
and chemical reactions occurring in the combustion zone, where the bed depth
and the gas flow have the most critical effect during sintering, were omitted.
Hence, the effective contribution of this work on bed depth technology is
original.
These fundamental reasons governed sintering technology in a world wide sense during
the 1950’s and 1960’s and the development of bed depth technology was restricted.
Nevertheless, Japanese sintering plants operated at bed depths between 275 and 410mm
by the beginning of the 1970‘s.
However, the experimental work of this thesis proved the feasibility of operating the
sintering process on an economical basis at a bed depth between 450 and 550mm.
Hence, this experimental work supported by the Port Kembla plant trials can be
considered a world first. Evidently, the success of this project was critically dependent
on the improvement of the raw material characteristics and the experimental equip
ment; and the introduction of Sintering Process Models II and III.
Details of the equipment and processes have already been presented in chapter 4,
sections 4.1, 4.2 and 4.3.
109
7.1.1 Presentation of Data and Results
Experimental data and results are shown in Tables 7.1.1-1 to 7.1.1-6 and Figures 7.1.1-1
to 7.1.1-6, and speak for themselves. Symbols and contractions are coded in a formal
nomenclature as near as possible to the tables, figures, or equations in which they are
used.
The traditional Sintering Process Model I, results of which are summarised in Table
7.1.1-1, was conducted using only two values for the bed depth: 254mm and 508mm,
at basicities of 1.2, 1.5 2.0 and 2.4. Both the productivity and sinter quality were
markedly improved, particularly the FeO content, which was decreased from 9.4% to
8.4% at a basicity of 1.2,8.4% to 6.3% at a basicity of 1.5; 5.7% to 3.2% at a basicity
of 2.0 and 4.5% to 2.6% FeO at a basicity of 2.4. Similarly, there is a gradual increase
in the degree of oxidation (DO) as the bed depth increases.
The results from Model II are illustrated in Figs 7.1.1-1 and 7.1.1-4 to 7.1.1-6. These
test series were conducted on the basis of four values for each main variable: 250,
350, 450 and 550mm bed depths at basicities of 1.2, 1.6, 1.8 and 2.2.
In this model, the microstructural components of the sinter follow the same gradient,
with one exception, the hematite at a basicity of 1.6. Obviously, the level of those
components varies in proportion to the basicity level.
The Sinter Process Model III is illustrated in Figs 7.1.1-2 and 7.1.1-3. These test series
were also conducted on the basis of the four values for the bed depth as above, but
only at a basicity of 1.2, using the feed mixes "M2" and "M3".
In both Sintering Process Models II and III, a linear regression analysis of each
microstructural sinter component and the respective productivity and sinter quality,
expressed as the cold sinter strength (SI, TI), FeO content and the reduction degradation
index (RDI) was performed as a function of the bed depth.
Table 7.1.2-1, sets out the comparative results of the Sintering Process Models II and
III, where the magnetite content in the sinter mineralogy follows a positive slope in
the Model II and a negative gradient in the Model III. Similar favourable gradient
reversals were achieved for the other components.
110
Furthermore, results contained in the preliminary report of this thesis [69] and
presented during the Sinter Technology Seminar on Highly Fluxed Sinter at Port
Kembla Steelworks, in July 1972, were strongly supported [70]. The sinter development
committee reported as follows:
1). "The experimental reactor tests indicated that the maximum productivity in
terms of iron units is achieved at basicities in the range 1.5 to 1.8, 450 to
508mm bed depth and 10 kPa suction. This had been confirmed during short
periods of operation at Port Kembla sintering plant".
2). "The quality of the sinter on the tumbler index improved markedly from 66
to 76%".
3). "The Port Kembla plant sinter had had encouraging results on No 3 blast
furnace, resulting in 2% reduction in coke rate and 4% increase in ironmaking
productivity using 40-50% sinter burden" [71].
4). "In this sense, the Port Kembla experimental work had produced a superior
quality sinter having a good balance between cold sinter strength and
reduction-degradation properties" [70, 71].
Ill
Table 7.1.1-1 Sintering Process Model I at basicities of 1.2 and 1.5

Feed mix "M3" RF level in mix (%)
Flux 100% ML Coke rate ini mix (%)
RF size (mm) -6.35 for all tests RFB, (ratio)
Suction (kPa) 10 Bed depth, 254, 508
Basicity 1.2, 1.5, 2.0, 2.4
Bed depth, mm 254 508 254 508

Basicity, C aO /Si0 2 1.2 1.2 1.5 1.5
RF in mix, % 40 37 39 37
Coke in mix, % 5.5 3.5 5.3 3.5
PR, t/m 2/d 31.4 34.6 30.9 33.9
PR, tFe/m 2/d 18.7 20.5 17.4 19.6
RFB, ratio 1.00 1.03 1.05 1.00
SI, +6 .4mm % 60 62 59 63
TI*, +6.4mm % 88 90 89 92
DO, % 93 94 95 96
Sinter Analysis
FeO % 9.3 8.4 8.4 6.3
Total Fe % 59.3 59.2 56.4 57.7
CaO % 7.0 7.3 8.7 9.0
S i0 2 % 5.7 6.1 5.7 5.9
a i 2o 3 % 2.9 2.8 2.8 2.8
CaO /Si0 2 - 1.23 1.20 1.53 1.53
Slag % 15.6 16.2 17.2 17.7
112
Table 7.1.1-1 (cont’d) Sintering Process Model I at basicities
of 2.0 and 2.4
Bed depth, mm 254 508 254 508

Basicity, C a 0 /S i0 2 2.0 2.0 2.4 2.4
RF in mix % 39 37 40 39
Coke in mix % 5.3 3.6 4.8 3.6
PR, t/m 2/d 30.2 34.7 29.5 31.6
PR, tFe/m 2/d 16.5 19.6 15.7 17.5
RFB, ratio 1.03 1.05 1.03 1.01
SI, +6 .4mm % 60 62 59 60
TI*, +6.4mm % 88 90 90 92
DO, % 96 98 97 98
Sinter Analysis
FeO % 5.7 3.2 4.5 2.6
Total Fe % 54.8 56.4 53.3 55.4
CaO % 10.3 10.6 11.4 12.5
Si0 2 % 4.9 5.6 4.7 5.3
a i 2o 3 % 2.7 2.7 2.6 2.7
CaO /Si0 2 - 2.10 1.89 2.43 2.36
Slag % 17.9 18.9 18.7 20.5
Code:
PR = Production rate /productivity = t/m 2/24hrs = t/m 2/d
RFB - Return fines balance = RF out/RF in
RF = Return fines level in the feed mix
SI= Size index, cumulative % retained on 6.35mm mesh
TI*= Tumbler index, cum % retained on 6.4mm, using 4.5 kg sample
TI - Tumbler index, cum % retained on 6.4mm, using 12 kg sample (std)
RDI = Nagoya reduction degradation index, % -2.8mm
DO = % degree of oxidation (Linder original state of oxidation)
Ca-ferrite = Calcium ferrite
Dica-silicate - Dicalcium silicate
Slag/gangue in sinter chemical composition = % (CaO + Si0 2 + A120 3)
Slag in sinter mineralogy = % (silicate glass + dicalcium silicate)
CB1 = Coffin Bay limesand flux
ML = Marulan limestone flux
113
Table 7.1.1-2 Sintering Process Model II at basicity of 1.2
Constant parameters: Variable parameters:

Feed mix "M2" RF level in mix (%)
Flux (%) 75 ML/25 CB1 RFB (ratio)
Suction (kPa) 10 Bed depth: 250, 350, 450, 550
Basicity 1.2, 1.6, 1.8, 2.2
Coke in mix (%) 5.0
Bed depth, mm 250 350 450 550

Basicity, C aO /Si0 2 1.2 1.2 1.2 1.2
RF in mix % 45 37 32 30
PR, t/m 2/d 29.1 30.5 32.0 31.7
RFB, ratio 0.99 1.03 1.03 1.04
SI, +6.4mm % 55.5 62.0 67.0 68.7
TI, +6.4mm % 60 64 68 69
RDI, -2.8 mm % 33 31 27 24
Sinter Mineralogy (vol %)
Ca- ferrite 15 15 15 14
Total hematite 29 26 22 18
Glass 16 14 13 15
Dica -silicate 0 0.1 0.5 0.9
114
Table 7.1.1-3 Sintering Process Model III at basicity of 1.2
Constant parameters: Variable parameters:

Feed mix "M2" Coke in mix (%)
Basicity 1.2 RFB: 1.09, 1.02, 0.97, 0.94
Suction, kPa 10 Bed depth: 250, 350, 450, 550
RF in mix, % 40
Bed depth, mm 250 350 450 550

Basicity, C aO /Si0 2 1.2 1.2 1.2 1.2
Coke in mix, % 5.5 5.0 4.0 3.5
PR, t/m 2/d 29.3 30.2 31.2 31.6
RFB, ratio 1.09 1.02 0.97 0.94
SI, +6.4mm % 56.6 59.1 61.1 62.3
TI, +6.4mm % 59 61 62 63
RDI, -2.8mm % 30 32 37 36
Ca -ferrite 5 8 10 13
Glass 19 17 16 14
Dica -silicate 0.4 0.3 0.3 0.2
Table 7.1.1-4 Sinter Process Model III at basicity of 1.2

Sintering Parameters: Feed mix "M3"
Others "as above"
Bed depth, mm 250 350 450 550

Basicity, C aO /Si0 2 1.2 1.2 1.2 1.2
Coke in mix, % 6.0 5.0 4.0 3.0
PR, t/m 2/d 29.6 30.1 31.8 34.3
RFB, ratio 1.02 0.97 0.90 0.94
SI, +6.4mm % 59.3 61.2 64.0 62.4
TI, +6.4mm % 58 60 63 62
RDI, -2.8mm % 31 34 38 36
Ca -ferrite 8 10 12 13
Glass 17 15 14 13
Dica -silicate 1.3 1.2 0.2 0
115
Table 7.1.1-5 Sintering Process Model II at basicity of 1.6

Feed mix "M2” RF level in mix, %
Suction, kPa 10 RFB, ratio
Basicity, 1.2, 1.6, 1.8, 2.2 Bed depth: 250, 350, 450, 550
Coke in mix, % 5.0
Bed depth, mm 250 350 450 550

Basicity, C aO /Si0 2 1.6 1.6 1.6 1.6
RF in mix, % 44 36 31 29
PR, t/m 2/d 31.3 32.6 32.9 32.5
RFB, ratio 0.94 0.99 1.02 1.02
SI, +6.4mm % 58.6 64.5 68.3 70.5
TI, +6 .4mm % 56 60 63 64
RDI, -2.8mm % 30 31 30 27
Sinter Mineralogy (vol °/o)
Ca -ferrite % 37 34 32 32
Total hematite % 32 33 34 34
Magnetite % 22 24 25 26
Glass % 8 8 7 5
Dica -silicate % 1.2 1.4 2.4 3.1
116
Table 7.1.1-6 (cont’d) Sintering Process Model II at basicities
of 1.8 and 2.2.
Bed depth, mm 250 350 450 550
Basicity 1.8 1.8 1.8 1.8
RF in mix, % 40 32 29 27
PR, t/m 2/d 30.6 32.0 32.9 31.9
RFB, ratio 0.95 1.03 1.04 1.03
SI, +6.4mm % 62.0 67.0 69.9 72.3
TI, +6.4mm % 58 61 65 66
RDI, -2.8mm % 31 28 24 22
Ca -ferrite 38 39 38 37
Glass 10 8 8 9
Dica -silicate 1.4 1.9 30 4.4
Bed depth, mm 250 350 450 550
Basicity 2.2 2.2 2.2 2.2
RF in mix, % 36 30 25 24
PR, t/m 2/d 30.1 31.5 31.8 31.7
RFB, ratio 1.01 1.04 1.04 1.11
SI, +6.4mm % 63.5 68.7 74.0 76.0
TI, +6.4mm % 55 59 62 63
RDI, -2.8mm % 28 25 21 18
Ca -ferrite 52 53 53 52
Glass 7 4 4 5
7.1.2 Interpretation of Data and Results
In general terms, there was a significant increase in productivity and sinter quality.
In the context of the sinter mineralogy, each regression equation and its correlation
coefficient shows a clear difference related to the sintering model used, the feed mix
and operating parameters.
The classification of the sintering process into three models is another original con
tribution to the elucidation of the sintering technology. Models II and III show a
clear difference in their effects on the sintering process, particularly on the
microstructural composition of the sinter. Hence, these models can be applied to the
sintering plant operating conditions and to the meeting of blast furnace requirements.
The Sintering Process Model III was reproduced using a different feed mix but the
same sintering parameters. The comparative mathematical relationships of the
microstructural components, and the other sinter properties, and productivity values
are summarized in Table 7.1.2-1.
117
Table 7.1.2-1 Linear correlation equations and squared correlation
coefficients for Sintering Process Models II and III
at basicity of 1.2 (CaO/Si02), using sintering feed
mix "M2" and "M3".
MODEL III MODEL II MODEL III

Feed mix "M2" "M2" "M3"
M Y= -.06X + 71.2 Y= .04X + 30.5 Y= -.04X + 60.75
R2= 0.99 R2= 0.99 R2= 0.98
TH Y= .045X + 8.25 Y=-.037X + 38.55 Y= .04X + 10.35
R2= 0.99 R2= 1.0 R2= 0.97
CF Y - .026X - 1.4 Y=-.003X + 15.95 Y= .02X + 3.95
R2= 0.99 R2= 0.60 R2= 0.98
G Y= -.016X + 22.9 Y= -.004X + 16.1 Y=-.01X + 19.95
R 2= 0.98 R2= 0.16 R2= 0.97
DS Y=-.0006X + 0.54 Y= .003IX - 0.83 Y=-.005X + 2.64
R2= 0.90 R2= 0.95 R2= 0.89
PR Y= .0079X + 27.41 Y= .0093X + 27.1 Y=-.02X + 25.13
R2= 0.97 R2= 0.83 R2= 0.93
RDI Y= .027X + 24.55 Y=-.0295X + 40.43 Y= .02X + 27.15
R2= 0.81 R2= 0,97 R2= 0.87
FeO N/A N/A Y=-.012X + 14.87
R2= 0.89
SI Y= .019X + 51.99 Y= .0446X + 45.46 Y= .01X + 56.89
R2= 0.97 R2= 0.94 R2= 0.82
TI Y= .013X + 56.05 Y= .031X + 52.85 Y= .02X + 54.75
R2= 0.97 R2= 0.95 R2= 0.76
RF Constant Y= -.05X + 56.0 Constant
R2= 0.93
Coke Y= -.01X + 7.3 Y= -.01X + 8.5
R2= 0.98 R2= 1.00
Code:
M - Magnetite DS = Decalcium silicate
TH = Total hematite X = Variable bed depth
CF - Calcium ferrite Y - Vector of obeservations
G = Silicate glass
118
S IN TE R IN G P R O C E S S M O D E L II
Feed Mix "M2" RF in mix (% )
Basicity 1.2 Bed Depth: 250, 350, 450, 550
Suction (kPa) 10 RFB: 0.99, 1.03, 1.03, 1.04
Coke in mix (% ) 5
B ed Depth (m m )
Fig 7.1.1 -1 Sinter mineralogy, productivity and strength as a function of

bed depth.
119
S IN TE R IN G P R O C E S S M O D E L III
Feed Mix "M2" Coke (% )
Suction (kPa) 10 RFB: 1.09, 1.02, 0.97, 0.94
RF in mix (% ) 40
20 63
19 62 - SI
V
18 - N . ^ 61 . y /
H 17 ■
N. 2 60 - /
^ 16 3
8 ~ 59 -
§ 15 ■ y - - 0 .0 1 6 X + 22.9 \ 8 /
\ "» 58 - y - 0 .019 x + 51.99
14 • r* - 0.98 ■
G
13 57 . r* - 0.97
56 ____ I____,____ .— .— .— i— .— i—
12
Fig 7 .1 .1 -2 Sinter mineralogy, productivity and strength as a function of

bed depth.
120
Feed Mix "M3" Coke (% )
Suction (kPa) 10 RFB: 1.02, 0.97, 0.90,0.94
RF in mix (% ) 40

bed depth.
121
Feed Mix "M 2ff RF in mix (% )
Suction (kPa) 10 RFB: 0.94, 0.99, 1.02, 1.02
Coke in mix (% ) 5
Hsmatits (Vol X)
Colclum Ferrits (Vol X)
Mognstits (Vol X)
Gloss (Vol X)
DS (Vol X)

bed depth.
122
Suction (kPa) 10 RFB: 0.95, 1.03, 1.04, 1.03
Coke in mix (% ) 5

bed depth.
123
Suction (kPa) 10 RFB: 1.01,1.04, 1.04, 1.11
Coke in mix (% ) 5

bed depth.
124
The conclusions on the effects of bed depth related to a constant basicity level and
feed mix for each test series are:
1. In the context of the sinter microstructural components, change of the positive

gradient of the magnetite content in the Sintering Process Model II to a negative
gradient in the Model III, and similar favourable gradient reversals were
achieved for the other components.
2. A lower content of high melting point components is more economical for blast
furnace operation and the desired quantitative distribution of the sinter
microstructural components may be achieved by using the above models.
3. Experiments at 250, 350, 450, 508 and 550mm bed depths, and 1.2 basicity
required high fuel consumption to give low productivities, and product sinters
containing high FeO, with markedly high magnetite content in the sinter
mineralogy. Hence, the sinter strength was low. Therefore, these parameters
resulted neither in suitable sintering plant operation nor acceptable blast
furnace burden material.
4 Experiments at 250 and 350mm bed depths, and 1.5 to 2.4 basicities showed
these parameters to be unsuitable for sintering plant operation because of high
fuel rate and hence high cost.
5. Experiments at between 450 and 550mm bed depth, and between 1.6 and 1.8
basicity resulted in high productivity with marginal loss in sinter quality (TI),
but there was a significant increase in the calcium ferrite and total hematite
content and a marked decrease in the magnetite content. Therefore, these are
suitable parameters for sintering plant operation on a productivity basis, and
this product accomplishes the required blast furnace burden material char
acteristics.
125
6. Experiments at between 450 and 550mm bed depth, and 1.8 and 2.2 basicity
resulted in high sinter quality with marginal loss in productivity. Under these
experimental conditions, the calcium ferrite content significantly increased
and the total hematite content gradually decreased, while the magnetite
markedly decreased. Therefore, these are another set of suitable parameters
for sintering plant operation on the basis of sinter quality as well as accom
plishing the required blast furnace burden material
7. Experiments also at 450, 508 and 550mm bed depths, but between 2.2 and 2.4
basicities, produced high sinter size distribution but slightly lower tumbler
index, and high gangue content and low iron content with low productivity.
Therefore, these parameters are not significantly beneficial either for sintering
plant operation or blast furnaces.
8. There is a significant decrease in coke rate related to the increase in bed depth,
As a result, the sintering process temperature decreases.
9. The linear regression analysis results illustrate that the bed depth influence
on sintering can be applied to increase or decrease the values of the RDI and
FeO. Hence, this is another alternative to control the thermal conditions, and
the sinter properties during the sintering process.
10. There is a clear relationship between the microstructural components, pro

ductivity and the factors determining sinter quality as a function of the bed
depth. However, there is an optimum bed depth (X) for each vector of
observations (Y).
126
8.0 E F F E C T S OF B A SIC ITY ON SINTERIN G
8.1 Overview
In a similar way to the bed depth, world wide searches were conducted in 1972 for
references to the effects of basicity on sintering.
Table 8.1-1 shows that acid, self-fluxing and super-fluxing sinters had been used as
14%-55% of blast furnace burden material, in the American continent. The basicity
was defined as the ratio (Ca0+M g0)/(Si0 2+Al20 3).
Table 8.1-2 and 8.1-3 illustrate that self-fluxing to super-fluxing sinters had been used
as 48%-85% of blast furnace burden material in Europe, while in Japan, super-fluxing
sinters had been used as 65%-89% of blast furnace burden material. The basicity was
defined as the C a0 /S i0 2 ratio in both Europe and Japan.
Furthermore, a literature survey showed that the sinter basicity had gradually
increased, so that most blast furnaces had operated using acid to self-fluxing (0.2- 1.0
CaO/Si02) agglomerated material (pellets and sinter). Sinters of this basicity continued
in use until the 1960’s [19,78 to 83].
However, in contrast, the UK and Sweden used self-fluxing and super-fluxing sinters
(0.9-1.5 CaO/Si02), as early as the 1930’s [11]. The use of this type of sinter gradually
extended over almost all the world by the end of the 1960’s [84 to 87]
In Australia, the Kwinana plant started producing super-fluxing (1.3-1.8 basicity)

sinter in 1970. Subsequently the Port Kembla plant commenced producing sinter with
a basicity of 1.6-1.8 CaO/Si0 2 in 1972. Currently, sinter with a basicity of 1.9-2.0 is
being produced.
127
The advantages of flux additions [19] to increase sinter basicity are summarised as
follows:
1). Low blast furnace fuel cost, due to the endothermic decomposition of calcium
carbonate which takes place externally to the blast furnaces,
2). Improved stabilization of the reducing potentials (C 0 /C 0 2 ratios) at different

temperature levels in the blast furnace due to the elimination of the endothermic
calcination reaction in the blast furnace stack,
3). Uniform-slag making conditions within the blast furnace, as the gangue
minerals in the sinter are slagged with lime externally,
4). Super-fluxing sinters have a greater resistance to disintegration during

reduction (RDI), and a higher reducibility than the low-lime-bearing sinter
type, and
5). Finally, there is an overall reduction of the combined coke rate of both the
sintering plant and blast furnaces.
128
Table 8.1-1 North American Blast Furnace Burdens 1970-1971.
Data from A.I.S.I and I.I.S.I. sources.
% Average Burden
No of SB
Company B Fees P S 0 Mise Ave
Algoma 1 48 50 2 - 1.17
Armco 4 72 16 3 9 2.08
Bethlehem 5 54 27 8 11 1.84
Colorado F.& I* 1 - 55 42 3 0.82
Monterry 1 - - 99 1 . -
Pacific Steel 1 - - 100 - -
Crucible Steel 1 65 - - 35 -
Detroit Steel 1 74 - - 26 -
Dofasco 1 87 - 2 11 -
Granite City S. 1 50 41 5 4 1.23

Inland Steel 2 76 10 10 4 4.07
Interlake 2 55 17 20 8 2.21
Kaiser Steel 1 - 38 56 6 1.80
Lone Star 1 16 21 39 24 0.53
National Steel 4 38 26 28 8 1.71
Republic 5 59 23 7 11 0.56
Sharon 1 97 - - 3 -
Stelco 1 65 22 3 10 3.40
Sydney Steel 1 77 15 8 - 2.40
U.S.S. Corp. 17 20 39 16 25 0.84
W. Pittsburgh 2 72 20 - 8 3.25
Youngstown S & T 2 70 14 9 7 0.80
Jones & Laughlin 3 31 43 20 6 0.98
Allan Wood 1 38 48 6 8 0.80
Code:
P - Acid pellets, 0.2-0.6 basicity S » Sinter
Flux - Limestone and dolomite O - Screened ore
Misc - Miscellaneous (slag, flue dust, mill scale, unscreened ore,etc.
SB - Sinter basicity: (CaO+MgO) / (Al20 3+Si02).
129
Table 8.1-2 1971 European Blast furnace Burdens 1970-1971
Data from A.I.S.I & I.I.S.I. sources.
% Ave Burden
No of Basicity Size
Country B Fees P S O Mise Ave mm
Austria 1 - 85 9 6 1.07 6x40
Belgium 12 5 54 38 3 1.39 6x52
France 5 3 67 28 2 1.40 15x88
Germany 9 7 57 30 6 1.23 5x41
Italy 4 32 48 20 - 1.54 7x43
Holland 2 37 53 7 3 2.04 6x40
Spain 3 5 57 36 2 0.98 8x60
Sweden 2 4 83 11 2 1.45 5x100
U.K 7 19 69 8 4 1.14 3x40
Table 8.1-3 1971 Japanese Blast Furnace Burdens
% Ave Burden
Size
Company P S O Mise Basicity mm
Kawasaki
Chiba 5 66 29 -
1.81 5x50
Kobe
Kakogowa 76 - 24 - -
N.S.C.
Muroran - 89 9 2 1.30 5x50
Kimitsu 12 65 22 1 1.59 6x50
Nagoya 5 86 8 1 1.65 6x50
Tobata 15 72 10 3 1.66 6x50
Sumitomo
Wakayama 12 66 22 - 1.69 8x50
Nisshin
Kure 19 70 11 - 1.30 6x75
N.K.K.
Fukuyama - 74 24 2 1.46 5x50
Code:
P - Acid pellets Misc = Miscellaneous
S = Sinter Flux = Limestone
O - Screened ore Sinter basicity = CaO/Si0 2 ratio
Size mm - Sinter size
130
Experimental data and results are illustrated in Table 8.1.1-1 to 8.1.1-3 and Figs 8.1.1-1
to 8.1.1-5. Symbols and contractions are coded in a formal nomenclature as near as
possible to the tables, figures, or equations in which they are used.
The results obtained applying Model I are shown in Fig 8.1.1-1. This test series was
performed at basicities of 1.2, 1.5, 2.0 and 2.4, using the feed mix "M3'\ No microscopic
observations were conducted, hence comparisons with the subsequent test series which
were done following Model II, were limited to sinter strength (SI and TI) and pro
ductivity.
Results for Model II are set out in Figs 8.1.1-2 to 8.1.1-5. These test series were also
conducted on the basis of four values: 1.2, 1.6, 1.8 and 2.2 basicities, using the feed
mix "M2".
When using Model I, among the usual five sintering tests to determine each observation
point, either the coke or the return fines level was adjusted to maintain the return
fines balance close to unity. As a result, some of the observation points are scattered
(Fig 8.1.1-1). However, the results of Model II illustrate a clear difference in the
effects of each basicity variable (X) on sintering, and on most of the vectors of
observation (Y).
In the context of the sinter mineralogical composition, the contents of calcium ferrite
and dicalcium silicate increased, while the magnetite, hematite and glass decreased.
These gradients followed a favourable trend, except that for the dicalcium silicate.
On the other hand, when the basicity levels were increased, the productivity gradually
decreased as the sinter quality increased, particularly the sinter size distribution.
131
Table 8.1.1-1 Sintering Process Model I at bed depth of 254
and 508mm.

Feed mix "M3" RF in mix, %
Flux 100% ML Coke in mix, %
Suction, kPa 10 RFB, ratio
Bed depth, mm 254 and 508 Basicity: 1.2, 1.5, 2.0 and 2.4
Basicity, C aO /Si0 2 1.2 1.5 2.0 2.4

Bed depth, mm 254 254 254 254
RF in mix, % 40 39 39 40
Coke in mix, % 5.5 5.3 5.3 4.8
PR, t/m 2/d 31.4 30.9 30.2 29.5
PR, tFe/m 2/d 18.6 17.4 16.5 15.3
RFB, ratio 1.00 1.05 1.03 1.03
SI, +6.4mm % 60 59 60 59
TI*, +6.4mm % 88 89 88 89
DO, % 93 95 96 87
Sinter Analysis
FeO % 9.3 8.4 5.7 4.5
Total Fe % 59.1 56.4 54.8 53.3
CaO % 7.0 8.7 10.3 11.4
Si0 2 % 5.7 5.7 4.9 4.7
AI0O3 % 2.9 2.8 2.7 2.6
C aO/Si0 2 ~ 1.23 1.53 2.10 2.43
Slag % 15.6 17.2 17.9 18.7
Basicity, C aO /Si0 2 1.2 1.5 2.0 2.4
Bed depth, mm 508 508 508 508
RF in mix, % 37 37 37 39
Coke in mix, % 3.5 3.5 3.6 3.6
PR, t/m 2/d 34.6 33.9 34.7 31.6
PR, tFe/m 2/d 20.5 19.6 19.6 17.5
RFB, ratio 1.03 1.00 1.05 1.01
SI, +6.4mm % 61.9 63.0 61.2 60.4
TI*, +6.4mm % 90 92 90 91
DO, % 94 96 98 98
Sinter Analysis
FeO % 8.4 6.3 3.2 2.6
Total Fe % 59.2 57.7 56.4 55.4
CaO % 7.3 9.0 10.6 12.5
Si0 2 % 6.1 5.9 5.6 6.3
AI0Ô3 % 2.8 2.8 2.7 2.7
CaO /Si0 2 - 1.20 1.53 1.89 2.36
Slag % 16.2 17.7 18.9 20.5
132
Table 8.1.1-2 Sintering Process Model II at bed depth of 250
and 350 mm

Feed mix "M2” RF in mix, %
Coke in mix, % 5.0 RFB, ratio
Suction, kPa 10 Basicity 1.2, 1.6, 1.8 and 2.
Bed depth, mm 250 and 350
Basicity, C aO /Si0 2 1.2 1.6 1.8 2.2

Bed depth, mm 250 250 250 250
RF in mix, % 45 44 40 36
PR, t/m 2/d 29.1 31.3 30.6 30.1
RFB, ratio 0.99 0.94 0.95 1.01
SI, +6.4mm % 55.5 58.6 62.0 63.5
TI, +6.4mm % 60 56 58 55
RDI, -2.8mm % 33 30 31 28
Ca -ferrite 15 37 38 52
Glass 16 8 10 7
Dica -silicate 0 1.2 1.4 2.4
Basicity, C aO /Si0 2 1.2 1.6 1.8 2.2
Bed depth, mm 350 350 350 350
RF in mix, % 37 36 32 30
PR, t/m 2/d 30.5 32.6 32.0 31.5
RFB, ratio 1.03 0.99 1.03 1.04
SI, +6 .4mm % 62.0 64.5 67.0 68.7
TI, +6.4mm % 64 60 61 59
RDI, -2.8mm % 31 31 28 25
Ca -ferrite 15 34 39 53
Glass 14 8 8 4
133
Table 8.1.1-3 Sintering Process Model II at bed depths
of 450 and 550 mm
Sintering Parameters: "as in Table 8.1.1-2"
Basicity, C aO /Si0 2 1.2 1.6 1.8 2.2
Bed depth, mm 450 450 450 450
RF in mix, % 32 31 29 25
PR, t/m 2/d 32.0 32.9 32.9 31.8
RFB, ratio 1.03 1.02 1.04 1.04
SI, +6.4mm % 67.0 68.3 69.9 74.0
TI, +6.4mm % 68 63 65 62
RDI, -2.8mm % 27 30 24 21
Ca -ferrite 15 32 38 53
Glass 13 7 8 4
Basicity, C aO /Si0 2 1.2 1.6 1.8 2.2
Bed depth, mm 550 550 550 550
RF in mix, % 30 29 27 24
PR, t/m 2/d 31.7 32.5 31.9 31.7
RFB, ratio 1.04 1.02 1.03 1.11
SI, +6.4mm % 68.7 70.5 72.3 76.0
TI, +6.4mm % 69 64 66 63
RDI, -2.8mm % 24 27 22 18
Ca -ferrite 14 32 37 52
Glass 15 5 9 5
134
S IN TE R IN G P R O C E S S M O D E L I
Bed Depth (mm) 254, 508 Coke in mix (% )
Suction (kPa) 10 Basicity: 1.2, 1.5, 2.0, 2.4
RFB-254: 1.00, 1.05, 1.03, 1.03
RFB-508: 1.03, 1.00, 1.05, 1.01
Productivity (t/m2/d)
Productivity (tF«/m 2/d)
FeO (X)
DO (X)
Slog X (CaO + Si02 + AJ203)
B a s ic ity ( C a O / S i0 2 )
Fig 8.1.1-1 Sinter productivity, FeO and slag content, degree of oxidation(DO)
and strength as a function of basicity.
135
Coke in mix (% ) 5 Basicity: 1.2, 1.6, 1.8, 2.2
Bed Depth (mm) 250 RFB: 0.99, 0.94, 0.95, 1.01
Suction (kPa) 10
Fig 8.1.1 2 Sinter mineralogy, productivity and strength as a function of

basicity.
136
Bed Depth (mm) 350 RFB: 1.03, 0.99, 1.03, 1.04
Suction (kPa) 10
33
y - - 2 7 . 6 9 x + 7 5 .5 8 32
. X . r* - 0 .8 4 31 - > N . .
30 -
* N. O
n 29
1 28
O N.
27 -
26 . y - - 6 . 3 5 x + 3 9 .5 4 Nv
\ M r* - 0 .8 5 N
25 • a
----*------ * 1------ 1------ «-------1—---1------ 1------ 1------ 1____1__ RDI

24
70
- a 69 - x
68 - y - 6 .9 2 x + 5 3 .7 8 sl
- y — -3 .2 7 x + 3 1 .8 1 r — 0 .9 7
r* - 0 .0 7 67
_ S? 66 -
------------------------ f
IS
in
- y '
«
---------- ■— ._______ h
' ----------? 64 *
- 63
■
62
Fig 8. 1. 1- 3 Sinter mineralogy, productivity and strength as a function of

basicity.
137
Bed Depth (mm) 450 RFB: 1.03, 1.02, 1.04, 1.04
Suction (kPa)
Cakium Ferrite (Vol X)
Magnatita (Vol X)
Hematite (Vol X)
Glaaa (Vol X)
DS O'01 *)
10

basicity.
138
SINTERING PROCESS MODEL III
Feed Mix "M2" R F in m ix (% )
Bed Depth (mm) 550 RFB: 1.04, 1.02, 1.03, 1.11
Suction (kPa)
Calcium Ferrite (Vol X)
10
1 .4 1.6 1.8 2 .0 1 .4 1.6 1.8

B a s ic ity ( C a O / S i0 2 ) B a s ic ity (C a O / S i0 2 )
Magnetite (Vol X)
78
- a 77 SI
76 - O
75 - y - 7 .3 7 X + 5 9 .3 5
Hematite (Vol X)
y - - 4 . 6 2 x + 2 8 .8 5
r* - 0 .0 5 S 74 - r1 - 0 .9 7
« 73 -
. ------- ---------_ _ _ _ _ f 72 ■
.
- O
" --------
1
a 771
0 -
69 ’ a
o H
68
' 67
66
Close (Vol X)
os CM x)
B a s ic ity (C a O / S I0 2 )

basicity.
139
1. In the context of the sinter microstructural components, the negative gradient

of the magnetite content, and the positive gradient of the calcium ferrite
particularly in the Model II was a significant achievement. However, further
experimental work is recommended using Model III with basicities between 1.8
and 2.2 and at bed depths between 500 to 550mm.
2. Experiments with all basicities ( 1.2 to 2.4 CaO/Si0 2 ratios) and at 250 and
350mm bed depths required high fuel consumption, to give product sinters
containing high FeO content and showing high magnetite content in the sinter
mineralogy. Therefore, these parameters resulted neither in suitable sintering
plant operation nor in acceptable blast furnace burden material.
3. Experiments with a basicity of 1.2 and at bed depths of 450 and 550mm required
high fuel consumption or high return fines level, producing high FeO content
and with markedly high magnetite content in the sinter mineralogy. Hence,
they were outside the sinter specifications. Therefore, these parameters resulted
neither in suitable sintering plant operation nor in acceptable blast furnace
burden material.
4. Experiments with 1.6 basicity, and at between 450 and 550mm bed depths resulted
in high productivity with marginal loss in sinter quality. Therefore, these
parameters are suitable for sintering plant operation on a productivity basis,
and this product meets the requirements for blast furnace burden material.
5. Experiments with 1.8 basicity, and at between 450 and 550mm bed depths resulted
in a balance of productivity and sinter quality. Therefore, these parameters
are suitable for sintering plant operation, and this product also accomplishes
the requirements of blast furnace burden material.
6. Experiments at between 1.8 and 2.2 basicities, and at between 450 and 550mm
bed depths produced a marginal loss in productivity, with the highest level of
calcium ferrite and the lowest magnetite content in the sinter mineralogy, using
the lowest fuel rate or the lowest return fines in the feed mix. Therefore, these
140
parameters are the most suitable for sintering plant operation in terms of sinter
quality, and this product is the most suited of those produced to date to meet
the requirements of blast furnace burden material.
7. All experiments have shown that the effects of basicity in the sinter mineralogy
is highly sensitive, changes of the order of 0.1% being significant. A minor
variation in the basicity changes the microstructural composition of the sinter
markedly.
141
9.0 E F F E C T S OF SU CTIO N ON SIN TERIN G
9.1 Overview
A literature survey on the effects of suction on sintering shows that the suction level
was gradually increased from 6.0 kPa (windbox) at the beginning of the 1950’s [72],
to 6.7 kPa by the end of the same decade [13, 19].
This increase in suction continued, so that most of the sintering plants in Europe,
Japan, Australia and in North America were operating at suctions between 7.5-8.5 kPa
by the middle of the 1960’s and at between 8-10 kPa by the end of that decade [54,
69, 88], so that the work reported in this thesis examined the effects of suction of up
to 15 kPa.
The never-ending technological advance continued, improving the sintering process,

particularly in Japan, Germany, Luxembourg and in Australia by the beginning of
the 1970’s [51, 84].
Experimental data and results are set out in Tables 9.1.1-1 to 9.1.1-4 and Fgs 9.1.1-1
to 9.1.1-3. Symbols and contractions are coded in a formal nomenclature as near as
possible to the tables, figures, or equations in which they are used.
Sintering Process Model III was applied for all test series in the study of the effects
of suction on sintering (except the first test series illustrated in Table 9.1.-1). Results
are illustrated in Figs 9.1.1-2 to 9.1.1-4. These test series were conducted on the basis
of three suction levels: 10, 12.5 and 15 kPa, using the sintering feed mix "M4".
Results show that an orderly position of each combination of parameters (X, Y) was
achieved by Model III, compared to Models I and II, in chapters 7 and 8.
142
In the context of the components of the sinter mineralogy, there was a clear difference
of the effects of the suction on the calcium ferrite and total hematite content, compared
to the magnetite and slag components at each constant bed depth. The calcium ferrite
and hematite regression lines at 550mm bed depth are above those for the 450mm bed
depth , while the regression lines for the magnetite and slag components at the same
550mm bed depth are below those for the 450mm bed depth. Both cases are repeated
for all test series at 1.6, 1.8 and 2.0 basicities.
Hence, at a bed depth of 550mm and basicities of 1.6, 1.8 and 2.0, the overall effects
of the suction levels at 10 and 12.5 kPa were:
a) From the sinter microstructural composition point of view:

- higher contents of hematite and calcium ferrite,
- lower contents of magnetite and slag,
b) From the productivity, FeO content and the cold sinter

strength point of view:
- lower content of FeO in the sinter, and low coke rate in
the feed mix,
- higher cold sinter strength, and
- higher productivity were observed.
The same effects were observed at a 450mm bed depth, and in all test results at 15
kPa suction.
As may be seen in Fig 9.1.1-1, the productivity reached the peak of 38.5 t/m 2/d at a
basicity of 1.6, 550mm bed depth and a suction level of 12.5 kPa. However, from this
point, the productivity gradually decreased at 1.8 and 2.0 basicities, and a marked
fall occurred at 15 kPa suction.
As a result, the regression line corresponding to the 550mm bed depth intercepted the
regression line of the 450mm bed depth, producing a cut-off point (Fig 9.1.1-3). This
illustrates that the maximun productivity was obtained at the optimum combination
of operating parameters: 12.5 kPa suction, 550mm bed depth and 1.6 basicity.
143
In addition, the experimental results showed that the suction of 15 kPa at 1.6 and 1.8
basicities at between 450 and 550mm bed depth was detrimental to sinter quality,
mainly in terms of FeO content and was costly in energy consumption.
Table 9.1.1-1 Sintering Process Model I at basicities of

1.2, 1.5, 2.0 and 2.4.

Feed mix "M3" Coke in mix. %
Flux 100 % ML RF in mix. %
Bed depth, 508 Suction, kPa
Suction, kPa 10.0 12.5 10.0 12.5

Basicity, C a0 /S i0 2 1.2 1.2 1.5 1.5
RF in mix, % 37 37 37 37
Coke in mix, % 3.5 3.7 3.5 3.9
PR, t/m 2/d 34.6 35.2 33.9 37.0
PR, tFe/m 2/d 20.5 20.8 19.6 21.5
RFB, ratio 1.03 0.93 1.00 0.93
SI, +6.4mm % 61.9 65.8 63.0 65.8
TI*, +6.4mm % 90 88 92 84
DO, % 94 94 96 97
Sinter Analysis
FeO % 8.4 8.5 6.3 6.5
Total Fe % 59.2 59.0 57.7 58.0
CaO % 7.3 7.4 9.0 8.7
S i0 2 % 6.1 6.1 5.9 5.8
A120 3 % 2.8 2.9 2.8 2.7
C aO /Si0 2 - ' 1.20 1.21 1.53 1.50
Slag % 16.2 16.4 17.7 17.2
Suction, kPa 10.0 12.5 10.0 12.5
Basicity, CaO/Si02 2.0 2.0 2.4 2.4
RF in mix, % 37 40 39 40
Coke in mix, % 3.6 3.9 3.6 4.1
PR, t/m 2/d 34.7 35.1 31.6 31.4
PR, tFe/m 2/d 19.6 19.8 17.5 16.8
RFB, ratio 1.05 1.02 1.01 0.90
SI, +6.4mm % 61.2 59.4 60.4 64.0
TI*, +6.4mm % 90 91 91 94
DO, % 98 98 98 98
Sinter Analysis
FeO % 3.2 3.4 2.6 3.0
Total Fe % 56.4 56.4 55.4 53.5
CaO % 10.6 11.2 12.5 14.6
S i0 2 % 5.6 5.4 5.3 6.2
A120 3 % 2.7 2.8 2.7 3.1
C aO /Si0 2 - 1.89 2.08 2.36 2.35
Slag % 18.9 19.4 20.5 23.9
144
Table 9.1.1-2 Sintering Process Model III at basicity
(C a0/S i02) of 1.6

Feed mix "M4" Coke in mix, %
RF in mix, % 30 RFB, ratio
Bed depth, 450 and 550 Suction: 10, 12.5 and 15
Basicity 1.6
Suction, kPa 10 12.5 15 10 12.5 15

Bed depth, mm 450 450 450 550 550 550
Coke in mix, % 4.2 4.5 5.5 3.8 4.2 5.0
PR, t/m 2/d 35.7 36.5 37.1 38.1 38.5 38.3
PR, tFe/m 2/d 20.7 21.1 21.4 20.7 21.0 21.2
RFB, ratio 1.06 1.07 1.08 1.00 1.01 1.03
SI, +6.4mm % 68.4 68.0 67.5 70.0 69.6 69.1
TI, +6.4mm % 64 65 66 65 66 67
Sinter Analysis
FeO % 6.6 6.8 7.8 6.1 6.3 7.3
Total Fe % 57.9 57.7 57.6 57.4 57.6 57.5
CaO % 9.6 8.6 8.4 9.5 8.7 8.6
S i0 2 % 6.0 5.4 5.4 6.0 5.4 5.3
AI2O3 % 3.0 2.9 2.9 3.2 2.9 2.9
MgO % .31 .33 .34 .33 .35 .32
CaO/SiOo - 1.60 1.59 1.56 1.58 1.61 1.62
Slag %
Ca -ferrite 27 29 30 29 31 33
Total hematite 35 29 23 37 33 27
Magnetite 26 29 33 23 25 28
Glass 9 10 11 8 8 9
Dica -silicate 3 3 3 3 3 3
Slag (G + DS) 12 13 14 11 11 12
145
Table 9.1.1-3 Sintering Process Model III at basicity of 1.8.

Feed mix "M4" Coke in mix, %
RF in mix, % 30 and 28 RFB, ratio
Basicity 1.8
Suction, kPa 10 12.5 15 10 12.5 15

Bed depth, mm 450 450 450 550 550 550
RF in mix, % 30 30 30 28 28 28
Coke in mix, % 3.8 4.5 4.8 3.5 4.2 4.5
PR, t/mtyd 35.6 36.2 36.5 37.3 37.7 37.1
PR, tFe/m 2/d 20.4 20.6 20.8 20.7 20.8 21.0
RFB, ratio 1.00 1.02 1.03 1.03 1.05 1.06
SI, +6.4mm % 70.0 69.5 69.0 71.2 70.6 70.3
TI, +6.4mm % 66 66 67 67 67 68
Sinter Analysis
FeO % 6.1 6.3 7.3 5.6 5.8 6.8
Total Fe % 56.9 56.8 57.2 57.3 57.4 57.6
CaO % 8.8 8.7 8.4 7.8 8.4 8.1
SiOo % 4.9 4.8 4.6 4.3 4.6 4.5
AI2O3 % 3.0 3.1 2.9 2.7 2.8 . 2.9
MgO % .26 .29 .27 .24 .21 .27
C aO /Si0 2 - 1.80 1.81 1.83 1.81 1.83 1.80
Slag %
Ca -ferrite 32 34 36 34 36 38
Magnetite 24 28 30 21 25 28
Glass 8 9 10 7 7 8
Slag (G + DS) 13 13 14 12 12 13
146
Table 9.1.1-4 Sintering Process Model III at basicity of 2.0

Feed mix "M4" Coke in mix,7 %'
RF in mix, % 35 and 30 RFB, ratio
Basicity 2.0
Suction, kPa 10 12.5 15 10 12.5 15

Bed depth, mm 450 450 450 550 550 550
RF in mix, % 35 35 35 30 30 30
Coke in mix % 4.0 4.4 4.8 3.8 4.2 4.6
PR, t/m 2/d 35.5 35.8 35.9 36.5 36.3 35.5
PR, tFe/m 2/d 20.3 20.2 20.0 20.1 20.0 20.0
RFB, ratio 0.91 0.93 0.94 0.97 1.04 1.06
SI, +6.4mm % 68.0 67.5 67.0 71.0 68.7 68.3
TI, +6.4mm % 65 65 66 66 67 69
Sinter Analysis
FeO % 5.6 5.8 6.8 5.1 5.3 6.3
Total Fe % 56.3 56.5 56.1 55.8 56.1 56.3
CaO % 8.1 8.6 8.4 8.0 8.2 8.5
S i0 2 % 4.0 4.4 4.3 4.0 4.0 4.3
AI2O3 % 2.9 2.8 3.0 2.9 2.9 2.9
MgO % .30 .41 .29 .30 .28 .27
C aO /Si0 2 - 2.03 1.95 1.95 2.00 2.05 1.98
Slag %
Ca -ferrite 39 41 43 41 43 45
Magnetite 20 24 26 17 20 24
Glass 6 6 7 5 6 6
Slag (G + DS) 12 12 13 11 12 12
147
SINTERING P R O CESS M ODEL III
Feed Mix "M4" Coke in mix (% )
RF in mix (% ) 30 Suction (kPa) 10, 12.5,15
Bed Depth (mm) 450,550 RFB-450: 1.06, 1.07, 1.08
Basicity 1.6 RFB-550: 1.00, 1.01,1.03
Suction (kPa)
Fig 9.1.1-1 Sinter mineralogy, productivity and strength as a function of

suction.
148
SINTERING PR O CESS M ODEL III
Feed Mix "M4" Coke in mix (% )
RF in mix (% ) 30 Suction (kPa) 10, 12.5, 15
Bed Depth (mm) 450, 550 RFB-450: 1.00, 1.02, 1.03
Basicity 1.8 RFB-550: 0.96, 1.05, 1.06

suction.
149
SINTERING P R O CESS M ODEL III
Feed Mix "M 4" Coke in mix (% )
RF in mix (% ) 30 Suction (kPa) 10, 12.5, 15
Bed Depth (mm) 450, 550 RFB-450: 0.91,0.93, 0.94
Basicity 2.0 RFB-550: 0.97, 1.04, 1.06
36.7
36.6 - \ y - - 0 .2 x + 38.60
36.5
. >. r* - 0.89
5 *36.4
> 3 6 .3
J 3 6 .2
3 3 6 .1
36.0 -
t 35.9
■ 450
• — y - 0.08 x + 34.73
35.4 r1 - 0.92 550
35.3
10 12 14
Suction (kPa)
10 12 14 7.0
Suction (kPa) 6.8 - 450“
6.6 ■ y - 0.24 x + 3.07
6.4 - r1 - 0.87 __ ^
550»
6.2 ■
6.0
5.8 . ^
5.6 - ■
5.4 ' .
5.2 ♦
5.0 ' ^ y - 0.24 x + 2.57
4.8 r* - 0.87
4.6 -
4.4
^ 13.5 ■ y - 0.2 x + 9.83

w 450
Q 13.0 r1 - 0.75 a
+
S 12.5
5 1 2 .0 - a ■ ..-----' ' ♦
K 550
S ” '5
g.11.0 - ♦
y - 0.2 x + 9.17
(A
10.5 - r1 - 0.75
10.0 X. . 1____
10 12 14
Suction (kPa)
12
Suction (kPa)

suction.
150
1. The sintering Process Model III, applied in the six test series for the assessment
of the effects of suction on sintering, verified its effectiveness by producing
an orderly positioning of each combination of parameters (X,Y). Therefore,
this is the most effective model of the sintering technology developed to date.
2. Experiments at the three suction levels, with a basicity of 1.6 and 450mm bed
depth, resulted in a low productivity and low sinter quality-particularly, the
FeO content which was outside specifications. Therefore, these parameters
resulted neither in suitable sintering plant operation nor in acceptable blast
furnace burden material.
3. Experiments at suction levels of 10 and 12.5 kPa, with a basicity of 1.6 and at
550mm bed depth, gave the highest productivity and a balanced sinter quality.
Therefore, these are suitable parameters for sintering plant operation, and this
product meets blast furnace burden material specifications.
4. Experiments at a suction level of 15 kPa, with a basicity of 1.6 and at 550mm

bed depth gave a low quality sinter-particularly, the FeO content, which was
outside specifications. Therefore, these parameters resulted neither in suitable
sintering plant operation (high power cost) nor in acceptable blast furnace
burden material.
5. Experiments at suction levels of 10 and 12.5 kPa, with a basicity of 1.8, and at
450 and 550mm bed depth, gave high sinter quality, particularly at 550mm bed
depth, and the FeO content was within specifications. Therefore, these para
meters are suitable for sintering plant operation, and this product accomplishes
the required blast furnace burden material.
6. Experiments at a suction level of 15 kPa, with a basicity of 1.8, and at 450 and
550mm bed heights, gave poor sinter quality, with the FeO content again outside
specifications. Therefore, these parameters resulted neither in suitable sintering
plant operation nor in acceptable blast furnace burden material.
151
7. Experiments at suction levels of 10 and 12.5 kPa, with a basicity of 2.0, and at
450 and 550mm bed heights, gave the highest quality, particularly at 550mm.
Here the FeO content was at the lowest level, the calcium ferrite content reached
the highest level and the magnetite content the lowest level, but a loss of 2-3
t/m 2/d productivity occurred. Therefore, these parameters are suitable for
sintering plant operation only in terms of sinter quality.
8. Experiments at a suction level of 15 kPa, with a basicity of 2.0 and, at 550mm

bed depth, gave poor productivity and poor sinter quality. Therefore, these
parameters resulted neither in suitable sintering plant operation nor in
acceptable blast furnace burden material.
152
10.0 SIN TERIN G PLA N T TR IA LS
10.1 Overview
Once optimum parameters were established in the experimental work, ironmaking

manager gave permission for plant trials. The first plant trial was planned to operate
for two days, but was terminated by the operators after 8 hours at the end of day
shift, on 2nd February, 1972. The changes in productivity and sinter quality were
insignificant on an average result basis.
However, after evaluation of this preliminary trial and further discussion, management
approved a revised plan for a one week uninterrupted plant trial with the condition:
"If the experimental mode shows a significant contribution to productivity and sinter
quality during the trial period, the sintering plant will indefinitely continue operating
at these conditions. Otherwise, trials will be suspended and normal operating conditions
resumed". .
One of the most critical and decisive motivations during this trial was the belief that
expert visual observations are the simplest and fastest system of communication, and
instrumental tests are thie proof of the facts. The trials were observed at all times by
a technical staff member of the Ores & Sinter Research Section, as an hourly
communicator and the supervisor of raw materials and product sinter hourly sampling.
Trials at 455mm depth, 9.5 kPa suction and with a basicity of 1.55 were carried out
from 25 February to 26 March 1972. These trials continued at basicities of 2.0, 2.4,
2.5 and 2.8, which finished on 25 May 1972. Subsequently, Sintering Machine No 2
was stabilized to operate at between 1.5-1.6 basicity, 450-500 bed depth and 10 kPa
suction. However, further trials were continued at a basicity of 1.8, between October
and November 1972.
153
10.1.1 Pre-Trial Sintering Plant Operating Conditions and
Results
Typical No 2 Sintering Machine operating conditions and results, between 14 and 26

December, 1971 were:
Table 10.1.1-1 Pre-trial Port Kembla plant sinters
Date, December 14 16 17 18 26.12.71

Bed depth, mm 250 254 254 254 254
WB suction, kPa 8.3 8.8 8.5 9.0 8.5
WB temp, °C 260 227 221 193 182
HS temp, °C 449 488 477 510 421
RF in mix, % 41 41 34 44 41
Coke in mix, % 4.3 4.3 4.7 4.1 4.5
Coke, kg/t-S 86 85 84 88 91
TI, +6.4mm % 64 65 66 65 66
Sinter Size (%)
+38mm 14 21 6 11 12
+ 10mm 68 67 40 65 44
+6.4mm 88 86 74 83 79
Sinter Analysis
FeO % 15.5 14.8 16.8 17.1 16.8
Total Fe % 59 59.0 58.8 58.9 58.3
CaO % 6.8 7.0 7.1 7.0 7.5
SiOo % 6.6 6.9 7.7 6.5 7.7
O
2.9 3.0 3.1 3.1 3.0

>
CO
bO
C aO /Si02 1.03 1.01 0.92 1.07 0.97

Calcium ferrite 3 6 1 2 2
Total hematite 19 13 11 20 14
Magnetite 59 62 68 57 69
Glass 19 19 20 21 15
Dica-silicate 0 0 0 0 0
154
10.1.2 Port Kembla Sintering Plant Trial and Results
Results of three months’ operation under trial conditions confirmed that it was possible
to produce a super-fluxing sinter in deep beds. As a result, significant benefits were
observed at Port Kembla Sintering Plant and at the blast furnaces.
Plant trial results are illustrated in Table 10.1.2-1, and the correlation of the
experimental reactor unit results with No 2 Sintering Machine are shown in Figs
10.1.2-1 to 10.1.2-4.
Micrographs of typical sinter products at basicities of 1.6, 1.8, 2.5 and 3.2, and at 460,
460, 483 and 355mm bed depths respectively are shown in Figs 10.1.2-5 to 10.1.2-8.
The phase identifications of Figs 10.1.2-5 to 10.1.2-7 representing the No 2 Sintering
Machine products, plant trials T-l, T-2 and T-4 are listed in Table 10.1.2-1.
Fig 10.1.2-8 shows the sinter at a basicity of 3.2 and at 355mm bed depth. This is the
only material produced through No 1 Sintering Machine in a plant trial conducted on
24 April 1972. The phase identification is:
(Vol %)
Ca-ferrite (CF) 59
Total hematite (TH) 12
Magnetite (M) 10
Glass (G) 4
Dica-silicate (DS) 15
The cut-off points between the sintering plant base operating conditions and the trial
conditions, in terms of the sinter mineralogical components, the FeO content and the
sinter strength (expressed as the rumble index and the -6.4mm sinter fines content
of the blast furnace feed) are set out in Figs 10.1.2-9 to 10.1.2-11.
Finally, the new position of the Port Kembla sinter compared to Kwinana, Newcastle,
Port Kembla pre-trial and trial, and Japanese sinters in terms of its microstructural
components are illustrated in Figs 10.1.2-12.
155
Table 10.1.2-1 Port Kembla Sintering Plant Trial operating
parameters and results
Plant Base T-l T-2 T-3 T-4 T-5

Bed depth, mm 305 455 464 432 483 355
WB suction, Kpa 8.7 10.3 9.9 9.7 9.3 10.5
Basicity: 1.10 1.55 1.80 2.00 2.50 2.80
FI temp, °C 132 105 95 121 132 123
WB temp, °C 202 140 122 163 118 211
HS temp, % 525 423 402 356 340 316
Feed mix: "M4" "M4" '"M2" "M3" "M3" "M2"
RF in mix % 38.3 26.0 31.1 33.4 39.7 32.5
Coke in mix % 4.3 3.7 3.5 4.0 3.4 2.3
Coke, kg/t-S 82 58 56 45 68 40
PR, t/m Vd 30.9 35.4 31.6 32.0 28.8 26.4
PR, tFe/m 2/d 18.0 20.1 18.5 17.8 16.0 14.5
TI, +6.4mm % 65 72 74 74 79 78
Sinter Sizing (%)

+30mm 13 14 13 8 18* 7*
+10mm 44 68 65 62 77 60
+6.4mm 80 86 87 82 92 90
Sinter Analysis
FeO % 11.9 7.7 8.7 6.8 4.9 5.2
Total Fe % 58.2 56.7 58.4 55.7 55.4 55.1
CaO % 8.6 10.3 9.7 11.5 13.0 13.9
S i02 % 7.1 6.5 5.4 5.7 5.2 5.0
A120 3 % 3.2 2.9 3.1 2.9 2.6 2.5
C aO /Si02 1.21 1.58 1.8 2.02 2.50 2.78

Calcium ferrite 14 31 44 47 57 57
Magnetite 36 27 23 23 18 11
Glass 12 7 7 4 5 7
Dica-silicate 0.3 3 5 4 10 14
Code: Base = From 2-24 January 1972

FI = Fan inlet T-l = Trial 25 Feb-26 Mar 1972
WB = Windbox T-2 = Trial 13 Oct-24 Nov 1972
HS = Hot screen T-3 = Trial 27 Mar-4 Apr 1972
* = +40mm % T-4 = Trial 15-25 May 1972
S = Sinter T-5 = Trial 20-24 Apr 1972
156
3 5
3-4 ^ -^ca
^ V
3 3 - - _. . X
- — \ N. \ o
•3. 3 2 - ch ^
\ V
3 1 \
S<•_t' 3 0 - \\ \ s
2 9 \ s
¿RI 2a
2 ~7
1 .2 . 1.-4- 1 . 6 1 . a 2 . 0 2 . 2 2 .-4 - 2 . 6 2 .3 3 .0 3 .2
B a s ic ! ty ( C a 0 / S I 0 2 )
2 0
1 9
^ i a " ^ ----------
S- S '- 11 76 :
.-2 7 ' 1 5
•-g 1- 4- - " " -0
*1 13 -
q: 12 - ----------___
i i
1o
1 . 2 1 .-4 - 1 . 6 1 .S 2 . 0 2 . 2 2.-4- 2 . 6 2 . S 3 . 0 3 . 2
S'«I
« -O
26
2 -4 _________ R i a n t
2 2 _____ Ex p . R o a c to r
20
i a
83
16
1 -4
1 2
1 O
a
6
1 . 2 1.-4- 1 . 6 1 . a 2 . 0 2 . 2 2 .-4 - 2 . 6 2 . a 3 . 0 3 . 2
B a s i c ! ty ( C a 0 / S I 0 2 )
Fig 10.1.2-1 Correlation between the experimental reactor unit and the PK sintering
Plant trial results as a function of the basicity (C a0/S i02), bed depth (mm) and suction
(kPa).
157
(%)
RF in mix
Coke in mix (% )
Coke Rate (k g /t-s)
Coke Saving (% )
Basicity ( C a 0 / S i 0 2 )
Fig 10.1.2-2 Correlation between the experimental reactor unit and the PK Sintering
Plant raw material parameters (RF level, coke rate and coke saving) as a function of
the basicity, bed depth and suction.
158
Sin te rin g Tim e (m in u te*)
Fig 10.1.2-3 Relationship between the experimental (reactor) air velocity (m/min) and
sintering time (min) as a function of bed depth (254-465mm), basicity (1.2-2.8 CaO/Si02)
and suction (9-10kPa).
159
t,
o 20 40 60
Sinter Size Distribution (m m )
Fig 10.1.2-4 Port Kembla plant sinter size distribution at bed depth of 287mm, 8.5kPa
suction and. 1.2 basicity compared to the product sinter at bed depth of 465mm, 10.2kPa
and 1.55 basicity.
160
Fig 10.1.2-5 Typical sinter at a basicity of 1.6 and at 460mm bed depth. Magnification
x 200.
x 200.
161
x 200.
Fig 10.1.2-8 Typical sinter at a basicity of 3.2 and at 355mm bed depth.
162
HEMATITE %
C a-FE R R IT E %
MAGNETITE % GLASS %
PLANT TRIAL (HRS)

Fig 10,1.2-9 Change of the sinter mineralogical composition from the base period at
a basicity of 1.1 and 300mm bed depth to the plant trial period at a basicity of 1.6
and 460mm bed depth.
163
FeO % MAGNETIC
PERMEABILITY
70
60
50
40
30
20
10
Fig 10.1.2-10 Change o f the sinter FeO content and magnetic permeability from the
base period to the tria l period.
164
- 6 .3 5 m m RUMBLER
PLANT TRIAL (HRS)

Fig 10.1.2-11 Change of the sinter fines content (shatter index) and the sinter strength
(R I or TI), from the base period to the trial period,
165
10 20 30 40 50 60 70 80 90
% H E M A T IT E --------->
— •— A Newcastle Sinter
— .— port Kembla Sinter before 25 Feb 1972
»—«-«.««.a K w inana Sinter
'»111,11:hm«»«!9 Japanese Sinter
t X Port kembla Sinter after 25 Feb 1972
(Calcium fe rrite + hematite + magnetite => 100%)
Fig 10.1.2-12 Relationships amongst the three major phase components in sinters from
Japan, BHP-Newcastle, K w inana and Port Kembla Steelworks.
166
The conclusions derived from the sintering plant trial based on the experimental work
in sintering the reactor (pot) testing and results from No 2 Sintering Machine are:
1. Sintering plant operations were satisfactory at any basicity level up to 2.80

C a 0 /S i0 2. Operation at between 1.5 and 1.8 basicity was satisfactory and since
this level was close to the desired maximum for blast furnace operation with
60-70% sinter burden, regular sinter was produced at this level
2. The bed depth sinter mineralogy changed from a high magnetite-glass bond to
high calcium ferrite-glass bond. The texture changed from homogeneous, highly
fused sinter to heterogeneous, less fused and more porous sinter.
3. A marked decrease in FeO content, hence a marked increase in the degree of

sinter reducibility, was achieved.
4. The rumbler index improved, from the usual 66-67 up to 75-76, indicating a
significant improvement in sinter strength.
5. The total return fines loading dropped from 38% to 27% during the trial. As
a result, sinter productivity was increased.
6. A significant reduction in fuel consumption was achieved, showing a potential

financial saving.
7. Sintering operating temperatures were reduced, with a resultant lowering of
costs for maintenance of hot screens, coolers and product handling conveyor
belts.
8. In terms of the sinter microstructural composition, the Port Kembla sinter

achieved the same level as the Japanese sinter.
167
R EFER EN C ES
[i ] Pietsch W., " Pressure A gglomeration - State of the Art". Agglomeration 77,
Vol 2, A tlanta Ga., 6-10, 1977, pp 649-677
[2] English A., & Frans R.D., "Developments in Pelletizing". Agglomeration 77

Vol 1, A tlanta Ga., March 6-10, 1977, pp 3-22. *
[3] Ilmoni P.A., et al., "Comparison of The Pelletizing Methods at LKAB".

Agglomeration, Nürnberg, 1981, pp B1-B22 (Figs 1-4).
[4] Thomson I.A., "Production of Acid Pellets in Australia". International Iron

and Steel Institute, Rio de Janeiro, Brazil, 1979, pp 51-72.
[5] Poetschka N.L., et al., "Open Circuit Disc Balling at Hamersley’s Dampier
Pelletizing Plant". Pellets and Granules, The Australasian Institute of Mining
& Metallurgy, Newcastle, 1974, pp 43-50
[6] Kinsey F.W., "Agglomeration of Waste Oxides in a Steel Mill".

Aggomeration 77, Vol 2, Atlanta Ga., March 6-10, 1977, pp 624-646.
[7] Czako C.A., et al., "Laboratory Testing as basis for Optimum Engineering of
Sintering and Pelletizing Plants". Agglomeration 77, AIME New York, 1977,
Vol 1, pp 243-271.
[8] Stieler F.K., "Development of Latest Experience in U pdraft Sintering of Lead

and Lead/Zinc Concentrates". Agglomeration 77, Vol 2, Atlanta Ga., March
6-10, 1977, pp 604-623.
[9] Green F.A., "The Port Pirie Smelters". BHAS, Published by the Broken Hill
Associated Smelters Pty Ltd., Melbourne-Australia, 1977, chapter 9, pp 89-107.
[10] Kristensen A., "Agglomeration". Technical Services Report, Australian Iron &
Steel Pty Ltd, Port Kembla, 1975, pp 1-14.
[11] Danielsson C., "Sintering Practice at Domnarfvet, Sweden". Symposium on

Sinter, The Iron & Steel Institute, London, 1955, pp 142-144.
[12] Hopkins R.J., "Short History of Sintering Developments at Port Pirie". Sintering
Symposium, Port Pirie, South Australia, September 1958, pp 5-20.
[13] Jennings R.F., et al., "Sinter Plant Assessment Trials at Dagenham and at
Cleveland". Symposium on Sinter, The Iron and Steel Institute, London 1955,
pp 110-120.
[14] Technical S taff, "Materials, Methods, Men and Maintenance in Sintering".

Sintering Symposium, The Broken Hill Associated Smelters Pty. Ltd., Port
Pirie, South Australia, September, 1958, pp 285-306.
[15] Peucker M.D., et al., "Gas Recirculation in Sintering of Lead and Zinc Con
centrates". Sintering Symposium, Port Pirie, South Australia. September, 1958,
pp 261-284.
168
[16] Hallams K.K. & Greenlees J., "Engineering Problems in Sinter Plants - Plant
Design, Maintenance, etc." Sintering Symposium, Port Pirie, South Australia.
September 1958, pp 325-381.
[17] Mikka R.A. & Bagnall E.J., "Rapid Sinter FeO Determination by Magnetic
Permeability - A new Method". 3rd International Symposium on Agglomeration,
Nürnberg, 6-8/5/1981, pp 185-195.
[18] Australian Iron & Steel Pty. Ltd. Port Kembla, Operational Report, May 1960.
[19] Bogan L.C. & Worner H.K., "Some Researches into the Sintering of Fine Hematite
Ores". Sintering Symposium, Port Pirie, South Australia, September, 1958, pp
[20] Barry P. & Drinkwater J., "Sinter Plant Data". BHP - Slab & Plate Products
Division - Port Kembla Steelworks. Personal discussion, October, 1990.
[21] Lathlean D., "Blast Furnace Data". BHP Steel - Slab & Plate Products Division
- Port Kembla Steelworks, personal discussion, October, 1990.
[22] Woolfrey J., et al., "Sinter Plant Data". BHP Steel - Rod & Bar Division,
Newcastle Steelworks. Personal communication, October 1990.
[23] American Corporation, "The Encyclopedia Americana". International edition

1975, Voi 15, pp 448-449.
[24] Ostwald J., "Mineralogy of Australian Iron ores". BHP Technical Bulletin, Vol
25 No.l, May 1981.
[25] McLeod W.N., et al., "Geology-Iron Ores of Australia". Eighth Commonwealth

Mining & Metallurgical Congress Australia and New Zealand, 1965.
[26] Facer R.A., "Iron in Australia". Department of Geology, The University of

Wollongong, Note prepared on 9 July, 1970.
[27] McLennan I., "Steelmaking Raw Materials in Australia". Third Annual Meeting
and Conference of the International Iron & Steel Institute of Tokyo, Japan,
October 12-15, 1969.
[28] Harms J.E., et al., "Pisolitic Limonite Deposits in North West Australia". Proc.
Aust. IMM., 1964, pp 212, 91.
[29] BHP S taff, "Yampi Sound Iron Ores in Economic Geology of Australia and
Papua - New Guinea" I. Metals Monograph 5, Aust IMM., 1975, pp 253.
[30] Miinive L., et al., "Microstructural and Sintering Investigation of the Kwinana
Sinter Plant Raw Materials". Research Dept. Report R. 29/971, 1971, AIS-PK,
Australia.
[31] Whitehead S., et al., "Petrological Notes on the Koolyanobbing Iron Ore For
mation". BHP Tech. Bulletin 1965, 9 (1), 18.
[32] Ayers D.E., "The Hematite ores of Mt. Tom Price and Mt. Whaleback, Ham-
mersley Iron Province". Proc. Aus. IMM, 1971, 238, 47.
169
[33] Lindley J.G., "Genesis of the Mt. Whaleback Iron Ore Deposit". Thesis, Uni
versity of Adelaide, 1975.
[34] Chukhrov F.V., et al., "New Data on Iron Oxides in Weathering Zone". Proceed.
1972, International Clay Conference, Madrid 1973, 333.
[35] Chukhrov F.V., "Mineralogycal and Geochemical Criteria in the Genesis of

Red Beds". Chemical Geology, 1973, 12, 67.
[36] Schwertmann U. et al., "The Influence of Fe (11), Si and pH on the formation

of Lepidocrocite and Ferrhydrite during oxidation of aqueous FeCLo Solutions".
Clay Min. 1976, 11, 189.
[37] CSR S taff, "Mineralogy of Clays". CSR Research Laboratories Report C l0/75,
1975.
[38] Willis W.B, "Marra Mamba Handbook Volume 1 Geology", Mt. Newman Mining
Co. Pty. Ltd., 1978, 1, 33.
[39] Bardossy G., "The Geochemistry of Hungarian Bauxites". Acad. Geol. Acad.
Sci. Hungary 1958, 5, 103.
[40] Trendall A.F., "Second Progress Report on the Brockman Iron Formation in
the Wittnoom - Yampire area, Western Australia". W.A. Geol. Survey Ann. Report
1975, 75.
[41] Gale M.J., "Mineralogy and Petrology of very Low Grade Metamorphic Archaen
Banded Iron Formations, Weld Range, Western Australia". Am. Miner. 1980, 65,
8.
[42] American C orporation," The Encyclopedia Americana". International Edition

1975, Vol 11, pp 470; Vol 17, pp 490.
[43] Boyton R.S., "Chemistry and Technology of Lime and Limestone". Edition 1966,
John Wiley & Sons, Inc. pp 1-458.
[44] Encyclopaedia Britannica Inc., "Enciclopaedia Britannica". Edition 1975

Macropaedia , Vol 10, pp 979-984.
[45] Miinive. L., "Potassium Salts and by-products from the Sechura Desert Brine
Deposits, using solar energy\ Minera Bayovar S.A. Peru. Research Reports
PPR 1-28/65-67, 1967.
[46] Miinive L., "Assessment of Current Coke Sampling & Processing Practices as a
Basis for Decision Making". BHP Steel - Slab & Plate Products Division,
Ironmaking Technology & Development Report, November 1985.
[47] Lowcock K. et al., "Comparison of Port Kembla versus Japanese Sintering

Practice", BHP Steel - Slab & Plate Products Division, Ironmaking Technology
& Development Report, June 1984, pp 3.
[48] Le Corre C. et al., "In situ mineralogical and morphological investigations on

industrial sinter cakes". 3rd International Symposium on Agglomeration,
Nürnberg, 6-8.5.1981, pp 121 -129.
170
[49] Nightingale R.J. et. al., "Cost Reduction in the Sintering Process. A.I.M.E.
Proceec*ings, v °l- 46, Pittsburgh meeting, March 29-April 1, ¡987,
[50] Wendeborn H. et. al., "Limits of the Physical aspects of the Sinter Process"
Agglomeration based on an International Symposioum held in PhiladelDhia
Pennsylvania, April 12-14, 1961, pp 1053-1058. *
[51] Fleming G. et. al., "Process Control Aspects of a Sinter Plant". Agglomeration
77, Volume 2, Atlanta, Ga., March 6-10, 1977, pp 568.
[52] Russell A. and Hamilton G., "Operations Report on High Fluxed-High Bed
Sinter". Australian Iron & Steel Pty. Ltd., Port Kembla, Sinter Plant Department
Report, 9 June, 1972, pp 1-24.
[53] Dieckmann R.J., "Comments on Microstructures of High Bed Height Trial

Sinters". Australian Iron & Steel Pty. Ltd., Port Kembla, 7 March, 1973, pp 1-3.
[54] Bogan L.C., "Summarising Report on the Port Kembla Seminar on the Devel
opment of Highly Fluxed Sinters". BHP. Co. Ltd. Iron & Steelworks, Newcastle,
12 October, 1972, pp 1-19
[55] Jennings R.F. & Grieve A., "The Basis of Sinter Plant Design". Huntington
Heberlein & Co Ltd., London, England. Agglomeration based on an Intenational
Symposium held in Philadelphia, Pennsylvania, April 12-14, 1961, pp 495-497.
[56] Simpson J., "Hot Testing Operations Manual". BHP Steel - Slab & Plate Products
Division, Raw Materials Investigation Section. August 1988, pp 1-7.
[57 Gadsden W.R., "G. 22/7 Qualities of Kwinana High Basicity Sinter" Australian
Iron & Steel Pty. Ltd., Port Kembla, letter report to Manager Quality Control,
18 September 1970.
[58] O’Brien R.J., et al, "An Investigation of the Properties of Kwinana Plant
Sinter", Australian Iron & Steel Pty, Ltd., Port Kembla, Research Department
Report No R.33/1070, 1st October 1970
[59] Laver P.J., "G22/7- Sintering Characteristics at Kwinana" Australian Iron &
Steel Pty, Ltd., Kwinana Steelworks- Letter report to General Manager Aus
tralian Iron & Steel Pty. Ltd., Port Kembla, 7 Janauary, 1971.
[60] Miinive L., "G22/7- Investigation of the Influence of flux size distribution on
sintering" Australian Iron & Steel Pty .Ltd., Port Kembla. Letter report to Iron
making managers, 19 October, 1972.
[61] Bagnall E.J., "Influence of Feed Material Properties on Sinter for Blast Fur
naces" Agglomeration 77, Volume 2, Atlanta, Ga., March 6-10, 1977, pp 587.
[62] Rigby G.R. & Green A.T., "The Thin-Section Mineralogy of Ceramic Materials",
Book published by the British Ceramics Research Association, Second edition,
1953
[63] Perry R.H. & Chilton C.H., "Chemical E n g in e e r Handbook" Fifth edition,
1973, McGraw-Hill Chemical Engineering Services.
171
[64] Ayl ward G.H & Findlay T.J.V., "SI Chemical Data", Second edition 1981, John
Wiley & Sons.
[65] Bert Phillips & Arnulf Muan, "Phase Equilibria in the System CaO-Iron Oxide
in Air and at 1 Atm. 0 2 Pressure," J. Am. Cerm. Soc.. 41,445-54 (1958).
[66] I.S., 'Melting Points of Iron Oxides on Silica; Phase Equilibria in the System
Fe-Si-0 as a function of Gas compostion and temperature", J. Am. Cerm. Soc
70, 2046-53 (1948). ’
[67] Bert Phillips and Arnulf Muan, "Phase Equilibria in the System CaO-Iron
Oxide-Si02 in Air", J. Am. Cerm Soc., 42, 413-23 1959.
[68] Gadsden W.R. & O’Brien R.J., "The Microstructure of A.I.S Sinters "Australian
Iron & Steel Pty. Ltd., Port Kembla, Research Department Report No R. 19/566,
31 May 1966.
[69] Munive L & Me Auley I., "The Influence of Return Fines Level and Size, Bed
Height, Suction and Basicity on Sintering" Australian Iron & Steel Pty. Ltd.,
Port Kembla, Raw Materials Research Report No R. 24/1172, 20 November
1972.
[70] Bogan L.C., "G.22/7- Sinter Technology Report on the Port Kembla Seminar
on Highly Fluxed Sinters", BHP. Co. Ltd., Newcastle, 12 October 1972, pp 3.
[71] Lowcock K. and Allen D.V., "Blast Furnace Operation on Deep Bed Sinter",
A.I.S. Pty. Ltd., Port Kembla, Process Chemistry Report, 6 July 1972.
[72] Voice, E.W., et al., "Factors Controlling the Rate of Sinter Production", Sym
posium on Sinter, the Iron & Steel Institute, London, March, 1955, pp 43-98.
[73] Wild, R. and Dixon, K.G., "Pressure and Water Gradients through a Sinter Bed",
Agglomeration, Philadelphia, Pennsylvania, April 12-14, 1961, pp 565-583.
[74] Voice, E.W., et al., "The Permeability of Sinter Beds", Symposium on Sinter,
the Iron & Steel Institute, London, March, 1955, pp 99-102.
[75] Jennings, R.F., "The Rating of Sinter Plant for Economic Output", Symposium
on Sinter, The Iron & Steel Institute, London, March, 1955, pp 126-134.
[76] Miinive L. & Felton W., "Characterization of Sintering Plant Ore Fines and a
Comparison between Predicted and Actual Plant Sintering Rates". A.I.S. Port
Kembla Works Research Department. Report T.C. 27/971, September, 1971.
[77] Munive L., "Cyclone Classification of Koolan and Cockatoo Fines and
Investigations of Gangue content and size Distribution on Sintering Char
acteristics". A.I.S. Port Kembla, Technical Services Department, Research &
Development. Research Department Report No R. 34/1070.
[78] McBriar, M.E. (Miss), et ai, "The Nature of Ironstone Sinter", The United Steel
Companies, Ltd., Research and Development Department, Rotherham, 'Sym
posium on Sinter, the Iron & Steel Institute, London, March, 1955, pp 145-152.
172
[79] Ban, T.E., et al., "The C ontinuous Sintering Process-Research and A pplications",
D w ight-L loyd R esearch L aboratories, McDowele Company Inc C leveland’
Ohio A gglom eration, P hilad ep h ia, Pennsylvania, A pril 12-14, 196l’ pp 511-140*
[80] Wada, K., et ah, "Automotion of the Sintering M achine", In stitu te of Y aw ata
Works, Y aw ata Iron & Steel Co., Ltd., A pril 12-14, 1961, pp 541-563.
[81]] K nepper, W.A., et ah, "Study o f the Properties of Self-Fluxing Sinters", U.S.
Steel C orporation, A pplied R esearch L aboratory, M onroeville, Pennsylvania.
A gglom eration, P hiladelphia, Pennsylvania, A pril 12-14, 1961, pp 787-807.
[82] W atanabe, A., "Fundam ental Studies of Self-Fluxing Sinter", K okura Steel
w orks, Sum itom a M etal Industries Ltd., K okura, Japan. A gglom eration,
P hilad elp h ia, Pennsylvania, A pril 12-14, 1961, pp 865-900.
[83] Bogan, L.C. and W orner, H.K., "Structures and Bonding M echanism in Sinters",
BHP C entral L aboratories, Shortland, NSW, A ustralia. A gglom eration, P hi
ladephia, Pennsylvania, ap ril 12-14, 1961, pp 901-929.
[84] Ish ik aw a Y., et ah, "Latest D evelopm ent on Sintering Technology", Nippon
Steel C orporation, Japan. A gglom eration 77, A lanta, Ga., M arch 6-10, 1977,
Vol.2, pp 503-525.
[85] Jeu lin , D., et ah, "A New Technique fo r the Q u an tificatio n of Iron Ore Sinter
M ineralogy", In stitu t de R echerches de la Siderurgie Française, IRSID., France.
A gglom eration 77, A lanta, Ga., M arch 6-10, 1977, vol. 2, pp 526-537.
[86] Young, P.A., "On-Strand Cooling of Sinter". Mining M ineral Science, U niversity
of Leeds, U K . A gglom eration 77, A lanta, Ga., M arch 5-10, 1977 vol. 2, pp
538-567.
[87] H ideaki Som a, et ah, "Studies on Melt Form ation of Iron Ore Sinter and
A pplication of Study R esults to P ractical Operation", N ippon Steel C orporation,
Japan. 3rd In te rn a tio n a l Symposium on A gglom eration, N ürnberg, 6-8. 5. 1981,
pp 12-135.
[88] R aush H and Cappel F., "Comparison between C onventional Hot A ir, and Mixed
F ired Sintering", Lurgi G esellschaft fu r chemic und H üttenw esen, F ra n k fu rt
am M ain, G erm any. A gglom eration, Philadelphia, Pennsylvania, A pril 12-14,
1961, pp 455-479.
[89] Josephson G.W., et ah, "Iron B last-Furnace Slag Production, Processing,

Properties, and Uses". U nited States D epartm ent of the In terio r Bureau of
Mines, 1949, pp 1-260.
[90] Ward R.G., "An Intro d u ctio n to the Physical Chem istry of Iron and Steel
Making". Steel Company of C anada Professor of M etallurgy McMaster U n i
versity, H am ilton, O ntario, 1962, pp 1-215.
[91] C raw ford D.P., "Blast F urnace Theory: A Review". BHP Ironm aking Technical
D evelopm ent, N ew castle, Dec 1981, pp 1-68.
173
A PPEN D ICES
A. P R ESSU R E AGGLOM ERATION PRO CESS
The basic principle of pressure agglomeration is very old and not well documented.
The first applications were most probably in medicine for the production of pills, and
in metallurgy for agglomerating fine particles of coal, and non-ferrous materials.
Agglomeration by extrusion commenced in the U.S.A. as early as 1848. At that time,

a patent for a method of converting coal fines into briquettes by pressure alone was
obtained. This technique was introduced into Germany for the briquetting of brown
coal in 1858.
The Grondal process was an important step in developing quality agglomerates. This
process was introduced in Sweden in 1902, to agglomerate 0.1mm size magnetic con
centrate under pressure into briquettes, which were then fired in a tunnel furnace.
The same year briquetting machines were introduced by the Broken Hill Proprietary
(B.H.P.) smelting works at Port Pirie, in Australia, to consolidate the roasted con
centrates mixed with some finely divided fluxes. The low porosity of briquettes and
their poor physical strength made them less acceptable than other agglomerates for
use as blast furnace burden material.
A modern method of briquetting is the FIOR process. This process consists of the
prereduction of high-grade iron ore fine particles or concentrates in a hot gaseous
reactor to produce partially metallized briquettes suitable for electric-arc steelmaking
furnaces.
Briquetting of iron ores became more frequent from 1905 and in 1912 there were
about 25 briquetting plants throughout the world. At that time, the agglomeration
process was used to treat fine ores, flue dust and sulphide residues mixed with a wide
range of binders.
175
Briquetting machines, classified as a pressure agglomeration process, may be considered
a predecessor of the present conventional sintering process at high temperatures, using
coarser fine ores.
The first successfully operating roll pressing machine was constructed by the Belgian
Loiseau and was installed in a briquetting plant at Port Richmond, in the U.S.A., in
the late 1870's. Similar machines were built shortly thereafter by Fouquemberg,
Zimmerman and Hanrez, Schuchterman and Kremer and others.
The first isostatic pressing technique was patented by the Westinghouse Lamp Co, in
the U.S.A. [1] in the late 1900's.
Pressure agglomeration techniques have become important in more and more com
mercial application of the technological advances of the second half of the 20th
century.
176
B. NODULIZING PRO CESS
Nodulizing consists of feeding fine ores, flue dust or both together mixed with solid
fuels, to a gas fired rotary kiln. The first nodulizing kiln was construced by Feliner
and Ziegler for the Ferniewerken in Giessen, in 1908. This particular kiln was fired
using pulverised coal.
By 1912, there were 11 plants, most of them situated in Germany and Austria.
177
C. D EVELO PM EN T OF T H E PELLETIZIN G PR O CESS
C .l Overview
It was in the middle 1940’s when the Mines Experimental Station of the University
of Minnesota in the U.S.A. made a major breakthrough in pelletizing. At the same
time, contributions were also made by the Royal Institute of Technology in Sweden.
From this exploratory success, the first large-scale pelletizing tests were conducted in
a pilot plant vertical shaft furnace by the Armco Steel Corporation at Ashland,
Kentucky, in 1950 and 1951. Following this successful operation, a four-shaft furnace
type commercial plant was built at Babbit, Minnesota, by Reserve Mining Company
in the U.S.A [2].
The straight grate pelletising machine was introduced by Arthur G. McKee & Company
and Allis-Chalmers, at the same time also by the Mines Experimental Station, in the
U.S.A. in 1952. Soon thereafter, Lurgi of Germany also became identified with
pelletizing, using the Lurgi design - a straight grate or sintering type machine.
By 1956, four systems of pelletizing became available:
i. Shaft furnace;
ii. Straight grate by A.G. McKee and Allis-Chalmers, which was designed later as
the updraft and downdraft system;
iii. Straight grate by the Mines Experimental Station, which was called the updraught
system; and
iv. The straight grate by Lurgi, designed as the downdraft - updraft system.
From these innovations, commercial pelletizing plants were developed on the Mesabi
Range, a large deposit of taconite (a siliceous iron formation consisting of fine-grained
silica (Si02) mixed with magnetite (Fe30 4) and hematite (Fe20 3)) in Minnesota, adopted
throughout the U.S.A., and expanded successfully world wide.
Later, the Surface Combustion Division of Midland-Ross Corporation developed

another pelletizing system called the "Heat Fast” process, in 1964. This consisted of
a drier and a circular-type grate integrated with a shaft-type cooler. A more recent
development was the circular grate pelletizing machine by Arthur G. McKee & Company
in 1970: the first commercial plant of this pyrometallurgical system was constructed
by them in Mexico in 1974.
Initially, pelletizing processes were restricted to very fine magnetite ore, minus 53
micron (-300 mesh), and acid pellets, because exothermic oxidation occurs during
induration and so fuel requirements are low. As the proportion of hematite (an
178
endothermic material) increases, there is a corresponding increase in the fuel
requirement. When self-fluxing pellets are produced, the fuel consumption also
increases according to the type of material that is initially being pelletized.
For these reasons, pelletizing process improvements over the years focussed on recovery
of heat and prevention of heat loss. These objectives have been achieved by better
equipment design and selective use of lining refractories, which provided the possibility
of pelletizing high-grade hematite and self-fluxing pellets during the 1960’s [2].
In excess of 300 million tonnes of pellets were produced by the pelletizing industry
in 1981, and the process is considered to be one of the greatest commercial developments
of our time.
The pelletizing process has been successfully applied not only to ferrous materials but
also in non-ferrous fields, such as the manufacture of cement, pelletizing of phosphates
for the manufacture of elemental phosphorus, manganese ore, etc.
179
SHAFT FUHNACE
Fig. C-l Shaft furnace [3].
Fig C-2 Straight grate process [3].
180
M odified G rete K iln , Svappavaara
Fig C-3 Grate Kiln process [3],
Fig C-4 Possible future pelletizing plant [3].
181
Fig C -l illustrates an LKAB’S shaft furnace, where induration takes place in a
counter-current process [3].
Fig C-2 shows the principle of the straight grate process, where a layer of green pellets
about 400 mm thick moves on an endless conveyor through the various zones of the
machine. The pellets lie still throughout the process until they have attained full
strength [3].
Fig C-3 outlines a grate kiln process which consists of three stages that take place in
separate units [3]:
- a drying and preheat stage carried out on a grate,
- a firing stage that takes place in a rotary kiln, and
- a cooling stage which takes place in a circular cooler.
Fig C-4 suggests a future pelletizing plant, with features drawn from experience, amd
from various sources [3].
C.2 Pelletising Plants in Australia
The pelletizing process is being used in Australia to treat iron ores at the following
locations:
C.2.1 Savage River
The Savage River magnetite ore (35% Fe) formations in Tasmania were discovered in
1876. The central ore body originally contained over 100 million tonnes of crude ore,
from which it is estimated that more than 45 million tonnes of concentrate at 69.2%
Fe content can be produced.
The pelletizing plant is situated at Port Latta. The annual pellet production is 2.5
million tonnes, using a magnetite concentrate wet ground to 85% minus 53pm. Balling
moisture ranges from 9.2% to 9.5%. The furnace feed rate is varied between 66 and
86 t.p.h. Sixty percent of the air to the furnace is preheated to 1300X1. The total iron
content of the pellets averages 67% and over 90% of the pellet product is between 16
mm and 9mm.
C.2.2 Hamersley
The Hamersley pelletizing plant is situated on the north west coast of Western Australia
at Dampier. Hematite ore fines, minus 6mm in size are dried to about 1% moisture
and then dry ground in ball mills to a Blaine surface area of 240m2/kg. A balling
moisture of approximately 7.5% is used. The green balls are indurated at a temperature
of 1320X, and 90% of the pellets are in the size range.between 16mm and 9mm. The
production rate is 3 million tonnes per annum, all of which is exported to Japan [4]
[5].
182
C.2.3 Whyalla
The Whyalla pelletizing plant is situated in South Australia. Feed to the plant is dried
in a rotary kiln dryer to about 0.8% moisture and then dry ground in ball mills to
give a product having a Blaine surface area of 170m2/kg. Balling moisture averages
8.5%. Induration firing temperature is 1350X and the production rate is 2.3 million
tonnes per annum.
C.2.4 Robe River
This pelletizing plant is situated at Cape Lambert in north west Western Australia.
The iron ore is limonitic, containing 57% Fe and 9% combined water. Pelletizing plant
feed is dried to less than 1% moisture in rotary kiln dryers. The dried ore is ground
in ball mills to a Blaine surface area of approximately 280m2/kg. The green balls,
sized between 12mm and 18mm are indurated at the maximum temperature of 1350X1.
Pellet product, mainly 16mm to 9mm in size, is exported to Japan [4].
183
Table C.2-1 Pelletising Plants in Australia.
Savage Robe
River Hamersley Whyalla River
Owners:
Savage Hamersley The Broken Hill Cliffs
River Iron Pty Ltd Proprietary Co Robe
Mines Limited (B.H.P.) River
Feed Materials:
Nat. Magnetite Natural Natural Goethite
concentrate hematite hematite limonite
Type of Pelletizing Process:

Shaft Straight Grate- Straight
furnace grate kiln grate
Design Capacity, Million tonnes per annum (Mtpa):

2.25 2.00 1.50 4.20
Current Capacity, (Mtpa):

2.50 3.00 2.00 5.00
184
D. D EV ELO PM EN T IN SIN TERIN G OF NON-FERROUS ORES
IN A U STR A LIA
D .l Overview
The modern art and science of sintering as applied to ore treatment may perhaps be
said to have started with the invention of the Huntington-Heberlien (H&H) batch
process in 1896. This process had a great influence on the efficiency of lead smelters
in many parts of the world [13].
The first sintering plant installed in Australia at Port Pirie was the H&H process in
1901. The smelters were owned and operated by the Broken Hill Proprietary Company
Ltd. (B.H.P.), and the H&H reactors were used to sinter non-ferrous ore (lead) which
had been desulphurised in mechanical roasting furnaces.
The next significant development in sintering was the continuous moving-grate

sintering machine, invented by two Americans - Dwight and Lloyd. The D&L process
was to become universally successful.
Ten machines, five in "A” preroasting section and five in "B” sintering section were
established at Port Pirie in 1912. The H&H sintering machines were gradually phased
out, and the last convertor was emptied in May 1926.
In 1915, when B.H.P virtually passed from "silver to iron and steelworks", the Broken
Hill Associated Smelters Proprietary Limited (B.H.A.S.) assumed control of operations
at Port Pirie. In the years ahead, the B.H.A.S. was to be a world leader in lead sintering
practice with their outstanding developments based on the original D&L sintering
process [9].
In 1918, an Experimental and Research Department was established at Port Pirie.

During the years 1920-1924, the ground work was developed for a comprehensive
study of the down-draught D&L process.
In 1921, following a fire, which completely destroyed the original installation, the
sintering section was rebuilt as a two-stage plant, seven "A" section and six "B" section.
These machines were all standard size, 1.07m wide by 6.71m long over the windboxes.
185
D.2 Fundamental Studies and Developments in Sintering
. Basic sintering tests were carried out at Port Pirie, in a fixed pallet 381mm wide by
457mm long, 102mm to 203mm deep. This work later was carried out on the large
scale plant machines, 1.07m wide by 6.71m long over the windboxes.
. Raw material characteristics, operating factors, and most aspects of the D&L process
were studied.
. Numerous changes to thè methods and equipment used for igniting the charge were
made.
. Changes were made in the design of the pallet grates and in the shape of the windboxes.
. The most significant innovation was the fines bucket elevator to return any fine
material, which had fallen through the grates, back to the feed hopper of the machine.
. Development was made on the preparation of the sinter feed charge through
improvement in the condition of the ingredients themselves, and through their more
accurate proportioning in the mixture.
.The chief contribution was made by the new system of storage bins erected in 1926,
which allowed a much more accurate charge to be prepared. A device was fitted to
each feeder conveyor discharge to give a continuous cascade of material onto the
collector belt instead of an intermittent drop.
. A constant weight delivery mechanism was fitted to the bin conveyors, and the whole
charge was regularly checked for accuracy.
In 1927, a major advance was made when the practice was initiated of returning
crushed preroast as an addition to the raw charge on the preroasting machines. Over
some years of trial, the proportion of added preroast was gradually increased to
5096-60% of the new charge (equivalent to 36% to 43% of the total charge). As a result,
the quality of the product was markedly improved and the production rate was greatly
increased.
In 1931, a detailed study was done of the behaviour of charge mixtures ranging from
100% raw material to 100% preroast. It was shown from this work that there was an
optimum charge composition of approximately 30% raw charge and 70% preroast,
where the rate of sulphur elimination reached a maximum.
In 1935, an investigation into factors controlling the permeability of the charge bed
was begun, and this was continued in 1937. In this work, the effects of moisture
content of the charge, proportion and sizing of returned preroast, proportion of
granulated slag, etc., upon permeability prior to ignition were studied.
186
It was shown that the decrease in permeability after ignition was largely due to a
change in the lower portion of the bed and not in the roasting zone. This change
apparently was a marked compression of the bed following condensation of moisture
from the upper layers and collapse of the nodular structure. The solution to this
problem was the provision of a grate layer of coarse crushed return preroast.
D.3 Development of the Era of Large Sintering Machines.
Dr. G.K. Williams [9] successfully introduced the principle of larger, and therefore
fewer, units into a sintering process. By this principle, a plant would consist ideally
of two machines only, one for preroasting and the other for final sintering.
In 1937, an experimental machine 3.05m wide was built to guide the way to a full
scale plant, and the feed distribution across the 3.05m width, B.H.A.S designed machine
was successfully operated.
. The new sintering plant, which was completed and commissioned in 1940, consisted
of two similar down-draught machines, each 3.05m wide by 16.76m long over the
windboxes and operated in series, that is, one "A" and one "B" machine, entirely of
B.H.A.S design. The new plant included two 7.62m diameter mixing tables: one table
fed the "A” machine, and the other fed the "B" machine.
. Crushing and cooling of preroast. The machine itself was modified to crush the
largely red hot preroast to -19mm size.
. In 1940, there was an early feeling that "for best blast furnace operation, as far as
possible all fluxes and additives should be incorporated in the sinter" [9].
. Later, coke economy was greatly improved by the inclusion of about 1% of fine coke
in the charge.
. There were refinements of technique with each stage. With increasing instrumen
tation, key operators were reduced to a single attendant, controlling the whole plant
operation from one central control room. Control of the process from the central
point was facilitated by the use of a "clear call" intercommunication system linking
all manned operating units.
. By planning and organization of the four factors: materials, methods, men and
maintenance, a better system of process control has resulted in improved production
and greater metallurgical efficiency [14].
. It is interesting to note that the B.H.A.S-Port Pirie works aimed at making the plant
"the best in the world, doing the best work, justly rewarding every effort towards
economy and improved efficiency, and providing a healthy, decently paid occupation
187
for its employees". Outstanding and unique developments in lead metallurgy have
been achieved, giving B.H.A.S. world standing in its own sphere, and a reputation for
technical leadership and business integrity [9].
D.4 Development of the Up-draught Sintering Technology.
The basic purposes in developing the up-draught sintering process at Port Pirie,
Australia by B.H.A.S., were the production of urgently needed sulphuric acid from
sintering gases for South Australia's fertilizer industry, and to accomplish full removal
of the lead. This forthright approach was initiated in 1948 by the S.A. Government
in order to remove the S 0 2 gas from the local atmosphere [9].
Dr. Williams' impressive observations of fluid bed heat exchanger tests and results,
and a conference on 13th September 1948, led to a decision to build a pilot plant. As
a result, the continuous ignition method of down-draughting followed by up-draughting
sintering process was developed in the laboratory in December 1948.
The pilot plant used a 0.61m wide by 3.66m long machine, which was locally built in
1949, and indicated that there were possibly substantial advantages in an up-draught
sintering process, to produce S 0 2 gas of suitable strength (6.5% to 7.0% at the blowers),
and particularly complete freedom from lead loss [9] [12].
The next step was to plan a large scale commercial machine 3.05m wide and 30.48m
long. The B.H.A.S. up-draught sintering plant commenced production in 1955, after
five years of pilot plant development, design and construction at Port Pirie.
In the development of the up-draught process, it became evident that successful

sintering depended on the efficiency of the ignition layer technique.
However, to ensure the continuous ignition layer of approximately 25mm depth, it

was necessary to screen the feed passing to the ignition hopper. The total bed depth
was 267mm. Return roll crushing was used to reduce 89mm to 6mm [14].
In the early years, proven equipment had to be imported from already established
industries but, as time went on, Port Pirie began to develop its own thoughts on
sintering.
It is difficult to give credit to all those taking part in the long and fascinating story
of sintering. From the foregoing story, it is clear that the process of sintering has
had a unique and interesting history in Port Pirie.
However, it is a high distinction to mention and remember that G. Rigg, W. Robertson,

F. Whitworth, G.K. Williams, A.R. Hogg, R.J. Hopkins, H.H. Wight, A.J. Keast, L.B.
Haney, W.R. Burrow, W. McA. Manson, M.L. Baillieu, A. Langstaff, C.H. Marsh, F.C.
188
Adams and K.L. Riddley, supported and provided valuable contributions for the
development in sintering in the era of large machines (1918 - 1940), and of the new
up-draught sintering technology (1948 - 1955).
The up-draught sintering process was developed by B.H.A.S. at Port Pirie, Australia,
independent from Lurgi Gesellschaft fur Chemie und Huttenwesen at Stolberg,
Germany [8] [15]. Both B.H.A.S. and Lurgi set world levels of technical leadership in
sintering. However, while Lurgi currently maintains its leadership in the world,
B.H.A.S/s pioneering effects, like most of our Australian inventions and discoveries,
remaind only as a remarkable record in the history of technology and science.
Fig D-l outlines the feed and ignition arrangements and the plough device for fluffing
the bed, on the D&L "A” machine [9].
Fig D-2 describes the downdraught-updraught method as applied on a full scale

machine, which was designed and built at Port Pirie, South Australia. [9].
Figs D-3 and D-4, indicate a typical flow sheet for lead-zinc updraught sintering, and
S 0 2 gas recirculation to produce sulphuric acid [14].
189
Fig D -l Feed and ignition arrangements on the Dwight-Lloyd machine[9].
IGNITION LAYER. MAINI ROAST LAYER
Fig D-2 Feed and ingition arrangements on the updraught sintering machine designed at
Port Pirie, South Australia [9].
190
Ra W
MATERIALS
STORE DROSS FROM IMPERIAL
SMELTING FURNACE
'AND ZINC REFINERY
WATER
7 - 8 % S02 GAS MIXING CONDITIONING
TO ACID PLANT DRUM DRUM
CADMIUM PLANT
CADMIUM
METAL ' Ÿv'XJr—
ELECTROSTATIC
PRECIPITATOR
SLUDGES TO i \ 1n
SINTER
CRUSHING UPDRAUGHT SINTER MACHINE
IMPERIAL SMELTING
CIRCUIT
á9
FURNACE • nD
^ mm _ _ ___________ '
<12 mr
SLUDGES FROM
VENTILATION.
FROM IMPERIAL % BLUE POWDER,
SMELTING FURNACE ACID PLANT AND
CHARGE SCREENS EFFLUENT TREATMENT
ob J WATER
COOLING
10% > 6 mm DRUM
Fig. D-3 T ypical Sintering P lant Flow Sheet fo r lead-zinc [14].
TO SULPHURIC ACID PLANT
Fig. D-4 D raughting arrangem ents and gas recirculation for

lead-zinc u p d rau g h t sintering [14].
191
E. D EVELO PM EN T OF SIN TERIN G OF FERROUS O RES IN
A U STR A LIA
B.H.P S T E E L -S L A B & P LA T E PRO D U CTS DIVISION,
PO RT KEM BLA.
Table E-l General Data for Port Kembla Nol Sintering Machine [18].
Width of machine: 2.5m
Suction length: 40.08m
Size, suction area: 100.0m2 (grate area)
Feed operation:
Primary Conical trommel
Secondary Drum supported and driven by
pneumatic tyres
Feed to machine: Feed roll
Type of seal: Steel mounted pallet supported
on rails with a greased metal
to metal seal.
Method of discharge: Lowering sprocket wheels.
Windbox arrangement: Single windbox - single flue.
Hot screens: Schenk design.
Cooler: Elevated suction area 126 m2
inclined 13° approximately.
Cooler size: 2.0m wide x 72m centres.
Method of cooling: 3 induced draught fans.
Main fans: "Howden" straight bladed fan
8015m3/imn at 10 kPa
and 150^.
Dust collection "Lurgi" dust traps and
from sinter machine impact dust collector.
flue gases:
Method of feed Table feeders, with Avery
proportioning: check belt weigher.
General dedusting: Lurgi type electrostatic
dry dust collector.
192
Table E-2 Guaranteed Production Figures
Feed mix % % %
Grey Yampi ore 89 59 39
B.F. flue dust 11 11 11
Whyalla ore nil 30 50
Total 100 100 100
Production rate:
tonnes/24hrs 1422 2448 3048 (maximum)
Circulating loads (return sinter fines) between 40% and 50%, and 5.5% to 7.0 % coke
breeze rod-milled to pass 3.2mm, on a total dry mix basis were used.
Table E-3 No 1 Sintering Machine Typical Raw Materials Data [19].
Cumulative % Passing % Assay

1.7mm 152micron 76micron Fe S i0 2 a i 2o 3
Grey Yampi 90 70 38 62 6.0 5.0

Flue dust 100 75 65 33 7.8 3.1
Whyalla (M) 38 15 8 60 4.5 3.5
Whyalla (P) 80 62 48 64 2.5 2.3
(M) From Monarch Quarries in the Middleback Ranges.

(P) From Prince-Baron Quarries in the Middleback Ranges.
193
Table E-4 General Data for Port Kembla No 2 Sintering Machine [18].
Width of machine: 4.0m

Suction length: 44.0m
Size, grate area: 176.m2 (suction area)
Maximum capacity: 5283 tonnes/24hrs.
Feed preparation:
Primary Conical trommel
Secondary Drum supported and driven by
pneumatic tyres.
Feed to machine: Feed roll.
Type of seal: Grease seal, spring loaded.
Method of discharge: Lowering sprocket wheels.
Windbox arrangement: Divided windboxes-two flues.
Hot screening: Schenk screens.
Cooler: Elevated, area 223.2m2,
inclined 15° approximately.
Cooler size: 3.5m wide by 75m centres.
Method of cooling: 3 induced draught fans.
Main fans: Two "Howdens", aerofoil
section, blading approximately
8015 m3/min at 10 kPa.
Dust collection
from sinter machine "Davidson" skimming type collector.
flue gases:
Method of feed Transweigh constant weight feeders
proportioning: and Lurgi table feeders.
General dedusting: Lurgi type electrostatic dry dust collector.
The No 2 Sintering Machine was enlarged to 187.5m2 in 1971
194
Table E-4 Ore Handling and Screening Plant (cont’d).
Raw ore handling: 3 x 6 1 0 tonne bins and 400 t/h r

apron feeders, 1.07m wide x 813
t/h r conveyor belt.
Raw ore screening 2 x 1.83m wide x 4.88m long
station: Gyrex scalping screens, with
76.2mm opening screen
mechanically vibrated.
The -76.2mm ore was passed on to
two sets of tandem screens,
which classified it into
-76.2 +19.1mm, -19.1 +6.4 for
BF or OH, and -6.4mm, for sinter
plant fine ore bins.
Fine ore storage 7 bins with 1016 tonnes capacity

bins: each.
Miscellaneous Materials Handling and Coke Crushing
Dumping station: 2 bins for flue dust and 2 for coke or limestone.
2 Gyrex 1.83m x 4.88m single deck, heavy duty
Screening station: vibrating screens, with a circle throw diameter
of 12.7mm and operating at 625 rpm.
4 x 102 tonne bins. One vibrating Gyrex with
3.2mm opening screen. The +3.2mm coke was
Coke and limestone crushed to pass 3.2mm. The limestone was
crushing station: crushed to 3% +3.2mm.
One 2.29m dia x 4.27m long Newells rod mill;
and two 2.90m x 4.27m Marcy rod mills.
Sintering Section
Mixing and rolling Mixing drums, 152 tonne capacity

of ore and coke: hoppers.
Sintering Machines: A hearth layer of 25.4mm deep,
consisting of -28.6 +19.1mm sinter.
195
Table E-5 Sintering Plant Equipment Description [20].
Raw Material Bins
Number and Size 12 x 350 m3 12x350m3, 1x300 m3,
1 x 50 m3
Mixing/Rolling Drum
Dimensions (L x <$: 13.25 m x 3.8 m 25 m x 4.4 m
Rotational Speed: 5.4 rpm 5.5 rpm
Residence Time: 4.5 min 4.5 min
Inclination: 1.7 degrees 4.5 degrees
Design Capacity: 515 tph 1,185 tph
Motor Power: 1 x 400 kW 2 x 400 kW
Water Addition: Inlet end only Inlet end only
Sinter Strand
Type: Lurgi Lurgi
Dimensions (W x L): 4 m x 50 m 5 m x 84m
Design Capacity: 27.5 t/m 2/d 40 t/m 2/d
Typical Productivity: - 27 t/m 2/d
Suction Area: 200 m2 420 m2
Suction: 10 kPa 16 kPa
Number of Pallets: - 139
Pallet W x L: 4m x 1.5 m 5m x 1.5 m
Windlegs 2 x 13 2 x 23 Windboxes
Windmain: 2 x 3 m<j) 2 x 5 m<j>
Bed Height (typical): 400 - 450 mm 480 - 530 mm
Feed Roll Diameter: 1.3 m 2 m (outside diam)
Feed Plate: Adjustable Adjustable
Grate Bars: 50 mm wide, Predominantly 40 mm,
high chrome high chrome
Ingition Hood
Dimensions (W x L): 4mx 5m 5mx 3m
Combustion Air: n/a Hot combustion air
ex cooler.
Burner: Urquart, three Direct impingment
per side (Kawasaki line burner)
Preheating Hood
Dimensions (W x L): n/a 5mx4m
Air: n/a Ex cooler.
196
Table E-5 Sintering Plant Equipment Description (cont’d).
Primary Crusher
Type: Single Spike Roll Single Spike Roll
Cooler
Type: Straight Line Grate Circular Grate
Effective Cooling
Area (W x L): 210m2 (3.5m x 60m) 380 m2
Cooler Fan
Characteristics: n/a (3 x) 12,000 m3/imn»
at SO'C and 310 mm
WG delta P
Heat Recovery: n/a Hot air for pre-drying
and ignition combustic
air.
Cold Screens
Type: Schenck Schenck
Size: Primary 2.2 m x 4.8 m (3 x) 1.2 m x 5.5 m
(W x L) Secondary 2.0 m x 3.0 m (3 x) 1.5 m x 6.5 m
Tertiary 2.2 m x 5.2 m (3 x) 3.5 m x 6.5 m
Design Capacity:
Primary 320 tph 508 tph
Secondary 220 tph 415 tph
Tertiary 160 tph 362 tph
Screen Aperture:
Primary 20 mm Tear Drop 18 mm Tear Drop
Secondary 6 mm Tear Drop 10 mm Tear Drop
Tertiary 12 mm Tear Drop 6 mm Tear Drop
Motor Power:
Primary 22.5 kW 1 x 15 kW
Secondary 37.5 kW 1 x 15 kW
Tertiary 37.5 kW 1 x 55 kW
197
Table E-5 Sintering Plant Equipment Description (cont’d)
Dust Collection
a) Waste Gas System: Mechanical Dedusting One Lurgi Dry
system consisting of Electrostatic
Whirlwall and cyclones precipitator/gas
main with 3 operating
fields
Design Exit Dust
concentration 200 mg/Nm 3 200 mg/Nm 3
Waste Gas Fan (2 X) (2 X) 21,000 m3/min at 140^

and 1650 mm WG delta P
b) Room Dedusting Two Lurgi dry One Lurgi dry

System: electrostatic electrostatic
precipitators precipitator with 2
with 2 operating operating fields
fields
Design Exit Dust 200 mg/Nm 3

Concentration 100 mg/Nm 3
18,900 m3/min at 75 X and
Room Dedusting Fan n/a 220 mmWG delta P
Coke Crushing 1 & 2 Mills

Equipment Newells Rod Mills (X4)
Type: n/a 4.3 m x 2.44 m
Size (L x <$: n/a 30 tph (coke)
Capacity: n/a
3 & 4 Mills
Equipment Marcy Rod Mills
Type: n/a 4.27 m x 3.05 m
Size (L x 4): n/a 40 tph (coke).
Capacity: n/a
198
F. B.H.P-RO D & BAR PRO D U CTS DIVISION, N EW CA STLE
SIN TER PLA N T
F .l Overview
The first ferrous sintering carried out in Australia was at the former Broken Hill
Proprietary Co. Ltd., Newcastle, in the early years of World War II. This was a pilot
plant for the sintering of blast furnace flue dust. The plant operated until a new
1.83m wide machine was commissioned in 1942. This plant had a feed proportioning
system, which enabled good quality control [ 10].
The current Lurgi sintering machine with a grate area of 208m2, generally similar
design to the Port Kembla No 2 machine, was commissioned in the early 1960's. There
have been countless improvements and additions over the years, many small in nature
individually, but also generally low in capital cost. However, there have been some
large projects, most notable of which have been as follows:
. The rebuild of the Sinter Cold Screening Section, 1968-1969,
. The installation of the Blending and Storage Yard complex, 1969-1970,
. The No 2 Screening and Crushing Plant, 1971-1972,
. High basicity sintering practice from 1975,
. Deeper bed practice from 1982,
. Installation of tertiary ore crushing facilities and the beginning of a major rebuild
of the dedusting plant in 1986,
. The beginning of a major program of conveyor chute rebuild in 1987,
. The installation of a miscellaneous materials dumping and proportioning system
in the Blending Plant in 1989.
The plant now can operate at overall production levels higher than any time in the
past, produces sinter with superior quality parameters and uses approximately 30% of
the gas, 50% of the coke consumption and operates at much higher levels of availability
compared to the 1970's [22].
F.2 Sintering Process
The process involves mixing weighed amounts of blended fine ores, sinter fines, fluxes
(limestone, dolomite and serpentine), and fuels - coke breeze (at times crushed
anthracite coal is utilised to supplement coke supplies). This mixture is then laid on
the sintering strand and passed under the ignition furnace. The fuel in the mixture
is ignited and the flame, along with air, is pulled through the moving bed by suction
on the underside of the strand - the suction being created by the main fan.
The heat generated by the flame (at 1290°C) moving through the sintering bed causes
the formation of a slag which binds the fine ore particles together into large incan
descent sinter cake.
199
F.3 Basic Operations
The entire sinter making section is controlled from a central control room. As may
be seen in Fig 1-2, blended fine ore, limestone and coke breeze are discharged by
weigh feeders to a common conveyor. These are combined with return sinter fines
(approximately 30% hot return fines and 70% cold return fines), mixed in the primary
mixing drum and transported to a granulating drum-secondary mixer. Water is added
at both mixers, with controlled addition at the secondary mixer to meet optimal
granulation requirements.
A 35mm depth hearth layer (-20mm + 6mm sinter) is fed onto the sintering machine,
and the granulated feed mix containing 5.8% water is fed onto the hearth layer bed
by a roll feeder fitted with gates and a levelling plate to accurately control the feed
rate and bed depth.
The bed is ignited by overhead burners in a coke ovens gas fired ignition furnace
operating at 1lOOX). Natural gas is utilised to enrich the coke-ovens gas and maintain
a constant calorific value.
Incandescent sinter is discharged from the end of the strand directly onto a spiked
roll crusher for size reduction to approximatly -200mm and then over a hot screen
which separates the hot return fines, nominally -6mm size. The oversize sinter is
discharged to a straight line forced draught cooler and then passes to a series of cold
screens which separate the -6mm cold return fines and hearth layer from the product
sinter. This is conveyed to the sintering material bins, and the oversize is transferred
to blast furnace storage bins. [22].
200
F.4 Major Sintering Plant Equipment
- Sintering Machine: Dwight-Lloyd downdraught type of

Lurgi design.
- Raw Material Bins: 7 x 1000 tonnes iron ore bins.
3 x 600 tonnes limestone bins.
2 x 350 tonnes coke breeze bins.
All bins discharge are by rotating feed tables.
- Primary Mixer.
- Secondary Mixer - granulating mixer.
- Sinter Cooler - Lurgi straight line cooling conveyor.
F.5 Typical Sintering Machine Feed and Operating Parameters

Machine feed %
Jron ore fines 84%
Blended iron ore 6 0 < ~ - -Recycled materials 12%
"Fluxes 4%
Limestone 8 Total 100%
Coke (3.8 - 4) 4
Return fines (22 - 30) 28
Total 100
Operating Parameters
Grate area 208 m2

Bed height 450 mm
Strand speed 2.6 m/min
Machine suction 11.0 kPa maximum
Ignition furnace temp. hoot:
Sintering temperature 1290 T maximum
Dry coke consumption 55 kg/t-S
Power consumption 28.6 kwh/t-S
Productivity 32 t/m 2/day
201
F.6 Quality Control
Product sinter quality is designed to meet blast furnace specifications dictated by

current slag and burden practices.
The majority of sinter chemistry requirements are met by regulation of the materials
added to blended fines piles. Ca0 /Si 02 ra-tio is controlled at the sintering machine
by regulated additions of -3mm limestone.
Sinter strength is regulated by the FeO content (a magnetic permeability scheme)

controlling the coke addition to the machine raw mix.
Table F-l illustrates Newcastle Sintering Plant Operating data: feed mix, operation
and production parameters and quality of the product sinter [22].
Table F-2 indicates Newcastle No 3 and No 4 blast furnaces ferruginous burdens

[ 22].
202
Table F-l Newcastle Sintering Plant Operation Data
Feed Mix 1990 Jan. Feb. Mar. Apr. May.

Blended fines % 59.5 59.2 59.2 57.6 59.5
Limestone % 8.9 8.3 9.0 8.8 9.4
Return fines % 28.0 29.0 28.3 29.9 27.3
Coke breeze % 3.6 3.5 3.5 3.7 3.8
Total % 100.0 100.0 100.0 100.0 100.0
Operation
Bed height mm 440 440 440 440 440
Strand speed m/min 2.6 2.5 2.5 2.6 2.6
Fan suction kPa 10.1 9.9 9.8 9.4 8.0
Return fines kg/t-S 486 526 504 541 476
Energy Consumption
Dry coke kg/t-S 55 54 52 47 57
C.O. gas Nm 3/t-S 5.2 5.3 5.1 5.0 5.0
C.O. gas MJ/t-S 104.9 102.4 95.0 82.5 91.6
Production
Production,tonnes 68021 139555 163357 166972 159780
Production rate, tph 264 253 255 259 261
Productivity, t/mfyd 30.5 29.2 29.0 29.8 30.1
Availability % 97.8 97.2 97.9 96.8 97.8
Overall avail. % 87.7 79.0 87.2 91.8 85.0
Sinter quality
Fe % 56.1 55.9 55.9 55.7 55.7
FeO % 5.2 5.4 5.7 5.7 5.1
SiOo % 6.1 6.1 6.2 6.3 6.3
AI5O3 % 1.9 2.0 2.0 2.0 2.0
Cab % 9.9 9.8 10.0 10.1 10.0
MgO % 1.6 1.6 1.5 1.6 1.6
Mn % 0.5 0.6 0.5 0.5 0.6
CaO /Si0 2 1.6 1.6 1.6 1.6 1.6
Sizing
+ 31.5 mm % 16.0 16.5 15.9 16.1 16.7
+ 8.0 mm % 79.7 77.3 80.1 79.2 81.3
+ 4.0 mm % 94.4 92.2 94.9 93.8 94.7
Arith mean size mm 21.3 21.4 21.1 21.4 21.8
Abrasion + 8.0 mm % 67.1 65.2 67.1 67.1 67.4
R.D.I % 33.0 33.8 32.2 32.8 32.5
Reducibility % 62.0 61.8 62.9 64.1 64.0
203
F.7 Blast Furnace Ferruginous Burden
Table F-2 Newcastle Blast Furnace Ferruginous Burden.
1990 Jan. Feb. Mar. Apr. May,

No4 Blast Furnace,
Prod’vity.IV.t/mVd 2.0 1.6 1.7 1.9 1.9
Coke rate, kg/tHM 445 454 451 474 434
Natural gas, kg/tHM 47 44 44 29 51
Fuel rate, kg/tHM 491 498 495 503 484
Ferruginous Burden
Sinter, % 77.2 77.4 77.4 77.5 77.5
Lump ore, % 22.8 22.6 22.6 22.5 22.5
Pellets, % 0.0 0.0 0.0 0.0 0.0
No3 Blast Furnace

Prod’vity,IV t/m 3/d 1.9 1.8 1.9 2.0 1.9
Coke rate, kg/tHM 463 470 468 463 463
Natural gas, kg/tHM 53 38 39 52 50
Fuel rate, kg/tHM 517 508 507 515 514
Ferruginous Burden
Sinter, % 77.1 77.6 77.1 77.4 77.4
Lump ore, % 16.0 15.6 15.9 14.6 12.8
Pellets, % 6.9 6.8 7.0 8.0 9.8
Code : C.O. = Coke Ovens

S = sinter
avail. = availability
IV. = inner volume
Prod’vity, = productivity
Ign. = ignition.
204
G. A U STRA LIA N IRON & S T E E L P TY. LTD. - KWINANA
STEELW O R K S
G .l Overview
Kwinana sintering plant operations commenced on 16th September, 1968, continuing

until ironmaking operations ceased during the economic downturn in 1982. This
operation was established on acid sinters. Coffin Bay limesand was introduced in
stages to 16% of the ore by 17th November, 1968, to establish parity with East Coast
sintering experience of 1.2 CaO/Si0 2 ratio. It was also the level specified for the
plant performance guarantee by Lurgi commissioning officers. Bed depth was set at
305 mm, fan suction limited to 11.2 kPa, with an ignition hood temperature of 1100
°C.
In 1970, the Kwinana sintering plant, one strand with a grate area of 100 m2 and
annual production of one million tonnes, was operating with a bed height of 381 mm
and a suction level of 15 kPa, using 100% Dowd’s Hill fines, 50/50% Japanese limestone
and Wanneroo limesand; 40% return fines and 3.7% coke rate, at a nominal basicity
of 1.6 CaO /Si0 2 ratio.
Reasons for the high sintering rate and the high quality of the Kwinana product
sinter were investigated (using raw material samples taken in December 1970) by L
Munive and his colleagues [30]. The basic hypothesis attributed the high quality sinter
and productivity of the Kwinana sintering plant to the bed depth and high suction:
hence, the thermal conditions prevailing during the sintering process [30].
The hypothesis postulated above is the basis of this thesis. In March 1971, permission
was sought to investigate the effects of bed depth, suction and basicity on sintering.
This was subsequently granted by the Australian Iron & Steel Pty Ltd.- Port Kembla
ironmaking management.
205
H. FLUX
H .l Origin ( Continued )
Fossiliferous, marine sedimentation in oceans and fresh bodies of water, consisting

of shells and skeletons of plants and animals, were gradually accumulated through
deposition, layer upon layer to form in some instances massive beds of limestone.
Some of this sediment was deposited by natural chemical reactions, such as the
infinitesimal dissolution of these calcium carbonate fossils through the solvent action
of carbon dioxide, forming calcium bicarbonate, which was subsequently precipitated
in carbonate form.
The chemical reactions that occur, are:
C 0 2 (g) + H20 (1) ~ H 2C 0 3 (1)

CaC0 3 (s) + HC0 3 «-* Ca(HC0 3)2 (1)
Ca(HC0 3)2 + heat «-> CaC0 3 + C 0 2 + H20
In still another category are carbonate bodies, such as reefs, that have been constructed
principally by skeletal secretions of sedimentary organisms. Huge coral reefs are
gradually accumulated in this manner over thousands and millions of years; they have
formed mountains in the interior of continents. Pressure and heat have supplemented
chemical precipitation in consolidating the very small carbonate particles into these
imposing compact masses [43] [44]. Present-day formation of the rock by corals is
exemplified by the Great Barrier Reef of Australia [42].
In a similar manner, inland streams and rivers are carriers of soluble calcium
bicarbonate, through the leaching effect of rainwater percolating through soils in
watersheds and the gradual dissolution of carbonate rocks as the streams cascade over
them. The co-reactant, carbon dioxide, must be present to effect this phenomenon.
H.2 Geochemical Evolution
Carbonate rocks represent a natural response to the geochemical evolution of the Earth
and to major steps in the course of biological evolution. A first milestone in the
carbonate history is the appearance of the first organisms that began to use carbon
dioxide and to liberate oxygen as a by-product of sugar synthesis. This removal of
carbon dioxide from solution in seawater must have increased the alkalinity and led
to the precipitation of calcium carbonate. The first carbonate fossils, stromatolitic
algal structures, are found all over the world in Precambrian rock formations, some
of which are more than 2,700 million years old.
206
During Paleozoic and early Mesozoic time, 570 to about 200 million years ago, limestones
and dolomites were deposited in a belt extending from North America through Western
Europe and the Soviet Union to Australia. These carbonates were largely the product
of sedimentation in shallow seas. In the late Mesozoic, 65 to 150 million years ago,
pelagic organisms burst out in an immense biological explosion all over the Earth.
This great biological event changed the Earth’s carbonate economy and shifted a large
part of the carbonate deposition from the continental shelves to the oceanic depths
[44].
H.3 Chemical Properties
Carbonates are not chemically reactive, their predominant chemical property is that
they can be thermally decomposed (calcined) into lime. The most important chemical
properties are:
Thermal Decomposition: This is the most important chemical property of limestone.

All carbonate rocks dissociate at high temperature, forming oxides of calcium and
magnesium through the expulsion of C 0 2 gas.
This is probably the most basic and apparently the simplest of all chemical reactions.
However, there are many complexities attendant to this reaction. Stated chemically
with molecular weights, this reversible reaction for both high calcium (h.c.) and
dolomitic quicklime (dol.qlime) is:
CaC0 3 (h.c limestone) + heat «-» CaO (h.c.qlime) + C02

100 (s) 56 (s) 44(g)
CaC03. MgC0 3 + heat <-» CaO.MgO + 2C0 2

100 84(s) 56 40(s) 88 (g)
(dol. limestone) (dol. qlime)
There are three essential factors in kinetics of limestone decomposition:

1) the stone must be heated to the dissociation temperature of the carbonates,
2 ) the minimum dissociation temperature must be maintained for a predetermined
length of time, and
3 ) the carbon dioxide released in the process must be rapidly removed from the
kiln.
Temperatures of 900 to 100 °C at atmospheric pressure of 101.3 kPa (760 mmHg), are
required to dissociate calcium carbonate. Dolomitic limestone dissociates at 725 °C,
and magnesium carbonate between 402 to 480 *€.
The range in temperatures varies due to the different impurities and to the vast
difference in geological origin and physical structure.
207
Most complicated of all is the fact that each stone has its own peculiarities, the exact
limitations and optimum conditions of which can only be ascertained by trials and
assessment of results [43].
In practice, to ensure complete dissociation of the limestone particles, temperatures

higher than the required minimum are used for an optimum heating period.
H.4 Physical Properties
Most of the physical properties vary, often considerably, depending on the type of
carbonate.
Colour : pure forms of the minerals, calcite and magnesite, are often white, a pale
creamy-yellow, tan or grayish colour, and other colours caused by the type of impurities.
Texture : x-ray diffraction has revealed that all limestone is crystalline. However,
the size, uniformity and arrangement of its crystalline structures or mineral grains
vary greatly, from a dense, hard, compact and massive fine-grained crystalline variety
(lithographic) limestone to oolitic limestone. This last, granulated oolitic limestone,
is called limesand in Australia.
Porosity : there is considerable variation in the degree of porosity. Total porosity

ranges between 0.3 and 12% of the stone. Porosity of marble is the lowest, 0.1%.
Bulk Density : because of the difference in porosity, the bulk density of limestone
ranges between 2002 and 2803 kg/m3. Stone that contains absorbed moisture weighs
slightly more than its dry counterpart.
Specific Gravity : most commercial limestone has a specific gravity range in values
2.65 to 2.75 for high calcium, and 2.75 to 2.90 for dolomitic limestone.
H.5 Classification of Limestone
There are more than forty varieties of limestone in common use, with different
nomenclatures and definitions. The main differences are in origin, impurity content
and mineralogy. The most commercial types of limestone are:
. Marble is a metamorphic, highly crystalline carbonate rock which may contain calcitic
and dolomitic crystals. It may be extremely pure or rather impure. It occurs in
virtually every colour, in varying mottled effects, and is the most beautiful form of
limestone. Because of its unique texture it can be cut more precisely and polished to
a smoother surface than any stone. It is usually very hard : some types approach even
granite in hardness.
208
. Metallurgical Grade Limestone is an almost pure form of limestone used as a flux
or purifier in metallurgical purposes. It can be a high calcium, magnesium, or dolomitic
type, providing the stone is low in impurities and contains at least 95% total carbonate.
H.6 Limestone Uses
Total tonnage of crushed limestone use all around the world, is prodigious and difficult
to comprehend. The major use is in construction aggregate (concrete and roadstone),
cement industries, flux and agriculture.
The second largest use of limestone is in chemical and metallurgical industries. The
prime function of limestone in metallurgy is for fluxing, where the crushed and
carefully classified limestone is incorporated into furnaces, particulary in the sintering
process, blended with other raw materials and coke breeze.
Impurities in limestone have a most pronounced effect on fluxing efficiency. Hence,

it is required that limestone be as pure as possible within economic limitations.
209
Allbook Bindery
91 Ryedale Road
West Ryde 2114
Phone: 807 6026

Research Online Research Online

Uploaded by

Copyright:

Available Formats

Research Online Research Online

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Research Online Research Online

Uploaded by

Copyright:

Available Formats

University of Wollongong

University of Wollongong Thesis Collection University of Wollongong Thesis Collections

The effects of bed depth, basicity and suction on sintering technology

Follow this and additional works at: https://ro.uow.edu.au/theses

LIBORIO MÜNIVE, B.E. (Chem. Eng.)

A Thesis submitted in fulfilm ent of

M ASTER OF ENGINEERING (HONOURS)

THE UNIVERSITY OF WOLLONGONG

I would like to acknowledge the support, encouragement and guidance of my supervisor

Appreciation is expressed to the Head of Department of Materials Engineering,

I would also like especially to thank the following:

. Mr M. Burns - Assistant General Manager Ironmaking; Messrs D. Lathlean, A. Talbot,

An improvement in productivity and sinter quality, and a marked changes in the

. Fuel consumption was significantly reduced, and sintering temperatures

a) Sintering is a highly complex process, so a broad introductory knowledge is

The properties of sinter are determined by two sets of factors:

3 0009 02982 2728

. Ore types and their genetic categories.

. Ore deposits, their characteristics and modes of occurrence; e.g.,

The systematic interpretation and correlation of these characteristics, added to the

Sinter is a major component of blast furnace burden materials, so it is important that

These observations were of significant assistance in understanding why limestone

i. Optimization of granulation of the green sintering feed, with the aim of

The basic raw materials investigated were:

1.3 Sintering Plant Trials.

Plant operating conditions were:

The results in summary, included:

Encouraged by these results, management approved extensive trials at the Sintering

1.4 Summary of Research Contributions

3. A marked decrease in FeO content, hence, a marked increase in the degree of

5. A significant reduction in fuel consumption was achieved, showing a potential

6. Sintering operating temperatures were reduced, with a resultant lowering of costs

RAW MATERIALS FUEL

Agglomeration products are extensively used in medicine, food technology, chemical

2.2 Methods of Agglomeration

The technology of the consolidation of fine particles is carried out at conditions

ii. Nodulizing by rotary kiln.

iv. Sintering process by

Details of pressure agglomeration, nodulizing and pelletizing processes are presented

2.3 The Sintering Process

The introduction of froth flotation technology required some form of pre-treatment

2) . Significant technological improvements in the sintering process were achieved in

Modern sintering plants located at steelworks provided :

- Economic gains, as sinter production cost was reduced to one-third to one-fourth

- An effective fully beneficiated blast furnace feed, and

v. A management system based on fundamentals of economic viability and

2.4 DEVELO PM ENT OF T H E SINTERING PRO CESS

The continuous down-draught Dwight-Lloyd (D&L) process was invented in U.S.A. in

2.4.2 The Huntington-Heberlien Process

2.4.3 The Greenawalt Process

Typical Greenawalt and Dwight-Lloyd Sintering Plants

Full-scale Greenawalt and D&L sintering plants at the Appleby-Frodingham Steelworks

. Six pans, each 7.9m x 3.0m

. The time between successive charges was 31 to 32min, of which about:

Table 2.4-1 Blast Furnace Ferruginous Burden in Sweden, 1920 to 1950.

Year Sinter burden %

A prerequisite for effective sintering is satisfactory gas permeability of the charge.

Development in sintering of non-ferrous ores in Australia is outlined in substantial

ORE BINS riOlOR

2.6 B.H.P S T E E L - SLA B & PLA TE PRO D U CTS DIVI

A variety of manganese oxide minerals occur as minor constituents of Western Aus