Engineering Practice

Troubleshooting Tube-Deterioration
Mechanisms in Direct-Fired Heaters
A practical step-by-step guide for reducing tube failures
Babak Maghbooli and
Hamidreza Najafi
Farayand Sabz Engineering Co.

F
orced outage of a direct-fired
heater is the worst nightmare
for any operator in a petro-
leum-refining or petrochemi-
cal plant. Since fired heaters are not
replaceable by spares, forced outage
in many cases means emergency
shutdown of the entire plant. In most FIGURE 1. Misunderstanding warning signs led to
cases, actual tube failure or even se- a dramatic internal coke build up, followed by high
tube metal temperatures, creep and finally tube
vere tube deterioration are responsi- rupture in one of the pass outlets of a visbreaker
ble for outage of a direct-fired heater. plant heater. Operators had unintentionally ignored
Tube-deterioration mechanisms are high tube-skin temperatures for one week
not only an economical concern, but
also a matter of safety for the plant plex metallurgical phenomena, and
and personnel. An unpredicted tube inspection engineers with degrees or
rupture may lead to disastrous explo- specialties in metallurgical engineer- FIGURE 2. In this example of good trouble-
sions and severe human casualties. ing should evaluate and explicate shooting, operators observed a failed tube hanger,
which triggered the alarm for possible roof tubes
In order to prevent such unwanted them; but after the inspection or eval- sagging and led to the heater shut down for main-
events and keep the fired heater tubes uation stage, the main root causes of tenance and hanger repair
functional as long as possible, many such problems are mostly simple and
petroleum refineries and petrochemi- can be explained to an operator with of internal coke buildup (fouling),
cal plants have organized a preven- the aid of fact sheets or flowcharts. such as back pressure in the speci-
tive program, which according to the If the operators are being kept out fied pass, maybe serious courses of
American Petroleum Institute [1], is of the loop, it is always possible for action would have been considered
referred to as a “reliability program.” personnel to misinterpret a sign of before raising the tube metal temper-
tube deterioration. And since heater ature to values that eventually caused
Reliability programs inspectors are not around all of the severe creep and tube rupture.
A good reliability program includes time, such misinterpretations could Explaining the root causes of the
various detailed inspection programs easily lead to a disastrous event. deterioration mechanisms to the op-
performed both during heater opera- erating team can prevent deteriora-
tion and overhaul or occasional main- Dangers of misunderstandings tion in the first place. It is important
tenance shutdown periods. Continu- As an example, consider Figure 1, that the inspection team arrange
ous assessments of the remaining which demonstrates a dramatic tube meetings in which operating person-
lifetime of tubes and the supervision rupture case that was the result of nel walk through the chain of events
of heater performance are also in- misunderstanding the warning signs and performance imbalances that
cluded in the program of many pe- and a lack of root-causes knowledge lead to a tube failure.
troleum refineries. Although reliability on the part of operating personnel. As another example, the situation
programs have proven successful In this particular case, the pass out- shown in Figure 2 was dealt with
during long periods of heater opera- let was near the internal header-box before the shield section tubes ex-
tion, many heaters are still victims refractory linings. Because of this, the perienced serious sagging, which
of tube-deterioration mechanisms. operators had unintentionally ignored could easily lead to tube rupture and
That is because heater operators or high tube-skin temperatures and as- possible explosion. As the operators
field engineers are often kept out of sumed that they were misreading the were aware of the possible threat of
the loop of a reliability program. refractory wall temperature by the in- roof tubes sagging (in case of fallen
It is true that deterioration mecha- frared pyrometer. If they were aware of or failed tube hangers), observing the
nisms are mostly categorized as com- other warning signs and root causes failed hanger at the floor of the heater

CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015 71

 TABLE 1. TYPICAL FIRED HEATER TUBE -DETERIORATION MECHANISMS AND ROOT CAUSES

Improper mechanical or steam air decoking procedures

Chemical composition of process fluid (critical species)
Heavy hydrocarbon molecules pyrolysis (coke buildup)

Refractory damages or improper refractory repair

Misplaced or failed tube support or hanger
High pressure drop across heater passes

Faulty roll procedures or workmanship

Heavy fuel-oil combustion problems

Improper two-phase-flow regime*

Upstream equipment malfunction
High content of impurities in fuel

Increased heater charge rate

Control systems malfunction

Undue force to close fittings

Low process fluid flowrate

Cyclic heat-load variations

Poor heater-casing design
Improper fuel pressure
Heat flux imbalances
Flame impingement

Mineral sediments

High firing rate

Thermal shock

Air Leakage
Root Cause
Problem
Problem category sub category
Hotspot --- • • • • • • • •
Internal fouling • • • • • • • • •
Fouling
External fouling • • • • • • •
Internal corrosion • • • • • •
Corrosion External dew-point • • • •
corrosion
Minor creep • • • • • •
Creep Bulging • • • • •
Bowing and sagging • • • • • • • • • •
Vibration --- • • • • • •
Erosion --- • • •
Thermal fatigue --- • • •
Mechanical --- • • • • •
deterioration
Note: * In this article, the term “improper two-phase-flow regime” refers to “slug flow” regime only and complex two-phase-flow phenomena caused by poor de-
sign, or improper operating conditions (problems like sudden flash evaporation or mist flow) are beyond the scope of the current study.

triggered an alarm, which led to the would ignore or misunderstand vital 2. Internal fouling
shutdown of the heater for mainte- signs of a possible tube-deteriora- 3. External fouling
nance and hanger repair. tion mechanism. 4. Internal corrosion
A good troubleshooting guide can In this article we propose a simple 5. Creep (general or minor creep)
demonstrate the fact that failed roof root-cause analysis and also a com- 6. Bulging
hangers are caused by overheating puterized troubleshooting algorithm 7. Bowing and sagging
in the convection section. In this vein, that directs the heater operator from 8. Vibration
by monitoring burner flame heights one root cause of a possible tube- 9. External dew point corrosion
and burner fuel pressures (control- deterioration mechanism to another, 10. Erosion
ling the amount of heat released), as in sequence. The algorithms indicate 11. Thermal fatigue
well as monitoring bridge wall tem- how each of the improper perfor- 12. Mechanical deterioration
perature, operators can prevent this mances for a given cause is to be
phenomenon in the first place. corrected, and continues until all Root-cause analysis
possible causes for tube-deteriora- It should be noted that before enter-
Troubleshooting tion mechanism can be corrected. ing the root-cause-analysis phase,
Even if a company offers proper one should not neglect the possibility
training courses for personnel, and Tube-deterioration mechanisms that the material of construction used
has an effective reliability program According to the valuable references for the heater tube can be a possible
and an effective schedule of visual listed at the end of this article [1, cause of some of the problems re-
inspections, these measures will not 3–4], as well as our field experience, ferred to above. Thus, readily available
reduce the importance of practi- typical fired-heater tube-deteriora- data for allowable stress and corro-
cal troubleshooting knowledge. An tion mechanisms that may lead to sion rates for various tube alloys, as
operator without proper knowledge tube failure can be classified into the a function of temperature, should be
and understanding of root causes following 12 main categories: made available from appropriate tube
and a troubleshooting sequence 1. Local or longitudinal hotspots manufacturers. These data should
72 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015
 A

FIGURE 3. Troubleshooting algorithm for

hotspots on the tubes Hotspots on tubes can be the result of internal or external problems.
First check external problems which are mainly caused by faulty burner
performances.

Do you experience flame impingement on tubes ?

Solve the problem according to the instructions given in burner Yes
troubleshooting guide found in Ref. 2. Caution: Since the flame of fuel gas is hardly visible, flame impingement
can be easily ignored by the operator. Inspect the firebox carefully and use
Yes Is your problem solved? sodium salts to make sparks and turn the flame pattern into a visible form.

No No

One of the main external reasons that leads to hotspots is heat-flux imbal-
Solve the problem according to instructions given in burner ances, mainly caused by non-asymmetric burner-firing patterns along the
troubleshooting guide found in Ref. 2. Yes
Yes firebox.
Is your problem solved? Do the burners frequently go out and stay ignored by the operator?

No
No

Adjust the fuel pressures to the same value on each burner, also No Are the fuel pressure gauges on
adjust the air and draft amounts accordingly. every burner showing the same
value?
Caution: If by doing so flame patterns were corrupted or
Yes misshaped, based on the problem refer to burner troubleshooting Yes
guide found in Ref. 2.

After solving the problem, adjust fuel pressures.

Is your problem solved?

No

Solve the problem according to instructions given In case of fuel oil firing, you may No If there is no external reason for
in burner troubleshooting guide found in Ref. 2. Yes suffer poor atomization the hotspots in the firebox, your
Yes problems causing misshaped tube may be suffering from an
Is your problem solved? flame patterns and firing rates internal fouling phenomena.
along the firebox.
No Do you have poor atomization
problems?
Go to Point B in Figure 4
Seek engineering consultancy.

End trouble shooting procedure.

be checked before taking any step actions, which are provided by com- A practical example
for understanding the root causes of prehensive algorithms described in Let’s assume that an operating
tube-deterioration mechanisms. the next section. team tries to prevent a future tube
In order to analyze each deterio- failure like the one shown in Figure
ration-mechanism category and its Troubleshooting algorithms 1. In order to achieve this goal, it is
subcategories in the most concise The comprehensive troubleshooting obvious that they need to trouble-
and efficient way, a root-cause- algorithm is based on the logic de- shoot the deterioration mechanisms
analysis table is proposed (see picted in Figures 3 to 6 (and contin- that are responsible for this event.
Table 1). ues in Figures 8–15 at www.chemeng First of all, an inspection team
By understanding the root causes online.com), and has been successfully should identify the nature of tube-
of tube-deterioration mechanisms, used to recognize what the tube-dete- deterioration mechanisms that have
one has passed through the primary rioration mechanism might be, indicate led to such a disaster. A preliminary
step of tube-damage troubleshoot- the cause of the problem and correct analysis would demonstrate that
ing. Now what matters most is the the heater defect, so as to achieve the severe tube creep accompanied by
proper sequence of troubleshooting proper tube operating lifetime. internal fouling (coke buildup) are to

CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015 73

 B FIGURE 4. Troubleshooting algorithm for internal fouling

Internal fouling may be caused by coke Low process-fluid flowrate is one of the If the problem is not caused by human
formation or other material sediments main causes of internal fouling. Low error, then in most cases, a malfunction
(usually minerals salts dissolved in crude flowrate can cause high skin temperature of the control system is responsible. In
Yes
oil). Follow the questions below to be and coke buildup as a result of pyrolysis such cases, perform immediate control-
able to differentiate between the two can occur. Did you experience low flowrate systems maintenance or even replacement
phenomena. in one or multiple passes of your heater? procedures.

No

Go to Point DC Consider coke buildup Did you experience any fire-side End
with high probability. Solve (external side) hotspot problems troubleshooting
the problem according to Yes (mentioned in Figure 3) like over procedure.
the instructions given in firing or flame impingement for
the burner troubleshooting long periods of time?
guide found in Ref. 2.
No

The fouling is 90% caused Is your fired heater in atmospheric crude

by mineral sediments. Yes distillation preheater service or is it located
downstream of a de-salter?

No

Fouling is 90% caused by Did the fouling happened right after Consider removing and
mineral sediments or even Yes overhaul or upstream equipment’s replacing the tubes with
small particles carried out by malfunction? supervision of inspection
fluid stream into the furnace department.
tubes. No
Yes

If your furnace is in heavy fluid service, like vacuum-distillation-tower charge heater or Follow spalling and decoking
visbreaker service or any vaporizing heavy feed service, consider your fouling problem with steam and air procedure
DC No
as coke buildup. according to the vendor’s
instructions.
Caution: Coke buildup in gas reactive services like olefin plant furnaces is beyond the Is your problem solved?
scope of this algorithm, look at the proper references.

Follow spalling with steam procedure according to vendors instructions.

No
Is your problem solved?

Yes

be blamed. This particular heater is ternal fouling, caused by rapid local 4. Heavy fuel-oil combustion
in visbreaker service, which can be coke formation, made an insulating problems
classified as a thermal cracking pro- layer, which in turn led to higher tube 5. Low process fluid flowrate
cess. In thermal cracking of heavy skin temperatures for long periods 6. Upstream malfunction of
petroleum cuts, coke formation is of time. In this vein, high tube-skin equipment
an unwanted side reaction that can- temperature led to severe creep and 7. Pyrolysis of heavy hydrocarbon
not be eliminated, but the reaction eventually disastrous tube rupture. In molecules (coke buildup)
rate and duration can be controlled order to solve this problem and pre- 8. Mineral sediments
and reduced by maintaining proper vent it from ever happening again, 9. High firing rate
operating conditions. In some cases the main cause — internal fouling With the exception (in some cases)
the interval between decoking pro- — should be clearly understood and of root-cause Number 9 (high firing
cesses can be extended by up to analyzed. By taking a look at Table 1, rate) [2], all the other causes can be
three to four years. one can find the root causes of inter- treated by the stepwise troubleshoot-
In this particular case, unwanted nal fouling, which are as follows (not ing procedure outlined in this article.
coke deposition had occurred dur- necessarily in order of priority): In order to solve the internal foul-
ing the first year of an operating 1. Flame impingement ing problem, the operator can start
period. This event was unpredicted 2. Heat flux imbalances with the procedure given in Figure
and surprising for the operators. In- 3. Improper fuel pressure 4. This procedure starts with the
74 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015
 C FIGURE 5. Troubleshooting algorithm for external fouling

There are two types of external fouling, metallic oxides and external scaling.
Follow the questions below to be able to differentiate between the two. End troubleshooting procedure.

Yes
The particles are metallic oxides Do the external particles
Seek engineering consultancy.
(MOs). Proceed to point MO. respond to a magnet?

No
No No

Yes Because of possible metallurgical changes, some particles do Metallic oxides are mainly Solve the problem
not respond to a magnet therefore be sure that tube’s surface is caused by over firing or high Yes according to instructions Yes
not scratched or thinned in the particle position. fuel gas temperature ranges. given in burner
Does the tubes thickness decreased in the particle position? Do you experience uneven heat troubleshooting guide
distribution or over firing found in Ref. 2.
No conditions?
Is your problem solved?
In case of gas firing, external No The external fouling is soot or
scales (other than oxides) scale. Do you use fuel oil in Yes MO
are caused by complex heat burners?
transfer phenomena that
needs advanced engineering
consultancy and determining
the reason is not a straight
forward step.
Yes

There is a possibility that your

No fuel oil has a high content of
Seek engineering consultancy. Solve the problem according Main cause of soot formation
vanadium. In high-temperature
to instructions given in burner Yes in fuel oil combustion is poor
ranges or unexpected thermal
troubleshooting guide atomization or combustion
shocks, vanadium oxide would
found in Ref. 2. issues with fuel oil.
attack tube material, rapidly
Do you experience burner poor
causing dangerous corrosion
Yes Is your problem solved? performance?
and reducing tube thickness.
Optimize your firing conditions
No and revise fuel composition if
possible.
Is your problem solved?

There is a possibility that your fuel oil has a high content of sulfur No
(in the form of mercaptans, H2S or any bound or unbound form). Yes
Analyze your fuel and in case of high sulfur content, consider fuel
change or upstream process modifications. No No
Do you frequently experience
Seek engineering
thermal shocks or high firing
Is your problem solved? consultancy.
rates in the fire box?

Yes

simplest solution, which is solving this cause, the operator is asked is not the case in this example,
low process-fluid flowrate in heater to check for burner firing problems, may happen in crude charge heat-
passes. Low process-fluid flowrate like flame impingement, heat-flux ers that are placed after de-salters
can cause high tube-skin tempera- imbalances, improper fuel-oil pres- and heat exchanger networks in
tures that can lead to internal fouling. sure or heavy fuel-oil combustion most refineries. Apparently, if the
It should be noted that human error problems. If any of these problems heater is in crude charge service,
and control-equipment malfunction have been observed for long oper- checking the de-salter’s malfunc-
are the main causes of low process- ating periods, one can almost be tion should be the next step in the
fluid flowrate. It is mandatory to pre- certain that the internal fouling is troubleshooting algorithm. If this is
pare a regular program for checking due to coke buildup. In order to not the case, upstream equipment
control equipment accuracy. Also, solve any burner-firing problem, malfunction can be the next cause.
assign experienced personnel to the operator can refer to the de- Small particles carried by a fluid
control sensitive equipment, such as tailed troubleshooting algorithms stream can plug heater tubes and
fired heaters. described in Ref. 2. reduce flowrate locally or in a whole
If the problem is not related to Mineral sediments fouling, which pass. Similar to low process-fluid

CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015 75

 D FIGURE 6. Troubleshooting algorithm for internal corrosion

Internal corrosion is mainly influenced by the chemical End troubleshooting Seek engineering
composition of the process fluid. Do you have sulfur procedure. DE consultancy.
No compounds present in your process fluid composition?

Yes Yes

Its possible that the inlet process fluid contains particulate

No Do you experience localized or general tube-wall thinning? matter. Particulate matter can increase the corrosion rate by
stripping away protective scale and exposing the fresh metal
to continuous corrosion process (see Ref. 1). Inspect upstream No
Yes process equipment and resolve any deficiency that can cause
particulate formation in process stream.
Is your problem solved?

Yes No
Sulfidic corrosion may have happened inside the tube walls,
beware of chemical components like chlorides and hydrogen Did you experience high corrosion defects after a recent
that can aggravate the corrosion rate. Also local flame overhaul or upstream equipment malfunction? Mostly de-salter
impingement and high skin temperatures can increase the malfunction or low performance?
corrosion rate.

Yes Yes Two-phase flow currents with high linear

Control the chemical composition of the process fluid and Go to Point velocity in either of the phases, or slug
lower critical species concentration. In case of frequent flame DE flow regimes can cause local or overall
impingements, solve the problem according to instructions corrosion and even enhance the corrosion
given in burner troubleshooting guide (Ref. 2). rate caused by other factors. Consult an No
expert engineering team and try to figure out
Is your problem solved? possible improper flow regimes in boiling
regions of tubes. Is your problem solved?
No
Yes

Do you have naphthenic acids present in No

No Is your fired heater in atmospheric or vacuum towers charge
your process fluid composition? heating service?

Yes

Do you experience tube wall thinning in turbulent areas, such as bends

No
or pass outlets?

Yes
No

Naphthenic acid corrosion may have occurred inside the tube walls. Control the chemical composition of Yes Go to Point
Beware that high skin temperatures can increase the corrosion rate. process the fluid and lower critical species DE
concentration if possible.

Is your problem solved?

Key:

A Hotspot B Internal fouling C External fouling D Internal corrosion

flowrate, this can lead to severe in- heavy feed is vaporizing, coke in some cases eliminate — coke
ternal fouling. Operators should be buildup is a likely possibility. As coke buildup phenomena.
warned about the possibility of this formation in these kinds of heaters Other steps of this algorithm are
problem, specifically after overhaul is highly anticipated, controlling op- dedicated to the decoking and spal-
or upstream equipment malfunc- erating conditions, especially the ling practices that should be fol-
tion and maintenance. items mentioned in previous steps of lowed, depending on the nature of
As the next step, the nature of this algorithm, is very important and the internal fouling problem.
the process should be noted. For vital. By following the right steps, Following this procedure would
visbreakers or any service in which operators may control — and even usually eliminate and even prevent
76 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015
 8,500
— Pass A — Pass C
Acknowledgment
8,000 — Pass B — Pass D The authors wish to thank the help
7,500
and contribution of Alan Cross. He
is always a mentor to us who gives
7,000 us the courage and motivation for
Heater pass flowrate, bbl/d

6,500 preparing and submitting our experi-

ences in engineering papers.
6,000

5,500 References
1. American Petroleum Institute, “Inspection of Fired Boil-
5,000 ers and Heaters,” Recommended Practice No. 573, API,
Washington, D.C., Feb. 2003.
4,500
Early warnings of internal fouling in 2. Maghbooli B., Najafi H., Bakhtiari A., Correcting Improper
4,000 pass A (rapid decrease in feed flowrate, Performance of Direct Fired Heaters: A practical, step-
accompanied by 80 psig pressure by-step approach for finding the root cause and trouble-
3,500 increase in pass inlet) shooting burner problems, Chem. Eng., May 2013, pp.
39–46.
3,000 3. American Petroleum Institute, “Damage Mechanisms Af-
0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92
Time, days fecting Fixed Equipment in the Refining Industry,” Recom-
mended Practice No. 571, API, Washington, D.C., April
FIGURE 7. This graph shows the operating history of the process fluid flowrate in four passes of a vac- 2011.
uum distillation unit charge heater. Ignoring early signs of internal fouling caused by rapid coke buildup,
4. Patel S., Typical Fired Heater Problems and Root Causes,
led to disastrous tube rupture, like the one depicted in Figure 1 Hydrocarbon Process., March 2007, pp.68–69.

The complete set of troubleshooting algorithms is

future internal-fouling problems, as of possible internal fouling, this available online at www.chemengonline.com
for the case history shown in Figure problem may not have endured for
1. Surveying the operating history such a long operating period that Authors
of the heater proved that long-time would lead to tube rupture. This is Babak Maghbooli is a
local flame impingement was the another reason for the necessity member of the research and de-
velopment team of Farayand
main cause of this event, and tuning of an operator-involved reliability Sabz Engineering Co. (No.117,
the draft and excess-air amounts program in refineries in addition to Somaye Street, Tehran, Iran;
could have solved the problem in good troubleshooting knowledge Email:maghbooli@xthermo.
the first place. and training. com). With more than six years
of experience in the field of tun-
As a final note, we should add ing process fired heaters and
that our experiences have proven Computerized algorithm troubleshooting their problems,
that in similar case histories, in- A computer program (computer wiz- he has provided consultancy services to more than
five petroleum refineries in the field of troubleshoot-
creasing the awareness and knowl- ard) can easily be developed based ing, maintenance and analyzing mechanical and ther-
edge of operating teams about the on logic described in Figures 3–6, mal behavior of fired heaters and boilers. He also has
initial signs of internal fouling phe- and 8–15. By using this program in practical and professional experience in the field of
process simulation and developing root cause analy-
nomena is as important as know- industrial environments, like petro- sis procedures for chemical process problems. He is
ing the troubleshooting sequences. leum refineries and petrochemical an advanced VBA programmer and has developed
For example, in a similar case that plants, operators will be able to un- many process simulation solutions in MS-Excel for
happened in a vacuum charge derstand and correct the root causes various consulting projects. He holds a B.Sc.Ch.E.
degree from Arak University and he is an associate
heater, operators were too much of a possible deterioration mecha- member of IAChE and CSChE.
dependent on infra-red pyrometer nism in more efficient ways. As a
measurements of tube-skin tem- result fewer maintenance operations Hamidreza Najafi is head of
peratures. That misreading of the or mandatory tube replacements will the research and development
skin temperature made them igno- be needed. team of Farayand Sabz Engineer-
rant to the actual problem that was It should not be forgotten that the ing Co. (same address as above;
Email:hrnajafi@xthermo.com). He
taking place. As depicted in Figure main cost saving and also safety is a professional object-oriented
7, the process fluid flowrate had benefits of this program will result programmer and his main interest
decreased during coke buildup and from the following: is developing simulator packages
for industrial processes. With
simultaneously, the heater pass • Fewer forced direct-fired-heater more than 12 years of experience,
inlet pressure increased 80 psig. outage events due to tube failure he has practical and professional experience in the
These two events are very obvious • Elimination of disastrous events, fields of thermal radiation, especially for design and
simulation of thermal cracking furnaces and process
early warnings of internal fouling in such as a heater explosion caused fired heaters, and also development of chemical kinet-
the specified heater pass (Pass A by unpredicted tube ruptures ics and thermodynamics framework for simulator soft-
of Figure 7). • Extending the operating lifetime of ware. He has provided numerous solutions for process
The operators had simply re- heater tubes problems in these fields for various oil and petrochemi-
cal companies. He has a B.Sc.Ch.E. degree from Sharif
lated the decreased flowrate to • The possibility of increasing University of Technology and a M.Sc.Ch.E. degree from
the control equipment malfunc- throughput and decreased down- Shiraz University. He is an associate member of IAChE
tion and ignored it. If the operators time for decoking operations ■ and CSChE.
were aware of other warning signs Edited by Gerald Ondrey

CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015 77