Improving Plant Operability

Focus on Safety
volume two

CHEMICAL
ENGINEERING
 Table of Contents
Tolerable Risk................................................................................................................................................. 5
While determined risk is generally well understood, tolerable risk can be the missing link to complete risk
assessments

Chemical Lifecycle Management...............................................................................................................11

Sustainability demands a higher level of oversight that seeks new ways to effectively and cost efficiently
procure, handle and dispose of, or recycle materials

Hearing Protection For CPI Workers.......................................................................................................... 15

As statistics on rising rates of noise-induced hearing loss in the workplace show, a hearing-conservation
program alone is not enough. Workers must understand the risks, have access to hearing protection and,
most importantly, use it

Designing for a Safe Process..................................................................................................................... 20

Incorporating safety considerations throughout process design lowers the risk of a hazardous event

Heat Transfer Fluid Leaks: Break the Fire Triangle.................................................................................... 24

Extreme processing temperatures present the greatest risk. Know where leaks are most likely to occur and
how to prevent them

Avoiding Static Sparks in Hazardous Atmospheres............................................................................... 32

Relatively simple steps can be taken to mitigate risks associated with static electricity in process plants

A Checklist for Safer Chemical Batch Reactions..................................................................................... 36

A good understanding of the reaction chemistry is needed for a safe process design

A Safety-Centered Approach to Industrial Lighting................................................................................ 38

The proper design and operation of lighting is essential to ensure plant safety and support good
maintenance practices

Designing Safer Process Plants................................................................................................................. 43

Several often-overlooked strategies to increase inherent safety are discussed here

A Safety Checklist for Laboratories........................................................................................................... 48

These nine best practices for managing change in laboratories can help ensure a safe workplace

Engineering for Plant Safety...................................................................................................................... 51

Early process-hazards analyses can lead to potential cost savings in project and plant operations

Chemical Lifecycle Management.............................................................................................................. 60

Sustainability demands a higher level of oversight that seeks new ways to effectively and cost efficiently
procure, handle and dispose of, or recycle materials

Clearing the Air About Respiratory Protection........................................................................................ 64

Learn the basics about selection and regulatory compliance for these potentially life-saving devices

Fire-Water Pumps for CPI Facilities........................................................................................................... 70

Follow this guidance to improve the selection, design and operation of pumps handling water for
firefighting and related systems
Eye-and-Face Personal Protective Equipment..........................................................................................74
Protecting the eyes and face in the workplace is imperative to preventing the estimated 1020% of work-
related eye injuries that result in temporary or permanent vision loss

Managing Compliant MSDSs and Labels................................................................................................ 78

Compliant MSDS programs depend on inventory management, regulatory knowledge and consistent
authoring

Effective Plant Safety Management.......................................................................................................... 81

Three critical junctures and seven critical steps for plant safety are outlined

Remote Thermal Sensing........................................................................................................................... 83

By making it easy to detect heat anomalies, thermal cameras and infrared thermometers support
preventive and predictive maintenance

Finding the Right Gloves to Fit the Application...................................................................................... 88

There is a wide range of gloves available for hand protection on the job. Matching gloves to their chemical-
resistance properties is one criterion for selection

Dust Control in the Chemical Processing Industries............................................................................... 92

The prevention of dust hazards in the CPI is integral to process-safety management wide-reaching
mitigation schemes must be implemented

The Integral Role of Waste Management in Capital Projects................................................................. 96

To avoid unnecessary efforts and mitigate risks, all phases throughout the design and implementation of
large capital projects should incorporate comprehensive planning for waste management

Why Bad Things Happen to Good Steam Equipment............................................................................103

Accounting for an entire steam-trap population is crucial to avoiding safety incidents and suboptimal
production high-priority consideration must be given to steam-system management

Rotating Machinery: What You Should Know About Operational Problems.......................................112

Follow this guidance to improve the operation, safety and reliability of rotating machinery in chemical
process plants

Chemical Protective Clothing...................................................................................................................117

ISO 16602 offers a much-needed, common global language for expressing protective clothing performance

Technology Profile: Propylene Production via Metathesis................................................................... 120

Preventing Fires and Explosions in Pilot Plants.................................................................................... 121

The issues and complications are different from and in many ways more challenging than those for
full-scale plants

Fire & Gas in Safety Systems.................................................................................................................. 127

Integrating fire-and-gas detectors and mitigation systems into overall process safety control can help
ensure fast responses to emergencies

Anti-surge Valves for Dynamic Compressors......................................................................................... 132

Follow this guidance to improve the selection and sizing of anti-surge valves for axial and centrifugal
compressors

Preventing Dust Explosions..................................................................................................................... 137

Risk management programs are critical for safe handling and processing of combustible dust as well as for
OSHA regulatory compliance
The Next Step Change in Process Safety............................................................................................... 140
Leveraging the convergence of operational and information technologies can aid in minimizing risk

Selecting Laboratory Exhaust Systems................................................................................................. 146

With upfront thought, exhaust systems can ensure personnel safety and improve building efficiency

Vacuum Systems: Recommendations for Safe Operation................................................................... 149

Follow this guidance to ensure that steam ejector systems, mechanical vacuum pumps and integrated
vacuum systems are designed, operated and maintained to ensure process safety

Facts at Your Fingertips: Chemical Protective Clothing........................................................................ 157

Facts at Your Fingertips: Static Electricity Discharge and Fire Prevention.......................................... 158

Facts at Your Fingertips: Liquefied- Industrial-Gas Safety.................................................................... 159

Things You Need to Know Before Using an Explosion-Protection Technique.................................... 160

Understanding the different classification methods is necessary to better select the explosion-protection
techniques that will be used

Advanced Control Methods for Combustion......................................................................................... 165

Advanced control techniques can raise efficiency and lower pollutant emissions in industrial combustion.
The capabilities and adoption of several methods are discussed

Flare Consolidation Considerations........................................................................................................ 171

Consolidating multiple flares can help mitigate emissions and maintenance

Burner Inspection and Maintenance........................................................................................................174

Burners and their components can be quite complex. Establishing maintenance and inspection best
practices encourages longterm operational reliability

Connecting Operations Personnel to Reliability Efforts....................................................................... 180

Seven methodologies are described to help operations staff take greater ownership of asset performance

Improve Rotary Equipment Reliability with Checklists........................................................................ 187

Design selection and commissioning of rotary equipment can benefit by following a structured, checklist-
based method and by promoting end user involvement

An Exercise In Fire Protection.................................................................................................................. 195

Asking the right questions about fire protection will prevent minor incidents from escalating
 What is aN SIS?
Various acronyms are used in discussions about
safety instrumented systems (SISs). While these
Environmental Manager acronyms are defined in the text of this article,
a little more explanation is warranted.

Tolerable Risk
The acronym SIL stands for safety integ-
rity level. Each SIL represents the difference
between the process risk without an SIF, and
the risk deemed tolerable. An SIF is a safety
instrumented function and consists of the sen-
sors, logic solvers, and final elements neces-
sary to detect a particular hazardous condition
While determined risk is generally well understood, tolerable and cause the process to go to a safe state. Its
important to note that SILs apply to each SIF,
risk can be the missing link to complete risk assessments not to the SIS as a whole.
An SIS is a collection of all the sensors,
logic solvers, and final elements necessary to
Mike Schmidt much has been written; and tolerable risk,
address all the SIFs in a process, hence the
Emerson Process Management which is frequently invoked but rarely word system. SILs establish the required reli-

S
afety instrumented systems (SIS)
are in the spotlight these days.
More companies have come to rec-
ognize the importance of recently
published consensus standards, and re-
cent fatal accidents tied to the failure of
instrumented safeguards have brought
further attention. The result is that facili-
ties all over the world are looking at their
safety interlocks and concluding that
what they really need is a full-blown SIS.
As some work to upgrade their SISs,
they find plenty of guidance available for
the early steps process hazard analysis
(PHA), consequence analysis, and layers
of protection analysis (LOPA) and also
guidance for the later steps component
selection, failure probability and SIL
(safety integrity level) verification calcu-
lations, and system-design guidelines. A
disconnect comes when end users have de-
veloped their lists of safety instrumented
functions (SIFs) and are asked to assign
an SIL to each one (Table 1).
Detailed design of an SIS cannot begin
until each SIF has been assigned an SIL.
This is the point in the process where
many are tempted to throw up their
hands in despair and want to arbitrarily
assign everything as SIL 3. This is not the
answer. At the very least, it fails to comply
with the letter and the spirit of the stan-
dards, but worse, it misallocates resources
to overdesign when these resources could
be better used to reduce risk elsewhere.
The disconnect occurs because SIL as-
signment depends on a comparison of
two values: determined risk, about which
Table 1. Safety Integrity Levels
Safety Probability of Risk Reduction
Integrity Failure on Demand Factor (RRF)
Level (PFDAVG)
SIL 4 10-4 > PFD > 10-5 10,000 < RRF 100,000
presented in terms that are useful for SIL
assignment. Instead, the literature usu-
ally contains some variation of this state-
ment: Each organization is responsible
for establishing its own risk-tolerance
criteria. Some organizations have corpo-
rate risk-tolerance criteria that speak to
the total corporate risk, but many have
no risk-tolerance criteria at all. It is still
a rare organization that has established
risk-tolerance criteria that are actually
useful at the level of a process unit, where
the SIS resides.
Recommended approach
A typical approach to SIL assignment,
but certainly not the only one, is to use
a risk matrix. Risk is a product of conse-
quence and likelihood. As shown in Table
2, consequence and likelihood categories,
like the SILs themselves, are separated
by orders of magnitude. The challenge for
each company then, is to define the conse-
quences that are considered serious and
the benchmark frequency (f).
Many companies already use some type
of matrix, usually to prioritize recommen-
dations from HazOps (hazard and opera-
bility reviews) or other PHAs. An example
of this is shown in Table 3, where a rat-
ing of 1 would mean something should be
done immediately, and a 10 would mean
that no action is required.
Typical likelihood categories include
terms such as frequent, occasional, sel-
dom, remote and unlikely but what do
they mean? A company may leave it to the
individual participants in the PHA to de-
cide for themselves.
Table 2. Risk Matrix
f > High
Likelihood
ability of an SIF. Once a safety instrumented
function is designed, reliability calculations
can show if the SIF is reliable enough to reduce
the risk to a tolerable level.
Safety integrity levels range from 1 to 4, but
it should be noted that SIL 4 is not recognized
in all SIS standards. Instead, there is an expec-
tation that the process will be re-designed to
reduce the risk. An SIS can also include SIFs
that are not SIL rated. Various users refer to
them as N/R (not rated) or SIL 0 handy
terms, albeit not defined in any standards.
Some companies define these terms, but
may use frequency intervals that are not
uniformly spaced. A typical set might be:
Event occurs once a year
Event occurs once every 10 years (~1
order of magnitude)
Event occurs once every 50 years (~
order of magnitude)
Event occurs once every 150 years (~
order of magnitude)
Not likely to occur
The same holds true for the severity in the
consequence categories; a typical (nonlin-
ear) arrangement might be:
Loss of life; damage over $1 million
Lost-time injury; damage over
$500,000
Medical treatment; damage less than
$500,000
Minor injury; near miss; poor quality
No injury, impact on process
Converting an existing matrix
To convert an existing matrix to an SIL-
assignment matrix, the first step is to ad-
just the likelihood categories so they are
Consequences (per event)
< 0.01 x > 0.01 x > 0.1 x > 1 x > 10 x
Serious Serious Serious Serious Serious
N/R SIL 1 SIL 2 SIL 3 SIL 4

SIL 3 10-3 > PFD > 10-4 1,000 < RRF < 10,000
High > f > 0.1 High N/R N/R SIL 1 SIL 2 SIL 3
SIL 2 10-2 > PFD > 10-3 100 < RRF < 1,000
f < 0.1 High N/R N/R N/R SIL 1 SIL 2
SIL 1 10-1 > PFD > 10-2 10 < RRF < 100
f =Benchmark frequency
Chemical Engineering www.che.com September 2007 69

69-74 CHE 9-07.indd 69 8/29/07 5:43:13 PM


Environmental Manager
one order of magnitude apart. After that,
consequence categories should be adjusted
so they are also one order of magnitude
apart. The intervals could be 1.5, 15 and
150 years, or 5, 50 and 500 years, as long
as they are an order of magnitude apart.
The next step is to define the conse-
quence categories. The universally used
scale and absolutely required by the
SIS standards is of consequences to per-
sonnel. But other important consequence
QVCMJD
/4
scales, like community, societal or envi-
ronmental consequences, may also show
up in one of these matrices. Just keep in
mind that as an SIF for reducing the risk
of a specific hazard is evaluated, the conse-
QVCMJD
,JOHE
quences of that hazard must be evaluated
on each scale. The event will have the same
frequency, regardless of the types of conse-
quences considered, and the consequence
that results in the highest SIL rules.
While some organizations want to con-
sider site and operability consequences of
a hazard in their analyses, there are seri-
ous disadvantages to keep in mind. One, it
puts a dollar scale on the same matrix as
environmental, health and safety (EHS)
consequences, and it is not a big leap for
a plaintiffs attorney to equate the EHS
consequences to the dollar amounts. Also,
functions that strictly protect against eco-
nomic loss dont need to be burdened with
the documentation requirements of the
SIS. Organizations are usually better off
to leave the economically driven functions
in the basic process control system and
avoid giving them SIL ratings.
Consequence categories
If classifying likelihoods into categories
that are separated by orders of magni-
tude is hard, its nothing compared to
separating consequences into categories.
Once a benchmark frequency is defined
as a number, dividing or multiplying it
by ten is straightforward. Development of
consequence categories, however, requires
agreement on what consequence is ten
times worse than another.
In the early twentieth century, H.W.
Heinrich [1] proposed the idea that for
each fatality there were many incidents
with less serious consequences, and that
for each incident with less serious con-
sequences, there were even more near
misses. From that simple observation
grew the notion that if we could be more
alert to near misses and reduce them, we
would necessarily reduce the incidence of
70 Chemical Engineering www.che.com September 2007
69-74 CHE 9-07.indd 70
Table 3. An example of a
matrix for PHAs
Severity (Consequence)
5 4 3 2 1
1 5 4 3 2 1
2 8 7 6 4 2
Likelihood
3 9 8 7 6 3
4 10 9 8 7 4
5 10 10 9 8 5
Table 4. Relative cost of
consequences
Fatalities 1.0x
Serious injuries 0.1x
Injuries (reportables) 0.01x
First aids (non-reportables) 0.001x
more severe consequences. This led to the
now familiar, safety pyramid (Figure 1).
While the premise of the safety pyra-
mid has recently been called into ques-
tion, Heinrichs original observation is
still valid: near misses, first aids, and
fatalities are each separated by about an
order of magnitude (Table 4).
Many have published cost estimates as-
sociated with increasing levels of injury,
with costs ranging from a few hundred dol-
lars for a first-aid injury to a million dol-
lars for a fatality. What the various scales
have in common is that severity of conse-
quences, when reduced to dollars, seems to
be separated by orders of magnitude.
Once likelihood and consequence cat-
egories that are separated by orders of
magnitude have been established, an SIL-
assignment matrix can be developed. An
example of a blank table is given in Table
5. All that remains is to associate an SIL
with each cell in the matrix.
What is a tolerable risk?
SILs represent the residual risk between
what is left after all other layers of protec-
tion have been applied to a hazard, and
the tolerable risk. But what is the toler-
able risk? Consider this: Whether or not
your company has established a tolerable
risk criterion for your facility, what do you
believe it should be? How infrequently
would a fatality occur for you to be able
to describe your facility as safe? Once a
year? Once every three years? Once every
5,000 years? Once every million years?
The ALARP principle (Figure 2) was de-
veloped by the Health and Safety Execu-
tive of the U.K. It states that risks should
be reduced to a level that is As low as
reasonably practicable. The ALARP prin-
ciple first divides risk into three areas:
Intolerable Risk at the high end, Negli-
gible Risk at the low end, and the Toler-
able Risk that falls between the two. Risk
in the middle region can be tolerated as
long as all cost-effective measures to re-
duce risk have been put into place. Any-
'BUBMJUJFT
%JTBCJMJUJFT
*OKVSJFT
'JSTU
BJET
/FBS
NJTTFT
FIGURE 1. The Accident Triangle or Safety
Pyramid depicts the simple observations of
H.W. Heinrich [1]
'BUBMJUJF
NBO

*OUPMFSBCMF
SJTL
NBOZFBS
ZFBS "VTUS
XPSLFS
)POH
,
"-"31
PS
UPMFSBCMF
SJTL
/FUIFSMB
SFHJPO
NBOZFBS
ZFBS 6O
XPSLFS
/FHMJHJCMF
SJTL
FIGURE 2. The ALARP principle, developed
in the U.K., states that risks should be reduced
to a level that is as low as reasonably practi-
cable
one operating a process with risks in the
tolerable risk region must demonstrate
that they have achieved the lowest risk
possible, taking into consideration cost
versus risk reduction.
Government mandates
Some countries have mandated upper and
lower levels for tolerable risk. Most coun-
tries that have set these levels use the
average level of risk faced by the entire
workplace population, regardless of oc-
cupation, as the upper limit for tolerable
risk. In many developed countries, that
is the equivalent of one fatality per year
per 1,000 workers. This upper bound-
ary represents the level of fatality rate
already tolerated in industries such as
underground mining or deep-sea fishing.
The lower boundaries are generally two
or three orders of magnitude lower.
The Netherlands took a similar ap-
proach, but then declared that workplace
risk should be 1% of general risk, because
it is not voluntary risk. This results in lev-
els that are 100 times lower than those
chosen by most countries that set levels.
These levels, as shown in Figure 3, are
sometimes called bright lines, especially
when discussed in the U.S., where there
has been no success in setting tolerable
risk levels. The argument was made that
a strict bright line approach to decision
making is vulnerable to misapplication
since it cannot explicitly reflect uncer-
8/29/07 5:43:54 PM
 Table 5: SIL Assignment Matrix,without SILs
5 4
< 0.1 in- 0.1 injuries 0.1 disability 0.1 fatality
juries 1 first aid
1 near miss 10 first aids 10 injuries
1. Frequent
(>1 event/1.5 years)
2. Occasional
Likelihood
(>1 event/15 years)
3/4. Seldom/Remote
(>1 event/150 years)
5. Unlikely
(<1 event/150 years)
'BUBMJUJFT
NBO
ZS        
"VTUSBMJB
/48
*
)POH
,POH
/FUIFSMBOET
6OJUFE
,JOHEPN
DPNQBOJFT
FIGURE 3. Government mandates defining
tolerable risk are often referred to as bright
lines. The color coding here refers back to
the 3 levels in the ALARP principle as shown
in Figure 2. The U.S. does not set tolerable
risk levels, nor offer guidelines
Table 6: Voluntary risks
Smoking 9.7 x 10-3/yr or 9,700 fatalities per year per million smokers
Automobile accident 1.4 x 10-4/yr or 140 fatalities per year per million people
Lightning strike 1.5 x 10-7/yr or 15 fatalities per 100 years per million people
tainty about risks, variation of suscep-
tibility within a population, community
preferences and values, or economic con-
siderations all of which are legitimate
components of any credible risk manage-
ment process. In other words, in the U.S.,
tolerable risk remains very much subject
to the political process.
CPI benchmarks
Almost all major companies in the chemi-
cal process industries (CPI) have guide-
lines for tolerable risk, but they are
generally considered highly confidential.
No company is going to announce to the
public, for example, that they would tol-
erate up to one fatality per year per ten
thousand workers.
Some companies, especially smaller
ones, do not have explicit guidelines. How-
ever, the risk they tolerate can be inferred
from the risk reduction measures they
have in place. Smaller companies with-
out explicit guidelines generally tolerate
greater levels of risk than their larger
counterparts and this isnt inappropriate.
A small company, struggling to establish
itself or to make payroll, is going to look at
the amount of risk that it can and must
tolerate to get through the next year very
differently from a company that is look-
69-74 CHE 9-07.indd 71
Consequences (per event)
3 2 1
1 fatality
1 Injury 1 disability 10 dis-
abilities
       
$PNQBOZ
$PNQBOZ
**
$PNQBOZ
***
4NBMM
FIGURE 4. CPI benchmarks show that
large, multinational companies tend to set
levels consistent with government man-
dates. Smaller companies tend to operate in
wider ranges and implicitly at higher levels
of risk
ing out over the next 100 years. A small
company with a single plant will look at
a single event that occurs once every 250
years and may very well decide that once
every 250 years might as well be never.
A large company, with 50 similar plants
around the world would look at that same
risk as once every 250 years for each plant,
but once every 5 years for the corporation,
and make a very different decision about
whether the risk is tolerable (Figure 4).
Voluntary and natural risks
The levels mandated by governments,
or self-imposed by companies, can best
be understood in terms of some of the
risks to which people voluntarily expose
themselves. That is not to say that risks
like these should ever be compared to
process risks in a public setting, which
would likely inflame the audience. But,
for ones own understanding, it is helpful
to know that the upper level for ALARP is
typically set at 1 x 103 per year, or almost
ten times lower than the risk to which
smokers expose themselves. If a chemical
process had the same risk as smoking, it
would be considered an intolerable risk.
On the other hand, dying from a light-
ning strike would generally be considered
a negligible risk. If you consider your own
Bioengineering
Inversina
the gentle way
of mixing.
The Inversina mixes solids or liquids
thoroughly and efficiently. The process
is clean, because mixing takes place in
closed containers that can be quickly
interchanged. The Inversina mixes a
diverse range of components rapidly
and in an extremely gentle way.
Segregation does not occur, even
after extended mixing times, by virtue
of the eversion phenomenon ( Paul
Schatz principle ) .
Applications for the Inversina: analyti-
cal labs, metal finishing shops, powder
metallurgy and nuclear industry,
manufacture of batteries, cement,
ceramics, cosmetics, dental products,
diamond tools, dyes and pigments,
electrical and electronic devices,
explosives and pyrotechnics, foods,
homeopathic products, household
products, medicines and pharmaceu-
ticals, plastics, printing inks and many
other products. The Bioengineering
Inversina is available with capacities
of 2, 20, 50, 100 and 300 L .
Bioengineering, Inc.
Waltham, MA 02451, USA
Bioengineering AG
8636 Wald, Switzerland
info@bioengineering.ch
www.bioengineering.ch
Circle 52 on p. 92 or go to
adlinks.che.com/6899-52
8/29/07 5:45:12 PM
 Table 7. completed risk matrix for example A
5 4 3 2 1
< 0.1 0.1 injuries 0.1 disability 0.1 fatality fatality
Environmental Manager injuries 1 first aid 1 Injury 1 disability 10 dis-
1 near miss 10 first aids 10 injuries abilities
1. Frequent N/R SIL 1 SIL 2 SIL 3 Redesign
(1 event/1.5 years)
personal risk management strategies,

chances are that you have not invested in
lightning protection at your home. Even
in your plant, the lightning protection is
for electrical equipment, not personnel.
The risk of a fatality in a car accident is
within the ALARP region. Of course driv-
ing has the potential for fatalities, but the
risk is tolerable because the convenience
or benefit is worth it. If it wasnt, society
would insist on lower speed limits and
more safety features in cars.
Example A: 1 fatality/100 years
Lets get back to SIL assignment. While
each of us has a different level of personal
tolerable risk, when asked how small the
average frequency for a fatality would
need to be for a plant to be considered
safe, it is not at all uncommon for people
to pick 100 years. This is longer than any
one persons career but short enough to be
meaningful in terms of the life of a facil-
ity. So consider a facility with a tolerable
fatality rate of one per 100 years.
Likelihood
2. Occasional N/R N/R
(1 event/15 years)
3/4. Seldom/Remote N/R N/R
(1 event/150 years)
5. Unlikely N/R N/R
Further assume that at this facility,
the workforce includes 250 workers who
would be exposed to process risks:
(1 year/250 man-years) (1 fatality/100
years) = 1 fatality/25,000 man-years
= 4 x 105 fatalities/man-year
= total tolerable risk
In this case the total tolerable risk is one
fatality per 25,000 man-years, or 4 x 105
fatalities per man-year, which falls in the
middle of the tolerable risk region.
Process risk is only part of risk
Note that the tolerable risk established
above is for all fatalities in the plant. How-
ever, not all fatalities result from process
safety risks, that is from process hazards.
There are still the slips, trips and falls,
the housekeeping issues, the electrical
SIL 1 SIL 2 SIL 3
N/R SIL 1 SIL 2
N/R N/R SIL 1
safety and confined-space entry issues,
and transportation-related fatalities.
For Example A, assume that half of the
risk comes from process safety risks or
process hazards. If total tolerable risk is
4 x 105 fatalities/man-year, then process
safety risk is half of that, or 2 x 105 fa-
talities/man-year.
Although weve assumed that no more
than half of the risk comes from the pro-
cess hazards, we need to consider all of
the process hazards that could lead to a
fatality. Unless there is only one poten-
tially fatal hazard, not all of the process
safety risk can be allocated to a single
hazard. One approach is to assume work-
ers are each exposed to a certain number
of potentially fatal hazards, say five, and
then distribute the risk budget over

XXXDIFDPN

5IF
XFCTJUF
GPS
$1*
1SPGFTTJPOBMT

XXXDIFDPN
PGGFST
M
"DDFTT
UP
UIF
NPTU
SFDFOU
JTTVF
PG
$&
M
$&
"SDIJWFT
o
FWFSZ
BSUJDMF
QVCMJTIFE

JO
$&
GSPN
UIF
QBTU

ZFBST
M
$&
/FXTMFUUFST
M
$&
8FCJOBST

M
*OEVTUSZ
FWFOU
JOGPSNBUJPO

M
"OE
NVDI
NPSF

'PS
NPSF
JOGPSNBUJPO
DPOUBDU
+PIO
4USB[[B


BU

KTUSB[[B!DIFDPN

XXXDIFDPN 

Circle 53 on p. 92 or go to adlinks.che.com/6899-53
72 Chemical Engineering www.che.com September 2007
#%7EB(OUSE!DINDD !-

69-74 CHE 9-07.indd 72 8/29/07 5:46:39 PM

 Table 8. comleted risk matrix for example b
5 4
, "   / ,
/  "
< 0.1 0.1 injuries 0.1 disability 0.1 fatality
injuries 1 first aid
1 near miss 10 first aids 10 injuries
1. Frequent SIL 2 SIL 3
(1 event/1.5 years)
Likelihood
2. Occasional SIL 1 SIL 2
(1 event/15 years)
3/4. Seldom/Remote N/R SIL 1
(1 event/150 years)
5. Unlikely N/R N/R
those five hazards in establishing SIL se-
lection criteria. Other hazards, although
not expected to be fatal, can then use the
same criteria. If total process safety risk
is 2 x 105 fatalities/man-year, then:
N/R: 4 x 106 fatalities/man-year
SIL 1: 4 x 105 fatalities/man-year
SIL 2: 4 x 104 fatalities/man-year
SIL 3: 4 x 103 fatalities/man-year
SIL 4 or re-design: > 4 x 102 fatalities/
man-year
Putting a stake in the ground
With the tolerable risk criteria settled,
then the next step is to assign an SIL to
one of the cells in the matrix. For this ex-
ample, lets look at cell 22 in Table 5.
We want to consider the risk at the mid-
dle of the cell. Since the scale is logarith-
mic, the middle of the cell on the likelihood
scale is the log (geometric) mean of 1.5
and 15 years, or 4.7 years. The log mean of
0.1 fatalities per event and 1 fatality per
event is 0.32 fatalities per event.
Doing the math, the center of our tar-
get cell has a risk of 2.7 x 104 fatalities
per man-year. Going back to our tolerable
risk criteria, that is less than 4 x 104, so
it translates to SIL 2.
Once we can assign SIL 2 to that cell,
completing the rest of the matrix is easy
(Table 7). The cell to the left is one order
of magnitude less risk, so it is SIL 1. Ev-
erything to the left of that is not rated.
Similarly, the cell below is one order of
magnitude less risk, so it is also SIL 1.
Everything below that cell is not rated.
Filling in the rest of the matrix follows
the same logic.
When were finished, we have a risk
matrix that can be used for SIL assign-
ment for all SIFs, whether or not fatal
consequences are anticipated.
Interestingly enough, when the risk
matrix is finished, it spells out the crite-
ria for assigning an SIL to an SIF without
3 2 1  
fatality
1 Injury 1 disability 10 dis- 1 / ,
  / ,
/  "
abilities
Redesign Redesign Redesign
"   / ,
/  "
SIL 3 Redesign Redesign
SIL 2 SIL 3 Redesign
i`}iV
SIL 1 SIL 2 SIL 3
tially fatal hazards. In this case, however,
senior management has declared that risk
needs to be reduced to the level consid-
ered negligible on the ALARP diagram
(Figure 2). That is, less than one fatality
per million man-years.
The portion of total risk to an indi-
vidual attributable to process safety risk
would again be half, or 5 x 107 fatalities
per man-year. Once again, with each in-
dividual worker exposed to about five
potentially fatal hazards, the threshold
for avoiding an SIL-rated SIF is 1 x 107
fatalities per man-year.
Likewise, SIL 1 is less than 106, SIL 2 is
less than 105, and SIL 3 is less than 104.
In the CPI, anything larger than 104 would
need to be redesigned to reduce the hazard
before a safety instrumented function for
risk reduction was even considered.
Looking at the middle of the same cell,
we still end up with a risk of 2.7 x 104
fatalities per man-year. The plant is the
same, so the risk is the same. Whats dif-
ferent is the tolerable risk. In the case of
Plant B, the tolerable risk is set at a much
lower level, and so in this case, the risk
is not quite low enough to meet the SIL
3 rating:
i>{`iV>`ii>iLii
(1 event/4.7 years) (0.32 fatalities per }iVVivi`iii
event) (1 yr/250 man-years) = 2.7 x 104 >``VviL>i`Vvi
fatalities/man-year ViV>>`>>ViV>`/
}>`} iiiVi ii> v
SIL rating = SIL 4 or redesign >`i>LiVi>
vi>ViiViL>ii
As in the first example, the rest of the `}`Vvi}i
matrix is easily populated. Because of the >>`>iv>Vi
difference in tolerable risk, the SIL as-
signments have shifted toward the lower Please visit us at the Chem Show. Booth #431
left-hand corner of Table 8.
Variations to consider
A variation to consider, depending on risk
philosophy, is what part of a cell to bench-
mark to. In our examples, we took the cen-
ter of a cell. If the upper right corner is 
,"9
,L
ever explicitly acknowledging a tolerable
risk. The tolerable risk is only implied.
Example B: 1 x 106 fatalities/man-yr
Now consider Plant B. It is very similar to
Plant A, with the same number of exposed
used instead, the result will be an SIL as- ,i}>>i
signment that is higher by one level. This x7iL>`ii>
is clearly a more conservative approach. /i{xn
Another variation to consider is the
vJV`>``i
number of categories for both likelihood
and for consequence. Typically, a 5 x 5 ma-
777
,"9
,
"
workers and the same number of poten- trix is the largest workable matrix, with
Circle 55 on p. 92 or go to
adlinks.che.com/6899-55
55,6x254_baer_engl.indd 1 16.02.2007 9:35:53 Uhr

69-74 CHE 9-07.indd 73 8/29/07 5:49:12 PM

 Environmental Manager
a larger matrix, the team doing risk as-
sessment will probably spend more time
debating categories than is warranted. By
the same token, if the matrix is smaller
than 3 x 3, it is unlikely to have enough
detail to be worthwhile.
In general, it is a good idea to extend
the matrix far enough to the left to get a
column of all non-rated SIFs, or else all the
SIFs, regardless of risk, will end up being
SIL rated. By the same token, there is no
reason to extend the matrix farther to the
left than a column that is all N/R, or far-
ther to the right than is all redesign.
Business results achieved
When a company establishes a risk
matrix for SIL assignment, the matrix
receives a fairly broad distribution. To
the extent that the company is uncom-
fortable publicizing its tolerable risk
criteria, the risk matrix approach to
SIL assignment does not require an ex-
plicit statement of tolerable risk.
Because likelihood and consequence
Problems with
poor flowing
powder or
bulk solids?
ENGINEERED SOLUTIONS JENIKE & JOHANSON
Improving your bulk solids handling efficiency
T 1 (978) 649-3300
www.jenike.com
Circle 56 on p. 92 or go to adlinks.che.com/6899-56
74 Chemical Engineering www.che.com September 2007
69-74 CHE 9-07.indd 74
only need to be estimated to the nearest less under-specification with intolerably
order of magnitude, most SIL assign- high risk, and much less over-specifica-
ments can be made without using de- tion with resulting misallocation of lim-
tailed and expensive quantitative risk- ited safety resources.
analysis techniques. It is usually easy to Edited by Dorothy Lozowski
get agreement, based solely on the expe- Reference
rience of the risk-assessment team, that 1. Heinrich, H.W., Industrial Accident Prevention: A Sci-
an event is likely to occur between 10 and entific Approach, 4th ed., McGraw-Hill, N. Y., 1959.
100 years, rather than spend the time de- Author
termining whether it will be once every Mike Schmidt is a principal
consultant with Emerson Process
37 years or once every 54 years. Management at the Refining and
Since it is usually easy to get agree- Chemical Industry Center in St.
Louis (641 Lambert Pointe Drive,
ment in the middle regions of consequence Hazelwood, MO 63042; Phone:
314-872-6069; Fax: 314-872-8686;
and likelihood, quantitative risk analy- Email: mike.schmidt@emerson
sis can be reserved for the more extreme process.com). His responsibilities
include facilitating HazOps and
situations, such as hazards that are either other PHAs, consequence analy-
expected to be very, very rare, or to have sis, facilitating LOPAs, reviewing
and preparing SRSs (SIS safety requirements speci-
very, very large consequences. And even in fications), and performing SIL calculations. Schmidt
those cases, the quantitative analysis only consults on process design and optimization and teaches
in all these areas. In addition to being a registered PE
needs be done if there is a question about in several states, Schmidt is a CFSE. He has been with
Emerson for over ten years, and has been working in the
the SIL assignment of a particular SIF. CPI since 1977. The majority of his career has been in
Finally, risk matrix is easy to apply, operating companies, including Union Carbide, Rohm
and will result in hazards and their asso- and Haas, and Air Products. He is the author of several
articles on process design and process safety, and is cur-
ciated risks being treated similarly from rently serving on the CCPS committee that is preparing
a guideline book
Ad 385-3c ChemEng:Layout on tolerable risk
1 7/20/07 and onPM
2:08 the API
Pagecom- 1
project to project. This will result in much mittee on overfill protection.
For over 65 years, Mueller has earned
distinction for delivering outstanding
equipment and unique solutions to the
process industries.
Your production plant is an integrated
system that requires a broad range of
disciplines to optimize.
Paul Mueller Company has the technical
expertise, innovative engineering, and
manufacturing resources to implement a
N ow is the time to make the
most of your investment by
process system specific to your needs.
using Jenike and Johanson as your From modular systems to complete
engineered solution provider for turnkey plants, when you think of
reliable and safe solids handling process, think
and processing systems. Over 40
MUELLER
years of proven technology with
hands-on experience to improve
your bottom line.
Your Systems Specialists!
F 1 (978) 649-3399
2007 Paul Mueller Company 385-3c
Circle 57 on p. 92 or go to adlinks.che.com/6899-57
8/29/07 5:50:14 PM
 Environmental Manager

Chemical Lifecycle Management

Sustainability demands Waste diversion/disposal options

a higher level
Manufacturer / waste generator

of oversight that seeks Production Laboratory

new ways to effectively Carbon,

chlorine and SAA SAA SAA
other offsets
and cost efficiently MSA

procure, handle and Reprocess

dispose of, Waste

Toll Recycle Reuse Disposal

or recycle materials management

disposal Secondary market Fuel
blending
Landfill Incineration
services
(recycle) Treatment
Victor Belenchia
Clean Harbors Input to manufacturing process

C
hemicals in production and
laboratory environments
alike have three distinct
lifecycle phases: the first is
procurement and inventory storage;
the second is use in a manufacturing
process or research program; and the
third is post-use, including onsite han-
dling, removal and beneficial reuse,
recycle or disposal.
A number of considerations includ-
ing regulations, safety, material uti-
lization efficiency, costs and sustain-
ability govern all of the steps in this
lifecycle. The challenge for laboratory
and production managers is to balance
these sometimes competing forces to
create an effective process that meets
the organizations overall needs.
Proper chemical handling during
processing is typically emphasized in
production facilities, but how to han-
dle chemicals post-use is often given
less attention. While this article ex-
amines the full chemical lifecycle, it
focuses on this post-use phase.
Sustainable lifecycles
Every laboratory and chemical pro-
cessor must follow environmental,
health and safety (EH&S) proce-
dures and regulations. This is often
a challenge, especially for smaller
operations, since implementation fre-
quently rests on individual managers,
44 Chemical Engineering www.che.com December 2010
Figure 1. Managing post-use chemicals is a vital part of chemical lifecycle man-
agement. Recycling waste chemicals is an option that can contribute greatly to
sustainability goals
local teams and staff who have other
primary responsibilities. Regardless
of size, laboratory and production
managers must periodically review
their policies, conduct training and
audit their chemical inventories and
waste materials. Requirements, pro-
cesses and regulations change, and
it is managements responsibility to
remain in compliance, while meeting
everyday production demands.
Environmental sustainability con-
tinues to increase in importance and
is embraced as a corporate goal at
many manufacturing organizations.
This takes chemical management be-
yond the traditional regulatory and
safety considerations. The chemical
process industries (CPI) have become
even greater stewards of the materials
throughout their entire lifecycles as
they seek ways to further reuse or re-
cycle materials that were historically
disposed of as waste (Figure 1).
Sustainability requires the mini-
mization of consumption and waste
in order to reduce environmental im-
pact through process optimization and
reuse or recycling whenever possible.
This can require additional record-
keeping and administrative steps. The
challenge, when it comes to chemical
lifecycle management, is to find op-
erationally efficient and cost-effec-
tive ways to reduce the overall waste
stream through process optimization,
efficient material use, tolling or sell-
ing recyclable materials to a second-
ary market.
Purchasing and storage
Effective chemical management
starts before the chemicals arrive at
the site. Laboratory and operations
personnel need to work carefully with
purchasing staff to order the correct
amounts, minimize order errors and
ensure that proper receiving and
storage provisions are in place. For
operations personnel, it is a balance
between anticipating customer orders
and manufacturing volume in order
to eliminate production interruptions,
minimize storage and other inventory
costs, and to avoid having to dispose
of outdated materials. Purchasing
staff can consolidate orders to lower
costs, but they must have an under-
standing of the chemical properties
and intended uses in order to ensure
that materials are consumed before
they begin to degrade and negatively
affect chemical reactions or final prod-
uct properties. Overall, the CPI have
done a good job utilizing continuous
improvement techniques to manage
their supply chains.
Once received, the materials must
be stored in accordance with local,
state and provincial regulations, or
the International Code Councils
(ICC) International Building and In-
ternational Fire Codes, and handled
in accordance with company EH&S
procedures. In the U.S., handling
must follow the Occupational Health
& Safety Admin.s (OSHA) Code of
Federal Regulations (Title 29 CFR
1910.1200), which covers Hazard
Communication Standards (HCS),
the federal Risk Management Pro-
gram for certain highly toxic chemi-
cal compounds regulated under the
Clean Air Act, and the Chemical Hy-
giene Standard for laboratories (Title
29 CFR 1910.1450).
Production and laboratory use
As materials move into the second
phase of the lifecycle use in a pro-
duction environment or laboratory
the potential for handling errors in-
creases. Although industrial users and
researchers are concerned with, and
trained in, safe handling procedures
for hazardous materials, they are less
likely to focus on storage and removal.
This leads to some potential problems
in chemical lifecycle management.
Chemicals are generally well man-
aged in production environments
since production procedures are well-
defined and engineered. Failure most
often results from human error or
system breakdowns. Although the
results can be catastrophic, they are
largely preventable through adher-
ence to established EH&S procedures,
effective maintenance practices and
frequent training.
Research environments are less
systematized, which presents greater
opportunity for errors. Researchers
often work independently on projects
and they use smaller quantities of a
greater number of chemicals. This
presents an exponentially higher
number of possible chemical reactions
and storage issues.
Working alone or on small teams,
researchers often neglect to properly
label chemicals in secondary containers
at their stations. Unlabeled or improp-
erly labeled chemicals violate hazard-
ous communication (Hazcom) regula-
tions. Informal practices also lead to
the potential for abandoned chemicals
as researchers move on to other jobs or
projects. Unidentified chemicals may
require testing, special handling and
expensive disposal if they cannot be
verified by the researcher.
Consistency and training, sup-
ported by accurate, up-to-date la-
beling and recordkeeping, are key
ingredients to a successful chemical
management program.
Post-use
Handling of solvents, catalysts and
other chemicals can become much
more problematic once they have
been used in a production or labora-
tory process. They are no longer in
their original containers and are often
combined with other substances. In
the U.S. for example, OSHA, the En-
vironmental Protection Agency (EPA),
Dept. of Transportation (DOT) and, in
some cases, Drug Enforcement Admin.
(DEA) regulations also come into play.
A few important fundamental facts
must be kept in mind at all times: In
the U.S., all chemical storage areas
fall under Resource Conservation and
Recovery Act (RCRA) regulations and
are subject to inspection by regulators.
To comply with that, a limited number
of conveniently positioned and prop-
erly organized satellite accumulation
areas (SAAs), under the control of the
EH&S or chemical hygiene manager,
should be set up near production or
research areas.
There are a number of RCRA rules
for SAAs. For instance, containers
must be in good condition without
rust, dents, or cracks; they cannot be
stored near drains or other structures
that could pose an environmental
risk; hazardous waste containers can-
not exceed 55 gal; acutely hazardous
waste cannot exceed one quart; SAAs
must be inspected weekly. There are
additional guidelines including re-
quirements for clear and visible labels
stating hazardous waste and listing
the container contents, as well as the
three-day rule that requires that the
containers are immediately marked
Chemical Engineering www.che.com December 2010
with the current date when full, and
removed within three days to the main
storage area (MSA).
The MSA is governed by additional
rules. One of the most important re-
lates to allowed accumulation time
once materials enter the MSA. Small
quantity generators (SQGs) organi-
zations that generate more than 100
kg but less than 1,000 kg of hazard-
ous waste per month can accumu-
late materials for 180 days from the
start date.1
Large quantity generators (LQGs)
those that generate 1,000 kg or
more of hazardous waste per month,
or more than 1 kg per month of acutely
hazardous waste are subject to a
90-day accumulation rule.2
Regardless of size, the MSAs must
be secured against unauthorized entry;
hazardous waste and container con-
tents labeling rules must be followed;
containers must be inspected weekly
with reports kept on file; and pre-
paredness and prevention equipment
(for example an emergency phone,
alarm, fire suppression and spill pre-
vention equipment) is required.
There are additional regulations
and exceptions to the rules noted
above. For instance, peroxide form-
ers, such as ethers and dioxane, are
generally managed under a peroxide-
former program that tracks the ma-
terial based on the manufacture and
retention dates from the day it arrives
onsite until the day it is consumed or
disposed of. SAA storage times and
90/180-day MSA rules are all subordi-
nate to the peroxide-former deadlines.
So, if the earliest peroxide-former re-
tention date is 45 days in the future,
that is the containers deadline. Even
the waste disposal company must
track peroxide formers and follow ex-
pedited destruction schedules based
on the retention date.
Managing chemical storage areas.
A minimum number of properly
trained staff should be assigned to
transport materials from the SAAs to
the MSA and manage the materials in
all of the storage areas. Some opera-
tions are large enough to have a dedi-
1. http://www.epa.gov/waste/hazard/generation/
sqg/index.htm
2. http://www.epa.gov/solidwaste/hazard/
generation/lqg.htm
45

Environmental Manager
cated full-time staff, while for others it
is a part-time function (that requires
the same annual refresher training).
Properly organizing the handling
and in-house movement of chemicals
removes the responsibility from re-
searchers and substantially reduces
training requirements.
Another benefit of using dedicated
staff is that it puts a third party in po-
sition to regularly inspect warehouse,
production and laboratory areas to
ensure that chemicals are properly
stored and handled in accordance with
regulatory and company policies. Here
are some best practices that we have
seen in the field and that we practice
at our own sites and for customers:
Ensure that proper procedures
(such as labeling and tracking) are
being followed at each production
and laboratory location, and that
waste chemicals are stored only at
approved sites
Conduct periodic safety training for
research and production personnel
Establish collection schedules to ef-
ficiently move materials from SAAs
to the MSA in a timely way
Ensure that SAA containers are in
good condition, properly closed and
placed within secondary containers,
and that incompatible materials are
properly segregated
Catalog, store and consolidate MSA
materials in bulk containers and
pack smaller quantities, elbow bot-
tles, and so on, for transportation
Identify unknowns before shipping
Follow up on the disposal or recov-
ery of chemicals for company and
regulatory reporting
Software. Most large companies
utilize chemical inventory manage-
ment systems, while many laborato-
ries still use spreadsheets to manage
their inventories. Spreadsheets can be
helpful, but they are not the answer
for corporate-wide or comprehensive
lifecycle-management programs. They
are generally home-grown standalone
applications without built-in report-
ing mechanisms. They do not extend
beyond the department or integrate
with corporate systems. They do not
tie into industry resources and docu-
mentation, such as material safety
data sheet (MSDS) databases, that are
necessary to build chemical profiles.
46 Chemical Engineering www.che.com December 2010
TRI Chemicals
A
n added benefit of recycling and reusing chemicals can be reducing the volume
of materials listed in Toxics Release Inventory (TRI) reports. This can contribute to
reduced taxes and annual fees associated with RCRA regulations. TRI chemicals
have included the following:
Aluminum (fume or dust)
Aluminum oxide (fibrous forms)
Ammonia (anhydrous and aqueous)
Asbestos (friable)
Hydrochloric acid (acid aerosols)
Nitrate compounds (water dissociable;
aqueous solution)
Phosphorus (yellow or white)
Sulfuric acid (acid aerosols)
Vanadium (except when contained in an
alloy)
Zinc (fume or dust)
SOURCE: U.S. Environmental Protection Agency: EPA 260-R-09-006 October 2009 Toxic Chemical Release
Inventory Reporting Forms and Instructions Revised 2009 Version (http://www.epa.gov/TRI/report/rfi/
ry2009rfi121709.pdf). See Table II-3, p. 143 for the full alphabetical list of chemicals and qualifiers
Perhaps, most importantly, they may
not stand up to regulatory scrutiny if
the need arises.
Specialized software and online
chemical-waste-management systems
handle all of these issues and also
address many other important cri-
teria, such as chemical movements,
90/180-day storage-time limits, chemi-
cal expiration dates and reporting.
More sophisticated software and on-
line tracking systems assign cost cen-
ter allocations, and facilitate move-
ment, inventorying and packing, as
well as documenting the transfer of
the material to a certified waste dis-
posal company.
Waste disposal companies
Given the complexity of effective
waste-chemical management, many
companies find that it is more ef-
ficient and cost effective to turn to
waste disposal companies to handle
their chemical waste throughout the
entire post-use lifecycle phase. Service
companies manage the entire process
including onsite movements, storage,
packing and removal. In that way,
generators do not have to invest as
heavily in RCRA training and can rely
on the contractor to apply its exper-
tise, provide staff and backup staff as
necessary, monitor regulatory federal,
state and local developments, and im-
plement best practices.
This allows plant and laboratory
management to properly oversee the
function without adding headcount. It
also enables staff to focus on their core
responsibilities and not be burdened
with additional training or oversight
of chemical management functions.
The TRI listing for the following three
chemicals is based on the chemical activity
rather than the form of the chemical:
1. Dioxin and dioxin-like compounds
(only if they are manufactured at the fa-
cility; or are processed or otherwise used
when present as contaminants in a chemi-
cal but only if they were created during the
manufacture of that chemical)
2. Isopropyl alcohol (only if it is being
manufactured by the strong acid process)
3. Saccharin (only if it is being manu-
factured)
Using the same outsourcer to handle
materials in-house as well as for re-
moval and disposal eliminates inter-
organizational hand-offs and can fur-
ther reduce costs.
Communications between waste
generators and chemical disposal
companies. Before a waste disposal
company can remove hazardous
chemicals from a generator, all paper-
work must be in order, the materials
must be properly labeled and pack-
aged and ready to go. It is important
that chemical waste is properly char-
acterized in order to avoid unneces-
sary expense or, in a worst case, addi-
tional transshipping to another waste
disposal site. Confusion creates lost
time and increased costs for chemical
characterization tests and EPA and
DOT documentation.
Communications with the waste
disposal company are key. Direct links
between the generator and the waste
disposal company are efficient since
any materials or paperwork that are
not in compliance will be identified
early, waste profiles and DOT mani-
fests can be generated, and placards
assigned before the disposal company
arrives on site.
A direct link to the waste disposal
company can also report contain-
ers that are ready for packing and/or
pickup, enabling the removal company
to cost-effectively schedule pickups.
Automated procedures also help com-
panies stay in compliance with RCRA
90/180-day rules or special handling
requirements for peroxide formers
and other acute hazards.
Regardless of the degree of auto-
mation and communications, under-
 Revise #1 - 2/4/10
solvent Recycle and reuse
6:00 PM
S
ome waste disposal companies offer more sophisticated services, such as the ability
to recycle solvents. Examples of solvents that are typically recycled for reuse include
the following:
Centrifuge & Drying
1-Ethyl-2-pyrrolidone Methyl ethyl ketone
Technologies
Acetone Methylene chloride
Decalin (Decahydronaphthalene) N,N-Dimethyl acetamide Inverting Filter Centrifuge
Dimethyl sulfoxide Naptha
Ethanol solutions N-Methyl-2-pyrrolidone
(excluding specially denatured N-Propyl bromide
alcohols) Perchloroethylene
Ethyl acetate Petroleum distillates
Ethylene glycol Propylene glycol
Isoamyl alcohol Tetrachloroethylene
Isopropyl alcohol Tetrahydrofuran
Isopropyl alcohol (high water) Toluene
Methanol Xylenes (ortho-, meta-, para-)

Cutting edge centrifuge technology for

standing the chemical waste disposal
process minimizes costs up front by
smoothing the transition from the
MSA to disposal and reuse. Com-
munication with the waste disposal
company has become even more criti-
cal over the last few years due to in-
creased opportunities for chemical
reuse, recycling and tolling.
Reuse and recycle. Some hazardous
and non-hazardous waste and byprod-
ucts with residual value can be di-
rectly reused in other processes by an-
other manufacturer. In the past, these
materials may have been landfilled
or incinerated because it was less ex-
pensive and there were few disposal
alternatives. Today, both environmen-
tal and financial incentives encourage
materials reuse.
Beneficial reuse and recycling pro-
vide additional revenue streams, or
at least the potential for cost reduc-
tions, for the generator; however, they
add complexity to the removal-and-
disposal process. Rather than a one-
way transaction where the disposal
company removes and disposes of the
material, reuse and recycling may now
include waste removal discounts for
materials sold on the secondary mar-
ket. Tolling, where reprocessed ma-
terials are returned to the generator,
also results in more complex price ne-
gotiations based on fluctuating rates
for virgin materials.
Recycling, where the waste dis-
posal company reprocesses chemicals
for new uses, can also substantially
advance generators sustainability
performance. The materials stay out
of the waste stream and become feed-
stock for another companys process,
as outlined in Figure 1.
There are several other benefits
that accrue from the reuse and recy-
cling of chemical materials. In some
cases reuse and recycling operations
reclassify wastes, removing them from
the generators hazardous waste man-
ifest, which reduces the volume of ma-
terials listed in Toxics Release Inven-
tory (TRI) reports and advances waste
minimization goals. They contribute
to reduced taxes and annual fees as-
sociated with RCRA regulations (see
box titled TRI Chemicals) and other
corporate responsibility and environ-
mental reports.
Clearly, sustainable disposal, reuse
and recycling require a more sophis-
ticated service level from the waste
disposal company. Only a few are able
to analyze the waste with an eye to-
ward potential uses and the secondary
market. They also have knowledge of
the underlying regulations and access
to the separation and purification re-
sources necessary to reprocess the ma-
terials (see box titled Solvent Recycle
and Reuse). Companies that offer these
services along with in-house process-
ing, tolling and resale capabilities, are
best suited to develop comprehensive
waste management programs that
fully leverage sustainable secondary
uses for waste chemicals.
It makes financial and environmen-
tal sense to explore these opportuni-
ties with a waste processor.
Edited by Dorothy Lozowski
Author
Victor Belenchia is vice
president, chemical indus-
try for Clean Harbors (42
Longwater Dr., Norwell, MA
020619149; Email: belenchia.
victor@cleanharbors.com;
Phone: 7817925471; Web-
site: www.cleanharbors.com).
He holds a B.S. in biology and
chemistry from the University
of Central Connecticut and an
MBA in marketing from the
University of Connecticut. He spent more than
20 years in the chemical industry with Ameri-
can Cyanamid and Cytec Industries in various
technical and commercial positions and has been
with Clean Harbors for the last six years in his
current role.
filtration, washing and drying of solid/liquid
suspensions
t Widest range of applications - hardest to
easiest filtering products can be handled
t No residual heel for exact repeatable
batches and no loss of product
t PAC technology allows drying of the
product inside of the centrifuge
t Thin cake filtration operation allows for
improved quality and production rates
t Full containment eliminates operator
exposure
t Effective automated CIP
Kilo-Lab Conical
Vacuum Dryer-Mixer
Advanced technology
for real Kilo size
drying research
and development
t Utilizes interchangeable agitator systems
either orbiting arm screw or central shaft
t Flexible small scale volume of 150ml to
1500ml
t Plastic view through vessel available
t Designed for real laboratory requirements
of size, with full instrument & data
recording
t Direct scale up to production or pilot size
units
Horizontal & Vertical
Centrifuges
t Size ranges from 200mm to 1800mm
t Wide range of standard & custom designs
t Laboratory size equipment
Lab Testing Available
Rental & Lease Machines Available
www.heinkelusa.com
Tel: 856-467-3399
Circle 11 on p. 58 or go to adlinks.che.com/29257-11

Chemical Engineering www.che.com December 2010 47

 Cover Story

Hearing Protection
For CPI Workers &NQMPZFF
BVEJPHSBN

As statistics on rising rates 

-FGU

of noise-induced hearing loss



)FBSJOH
MFWFM
JO
E#
 3JHIU

in the workplace show, 



a hearing-conservation 


program alone is not enough. 



Workers must understand the risks, 



have access to hearing protection    

'SFRVFODZ
)[
 

and, most importantly, use it FIGURE 1. When provided with a copy of the audio-
gram showing a standard threshold shift (STS) in their
hearing, many workers are surprised: the decline in
hearing may be so subtle that the workers are unaware
Brad Witt of measurable effects to their hearing
Bacou-Dalloz

O
ccupational dangers in the
chemical process industries
(CPI) are many, with falls,
respiratory hazards, and
explosion risks confronted daily. One
hazard, however, causes no pain nor
visible scars. Its effects accumulate
with each overexposure and can go
unnoticed, often for decades. If statis-
tics are any indication, it is the most
common permanent and preventable
occupational injury at the plant. This
is by no means a silent hazard in
fact, this hazard is noise.
What makes noise-induced hearing
loss (NIHL) one of the most insidious
occupational hazards is that there are
no outward signs of damage. Loud
noise affects nerve cells deep inside
the ear, and once those nerve cells are
gone, they do not come back. NIHL is
permanent. There is no pill, surgery,
nor treatment to restore hearing that
has been lost to noise exposure.
Despite years of regulation, the prob-
lem of NIHL is growing. According to
the U.S. National Institute for Occu-
pational Safety and Health (NIOSH;
Washington, D.C.), approximately
30-million workers in the U.S. are ex-
posed to hazardous noise on the job,
and the hearing of an additional nine
million is at risk from other agents,
including ototoxic chemicals, such as
solvents and heavy metals [1]. Indeed,
according to the U.S. Centers for Dis-
ease Control and Prevention (Atlanta,
Ga.), NIHL is the number one hidden
disability [2] and the most common
occupational illness in North America.
It is also on the rise globally, with the
U.N. World Health Organization (Ge-
neva, Switzerland) calling it the most
common, permanent and preventable
occupational injury in the world.
Hearing hazards
Noise hazards in the CPI are typically
continuous (as opposed to intermit-
tent or fluctuating), and can be caused
by processing units, transfer pumps,
high-speed mixers, grinders and pul-
verizers, blowers, dryers and other
equipment.
The U.S. Occupational Safety and
Health Administration (OSHA; Wash-
ington, D.C.) defines 85 dBA (decibels
A-weighted) as the noise level at which
precautionary measures must be im-
plemented, or the OSHA Action Level.
It is not uncommon, however, for noise
levels in close proximity to equipment
at a chemical plant to reach 100 dB or
higher nearly 30 times the OSHA
Action Level1.
Growing evidence shows the CPI to
be particularly vulnerable to hearing
loss. Exposures to certain compounds
(those containing lead, toluene, n-butyl
alcohol and carbon monoxide), alone,
have been linked to increased hearing
loss. It is now known that these ex-
posures also combine synergistically
with noise to cause increased hearing
loss. Individual susceptibilities appear
quite varied, and the offending noise
levels are difficult to generalize. Hear-
ing professionals recommend increas-
ing the frequency of audiometric test-
ing for noise-exposed employees who
are also exposed to significant levels
of ototoxic solvents, heavy metals and
other chemical compounds.
hearing regulations
In the U.S., OSHAs Occupational
Noise Exposure Standard and Hearing
Conservation Amendment (29 CFR
1910.95) outlines a rigorous, worker-
oriented program to mitigate risks of
1. Noise is measured using a logarithmic decibel
(dB) scale. A 3 dB increase represents a doubling
of the sound energy

34 Chemical Engineering www.che.com March 2007

34-38 CHE 3-07.indd 34 3/1/07 6:57:30 PM


FIGURE 2. Combining new material with
an entirely new design concept, these sin-
gle-use earplugs offer users a range
of benefits, including multiple
attenuation levels, improved
fit and comfort, easier inser-
tion without rolling, and the
ability to block loud noise
while allowing voice fre-
quencies to be heard
more naturally
FIGURE 3. Multiple-
use earplugs, such as
these, use body heat to adapt the earplugs
to the shape of each wearers ear canal.
This innovative technology provides op-
timum protection and comfort for longer
wear time and increased compliance
noise exposure in the workplace. It be-
gins with engineering and administra-
tive controls on equipment and manu-
facturing processes to abate noise. If
those controls are impractical or infea-
sible, OSHA mandates that personal
protective equipment must be provided,
and that employers shall administer a
continuing, effective hearing conserva-
tion program when worker exposure
reaches damaging levels.
Heres a look at the five main com-
ponents of an OSHA-approved hear-
ing conservation program and some
practical advice on implementation.
Noise monitoring
The first critical step to a hearing
conservation program is measur-
ing noise. Monitoring can be accom-
plished with sound-level meters,
which measure area sound levels, or
with personal dosimeters, which re-
cord an individuals noise exposure
over time. While area sampling pro-
vides a good understanding of general
noise levels throughout a facility, per-
sonal sampling documents a workers
complete exposure during the course
of the entire workday. Since workers
move about during the day and can
34-38 CHE 3-07.indd 35
be exposed to a variety of noises, do-
simetry provides the best indication
of an individuals time-weighted av-
erage (TWA) exposure.
Noise monitoring can be performed
in house, but many employers choose
to contract with outside consultants.
Some employers utilize their liability
insurance carriers to perform this ser-
vice and make the appropriate recom-
mendations. OSHAs consultation ser-
vices also provide noise monitoring at
no cost to many employers.
Whichever path you decide to
choose, here are several points to keep
in mind:
Perform noise monitoring any time
there is a noticeable change in ma-
chinery or manufacturing processes
Identify noisy areas and post a
noise map of the facility in an easy-
to-access location to give workers a
visual reference to noise-hazardous
sites. Posting specific decibel levels
in work areas is also an effective
way to identify the need for hearing
protectors and help workers select
the appropriate personal protective
equipment
Include the TWA of an employees
noise exposure in his job description
and/or employment file. This can help
an audiologist better understand a
workers occupational noise history
when interpreting audiograms
Audiometric testing
Under OSHA regulations, new em-
ployees who are exposed to hazardous
noise must undergo an audiometric
test within six months of employment,
and all noise-exposed workers must be
tested on an annual basis. Individual
test results are compared from year to
year to determine if hearing has re-
mained stable.
Audiometric testing must be con-
ducted by a trained professional, and
can be done in house or outsourced to
industrial clinics and audiology prac-
tices. Many practices offer mobile test-
ing services that are economical and
can accommodate employee schedules
especially if a company works three
shifts to minimize worker down-
time. One source list of competent
mobile-testing services is maintained
by the National Hearing Conservation
Association (Greenwood Village, Colo.;
Chemical Engineering www.che.com March 2007
www.hearingconservation.org).
A good idea is to use an audiomet-
ric testing provider that can combine
required employee training with the
annual audiometric test.
Make certain that your testing ser-
vice provides the required baseline
comparisons, and that the follow-up re-
ports are understandable. Often these
providers can send out notifications as
required by the OSHA standard. Re-
search has shown that when employ-
ees receive copies of their audiograms
(see Figure 1), or explanations of the
results at the time of testing, hearing-
protector usage increases and rates
of noise-induced hearing loss in that
workforce decrease.
Hearing protection
OSHA requires employers to provide
employees with a variety of suitable
hearing protectors, whenever noise
levels meet or exceed the 85 dBA Ac-
tion Level. Workers with normal hear-
ing are required to wear hearing-pro-
tective devices (HPDs) when exposure
levels exceed 90 dBA, while work-
ers with an identified impairment
(a standard threshold shift, or STS)
must wear HPDs when noise exposure
reaches 85 dBA.
While OSHA does not define the
term variety, it is a good practice to
provide a robust selection of HPDs.
Everyones ears are different, and one
earplug or earmuff style may not be
comfortable for an entire workforce.
A wide variety of HPDs are available
to meet specific applications and/or
worker preferences, including ear-
plugs packaged in paper bags for pro-
cess industries, banded earplugs that
can be inserted quickly during inter-
mittent noise, and dielectric and cap-
mounted earmuffs.
Offer workers several different
styles, including single-, multiple-
use and banded earplugs (see Fig-
ures 2 and 3), as well as earmuffs
(Figure 4). Dielectric earmuffs are
made of materials with negligible
electrical and thermal conductiv-
ity. Cap-mounted earmuffs (Figure
5) offer snap-in adapters that allow
them to fit most hard hats, and are
height adjustable.
Including a group of workers from
different areas in the selection process
35
3/1/07 6:58:10 PM
 Cover Story
is a good idea to improve worker buy-
in and compliance. Here are some ad-
ditional thoughts:
Make hearing protectors available.
Compliance begins with access; if
workers cannot easily obtain a pair
of earplugs or earmuffs on the job-
site, they will most likely go with-
out them. Simple actions such as
placing single-use earplug dispens-
ers (Figure 6) by main entrances to
the jobsite, by the time clock, in the
cafeteria or locker room, or keeping
earmuffs at a supervisors worksta-
tion make access to HPDs more
convenient and encourage compli-
ance. Purchasing agents should be
kept in the loop to ensure an ade-
quate supply. If HPDs arent avail-
able, they cant be worn
Praise workers who always wear
their HPDs. Workers who are pub-
licly recognized are more likely to
wear their earplugs or earmuffs
and wear them properly than
those chastised for improper use
Encourage workers to take extra
earplugs home. NIHL is not solely a
workplace issue: it can also happen
off the job. Many workers use power
tools, attend loud rock concerts or
sporting events, or participate in
shooting sports, which are all oppor-
tunities for exposure to hazardous
noise levels. Prevention is the key,
on and off the job
Set a good example. Include all lev-
els of management in the audiomet-
ric testing program and make sure
managers/supervisors are pro-ac-
tive in wearing protection, even if
they are only in noise-hazard areas
for a few minutes. When employ-
ees see the managers taking their
hearing seriously, it sends a positive
message throughout the workforce
Training
Research has confirmed that one-
on-one training is more effective
than group training. There is no bet-
ter time to motivate workers to use
hearing protection than immediately
after their audiometric test, and the
subsequent explanation of results.
Mobile audiometric services often re-
quest that workers bring their hear-
ing protectors at the time of testing,
so that an individual fit check can be
36 Chemical Engineering www.che.com March 2007
34-38 CHE 3-07.indd 36
FIGURE 4. Designed to provide reliable FIGURE 5.
protection for intermittent or marginal noise Cap-mounted
hazards, earmuffs with an ultraslim-earcup design earmuffs fit
offer protection with less weight and bulk than most hard-hat
standard earmuffs. This particular design is available brands, and
in both folding and neckband models to provide noise- are height
rated protection of NRR 23 and NRR 22, respectively adjustable
performed. Improperly fitted earplugs instructions, noise thermometers and
provide little to no protection. posters that demonstrate the effects
Training must include information of NIHL. Many HPD manufacturers
on the effects of noise exposure; the offer such posters for free and can pro-
use, selection and fitting of hearing vide additional aids to reinforce the
protectors; and audiometric testing importance of hearing protection and
procedures. Audiologists, audiometric conservation.
technicians, occupational hearing con-
servationists or even videos are good Recordkeeping
resources to perform this training, but As a rule, government regulations re-
it is the safety officers responsibility quire adequate recordkeeping, and 29
to see that training occurs in a timely CFR 1910.95 is no exception. Record-
manner. keeping also just makes good sense:
Although only workers exposed to the without accurate records, you have no
Action Level and above are required to way of assessing progress or rectify-
participate, everyone in the organiza- ing problems. Thus, accurate records
tion can benefit from hearing conser- of noise exposure levels and of all
vation training. While some workers audiometric tests are necessary. And
may not be exposed to hazardous noise dont forget to post a copy of OSHAs
levels on the job, they may still be at hearing regulations at the worksite.
risk elsewhere. Hearing conservation Nearly three-fourths of OSHAs re-
training helps everyone recognize haz- cordkeeping violations in hearing con-
ardous noise situations at work and at servation in the past five years were
home, and encourages them to protect for simply not having a copy of the
themselves appropriately. hearing conservation standard posted
A number of available resources in the workplace.2
can be utilized for training. Ideally, a
training program will include motiva- Motivation
tional materials to help make the case The safety-savvy employer will know
that workers are susceptible to noise that motivation is the keystone to pre-
damage, demonstrate the future risk venting NIHL. Motivated employees
of hearing loss, and encourage work- take responsibility for protecting their
ers to wear hearing protectors. hearing both on and off the job, rather
In addition to formal training, it is than just viewing hearing protectors
helpful to hang motivational and in- as a workplace compliance issue. There
formational posters in common areas
or near hearing protection sources. 2. Posters of the standard are available from
several manufacturers, including that of the
These visual aids can include fitting author
3/1/07 6:58:57 PM

FIGURE 6. Earplug dispensers
provide an easy and convenient
way for workers to access hearing
protectors. Studies have shown
that workers are more likely to use
hearing protection when it is conve-
niently available
are three ways to instill the level of
risk awareness that motivates work-
ers to take proper defensive action by
wearing their hearing protectors.
The three motivators
Noise affects everyone. The first
thing to do is to dispel the illusion of
invulnerability. When it comes to ig-
noring hazardous noise, its not only
the young who feel invulnerable. Even
seasoned workers will claim that the
loud noise does not bother them be-
cause they are used to the noise.
While the brain may grow accustomed
to constant noise, anatomically, the
ears can never toughen up against
hazardous noise. Ears respond by los-
ing hearing.
Dispelling the illusion can be done in
several ways. One of the most effective
is to show workers exactly how noise af-
fects them. Several studies have found
that the incidence of NIHL drops sig-
nificantly when workers are provided a
copy of their annual audiometric tests,
with an explanation of the results. It
is hard to argue with an objective test
that historically shows the progression
of hearing loss from year to year.
Look ahead. Demonstrating fu-
ture risk also helps bring home the
need for hearing protection. We live
in a culture of the here and now, but
NIHL typically develops over years of
exposure. Many employers use audio
demonstrations to simulate hearing
loss, so that the worker has a clear un-
derstanding of the future risk and the
need for adequate protection today.
Another effective tool is to have em-
ployees talk to older workers who have
suffered hearing loss and regret their
disregard for hearing protection.
Make it simple and comfortable.
The third motivating tool is to re-
move the barriers to wearing hearing
protection. This can be as simple as
ensuring an adequate supply of ear-
plugs, but oftentimes the barriers run
deeper. Studies have shown that work-
ers resist wearing hearing protection
if the devices are not comfortable, or
if they interfere with communication
and job performance (Figure 7).

Excellence in solids-liquid separation and vacuum drying

PRESSOFILTRO No heel thanks to

Nutsche Filter-Dryer Whirl Snake
High containment high efficiency turbine
OEL < 1 g/m

COMBER COMBER USA, INC.

Agrate Brianza (Milan) / Italy Charlotte, North Carolina
Ph.: +39 039 64214.1 Ph.: 704 393-5322
Fax.: +39 039 64214.20 Fax.: 704-393-5374
E.mail: salesdept@comber.it E.mail: info@comberusa.com
Circle 63 on p. 78 or go to adlinks.che.com/6893-63 Circle 64 on p. 78 or go to adlinks.che.com/6893-64
Chemical Engineering www.che.com March 2007 37

34-38 CHE 3-07.indd 37 3/1/07 7:00:16 PM

 Cover Story
It is hard to blame workers for
being dissatisfied with their hearing
protectors when the only choices of-
fered are earplugs that overprotect
and isolate the worker. Todays hear-
ing protectors address workers con-
cerns of overprotection and communi-
cation interference.
Earplugs offer different attenu-
Circle 65 on p. 78 or go to adlinks.che.com/6893-65
38 Chemical Engineering www.che.com March 2007
34-38 CHE 3-07.indd 38
ation ratings3 for different noise
levels. Some are made of special
high-tech material that conforms to
the shape of the wearers ear canal.
Several earplugs and earmuffs have
been designed to maximize com-
3. The noise reduction rating (NRR) represents the
amount of attenuation, in decibels, that the device
can be expected to provide when properly fitted
FIGURE 7. Although noise reduction
is one factor to consider in selecting
appropriate hearing protection, it is not
necessarily the most important. Other
factors include comfort, size, and com-
patibility with other personal protective
equipment such as hard hats and safety
eyewear
munication through uniform at-
tenuation, allowing wearers to hear
important sounds more naturally
while still protecting from harm-
ful noise levels. Electronic earmuffs
can also enhance communication by
amplifying ambient sounds (includ-
ing speech and warning signals) to a
safe level, while protecting against
louder, more damaging noise
Edited by Dorothy Lozowski
References
1. U.S. Dept. Of Health And Human Services,
National Center For Health Statistics,
Hyattsville, Md., Summary Health Statistics
for U.S. Adults:National Health Interview
Survey, 2002, DHHS Publication No. (PHS)
2004-1550 Vital and Health Statistics, Series
10, Number 222, July 2004.
2. Franks, J.R., Stephenson, M.R., Merry, C.J.,
eds., U.S. Department Of Health And Human
Services Public Health Service, Centers for
Disease Control and Prevention, National In-
stitute for Occupational Safety and Health,
Division of Biomedical and Behavioral Science,
Physical Agents Effects Branch, Preventing
Occupational Hearing Loss A Practical
Guide, June 1996.
Author
Brad Witt, MA, CCC-A is
currently audiology and regu-
latory affairs manager for the
Bacou-Dalloz Hearing Safety
Group (7828 Waterville Road,
San Diego, Calif.; Phone: 800-
430-5490; Email: BWitt@
bacou-dalloz.com). He has
been an industrial audiologist
for 23 years, is a former presi-
dent of the National Hear-
ing Conservation Association
(NHCA), and was recently awarded the Out-
standing Lecture award by the NHCA. Prior
to joining Bacou-Dalloz, Brad served as occupa-
tional audiologist for Saudi Aramco, the worlds
largest oil company, in Dhahran, Saudi Arabia,
where he managed the hearing conservation
program for 12,000 noise-exposed employees.
Brad also served for fourteen years as a safety
consultant to industry throughout the western
U.S., providing audiometric testing, monitoring
noise levels, and teaching employee/manage-
ment training classes at over 175 locations. A
frequent writer and speaker on the subject of
hearing conservation, Brad Witt is a member of
the American Speech-Language-Hearing Asso-
ciation, and holds degrees from Brigham Young
and Northwestern Universities.
3/1/07 7:00:57 PM
 Feature
Cover Story
Report
Designing for
A Safe Process
Phil Leckner
CH2M Hill Lockwood Greene
A
ccidents do happen. While not
everything can be predicted,
addressing safety concerns
throughout the design of a pro-
cess can help to prevent accidents
from occurring. Designing with safety
in mind can also help to minimize po-
tentially serious consequences that
would result if an accident did occur.
On April 12, 2004, toxic allyl alcohol
and allyl chloride were released from
a reactor at a facility in Dalton, Ga.
The consequences included injuries
and chemical contamination to people
and property in the surrounding area.
According to their report [1], the U.S.
Chemical Safety and Hazard Inves-
tigation Board (CSB) concluded that
better process design, engineering,
and hazard analysis would likely have
prevented the 2004 runaway chemi-
cal reaction and toxic vapor cloud re-
lease...
On March 23, 2005, an explosion at
a refinery in Texas City, Tex. killed 15
workers and injured 180 others when
flammable liquid and vapor overfilled
a blowdown drum during the startup
of the refinerys isomerization unit [1].
All of the fatalities and many of the
injuries occurred in and around trail-
ers that had been positioned near the
isomerization unit to support mainte-
nance activities on adjacent refinery
units. The CSB report on this incident
[1] recommended that new guidelines
be developed for the placement of
these and similar temporary struc-
tures around hazardous areas.
Having a procedure in place to pro-
mote safe process design can help
ensure that safety concerns are con-
sidered at appropriate phases in the
design. The methodology presented
here, as outlined in Table 1, may help
prevent accidents such as those in
30 Chemical Engineering www.che.com December 2006
30-33 Ce 12-06.indd 30
'FFE %JTUJMMBUJPO
)1
TUFBN
$POEFOTBUF 3FCPJMFS
#PUUPNT
&YBNQMF
PG
B
CMPDL
GMPX
EJBHSBN
Figure 1. Block flow diagrams give a
simple overview of the major steps in a process
Dalton, Ga. and Texas City, Tex. This
methodology applies to process de-
signs throughout the chemical process
industries (CPI), which include not
just chemical production, but also, for
example, wastewater-treatment fa-
cilities, pharmaceutical and food-and-
beverage plants. While following these
guidelines will cost time and money,
the practice can be a very inexpensive
way to help prevent the much more
costly consequences of not providing
the safest design possible.
Basic Engineering
Process flow diagrams
As a process engineer, there are two
types of flow diagrams that interest
me: the block flow diagram (BFD) and
the process flow diagram (PFD).
The BFD (Figure 1) presents an
overall picture of the process, show-
ing only major process steps. These
steps are shown as black boxes with
simple descriptions. Equipment can
be depicted singly or grouped together
as a system.
The PFD (Figure 2), meanwhile,
depicts major and minor equipment
with specific symbols that are typi-
Part 1
Incorporating safety
considerations
throughout process
design lowers the risk of
a hazardous event
-1
TUFBN
8BUFS
$POEFOTFS
7FOU
3FDFJWFS
1VNQ
%JTUJMMBUF
cally used in the CPI. Equipment is
usually identified and shown with an
alphanumeric designation. The PFD
includes major and some minor pro-
cess streams as well as utility streams,
such as steam, condensate and cooling
media. This diagram can also be used
to show process safety requirements,
such as proposed locations for relief
valves. Often, a heat-and-material
balance and major control loops are
included.
The PFD should be used as the basis
for generating the more detailed pip-
ing-and-instrument diagram (P&ID).
Prior to P&ID preparation, the PFD is
reviewed by the design team and is-
sued for design (IFD), with a revision
number of zero.
At a minimum, the basic design
package should include the BFD and
the PFD. However, I believe that an-
other document, which takes the PFD
to a new level, better promotes a safe
process design. This document, which
I call the process definition drawing
(PDD), is not ordinarily a part of the
basic design package, but is a great
tool for the process design engineer.
The PDD includes operating-and-
11/22/06 4:35:40 Pm

&YBNQMF
PG
B
QSPDFTT
GMPX
EJBHSBN
'FFE
%8(
/P  1*$
5*$
'*$
R 5
)14
Y
MCIS
)&
-*$
5
)1$
%JTUJMMBUF
%8(
/P  BFDs, and include at least
design conditions for each equipment
item, each control-valve station and
all relief devices. The benefit of this
document is that the process engineer
can quickly recognize inconsistent or
conflicting operating-and-design con-
ditions. It helps the engineer think
about the process in terms of how
it will operate and what needs to be
done to make it safe.
Detailed engineering work is also
aided by the PDD, since much of the
information included on this docu-
ment will be used in the generation of
process-equipment, instrumentation
and safety-device duty specifications.
The PDD is a living document and
will change as the process design pro-
gresses. As such, it is often sketched
using rudimentary drawing software
rather than CAD, thereby making it
simple to construct and maintain.
Preliminary safety review
After the PFDs are created and ap-
proved for design, a preliminary safety
review (PSR) is undertaken.
The PSR is the gathering of docu-
mentation with an emphasis on pro-
cess safety. Included are items such as
the project scope definition, the pro-
cess design basis, the process descrip-
tion and a material safety data sheet
(MSDS) for each substance used. An
extensive list of documentation that
may be included in the PSR can be
found in Ref. [2].
The MSDS is an important docu-
30-33 Ce 12-06.indd 31
'-
5 R
'*$
Y
MCIS
1
"
-*$
1
'*$
1
ment that provides a wealth of in-
formation including the proper han-
dling of a substance, special storage
requirements (such as keep out of the
sun) and required personal-protection
equipment (PPE), such as breathing
apparatuses.
The MSDS also provides basic phys-
ical-property and toxicity data, expo-
sure limits and flammability ranges.
It may describe what to do in case of
spills. The manufacturer or supplier
of the raw materials and various web-
sites on the Internet are sources for
MSDSs. If the facility is producing a
finished product, then the plant owner
will have to develop an MSDS for
that product and make the document
available to potential users. Table 2
lists information typically found in an
MSDS.
Once the documentation is gath-
ered, the PSR is assembled into a
formal report and issued to all perti-
nent members of the design team for
comment. The report is given to proj-
ect management for distribution as a
revision 0 issue. Note that the PSR
is a living document and is subject to
change as the design progresses. The
PSR document is issued as soon as
feasibly possible because it will form
the foundation for the balance of the
safety review of the project.
Design safety review
Once the PSR is issued, the process
can be reviewed for major safety con-
Chemical Engineering www.che.com December 2006
Y
MCIS
-14
-*$ -1
#'8
)&
#%
'-
1 5
/
5
#PUUPNT
 %8(
OP
Figure 2. Process flow dia-
grams give more detail than
major process streams and
equipment items. Symbols that
are typical for the CPI are used
cerns in the design safety review.
Using the most up-to-date version
of the PSR, all features associated
with safety, environmental and layout
issues are reviewed, including the fol-
lowing:
Defining the hazardous location
classification (HLC), or the electri-
cal area classification as it is some-
times called The HLC is used to
determine electrical-design criteria,
such as equipment that may require
explosion-proof motors. The HLC
boundaries should be shown on ap-
propriate documents such as the
PFD or equipment-layout drawings
(if available). One source of defini-
tions for HLCs can be obtained from
Ref. [3].
Locating major pressure-relief de-
vices, such as relief valves and rup-
ture disks, explosion panels and
flame arrestors These devices
should be indicated on the PFD
and, in more detail on the PDD. The
destination of the vent from these
particular safety devices needs to
be considered. The applicability and
use of alternate safety systems, such
as safety-instrumented systems
(SIS) is also evaluated at this time.
(For more, see Part 2 of this report,
p.34)
Evaluating the layout with respect
to minimizing hazards As noted
earlier in the Texas City, Tex. ac-
cident, personnel placement is a
very important consideration. For
31
11/22/06 4:36:16 Pm
 Cover Story
equipment placement, the American
Petroleum Institute (API; www.api.
org) recommends that all equipment
with a ground area of 2,500 to 5,000
ft2 should be considered part of the
same fire-relief scenario [4]. Deter-
mine if larger-sized equipment can
be moved outside the common fire
zone to reduce overall relieving ca-
pacity. Also be cognizant of where
chemicals are stored in proximity to
each other. You may not want acids
stored near bases, for example.
Locating flares in safe areas, taking
into account their radiation effects
at ground level.
Evaluating the need for and place-
ment of, fire and gas detectors (for
more, see p.18).
With the major pressure-relief de-
vices located and the operating and
design conditions fully defined, it is
appropriate to evaluate modifications
that might make the system safer.
For example, determine if there are
safety advantages to changing stor-
age requirements from concentrated
solutions to more dilute solutions, or
vise-versa. Evaluate the advantages
and disadvantages of breaking unit
operations into smaller, more discrete
pieces to make process equipment,
such as heat exchangers and reactors,
smaller.
Preliminary hazard analysis. Fi-
nally, some type of preliminary hazard
analysis (PrHA) should be undertaken
before moving into the detailed pro-
cess engineering phase of the project.
During a PrHA, team members vi-
sualize ways in which a process design
can malfunction or be operated incor-
rectly. The PrHA can take one of many
forms, such as a pre-HAZOP, a what
if, failure mode-and-effects analysis
(FMEA) or FMEA check list. The vari-
ous types of hazard analyses that are
acceptable to the U.S. Occupational
Safety and Health Administration
(OSHA) are outlined in Refs. [2] and
[5]. The PrHA is performed on the de-
tailed PFD that is issued for design.
The documents collected during the
PSR and the PDD provide reference
materials. Note that a preliminary or
IFD issue of the P&ID may have been
developed by this time and if so, would
be part of the PrHA.
At the conclusion of the design safety
32 Chemical Engineering www.che.com December 2006
30-33 Ce 12-06.indd 32
review, all documents and results are
collated and issued. This compilation
is included as part of the front-end de-
sign package and the design proceeds
to the next phase of the project, the
detailed process engineering.
Detailed Engineering
Piping and instrument diagram
At the start of the detailed process-
engineering phase of design, all out-
standing issues and especially those
brought up during the design safety
review are examined and addressed.
The P&IDs, PFDs and the PDDs are
updated as required. The P&IDs are
then reviewed in a formal setting as
a team, which should include the pro-
cess, mechanical, and piping and in-
strumentation engineers.
The review will expose any last min-
ute safety and design issues that must
be addressed before the more detailed,
and required, process hazard analysis
(PHA) is undertaken. Changes are
documented by again updating the
P&IDs and PFDs. I suggest that these
documents be issued with a separate
revision number established exclu-
sively for the PHA.
The process hazard analysis
The PHA evaluates the design in
terms of both safety and operability.
The analysis should be performed on
the process, as well as instrumenta-
tion and control systems, such as the
digital control system (DCS).
PHAs are mandatory for all plants
that fall within the scope of OSHA 29
CFR 1910.119 [5]. This scope applies
to plants that meet the following two
criteria:
1. Those whose processes involve one
or more of certain chemicals (listed
in appendix A of the regulation), and
in quantities at or above the thresh-
old given
2. Those whose processes involve flam-
mable liquid or gas onsite in one lo-
cation in a quantity of 10,000 lb or
more with the exception of:
a. Hydrocarbon fuels used solely for
workplace consumption as a fuel
b. Flammable liquids stored in at-
mospheric tanks and kept below
their normal boiling point with-
out the aid of chillers or refrigera-
tion
Table 1.
A methodology to promote
a safe process design
I. Basic process engineering
Create the process flow diagram
Perform a formal review of the pro-
cess flow diagram
Conduct a preliminary safety re-
view
Perform a design safety review
II. Detailed process engineering
Create the piping and instrumenta-
tion diagram
Conduct a formal P&ID review
Perform the process hazard analy-
sis on the process
Perform the process hazard analy-
sis on the control system
III. Implement a management of
change procedure
Facilities that do not fall within 29 CFR
1910.119 would still benefit greatly
from the PHA if not for the safety as-
pects of the process, then for process
operability. I cant stress this enough:
not being required to do a PHA by law
should not exclude you from doing one
to ensure that your process is indeed
safe in design and operation.
The team make-up. The PHA team
should include a facilitator and a scribe
(who serves the function of record-
ing secretary), the design firms area
process engineer, the plant process
engineer and representatives from
the plants safety-and-environmental
and operations-and-maintenance de-
partments. If vendor packages are in-
volved, a vendor representative is also
advised as a team member. An exten-
sive list of possible participants can be
found in Ref. [2]. The PHA team make-
up and the extent of their efforts will
vary based on the nature and complex-
ity of the process design.
The facilitator should be chosen
with the following criteria in mind:
The person should be knowledgeable
in the type of PHA to be performed
The person should not be intimately
involved in the process design (thats
what the process engineers are for)
The person does not necessarily need
to be very knowledgeable about the
particular process to be reviewed
The facilitators function is to guide
11/22/06 4:37:10 Pm
 Table 2. INFORMATION typically found in an msds
Item Remarks
Chemical name and any The common name will be the same name on
common names the label
Date of preparation or revision The date the MSDS was prepared or revised
List of contacts that can pro-
vide more information
Physical characteristics Includes smell, color, appearance, flash point
and vapor pressure
Physical hazards For example, if the substance is subject to vio-
lent reactions such as explosions or fires
Health hazards Describes if and how a substance can cause
harm to human health and also provides
symptoms of exposure
Route of entry information Describes how the substance can enter the
body, for example by ingestion or inhalation
Exposure limits The maximum amount of exposure a person
should have to the substance
Carcinogenic status Whether the substance causes cancer
Safe handling and use Explains precautions and protective measures
needed when using and handling, including
spill control
Control measures Suggested engineering controls, work practices
and personal protective equipment
Emergency and first aid pro- How to deal with releases and exposure
cedures
the team, keep the PHA on track and on the earlier steps mentioned in this
motivate participation. The facilitator methodology can reduce the cost and
is not the person who brings up all of duration of the PHA.
the issues. A facilitator who is too in- Documenting the PHA. One com-
volved with the process is analogous mon problem is that many PHAs al-
to someone proofreading his or her most become P&ID reviews and fix it
own material things are going to be sessions. Any problems identified with
missed that would otherwise not. the P&ID that need fixing should be
Preparation. Having the right docu- recorded for further action and not
mentation available is key to a smooth discussed in detail during this analy-
PHA. The team should at a minimum sis. Note that once the P&ID fix is
have all the documents from the PSR implemented, the change needs to be
and the DSR, as well as the P&IDs re-evaluated.
and PFDs, plot plans, equipment lay- After the PHA is completed, a re-
outs, hazardous classification draw- port is issued documenting what was
ings, operating-and-maintenance pro- checked and any actions that need to
cedures, batch sheets (if applicable), a be addressed, such as additional relief
summary of relief-device calculations devices, changing instrumentation
and specifications for equipment, in- and adding information to what will
strumentation and piping. eventually become standard operating
A PHA can be very time consum- procedures. These action items are to
ing and expensive to implement. Poor be addressed in a timely manner and
documentation and the failure to per- the plant design should be revised as
form a comprehensive P&ID review required. Once all issues are addressed
prior to the PHA can contribute sig- and design changes implemented, the
nificantly to the cost. More time spent design becomes fixed in terms of
References
Practice 521, Guide for Pressure-Relieving
1. The U.S. Chemical Safety and Hazard Inves-
tigation Board website (www.csb.gov) and Depressuring Systems, 4th ed., March,
1997.
2. The Center for Chemical Process Safety,
Guidelines for Hazard Evaluation Proce- 5. The Occupational Safety and Health Ad-
dures with Worked Examples, 2nd ed., ministration, Process Safety Management
AIChE, 1992. of Highly Hazardous Chemicals, in 29 CFR
1910.119, OSHA, Washington, D.C., 1992.
3. National Fire Protection Association, NFPA
70, National Electric Code, Chapter 5, 6. The Center for Chemical Process Safety,
Plant Guidelines for Technical Manage-
2005.
ment of Chemical Process Safety (Revised
4. American Petroleum Institute, Recommended Edition), AIChE, 1992, 1995.
Chemical Engineering www.che.com December 2006
30-33 Ce 12-06.indd 33
safety. The design documents should
be issued as process safety manage-
ment approved.
More specifics on PHAs can be found
in Refs. [2] and [5] and in an abundant
number of books and government pub-
lications (OSHA), that the reader is
encouraged to research and review.
Management of change
As in all designs, things change even
after the PHA is completed. These
changes must be captured as they
can affect the safety of the design and
create new problems. Management
of change (MOC) procedures capture
these changes and should be strictly
followed.
A MOC procedure is simply a writ-
ten way of documenting and inform-
ing people about changes made to the
design after the PHA. The MOC may
even outline when a change neces-
sitates a new PHA. For example, if a
valve is added to a line that was not
there when the original PHA was per-
formed, a PHA must be performed on
this area of the design to ensure that
no additional safety hazards were
introduced, or if they were, are ad-
dressed accordingly.
There is no single way to design
and implement a MOC procedure. It
is up to the discretion of the project
team as to how it should be done. The
MOC complexity will depend on the
complexity of the process. The key is
to ensure that the MOC procedure is
easy to follow and that it allows easy
documentation of all required activi-
ties and includes a system for approv-
als by the appropriate project team
members [6]. n
 Edited by Dorothy Lozowski
Author
Phil Leckner is a senior
process engineer with CH2M-
HILL Lockwood Greene
(Phone: 732-868-2277; Email:
Philip.leckner@ch2m.com).
Phil has over 30 years expe-
rience in process design and
project engineering, and com-
missioning and startup for the
chemical, petrochemical, food-
and-beverage and bio-pharma-
ceutical industries. Over the
past 10 years, he has been deeply involved with
process-safety issues with emphasis on relief-
system design. He has been involved in a number
of PHAs, including serving as HAZOP and What
if? facilitator. Phil received his B.S.Ch.E. from
Lowell Technological Institute, which is now part
of the University of Massachusetts.
33
11/22/06 4:38:02 Pm
 Feature
Cover Story
Report
Heat Transfer
Fluid Leaks:
Break the
Fire Triangle
Extreme processing temperatures
present the greatest risk.
Know where leaks are most likely to
occur and how to prevent them
Conrad E. Gamble and Matthias Schopf
Solutia Inc.
M
anagement of process plants
requires the understand-
ing and application of the
fundamental tenets of safe
design, construction, and operation
of each facility. In that context, many
years of practical experience with sys-
tems that use very high-temperature
organic heat-transfer media has led
to much knowledge regarding de-
sign features to adopt and those that
should not be repeated.
For the media itself, the severe
stresses of extreme operating temper-
atures involved can sometimes nar-
row the number of candidate fluids to
one: the eutectic mixture of diphenyl
and diphenyl oxide (DP:DPO; Figure
1). This article shares best practices
that have been developed over the
years to ensure the safest possible use
of this heat transfer fluid (HTF). Ef-
fective methods to identify and deter-
mine the order of magnitude of leaks
are discussed. Meanwhile, a review of
the concepts necessary for effectively
maintaining a leak-tight system is
presented. And, for occasions in which
unexpected leaks do occur, a thorough
26 Chemical Engineering www.che.com December 2010
explanation of the fire risk is offered
and approaches that can minimize ig-
nition potential and property damage
are addressed.
Introduction
Basics of system requirements
Process plants require heating and
usually cooling capabilities. A typical
HTF system is composed of an energy
source, such as fired heaters or waste
heat recovery systems; pumps to force
the fluid flow; an expansion tank to
absorb the volume expansion of the
fluid; and last but not least, a heat
consumer. High-temperature heat-
transfer systems are usually closed
systems, and hence, a release of fluid
can typically only occur in case of ac-
cidents or malfunctions. Heat transfer
media used in such systems are usu-
ally water-and-steam, water based
fluids, mineral oils or synthetic heat-
transfer fluids specially designed for
this purpose. Based on their physical
and chemical properties, such fluids
imply certain handling risks in addi-
tion to the general risk of operating an
HTF system.
Besides classical process plants in
the chemical, petrochemical and plas-
tic industries, two other applications
have recently come into the focus by
utilizing large volumes of DP:DPO
HTFs (see box entitled Large volume
applications): electrical energy pro-
duction by concentrating solar power
(CSP) and converting natural gas into
synthetic oils in a process called gas to
liquids (GTL).
As with all HTF systems, the de-
sign must accommodate the volume
expansion of the heating fluid, which
is typically provided by an expansion
or surge tank of sufficient volume and
headspace. For plants using DP:DPO
near its maximum temperature of
400C (750F), the volume expansion
from ambient temperature is greater
than 30% [1]. The expansion vessel is
typically pressurized to keep the HTF
in liquid phase and is equipped with
a vent system to permit the exit of
nitrogen, degradation products, and
a portion of DP:DPO into the ullage
vent-collection system. The extensive
network of HTF handling piping, in-
struments, and vessels, combined with
 Large volume applications
CSP plants
Concentrating solar power (CSP) plants require a reliable heat-transfer fluid
to efficiently absorb the suns energy and transfer it to the steam generator.
The HTF is the single component of the system, which traverses the solar-
energy collector field (mirror arrays) and the power generation unit (power
block). The uninterrupted circulating flow of the fluid and thermal energy is
essential for system operation. Any resistance to flow reduces demonstrated
efficiency from theoretical levels, and interruptions in flow can take the entire
system offline.
The designs of existing and planned CSP plants that utilize DP:DPO heat
transfer fluid have similar features. Large circulation pumps capable of
delivering flowrates up to 10,000 gal/min or more provide the motive
force for circulating the fluid through field supply headers of decreasing
diameters extending to the farthest mirror rows. HTF flowrates through the
individual heat-collection element (HCE) rows are regulated to maintain
the desired HTF temperature in the return headers. The heated fluid is used
in large shell-and-tube heat exchangers to provide superheated steam,
which is then utilized in traditional power turbines. The HTF continually
circulates throughout the operational hours to support the movement of
solar energy from direct solar and thermal storage systems, if provided.

Figure 1. Extreme operating temperatures some-
times narrow the number of candidate fluids to one:
the eutectic mixture of diphenyl and diphenyl oxide
(DP:DPO)
GTL-processes
In the gas-to-liquids (GTL) process, natural gas is chemically converted
into fuel liquids. The base is the Fischer-Tropsch (F-T) process of converting
mixtures of hydrogen and carbon monoxide into liquid hydrocarbons. In
general, GTL technologies are three-step processes. The first step is conver-
sion of the natural gas into a hydrogen and carbon monoxide mixture usu-
ally called synthesis gas or syngas. The second step is a catalytic reaction
of this syngas into a mixture of hydrocarbons (F-T). Finally, the last step
is a cracking and conversion of the hydrocarbons into the desired final
products [2]. Due to high temperature requirements of some of the process
steps, DP:DPO heat transfer fluid is utilized for process heat control.

elevated vapor pressures, can increase
the potential for leaks from systems
if adequate design and maintenance
measures are not also incorporated.
Incident history of leaks, fires
Examination of the leak history from
DP:DPO handling systems indicates
that the primary sources are flanged
connections, flexible connectors or
rotary joints, and pump seals. Fire-
resistant gaskets used in DP:DPO
service can require substantial com-
pressive force on the gasket face. In
cases where insufficient flexibility is
provided in piping networks (Figure
2), the resulting force applied to the
flange pair reduces compression on a
portion of the gasket, leading to leak-
age. Such leaks can sometimes be ad-
dressed by tightening sufficiently, but
only as a temporary repair. Permanent
repair will require gasket replacement
and improved piping flexibility.
Stainless-steel flexible hoses were
first installed on the early solar en-
ergy generating systems (SEGS) in
the California Mojave Desert in the
1980s. Some of the original flex hose
remain in service today, while most has
been replaced over time. Experience in
the mode of failure finds that the hose
develops small cracks through which
the DP:DPO fluid seeps slowly at first.
The leakage can be visually observed
as a darkened area on the outside of
the hose assembly. Prompt mainte-
nance can permit the piping section
to be shut down and isolated for hose
replacement prior to a rupture. When
proper experience and care is applied
during hose installation, the hose can
be kept free of the torsion and mis-
alignment that reduce its service life.
It is important to follow hose manu-
facturer recommendations on inspec-
tion and replacement frequency, as the
hose has smaller wall thickness than
rigid pipe.
Rotary joints have the advantage of
using greater wall thickness for a more
robust joint, yet have the flexibility to
support the daily cycling demands of
the mirrored rows of heat collection
elements (HCEs). Designs can incorpo-
rate high-temperature, metal-to-metal
seals and can include provision for
injectable, graphite-based packing for
Chemical Engineering www.che.com December 2010
maintaining adequate sealing. Some
users, however, have reported that
some of the rotary joints failed more,
and without warning.
Whether from flexible hose or rotary
joints, the close proximity of these de-
vices to nearby ignition sources can
lead to the ignition, or even autoigni-
tion of a released cloud of HTF mist if
surface temperatures exceed 1,100F
(593C). For safety of personnel, re-
sponse typically involves remote isola-
tion of the piping circuit to allow the
fire to burn itself out in remote or non-
congested areas. Alternate measures
may be appropriate to protect against
peripheral damage in more congested
areas. Extinguishing the fire prior to
stopping the HTF release may result
in later, repeated ignition. Operating
practices must incorporate the fre-
quent monitoring of the integrity of
field piping systems through regular
inspections and/or remote video sur-
veillance, or both.
Pumps in DP:DPO service can have
single or double mechanical seals.
Some of the pumped liquid may be
required to adequately lubricate seal-
27
 Cover Story

ing surfaces during operation. Exces-

sive temperatures at the seal faces

Fo
can also vaporize the HTF resulting

in no lubrication and ultimately cause
mechanical damage to seal face mate-
rials. This mechanism can create par-
ticulate matter that erodes the seal
face and create a separation of seal
faces, thereby resulting in leakage. A
seal flush can be effective in minimiz-
ing excessive seal wear by removing
harmful corrosive or fouling deposits
with filtered fluid [3].
It should be noted that while infre-
quent insulation fires are experienced
in very-high-temperature organic
HTF systems, it is extremely rare
in DP:DPO systems. DP:DPO has a
lower relative-normal boiling point
than other high temperature HTFs,
permitting its more rapid evaporation
from soaked insulation before it can
be heated to its autoignition tempera-
ture (AIT) by the oxidation exotherm
within porous insulation. Cellular
glass insulation has proven most ef-
fective in interrupting the mechanism
of HTF insulation fires among most
organic HTF chemistries.
Consequences of leaks, fires
There are no positive consequences
of leaks and fires, except perhaps the
heightened awareness of design or
construction flaws, which leads to im-
provement. The immediate negative
consequences include risks to human
life and health from released fluid
ignition and fire fighting response,
or inhalation of partially combusted
hydrocarbons. Personnel should be
trained and equipped to respond by
remotely isolating systems, and stay-
ing out of the path of smoke, vapors
and liquid runoff. In CSP plants, the
amount of fluid contained in a 100-m
mirrored row with 70-mm O.D. heat-
collection elements will be limited
to nearly 0.33 m3 (87 gal), plus the
volume contained in piping up to the
isolation valve(s). Much of the HTF
released from piping containment
will flash into the vapor phase, and
then quickly condense into a mist
cloud. The balance will rapidly cool
in contact with equipment and earth,
remaining in liquid phase. Contami-
nated soils will require disposal or
treatment per regulatory require-
28 Chemical Engineering www.che.com December 2010
rce
Reduced
n compression Fo
rc
io
u lat of gasket n e
Ins io
ct
fle
De
Fixed restraint
Figure 2. Poor piping flexibility can cause leaks
Table 1. DP:DPO Key Fire-Related Properties
Characteristic Method C F Vol. % in air
Flash point (open cup) ASTM D-92 124 255 ---
Fire point ASTM D-92 127 260 ---
Flash point (closed cup) ASTM D-93 110 230 ---
Autoignition temperature ASTM E-659 601 1,114 ---
Lower combustion limit --- 149 300 0.8
260 500 0.5
Upper combustion limit --- 149 300 3.3
260 500 6.2
ments, possibly including incineration Fluid Chemistry and
or bioremediation. Hot fluid released Inherent Properties
and collected should not be reused, as Formulation
it will have become oxidized and per- DP:DPO heat transfer fluids are a eu-
haps contaminated. Equipment wet- tectic mixture of diphenyl and diphenyl
ted with HTF condensate should be oxide. These two components exhibit
cleaned for proper housekeeping, and the highest thermal stability available
to remove residue that could poten- among organic heat-transfer media.
tially fuel a future fire. Their eutectic ratio ensures the lowest
In areas where leakage is of high- freeze point possible, thereby creating
est risk (such as flex hose or rotary the widest working range of tempera-
joints, pump seals, valves or instru- ture. As diphenyl and diphenyl oxide
ment manifolds, and so on) efforts exhibit tremendous thermal stability,
should be made to limit the amount the weakness of the specific fluid formu-
of instrumentation and wiring poten- lation would then lie in its impurities.
tially exposed to fire. Doing so will Therefore, it is most desirable to have
minimize the downtime required to very high assay of the two primary con-
return the equipment into operation. stituents, and limit the percentage of
Any equipment, piping, instrumenta- impurities possible. Doing so protects
tion, insulation or controls that have thermal stability of the fluid and also
been exposed to fire or DP:DPO spray ensures that the physical properties of
should be inspected by qualified engi- the HTF will have minimal deviation
neers and technicians for reliable op- from its key components.
erating integrity, prior to returning it Low chlorides content (< 10 ppm)
into service. ensures long life of stainless-steel

Figure 3. Pipe flanges are potential sources of leaks. Design considerations are
important here
system components without exces-
sive risk of stress-corrosion cracking.
There are SEGS systems, for example,
that have been utilizing low-chloride-
content heat-transfer fluid since the
1980s with no reported cracking is-
sues. Low chlorides content should
therefore be a requirement of DP:DPO
fluids used in plants to help maintain
longterm mechanical integrity.
Tendency toward leakage
Operators of high temperature, or-
ganic heat-transfer fluids should be
aware of the potential for leakage of
DP:DPO from components that may
be leak-tight when pressure tested
with water. It has been noted that,
the physical and solvent properties
of organic fluids allow the fluids to
penetrate ordinary valve and pump
packing and bleed through porous ma-
terials including some cast irons [4].
For piping flanges, fire-resistant gas-
kets are preferred, which can require
higher compression to adequately seal
as compared to softer gasket materi-
als. To help minimize leak potential
and keep the DP:DPO confined within
the system, engineers and safety pro-
fessionals have developed key prac-
tices as outlined later in the section on
preventing leakage.
When DP:DPO eutectics freeze,
the material contracts in volume by
over 6%. If the product then melts be-
tween frozen plugs of product or other
mechanical boundaries, tremendous
pressure can result, possibly leading
to release of the product through the
weakest constraint, such as flanged
connections, valve stems, pump seals,
and so on. When thawing a frozen sec-
tion of piping or equipment, it is very
important to accommodate the expan-
sion in volume into unobstructed pip-
ing or equipment.
Key fire-related properties
Key properties of interest in gauging
the potential for fire with organic heat
transfer fluid are provided for DP:DPO
in Table 1.
Minimum ignition energy (MIE)
is a measure of an amount of energy
below which an explosive mixture in
air will not be affected, and above
which the mixture can be affected.
Measured values for heat transfer
media are not readily found in pub-
lished literature. As T. H. Pratt ex-
plains, the MIE concept applies only
to capacitive spark discharge and
that, based upon numerous factors,
one must recognize that MIEs vary
with most everything [5]. Taking this
into account, NFPA 77-2007 states
that the MIE for most saturated hy-
drocarbon vapors is near 0.25 mJ, and
that mists can have MIE values one
or two orders of magnitude higher [6].
This value can have considerable de-
viation from the MIE experienced in
actual field environments.
Chemical Engineering www.che.com December 2010
Preventing leakage,
component selection
Reliable system design
Complete guidance on system reli-
ability cannot be adequately covered
within the constraints of this article;
however, many important concepts
are provided herein in support of de-
signers and operators. Economics of
projects often require use of the low-
est cost materials that are fully com-
patible for the service to be utilized.
Since the mid-1970s and the devel-
opment of the first low-chloride (<
10 ppm) formulation of heat transfer
fluids, stainless-steel metallurgy for
instrument tubing and specialized
equipment could be used without fear
of chlorides-related stress cracking
potential. It is important to note that
any significant extension of operating
temperatures above those currently
used in DP:DPO HTF systems would
require more exotic and costly alloys
for piping and equipment, as well as
the development of satisfactory, new
heat-transfer fluids. Useful reference
standards and codes are included in
the References section [714].
With the HTF being significantly
above its normal boiling point during
operation, adequate overpressure pro-
tection in the form of pressure relief
valves (PRVs) and rupture disks must
be provided with adequate capacity
for relief between selected points of
isolation. With proper design consid-
erations, the PRVs can be installed in
series with rupture disks to minimize
fugitive emissions from the system.
General good practices for HTF system
design are typically available from the
HTF manufacturer, including informa-
tion on system components, filtration
and expansion tank design.
Piping, flanges and gaskets
Seamless carbon steel has been demon-
strated as an appropriate piping mate-
rial for use with organic HTFs in plants
up to their maximum bulk operating
temperatures. Threaded fittings in pip-
ing are to be avoided in preference to all-
welded construction. Graphite-based,
paste thread sealants have demon-
strated marginal success with threaded
connections that cannot be avoided,
such as instrument connections, pump-
casing drain plugs, and so on.
29
 Cover Story
With any metallurgy employed, the
linear expansion and contraction of
the piping must be accommodated by
use of expansion loops or flexible con-
nection members, and pipe supports
should be generously spaced to pre-
vent sagging. When thin-walled expan-
sion joints are selected, manufactur-
ers recommendations for inspection
and replacement frequency should be
strictly followed. Piping should be pro-
vided with shoes, which can slide un-
restrained axially on the pipe supports
with thermal cycles. Where flanges are
necessary, Class 300 and Class 600
ring-joint flanges or raised-face flanges
are commonly used. While the number
of flanged pairs should be minimized
to reduce the number of potential
leak points and fugitive emissions,
designers should consider maintain-
ability and install such connections to
all major pieces of equipment and in
other key locations for ease of servic-
ing equipment (Figure 3).
When used with raised-face flanges,
gaskets should have a metal ring for
blow-out resistance and graphite-filler
within 316 stainless-steel, spiral wind-
ings for fire resistance. A number of
manufacturers offer such gaskets, and
some designs provide more-exotic-alloy
spiral windings for improved memory
and longer use-life for cycling tem-
perature services. A key necessity for
flanged connections is to ensure that
adequate and uniform sealing compres-
sion (seating stress) is provided on the
gasket faces. Data on compression force
of gasket surfaces are available from
gasket manufacturers, which will then
determine proper bolt/stud/nut selec-
tion and torque requirements.
For non-circular gaskets, such as
in some pump casings and valve bon-
nets, foil-inserted graphite has been
successfully used. Where larger gas-
kets are required for heat exchanger
heads with partitions, and so on, foil-
inserted graphite, tang or corrugated
metal-inserted graphite can be used.
With any gasket type selected, proper
installation techniques prescribed by
the manufacturer should be followed.
Pumps and pump seals
Pumps in HTF service are gener-
ally of cast steel for withstanding the
stresses of thermal shock. Pumps in
30 Chemical Engineering www.che.com December 2010
high temperature service can have
double mechanical seals, or can be of
a sealless design. The pumped fluid is
used to lubricate and cool bearings in
either type. Excessive temperatures
at mechanical seal faces (typically
tungsten carbide, stellite, or silicon
carbide) can vaporize the HTF result-
ing in no lubrication and mechanical
damage to seal face materials. This
mechanism can create particles that
can erode the seal face and create a
separation of seal faces resulting in
leakage. Therefore, stuffing box and
seal gland cooling is important to
maintain lower temperatures and also
improved lubricity of the HTF. A seal
flush can be effective in minimizing
excessive seal face wear and particle
accumulation on flex metal bellows by
removing harmful deposits with fil-
tered fluid [15].
Valves
Valves in DP:DPO service may in-
clude forged- or cast-steel, or stain-
less-steel bodies, balls, plugs and
disks. 13-Chrome (Cr) stems and
13-Cr or stellited seats should be
considered. Fire-resistant graphite
packing should be specified in select
configurations available. Bellows seal
designs can provide physical barri-
ers for reduced emissions and leaks.
With frequent updates of valve fea-
ture availability, the designer should
always consult with the valve manu-
facturer to consider the latest inno-
vations available. Soft-seat materials
should be avoided since they can burn
out in case of fire, potentially adding
to the complexity of the HTF release.
Small valves with welded end-con-
nections should be considered to re-
duce potential leak points, and larger
valves should be considered with
flanged end-connections. Connecting
piping should have adequate flexibil-
ity to minimize torque induced from
thermal expansion and contraction of
piping applied to flanges, which may
develop leaks.
Maintenance practices
Simply put, three keys for adequate
maintenance practice are as follows:
1. Respond to identified leaks promptly
with repairs
2. Repairs should address the cause of
the failure, and not just the conse-
quences of the failure
3. Learn, document, and practice the
recommended preventive mainte-
nance of the equipment
Leak Detection
Human senses
DP:DPO fluid can be operated in liq-
uid or vapor phases. When there is
leakage, even above its normal boiling
point, the vapor emitted can quickly
condense to form a visible, near-white
mist cloud. The cloud quickly increases
its transparency as it dissipates. For
small leaks it may be possible to ob-
serve liquid droplets present at the
source of the leak, such as valve stems,
pump seals, flanges, and so on, as well
as on the ground. As mentioned ear-
lier, a small leak from a flex hose can
appear as a darkened, wetted area.
When cool ambient conditions permit,
frozen DP:DPO may be observed as
white crystalline solid. Visual detec-
tion capability for leaks can be greatly
enhanced by use of remote video-sur-
veillance monitoring.
The DP:DPO eutectic mixture
has been measured to have an odor
threshold of 9 parts per billion (ppb) in
air, making detection by odor possible
without exceeding established airborne
exposure limits for either component.
In indoor areas, the airborne HTF may
not dissipate readily, preventing easy
tracing back to the leak point. In out-
door environments, the odor is more
closely localized around the leak point
due to the more rapid dissipation of
the components in air. While odor may
be a sensitive detection technique, the
use of instrumentation is a preferred
detection-and-measurement method
with superior quantifying capability
for leak tracing.
Specialized instrumentation
Instruments using the principal of
photoionization are well suited for use
in CSP and process plants. Typically,
these instruments can readily detect
DP:DPO concentrations in air as low
as the ppb range. For realtime per-
sonnel exposure monitoring, the ppb
sensitivity is appropriate to quantify
exposure at levels below established
occupational exposure limits for the
fluids components. For maintenance
 Water fog mist suppression system
Vapor mist
Heat
transfer fluid
piping
Figure 4. Vapor or mist detection systems automatically detect the presence
of clouds of DP:DPO and can automatically trigger water-fog deluge systems and
alarms, initiate forced ventilation, and de-energize area equipment
needs, parts-per-million (ppm) sensi-
tivity is sufficient to determine orders
of magnitude of identified leak sources
so that repairs can be scheduled on
a prioritized basis. The instruments
can be purchased as handheld units
for lightweight mobility, or as fixed-
mount, continuous area monitoring
stations. The devices pull air samples
through the instrument for measure-
ments. Drawbacks include inability to
distinguish DP:DPO vapor from other
measureable organics and combusti-
bles (although, this is not a problem in
most CSP plant areas), and tendency
to condense vapors within the sample
inlet when sampling high concentra-
tion leak points. The internal conden-
sation can require the instrument to
be offline until disassembled for thor-
ough cleaning. Experience in han-
dling the unit can quickly overcome
this issue. Cost of these instruments
currently varies from approximately
$3,000 for small, ppm handheld mod-
els, to $5,000 and higher for the ppb
and fixed-mounted systems. They can
be purchased with intrinsically safe
certifications for use in hazardous
areas.
To prevent the potential ignition of
vapor- and mist-cloud leaks within in-
door areas with reduced ventilation,
custom-developed instrumentation
has been installed to automatically de-
tect the presence of clouds of DP:DPO
with the capability of automatically
triggering water-fog deluge systems
and alarms, initiating forced ventila-
tion, and de-energizing area equip-
Fog nozzles
Photocell
detector
I/P
Water
header
Deluge
valve
ment with a continuous monitoring ca-
pability. This design has incorporated
the use of multiple-sensor actuation
to avoid accidental activation from
only one monitor (Figure 4). The use
of water-fog systems has been found
in testing to rapidly reduce airborne
DP:DPO concentrations to below com-
bustible concentrations in air [16].
It is important to follow the manu-
facturer recommendations with re-
spect to the use and maintenance of
these instruments.
Process level indications
Todays facilities will typically have
installed instrumentation for the
routine measurement of liquid level
in process vessels, including bulk
storage tanks, surge or expansion
tanks and ullage vessels, conden-
sate tanks, and others. Modern in-
strumentation often permits excel-
lent level-trend capability so that
response time to unexpected level
changes is greatly enhanced by use
of configured deviation alarms with
a distributed control system (DCS).
Common suitable instrumentation
types can include displacer, differen-
tial pressure and radar within their
respective design limitations of tem-
perature. Externally mounted floats
within stainless-steel chambers can
provide 0 to 100% level indication
and remote, continuous monitoring
on DCS screens. Changes in liquid
level in vessels can be a somewhat
crude, but important component of
detection material loss.
Chemical Engineering www.che.com December 2010
Potential fire scenarios
The prerequisite to a fire is a leak from
point sources, such as described in the
earlier section on incident history.
Insulation fires. Insulation fires are
not uncommon with high-temperature,
organic heat-transfer fluids. However,
in one extensive study of such fires by
Britton, he concludes, Least prone
should be DP:DPO types, which have
very high AITs (above 500C) and
relatively low flash-points of around
130C [17]. DP:DPO-soaked insula-
tion apparently avoids the insulation
fire mechanism by its more rapid rate
of evaporation out of the insulation
than other higher-boiling heat-trans-
fer-fluid chemistries, while avoiding
the close approach of its autoignition
temperature.
Spark- or flame-produced ignition.
With DP:DPO fluid, a more expected
cause of fire would be from spark- or
flame-produced ignition. Releases of
liquid DP:DPO above its fire point of
127C (260F) are susceptible to igni-
tion from area sources, such as exposed
electrical contacts, spark-producing
tools, open flames and so on. However,
hot HTF can rapidly cool once released
into its environment, both by evapora-
tive cooling and by conducting thermal
energy into the heat sinks provided by
the ground, equipment and so on. A
good design practice is to provide slop-
ing of underlying surfaces such that
any liquid release can drain away from
process equipment, thereby minimiz-
ing potential ignition. Also, once any
release of HTF is observed, all spark-
producing work should be halted until
the situation has stabilized.
Autoignition. Autoignition of a
DP:DPO cloud has been reported in
one CSP plant where the HTF leaked
in close proximity to the heat collection
element (HCE). When a vapor or mist
cloud of DP:DPO within its combus-
tible limits of concentration in air con-
tacts a surface at temperatures above
the autoignition point of the HTF, it
can ignite. The cloud is consumed rap-
idly and can continue to burn at the
source of ongoing leakage until its flow
is stopped. Where possible, designers
are encouraged to consider incorporat-
ing remote isolation capability for pip-
ing circuits, so that the fuel to the fire
can be safely interrupted.
31
 Cover Story
Fire prevention &
fighting techniques
Breaking the fire triangle
Three necessary components make up
the fire triangle (Figure 5) for an HTF
system: oxygen, an ignition source and
fuel (DP:DPO). How do we address
each of these successfully to avoid
their fiery combination?
Oxygen. Within process vessels, pro-
vide blanketing with dry, inert gas. Ni-
trogen is typically used, and should be
a minimum of 99% purity.
Ignition sources. HCEs of CSP
plants are the previously identified
point of AIT concern, as the surfaces
can reach temperatures above the
DP:DPO AIT, particularly when leak-
age reduces the flowrate of the HTF
through the HCE. To date there has
been no known provision made to avert
the potential drifting of a DP:DPO
cloud across the HCE. Upon the first
detection of leakage from piping com-
ponents in the proximity of potential
ignition sources, prompt maintenance
should be scheduled for repair or re-
placement. Potential for open flames
(such as smoking, welding and cut-
ting torches) can be addressed by
strict compliance with adequate ad-
ministrative controls. Electrical area
classification design should be prop-
erly addressed so that it conforms to
applicable code requirements. This
should establish a minimum radius
around each potential DP:DPO leak
point, and should be determined ac-
cording to specified standards for the
requirements of electrical enclosures,
switches, motors and so on.
Fuel (DP:DPO). Use of HTFs at very
high temperatures often requires an
organic heat-transfer fluid of high-
References
1. Therminol VP-1 Heat Transfer Fluid Vapor
Phase / Liquid Phase Heat Transfer Fluid,
Pub. No 7239115C, Solutia Inc.
2. Apanel, George J., Process Economics Pro-
gram Report 247, Gas To Liquids Update,
SRI Consulting, 2002.
3. Systems Design Data, pp. 4.3 4.4, Pub.
No. 9239193, ver. C., Solutia Inc.
4. Organic Heat Transfer Fluids and Equip-
ment, Ind. Risk Insurers, IM.7.1.5, p. 1,
June 1, 1975.
5. Pratt, Thomas H., Electrostatic Ignitions of
Fires and Explosions, Wiley-AIChE, Center
for Process Safety, July 15, 1997.
6. NFPA 77, Recommended Practice on Static
Electricity, National Fire Protection Assoc.,
2007.
32 Chemical Engineering www.che.com December 2010
est thermal stability, which is
the DP:DPO eutectic fluid. The
proper approach to eliminating
the presence of DP:DPO outside
of its containment is by careful
design and installation of the
system and its components,
and incorporating a culture of
effective preventive mainte-
nance and inspections for the
systems proper operational
integrity. For the unexpected
leaks, designs incorporating
remote isolation capability can
help minimize DP:DPO release
Figure 5. Oxygen, an ignition source and fuel
and personnel exposure during (DP:DPO) are the three necessary components of
these events. the fire triangle for an HTF fire scenario
Fire fighting response
For small or incipient-stage fires with
a limited and minimal source of fuel
(DP:DPO), the use of a handheld fire
extinguisher can be an effective and
appropriate selection to extinguish
a fire. Approach to the fire should
be from upwind to avoid breathing
partially combusted materials, with
standard sweeping discharge of dry-
chemical or carbon-dioxide media.
Consider placement of 2030 lb (914
kg) dry-chemical fire extinguishers in
easily accessible locations at each level
of structure where the fire potential
exists in HTF handling areas [12].
Personnel must be properly trained in
the operation of selected extinguishers
and techniques for approaching and
extinguishing HTF fires.
Larger dry-chemical units are avail-
able on trailers and can be quickly
transported to distant locations within
the expanse of a facility. It is recom-
mended that personnel intended to
operate such equipment receive spe-
7. NFPA Standard 30, Flammable and Com-
bustible Liquids Code, NFPA, 2008.
8. ASME B31.1 Power Piping, American
Soc. of Mechanical Engineers, New York, NY,
2004.
9. ASME B31.3 Chemical Plant and Petro-
leum Refinery Piping, ASME, New York,
NY, 2004.
10. ASME Code Sec. VIII Div. I Pressure
Vessels, ASME, New York, NY, 2004.
11. ASME Code Sec. VIII Div. II Pressure
Vessels-Alternative Rules, ASME, New
York, NY, 2004.
12. Loss Prevention Data Sheets 7-99, Organic
Heat Transfer Fluids, Factory Mutual,
2009.
EN
HE
YG
AT
OX
FUEL
cialized training and certification
prior to being authorized to respond.
Onsite fire brigades can address fires
much more rapidly than offsite fire de-
partments, and can help to minimize
potential damage, system downtime
and repairs.
As mentioned earlier, fires involv-
ing larger releases of HTF in non-con-
gested areas of a facility that are be-
yond the capability of incipient-stage
response techniques are perhaps best
managed by remote isolation of the
DP:DPO source, and allowing the fire
to burn out without risk to employees.
Runoff of liquid should be directed
away from areas that might result in
costly damage, extensive downtime
and repairs. Water monitor stations
equipped with water fog nozzles may
provide needed protection and cooling
of equipment in certain areas, which
should be assessed in planning for a
fire-water management scheme.
Indoor areas may require sprinkler
protection when potential leak points
13. National Board Inspection Code, National
Board of Boiler and Pressure Vessel Inspec-
tors, Columbus, OH, 2007.
14. API 510, Pressure Vessel Inspection Code,
Am. Petroleum Institute, Washington, D.C.,
2006.
15. Systems Design Data, Pub. No. 7239193
ver. C, Therminol Heat Transfer Fluids, So-
lutia Inc.
16. Vincent, G. C., Nelson, R. C., Russell, W. W.,
Hydrocarbon Mist Explosions Part II, Pre-
vention by Water Fog, Loss Prevention, (Vol.
10), American Institute of Chemical Engi-
neers, 1976.
17. Britton, L. G. , Spontaneous Fire in Insula-
tion Plant/Operation Progress (Vol. 10, No.
1), pp. 33, Jan., 1991.
from HTF piping and equipment
exist. Consideration may be given to
automatic leak-detection systems,
coupled with the capability to de-en-
ergize area electrical equipment and
activate zones of water fog deluge sys-
tems and alarms. If ventilation is acti-
vated, it may negate the desired effect
of water fog sprays in scrubbing the
HTF mist from the air. Pooled or con-
tained DP:DPO liquid settles to form
a bottom layer once it has cooled to a
temperature at which its density is
greater than the density of the water.
For any of the above approaches for
fire response, consultation with insur-
ance providers is recommended.
Final Thoughts
The primary intent of this paper is to
stress the importance of good design,
installation, and maintenance of equip-
ment in DP:DPO handling facilities in
preventing HTF release and fire events.
By keeping the HTF within the pip-
SPEC IT AND
FORGET IT.
Check-All Valve is your
one-stop supplier for the
check valves you need in the
materials you require. Better
yet, every valve includes the
experience, engineering, and
application know-how you need
for spec-it-and-forget-it reliability.
After all, you have better things to
do, and check valves are all we do.
Call us at 515-224-2301 or
e-mail us at sales@checkall.com.
www.checkall.com
Circle 3 on p. 58 or go to adlinks.che.com/29257-03
ing as intended by robust design, the
frequency, severity, and consequences
of leaks and fires can be minimized or
even prevented altogether. All compo-
nents of a complex HTF system cannot
be addressed within the context of this
paper, therefore the designer and facil-
Authors
Conrad E. Gamble is the
product steward for Ther-
minol heat transfer fluids
at Solutia Inc. (702 Clydes-
dale Ave., Anniston, AL
36201; Phone: 256-231-8525,
Fax: 256-231-8553; Email:
cegamb@solutia.com). He cur-
rently supports the Therminol
business in product develop-
ment, customer technical sup-
port, and providing technical
presentations in various venues.Gamble joined
Solutia (then Monsanto Co.) in 1985 as a process
engineer. He has advanced in various manufac-
turing and engineering roles, until his current
assignment as technical service lead. A licensed
professional engineer and recognized expert in
the industry, Gamble has served the chemical
industry for more than 20 years. He is also a
member of the American Institute of Chemical
Engineers. Gamble holds a B.S.Ch.E.from the
University of Alabama, Tuscaloosa, AL.
Chemical Engineering www.che.com December 2010
ity operator should consult with knowl-
edgeable specialists representing sup-
pliers of the heat transfer fluid, piping
components and specialty fittings, and
insurance providers to properly protect
against leaks and their consequences.
Edited by Rebekkah Marshall
Matthias Schopf is the MTS
manager of Solutias Spe-
cialty Fluids business in Eu-
rope (Solutia Europe BVBA/
SPRL, Rue Laid Burniat 3,
1348 Louvain-La-Neuve, Bel-
gium; Phone: +32- 10-48-1579;
Fax: +32-10-48-5579; e-mail:
matthias.schopf@solutia.
com). His responsibilities
include providing technical
service to Therminol heat
transfer fluid customers as well as leading activ-
ities in new technologies like renewable energy
production. Prior to this role, he worked as sales
engineer for heat transfer fluids and aviation
hydraulic fluids in Europe and the Middle East
for more than 10 years. Before joining Solutia
in 1996 he worked as an engineer for machine
monitoring systems for power plants and in low
temperature thermometry research at the Ger-
man institute PTB. Schopf received a Master of
Science degree in Physics in 1989 from Hum-
boldt University in Berlin, Germany.
Since 1958
West Des Moines, Iowa, USA
33
 Solids
Environmental
Processing
Manager
Avoiding Static Sparks
In Hazardous Atmospheres
Relatively simple steps can be taken to mitigate risks
associated with static electricity in process plants
Graham Tyers, Newson Gale Inc.
W
e are all familiar with the
effects of static electricity in
everyday life the sharp
spark we experience as we
approach a metal file cabinet or the
crackle we feel as we remove a garment
made of synthetic material. While these
effects are largely harmless, in the
chemical process industries (CPI),
whenever flammable or combustible
atmospheres are present, uncontrolled
discharges of static electricity can be
potentially disastrous.
The National Fire Protection Assn.
(NFPA; Quincy, Mass.; www.nfpa.org)
tracks the causes of all industrial
fires and explosions reported to U.S.
fire departments. Roughly 280 inci-
dents/year in the U.S. are attributed
to static electricity, and these result
in significant damage to property, lost
production, injuries and even fatali-
ties. According to the U.K. Institute
of Chemical Engineers, roughly 350
incidents per year are caused by static
electricity. And this figure is probably
conservative, as more such incidents
likely occur but go unreported, or do
not have an electrostatic cause sub-
stantiated during the investigation.
Case in point
One major industrial fire at a tote-
filling operation at a chemical distri-
bution facility in Des Moines, Iowa,
was investigated in 2008 by the U.S.
Chemical Safety Board (CSB; Wash-
ington, D.C.; csb.gov), and the cause
was determined to be static electricity.
(the full report is available at csb.gov).
CSB concluded that a combination of
technical and procedural factors led to
an uncontrolled static discharge, which
created a fire that spread rapidly to a
wood-framed warehouse, igniting a
large volume of flammable and com-
44 Chemical Engineering www.che.com JUNE 2009
Figure 1. Many self-testing ground-
bustible liquids. One employee and ing clamps also monitor cable condi-
one firefighter were injured. The main tion back to the designed ground point,
warehouse structure was destroyed, and will alert the operator if the cable
the operators business was signifi- has worked itself loose or is broken
cantly interrupted, and surrounding
businesses were evacuated. tance of the object, this may result in
significant levels of energy being avail-
Best practices able for discharge well above the
Anyone responsible for, or working minimum ignition energy (MIE) of the
in, potentially explosive atmospheres surrounding flammable atmosphere.
(Class I, II or III) is aware of the fact Typical MIE values (Table 1) vary ac-
that all potential sources of ignition, cording to whether the flammable at-
including naked flames, hot surfaces mosphere comprises vapor, dust or gas,
and electrical sparking must be elimi- but many commonly used solvents and
nated. Today, a variety of safety codes other flammable chemicals have MIEs
and guidelines (discussed below) spec- that are relatively low on the order
ify safe working practices in hazardous of 1 mJ or less. If the isolated conduc-
areas, and provide specific guidance tor then comes into proximity with an-
on how to control static electricity. other object at a lower potential, much
The risk of static electricity is ever- of this energy could be released in the
present, as it is generated continu- form of an incendive spark.
ously, whenever surfaces come into Of course, in order for there to be
contact and then separate. Examples static ignition of the flammable atmo-
include liquids flowing through pipe- sphere, there would need to also be a
lines or filling into drums and tanks, suitable concentration of fuel (vapor,
powder dropping down a chute, and dust or gas) in air, but for the purposes
even a person walking across an insu- of safe plant design, the very fact that
lating floor. there is an identified flammable atmo-
The extent of charge generation in sphere should suggest that this is pos-
industrial processes is usually very sible or likely.
low, typically no greater than 1 Despite these potential hazards, the
104 Amp. However, these generated danger associated with static electric-
charges become a problem in hazard- ity can be effectively mitigated by fol-
ous areas when allowed to accumu- lowing national and industry codes
late on objects that are not at ground of practice, technical standards, and
(earth) potential. Unfortunately, many working practices such as NFPA
of today's paints, coatings, gaskets, 77 (Recommended Practice on Static
seals and other non-conductive mate- Electricity), and NFPA 30 (Flammable
rials are sufficiently insulating so and Combustible Liquids Code) that
they actually prevent the proper dis- specify proper grounding and bonding
sipation of static charge. techniques. Such techniques work on
Without proper engineering and the assumption that if a) conductive
worker training, charges can quickly or static-dissipative plant, equipment
build up to a very high potential, with and materials are used, and b) these
voltages in excess of 30 kV not being materials are properly bonded and
uncommon. Depending on the capaci- grounded, it will be impossible for
 Table 1. Typical MIE values
Material Minimum
(gas/vapor or Ignition Energy
powder/dust) (MIE), mJ
Carbon disulfide 0.009
Methanol 0.14
Xylene 0.20
Toluene 0.24
Propane 0.25
Ethyl acetate 0.46
Zirconium 5.00
Epoxy resin 9.00
Aluminum 10.00
Sugar 30.00 Figure 3. Static ground-verification and interlock
Wheat flour 50.00 systems are essential for certain critical applications
Figure 2. FM/ATEX- such as loading/unloading tanker trucks and rail cars,
Note 1: Minimum ignition energy (MIE) is defined approved grounding
as the minimum energy that can ignite a mixture IBCs, fluid bed dryers, mixers, transfer equipment and
of a specified flammable material with air or oxy- clamps, when properly other process equipment that have a high explosion
gen, measured by a standard procedure. bonded and grounded, potential, applications with a high likelihood of static
Note 2: MIE values are provided for guidance can control static charge accumulation in a very-low-MIE flammable atmo-
only specific MIE data for any material buildup sphere, and situations in which low-conductivity liquids
should be verified. Source: NFPA, IchemE
or highly resistive powders are being handled

dangerous levels of static electricity
to accumulate and result in an uncon-
trolled discharge.
NFPA 77 and NFPA 30 are ex-
tremely useful, in that they provide
practical examples of common op-
erations, and detail effective ways
to eliminate, control or mitigate the
problem. Where today's recommenda-
tions tend to converge is in the recom-
mendation to always use conductive or
static dissipative materials, and to en-
sure effective bonding and grounding.
In this context, the term conductive
would apply to metal materials, such
as stainless or carbon steel, alumi-
num and others, and static-dissipa-
tive may indicate rubber or plastics
that have been formulated with some
added semi-conductive additives.
Bonding means linking these ob-
jects together by means of a suitably
strong conductor (wire), and ground-
ing refers to a true ground/earth
connection that is applied to one or
more of the bonded objects. When one
or both of these techniques is applied,
and while a low resistance connection
between the objects and ground is
maintained, operators are able to pre-
vent dangerous levels of static charge
from accumulating.
Hidden dangers
In any type of flammable atmosphere
there may be hidden dangers lurking
in the form of so-called isolated conduc-
tors. These are conductive objects that
are either inherently or accidentally
insulated from earth. The insulation
effectively keeps any static electric-
ity buildup from safely discharging,
thereby resulting in accumulation of
charge on the object.
Isolated conductors may arise from
metal flanges, fittings or valves in
pipework systems; portable drums,
containers or vessels; tanker trucks,
rail cars and intermediate bulk con-
tainers (IBCs); and even people. Dur-
ing day-to-day operations at industrial
facilities, isolated conductors are prob-
ably the most likely source of static ig-
nition incidents.
Fortunately, the problems associ-
ated with isolated conductors can be
remedied by effective grounding and
bonding. In the case of fixed installa-
tions such as pipe work, storage tanks
and so forth, this is relatively simple
to implement. However, these preven-
tive measures are more difficult to im-
plement with mobile/portable objects,
such as drums, IBCs and tankers. In
these instances, purpose-designed
temporary grounding and bonding de-
vices must be used, with strict proce-
dures to ensure that they are always
in place prior to starting the process. jects, resistance levels higher than
For instance, specific types of
clamps and devices for grounding
and bonding portable or mobile plant
equipment, drums and containers are
recommended in NFPA 77, and such
grounding clamps and devices gen-
erally should employ sharp contact
points. These contact points should
be made of a wear-resistant material,
have positive spring pressure, and be
universally adaptable to a wide range
of plant objects.
In some cases, these grounding
clamps and devices may have addi-
tional third-party notified-body test-
ing such as Factory Mutual (FM) or
the European ATEX Directive (Figure
2). If these units are properly specified
and correctly used, in most cases the
user can be reasonably sure of effec-
tive static control.
Chemical Engineering www.che.com JUNE 2009
In all situations it is also impor-
tant to periodically test the control
measures used, to check the condition
of the clamp, contact and cable, and
to verify the all-important connec-
tion back to the grounding point (bus
bar). Instruments like Ohm testers or
multi-meters are useful for perform-
ing such verifications, but, of course,
these will need to be approved as in-
trinsically safe instruments if you are
working live in a hazardous area.
According to the recommendations
in the latest edition of NFPA 77, the
maximum resistance to ground in
conductive (metal) bonding systems
should be tested to 10 Ohms or less.
With all-metal systems and tight con-
nections, it should be easy to achieve
this level of resistance to ground.
While higher resistance levels may
be theoretically low enough for static
charge removal from conductive ob-
two-digit figures in all-metal systems
usually indicate problems, such as cor-
rosion, loose or broken connections or
a buildup of coating or insulating ma-
terials somewhere in the system. Any
such readings should be immediately
investigated and rectified.
The presence of semi-conductive or
so-called static-dissipative materials
may also exhibit higher resistance
levels, owing to their varying charac-
teristics (which in some cases may be
as high as 1 108 Ohms).
Additional considerations
Even when the appropriate static
safety equipment has been specified,
there are some further concerns that
must be addressed by all those respon-
sible for operations within hazardous
areas. In operational terms, attaching
45
 Environmental Manager

Figure 4. If static-
a grounding clamp to a plant ob- Recommended dissipative footwear is
ject is always a physical action. resources specified, it can also
Thus, there is always a risk that be tested for ongoing
even if operators diligently carries Useful guidance on minimizing fire and effectiveness using
explosion hazards, and managing static test stations that test
out their duties per the company-
electricity in process plants, can be found the ground leakage
recommended safety procedures, at these websites. from the individual,
no one can ever be sure that the via the footwear, to
clamp has made good enough NFPA77, NFPA30 www.nfpa.org ground. These devices
contact with the object. For in- Chemical Safety Board www.csb.gov help prevent incor-
IchemE www.icheme.org rect shoes from being
stance, the clamp has to make
OSHA www.osha.gov used in static-sensi-
a low enough resistance contact tive environments
with the conductive object to en-
able any static that is generated to be forming its intended function of safely case, the process cannot be initiated
safely dissipated to ground before it dissipating any static electricity that until proper grounding is established,
can accumulate to dangerous levels. is generated. thereby ensuring that any generated
The fact remains that many conduc- These specialized clamps employ static electricity is safely dissipated to
tive objects that are capable of accumu- intrinsically safe electronic monitor- ground. These systems are generally
lating high static charges also have in- ing circuits that are powered from a fed from a line feed 110/120-V a.c.
sulating layers on their surfaces, which low-energy battery. The circuit is only power supply, and employ approved
may prevent this low resistance con- completed when the clamp achieves intrinsically safe barriers to limit the
tact. Examples include paints and coat- a low resistance contact onto the ob- monitoring circuit down to safe levels.
ings on drums, tank vehicles and other ject to be grounded, and the operator They must also have proper hazard-
mobile plant equipment, or product receives visual confirmation of this ous location and safety certification.
buildup caused by normal working con- via a light/indicator (usually a flash- Static ground-verification and inter-
ditions (for instance where insulating ing LED). The self-testing grounding lock systems are typically used in criti-
liquids, powders and other materials clamp also monitors cable condition cal/ultra-safety applications, such as
are part of the operation or process). As back to the designed ground point, loading/unloading tanker trucks and
a result, many grounding and bonding and will also fail to register a permis- rail cars, IBCs, fluid bed dryers, mixers,
clamps show very high resistance read- sive signal if the cable has worked transfer equipment and special process
ings when clamped to conductive objects itself loose or is broken (Figure 1). In machines that have a high explosion
with insulating surfaces. Worse still, if order to be safely used in hazardous potential. They are also useful wher-
the company tries to cut costs by using locations, these clamps should be ap- ever there is a high likelihood of static
standard welding clamps or lightweight proved as intrinsically safe devices charge accumulation in very low MIE
alligator clips for static grounding and carry the appropriate Hazardous flammable atmospheres, or where low-
(instead of purpose-designed clamps), Area Certification, for instance, from conductivity liquids or highly resistive
these devices have an even higher Factory Mutual (FM), Underwriters powders are being handled (Figure 3).
failure rate with potentially grave Laboratories (UL), or the Canadian Static ground-verification clamps
consequences. Standards Authority (CSA). and interlock systems also tend to
During an internal audit (using an To move to an even higher level of have an important beneficial effect
approved Ohm meter) of ground con- security, static ground verification sys- on the operators using them. Since
nections in its flammable-liquids area, tems are available that not only give they build an additional check into
one major chemical company actually visual verification to the operator, but the operation, their use helps to re-
determined that only 85% of the de- that provide interlock switching con- inforce the static safety procedures of
vices had acceptable low resistance tacts that may be linked to process the company in short, the operator
levels, or proper ground. Therefore, pumps, valves, alarm/shutdown/con- is more likely to observe the correct
15% of their ground connections failed trol systems, as well. This means that procedures, as he or she is kept aware
to meet the proper levels, and as a re- the process cannot be started until of the need to properly control static
sult were at risk to the process area the conductive object has been safely electricity on a daily basis.
and operating personnel. grounded, and, if at any time during the One final element of the static safety
To solve this problem, NFPA 77 sug- operation the condition changes (due to audit is the static accumulating char-
gests that intrinsically safe, self-check- a clamp falling off or wire breaking for acteristics of people working in the
ing grounding and bonding clamps be instance), the system automatically hazardous area. If the human body is
used. From an operators point of view, switches to the non-permissive state insulated from natural contact with
these devices are employed in exactly and shuts down the process. the ground, either by non-conducting
the same way as conventional ground- Systems employing interlocks can shoes, insulating floors or both, then
ing clamps. Where they differ is in also prevent accidents that would oth- static charge can start to accumulate
the way that they reassure the opera- erwise result when operators approach as the person walks along the floor.
tor that the clamp has not only been plant objects that are already carrying After just a few steps, a fairly high po-
physically attached, but is also per- accumulated static charges. In this (Continues on p. 49)
46 Chemical Engineering www.che.com JUNE 2009
 Environmental Manager
Avoiding Static Sparks In Hazard-
ous Atmospheres (Cont. from p. 46)
tential may develop (especially in dry,
low-humidity atmospheres).
If the person now approaches a con-
ductive material at either ground or a
lower electrical potential, a spark may
be promoted. It is generally thought
that sparks from people can reach en-
ergies as high as 30 mJ sufficient
to ignite almost all flammable vapors
and gases, and many sensitive com-
bustible powder (dust) atmospheres.
In order to control this, it is neces-
sary to ground the employees working
in sensitive hazardous areas. This can
take the form of ground wrist-straps
for sedentary workers, or specialized
static-dissipative (SD) safety shoes for
personnel who need to move around to
carry out their tasks. In the case of the
latter, it is also necessary for the floor
materials to be sufficiently conduct-
ing to allow a ground leakage path
to safely dissipate any static charge as
each foot contacts the ground.
In the U.S., SD footwear is manu-
factured according to ASTM F2413-05,
which specifies maximum and mini-
mum levels for ground leakage via the
shoe. According to ASTM, it should be
sufficiently low to allow static charges
to be safely dissipated to ground, but
sufficiently high to afford some pro-
tection against electric shock. If SD
footwear is specified, it can be tested
for ongoing effectiveness using test
stations that test the ground leakage
from the individual, via the footwear, to
ground. These devices help prevent in-
correct shoes from being used in static-
sensitive environments (Figure 4).
In general, proper mitigation of
static electricity risk in hazardous
areas demands a holistic approach to
plant, process and personnel safety.
Any multi-part control program is
only as good as the weakest link in the
chain. As the speed and scale of mod-
ern manufacturing and distribution
techniques increase, and the range of
materials used and processed grows,
this basic approach to safety will be-
come even more important.
More detailed information on this
subject may be found in industry
safety guidelines, such as NFPA 77,
and OSHA Regulations, Section 1910,
which provide safety managers with
practical guidance for specific situa-
tions. It is also important to remember
that proper grounding and bonding may
not be enough on their own to ensure
completely safe operations. Depending
on the materials involved it may some-
Chemical Engineering www.che.com JUNE 2009
times be necessary to also employ other
protection measures like static charge
neutralization, inert gas blanketing (to
remove the flammable atmosphere),
and to look at process related ways to
minimize charge generation and maxi-
mize charge dissipation rates.
Controlling static electricity in haz-
ardous areas may be simple in theory,
but in practice requires a thorough
knowledge of processes and operations,
sound engineering controls, properly
specified safety equipment and prop-
erly trained operational staff.
Edited by Suzanne Shelley
Author
Graham Tyers is president
of Newson Gale Inc. (www.
newson-gale.com), a manufac-
turer of static control equip-
ment for hazardous areas (460
Faraday Ave., Unit B, Suite 1,
Jackson, NJ 08527). He has
worked in the specialist field
of electrostatic control and
hazardous atmospheres for
nearly 20 years. His industry
background is in monitoring
and control systems, and general electrical equip-
ment for potentially explosive atmospheres. Tyers
is a member of the NFPA, ESDA, and ISA. He reg-
ularly conducts training and presents to industry
and safety groups on the subject of static electric-
ity and hazardous areas. In addition to his tech-
nical subject background, he holds a Diploma in
Business Management (Open University, United
Kingdom). As part of the management team of
Newson Gale in the U.K., Graham was awarded
the Queens Award for Enterprise in 2005. He now
heads up Newson Gales North American Opera-
tions, based in New Jersey.
49
 Solids
Environmental
Processing
Manager
A Checklist for
Safer Chemical
Batch Reactions
A good understanding of the reaction
chemistry is needed for a safe process design
Richard Kwasny , Chilworth Technology Inc.
I
t is a fact that safety-related in-
cidents do occur in the chemical
process industries (CPI). These
incidents have, in part, led to in-
creased attention to reactive-chemical
issues by industry, government and
other stakeholders. We know that good
safety-management systems take the
reactivity of chemicals as well as the
energetics of both desired and adverse
reactions into consideration.
This article provides a safety check-
list that can be used as a guide for the
design of a new process. Alternatively,
it can be used to identify information
gaps when existing processes undergo
periodic review.
Runaway reactions
It is known that in many cases, ther-
mal runaway reactions occur due to
the following factors:
1. Lack of understanding of the pro-
cess chemistry and energy for the
desired reaction
2. Inadequate heat transfer capacity
at the plant level
3. Insufficient control and inadequate
plant-safety back-up systems, in-
cluding emergency venting
4. Ineffective plant operational proce-
dures and inadequate training
The reader is encouraged to add
plant-specific items to this list, as
needed. A hazard analysis can then
be used to identify appropriate con-
trols and safeguards to reduce risk. It
is important to note that as a process
undergoes changes, the safety informa-
tion also needs to be updated. The final
process-safety package should be at
such a level that it can be used for tech-
nology-transfer purposes at the R&D
or commercial-production stage by out-
sourcing contractors and by in-house
personnel. When developing safety
documentation, it is important to keep
in mind that it must comply with com-
pany policies and procedures as well as
country and local regulations.
Process safety checklist
The following items should be consid-
ered in relation to process safety.
Preliminary hazard assessment:
Determine the thermal stability of
all reaction components within the
minimum and maximum process
temperatures attainable under a
worst-case scenario
Identify unwanted interaction be-
tween reagents and solvents
Identify potential reaction contam-
inants that may have an inhibitory
or catalytic effect on the desired
reaction
Quantification of desired reactions:
Determine the heat of reaction and off-
gas rates for the desired and quench
reactions, including the heat resulting
from accumulation of reagents or slow
forming intermediates
Determine the maximum adiabatic
temperature for the reaction, and
determine the basis of safety rela-
tive to the estimated boiling point of
the reaction mixture
Understand the relative rates of all
chemical reactions
Quantification of adverse reactions:
Assess the thermal stability of the
reaction mixture over a wide tem-
perature range
When optimizing the robustness of
the process, consider other reaction
variables, such as pH, concentra-
tion, conversion rate, off-gas rate,
stability of starting and product sub-
Chemical Engineering www.che.com April 2008
strates in solution and as a slurry
Consider the potential and impact of
unwanted vapor-phase reactions
Develop a chemical-interaction ma-
trix for materials present in the
reaction mixture, classify the reac-
tivity, and communicate this infor-
mation to operational personnel
Plant considerations:
Conduct a basic energy balance to
consider the heats during various
additions, heat generated during the
chemical reaction, and the heat re-
moval capability of the plant reactor
system. Remember to include reac-
tor agitation as a source of energy
(~2250 Btu/h/hp)
Consider the impact of possible de-
viations from the intended reactant
charges and operating conditions
Identify all heat sources connected
to a reaction vessel and assume
the maximum possible worst-case
scenario
Determine the effect of the lowest
possible temperature to which the
reactor heat-transfer fluid could cool
the reaction mixture
Consider the impact of temperature
gradients and other issues, such as
increased viscosity, freezing at re-
actor walls, fouling, and so on, in
plant-scale equipment
General chemistry and engineer-
ing design concepts:
Design reactions that occur fairly
rapidly
If possible, avoid batch reactions
in which all the potential chemical
energy is present at the onset of the
reaction
Use semi-batch processes for exo-
thermic reactions in which the batch
temperature and any off-gassing can
61
 Environmental Manager

be maintained through controlled
addition of the reagent
For highly exothermic reactions,
avoid using temperature control
of the reaction mixture as the only
means for limiting the reaction rate
When scaling up a reaction, account
for the impact of vessel size on heat
generation and heat removal: The
volume of the reaction mixture in-
creases by the cube of the vessel
radius but the heat-transfer area in-
creases by the square of the radius
A comprehensive hazard evalua-
tion should be conducted using appro-
priate estimation and experimental
techniques to identify potential reac-
References
1. Hendershot, Dennis C., A Checklist for In-
herently Safer Chemical Reaction Process
Design and Operation, Center for Chemical
Process Safety International Conference and
Workshop on Risk and Reliability, 2002.
2. Kwasny, Richard S., Hazard Assessment
Strategies for Reduction Reactions, South-
Chem_Eng1-1_177x123_M_April08
bank University, London, 1999. 07.12.2007
tion hazards in materials, as well as
the desired and adverse reactions. We
use estimation techniques, differential
scanning calorimetry (DSC), Carius
Tube, and reaction calorimetry as
needed. Identify any adverse or ther-
mal runaway reactions and, if needed,
characterize them using adiabatic
calorimetry, such as ARC (accelerat-
ing rate calorimetry) or an Adiabatic
Dewar Calorimeter. If required, the
vent size can be determined using
Design Institute for Emergency Relief
Systems (DIERS, an AIChE industry
alliance) methodology with data gen-
erated using an adiabatic dewar or
Vent Sizing Package (VSP; a special-
3. Barton, J. and Rogers, R., Chemical Reaction
Hazards, 2nd ed., Gulf Publishing, 1997.
4. Brethericks Handbook of Reactive Chemical
Hazards, 6th ed., Butterworth Heinemann,
1999.
5. Guidelines for Chemical Reaction Hazard
Evaluation and Application to Process De-
sign, Center for Chemical Process Safety of
11:33
the AIChE, 1995.Uhr Seite 1
ized adiabatic calorimeter that uses
temperature, pressure, and rate data
to allow for sizing emergency vents).
The references given below were used
to develop this article and are an ex-
cellent source of information. n
Edited by Dorothy Lozowski
Author
Richard Kwasny is the as-
sociate director of the Pro-
cess Safety Laboratory at
Chilworth Technology Inc.
(250 Plainsboro Road, Build-
ing 7, Plainsboro, N.J. 08536;
Email: rkwasny@chilworth.
com; Phone: 6097994449),
where he has responsibility
for the quality, safety and
productivity of the thermal
hazard laboratory. His exper-
tise encompasses all areas of chemical reaction
hazards and flammability of dusts, vapors and
gases. Kwasny completed his Ph.D. at London
South Bank University. His research developed,
in part, an assessment strategy to allow for the
safe scale up of reduction reactions. Before his
current position, Kwasny headed the Process
Safety & Hazard Identification group for Scher-
ing-Ploughs Chemical and Physical Services
Dept. Previously, he managed the Dow Chemical
Reactive Chemicals testing function for Canada.
Kwasny is a member of the ASTM Committee
E27 on hazard potential of chemicals.

On pages 2 to 31 we show how

you can turn a storage and shipping drum into
a process vessel in next to no time.
Everyone who opts for MLLER gets
an original. We offer more than just
a drum: our extensive range of acces-
rum and
Mller D Systems
sories makes it a multifunctional
vessel - systematically.
er
Contain Extendible whenever you want to fit
your concept and meet your needs
008 exactly. Dont take any chances: only
CK 2
T E RPA o rf a Mller original embodies the expertise
IN sseld 8 gained over more than a century.
in D 30.4.200 0 Ask for our brochure Mller Drum and
m 24.- 7 0 B2
fro tand Container Systems today.
7.0, S
Hall Mller GmbH, Industrieweg 5
D-79 618 Rheinfelden, Germany
Phone: + 49 (0) 76 23 / 9 69 - 0
Fax: + 49 (0) 76 23 / 9 69 - 69
E-mail: info@mueller-gmbh.com
A Company of the Mller Group

1.1
Lidded Dr
ums
th bungs
/ hopper
cans

Innovation in Stainless Steel

Drums wi
Hoppers
Silos
Containe
rs
www.mueller-gmbh.com

Valves
Butterfly
ucks
Drum Tr

Circle 20 on p. 76 or go to adlinks.che.com/7371-20
62 Chemical Engineering www.che.com April 2008
 Engineering Practice
A Safety-Centered Approach
To Industrial Lighting
The proper design and operation of lighting
is essential to ensure plant safety and
support good maintenance practices
Boris Viner
Humatrack SafeLight
T
he use of appropriate hazardous-
area lighting in the chemical
process industries (CPI) is a key
factor when it comes to improv-
ing overall plant safety and worker
productivity, and supporting the best
maintenance practices. The poten-
tially hazardous conditions associ-
ated with many industrial settings
place high demands on the lighting
fixtures, and it is essential that path-
ways, walkways and stairways in
multi-level platforms or facilities be
properly illuminated.
This article discusses the impor-
tance of using the most effective
lighting practices in hazardous envi-
ronments, and presents a conceptu-
ally new lighting approach dubbed
safety-centered lighting that seeks
to help facility operators choose and
install the best hazardous-area light-
ing throughout the facility by combin-
ing knowledge of human physiology
with the latest advances in lighting
technologies, area-illumination tech-
niques and installation practices.
Todays lighting fixtures based
on light-emitting-diode (LED) tech-
nology offer significant improve-
ments over conventional options
for hazardous-area lighting such as
incandescent, fluorescent or high-
pressure sodium (HPS) lamps. As
discussed below, LED fixtures can
provide direct improvements in
overall safety, while reducing main-
tenance and energy requirements,
as well.
Color B C
Candela
A
Human
factor
D E
Coverage Configuration
A better approach
An appropriately de- F G
signed industrial light-
Longevity Integrity
ing program requires
a keen understanding
of both the human fac- Figure 1. When devising the best lighting scenarios
for CPI facilities, all of these factors which influence
tors that dictate how a the true impact that the selected lighting options will
person processes visual have in the workplace must be considered
cues and lighting, and
the nature of the lighting scenario This has potentially serious implica-
itself. Today, sufficient data are avail- tions in industries where employees
able to substantiate that proper light- may be exposed to hazardous chemi-
ing is a key contributing factor to over- cals or fire, because workers who may
all plant safety, and that well-lighted be inadvertently exposed to chemical
facilities tend to have better safety compounds in the workplace must be
and productivity records compared to able to get to the emergency eyewash
poorly lit ones. or safety shower as quickly as possible
Unfortunately, while chemical to wash the affected areas.
plants, petroleum refineries and other Because the first few seconds after
types of CPI facilities often expend sig- exposure to a hazardous substance
nificant time and effort to incorporate are critical to avoid more severe injury
improved operating practices that are or permanent scarring, eyewashes
intended to keep their workers safe, and safety showers must not only be
opportunities to improve the type and located within reach of employees,
nature of the lighting throughout the but they must be easily identified and
facility are often overlooked. properly illuminated and accessible
One of the reasons for this dilemma during both daytime or nighttime con-
is that conventional lighting fixtures ditions, so that affected workers can
have limitations in terms of how the reach them without hesitation.
light fixures function and how the re- As discussed below, LED-based
sulting light is perceived by the end lighting can deliver an immediate
user. For instance, incandescent or and valuable benefit over prevailing
fluorescent lamps are commonly used incandescent or fluorescent fixtures
to identify the locations of emergency during the precious seconds needed to
safety systems, such as industrial locate safety showers and eyewashes,
showers and eyewash stations. How- because LED-based fixtures are bet-
ever, these fixtures provide practically ter able to focus the light at an op-
no useful signaling during the day be- timal, targeted color range, making
cause the lighting blends into sunlight. the safety shower or eyewash loca-
Chemical Engineering www.che.com April 2010 53
 100 680

Engineering Practice 555 nm

Day
100

Relative eye sensitivity, V(M)

10-1
tion light visible at any time of the 507 nm
day or night. Night

Efficacy, Im/W
Meanwhile, with regard to area-illu- 10
mination lighting, it has been widely 10-2 LED HPS
accepted that where artificial light is
used, lighting level calculations could
ignore the color of the light source. Mesopic
1
As a result, HPS light sources (whose 10-3
light falls on the yellow portion of the

Orange
Yellow
Green
Violet
spectrum) are popular in industrial fa-

Blue

Red
cilities, even though they emit rather
poor quality of light at night. 0.1
10-4
By contrast, the newer, LED-based 300 400 500 600 700 800
lighting technologies are better able Wavelength, nm
to provide appropriate lighting that Lm/W = Luminous efficacy, which is the light output of a light source divided by the
takes advantage of the capabilities total electrical power input to that source, expressed in lumens per watt
nm = nanometers; All wavelengths in the color spectrum, as perceived by human
and limitations of the human visual
sight, are expressed in nm
system. The remainder of this article The V (M) function represents therelativeeye sensitivity, in terms of the wavelength
presents a new seven-point lighting of the color produced by the light
profile that provides a unique per-
spective on the complex interactions Figure 2. This chart shows relative eye sensitivity as a function of light wavelength,
between the human visual system and and can be used when evaluating competing industrial area-illumination options. The
key aspects of artificial industrial-illu- goal is to maximize effective lumens (those that can be appropriately utilized by the
human eye). Note that the optimum wavelength selection for safety showers would be
mination and signal-lighting options. green location lights, since the wavelength falls in the mesopic region between peak
sensitivity for both day and night vision
A new lighting profile
As shown in Figure 1, the seven-point
lighting profile presented here con-
sists of the following measurable at-
tributes: a) human factor, b) color, c)
candela, d) coverage, e) configuration,
f) longevity, and g) integrity. Each is
discussed below.
Human factor. As the biggest part of
the puzzle, the human factor is dis-
cussed first and at the greatest length,
because ultimately, many of the driv-
ers for developing the most physi-
ologically appropriate lighting relate
to the human eye. In essence, the best
lighting technologies those that
contribute greatly to overall facility
safety are those that are optimized
for the human eye.
It is important to realize that all
forms of artificial light are not the
same, and different types of light are Figure 3. LED-based location lighting is useful for signaling the location of safety
interpreted in different ways as a showers and eyewashes, since they remain visible in daylight conditions
function of the complex physiology of
the individual viewer. For instance, nighttime vision, and therefore the object, we are using our cone recep-
the retina acts like a light-sensitive sensitivity of the individual human tors. Cone-activated vision is defined
screen at the back of the eye, and it eye varies at different light levels. as photopic (day vision).
has many light receptors that convert The very central part of the retina, The rods are responsible for human
light into electrified signals sent to the the fovea, contains only cones while vision at low light levels, and are prev-
vision centers of the brain. Because of the rest of the retina contains both alent in the peripheral field of view,
their shapes, the two major categories rods and cones. Cones are active at away from our direct line of sight.
of light receptors are called cones and high light levels and are most densely Rod-activated vision is defined as sco-
rods. Cones are responsible for day vi- situated in the central part of the field topic (night vision).
sion, rods play an important role in of view, so when we look directly at an Both in theory and in practice, the
54 Chemical Engineering www.che.com April 2010
 Isofootcandle chart, 8-ft mounting
3 10.5-11.0
Ratio of distance to mounting height 10.0-10.5
9.5-10.0
2.5
9.0-9.5
8.5-9.0
2 8.0-8.5
7.5-8.0
7.0-7.5
1.5
6.5-7.0
1 6.0-6.5
5.5-6.0
5.0-5.5
0.5 4.5-5.0
4.0-4.5
0 3.5-4.0
3.0-3.5
2.5-3.0
-0.5
2.0-2.5
1.5-2.0
-1 1.0-1.5
3 2.5 2 1.5 1 0.5 0 -0.5 -1 -1.5 -2 -2.5 -3
0.5-1.0
Ratio of distance to mounting height
0.0-0.5
The isofootcandle chart is derived from the candlepower data, and shows exact plots
of equal footcandle levels on the work plane when the fixture is at a designated mounting height.
The color legend represents the intensity (candelas) of a cross-sectional photometric distribution
curve.Isofootcandle chart, 8-ft mounting
Figure 4. The photometric data shown here for an LED-based fixture installed at
an 8-ft (2.4-m) mounting height demonstrate how such fixtures can effectively focus
artificial, industrial white lighting on straight lines and avoid spillage, thereby contrib-
uting to optimum illumination, especially in multi-level platforms
determination of lamp lumens involves ing level calculations could ignore
knowing both the spectral power dis- light source color, relying only on a
tribution (SPD) of the lamp and the vi- certain lamp lumen rating (usually
sual response of the eye. Interestingly, as provided by the manufacturer). In
it is not simply defined as a form of these instances, calculated values of
energy in the same way as other forms candelas, lumens, or lux (lumens per
of radiation; rather, it is defined as en- square meter) are not dependent upon
ergy as evaluated by the human eye (a whether the light source is white, blu-
reflection of the visual effect which ish white, yellow or pink in nature.
differs by individual that is created However, published research in-
by that light energy source). dicates the importance of lamp SPD.
Color. When it comes to how the eye It is only recently that the lighting
responds to color, vision scientists industry has begun to acknowledge
have known for most of the twentieth the magnitude of the color effects in-
century that lighting conditions play volved, and to show serious interest in
a big part. While its been widely ac- considering the true properties of the
cepted that the cones in the eye handle light. With this in mind, some light-
day vision and the rods are designed ing experts have started to use the
for night vision, up until now light- term effective lumens to define the
ing manufacturers have continued modified lumen output of a lamp, tak-
to rely on light meters to measure a ing into account the shifting color sen-
lamps lumen output that is calibrated sitivity of the eye at low light levels.
by examining the eyes sensitivity to For example, until recently the
only cone-activated (photopic) vision lumen has been defined as the amount
completely ignoring the effect of of light as perceived by the eye under
rod-activated (scotopic) vision. Unfor- photopic (day) conditions. Based on
tunately this approach represents a this definition, HPS lamps tend to
gross oversimplification of how human have high lumen ratings.
vision works, and signals a missed op- However, due to the yellow color
portunity in terms of optimizing in- of the source light produced by HPS
dustrial lighting options. lamps, at least half of the light cant
This is perhaps one of the most fas- be effectively utilized by human eyes
cinating and important subjects to be at night. So, it is not as important that
addressed for efforts to improve in- the sodium lamp produces a high out-
dustrial lighting in recent years. In put of energy, but rather that its en-
the past, it has been widely accepted ergy peak is near the maximum pho-
that under virtually all circumstances topic sensitivity wavelength of the eye
where artificial light is used, light- (see Figure 2, yellow region).
Chemical Engineering www.che.com April 2010
As a result, one should really deter-
mine the effective lumens of the HPS
light source at scotopic (night) levels.
To do this, the lamp power at each
wavelength would need to be multi-
plied by the scotopic eye sensitivity
at each wavelength, and the values
summed. The effective lumens value,
therefore, will be different from the
conventional lumens value (which, as
noted, is considered only under photo-
pic or daytime conditions).
Because a relatively small portion of
the overall energy output of the HPS
lamp occurs at wavelengths shorter
than the peak, the effective lumens
available for scotopic (night) condi-
tions is greatly reduced. This shows
that sodium lamps (being a source of
yellow light) produce relatively little
useful light (effective lumens) for
nighttime conditions, and therefore
their effectiveness under low light lev-
els when it is essential to have ad-
equate industrial lighting is drasti-
cally reduced.
By contrast, for a white light source,
such as that produced by an LED-
based fixture, the energy output lies in
the high-sensitivity region of the eye
for low light levels (Figure 2).
As a result, the effective lumens in-
crease as the light level is reduced at
night and the eye shifts to a greater
blue/green peak sensitivity.
As shown in Figure 2, the peak sen-
sitivity of the human eye for night vi-
sion is a wavelength of 507 nanome-
ters (nm), and the peak sensitivity of
the human eye for daytime vision is a
wavelength of 555 nm. The region be-
tween these two peaks is referred to
as mesopic viewing conditions. Even
under mesopic viewing conditions,
yellow light sources have reduced ef-
fectiveness, while LED sources have
superior overall effectiveness.
To provide a mechanism to deter-
mine the true quality of light, the
author proposes the concept of a ratio
called the Quality of Light Factor
(QLF) to determine a rating for the
illumination provided by a fixture that
reflects actual lumens utilized by the
human eye. The author proposes that
the value of QLF can be simply defined
as a mininum ratio value for com-
parative purposes. For instance, if an
HPS lamp is rated with a QLF equal
55
 Dialight

Engineering Practice

to one, then the QLF of the LED lamp
would need to be rated as a minimum
of 2, as the strength of the scotopic eye
sensitivity of LED light would be a
number of times greater than the sco-
topic eye sensitivity of HPS source.
It is beyond the scope of this article
to recommend actual QLF values that
should be used for different sources
under different conditions. Nonethe-
less, as a general rule of thumb, the
strength of the scotopic eye sensitivity
of an LED is a number of times higher
than that of an HPS source. Thus,
when purchasing industrial lighting
systems, the QLF of an LED-based
system should be rated much higher
than the QLF of a conventional system,
such as one based on HPS lamps.
When it comes to safety shower and
eyewash locations, lights must be visi-
ble under both daytime and nighttime
conditions. The color chart provided
in Figure 2 shows that the best type
of lights for signaling safety showers
and eyewashes are those that provide
a wavelength in the mesopic range (as
described above). As shown in Figure
3, an LED-based light fixture is per-
fectly visible during daytime condi-
tions.
Candela. Candela is a measure of the
intensity of a light source. It relates
to the light focusing capabilities
of a given light fixture and is thus a
function of the light fixtures optical
design.
Coverage. Coverage is a measure of
the amount of effective light that is
spread out for the intended area-illu-
mination application. Adequate cover-
age is achieved through optical design
of the LED fixture, to ensure optimum
light distribution for industrial white
lighting (Figure 4).
Configuration. The configuration
of a balanced lighting layout refers
to the exact positioning of the fix-
tures (Figure 6). Proper configuration
within the industrial space is essen-
tial, and it is achieved by proper spac-
ing between the stanchion mounts of
industrial lights. The goal is to ensure
optimum area illumination that usu-
ally contributes to safety.
Longevity. Longevity is a measure
of how long-lasting a light source is,
and it is a direct contributing factor
to overall facility safety, since it re-
56 Chemical Engineering www.che.com April 2010
duces the need for maintenance
personnel to climb structures to
maintain or repair light fixtures
or replace bulbs. With this in mind,
the use of state-of-the-art light fix-
tures, such as those based on LED
technology, which tends to have
longer service life compared to
other types of conventional lamp
sources (incandescent, fluorescent
and HPS), typically translates
into improved safety and reduced
maintenance and lifecycle costs.
LED-based lights also tend to have
reduced energy requirements com-
pared to conventional lamp types,
and this helps to provide a rapid
return-on-investment.
Integrity. The integrity of the cho- lighting fixtures typically include cooling
sen lighting solution relates to a operation
fixtures optimized mechanical and
electrical design. For instance, LED
fixtures provide greater integrity
thanks to several modernizing ad-
vances that are now widely available,
such as universal input voltage capa-
bilities and much cooler temperature
ratings. Both of these aspects contrib-
ute to greater integrity, multi-applica-
tion capabilities and improved safety.
LED fixtures are also particularly
useful as a robust light source be-
cause they have no filaments, which
makes them particularly well-suited
for areas that may be exposed to
high vibration.
Additionally, in an LED system,
lumen maintenance is directly propor-
tional to thermal management. LED The functional advantages described
systems are typically designed with
cooling fins to compensate for heat
buildup during use (Figure 5).
Recommended practices
Given the discussion presented above,
the author offers the following rec-
ommendations for optimizing light
source efficiency using LED-based
fixtures for hazardous-area illumina-
tion. Although the example discussed
below focuses on a flare-gas-recovery
plant, the following recommendations
can be considered by most chemical
plants, petrochemical units and other Unlike conventional lights, todays
CPI plants.
LED technology provides better
directional control of light than is
possible using conventional incan-
descent, fluorescent or HPS lighting
Figure 5. Todays LED-based industrial
fins to help disseminate heat buildup during
options. This is important in appli-
cations such as gas-recovery units
where numerous railings straight
walkways and multi-level structures
need to be properly illuminated. Un-
like conventional lights, some of to-
days patented LED fixtures contain
several reflectors to further optimize
light distribution. The ability to ef-
fectively focus the light on straight
lines where it is needed helps to
reduce spillage of light beyond the
target focus area to just a few feet.
LED fixtures also require lower
wattage than conventional lighting
compared to incandescent, fluores-
cent or HPS lighting options
above also enable LED lights to be
spaced at a greater distance than
what would be required using con-
ventional light fixtures. As a result,
the use of LED fixtures can help
users to reduce the number of light-
ing fixtures required to illuminate
the same area by as much as 50%
compared to the use of conventional
lighting options. The ability to use
fewer lights reduces capital costs,
installation time and costs, mainte-
nance costs and overall energy re-
quirements
LED fixtures can easily be installed
using multiple mounting options
(that is, they can be wall-mounted or
stanchion-mounted and can be used
as side-by-side modules (Figure 6).
 ing and selecting competing lighting
options that are most appropriate for
industrial applications under both
daytime and nighttime applications.
With the introduction of new LED
lighting technologies in recent years,
it is now possible to provide an ideal
light source one that is designed to
take advantage of maximum retinal
efficiency in the human eye.
The recommendations provided
here aim to help CPI facilities to em-
ploy optimum color and light sources
that would be most compatible with
the physiology and functionality of the
human eye under daytime and night-
time conditions. As discussed, choos-
Note: Luminaire refers to a lighting fixture that is complete with installed lamps and other accessories ing the most appropriate color during
the selection of visual signal lights can
Figure 6. Shown here is the appropriate positioning of LED-based light fixtures ensure that workers will have clear
throughout a typical flare-gas-recovery unit at a petrochemical facility and easy access to safety showers and
other critical safety systems at any
time of the day or night.
LED lighting technology lends itself
well to industrial applications that are
considered classified hazardous areas.
A well-designed LED fixture takes
advantage of the directional charac-
terization of the LED and will deliver
more effective lumens to the targeted
application, therefore increasing the
overall system efficacy for area illumi-
nation at CPI plants.
The practical examples presented
demonstrate how the new LED light-
ing technologies in conjunction with
the QLF can contribute to improve-
ment in safety and maintenance and
save energy.
Edited by Suzanne Shelley
Acknowledgement
Figure 7. As seen in this photo, taken in the field at a gas-recovery unit, the use of
new LED-based lighting technologies can lead to tremendous visual acuity improve- The author wishes to thank Dialight
ments at petrochemical facilities by providing adequate light where it is needed the Corp. (Farmingdale, N.J.; dialight.
most (throughout the multi-level platform). Notice that the penetration of the light com) for its assistance during the de-
from the higher elevation of the structure down to the lower floor, and the distribution
of light throughout the entire structure provides optimum overall illumination velopment of this article.

Large facilities often stock a few
dozen types of lighting fixtures to
be used at different locations, so the
ability to standardize on such multi-
purpose LED fixtures can drasti-
cally reduce the amount of inven-
tory the facility must manage while
still providing the desired lighting
at different locations throughout
the facility
Appropriate lighting plays a critical
role in supporting overall industrial
safety (Figure 7). Understanding the
complex properties of the human vi-
sual system provides us with the an-
swers we need to develop better and
safer lighting fixtures that will have
significant impact on overall facility
safety throughout the CPI.
The lighting profile discussed here
which considers the complex inter-
actions between competing lighting
options and the human eye pro-
vides a new methodology for evaluat-
Author
Boris Viner is president of
Humatrack SafeLight (P.O,
Box 107, Avon, CT 06001;
Phone: 860-673-3498; Email:
boris.viner@humatrack.com).
He has worked in the light-
ing industry for more than
25 years. He holds a Masters
degree in electrical engineer-
ing and a Masters degree
in management Rensselaer
Polytechnic Inst. (RPI; Troy,
N.Y.). Viner has presented numerous papers at
technical conferences and engineering seminars,
and is a hazardous-area-lighting consultant
to industrial lighting company Dialight Corp.
(www.dialight.com).

Chemical Engineering www.che.com April 2010 57


Feature Report
Engineering Practice
Designing
Safer
Process Plants
Several often-overlooked strategies
to increase inherent safety are discussed here
Victor H. Edwards, P.E., Aker Solutions
M
any individuals and organi-
zations have made impor-
tant contributions to the cre-
ation of inherently safer (IS)
products, processes and process plants
[13]. A brief survey of successful case
histories shows that most reported
applications relied on only a few of
the core IS principles. This paper em-
phasizes the opportunities presented
by three particular and often-over-
looked possibilities for inherently
safer processes.
The methods proposed here ensure
integration of IS methods beginning
with process conception and continu-
ing through process plant engineering
design. Particular emphasis is given
to matching the IS principles with the
state of the project. For example, sub-
stitution is best applied during prod-
uct and process research, while limita-
tion of effects is most effective during
plot plan layout and equipment ar-
rangement.
The chemical process industries
(CPI) face the challenge of working
with processes and products that
present many hazards, such as
the following:
The manufacture of fuels uses and
produces products that burn with
significant energy release
Certain basic chemicals, such as
mineral acids and halogens are toxic
and/or corrosive
Many manufacturing processes ei-
ther release or require significant
44 Chemical Engineering www.che.com April 2011
energy transfer to achieve chemical
transformation
Some manufacturing processes pro-
duce benign products but require
hazardous chemical intermediates
in their manufacture
For these reasons, rigorous process
and product safety practices must be
used throughout the lifecycle of pro-
cess plants and must be applied to
their associated raw materials and
products. In recent years, this has
led to major efforts in green chemis-
try and engineering to develop prod-
ucts, manufacturing processes, and
plants that are safer for both people
and the environment.
Before green chemistry and engi-
neering achieved prominence, there
were pioneering insights in the de-
sign of safer process plants. Early ap-
proaches to safer processes often em-
ployed additional instrumentation and
procedures. These measures were often
helpful and necessary, but instrumen-
tation and operators can fail, especially
when faced with complexity.
Trevor Kletz [1] recognized that
What you dont have cant leak, when cluding with the detailed design. No
he first proposed the concept of the
inherently safer chemical processes
in 1977. His approach placed an em-
phasis on the inherent nature of the
process. Since then, important related
concepts such as product design for
safety and safer products, process and
plant lifecycles have also advanced.
Creation of IS processes has been the
8
6
5
4
3
2
1
1. Process design
2. Basic controls, process alarms and
operator supervision
3. Critical alarms, operator supervision
and manual intervention
4. Automatic action safety-instrumented
systems (SIS) or ESD
5. Physical protection (relief devices)
6. Physical protection (dikes)
7. Plant emergency response
8. Community emergency response
FIGURE 1. Shown here are some
typical layers of protection that can be
employed in a modern process plant [4].
At the core is an inherently safe process
design. Moving outward from the core,
the proposed options move through the
spectrum from inherent to passive to
active to procedural or administrative
controls, which are considered to be
progressively less reliable
objectives of a number of creative indi-
viduals and organizations since Kletzs
path finding proposal, with many no-
table successes.
Complete coverage of the entire prod-
uct/process/plant lifecycle is needed to
assure optimum health, safety and en-
vironmental performance of a chemi-
cal enterprise.
This article focuses on how to en-
sure maximum incorporation of IS
processes into the creation of a pro-
cess plant by beginning at the product
and process research stages and con-
effort is made to address the applica-
tion of inherently safer principles be-
yond plant design, although these are
also important.
Layers of protection
The classical onion diagram (Figure
1) illustrates the safety layers that
technical professionals throughout

FIGURE 2. The Flixborough tragedy ushered in a new era in process safety [6]
the CPI use to prevent process plant
incidents. This diagram helps to ex-
plain the following four basic process
risk-management strategies: Inher-
ent, passive, active, and procedural or
administrative
Inherent safety is at the core of the
onion the process design. A process
that cannot have a major fire, explo-
sion or toxic release is inherently safer
than one that could if one or more lay-
ers of protection were to fail.
Passive safety layers represent the
addition of such safety features as a
dike or a blast wall. Because passive
layers of protection require no active
intervention by a human or by a ma-
chine, they are deemed more reliable
than active layers of protection or
procedural layers of protection. None-
theless, the ability to make an explo-
sion impossible when possible is
clearly better than trying to mitigate
the effects of a potential explosion by
adding a blast wall.
Active layers of protection repre-
sent such features as the basic process
control system, a safety-instrumented
system, and mechanical interlocks.
Procedural or administrative
safety layers are generally considered
to be the least reliable and include op-
erating procedures and operator inter-
vention. Depending on the site-specific
hazard, procedural or administrative
controls may be entirely appropriate.
In general, the preferred ranking
of methods to control process risks is
shown below:
Inherent > passive > active > proce-
dural or administrative
Basic concepts
Inherently safer process concepts are
summarized below [1]:
Substitution
Minimization or intensification
Moderation or attenuation
Simplification
Limitation of (hazardous) effects
Avoiding knock-on effects
Making incorrect assembly impossible
Make status clear
Tolerance of error
Ease of control
Administrative controls or proce-
dures
In 2007, the Center for Chemical
Process Safety (CCPS) of the Ameri-
can Institute of Chemical Engineers
(AIChE) concluded that these eleven
basic concepts could be reduced to the
following four principles [2]:
Minimize
Substitute
Moderate
Moderate and simplify
This more concise set of principles
makes IS practices simpler to under-
stand and easier to apply. The excel-
lent new CCPS book (2009) goes on to
distinguish between first-order and
second-order IS:
First-order IS efforts change the
chemistry of a process
Second-order IS efforts change the
process variables
As can be seen by a survey of the pro-
cess safety literature, most published
work has applied one or more of the
first four concepts of the eleven cited
by Kletz and Amyotte [1] For this
reason, this article emphasizes three
other promising concepts.
Chemical Engineering www.che.com April 2011
Often-overlooked IS concepts
Three underutilized IS concepts are
presented here and illustrated with
examples:
1. Hybridization or transforma-
tion. One relatively new IS concept is
based on the recent innovative work
by Chen [5] who reports an inherently
safer process for the partial oxidation
of cyclohexane. Partial oxidation pro-
cesses often involve hazardous condi-
tions, as illustrated by the Flixborough,
England, tragedy in 1974 which
killed 28 people, destroyed a plant, led
to new process safety regulations, and
inspired Trevor Kletz to propose his
inherently safer design concept. The
Flixborough plant carried out liquid-
phase oxidation of large inventories of
hot cyclohexane in large pressurized
vessels. When containment was lost, a
large flammable vapor cloud formed,
ignited, and exploded with devastating
effect (Figure 2, from Mannan [6]).
The traditional cyclohexane-oxida-
tion process to produce a mixture of cy-
clohexanone and cyclohexanol (K/A oil
or ketone/alcohol oil) was operated at
low conversion rates (typically 35%)
to avoid formation of unwanted byprod-
ucts. The K/A oil was subsequently con-
verted into adipic acid and caprolactam
for the production of nylon.
Oxidation of cyclohexane with air
instead of oxygen is common practice
to reduce risks of transition from a
partial oxidation reaction to an un-
controlled deflagration in bubbles
or in the vapor space in the reactor.
Low conversions and reaction
rates led to large inventories of liquid
cyclohexane.
During systematic research on the
flammability and deflagration haz-
ards of cyclohexane, air and oxygen
mixtures, Chen [5] discovered that the
addition of a small amount of water
which is inert and does not par-
ticipate in the reaction helped to
inert the otherwise flammable vapors.
Cyclohexane and water are known to
form minimum-boiling azeotropes.
The increase in the vapor pressure of
the cyclohexane/water liquid results
from the increased vapor pressure of
the water. The water vapor inerts the
vapor mixture by lowering the upper
flammable limit of the vapor [5].
Chens work suggests that it will be
45

Engineering Practice
safe and practical to use pure oxygen
for cyclohexane oxidation. Benefits in-
clude both IS operation and improved
productivity. They also suggest that
this approach could be extended to
safer processes for partial oxidation of
other liquid hydrocarbons using pure
oxygen.
Chens approach is a first-order IS
process innovation because it changes
the chemistry of the gas phase in a
gas-liquid reaction and prevents the
unwanted side reaction of combustion
from occurring in the gas phase.
Although reference [5] did not claim
to have demonstrated a new IS con-
cept, Chens work is different from the
classical definition of the Substitute
principle because the same reactants,
chemical reactions, and products are
involved. If the name Substitute were
broadened to names such as Change
in Chemistry or Hybridize, then it
could be lumped in with the many suc-
cessful applications that are possible
when using the Substitute concept.
Chens innovation permits rapid cy-
clohexane oxidation at lower tempera-
tures and pressures, and could thus
be said to be an example of the inher-
ently safer principle Moderate. How-
ever, Chens approach enables more
moderate conditions by narrowing the
flammability limits through the addi-
tion of a new component, water. It is
thus an example of supplementation
or hybridization.
Although not proposed by Chen [5]
himself, his work suggests that there
may be many other opportunities
for transformation or hybridization
of other potentially hazardous reac-
tions to make them inherently safer.
Although water would be high on any-
ones list as a potentially transform-
ing additive, it probably will not help
many potentially hazardous reactions.
However, there are many other chemi-
cals that may be inert to the reaction
and thus also be capable of inerting the
vapor phase involved in an otherwise
reactive liquid-vapor reaction. For in-
stance, there are many examples of
azeotropic mixtures in the literature
and there are many compounds that
could prove inert to oxidation reac-
tions (such as, certain halocarbons).
Applications are not limited to
partial oxidation with air or oxygen;
46 Chemical Engineering www.che.com April 2011
Qheat generated
E
B
Qout
Qout Heat removed
3
Qout
Btu/h
D
other oxidations include
1
chlorination and bromi- 2
nation reactions, for ex- C
ample. And there may be A
other examples of vapor-
liquid reactions, such as
Reactor temperature
hydrogenation reactions,
where addition of a new FIGURE 3. Heat-generation (Q
heat generated) and heat-
chemical could improve removal (QOut) rates as a function of reactor temperature
the safety of the process. for three different heat-removal designs [9]. Heat gen-
Addition of an ad- eration is equal to heat removal at points A, C, D, E, and
ditional compound to a B, so steady state operation is possible. However, the
reactor is not stable at point D without the addition of
reaction mixture to min- controls or a modification of the design
imize hazardous reac-
tions may add complexity to the puri- in the process variables and is a form
fication process, but it may be justified of Moderate, a second-order inherently
by the increased safety. safer design.
Chens [5] paper on cyclohexane CCPS [2] also cites the work of
oxidation illustrates transformation Luyben and Hendershot [8] that high-
or hybridization, in which the basic lights how minimization or intensifi-
chemistry is maintained, but the ad- cation in a reaction system that is in-
dition of another chemical component tended to improve process safety may
transforms a potentially hazardous re- lead to less robust processes with the
action process into a much safer one. opposite effect.
2. Create a robust process to sta- I propose here that Stabilize or En-
bilize or ensure dynamic stability. sure Dynamic Stability be added to
Not all process designs are inherently the list of IS concepts to be sure that
stable, and if the process design is to be it is not overlooked in the quest for in-
safe, the process engineer must ensure herently safer processes.
dynamic stability as well as ensuring Application of some of the other IS
that the steady-state mass and energy principles can adversely affect the dy-
balances are achieved. A number of namic stability of a process. For exam-
processes exist that have narrow safe- ple, reduced liquid inventories (Mini-
operating limits but have been made mize) in a distillation train make the
stable by the addition of control sys- process inherently safer from one per-
tems. Dynamic stability and control spective because the smaller process
of chemical processes has been exten- inventory decreases the consequences
sively studied [7]. of loss of containment. However, the
Designing the process to be more smaller inventory also shortens the
inherently stable to process upsets response time of the distillation sys-
with and without control systems is tem to process upsets, increasing the
clearly inherently safer, although this risk that the basic control system will
principle is not addressed in most dis- not be able to restore the distillation
cussions of IS. The IS principle Ease of system to the desired operating condi-
Control has usually been interpreted tions and avoid a potentially unsafe
to mean a process with a control sys- operating condition and/or an un-
tem that the operator can understand scheduled process shutdown [2].
clearly and manage effectively. Chemical reactors carrying out
CCPS briefly mentions the advan- exothermic chemical reactions are
tages of designing processes that are perhaps the best known examples of
inherently more stable or robust [2]: processes that can be dynamically
It is inherently safer to develop unstable. Harriott [9] provides the il-
processes with wide operating limits lustration of an irreversible first-order
that are less sensitive to variations in chemical reaction being conducted in
the operating parameters...Sometimes a continuous-flow, stirred-tank reactor
this type of process is referred to as a (CSTR). Figure 3 shows the heat-gen-
forgiving or robust process. eration rate by the chemical reaction
Designing a robust process increases as a function of reactor temperature.
inherent safety by imposing a change Heat-generation rates are low at low
 Tools for Inherently Safer Process Plant Design
P rocess hazards reviews
Chemical interaction matrices
Dow Fire and Explosion Index and
Chemical Exposure Index
Fire, explosion and toxic-release product and process research, and safety instrumented systems
consequence modeling and risk development and design
assessments
Layer of protection analysis
Spacing tables for units and for computer models
process equipment
temperatures, but as temperature
increases, the reaction rate increases
rapidly because of the exponential
dependence of the reaction rate co-
efficient on temperature. At higher
reactor temperatures, the shrinking
concentration of reactant (due to con-
version to product) reduces the reac-
tion rate and partially overcomes the
still-increasing reaction-rate coeffi-
cient. The heat-generation rate even-
tually reaches a constant maximum
value when the reaction has reached
complete conversion.
Figure 3 also shows three different
straight lines for the heat-removal
rate from the reactor for three differ-
ent reactor-cooling-system designs. To
achieve a steady-state energy balance,
the rate of heat generation (Qheat gen-
erated) by the chemical reaction must
equal the rate of heat removal (Qout)
by the reactor cooling system. That
energy balance occurs when the heat
generation curve intersects the heat
removal curve (where Qheat generated
= Qout). In Figure 3, the three differ-
ent heat-removal-rate lines intersect
the reactor heat generation rate curve
at five points. At four of these points
(A, B, C, E), the steady-state energy
balance solution is stable. At each of
these points, if there is an increase in
temperature, the rate of heat removal
increases more rapidly than the rate
of heat generation by the reaction and
the reactor temperature tends to re-
turn to the desired operating point.
Similarly, if the temperature drops
slightly at one of these four operating
conditions, the rate of heat removal
decreases more than the rate of heat
generation by the reactor and the tem-
perature trends back up to the desired
operating condition.
In contrast, point D in Figure 3 is an
inherently unstable operating condi-
tion even though the steady state rate
of heat generation by the reactor equals
the rate of heat removal by the reactor
cooling system. At point D, an increase
in reactor temperature increases the
rate of heat generation by the reactor
Dynamic process simulation
Inherent safety analysis
P eriodic design reviews during Safety integrity level assessments
R
 eviews of plant siting, plot plan,
equipment arrangement and 3-D Safety case development
more than it increases the rate of heat
removal by the reactor cooling system,
so the reactor temperature increases
more instead of cooling back to the de-
sired operating point.
This further increase in reactor tem-
perature then leads to an even larger
rate of heat generation rate by the
reactor and additional heating of the
reactor. Without any effective control
actions, the reactor temperature will
tend to increase to point E in Figure 3
before it stabilizes.
Similarly, in Figure 3 a decrease in
reactor temperature at point D could
eventually lead to the reactor temper-
ature and conversion dropping back to
point C.
Clearly, of the three reactor cooling-
system designs represented by the
three straight lines in Figure 3, the
reactor cooling system represented by
line CDE is the least desirable from
a dynamic-stability perspective. Ad-
dition of an effective control system
might be able to provide dynamic sta-
bility but at the cost of installation
and maintenance of the control sys-
tem and at the cost of residual risk if
the control system fails.
Another example of potential
sources of process instability results
from efforts to improve energy effi-
ciencies in distillation trains through
heat integration. In these cases, the
feed to a column may be preheated
by the bottoms product of a second
downstream column. This may in-
crease the risk of process upsets due
to increased interactions between the
two columns.
While avoidance of add-on controls
has always been a goal of inherently
safer design, achievement of that goal
has seldom mentioned the concepts
of Ensure dynamic stability or Stabi-
lize as tools of the process engineer. It
should be considered when consider-
ing other means to assure inherently
safer processes during process design.
The process engineer should work
closely with the control systems engi-
neer to address the dynamic stability
Chemical Engineering www.che.com April 2011
O ccupied building evaluation
and design
Area electrical classification
Human factors reviews
Ergonomics reviews
The design process itself
of both the uncontrolled process and
the controlled process to ensure a ro-
bust process.
3. Limit hazardous effects during
conceptual and detailed engineer-
ing. David Clark published a seminal
paper [10] on the limitation of effects
when siting and designing process
plants. He reminds us that there is a
strong, non-linear decrease of fire, ex-
plosion, and toxic effects with separa-
tion distance. Comparatively small de-
creases in separation distance have a
major effect, while larger increases in
separation offer diminishing returns.
Methods, such as the Dow Fire and
Explosion Index [11] and the Dow
Chemical Exposure Index [12, 13], pro-
vide quantitative screening estimates
of the hazards from various parts of a
chemical process. Other indices have
been developed and evaluated to per-
form a similar objective to the Dow in-
dices [1, 2, 14]. These screening tools
can identify those parts of a process
where increased separation distances
are needed to limit potential escala-
tion of an incident.
In one typical plant design, a 10%
increase in separation distances for
all units increases total plant invest-
ment cost by only 3%. Similarly, dou-
bling the separation distance for a
hazardous unit representing 10% of
the investment cost of the plant would
cost only 3% more. Because of the non-
linear effect of separation distance,
doubling the separation distance for a
hazardous unit could reduce explosion
overpressures on the adjacent units
by a factor of four or more.
The strong decrease in hazardous
effects with modest increases in sepa-
ration distances will often more than
justify increased capital cost.
Spacing also offers important ben-
efits in crane and other maintenance
access, ergonomic advantages and
decreased risk of incident escalation.
Future plant expansions or process
improvements are also facilitated, al-
though expansions that decrease spac-
ing may increase hazardous effects.
47
 Engineering Practice
Applying different IS principles
As discussed, the different IS prin-
ciples are best applied at different
stages of the process plant timeline.
Although IS checklists are often used
at the screening process hazards anal-
ysis (PHA) level, much more is needed
throughout the development and de-
sign of a process plant.
For example, Substitute is best
done during the product and process
research phases before significant
investments of time and resources in
a particular product and process are
made. Hybridize or Transform is best
done during process research and de-
velopment, as is Moderate.
Minimize, Simplify, and Error tol-
erance have the best result when ap-
plied during the process development,
conceptual design and detail design
phases. Stabilize or Ensure Dynamic
Stability is also best done during de-
sign development.
Limitation of effects, which is closely
related to passive protection, has its
greatest impact during development
of the plot plan and equipment ar-
rangement.
IS processes and plants
As mentioned previously, the CCPS [2]
defines two levels of inherent safety:
First-order inherent safety results
References
1. Kletz, Trevor A., and Amyotte, Paul, Process
Plants a Handbook of Inherently Safer De-
sign, 2nd Ed., Taylor and Francis, Philadel-
phia, PA, 2010.
2. Center for Chemical Process Safety (CCPS),
Inherently Safer Chemical Processes A
Life Cycle Approach, 2nd Ed., AIChE, New
York, NY, 2009.
3. Hendershot, Dennis C., An overview of inher-
ently safer design, Process Safety Progress,
Vol. 25, No. 2, 98107, June 2006.
4. Dowell, III, Arthur M., Layer of protection
analysis and inherently safer processes, Pro-
cess Safety Progress, Vol. 18, No. 4, 214220,
Winter 1999.
5. Chen, Jenq-Renn, An inherently safer process
of cyclohexane oxidation using pure oxygen
An example of how better process safety leads
to better productivity, Process Safety Progress,
Vol. 23, No. 1, 7281, March 2004.
6. Mannan, Sam, Ed., Lees Loss Prevention in
the Process Industries, 3rd Ed., Elsevier But-
terworth Heinemann, Oxford, U.K., 2005.
7. Edgar, Thomas F., and others, Process Control,
Section 8 in Perrys Chemical Engineers Hand-
book, 8th Edition, Don W. Green, Editor-in-Chief,
McGraw-Hill Book, New York, NY, 2008.
8. Luyben, W.L., and Hendershot, D.C., Dy-
namic disadvantages of intensification in
inherently safer process design, Industrial
Engineering Chemistry Research, Vol. 43, No.
2 (2004) cited in CCPS, 2009.
9. Harriott, Peter, Process Control, McGraw-
Hill, New York, NY, 1964.
48 Chemical Engineering www.che.com April 2011
from changes in the chemistry of a
process that reduces the hazards
of the chemicals used or produced.
Substitute or Hybridize efforts lead
to first-order inherent safety
Second-order inherent safety results
from changes in the process vari-
ables. Examples include Minimize,
Simplify and Stabilize the opera-
tions.
It is also helpful to distinguish be-
tween IS processes and IS plants.
Even when hazards cannot be
eliminated from the chemistry of
the process, the plant using the po-
tentially hazardous process can be
made inherently safer through ju-
dicious design.
Note also that even with IS process
chemistry, it is essential to employ
IS principles during the process and
plant design to ensure an IS plant.
Tools for IS plant design
There are a number of tools available
to aid in designing process plants that
are inherently safer (Box, p. 18). Al-
though inherently safer reviews are a
valuable tool for identifying opportu-
nities for improvement, it is important
to keep the principles of inherently
safer in mind throughout the design
process. n
Edited by Suzanne Shelley
10. Clark, David G., Applying the limitation of ef-
fects inherently safer processing strategy when
siting and designing facilities, Process Safety
Progress, Vol. 27, No. 2, 121130, June 2008.
11. Dows Fire and Explosion Index Hazard Clas-
sification Guide, 7th Ed., American Institute
of Chemical Engineers, New York, NY, 1994.
12. Dows Chemical Exposure Index Guide,
American Institute of Chemical Engineers,
New York, NY, 1994.
13. Suardin, Jaffee, Mannan, M. Sam, and El-
Halwagi, Mahmoud, The integration of Dows
Fire and Explosion Index (F&EI) into process
design and optimization to achieve inherently
safer design, Journal of Loss Prevention in the
Process Industries, Vol. 20, pp. 7990, 2007.
14. Khan, Faisal I., and Amyotte, Paul R., How
to make inherent safety practice a reality,
Canadian Journal of Chemical Engineering,
Vol. 81, No. 2, 216, February 2003.
Additional suggested reading
1. Edwards, David, Editorial Special Topic
Issue Inherent safety Are we too safe for
inherent safety?, Process Safety and Envi-
ronmental Protection Transactions of the
Institution of Chemical Engineers Part B,
Vol. 81, No. B6, 399400, November 2003.
2. Englund, Stanley M., Inherently safer plants:
Practical applications, Process Safety Prog-
ress, Vol. 14, No. 1, 6370, January 1995.
3. French, Raymond W., Williams, Donald D., and
Wixom, Everett D., Inherent safety, health, and
environmental (SHE) reviews, Process Safety
Progress, Vol. 15, No. 1, 4851, Spring 1996.
Acknowledgments
I gratefully acknowledge the process safety in-
sights from my colleagues at Aker Solutions and
at the leading operating companies whose facili-
ties we have helped to design, from Professors
Sam Mannan, Trevor Kletz, Ron Darby, Harry
West and the Mary Kay OConnor Process Safety
Center at Texas A & M University, and from
many others in the community of process safety
professionals. The financial support of Aker So-
lutions is also appreciated.
Author
Victor H. Edwards, P.E.,
is director of process safety
for Aker Solutions Ameri-
cas Inc., (3010 Briarpark
Drive, Houston, TX 77042;
Phone: 713-270-2817;
Fax: 713-270-3195; mail:
vic.edwards@akersolutions.
com). In his 28 years with
Aker, Edwards experience
includes process engineering,
safety management and pro-
cess, biochemical and environmental technolo-
gies. He has received numerous accolades in the
areas of safety and environmental engineering,
including five DuPont awards, and has contrib-
uted extensively to the engineering literature.
His earlier experience includes assistant pro-
fessor of chemical engineering at Cornell Uni-
versity, an assignment at the National Science
Foundation, pharmaceutical research at Merck,
alternate energy research at United Energy
Resources, visiting professor at Rice University
and process engineering at Fluor Corp. Edwards
earned his B.A.Ch.E from Rice University and
his Ph.D. in chemical engineering from the Uni-
versity of California at Berkeley. A registered
professional engineer in Texas, he is an AIChE
Fellow, and a member of ACS, AAAS, NFPA,
NSPE, and the N.Y. Academy of Sciences.
Note: This article is based on a paper presented
at the Mary Kay OConnor International Sym-
posium, Texas A & M University, October 27-28,
2009.
4. Gupta, J.R., and Edwards, D.W., Inherently
safer design Present and future, Process
Safety and Environmental Protection Trans-
actions of the Institution of Chemical Engineers
Part B, Vol. 80, 115125, May 2002.
5. Gupta, J.R., Hendershot, D.C., and Mannan,
M.S., The real cost of process safety A clear
case for inherent safety, Process Safety and
Environmental Protection Transactions of
the Institution of Chemical Engineers Part
B, Vol. 81, No. B6, 406413, November 2003.
6. Hendershot, Dennis C., et al., Implementing in-
herently safer design in an existing plant, Process
Safety Progress, Vol. 25, No. 1, 5257, March 2006.
7. Kletz, Trevor A., Inherently safer design: The
growth of an idea, Process Safety Progress,
Vol. 15, No. 1, 58, Spring 1996.
8. Lutz, William K., Take chemistry and phys-
ics into consideration in all phases of chemi-
cal plant design, Process Safety Progress, Vol.
14, No. 3, 153160, July 1995.
9. Lutz, William K., Advancing inherent safety
into methodology, Process Safety Progress,
Vol. 16, No. 2, 8688, Summer 1997.
10. Maxwell, Gary R. Edwards, Victor H., Robert-
son, Mark, and Shah, Kamal, Assuring process
safety in the transfer of hydrogen cyanide man-
ufacturing technology, Journal of Hazardous
Materials, Vol. 142, pp. 677684, June 2007.
11. Overton, Tim and King, George M., Inher-
ently safer technology: An evolutionary ap-
proach, Process Safety Progress, Vol. 25, No.
2, 116119, June 2006.
12. Study, Karen, A real-llife example of choosing an
inherently safer process option, Process Safety
Progress, Vol. 25, No. 4, 274279, December 2006.
 Solids
Environmental
Processing
Manager
A Safety Checklist
For Laboratories
These nine best practices for managing change in
laboratories can help ensure a safe workplace
Lori Seiler
The Dow Chemical Company
L
aboratories, such as those
used for quality control in a
production environment and
those in research and devel-
opment, are an integral part of the
chemical process industries (CPI).
Everyday demands of business can
easily overtake safety when it comes
to setting priorities in the laboratory.
Deadlines must be met. Samples
must be analyzed. Questions must
be answered and quickly.
But even for the most experienced
chemists and laboratory workers,
safety must remain at the forefront.
This is especially true when new
chemicals, employees or equipment
are introduced. Changes such as
these are part of life in the labora-
tory, yet change represents risk. And
as such, it is an element that must
be thoroughly incorporated into
every CPI laboratorys safety plan.
Read on for nine best practices
from identifying hazards to com-
municating effectively to ensure
that your laboratory is as safe as
possible amid inevitable and ongo-
ing change.
1. Tackle change upfront
As the old saying goes, nothing is
constant but change. And thats
doubly true for work in the labora-
tory. In fact, change defines chemis-
try. So how do we plan for it?
Take time upfront to carefully as-
sess the impact that every change
will have on safety, health and en-
vironmental issues. Laboratory
operations designed for safety also
generate better, more reproducible
results, without placing people or
equipment at risk.
54 Chemical Engineering www.che.com August 2014
All project changes that are
made in the laboratory should be
first evaluated for potential haz-
ards. Having a process in place to
evaluate and manage change is
extremely helpful. Inviting oth-
ers to get involved as you work
through the process is smart, as
well. Your supervisors, advisors,
co-workers and environmental,
health and safety (EH&S) ex-
perts can put their knowledge and
experience to work by helping to
identify potential hazards and of-
fering ways to mitigate those risks.
2. Identify potential hazards
You cant manage what you dont
know or understand. Using a
change-management process, you
can identify potential hazards and
mitigation solutions associated
with change. Reviewing predefined
triggers or typical changes that
are known to create hazards can
help as you go through laboratory
changes step-by-step. The triggers
quickly highlight changes that need
further consideration.
A Hazard Assessment Trigger
Grid and Safe Operations Card
(SOC) are two tools that can be
used to identify potential hazards
[1]. SOCs allow you to evaluate haz-
ards as you plan your laboratory
project, thereby helping to ensure
that any potential hazards are fully
identified and assessed early on.
They also document appropriate
operating ranges, conditions and
emergency response in the event of
an accident.
Documenting potential hazards
and evaluating how you will miti-
gate the risks are key components
in managing process change. This
ensures that others know about
the changes, as well as their poten-
tial hazards and impact. The more
everyone in the laboratory under-
stands, the safer the work environ-
ment in the laboratory becomes.
3. Rely on a second set of eyes
Your colleagues can often see what
you cant. The truth is that we be-
come so familiar with our own work
that we often overlook the obvious.
Inviting a co-worker or supervisor
to be your second set of eyes during
a change review can help identify
and address potential issues.
When sharing your situation with
a reviewer, dont just sit in an office
and talk about your changes with
the other person. Go where the ac-
tion is, like the laboratory hood or
bench. Tell and show your review-
ers exactly what you will be doing
(Figure 1). With everything set up
and in sight, you and your review-
ers may see things that can be
modified to further enhance safety.
Reviewers should ask open-ended
questions to best understand the
work to be completed and potential
worst-case scenarios, as well as to
 Safety Inspections: The Kindness Factor

I
t is amazing what you can find when you take time to look, ask
and connect with those around you. During times of change,
which can introduce extra stressors into the environment, this
element of connection becomes extremely important in ensuring
that communication is clear and effective. This is true throughout
laboratory work, but is notably important during inspections.
Inspections help identify and mitigate safety discrepancies, and
they can be formal or informal, depending on what is needed. Yet
inspections are more than checking off items on a checklist to see
what is good or bad. They are an opportunity to ensure that ment every step of the way.
everyone in the laboratory clearly recognizes and understands
what is needed to maximize laboratory safety.
In fact, one of the most important components of inspections is
the interaction and dialogue that will take place between you,
your co-workers and supervisor. That is why it is a must to make
a personal connection during the inspection. Psychologist E.
Scott Geller says that actively caring means practicing system-
atic and purposeful acts of kindness to keep other people safe
and healthy.
When you and your co-workers care about safety and each
other you will want to learn how to improve. You will be ac-
tively involved. So create an environment in your workplace where
people are free to ask questions and know how to escalate issues
that become a problem. Make sure that everyone in the laboratory
knows that no matter what, safety must be planned into the experi-
Finally, after taking the time for inspections and improvements,
remember to recognize success. As improvements are made, or if
a minimum number of safety discrepancies are found, let everyone
know. Be loud and proud about your successes in incorporating
new chemicals, materials, equipment and new people into your
work environment. 

understand the layers of protection
and safety measures in place to mit-
igate risks.
Youll find that both you and your
project will benefit when you involve
a variety of people with different
expertise and experience early on.
When considering who to approach
for this role, seek to match your re-
viewers with the type of change that
youre planning. For instance, if you
are installing new equipment, get
people who are familiar with that
type of equipment. If you are using
new chemistries, get people who
are knowledgeable about those. Use
their knowledge to your advantage.
An added bonus to the review is
that this type of cross-functional
involvement creates networking
opportunities. You will get to know
people throughout your department
and in other functions, as well.
4. Understand the risks
When new chemicals enter the
laboratory, the risks increase. This
situation calls for special planning
to account for how you will plan
your experiment including what
equipment youll use, how you man-
age your waste and decontamina-
tion considerations.
To provide an example, lets say
that you are beginning new labora-
tory work that involves nanomate-
rials. Your preplanning research
would tell you that these ultra-
small materials could be absorbed
by the body faster than other ma-
terials, thereby resulting in greater
exposure. Also, they may travel to
internal organs that were not previ-
ously accessible to the larger-scale
particles of the same material.
As part of the preplanning pro-
cess, you would want to assess
potential worker exposure to any
nanomaterials that will be used by
considering tasks that may poten-
tially expose you or fellow labora-
tory workers to nanomaterials. You
would consider the dustiness of the
material, how it will be used, how
much will be used, and the duration
and frequency of the task.
Once the preplanning is finished,
move on to safety equipment con-
siderations. Will a fume hood be
needed? Is your standard personal
protective equipment (PPE) enough
to keep you safe? Continuing with
the nanomaterials example, you
would want to ensure that you keep
the materials off your skin and out
of your lungs and eyes. As such, ven-
tilation systems would be needed to
capture and remove airborne nano-
materials before they are inhaled
by workers. Depending on the scale
of the operations, ventilated enclo-
sures, local exhaust ventilation or
other types of ventilation could be
needed. Go through this planning
process for any new material you
introduce, devoting extra time to
evaluate your safety when dealing
with pyrophoric materials, hazard-
ous chemicals, cryogenic liquids and
reactive chemicals.
When considering disposal of new
chemicals or materials, always be
sure that different categories of waste
are stored in different containers
and are clearly separated, and that
incompatible chemicals are stored
separately. The laboratory is not the
place for unexpected reactions.
5. Know your equipment
When a new material is introduced
to the laboratory, new equipment
Chemical Engineering www.che.com August 2014
often must be used. From fume
hoods and glassware to gas cylin-
ders, nearly every piece of equip-
ment in a laboratory can pose a
hazard when used inappropriately.
Be sure to think about and plan for
how youll handle tasks like equip-
ment set-up, maintenance and
cleaning. The hazards associated
with these types of support tasks
(such as ergonomics, decontamina-
tion, energy control and more) are
often underestimated.
Not only must you make sure that
you are completely up to speed on
the safest way to operate the equip-
ment especially if its been some
time since youve received training
but you also must ensure that
others, including new employees,
are up to speed as well. In the rush
to find results for an experiment,
shortcuts may seem appealing, but
proper care always must be taken
even more so when change is afoot.
If you are uncertain about how to
use a piece of equipment, seek out
training. For quick refreshers, re-
sources are available on a variety of
topics [2].
6. Take care when working alone
On a busy day at work on a new
project, you may be tempted to stay
late to make additional progress. If
you find yourself in this situation,
you must first look at the work
youll be doing and evaluate its haz-
ard potential before deciding that it
is safe to work alone.
If what youre doing is classi-
fied as a high-hazard operation,
its never safe to work alone, even
if special precautions are in place.
Some examples of high-hazard op-
erations include work with process
55
 Environmental Manager
plants, any equipment that com-
presses, and work that involves
high quantities of flammable or
extremely hazardous materials.
These kinds of operations require
two or more people, including a
safety observer or someone who
can provide immediate help in case
of an emergency.
But even low-hazard operations
come with risks. In general, low-
hazard operations include routine
analytical work, such as titration,
handing samples for routine analy-
sis, minor maintenance, general
laboratory work and office work.
These operations dont present any
major risk to safety and are gener-
ally safe, as long as you have a plan
in case of a problem. A mechanism
like a lone operator alarm or emer-
gency radio should be in place.
7. Practice good housekeeping
In the laboratory, everything you
do, or dont do, makes a difference
in the outcome of your experiments
and your safety. That includes keep-
ing the laboratory clean. Whether
youre introducing change into the
laboratory or not, youre much more
likely to notice when something is
amiss when everything is in order.
Use a checklist to perform in-
spections on at least a monthly
basis. Ensure your space is free
of tripping hazards, unused ma-
terials, excess paper and clutter.
Check to make sure that the covers
for drains and trench drains are in
place. Also, pay special attention
to surfaces including benches,
floors and operating equipment
to see that they are free of hazard-
ous dust accumulation.
Part of your checklist should
focus on safety equipment: Is ev-
erything in place and function-
ing as it should be? Be especially
vigilant when it comes to personal
protective equipment, respirators,
eyewash, safety shower, fire ex-
tinguishers and telephone. Once
your laboratory inspection is done,
openly communicate any problems
you see, especially unsafe prac-
tices and procedures. (See side-
bar. p.55: Safety Inspections: The
Kindness Factor.)
56 Chemical Engineering www.che.com August 2014
8. Communicate clearly
Communication is critical to ef-
fective laboratory safety in any
situation, but especially when any
change is introduced to your labora-
tory. While this may seem obvious,
poor communication is the reason
for far too many laboratory acci-
dents. Thats why it is absolutely es-
sential that you communicate with
everyone who could be impacted by
any change.
Your plans must always be shared
with anyone who could be impacted
by them. Be diligent about this.
Discuss the experiments you will
be conducting, the potential haz-
ards, worst-case scenarios and any
planned emergency response. As
project owners, be receptive to the
advice and safety improvements sug-
gested by colleagues. As reviewers,
interact and share your expertise
in a way that influences safety best
practices. The engagement should
be a positive interaction where risks
are identified and solutions to miti-
gate these risks are identified.
The aforementioned SOC is a
great tool for helping to plan for
and communicate potential haz-
ards connected to process change.
SOCs establish accurate commu-
nication between you and every-
one else in the laboratory and
that is one of the most important
things you can do to ensure a safe
working environment.
9. Make change part of the culture
In order to ensure that change
management is always part of the
plan, a strong safety culture must
permeate the laboratory. And while
everyone is essential to the safety
culture, it starts with leaders.
Experience shows that leaders
FIGURE 1.
Pre-startup
safety reviews
are an important
tool to help en-
sure laboratory
safety
who actively engage with their
teams help deliver outstanding re-
sults and improved performance.
They are able to build, strengthen
and engage their teams. This is cer-
tainly true for safety performance.
Even if you are not a laboratory
leader, you have a big role in pro-
moting leadership engagement.
First, help your safety leader in
identifying key systems needed,
such as programs, practices, proce-
dures and training. As your leader
identifies key behaviors required to
achieve the desired goals, be open
and willing to adjust, use and cham-
pion those behaviors.
Change and safety are constant
Your personal safety, as well as the
safety of your co-workers, depends
on your commitment to manag-
ing change wisely. Remember that
change is the main reason to stop
and evaluate potential hazards.
Whether youre changing chemi-
cals, equipment or people, you can
manage it successfully and safely.
Edited by Dorothy Lozowski
References
1. Free downloads of these tools are available
under Resources at the Dow Lab Safety
Academy, http://safety.dow.com).
2. The Dow Lab Safety Academy (http://safety.
dow.com) offers short tutorial videos on a va-
riety of topics.
Author
Lori Seiler is the associate
director for environmental
health & safety for Dow Re-
search and Development at
The Dow Chemical Company
(2030 Dow Center, Midland,
MI 48674; Email: lseiler@dow.
com). Seiler is a certified in-
dustrial hygienist (CIH) and
holds a Masters degree from
the University of Massachu-
setts at Amherst. She has
been closely involved in Dows university labora-
tory safety initiative and the creation of the Dow
Lab Safety Academy (http://safety.dow.com).
 Feature Report

Engineering for Plant Safety

Early process-hazards analyses can lead to potential cost savings in
project and plant operations
Sebastiano
Giardinella
and Alberto
Baumeister
Ecotek group of
companies
Mayra
Marchetti
Consultant

In Brief
CPI project lifecycle

Process Hazard
Identification

T
When to use a given
he chemical process industries (CPI)
method
handle a wide variety of materials,
Safe-design options many of which are hazardous by na-
addressing hazards ture (for example, flammable, toxic
early or reactive), or are processed at hazardous
conditions (such as high pressures or tem-
final remarks
peratures). The risks associated with CPI
facilities not only extend to the plant person-
nel and assets, but can potentially affect the
surrounding population and environment
sometimes with consequences having re-
gional or international scale, as in the case of
toxic vapor or liquid releases.
It is for this reason that process safety is
recognized as a key element throughout the
entire life of the plant, and several industry and
professional associations and government
authorities have issued norms, standards and
regulations with regards to this subject.
Process safety, as defined by the Center
for Chemical Process Safety (CCPS), is a
discipline that focuses on the prevention and
mitigation of fires, explosions and acciden-
tal chemical releases at process facilities.
Excludes classic worker health and safety
issues involving working surfaces, ladders,
protective equipment and so on. [1] Process
safety involves the entire plant lifecycle: from
visualization and concept, through basic
and detailed engineering design, construc-
tion, commissioning, startup, operations, re-
50 Chemical Engineering
vamps and decommissioning.
In each of the plant life phases, different
choices are made by engineers that have a
direct impact on the overall risks in the fa-
cility; however, the highest opportunities for
cost-effective risk reduction are present in
the earlier phases of the project. In contrast,
the cost of implementing changes in the later
stages of the project increases dramatically.
Hence, it is important for the design team to
identify risks, and implement effective design
solutions as early as possible.
This article covers some of the typical de-
cisions that the project design team has to
make over the course of a project, with ex-
amples of how the incorporation of process
safety throughout the entire design process
can significantly reduce the risk introduced
by a new CPI facility, while also avoiding po-
tential cost-overruns, or unacceptable risk
scenarios at later stages.
CPI project lifecycle
A project for a new chemical process facility
usually involves different phases, which are
outlined here:
A screening or visualization phase. In this
phase, the business need for the plant is
assessed. Typical choices at this stage in-
volve defining plant throughput, processing
technology, main blocks and plant location
www.chemengonline.com august 2015
Project Life Figure 1. The relative influence of decisions on
total life cost, and cost of implementing changes
throughout the project lifecycle

Influence of
earlier in the project lifecycle have the
decisions on greatest impact on the total plant life
total life cost
cost; in contrast, the cost of imple-
menting changes in the later stages
Cost of of the project increases dramatically,
implementing
changes
as can be seen on Figure 1.
The same holds true for overall
plant risk, as the impact of decisions
on overall facility risk is greatest in
the earliest stages of the project.

Risks and hazards

A risk can be defined by a hazard,
its likelihood (or probability) of occur-
rence, and the magnitude of its con-
sequence (or impact).
Visualization Conceptual Basic Engineering, procurement & construction Operations A hazard, as defined by the Center
engineering engineering
for Chemical Process Safety (CCPS),
(high-level), with the goal of develop- bulk materials, the execution of the is an inherent chemical or physical
ing a high-level project profile, and a construction work, the pre-commis- characteristic that has the potential
preliminary business case based on sioning, commissioning and startup for causing damage to people, prop-
ball-park estimates, benchmarks of the facilities. erty or the environment [2].
and typical performance ranges, in Table 1 shows typical engineering Process hazards can be classified
order to identify project prospects. deliverables, along with their degree in terms of the following:
A conceptual engineering phase. of completion, for each phase of 1. Their dependence on design
In this phase, the design team fur- project development. choices:
ther develops the concept of the After the plant construction is fin- Intrinsic not dependent on
plant, leading to a more-defined ished, the facility enters the opera- design decisions (that is, always
project description, an improved tions phase. At the end of its life, the associated with the operation or
capital-cost estimate, and a more- plant is decommissioned. process). For instance, hazards
developed business model. At this It is a generally accepted fact in proj- associated with the chemistry of
stage, the process scheme is de- ect management that decisions made the materials being handled (flam-
fined, along with the characteristics
of the major pieces of equipment and TABLE 1. Typical Engineering Deliverables and Status per Project Phase
their location on the layout (which Deliverable V CE BE DE
would ideally be set over a selected Project scope, design basis and criteria S P C C
terrain). The needs for raw materials, Soil studies, topography, site preparation S C C
intermediate and final product inven- Construction bid packages P C
tories, as well as utility requirements
Process block diagrams S/P P/C C C
are also established.
Plot plan S P/C C
A basic engineering, or front end
engineering design (FEED) phase. Process and utility flow diagrams (PFDs / UFDs) S/P P/C C
This sets the basis for the future P&IDs S P/C C
engineering, procurement and con- Material & energy, utility balances S P/C C
struction (EPC) phase, by gener- Equipment list S/P P/C C
ating a scope of work that further Single line diagrams S/P P/C C
develops the process engineering, Data sheets, specifications, requisitions S P/C C
and includes the early mechanical,
Mechanical equipment design drawings and documents S P/C C
electrical, instrumentation and civil/
Piping design drawings and documents S/P C
structural documents and drawings.
This phase also serves to generate a Electrical design drawings and documents S/P C
budget for the construction. Automation and control drawings and documents S/P C
An EPC phase. The EPC phase also Civil / structural / architectural design drawings and documents S/P C
includes the detailed engineering for Cost estimate C5 C4 C3 C2/C1
the development of the for con- Key: V = visualization; CE = conceptual engineering; BE = basic engineering; DE = detailed engineering; S =
struction engineering deliverables, started; P = preliminary; C = completed; C5, C4, ..., C1 = Class 5, Class 4, ..., Class 1 cost estimate (AACE)
the procurement of equipment and

Chemical Engineering www.chemengonline.com august 2015 51

 Studies Scope definition Engineering, procurement & construction

Visualization Conceptual Basic Detail Precomm.,

engineering engineering engineering Construction comm & startup Operations Decommissioning

Expert judgement HAZID HAZOP HAZOP Inspections Preventive HAZID

High level risk What-if LOPA Constructability Materials and and corrective
identification Consequence QRA review equipment tests maintenance checks
analysis FAT & SAT Periodically check
Hydrostatic tests instrument and relief-
valve calibration
Periodic hazards
Figure 2. Typical hazards analyses that are used analysis
throughout a CPI project lifecycle

mability, toxicity, reactivity and so with the operating conditions material embrittlement
on); these properties cannot be (pressure, temperature), and ma- higher material inventories

separated from the chemicals terial inventories. As general rules: increase the impact of poten-
Extrinsic dependent on de- higher pressures increase the tial releases, whereas lower
sign decisions. As an example: impact of potential releases, material inventories reduce
hazards associated with heating whereas vacuum pressures response times in abnormal
flammable materials with direct increase the probability of air operating conditions
burners can be avoided by using entering the system Equipment failures associated
indirect heating higher temperatures increase with damages to plant equipment
the energy of the system (and Utility failures associated with
2. Their source: hazards, especially when failures in utilities supplied to the
Process chemistry associated near the flashpoint or self-ig- facility, such as electricity, cooling
with the chemical nature of the nition temperature), whereas water, compressed air, steam, fuel
materials (for example, flammabil- very low temperatures could or others
ity, toxicity, reactivity and
Mu?llerGmbh_Chemical so on)
Engineering e pose the risks of freezing,
86x123_2011.qxd:Mull Human activity associated with
Process variables associated Chem eng 1-4pgKrytox Ad 11-9-2014.qxp_Layout 2 11/15/14 12:46 PM Page
Chemical Engineering e UC 86x123 formation
03/2011of hydrates, or activities by humans over the facil-

Ultra-Clean Fluorinated Oils, Greases,

The new cGMP-drum
offers process reliability by
validated cleaning procedures
Details of the Ultra-Clean line:
Sanitary welded edging
Geometry of beads and bottom
optimized for clean discharge
of product and for drum cleaning
Body, base and lid in stainless
steel AISI 316
FDA-approved silicone elastomer
seal ring, USP Class VI
Choose from a range
of 20 different sizes
PTFE and Dry Film Lubricants
Miller-Stephenson offers a complete line of inert high
performance fluorinated lubricants that include
DuPont Krytox oils and greases, as well as a fami-
ly of PTFE Dry Lubricants. They provide superior
lubricity while being thermally stable, nonflammable,
non-migrating, and contain no silicone. Krytox lubri-
cants offer extreme pressure, anti-corrosion and anti-
wear properties, as well oxygen compatibility and low
outgassing. Our PTFE creates a superior release for
plastics, elastomers and resins with multiple releases
between applications.
For technical information and sample, call 800-992-2424 or
203-743-4447.
m
TM

s
Compliant with FDA and cGMP
guidelines

Mller GmbH - 79 618 Rheinfelden (Germany)

Industrieweg 5 - Phone: +49 (0) 76 23 / 9 69 - 0 - Fax: +49 (0) 76 23 / 9 69 - 69 Connecticut - Illinois - California - Canada
A company of the Mller group supportCE@mschem.com
info@mueller-gmbh.com - www.mueller-gmbh.com miller-stephenson.com

Circle 20 on p. 74 or go to adlinks.chemengonline.com/56200-20 Circle 19 on p. 74 or go to adlinks.chemengonline.com/56200-19

52 Chemical Engineering www.chemengonline.com august 2015
 Studies Scope definition Engineering, procurement & construction

Visualization Conceptual Basic Detail Precomm.,

engineering engineering engineering Construction comm & startup Operations Decommissioning

Define plant Define process

Select codes Define Analyze plant Develop Conduct inspections Perform preventive Assess hazards
capacity scheme and standards process and hazards and construction Test piping and and corrective Follow work
Select technology Define for design controls operability drawings materials maintenance procedures
Define process equipment Define basis Define design Identify layers Verify plant Perform factory Periodically Document and
blocks and buildings of design conditions of protection hazards and acceptance and site check instrument signal abandoned
Decide plant location (layout) Define Assess risk risks acceptance tests (FAT and relief valve facilities (for example,
location Define raw electrical area Identify Finalize & SAT) calibration underground piping,
materials, classification additional safeguards Calibrate instruments Train new ducting, and so on).
products and Select safeguard design and relief valves operations
intermediate materials needs Define Perform hydrostatic personnel
product Design/ commissioning tests Follow work
inventories specify and startup Train operations procedures
equipment procedures personnel Periodically assess
Design Conduct hazards
buildings constructability Repeat previous
Define review activities for
control and revamps /
Figure 3. Typical design decisions af- emergency expansions
systems
fecting cost and risk throughout a CPI Design
project lifecycle preliminary
relief system
ity (for example, operator error, quency or probability of occurrence. Onshore Reliability Data (OREDA),
tampering with facilities, security This likelihood can be either relative American Institute of Chemical En-
threats and so on) (low, medium, high), or quantita- gineers (AIChE), Center for Chemical
Environmental associated with tive (for instance, 1 in 10,000 years). Process Safety (CCPS), American
environmental conditions (for ex- Quantitative values of the likelihood of Petroleum Institute (API), U.K. Health
ample, earthquakes, hurricanes, different categories of risk, or equip- and Safety Executive (HSE), Nether-
freezing, sandstorms and so on) ment failures, as well as risk tolerabil- lands Committee for the Prevention
The likelihood of a risk can be ex- ity criteria, can be obtained from liter- of Disasters by Dangerous Materials
pressed in terms of an expected fre- ature sources, such as Offshore and (CPR), or local government agencies,

PROTECT PUMPS
DRY RUNNING CAVITATION BEARING FAILURE OVERLOAD

MONITOR PUMP POWER COMPACT EASY MOUNTING

Best Sensitivity Only 3.25" x 6.25" x 2"
Digital Display Starter Door Panel
Raceway Wall
TWO ADJUSTABLE SET POINTS
Relay Outputs UNIQUE RANGE FINDER SENSOR
Adjustable Delay Timers Works on Wide-range of Motors
Simplies Installation
4-20 MILLIAMP ANALOG OUTPUT

WHY MONITOR POWER INSTEAD OF JUST AMPS? PUMP POWER

VALVE CLOSING

PUMPING
POWER

AMPS

VALVE OPENING NO FLUID

No Sensitivity
Power is Linear-Equal Sensitivity For Low Loads
at Both Low and High Loads
NO LOAD FULL LOAD NO LOAD FULL LOAD

CALL NOW FOR YOUR FREE 30-DAY TRIAL 888-600-3247

WWW.LOADCONTROLS.COM

Circle 18 on p. 74 or go to adlinks.chemengonline.com/56200-18

Chemical Engineering www.chemengonline.com august 2015 53

 TABLE 2. Examples of Changes in Design as Result of Process Hazards analyses in method to quantitatively assess the
different Project phases consequences of hazardous material
Impact on Project Conceptual Basic Detailed releases. Release rates are calculated
description engineering engineering engineering
for the worst case and alternative sce-
Process Industrial 1. Preliminary 2. Tower diameter, reboiler, 3. Line routings narios, end toxic points are defined,
definition solvents process design condenser, and pump changed after
manufactur- and equipment capacities changed, constructa- and release duration is determined.
ing facility characteristics spare equipment, alter- bility review, Hazard identification analysis
were defined nate lines and valves adding pres- (HAZID). HAZID is a preliminary
based on process added following What-If sure drop,
simulations and analysis. which altered
study that is performed in early proj-
best engineering pumps and ect stages when hazard material,
practices. control valves. process information, flow diagram
Plant layout / High-pressure 1. Preliminary plot 3. Relief systems design 5. Location of and plant location are known. Its
area gas plant plan was arranged required further modifi- some lines generally used later on to perform
based on available cations to plot plan, and and equip-
terrain and recom- an additional 10% of ment was
other hazard studies and to design
mended equipment space for flare exclusion slightly the preliminary piping and instru-
spacing. area. changed as mentation diagrams (P&IDs).
2. After conse- 4. After QRA, proper safe- result of con- What-if. This is a brainstorming
quence analysis, guards were selected structability
plant area was in order to reduce risk review, to method that uses questions start-
increased by 50% contours to tolerable allow early ing with What if..., such as What if
and equipment and levels in occupied build- operations in the pump stops running or What if
buildings were re- ings and public spaces, parallel with
located to prevent hence reducing space construction.
the operator opens or closes a cer-
impact areas from requirement by 25% tain valve? It has to be held by ex-
reaching occupied versus that required by perienced staff to be able to foresee
buildings and pub- consequence analysis. possible failures and identify design
lic spaces.
alternatives to avoid them.
Automation Crude-oil cen- 1. Only summary de- 2. Control system designed 5. Some addi- Hazard and operability study
and controls tral process- scription of major according to P&IDs. tional modifi-
ing facilities control system 3. Approximately 30% cations were (HAZOP). This technique has been
items developed in more instruments and required after a standard since the 1960s in the
conceptual engi- control loops added as reception of chemical, petroleum and gas Indus-
neering. result of HAZOP. vendor infor-
4. The overall system was mation.
tries. It is based on the assumption
increased from SIL-1 that there will be no hazard if the
to SIL-2 after LOPA, as plant is operated within the design
result of one section of parameters, and analyzes deviations
the plant handling light
ends. of the design variables that might
lead to undesirable consequences
and they can be especially valuable factor in ensuring that plant hazards for people, equipment, environment,
when performing quantitative, or can be avoided or reduced as early plant operations or company image.
semi-quantitative studies. as possible in the design. If a deviation is plausible, its conse-
The consequence of a risk can be Aside from the experience of the quences and probability of occur-
expressed in terms of its impact on team, it is generally accepted that rence are then studied by the HAZOP
several recipients, such as assets, different methodical approaches can team. Usually an external company
personnel, society and environment. be applied in a timely manner to the is hired to interact with the operator
The combination of likelihood and engineering design process, in order company and the engineering com-
consequence defines the risk. The to detect possible hazards that were pany to perform this study. There are
risk is then analyzed versus tolerabil- not addressed by the design team. at least two methods using matrices
ity criteria, either qualitatively (for ex- These structured reviews are called to evaluate the risk (R): one evalu-
ample, in a risk matrix), or quantita- process hazards analyses (PHAs), ates consequence level (C) times
tively (for example, in risk contours). and are often conducted or moder- frequency (F) of occurrence; and the
Company management and the de- ated by a specialist, with participa- other incorporates exposition (E) as a
sign team may then select measures tion of the design team, owners em- time value and probability (P) ranging
to eliminate or reduce individual risks, ployees or experienced operators. from practically impossible to almost
if they are not in the tolerable range. Several methodologies exist for sure to happen, in this method, the
conducting a PHA, each suitable for risk is found by Equation (1):
Process hazards identification specific purposes, processes, and
An experienced engineering design for certain phases of project devel- R = E P C (1)
team, with proper design basis doc- opment and plant lifecycle (Figure 2).
umentation, and working under ap- Below is a brief description of some of Layer-of-protection analysis
proved industry standards and best the most used PHAs in the CPI. (LOPA). This method analyzes the
engineering practices, is the first Consequence analysis. This is a probability of failure of independent
54 Chemical Engineering www.chemengonline.com august 2015
protection layers (IPLs) in the event
of a scenario previously studied in
a quantitative hazard evaluation like
HAZOP. It is used when a plant uses
instrumentation independent from
operation, safety instrumented sys-
tems (SIS) to assure a certain safety
integrity level (SIL). The study uses a
fault tree to study the probability of
failure on demand (PDF) and assigns
a required SIL to a specific instru-
mentation node. For example in pe-
troleum refineries, most companies
will maintain a SIL equal to or less
than 2 (average probability of failure
on demand 103 to <102), and a
nuclear plant will tolerate a SIL 4 (av-
erage probability of failure on demand
105 to <104).
Fault-tree analyses. Fault-tree anal-
ysis is a deductive technique that
uses Boolean logic symbols (that is,
AND or OR gates) to break down
the causes of a top event into basic
equipment failures or human errors.
The immediate causes of the top
event are called fault causes. The
resulting fault-tree model displays the
logical relationship between the basic
events and the selected top event.
Quantitative risk assessment
(QRA). QRA is the systematic de-
velopment of numerical estimates of
the expected frequency and conse-
quence of potential accidents based
on engineering evaluation and math-
ematical techniques. The numerical
estimates can vary from simple val-
ues of probability or frequency of an
event occurring based on relevant
historical data of the industry or
other available data, to very detailed
frequency modeling techniques [4].
The events studied are the release
of a hazardous or toxic material, ex-
plosions or boiling liquid expanded
vapor explosion (BLEVE). The results
of this study are usually shown on
top of the plot plan.
Failure mode and effects analysis
(FMEA). This method evaluates the
ways in which equipment fails and the
systems response to the failure. The
focus of the FMEA is on single equip-
ment failures and system failures.
When to use a given method
Some studies have more impact
in some phases than in others. For
Chemical Engineering www.chemengonline.com
example, if a consequence analysis
is not performed in a conceptual or
pre-FEED phase, important plot plan
considerations can be missed, such
as the need to own more land to
avoid effects over public spaces; or
the fact that the location might have
a different height with respect to sea
level than surrounding public places
impacted by a flare plume. PROVEN
Some other studies, like HAZOP,
cannot be developed without a con- PERFORMANCE
trol philosophy or P&IDs, and are
performed at the end of the FEED or
detailed engineering (for best results,
at the end of both) to define and vali-
date pressure safety valves (PSVs)
location and other process controls ROTOFORM
and instrument safety requirements.
QRA or LOPA (or both) are done GRANULATION
after HAZOP to validate siting and FOR PETROCHEMICALS
define safety instrumented systems AND OLEOCHEMICALS
SIL levels, and finally meet the level
required by the plant.
Figure 2 shows the typical CPI
project phases, with a general indi-
cation of when it is recommended
to conduct each study; however,
this may vary depending on the
specific industry, corporate prac-
tices, project scope and execution
strategy. AIChEs CCPS [2] has an High productivity solidication of
Applicable PHA technique table that products as dierent as resins, hot
indicates which study to perform in melts, waxes, fat chemicals and
each project phase, which also in- caprolactam has made Rotoform the
cludes research and development granulation system of choice for
(R&D), pilot plant operations, and chemical processors the world over.
other phases not covered in the Whatever your solidication
present article. requirements, choose Rotoform for
Table 2 includes some real-life ex- reliable, proven performance and a
premium quality end product.
amples of how the results of some of
these studies can impact the devel-
High productivity
opment of the plant design at differ- on-stream factor of 96%
ent project phases. Proven Rotoform technology
Out of the previously mentioned nearly 2000 systems installed
studies, a properly timed HAZOP, in 30+ years
at the end of the basic engineering Complete process lines or
phase, is key to identifying safety retrot of existing equipment
and operability issues that have been Global service / spare parts supply
overlooked by the engineering de-
sign team, especially when involving
an experienced facilitator and plant
operators in the study, given that
they have a fresh, outsiders view
Sandvik Process Systems
of the project, and they can provide Division of Sandvik Materials Technology Deutschland GmbH
input on daily operating experience. Salierstr. 35, 70736 Fellbach, Germany
Tel: +49 711 5105-0 Fax: +49 711 5105-152
Also, the deviations identified in the info.spsde@sandvik.com
HAZOP can serve to detect the need www.processsystems.sandvik.com
for additional safeguards that were
Circle 26 on p. 74 or go to adlinks.chemengonline.com/56200-26
august 2015 55
SANDVIK_Chemical_ad_55.6x254_MASTER.indd 1 09/02/2015 14:48
 TABLE 3. Additional Costs of Changes Associated with HAZOP Recommendations
during EPC Phase
Sample Project description
project
number
1 Gas dehydration unit
2 Gas compression unit
3 Crude oil atmospheric
unit
4 Fuel storage tank farm
5 Petrochemical plant relief
and flare systems
6 Crude oil dehydration
station
7 Crude oil evaluation
facilities
8 Heavy crude oil dehydra-
tion unit
9 Propane/air injection
plant
10 Oil pipeline + two gas
compression units
11 New flare system in ex-
isting refinery
12 Refinery gas concentra-
tion unit revamp
13 Extra-heavy oil deasfalt-
ing unit
14 Demineralized water
plant
15 Hydrogen compression
unit
not considered by the design team.
When the recommendations are
implemented correctly, and no other
changes to the process or plant are
done between the preparation of the
basic engineering design book and
the EPC phase, then a HAZOP sig-
nificantly reduces the probability of
significant cost impacts in the latter
as a result of changes due to addi-
tional PHAs.
Even though what-if, HAZID
and consequence analyses have
impact on the capital cost of the
project, the cost of implementing
their modifications to the design are
typically included on the EPC bid-
ding process, as they are realized
at the beginning of the project life-
cycle. Fault-tree analysis and LOPA
are used to define the redundancy
level of controls and instrumenta-
tion. The changes derived from
these studies generally represent a
minor portion of the total capital ex-
penditure. That leaves HAZOP and
QRA as the most important studies
56
The design options include, in de-
scending order of reliability: inherently
Estimated cost of changes PHAs and proper safer design, engineering controls
associated with HAZOP safe-design practices
recommendations in EPC phase implemented in previous (passive and active) and administra-
(as % of approved budget) design phases? tive controls (procedural).
3% Yes Inherently safer design involves
3% Yes avoiding or reducing the likelihood of
1% Yes
a hazard in a permanent or insepa-
rable fashion. For example, when de-
2% Yes signing a centrifugal pump discharge
2% Yes
system, an inherently safer design
would be to specify the design pres-
1% Yes sure at the centrifugal pump shut-off
pressure, thereby largely reducing
1% Yes the risk that an increase in the pump
discharge pressure (for example, due
3% Yes to a blocked outlet) could cause a
rupture in the pipes with consequent
1% Yes loss of containment.
Engineering controls are features
1% Yes incorporated into the design that re-
duce the impact of a hazard without
1% Yes requiring human intervention. These
can be classified as either passive
7% No (not requiring sensing and or active
response to a process variable) or
5% No
active (responding to variations in
process conditions). In the previ-
13% No
ous centrifugal pump example, a
35% No
passive solution would be to con-
tain possible leaks within dikes, and
with adequate drainage. Examples
to identify design improvements to of active solutions could be: a) pro-
prevent process hazards in the lat- viding a high-pressure switch asso-
ter project phases. ciated with an interlock that shuts
the pump down; and b) providing
Safe-design options a pressure safety valve (PSV) de-
At the early project phases, it is not signed for blocked outlet.
possible to identify all possible risk- Administrative controls require
reduction measures that could be human intervention. These are the
included in the design. However, a least reliable, because they depend
safety-oriented design team might on proper operator training and re-
be able to pinpoint sources of project sponse. In the previous example, an
risk due to lack of data, and oppor- administrative control would be to
tunities for risk reduction that could require operators to verify that the
be evaluated in later stages, as the valves in the pump discharge lines
design progresses and further details are open.
are known. Throughout the engineering
Some large organizations have col- phases leading to the EPC phase,
lected the pool of their experiences different safe-design choices can be
within risk checklists and proprietary made, as further information is made
design standards, thus paving the available. Figure 3 shows some of
way for future work. Where organiza- the typical design choices made by
tions have not established their own the engineering team throughout
standards and engineering practices, a chemical process plant lifecycle,
the design team should look for ac- which have direct impact on lifecycle
cepted codes and standards that are cost and risk.
the result of best engineering prac- In the visualization phase, safety
tices in a particular field or industry. can be included in the analysis as a
Chemical Engineering www.chemengonline.com august 2015
factor to decide key items, such as
production technology and plant lo-
cation. These key items are typically
selected based on other technical
criteria, such as overall efficiency,
production cost, or vicinity to either
raw materials, or markets (or export
facilities). For instance, when select-
ing a technology, health, safety and
environmental concerns could be
included as a criteria on the evalua-
tion matrix, by adding positive points
to technologies that reduce risks to
their environment by using less-toxic
materials, operating at lower pres-
sures or temperatures, or yielding
non-toxic byproducts. When se-
lecting a high-level plant location,
management could opt to locate
the plant away from large population
centers, in order to minimize risks to
communities. In this case, planning
authorities also have an important
role in defining allowable land-uses.
In the conceptual engineering phase,
safety can be included in the analysis,
for example, in the following ways:
1. Defining a simple, yet functional
process scheme, as relatively
simple processes have less equip-
ment and consequently lower fail-
ure probability (this can conflict
with other design goals); also, the
types of equipment selected can
have an important effect on pro-
cess safety (for example, selecting
indirect over direct heating).
2. Including safety concerns in the
early layout definition. For instance,
a design by blocks keeping the
main process, storage, and utility
areas separate from each other
can reduce overall risk. Other good
practices include: maintaining an
adequate separation between
pieces of equipment; separating
product inventories taking into ac-
count their flammability, toxicity or
reactivity, and considering dikes
around tanks containing danger-
ous materials; placing flares and
vents in locations separate from
human traffic, taking into account
wind direction (for example, so that
flames or plumes are directed far-
ther from personnel or population);
and allowing sufficient plot space
for an adequate exclusion area.
3. Keeping flammable and toxic ma-
Chemical Engineering www.chemengonline.com
terial inventories to the minimum
required to maintain adequate
surge/storage capacity and flex-
ibility in shipping.
In the basic engineering or FEED
phase, many design choices are
made over the specific mechani-
cal, piping, electrical, automation
and civil design that impact on the
overall facility risk. The first deci-
sion involves selecting the codes
and standards that will be used
for design, and defining the design
basis and criteria for each engineer-
ing discipline. Then, throughout the
design, some other decisions may
include: selecting between auto-
mated and manual operation, set-
ting equipment and piping design
conditions, defining the electrical
area classification, designing or
specifying equipment, structures
and buildings, defining control and
emergency systems (including ap-
propriate redundancy, where appli-
cable), and designing appropriate
relief systems, among others. Then,
there are equipment and system-
specific hazards and available safe-
guards that need to be considered.
Ref. 2 contains a comprehensive list
of hazards and safeguards for vari-
ous types of unit operations.
When hazards have been prop-
erly identified and addressed in the
earlier design phases, this reduces
the probability of significant costly
changes being made during the
EPC phase as a result of unsafe
process conditions.
Addressing hazards early
When hazards are identified, and
proper design choices are taken
early in the engineering design to ad-
dress them, significant benefits can
be obtained.
Table 3 compares the additional
cost of changes arising from recom-
mendations made during a HAZOP
at the EPC phase. The costs are
expressed as a percentage of the
budget that was approved dur-
ing the bidding stage, of projects
of different scope and plant type,
executed by different companies
in different countries, including the
U.S. and Latin America, with ap-
proved budgets between $5 million
august 2015
and $200 million.
The projects are divided into two
categories: a) projects where the
design contractor applied best en-
gineering standards and employed
PHAs at optimum points during the
conceptual engineering and FEED
phases; and b) projects where ad-
equate PHAs and safe-design prac-
tices were not applied in the previous
design phases.
As can be seen in Table 3, there
is a significant difference between
the cost of the changes arising from
HAZOP recommendations when
proper safe-design practices and
PHAs were applied during the FEED
phase, and when they were not.
For the first category, changes
were typically in the range of 1 to
3%. In the upper end of this cate-
gory, changes were higher when the
owner requested some minor modi-
fications to the FEED design without
properly assessing the risks associ-
ated with said changes.
As an example, the heavy crude
oil dehydration unit (Project 8) was
designed according to best engi-
neering practices, and adequate
analyses (HAZOP, LOPA) were
conducted during the engineering
phase. However, the owner de-
cided to implement changes in the
design in order to compress the
schedule, by removing several long-
lead items that included emergency
shutdown system (ESD) valves and
components, without updating the
PHAs. With the unit in operation,
the owner asked the contractor to
include the ESD items that were in
the original design.
For the second category, changes
exceeded 5%, and in one case
reached as high as 35% of the ap-
proved budget. Below is a descrip-
tion of what went wrong in each of
these projects:
The refinery gas concentration unit
revamp (Project 12) FEED consid-
ered hand operations in key pieces
of equipment. As a result of a HAZOP
during the EPC, the operations had
to be automated, which changed
the equipment specifications and
design. The number of loops added
after the HAZOP exceeded the ca-
pacity of the controller, and another
57

Now Available:
Chemical Engineering
Features Report
Guidebook-2014
Chemical Engineerings Feature
Reports provide concise, factual
information that aids in solving
real problems. The practical,
how-to orientation of these
articles ensures that they can
be directly applied to chemical
engineers daily jobs. Each article
applies to a relatively broad
section of the chemical process
industries as opposed to focusing
on just one niche.
ts
r e R ep or n
FeatuCompilati
o identify hazards, and address them
2014
25139
Find this and other related
reference material at
store.chemengonline.com
58
one had to be installed.
The extra-heavy oil deasphalting unit
(Project 13) was designed during the
basic engineering phase as a mostly
hand-operated facility, with minimum
supervisory controls. As a result of a
HAZOP during the EPC, the risk was
not tolerable to the owner, and the
whole unit had to be automated.
The demineralized water plant
(Project 14) was delivered by the ven-
dor as a package unit, and no PHAs
were conducted by the vendor. When
received, the plant had many safety
and operability issues and a number
of important modifications had to
be made, including: additional lines,
block and control valves, relief valves
and associated lines, among others.
Aside from the costs associated with
the changes, the project was delayed
by six months.
The hydrogen compression unit
(Project 15) basic engineering design
did not address all of the safety con-
siderations associated with hydro-
gen handling. Some of the modifica-
tions recommended by the HAZOP/
LOPA studies during the EPC phase
included changing the compressor
specification, and increasing the SIL
of the SIS from SIL-1 to SIL-3.
Final remarks
Hazards are present in the CPI;
some are avoidable, while others
cannot be separated from the plant,
as they are tied to the very nature
of the chemicals or the unit opera-
tions, or both. However, a proper
design team, one that is trained to
using the best engineering practices
in safe-design from early on in the
project lifecycle, along with prop-
erly timed and executed PHAs, can
be very valuable in avoiding costly
changes during the EPC phase, or
even worse: potential damages to
persons and the environment. n
Edited by Gerald Ondrey
References
1. Center for Chemical Process Safety (CCPS), Guidelines
for Investigating Chemical Process Incidents, 2nd edi-
tion, CCPS, AIChE, New York, N.Y., 2003.
2. CCPS, Guidelines for Engineering Design for Process
Safety, 2nd ed., CCPS, AIChE, New York, N.Y., 2012.
3. AACE International Recommended Practice No. 18R-
97, Cost Estimate Classification System As Applied
in Engineering, Procurement, and Construction for the
Process Industries.
Chemical Engineering www.chemengonline.com
4. American Petroleum Institute (API) Recommended
Practice (RP) 752, Management of Hazards Associ-
ated with Location of Process Plant Permanent Build-
ings, 3rd ed., 2009.
5. U.S. Environmental Protection Agency (EPA), Risk Man-
agement Program Guidance For Offsite Consequence
Analysis, March, 2009.
6. EPA, Chemical Emergency Prevention & Planning News-
letter, Process Hazard Analysis, July August, 2008.
7. Occupational Safety and Health Administration (OSHA)
29 CFR 1910.119. Process Safety Management of
Highly Hazardous Chemicals.
Authors
Sebastiano Giardinella is
the vice president and co-owner
of the Ecotek group of companies
(The City of Knowledge, Bldg.
239, 3rd floor, offices A and B,
Clayton, Panama City, Republic of
Panama; Phone: +507-203-
8490; Email: sgiardinella@
ecotekgrp.com). He has experi-
ence in corporate management,
project management, project engineering and process
engineering consulting in engineering projects for the
chemical petrochemical, petroleum-refining, oil-and-
gas and electrical power-generation industries. He is a
certified project management professional (PMP), has
a M.Sc. in renewable energy development from Heriot-
Watt University (Scotland, 2014), a masters degree in
project management from Universidad Latina de Pan-
am (Panama, 2009), and a degree in chemical engi-
neering from Universidad Simn Bolvar (Venezuela,
2006). He is also professor of project management at
Universidad Latina de Panam, and has written a num-
ber of technical publications.
Mayra Marchetti is a senior
process engineer, currently working
as independent consultant (Coral
Springs, Fla.; Email: mmarchetti@
ecotekgrp.com), with more than ten
years of experience in the oil-and-
gas, petrochemical, petroleum-re-
fining and pharmaceutical indus-
tries, and has participated in the
development of conceptual, basic
and detail engineering projects. She specializes in pro-
cess simulation, plant debottlenecking and optimization,
and relief systems design. She has a masters degree in
engineering management from Florida International Uni-
versity (Florida, 2008), and a degree in chemical engi-
neering from Universidad de Buenos Aires (Argentina,
1996). She has published articles and delivered world-
wide seminars focused in the use of simulation tools for
the process industry.
Alberto Baumeister is the
CEO and co-owner of the Ecotek
group of companies (same address
as above; Email: abaumeister@
ecotekgrp.com). He has experience
in corporate management, project
management, and senior process
consulting in engineering projects
for the chemical, petrochemical,
petroleum-refining, oil-and-gas,
electrical power-generation and agro-industrial indus-
tries. He has a specialization in environmental engineer-
ing (gas effluents treatment) from the Universidad Miguel
de Cervantes (Spain, 2013), a masters diploma in water
treatment management from Universidad de Len (Spain,
2011), a specialization in management for engineers at
Instituto de Estudios Superiores de Administracin (Ven-
ezuela, 1990), and a degree in chemical engineering
from Universidad Metropolitana (Venezuela, 1987). He
was professor of the Chemical Engineering School at
Universidad Metropolitana between 1995 and 2007, and
has written a number of technical publications.
august 2015
 Environmental Manager

Chemical Lifecycle Management

Sustainability demands Waste diversion/disposal options

a higher level
Manufacturer / waste generator

of oversight that seeks Production Laboratory

new ways to effectively Carbon,

chlorine and SAA SAA SAA
other offsets
and cost efficiently MSA

procure, handle and Reprocess

dispose of, Waste

Toll Recycle Reuse Disposal

or recycle materials management

disposal Secondary market Fuel
blending
Landfill Incineration
services
(recycle) Treatment
Victor Belenchia
Clean Harbors Input to manufacturing process

C
hemicals in production and
laboratory environments
alike have three distinct
lifecycle phases: the first is
procurement and inventory storage;
the second is use in a manufacturing
process or research program; and the
third is post-use, including onsite han-
dling, removal and beneficial reuse,
recycle or disposal.
A number of considerations includ-
ing regulations, safety, material uti-
lization efficiency, costs and sustain-
ability govern all of the steps in this
lifecycle. The challenge for laboratory
and production managers is to balance
these sometimes competing forces to
create an effective process that meets
the organizations overall needs.
Proper chemical handling during
processing is typically emphasized in
production facilities, but how to han-
dle chemicals post-use is often given
less attention. While this article ex-
amines the full chemical lifecycle, it
focuses on this post-use phase.
Sustainable lifecycles
Every laboratory and chemical pro-
cessor must follow environmental,
health and safety (EH&S) proce-
dures and regulations. This is often
a challenge, especially for smaller
operations, since implementation fre-
quently rests on individual managers,
44 Chemical Engineering www.che.com December 2010
Figure 1. Managing post-use chemicals is a vital part of chemical lifecycle man-
agement. Recycling waste chemicals is an option that can contribute greatly to
sustainability goals
local teams and staff who have other
primary responsibilities. Regardless
of size, laboratory and production
managers must periodically review
their policies, conduct training and
audit their chemical inventories and
waste materials. Requirements, pro-
cesses and regulations change, and
it is managements responsibility to
remain in compliance, while meeting
everyday production demands.
Environmental sustainability con-
tinues to increase in importance and
is embraced as a corporate goal at
many manufacturing organizations.
This takes chemical management be-
yond the traditional regulatory and
safety considerations. The chemical
process industries (CPI) have become
even greater stewards of the materials
throughout their entire lifecycles as
they seek ways to further reuse or re-
cycle materials that were historically
disposed of as waste (Figure 1).
Sustainability requires the mini-
mization of consumption and waste
in order to reduce environmental im-
pact through process optimization and
reuse or recycling whenever possible.
This can require additional record-
keeping and administrative steps. The
challenge, when it comes to chemical
lifecycle management, is to find op-
erationally efficient and cost-effec-
tive ways to reduce the overall waste
stream through process optimization,
efficient material use, tolling or sell-
ing recyclable materials to a second-
ary market.
Purchasing and storage
Effective chemical management
starts before the chemicals arrive at
the site. Laboratory and operations
personnel need to work carefully with
purchasing staff to order the correct
amounts, minimize order errors and
ensure that proper receiving and
storage provisions are in place. For
operations personnel, it is a balance
between anticipating customer orders
and manufacturing volume in order
to eliminate production interruptions,
minimize storage and other inventory
costs, and to avoid having to dispose
of outdated materials. Purchasing
staff can consolidate orders to lower
costs, but they must have an under-
standing of the chemical properties
and intended uses in order to ensure
that materials are consumed before
they begin to degrade and negatively
affect chemical reactions or final prod-
uct properties. Overall, the CPI have
done a good job utilizing continuous
improvement techniques to manage
their supply chains.
Once received, the materials must
be stored in accordance with local,
state and provincial regulations, or
the International Code Councils
(ICC) International Building and In-
ternational Fire Codes, and handled
in accordance with company EH&S
procedures. In the U.S., handling
must follow the Occupational Health
& Safety Admin.s (OSHA) Code of
Federal Regulations (Title 29 CFR
1910.1200), which covers Hazard
Communication Standards (HCS),
the federal Risk Management Pro-
gram for certain highly toxic chemi-
cal compounds regulated under the
Clean Air Act, and the Chemical Hy-
giene Standard for laboratories (Title
29 CFR 1910.1450).
Production and laboratory use
As materials move into the second
phase of the lifecycle use in a pro-
duction environment or laboratory
the potential for handling errors in-
creases. Although industrial users and
researchers are concerned with, and
trained in, safe handling procedures
for hazardous materials, they are less
likely to focus on storage and removal.
This leads to some potential problems
in chemical lifecycle management.
Chemicals are generally well man-
aged in production environments
since production procedures are well-
defined and engineered. Failure most
often results from human error or
system breakdowns. Although the
results can be catastrophic, they are
largely preventable through adher-
ence to established EH&S procedures,
effective maintenance practices and
frequent training.
Research environments are less
systematized, which presents greater
opportunity for errors. Researchers
often work independently on projects
and they use smaller quantities of a
greater number of chemicals. This
presents an exponentially higher
number of possible chemical reactions
and storage issues.
Working alone or on small teams,
researchers often neglect to properly
label chemicals in secondary containers
at their stations. Unlabeled or improp-
erly labeled chemicals violate hazard-
ous communication (Hazcom) regula-
tions. Informal practices also lead to
the potential for abandoned chemicals
as researchers move on to other jobs or
projects. Unidentified chemicals may
require testing, special handling and
expensive disposal if they cannot be
verified by the researcher.
Consistency and training, sup-
ported by accurate, up-to-date la-
beling and recordkeeping, are key
ingredients to a successful chemical
management program.
Post-use
Handling of solvents, catalysts and
other chemicals can become much
more problematic once they have
been used in a production or labora-
tory process. They are no longer in
their original containers and are often
combined with other substances. In
the U.S. for example, OSHA, the En-
vironmental Protection Agency (EPA),
Dept. of Transportation (DOT) and, in
some cases, Drug Enforcement Admin.
(DEA) regulations also come into play.
A few important fundamental facts
must be kept in mind at all times: In
the U.S., all chemical storage areas
fall under Resource Conservation and
Recovery Act (RCRA) regulations and
are subject to inspection by regulators.
To comply with that, a limited number
of conveniently positioned and prop-
erly organized satellite accumulation
areas (SAAs), under the control of the
EH&S or chemical hygiene manager,
should be set up near production or
research areas.
There are a number of RCRA rules
for SAAs. For instance, containers
must be in good condition without
rust, dents, or cracks; they cannot be
stored near drains or other structures
that could pose an environmental
risk; hazardous waste containers can-
not exceed 55 gal; acutely hazardous
waste cannot exceed one quart; SAAs
must be inspected weekly. There are
additional guidelines including re-
quirements for clear and visible labels
stating hazardous waste and listing
the container contents, as well as the
three-day rule that requires that the
containers are immediately marked
Chemical Engineering www.che.com December 2010
with the current date when full, and
removed within three days to the main
storage area (MSA).
The MSA is governed by additional
rules. One of the most important re-
lates to allowed accumulation time
once materials enter the MSA. Small
quantity generators (SQGs) organi-
zations that generate more than 100
kg but less than 1,000 kg of hazard-
ous waste per month can accumu-
late materials for 180 days from the
start date.1
Large quantity generators (LQGs)
those that generate 1,000 kg or
more of hazardous waste per month,
or more than 1 kg per month of acutely
hazardous waste are subject to a
90-day accumulation rule.2
Regardless of size, the MSAs must
be secured against unauthorized entry;
hazardous waste and container con-
tents labeling rules must be followed;
containers must be inspected weekly
with reports kept on file; and pre-
paredness and prevention equipment
(for example an emergency phone,
alarm, fire suppression and spill pre-
vention equipment) is required.
There are additional regulations
and exceptions to the rules noted
above. For instance, peroxide form-
ers, such as ethers and dioxane, are
generally managed under a peroxide-
former program that tracks the ma-
terial based on the manufacture and
retention dates from the day it arrives
onsite until the day it is consumed or
disposed of. SAA storage times and
90/180-day MSA rules are all subordi-
nate to the peroxide-former deadlines.
So, if the earliest peroxide-former re-
tention date is 45 days in the future,
that is the containers deadline. Even
the waste disposal company must
track peroxide formers and follow ex-
pedited destruction schedules based
on the retention date.
Managing chemical storage areas.
A minimum number of properly
trained staff should be assigned to
transport materials from the SAAs to
the MSA and manage the materials in
all of the storage areas. Some opera-
tions are large enough to have a dedi-
1. http://www.epa.gov/waste/hazard/generation/
sqg/index.htm
2. http://www.epa.gov/solidwaste/hazard/
generation/lqg.htm
45

Environmental Manager
cated full-time staff, while for others it
is a part-time function (that requires
the same annual refresher training).
Properly organizing the handling
and in-house movement of chemicals
removes the responsibility from re-
searchers and substantially reduces
training requirements.
Another benefit of using dedicated
staff is that it puts a third party in po-
sition to regularly inspect warehouse,
production and laboratory areas to
ensure that chemicals are properly
stored and handled in accordance with
regulatory and company policies. Here
are some best practices that we have
seen in the field and that we practice
at our own sites and for customers:
Ensure that proper procedures
(such as labeling and tracking) are
being followed at each production
and laboratory location, and that
waste chemicals are stored only at
approved sites
Conduct periodic safety training for
research and production personnel
Establish collection schedules to ef-
ficiently move materials from SAAs
to the MSA in a timely way
Ensure that SAA containers are in
good condition, properly closed and
placed within secondary containers,
and that incompatible materials are
properly segregated
Catalog, store and consolidate MSA
materials in bulk containers and
pack smaller quantities, elbow bot-
tles, and so on, for transportation
Identify unknowns before shipping
Follow up on the disposal or recov-
ery of chemicals for company and
regulatory reporting
Software. Most large companies
utilize chemical inventory manage-
ment systems, while many laborato-
ries still use spreadsheets to manage
their inventories. Spreadsheets can be
helpful, but they are not the answer
for corporate-wide or comprehensive
lifecycle-management programs. They
are generally home-grown standalone
applications without built-in report-
ing mechanisms. They do not extend
beyond the department or integrate
with corporate systems. They do not
tie into industry resources and docu-
mentation, such as material safety
data sheet (MSDS) databases, that are
necessary to build chemical profiles.
46 Chemical Engineering www.che.com December 2010
TRI Chemicals
A
n added benefit of recycling and reusing chemicals can be reducing the volume
of materials listed in Toxics Release Inventory (TRI) reports. This can contribute to
reduced taxes and annual fees associated with RCRA regulations. TRI chemicals
have included the following:
Aluminum (fume or dust)
Aluminum oxide (fibrous forms)
Ammonia (anhydrous and aqueous)
Asbestos (friable)
Hydrochloric acid (acid aerosols)
Nitrate compounds (water dissociable;
aqueous solution)
Phosphorus (yellow or white)
Sulfuric acid (acid aerosols)
Vanadium (except when contained in an
alloy)
Zinc (fume or dust)
SOURCE: U.S. Environmental Protection Agency: EPA 260-R-09-006 October 2009 Toxic Chemical Release
Inventory Reporting Forms and Instructions Revised 2009 Version (http://www.epa.gov/TRI/report/rfi/
ry2009rfi121709.pdf). See Table II-3, p. 143 for the full alphabetical list of chemicals and qualifiers
Perhaps, most importantly, they may
not stand up to regulatory scrutiny if
the need arises.
Specialized software and online
chemical-waste-management systems
handle all of these issues and also
address many other important cri-
teria, such as chemical movements,
90/180-day storage-time limits, chemi-
cal expiration dates and reporting.
More sophisticated software and on-
line tracking systems assign cost cen-
ter allocations, and facilitate move-
ment, inventorying and packing, as
well as documenting the transfer of
the material to a certified waste dis-
posal company.
Waste disposal companies
Given the complexity of effective
waste-chemical management, many
companies find that it is more ef-
ficient and cost effective to turn to
waste disposal companies to handle
their chemical waste throughout the
entire post-use lifecycle phase. Service
companies manage the entire process
including onsite movements, storage,
packing and removal. In that way,
generators do not have to invest as
heavily in RCRA training and can rely
on the contractor to apply its exper-
tise, provide staff and backup staff as
necessary, monitor regulatory federal,
state and local developments, and im-
plement best practices.
This allows plant and laboratory
management to properly oversee the
function without adding headcount. It
also enables staff to focus on their core
responsibilities and not be burdened
with additional training or oversight
of chemical management functions.
The TRI listing for the following three
chemicals is based on the chemical activity
rather than the form of the chemical:
1. Dioxin and dioxin-like compounds
(only if they are manufactured at the fa-
cility; or are processed or otherwise used
when present as contaminants in a chemi-
cal but only if they were created during the
manufacture of that chemical)
2. Isopropyl alcohol (only if it is being
manufactured by the strong acid process)
3. Saccharin (only if it is being manu-
factured)
Using the same outsourcer to handle
materials in-house as well as for re-
moval and disposal eliminates inter-
organizational hand-offs and can fur-
ther reduce costs.
Communications between waste
generators and chemical disposal
companies. Before a waste disposal
company can remove hazardous
chemicals from a generator, all paper-
work must be in order, the materials
must be properly labeled and pack-
aged and ready to go. It is important
that chemical waste is properly char-
acterized in order to avoid unneces-
sary expense or, in a worst case, addi-
tional transshipping to another waste
disposal site. Confusion creates lost
time and increased costs for chemical
characterization tests and EPA and
DOT documentation.
Communications with the waste
disposal company are key. Direct links
between the generator and the waste
disposal company are efficient since
any materials or paperwork that are
not in compliance will be identified
early, waste profiles and DOT mani-
fests can be generated, and placards
assigned before the disposal company
arrives on site.
A direct link to the waste disposal
company can also report contain-
ers that are ready for packing and/or
pickup, enabling the removal company
to cost-effectively schedule pickups.
Automated procedures also help com-
panies stay in compliance with RCRA
90/180-day rules or special handling
requirements for peroxide formers
and other acute hazards.
Regardless of the degree of auto-
mation and communications, under-
 Revise #1 - 2/4/10
solvent Recycle and reuse
6:00 PM
S
ome waste disposal companies offer more sophisticated services, such as the ability
to recycle solvents. Examples of solvents that are typically recycled for reuse include
the following:
Centrifuge & Drying
1-Ethyl-2-pyrrolidone Methyl ethyl ketone
Technologies
Acetone Methylene chloride
Decalin (Decahydronaphthalene) N,N-Dimethyl acetamide Inverting Filter Centrifuge
Dimethyl sulfoxide Naptha
Ethanol solutions N-Methyl-2-pyrrolidone
(excluding specially denatured N-Propyl bromide
alcohols) Perchloroethylene
Ethyl acetate Petroleum distillates
Ethylene glycol Propylene glycol
Isoamyl alcohol Tetrachloroethylene
Isopropyl alcohol Tetrahydrofuran
Isopropyl alcohol (high water) Toluene
Methanol Xylenes (ortho-, meta-, para-)

Cutting edge centrifuge technology for

standing the chemical waste disposal
process minimizes costs up front by
smoothing the transition from the
MSA to disposal and reuse. Com-
munication with the waste disposal
company has become even more criti-
cal over the last few years due to in-
creased opportunities for chemical
reuse, recycling and tolling.
Reuse and recycle. Some hazardous
and non-hazardous waste and byprod-
ucts with residual value can be di-
rectly reused in other processes by an-
other manufacturer. In the past, these
materials may have been landfilled
or incinerated because it was less ex-
pensive and there were few disposal
alternatives. Today, both environmen-
tal and financial incentives encourage
materials reuse.
Beneficial reuse and recycling pro-
vide additional revenue streams, or
at least the potential for cost reduc-
tions, for the generator; however, they
add complexity to the removal-and-
disposal process. Rather than a one-
way transaction where the disposal
company removes and disposes of the
material, reuse and recycling may now
include waste removal discounts for
materials sold on the secondary mar-
ket. Tolling, where reprocessed ma-
terials are returned to the generator,
also results in more complex price ne-
gotiations based on fluctuating rates
for virgin materials.
Recycling, where the waste dis-
posal company reprocesses chemicals
for new uses, can also substantially
advance generators sustainability
performance. The materials stay out
of the waste stream and become feed-
stock for another companys process,
as outlined in Figure 1.
There are several other benefits
that accrue from the reuse and recy-
cling of chemical materials. In some
cases reuse and recycling operations
reclassify wastes, removing them from
the generators hazardous waste man-
ifest, which reduces the volume of ma-
terials listed in Toxics Release Inven-
tory (TRI) reports and advances waste
minimization goals. They contribute
to reduced taxes and annual fees as-
sociated with RCRA regulations (see
box titled TRI Chemicals) and other
corporate responsibility and environ-
mental reports.
Clearly, sustainable disposal, reuse
and recycling require a more sophis-
ticated service level from the waste
disposal company. Only a few are able
to analyze the waste with an eye to-
ward potential uses and the secondary
market. They also have knowledge of
the underlying regulations and access
to the separation and purification re-
sources necessary to reprocess the ma-
terials (see box titled Solvent Recycle
and Reuse). Companies that offer these
services along with in-house process-
ing, tolling and resale capabilities, are
best suited to develop comprehensive
waste management programs that
fully leverage sustainable secondary
uses for waste chemicals.
It makes financial and environmen-
tal sense to explore these opportuni-
ties with a waste processor.
Edited by Dorothy Lozowski
Author
Victor Belenchia is vice
president, chemical indus-
try for Clean Harbors (42
Longwater Dr., Norwell, MA
020619149; Email: belenchia.
victor@cleanharbors.com;
Phone: 7817925471; Web-
site: www.cleanharbors.com).
He holds a B.S. in biology and
chemistry from the University
of Central Connecticut and an
MBA in marketing from the
University of Connecticut. He spent more than
20 years in the chemical industry with Ameri-
can Cyanamid and Cytec Industries in various
technical and commercial positions and has been
with Clean Harbors for the last six years in his
current role.
filtration, washing and drying of solid/liquid
suspensions
t Widest range of applications - hardest to
easiest filtering products can be handled
t No residual heel for exact repeatable
batches and no loss of product
t PAC technology allows drying of the
product inside of the centrifuge
t Thin cake filtration operation allows for
improved quality and production rates
t Full containment eliminates operator
exposure
t Effective automated CIP
Kilo-Lab Conical
Vacuum Dryer-Mixer
Advanced technology
for real Kilo size
drying research
and development
t Utilizes interchangeable agitator systems
either orbiting arm screw or central shaft
t Flexible small scale volume of 150ml to
1500ml
t Plastic view through vessel available
t Designed for real laboratory requirements
of size, with full instrument & data
recording
t Direct scale up to production or pilot size
units
Horizontal & Vertical
Centrifuges
t Size ranges from 200mm to 1800mm
t Wide range of standard & custom designs
t Laboratory size equipment
Lab Testing Available
Rental & Lease Machines Available
www.heinkelusa.com
Tel: 856-467-3399
Circle 11 on p. 58 or go to adlinks.che.com/29257-11

Chemical Engineering www.che.com December 2010 47

 Cover Story
Figure 1. When
engineering controls
are not enough to
reduce or eliminate
airborne hazards,
respirators are
needed for protection
Dennis Capizzi
MSA Safety
A
irborne respiratory hazards are a
very real threat in the chemical
process industries (CPI). Exist-
ing in a variety of forms includ-
ing gases, vapors, dusts, mists, fumes,
smoke, sprays and fog, such hazards can
cause illnesses including cancer and
lung impairment, or even death. Ex-
amples of respiratory hazards include
combustion byproducts; toxic fumes or
dust created by metal melting; ozone
and nitrogen oxides from processes
involving an electric discharge in air;
and dust particles released by grinding
(notably dry grinding such as blast-
ing) and crushing applications, and the
transport, sieving, mixing or screening
of any dry material.
The specific hazardous gases pres-
ent in a workplace will, of course,
vary according to the processes of the
facility, but commonly include chlo-
rine, chlorine dioxide, ammonia, ni-
trogen dioxide, nitric oxide and vola-
tile organic compounds (VOCs). Some
hazardous gases, such as carbon
monoxide, act rapidly and can cause
unconsciousness or death within min-
utes, while other toxic gases can take
28 Chemical Engineering www.che.com December 2012
Clearing the
Air About
Respiratory
Protection
Learn the basics about selection
and regulatory compliance for these
potentially life-saving devices
years to produce noticeable harm.
Where toxic substances are pres-
ent in the workplace and engineering
controls (such as enclosing or confin-
ing the contaminant-producing opera-
tion, exhausting the contaminant, or
substituting with less toxic materials)
are inadequate to reduce or eliminate
them, it is time to turn to respirators
(Figure 1).
What is a respirator?
A respirator is a protective device that
covers the nose and mouth or the en-
tire face or head to guard the wearer
against hazardous atmospheres. Res-
pirators may be either tight-fitting or
loose-fitting. There are two types of
tight-fitting respirators: half masks
cover the wearers mouth and nose,
while full facepieces cover from the
hairline to below the chin (Figure
2). Loose-fitting respirators include
hoods and helmets that cover the
head completely.
The OSHA standard
The appropriate use of respiratory pro-
tection is not only important for worker
health and safety in the U.S., it is
the law. The U.S. Occupational Health
and Safety Administration (OSHA) is
the main federal agency charged with
the enforcement of safety and health
legislation in the U.S. Much of this leg-
islation is written in the form of stan-
dards. OSHA standards are rules de-
signed to provide a work environment
free from known dangers or hazards.
They require the use of certain safe
practices and equipment, and assign
employers the responsibility of moni-
toring hazards and keeping records of
workplace injuries and illnesses. Com-
pliance with OSHAs standards is man-
datory and is stringently enforced.
In 2011, Respiratory Protection,
General Industry (29 CFR 1910.134)
was OSHAs fourth most frequently
cited standard violation. Standard 29
CFR 1910.134 covers the respiratory
protection division of the personal
protective-equipment category. Essen-
tially, it details the requirements for
compliance in workplaces where respi-
ratory hazards are a potential threat
(Figure 3). The aforementioned statis-
tic reveals that too many workers are
not employing appropriate protection
from respiratory hazards on the job.
According to OSHA, an estimated 5
million workers are required to wear
respirators in 1.3 million workplaces
throughout the U.S.

FIgure 2. This is an example of a
full-facepiece respirator that covers the
wearers face from the hairline to below
the chin
The role of ANSI
In addition to OSHA, another driv-
ing force in the regulatory arena is
the American National Standards In-
stitute (ANSI). As a voice of the U.S.
standards and conformity assessment
system, ANSIs mission is to enhance
both the global competitiveness of
U.S. business and quality of life by
promoting and facilitating voluntary
consensus standards and conformity
assessment systems, and safeguard-
ing their integrity, in an effort to pro-
tect workers.
ANSI thus provides a neutral
forum for the development of policies
on standards issues and serves as a
watchdog for standards development
and conformity assessment programs
and processes [2]. ANSI/AIHA Z88.2
Standard Practices for Respiratory
Protection is a voluntary consensus
standard drafted by the ANSI Z88
Committee on Respiratory Protection
[3]. This standard is significant, as it
addresses the issue of assigned protec-
tion factors (APFs). APFs are the main
factor used by many employers when
they select respirators for worker pro-
tection from airborne hazards. The
ANSI standard recommends protec-
tive factors that are very similar to
those of OSHA, except for a more con-
servative APF for filtering facepieces
(disposable respirators). The standard
also addresses other key issues re-
garding the proper use of respiratory
protection in the workplace, including
the following:
Respiratory protection program re-
quirements
Standard operating procedures for
programs
Respirator selection, limitation and
use
Training
Fit testing
Maintenance, inspection, storage
and disposal
Breathing air and oxygen-deficient
atmospheres
Recordkeeping
Establishing a program
The fundamental goal of any respira-
tory protection program is to control
occupational diseases and injuries
caused by breathing air contaminated
with harmful dusts, fogs, fumes, mists,
gases, smoke, sprays and vapors. The
defense against these contaminants is
simple: keep them out of the workers
breathing air.
Engineering and administrative
controls, which have the potential to
completely eliminate the hazard, must
always be implemented first. These
include actions such as enclosing or
confining the contaminant-producing
operation, exhausting the contami-
nant, or substituting the contami-
nant with a less-toxic material. Work
practices such as limiting employee
exposure should also be considered.
If these controls do not eliminate the
hazards, then employers must provide
appropriate respiratory protection for
every employee who might be exposed
to them.
Whenever OSHA standards or em-
ployers require respirator use, there
must be a complete respiratory protec-
tion program in place. Employers must
have written operating procedures to
ensure that employees use the respi-
rators safely and properly. Users must
be familiar with these procedures as
well as the types of respirators avail-
able and their limitations.
A qualified program administra-
tor. First of all, a qualified program
administrator must be responsible for
the program. This person must know
enough about respirators to supervise
the program properly. Larger plants
or companies with industrial-hygiene,
Chemical Engineering www.che.com December 2012
in-house medical, safety-engineering
or fire-prevention departments should
administer the program in liaison with
the program administrator. In smaller
plants that do not have specialists, an
upper-level superintendent, foreman
or qualified person must serve as the
program administrator.
Employee training. Any respira-
tor program should stress thorough
training of all respirator users. Em-
ployees must be aware that a respi-
rator does not eliminate the hazard.
If the respirator fails, the user will be
overexposed to dangerous substances.
To reduce the possibility of failure, the
respirator must fit properly and be
maintained in a clean and serviceable
condition. Employers and employees
must understand the respirators pur-
pose and limitations. Users must not
alter or remove the respirator even
for a short time, regardless of its po-
tential discomfort.
Regular evaluation. Employers
must evaluate the effectiveness of
their companys respirator program
regularly and modify the written op-
erating procedure as necessary to
reflect the evaluation results. Work-
place conditions that affect respira-
tory hazards and respirator use may
change over time. This would include
new work processes or techniques,
the use of new or different materials
or chemicals, changes in the amount
of a respiratory hazard present in the
workplace and changes in the types
of respirators being used. The written
program must therefore be updated as
necessary to reflect and address these
changes. Employees should always no-
tify their employer if a change in the
workplace occurs that conflicts with,
or may not be covered by, the exist-
ing respirator training or established
workplace policies or procedures.
Choosing the correct respirator
Selecting the right equipment is a
three-step process:
1. Define the hazard and determine its
extent.
2. Evaluate user factors that affect res-
pirator performance and reliability.
3. Select an appropriate respirator that
is certified by the National Institute
for Occupational Safety and Health
(NIOSH) [1]. Note that equipment
29

Cover Story
must always be used in accordance
with the specifications accompany-
ing the NIOSH certification.
When selecting respirators, employ-
ers must consider the chemical and
physical properties of the contami-
nant, as well as the toxicity and con-
centration of the hazardous material
and the amount of oxygen present.
Other selection factors are the nature
and extent of the hazard, employee
work rate, the size of the area to be
covered, worker mobility, work re-
quirements and conditions, and the
limitations and characteristics of the
available respirators.
Air-purifying respirators use filters
or sorbents to remove harmful sub-
stances from the air. They range from
simple disposable masks to sophis-
ticated devices. They do not supply
oxygen and must not be used in oxy-
gen-deficient atmospheres or in other
atmospheres that are immediately
dangerous to life or health (IDLH).
Atmosphere-supplying respirators
are designed to provide breathable air
from a clean air source other than the
surrounding contaminated work at-
mosphere. They include self-contained
breathing apparatus (SCBA) units
(Figure 4a) and supplied-air respira-
tors (SARs; Figure 4b).
The time needed to perform a given
task, including the time necessary to
enter and leave a contaminated area,
is an important factor in determin-
ing the type of respiratory protection
needed. For example, SCBAs, gas
masks, or air-purifying chemical-car-
tridge respirators provide respiratory
protection for relatively short periods.
On the other hand, an atmosphere-
supplying respirator that supplies
breathable air from an air compressor
through an air line can provide protec-
tion for extended periods.
If the total concentration of airborne
particulate matter is low, particulate-
filter air-purifying respirators can
provide protection for long periods
without the need to replace the filter.
Where there are higher concentra-
tions of contaminants, however, an
atmosphere-supplying respirator such
as the positive-pressure SAR offers
better protection for a longer time.
Since they use an independent air
source, SARs eliminate the need for
30 Chemical Engineering www.che.com December 2012
Voluntary respirator use
sec (c)(2)
Filtering facepiece Traditional facepiece
sec (c)(2)(ii)
concern about filter breakthrough
times, change schedules, or the use of
end-of-service-life indicators (ESLIs)
for airborne toxic materials factors
that must be considered when using
air-purifying respirators.
Respirators must not impair the
workers ability to see, hear, commu-
nicate or move as necessary to per-
form the job safely. For example, at-
mosphere-supplying respirators may
restrict movement and present other
potential hazards. SARs, with their
trailing hoses, can limit the area
the wearer can cover and may pres-
ent a hazard if the hose comes into
contact with machinery. Similarly, a
SCBA that includes a back-mounted,
compressed-air cylinder is both large
and heavy. This may restrict climb-
ing and movement in tight places,
and the added weight of the air cyl-
inder presents an additional burden
to the wearer.
Another factor to consider when
using respirators is the air-supply
rate. The wearers work rate deter-
mines the volume of air breathed per
minute. The volume of air supplied to
meet breathing requirements is very
significant when using atmosphere-
supplying respirators, such as self-
contained and air-line respirators that
use cylinders, because this volume de-
termines their operating lives.
Peak airflow rate is also important
in the use of a constant-flow SAR.
The air-supply rate should always be
greater than the maximum amount
of air being inhaled in order to main-
tain positive pressure in the respira-
tory enclosure.
The increased breathing resis-
tance of air-purifying respirators
under conditions of heavy work may
cause the user breathing difficulty,
especially in hot, humid conditions.
To avoid placing additional stress on
the wearer, use the lightest respira-
tor possible that presents the least
Written respirator program
sec (c)
Hazard assessment
sec (d)
Select type of protection
sec (d)
Medical evaluation
sec (c)(2)(i) sec (e)
Fit testing
sec (f)
Training
sec (k)
Respirator use
sec (g)
Cleaning and maintenance
sec (h)
Program evaluation
sec (l)
Record keeping
sec (m)
Figure 3. This simplified illustration
shows the requirements that employers
must follow to comply with the 29 CFR
Part 1910.134 Standard
breathing resistance.
SCBAs and some chemical canis-
ter respirators provide a warning of
remaining service time. This may be
a pressure gage or timer with an au-
dible alarm for SCBAs or a color ESLI
on the cartridge or canister. The user
should understand the operation and
limitations of each type of warning
device. For the many gas masks and
chemical-cartridge respirators with-
out ESLI devices, the employer must
establish and enforce a cartridge or
canister change schedule. Employees
should begin each work shift with new
canisters and cartridges.
Ensuring proper fit
Different types of respirators, and
even different brands of the same type
of respirator, have different fit char-
acteristics. No one respirator will fit
everyone. Some employees may be un-
able to get an adequate fit with certain
respirator models of a particular type
of respirator. This is why employers
must provide a sufficient number of
respirator models and sizes to ensure
 ADDITIONAL RESOURCES

A
dditional information on protecting yourself from airborne
hazards can be obtained from the U.S. Occupational Safety
and Health Administration (OSHA), the National Institute for
Occupational Safety (NIOSH), and the American National Stan-
dards Institute (ANSI). Safety product manufacturers websites can
also be a helpful resource. Here are a few useful website links:
www.osha.gov/law-regs.html For OSHAs standards
www.cdc.gov/niosh The home page for the and Healths
National Personal Protective Technology Laboratory, which con-
tains a variety of information and links related to respiratory
protection and other personal protective equipment
www.ansi.org The home page for ANSI
www.osha.gov/dte/library/materials_library.html OSHA
Directorate of Training and Educations Training and Refer-
ence Materials Library. Provides outreach training materials
(such as slide presentations) for OSHAs respiratory protection
standard, as well as links to additional resources about respi-
ratory protection
webapps.msanet.com/ResponseGuide/ Website from au-
thors company that contains two interactive tools: Chemical
Database & Respirator Selection and Cartridge Service Life Cal-
culator, as well as a complimentary download entitled Manag-
ing a Respiratory Protection Program, a guide to simplifying
compliance with 29 CFR Part 1910.134
w ww.osha.gov/dts/shib/respiratory_protection_bulletin
_2011.html OSHA bulletin, General Respiratory Protection
Guidance for Employers and Workers (2011), which provides
basic information to employers and workers who may find
themselves using respiratory protection for the first time. It also
provides information on what respirators are, how they work,
and what is needed for a respirator to provide protection
www.osha.gov/SLTC/etools/respiratory/index.html OSHAs
eTool, Respiratory Protection (1998), which provides employ-
ers with instruction on the proper selection of respiratory pro-
tection and the development of change schedules for gas/
vapor cartridges
www.osha.gov/Publications/OSHA3079/osha3079.html
OSHA guidebook, Respiratory Protection, publication no.
3079, (2002). This booklet clarifies key provisions of the respi-
rator protection standard, which was revised in January 1998.
It also provides specific guidance on respirator selection, fit test-
ing, hazard evaluation and medical evaluation
w w w. o s h a . g o v / P u b l i c a t i o n s / 3 2 8 0 - 1 0 N - 0 5 - e n g l i s h
-06-27-2007.html OSHA quick card, Respirators Quick
Card, publication no. 3280 (2005), which provides employ-
ees with a brief description of the different types of respira-
tors available
w ww.cdc.gov/niosh/npptl/topics/respirators/disp_part/
RespSource.html NIOSH Web page, NIOSH Respirator
Trusted-Source Information Page, which provides information
to understand the types of respirators, how to identify approved
models and outlets for purchase, a listing of all NIOSH-approved
and U.S. Food and Drug Admin. (FDA) -cleared surgical N95
respirators, a listing of recently revoked respirator approvals
and relevant user notices. It also contains information on how
to implement the use of respirators in the workplace and use
them appropriately, including commonly asked questions and
answers (fact sheets), respirator myths, the science of respirator
function and performance, and respiratory protective devices
not approved by NIOSH

that every employee can select an ac-
ceptable respirator that fits properly.
Corrective eyeglasses worn by
employees can also present a prob-
lem when fitting respirators. Special
mountings are available to hold cor-
rective lenses inside full facepieces.
A qualified individual must fit the
facepiece and lenses to provide good
vision, comfort and proper sealing.
Tight-fitting respirators cannot pro-
vide proper protection without a tight
seal between the facepiece and the
wearers face. Consequently, beards
and other facial hair, the absence of
normally worn dentures, facial defor-
mities, jewelry or head gear that proj-
ects under the facepiece seal can also
seriously affect the fit of a facepiece.
To ensure proper respiratory protec-
tion, check the facepiece each time you
wear the respirator. You can do this by
performing either a positive-pressure
or negative-pressure user-seal check.
Detailed instructions for performing
these checks are in Appendix B-1 of
OSHA 29 CFR 1910.134.
Qualitative versus quantitative
fit testing. Fit testing is required for
tight-fitting facepiece respirators. You
can test the effectiveness of the fit of
the facepiece two ways: qualitatively
and quantitatively.
Qualitative fit testing involves the
introduction of a harmless, but strong,
smell or irritating substance into the
breathing zone around the respirator
being worn. If no odor or irritation is
detected by the wearer, then this indi-
cates a proper fit.
Quantitative fit testing offers more
accurate, detailed information on res-
pirator fit. While the wearer performs
exercises that could induce facepiece
leakage, a fit-testing instrument nu-
merically measures the amount of
leakage into the respirator. This test-
ing can be done either by generating
a test aerosol as a test atmosphere,
using ambient aerosol as a test agent,
or using controlled negative pressure
to measure any leakage. Detailed in-
structions for performing both quali-
tative and quantitative fit testing are
contained in Appendix A of OSHA 29
CFR 1910.134.
Proper inspection and care
It is important to inspect all respi-
rators for wear and tear before and
after each use, giving special atten-
tion to rubber or plastic parts that
can deteriorate or lose pliability. All
parts including the facepiece, head-
band, valves, connecting tube, fittings
and cartridges, canisters or filters
must be in good condition. A respi-
rator inspection must include check-
Chemical Engineering www.che.com December 2012
ing the tightness of the connections.
SCBAs should be inspected at least
monthly, and air and oxygen cylinders
should be fully charged according to
the manufacturers instructions. This
inspection should include a check of
regulator and warning devices to en-
sure proper functioning. Employers
must keep a record of inspection dates
and results.
Chemical cartridges and gas mask
canisters should be replaced as neces-
sary to provide complete protection,
following the manufacturers recom-
mendations. Additionally, mechanical
filters should be replaced as necessary
to avoid high breathing resistance.
Only an experienced person is per-
mitted to make repairs, using parts
specifically designed for the respira-
tor. This person must consult the man-
ufacturers instructions for any repair,
and no attempt should be made to re-
pair or replace components or make
adjustments or repairs beyond the
manufacturers recommendations.
Employers must ensure that res-
pirators are cleaned and disinfected
as often as necessary to keep them
sanitary. In addition, the employer
must have emergency-use respirators
cleaned and disinfected immediately
after each use.
Respirators should be washed in
31
 Cover Story
a detergent solution and then disin-
fected by immersing them in a sani-
tizing solution. Cleaner-sanitizers
that effectively clean the respirator
and contain a bactericidal agent are
commercially available. Strong clean-
ing and sanitizing agents and many
solvents can damage rubber or plastic
respirator parts. Use these materials
with caution or after consultation with
the respirator manufacturer.
Store respirators in a manner that
provides protection against dust, sun-
light, heat, extreme cold, excessive
moisture and damaging chemicals.
When packed or stored, each respira-
tor should be positioned to retain its
natural configuration. Facepieces and
exhalation valves should rest in a nor-
mal position to prevent the rubber or
plastic from deforming.
Manufacturers tools
There are three especially valuable
tools that can assist in the develop-
ment, administration and mainte-
nance of a workplace respiratory pro-
tection program:
Respirator selection guide
Cartridge life expectancy calculator
Manufacturer training programs
Many respirator manufacturers
websites contain these tools. OSHAs
website contains a respirator selec-
tion guide.
Respirator selection guide. A res-
pirator selection guide is designed
to help users identify the proper
respirator solution to meet their in-
dividual needs. Choices include ev-
erything from SCBAs to half- and
quarter-type facemasks. In addition
to selecting the proper respirator, a
respirator selection guide also helps
users choose the correct filtering
element if their exposure situation
allows for the use of an air-purify-
ing respirator. The following steps
should be taken to use a respirator
selection guide.
Step 1: Assess your environmental
conditions to determine the hazard(s).
To determine an atmospheres oxygen
content or concentration level of gas-
eous contaminants, proper air sam-
pling must be conducted. The results
of this sampling will determine what
level of protection is required. Gener-
ally, respirator and cartridge selection
32 Chemical Engineering www.che.com December 2012
FigureS 4a and 4b. Supplied-air respirators are designed to provide breathable air
from a clean air source other than the surrounding work atmosphere. They include self-
contained breathing apparatus units (left) and air-line respirators (right)
is based on three factors:
1. The results of the atmospheric mon-
itoring or sampling program
2. The accepted American Conference
of Industrial Hygienists (ACGIH),
OSHA or NIOSH exposure limits for
the substance(s) present
3. The maximum use concentration of
a substance for which a respirator
can be used
An additional part of the hazard as-
sessment process is determining
whether the contaminant has ad-
equate warning properties. This af-
fects respirator selection because
chemical cartridge respirators and
gas masks should only be for routine
use against gases and vapors with
adequate warning properties, unless
the cartridge is equipped with an
end-of-service life indicator.
Step 2: Check chemical exposure
limits. After finding out what hazards
exist, compare the chemical concentra-
tion at your facility (determined by air
sampling) with the chemicals expo-
sure limits. Note that most respirator
selection guides use the lowest expo-
sure limit published by ACGIH, OSHA
or NIOSH as a baseline to determine
the need for respiratory protection.
One exception to this rule is when an
OSHA substance-specific standard
exists with applicable respirator use
limitations for protection against the
chemical of interest. In this case, the
OSHA requirement is specified.
Step 3: Select a respirator and car-
tridge. Now it is time to select a res-
pirator and cartridge. During this
process it is important to note that
recommendations in a respirator se-
lection guide are typically based on
appropriate levels of respiratory pro-
tection. If you choose, a higher level of
protection can be used by selecting a
respirator with a higher maximum-
use concentration or, if applicable, a
higher cartridge-filter efficiency. This
is advisable if the atmosphere in the
work area is prone to changes, since
the contaminant concentration has
the potential to become higher.
Cartridge life-expectancy calcu-
lator. Lets assume that you deter-
mined that the respirator of choice
should be a full-facepiece with a
multi-gas cartridge. It is now time to
use the cartridge life-expectancy cal-
culator. This tool is interactive and
is designed to provide guidance as to
how long a specific cartridge should
be used before it needs to be replaced.
The determination is called time of
breakthrough and refers to the point
at which a hazardous chemical will
begin to break through the cartridge
or filter without being absorbed or
collected by the protective barrier. ditions, accuracy in determining the to onsite training, some respirator
OSHA requires that respirator car- ambient conditions and user training manufacturers offer online programs
tridges be replaced at pre-determined experience. Ultimately, the appropri- through their websites. As part of an
intervals based on worksite-specific ate change-out schedule must be de- effective training program, the user
usage conditions [4]. veloped by a qualified professional. may be required to pass a test about
A cartridge life-expectancy calcu- Manufacturer training programs. the respirator, including how to wear
lator is very easy to use. Simply an- The third tool is the training pro- and maintain it. An employee training
swer a minimal number of questions grams that are available from res- program is required by OSHA.
by entering key environmental and pirator manufacturers. In addition Edited by Dorothy Lozowski
usage factors. The calculator then
provides a suggested maximum ser-
vice time for the cartridge(s) being
used in that specific environment.
Although cartridge calculators may
vary somewhat, they all require the
same basic input:
Chemical hazard and concentration
Temperature
Relative humidity
Atmospheric pressure or altitude
Type of respirator and cartridge
Type of work and average breathing
rate (light, moderate or heavy)
Breakthrough concentration pref-
erence
Safety factor
When using a cartridge life-expectancy
calculator you must remember that
this tool provides a guideline for your
final decision on cartridge replace-
ment. Several other factors should
also be considered, such as changes
in conditions during the work shift,
changing climate conditions (seasonal
changes), condition of the cartridge
or respirator, cartridge storage con-

References
1. A list of NIOSH-approved respirators can be
found at http://www.cdc.gov/niosh/npptl/top-
ics/respirators/cel/ Get more out of your coal.
2. About ANSI Overview, http://ansi.org/about_ Only too often do we fail to see the treasures that are right in front of us.
a n s i / o v e r v i e w / o v e r v i e w. a s p x ? m e n u i d With our solids gasification technology you can make more out of any
=1 (accessed September 18, 2012)
feedstock. Why not contact us: you might just be surprised!
3. The ANSI/AIHA Z88.2 Standard Practices
for Respiratory Protection is available on As a leading EPC contractor, we also have a proprietary portfolio of
the ANSI website, www.ansi.org technologies. And we network intelligently within the ThyssenKrupp Uhde
4. Federal Register, p. 1272, col.1 OSHA group based on our business philosophy Engineering with ideas.
1910.134(d)(3)(iii)

Author
Dennis Capizzi is the prod-
uct line manager for air pu-
rifying respirators at MSA
Safety (1100 Cranberry
Woods Dr., Cranberry Twp.,
PA, 16066; Phone: 724-776-
8600; Email: dennis.capizzi@
MSAsafety.com), where he
has worked since 2007. He
serves as head of the North
American Air-Purifying Res- ThyssenKrupp Uhde
pirator Steering Team, and www.uhde.eu
has a seat on the International Safety Equip-
ment Assn.s Committee for Respiratory Protec-
tive Escape Devices.
Circle 16 on p. 60 or go to adlinks.che.com/40274-16
Chemical Engineering www.che.com December 2012 33
 Solids
Environmental
Processing
Manager
Fire-Water Pumps
for CPI Facilities
Follow this guidance to improve the selection, design
and operation of pumps handling water for fire-
fighting and related systems
Amin Almasi
Consultant
F
ire-water pumps are criti-
cal machines that save lives
and prevent chemical pro-
cess industries (CPI) facili-
ties from damage more so than
any other plant components. Fire-
water pumps are nearly always cen-
trifugal pumps with capacity from
around 203,000 m3/h. Specific re-
quirements for fire-water pumps
are briefly noted in fire codes (such
as NFPA 20 [1]), but this may not
be sufficient for specifying high-
performance fire-water pumps in
a way that ensures good reliability
and operation as well as optimum
price. This article provides practi-
cal notes on fire-water pumps to ex-
pand upon the information that can
be found in existing fire codes.
Various styles and configurations
of fire-water pumps are available
at different prices. In addition to
proper selection, fire-water pumps
must be properly integrated into
the overall fire-water system, as an
integral part of the CPI facility.
Figure 1 shows an example of
a skid-mounted, diesel-engine-
driven fire-water pump package.
The performance and reliability of
a fire-water pumping system is an
important issue, and details of the
fire-water pumping system are usu-
ally a part of risk studies, HAZOP
and inspection activities.
Fire-water pumps are important
to different stakeholders, including
clients, investors in the CPI plant,
and insurance providers. Usually
around 2035% of the insurance-
deficiency rating points for a CPI
plant are related to inadequate fire-
water pumping systems. On aver-
age, 510% of all fire-water pumping
systems in CPI plants have failed to
provide satisfactory services at the
time required (as evidenced during
actual fire cases or drill-type exer-
cises). Thus, it is critical to design,
implement, operate and maintain
these critical systems beyond the
minimum requirements set forth in
published fire codes.
Fire-water system
Fire-protection efforts are categorized
as passive or active. The primary
passive measures for fire protection
include efforts to ensure sufficient
clearances, install protective barri-
ers, limit and protect fuel sources,
and other steps designed to reduce
fire risks (such as the use of less
hazardous materials, processes and
equipment). By comparison, active
fire-protection systems are designed
to detect and apply fire-protection
measures, which usually rely on some
effort to actually extinguish the fire.
Commonly used system components
include fire hydrants, monitors, hose
reels, water-spray systems, deluge-
type fire-protection systems or water-
exposure cooling systems. The fire-wa-
ter pump plays an important role in
most active fire-protection systems.
System design and sizing
There have been different sets of
rules to define the required flow
and head of fire-water pumps. In
other words, a variety of differ-
Chemical Engineering www.chemengonline.com DECEMBER 2014
FIGURE 1 Fire-water pump assem-
blies are typically skid-mounted to ease
installation and operation. Shown here
is an example of a skid-mounted, diesel-
engine-driven, fire-water pump package
ent guidelines have been used to
estimate the ideal water volume
and flow requirements for CPI fire
cases, depending on plant specifics,
applicable codes, regulations and
fire-fighting methods. As a result,
these specifications will vary based
on whether the plant is using fire
control, fire suppression, exposure
cooling and so on.
Fire-water demands are usually
calculated based on the maximum
rate of water that will be required for
a worst-case scenario typically a
potential scenario involving a large,
single-fire incident. The most remote
unit(s) from the fire-water pumps, or
the largest unit(s), are typically exam-
ined to identify the worse possible fire
scenario(s). The potential scenario of
a vast fire in the largest unit should
be used to define the capacity of fire-
water system. The most-remote fire
unit(s) should be used to define maxi-
mum rated pressure of the fire-water
pump.
CPI fire cases can be very differ-
ent, considering the different types
of materials handled and the types
of operations carried out at differ-
ent facilities. Today, computerized
simulations play a critical role in
identifying and modeling potential
fire scenarios, validating fire-fight-
ing methodologies, and estimating
the required water capacity.
When evaluating a potential
fire-fighting scenario, additional
pressure (a safety margin to the
calculated head) should be added
to maintain the fire-water pres-
sure in all remote units and critical
fire-fighting systems; this is nec-
essary to ensure that a fire-water
stream with adequate pressure can
be maintained to support all appli-
53
 Environmental Manager
cable fire-fighting systems should
there be a fire in a unit.
Fire-water and utility-water sys-
tems have sometimes been com-
bined in non-critical plants. In the
event of a fire, the connected util-
ity water system would be tripped.
However, these combined systems
are always risky. Various fire codes
recommend that no utility-water
connections be made to the fire-wa-
ter system. In some special cases,
the fire-water system may be used
for emergency process-cooling re-
quirements, but only as the second-
ary (reserve) supply.
Fresh (treated) water is always
preferred (over seawater, brackish
or untreated water, for instance) for
fire-water systems in all onshore
plants. Untreated or brackish water
can cause many issues such as cor-
rosion, which can potentially wreak
havoc on the system components.
In general, engineers should pur-
chase or construct the fire-water
pumping system and the fire-water
distribution system using proper
materials (for instance, selecting
suitable corrosion-resistant mate-
rials or proper protective coatings),
because untreated raw water (such
as seawater) could be used as the
secondary source for extra fire-wa-
ter capacity, in the case of an unex-
pected fire event. If this happens,
the fire-water system should be
flushed with treated water after the
incident, to remove residual traces
of untreated source water.
Selecting fire-water pumps
Centrifugal pumps with a rela-
tively flat characteristic perfor-
mance curve (a graph of head ver-
sus flowrate) are generally selected
for fire-water pumps. Ideally, the
head should rise continuously from
the rated point to the shutoff point,
with only a small increase of head
(say, a 915% rise of the head from
rated point to shutoff point). These
pumps can provide a steady, stable
flow of water at a relatively uniform
pressure over a wide range of re-
quired fire-water flowrates.
A relatively flat performance
curve is always encouraged for cen-
trifugal fire pumps for the following
reasons:
54 Chemical Engineering www.chemengonline.com DECEMBER 2014
The control of a fire event usu-
ally requires variable amount
of water at a relatively constant
pressure
Fire-water pumps are typically
operated in parallel. A relatively
flat curve ensures troublefree
parallel operation
Sometimes, a large amount of
water can be required by the fire-
water system to battle a vast fire;
in those cases, the required water
could be considerably larger than
the rated flow of the pump. In
this regard, the fire-water pump
overload point (the end operating
point at the right side of the pump
curve) should demonstrate a capac-
ity of more that 150% of the rated
capacity at a head more than 70%
of the rated point. In other words,
operation point could move to the
far right side of the rated point and
that point should offer sufficient
flow and head.
A steep pump curve should al-
ways be avoided. As a rough indica-
tion, the average slope of a fire-wa-
ter pump curve should preferably
be around 1020% (for instance, an
average slope of 1/10 up to 1/5).
Fire-water pumps can idle
against closed valves for a short
period of time. In other words, for a
short time, the pump should be able
to operate in a closed water system
without any fire-water application.
Check valves should be provided
at both the discharge and the suc-
tion. The rated pressure of a fire-
water pump could be 430 barg.
Single-impeller centrifugal pumps
(for applications that require pres-
sure below roughly 12 barg), and
multi-impeller centrifugal pumps
(for higher-pressure systems) are
also commonly used.
The differential pressure of a
pump is proportional to both the
square of the rotating speed and
the square of the impeller diameter.
A discharge pressure of around 10
barg can be obtained by a relatively
large, single-impeller pump (with a
suitable speed).
Overhung (OH) pumps have been
used for small- and medium-sized
fire-water pumps. Users should
consider the between-bearing (BB)
pump design when size, power rat-
FIGURE 2. Shown here are several
examples of fire-water pumps; an identi-
cal spare pump is commonly used to in-
crease the reliability of fire-water pump-
ing systems
ing and power-density exceed a cer-
tain level. As a rough indication,
this limit could be 400 kW.
As noted, the fire-water pumps in-
stalled at any given facility should
be able to operate in parallel. How-
ever, there are some challenges
and issues in ensuring parallel op-
eration. Even in certain conditions,
pumps designed to operate in paral-
lel could be subject to overheating
or damage. A well-known danger is
one pump operating at higher flow,
forcing another pump to operate at
lower flow; operation at lower flow
can be damaging to the pump.
When fire-water pumps are oper-
ated in parallel, the pump with the
lowest head may work at a reduced
flowrate. In this way, the pump
could work far from the best effi-
ciency point with a very low effi-
ciency, high friction and heat gener-
ation, which can result in damage.
Even in identical fire-water pumps,
pumps that have been in use for
more hours (and thus has probably
been subjected to more wear), pumps
with minor defects, and pumps with
slightly lower speed could all be
subjected to a reduced flow, which
can create problems during an ac-
tual fire event. Because of this ef-
fect, operators should rotate pumps
over time, so that each pump works
as the main fire-water pump for
some period of time; this can help to
ensure even wear patterns among
identical pumps in service. Individ-
ual protection against the minimum
flow (to ensure a minimum flow for
each pump) is recommended.
Monitoring of the differential
temperature of each pump can pro-
vide valuable insight for estimating
the parallel operation issue (the re-
 Environmental Manager
duced-flow problem) and resulting
inefficient operation (overheat). In
case of reduced flow, the differential
temperature (the discharge temper-
ature minus the suction tempera-
ture) would rise and could indicate
such a malfunction.
Because of small leakage and
small consumption of fire water, the
pressure in a fire-water network
could decrease slightly. Fire-water
pump systems are usually designed
in a way that spare pumps should
be started if fire-water pressure
dropped below a certain level. How-
ever, a slight pressure drop should
not lead always to the startup of a
large fire pump, as this could result
in many unnecessary on-off oper-
ating cycles of the main fire water
pump (Figure 2).
On the other hand, small pres-
sure changes resulting from varia-
tions in fire-water consumption
during a fire incident can result in
an unstable operation of the main
fire-water pump(s). For instance,
this may lead to unnecessary fast
changing of the operating point of a
large pump, which can result in per-
formance and reliability issues.
Smaller-capacity pumps (known as
jockey pumps) are usually employed
in conjunction with the main pump(s)
to maintain a relatively constant fire-
water pressure. Jockey pumps usu-
ally initiate operation after a rela-
tively small pressure drop (say 0.51
bar) in a fire-water system.
Main fire-water pumps are typi-
cally electrically driven and the
spare (backup or reserve) fire-water
pumps are typically driven by die-
sel engine. A commonly used ar-
rangement for critical CPI facilities
is to install six fire-water pumps,
including two electric-motor-driven
pumps, two diesel-engine-driven
pumps and two jockey pumps. Fire-
water pumps are nearly always pro-
vided on a prefabricated skid. This
packaging concept can help to ease
the alignment and installation is-
sues and ensure high reliability.
As noted earlier, the fire code
NFPA 20 is dedicated to fire-water
pumps. It specifies proper require-
ments for pump tests, pump per-
formance curves, pump accessories
and auxiliaries, and some packag-
ing details. However, in this au-
thors view, NFPA 20 should be con-
sidered as a minimum requirement
for a fire-water pump. For critical
fire pumps, the well-known API-
610 pump standard [2] is addition-
ally applied.
API-610 fire-water pumps
The API-610 pump standard is used
to ensure the reliable operation of
high-performance pumps, mainly in
the oil-and-gas, petroleum refinery
and petrochemical sectors. The API-
610 is usually considered to be the
minimum specification for pumps
that handle hazardous, flammable,
toxic and explosive liquids since
any reliability issue associated with
these pumps could result in a poten-
tial disaster. API-610 pumps are also
very popular in applications with
extreme temperatures, including
pumps for both high-temperature
service (such as boiler-feed-water, or
BFW, pumps) and low-temperature
applications (for example, pumps
used in liquid petroleum gas (LPG)
liquefied oxygen, and liquefied natu-
ral gas (LNG) service). For critical
(high-risk) CPI units, fire-water
pumps are usually specified to com-
ply with the API-610, to ensure that
they are able to achieve a high re-
liability level the same as other
pumps in the unit.
Engineers often struggle with
whether or not to use API-610-
compliant fire-water pumps for
a CPI plant. This decision would
depend on the application, pump
head, power rating, capacity, pump
speed and expected reliability. The
main variable is CPI service (the
CPI plant and expected reliability).
For instance, for critical units han-
dling flammable liquids and gases,
API-610-compliant fire-water pumps
are often preferred.
For a fire-water pump with dif-
ferential pressure more than 20
bar, API-610 is usually specified.
The pump power rating is a bit
tricky, since there are many non-
API fire-water pumps available
(with successful references) that
are intended for high power ranges
in a wide array of industry applica-
tions. As a general rule of thumb
for many CPI plants, API-610 can
Chemical Engineering www.chemengonline.com DECEMBER 2014
be considered for pumps above the
350-kW range.
Fire-water pump arrangement
The location for fire-water pumps
should be selected carefully to mini-
mize various risks and potential
hazard situations. Explosions or
high-hazard fires are major con-
cerns, which can disable fire-water
pumps. Ideally, there should be
4080 m of clearance between fire-
water pumps and a hydrocarbon
or chemical process unit or storage
area. This limit should also be re-
spected for some utility areas, such
as a power-generation units, gas-
compression units, oxygen-genera-
tion units, and similar.
The possibility of an unconfined
vapor-cloud explosion is one of the
main concerns, as this could disrupt
utilities, damage major support fa-
cilities, and damage the fire-water
pumping system. Generally, there
is a great possibility of the electri-
cal network or the steam-distribu-
tion system failing in the event of
a major explosion or extensive fire
event. This underscores the critical
role of independent, diesel-engine-
driven fire-water pumps. Fire-wa-
ter diesel engines should generally
comply with NFPA 37 [3].
Regarding the diesel fuel-tank ca-
pacity, typically, a 12-h duration is
specified as the minimum require-
ment. However, some critical CPI
plants require 24-h-capacity fuel
tanks for each fire-water pump-die-
sel engine.
Meanwhile, each diesel engine
should be provided with independent
auxiliaries and accessories, includ-
ing a dedicated fuel system and fuel
tank. The startup of the engine is
commonly managed by a battery sys-
tem (with two independent barriers).
The failure of a diesel engine is
usually the result of a problem with
one of the auxiliary systems. Major
reasons for such a failure include
fuel-system issues, a lubrication-
system problem, a starting issue,
a wiring problem or component fa-
tigue. Only clean, high-quality die-
sel fuel should be used, and special
attention is required for the lubri-
cation oil selection and supply.
Proper overhauls and repairs
55
 Environmental Manager
are required, just like for any other
properly designed combustion en-
gines. Experience has shown that
the diesel-engine-driven, fire-water
pump is the most reliable option
currently available for severe loss
incidents in a CPI plant. The reli-
ability and availability of micro-
turbines (small gas turbines in the
50400 kW range) could be higher
than diesel engines, but their ef-
ficiencies are relatively lower (in
terms of lower operating duration
with the same amount of fuel). Cur-
rently they are not popular for fire-
water pump systems.
Fire-water pumps are typically
arranged for both manual and au-
tomatic startup. Automatic startup
is expected to happen rapidly, in a
very reliable manner, once a fire
event has been detected. Fire-water
pumps are usually stopped manu-
ally at the pumps local control
panel. In other words, operator in-
tervention is usually used to turn
off the pump, once the situation has
stabilized and the fire is out.
A suitable enclosure (or building)
should be provided for fire-water
pumps. Sufficient reinforcement
should be considered for the fire-
pump enclosure. This is very im-
portant. For example, in the case
of a major earthquake, fire-water
pumping systems need to be fully
operational to respond to fire events
resulting from the earthquake. An
open-sided shelter is not recom-
mended. And, fire-water pumps
should be located at a higher eleva-
tion than the majority of the CPI
facility and upwind of it.
To provide another layer of protec-
tion (in order to avoid common fail-
ures), the main fire-water pumps,
and any other reserve or support-
ing fire-water pumps should not be
located immediately next to each
other. Locating fire-water pumps at
two separate locations can improve
both the fire-water system reliabil-
ity and overall fire-water network
hydraulic behavior.
Special fire-water pumps
For critical CPI plants, additional
emergency (or reserve) firewater
pumps should be provided in ad-
56 Chemical Engineering www.chemengonline.com DECEMBER 2014
dition to the conventional fire-water
pumps to supply seawater (for
CPI plants located at the coastal re-
gions) or other sources of untreated
raw water (such as untreated water
from a lake or water wells), to
quickly supply additional capacity
to the plants fire-water network.
For a seawater-based emergency
fire-water pumping system, the
pumps are usually submerged in
seawater. For some locations, the
seawater level may fluctuate from
7 m to +16 m. Considering that
there is often a long distance from
the sea to the CPI facility, these fire-
water pumps should be designed
to produce a relatively high head.
These special fire-water pumps are
usually multi-impeller pumps. Prop-
erly designed highly reliable pumps
are always specified for such com-
plex service. Electric-motor-driven
submersible pumps, or sometimes
hydraulic-driven pumps are used
for these special applications. These
are usually down-hole, vertical tur-
bine-type pumps.
In some cases, local regulations
or plant specifications require three
dedicated fire-water pumps (as the
minimum) for special CPI applica-
tions (such as CPI plants that han-
dle highly explosive or highly flam-
mable materials). In such cases, one
of the main concerns is personnel
safety during a major fire case and
provisions must be made to ensure
a safe personnel evacuation.
Installation and commissioning
The piping installation and connec-
tion to a fire-water pump can lead
to relatively high loads on the pump
nozzles and to the pump trains sen-
sitive components, such as bearings,
coupling and rotating assemblies.
The fire-water pumps are often
left on standby, therefore any high
nozzle loads or misalignment might
be left unchecked. This can poten-
tially wreck the pump in the first
hours of operation in the event of
fire. Periodic checks are important.
A well-known method to monitor
the movements and deformations of
critical machinery components dur-
ing the piping connection (piping
flange and machinery flange bolt
tightening) is to install dial indica-
tors (or other types of indicators)
that monitor movements in critical
parts of the machinery train. Usu-
ally, two dial indicators are used to
observe movement in each machin-
ery component (such as the driver,
the fire-water pump and the gear
unit, if any) compared to the base-
plate or foundation (the main pur-
pose is to identify improper support
of machinery, called soft-foot).
Two dial indicators can be used
to monitor critical bearing housing
movements in the fire-water pump
(usually in x and y directions). Ac-
ceptable movements should be
below 0.04 mm (40 micrometers) to
ensure a proper piping-pump con-
nection. A similar limit should be
applied to movements in all critical
pump train locations (such as bear-
ing housing, coupling, machinery
support and others), and suction and
discharge nozzle flanges (in terms
of limiting deformations in all direc-
tions). For special fire-water pumps,
depending on the machinery design,
speed, power rating and applica-
tions, a limit higher or lower than
the above-mentioned (0.04 mm)
may be specified. n
Edited by Suzanne Shelley
References
1 NFPA 20, Standard for the Installation of Sta-
tionary Pumps for Fire Protection, National
Fire Protection Assn. (NFPA), 2013.
2 API 610, Centrifugal Pumps, American Petro-
leum Inst. (API), 2009.
3 NFPA 37, Standard for the Installation and
Use of Stationary Combustion Engines and
Gas Turbines, National Fire Protection Assn.
(NFPA), 2014.
Author
Amin Almasi is a rotating-
equipment consultant in Aus-
tralia (Email: amin.almasi@
ymail.com). He previously
worked at Worley Parsons Ser-
vices Pty Ltd. (Brisbane, Aus-
tralia), Technicas Reunidas
(Madrid, Spain) and Fluor
Corp. (various offices). He
holds a chartered professional
engineer license from Engi-
neers Australia (MIEAust
CPEng Mechanical), a chartered engineer cer-
tificate from IMechE (CEng MIMechE), RPEQ
(registered professional engineer in Queensland)
and he also holds M.S. and B.S. degrees in me-
chanical engineering. He specializes in rotating
machines including centrifugal, screw and recip-
rocating compressors, gas and steam turbines,
pumps, condition monitoring and reliability. Al-
masi is an active member of Engineers Australia,
IMechE, ASME, Vibration Institute, SPE, IEEE,
and IDGTE. He has authored more than 60 pa-
pers and articles dealing with rotating machines.
 Solids
Environmental
Processing
Manager

Eye-and-Face Personal
Protective
Equipment
Gateway Safety

Protecting the eyes and face in the workplace FIGURE 1.

Safety glasses help
shield eyes from impact hazards
is imperative to preventing the estimated
1020% of work-related eye injuries OSHA compliance
that result in temporary or permanent vision loss OSHA defines minimum require-
ments for workplace PPE in Subpart I
of the Occupational Safety and Health
Victor J. DAmato Standards for general industry em-
Atrium Environmental Health and Safety Services ployers (29 CFR 1910), which include

E
ye injuries of all types occur
at a rate of more than 2,000
per day in the U.S., and half
of those occur in the work-
place. In 2006, the U.S. Bureau of
Labor Statistics (BLS) reported over
52,000 workplace injuries to the eyes
and face, and approximately 70% of
those were eye injuries [1]. The need
to protect the face and eyes in the
workplace is clear.
Implementing an eye-and-face
protection program is important for
several reasons. One is compliance.
Between October 2006 and Septem-
ber 2007, more than 450 citations re-
sulted in more than $215,000 in fines
and penalties to manufacturers for
failure to comply with the U.S. Occu-
pational Safety & Health Administra-
tion (OSHA) standard for eye and face
protection [2].
Another reason is the cost of eye
and face injuries. The financial cost
of these injuries is enormous and can
range from $300 to $3,000 per case.
More than $300 million per year is lost
in production time, medical expenses,
and workers compensation according
to OSHA. The return on investment
for implementing an eye-and-face
safety program is huge as its cost is
48 Chemical Engineering www.che.com February 2009
relatively inexpensive compared to
the potential expense of an injury. But,
no dollar figure can adequately reflect
the personal toll these accidents take
on injured workers. Prevent Blind-
ness America estimates that 1020%
of work-related eye injuries result in
temporary or permanent vision loss.
This article outlines the basics of
eye-and-face protection programs for
plant operations, including key com-
pliance requirements and practical
information for personal protective
equipment (PPE) selection.
The causes of injuries
Most impacts to the face come from
flying objects, usually particulate
matter. When a particle hits the face it
might cause a cut or scratch, and the
skin will repair itself over time given
proper medical attention. If the par-
ticle hits the eye, it causes direct dam-
age, and it is far more difficult for the
eye to repair itself.
BLS reports that of the nearly 36,000
injuries to workers eyes in 2006, nearly
70% resulted from contact with objects
like flying particles, falling objects or
sparks striking the eye. Contact with
chemicals was responsible for the re-
maining 30% of eye injuries.
the chemical process industries (CPI).
The general requirements of this stan-
dard include that employers select
PPE for their employees based on the
hazards present in the workplace, and
that employers train employees on the
proper use and limitations of the PPE
selected for use. In November 2007,
OSHA updated the standard to spe-
cifically require employers to pay for
the PPE required under the standard,
with some limited exceptions [3].
The OSHA Eye and Face Protection
Standard (29 CFR 1910.133) was estab-
lished specifically to protect employees
from eye and face hazards in the work-
place. To comply with this standard,
employers must ensure the following:
1. Employees who may be exposed to
eye or face hazards from flying parti-
cles, molten metal, liquid chemicals,
acids, caustic gases, vapors or poten-
tially injurious light radiation use
appropriate eye or face protection.
2. Employees who may be exposed to
a hazard from flying objects use
eye protection that provides side
protection.
3. Employees who require prescrip-
tion lenses while engaged in op-
erations that involve eye haz-
ards wear eye protection that
incorporates the prescription in
 Gateway Safety
FIGURE 2. Face
shields protect the
entire face, but do not
protect the eyes from
impact hazards when
worn alone. Safety glasses
or goggles should also be worn
its design, or that they wear eye
protection that can be worn over
the prescription lenses without
disturbing the proper position of
the prescription lenses or the pro-
tective lenses.
4. Eye-and-face personal protective
equipment is distinctly marked so
that the manufacturer is clearly
identified.
5. Employees who may be exposed
to injurious light radiation during
work process, such as welding, ther-
mal cutting, brazing, torch solder-
ing, and using lasers, use eye protec-
tion equipped with filter lenses that
have a shade number appropriate
for the work being performed.
implementation
Implementing a practical, effective
PPE program includes the following
straightforward steps, which are de-
scribed in more detail below.
1. Assess the workplace for hazards.
2. Select PPE based on these hazards.
3. Train workers on proper PPE use.
4. Determine how prompt emergency
care will be provided in the event of
an accident.
Hazard assessment
OSHA provides guidance on conduct-
ing hazard assessments, which are
summarized as follows [4].
1. The first step in the hazard assess-
ment for eye and face protection is
to walk through your facility and
identify the potential hazards that
can injure the eyes or face. Hazards
should be assessed by occupation or
trade, work area and work activity.
table 1. Hazard Assessment
Hazard type Examples of hazard
Impact Flying objects, such as
large chips, fragments,
particles, sand and dirt
Heat Anything emitting ex-
treme heat
Chemicals Splash, fumes, vapors
and irritating mists
Dust Harmful dust
Optical Radiant energy, glare
radiation and intense light
Source: U.S. Dept. of Labor, Occupational Safety & Health Administration (OSHA), Eye and
Face Protection eTool, http://www.osha.gov/SLTC/etools/eyeandface/ppe/selection.html
2. Identify the sources of the haz-
ards. Talk to employees as they
know best about their trades,
work areas and processes.
Consider all possible hazards,
even those that seem minor or
unlikely. Some examples of eye and
face hazards and typical sources are
summarized in Table 1.
3. Organize the data collected. De-
velop a table or spreadsheet listing
the occupations, work areas, and
processes you identify, the hazards
identified in each category, and the
sources of the hazards.
4. Analyze your data, considering the
probability and severity of each
hazard. Keep in mind that mul-
tiple hazards and sources may be
present. This will help in deciding
which PPE is appropriate.
5. Certify your assessment. OSHA re-
quires that every employer verify its
assessment, identifying the areas
assessed, who performed the assess-
ment, and the dates it was performed.
6. Review and update your assessment
on a routine basis, or if changes in the
workplace or processes are planned.
PPE selection
The American National Standards
Institute (ANSI) standard for eye and
face protection (Z87.1) sets forth cri-
teria related to the description, gen-
eral requirements, testing, marking,
selection, care and use of eye and face
PPE in the workplace. The 2003 edi-
tion specifies requirements for safety
glasses, safety goggles, face shields
and full-face and hooded respirators.
This ANSI standard is revised and
updated routinely to reflect current
technologies and the current state of
practice. While most PPE manufac-
turers have adopted the 2003 edition
of the standard in new products, eye
and face PPE currently being used
Chemical Engineering www.che.com February 2009
Common related tasks
Chipping, grinding, machining, masonry
work, wood working, sawing, drilling, chisel-
ing, powered fastening, riveting and sanding
Furnace operations, pouring, casting, hot
dipping and welding
Acid and chemical handling, degreasing,
plating and working with blood
Woodworking, buffing and generally dusty
conditions
Welding, torch-cutting, brazing, soldering and
laser work
may have been manufactured to
meet earlier editions of ANSI Z87.1.
OSHA requires that eye and face
PPE purchased after July 1994 be
manufactured to meet the 1989 edi-
tion of ANSI Z87.1.
Eye and face PPE that meets the cri-
teria specified in ANSI Z87.1 must be
marked with a manufacturers unique
marking and the Z87 marking. Spe-
cial purpose lenses and shaded lenses
must also be marked indicating the
purpose and shade. Eye and face PPE
must be selected based on the hazards
identified in the hazard assessment.
Some examples of the types of protec-
tion available for hazards are summa-
rized in the next sections.
Ultimately, the eye and face pro-
tection must be comfortable to wear
and not interfere with the employees
work. Poor-fitting PPE will not afford
the necessary protection, may be un-
comfortable to wear, and may not get
used. Since eye-and-face protection
devices can come in different sizes,
care should be taken to ensure that
the right size is selected.
Impact hazards. Safety spectacles
(Figure 1) are intended to shield the
wearers eyes from impact hazards,
such as flying fragments, objects,
large chips, and particles. Workers
are required to use eye safety spec-
tacles with side shields when there is
a hazard from flying objects. Non side-
shield spectacles are not acceptable
eye protection for impact hazards. The
selection of safety glasses is almost in-
finite as a host of manufacturers have
stepped up to meet employee demand.
Safety goggles are also intended to
shield the wearers eyes from impact
hazards. Goggles fit the face immedi-
ately surrounding the eyes and form
a protective seal around the eyes. This
prevents objects from entering under
or around the goggles.
49
 Environmental Manager
Face shields (Figure 2) are intended
to protect the entire face or portions of
it, but they do not protect employees
eyes from impact hazards when worn
alone. Face shields should be used in
combination with safety spectacles or
goggles for additional protection. Face-
shield windows are made of varying
types of transparent materials and
in varying thicknesses, both of which
should be considered when selecting
face shields for specific tasks. Window
and headgear devices are available in
various combinations to be compatible
with other PPE, such as hardhats.
Welding shields (Figure 3) provide
eye and face protection from flying
particulate matter during welding,
thermal cutting and other hot-work
activities. These shields are typically
equipped with shaded lenses that pro-
vide protection from optical radiation.
Heat hazards. Working with heat
requires goggles or safety spectacles
with special-purpose lenses and side
shields. However, many heat-hazard
exposures require the use of a face
shield in addition to safety spectacles
or goggles. When selecting PPE, con-
sider the source and intensity of the
heat and the type of splashes that may
occur in the workplace.
Chemical hazards. Safety goggles
(Figure 4) protect the eyes, eye sock-
ets, and the facial area immediately
surrounding the eyes from a vari-
ety of chemical hazards. The protec-
tive seal formed around the eyes is
especially important when work-
ing with or around liquids that may
splash, spray or mist. Several kinds
of cover-type goggles are available:
direct vented, indirectly vented or
non-vented. Its important to select
the right type. Vented goggles, for
example, may be less effective in pro-
tecting the eyes from splashes and
respiratory aerosols than non-vented
or indirectly vented goggles.
Face shields are intended to protect
the entire face from a variety of chemi-
cal hazards, particularly when pour-
ing chemicals or where splashes may
occur. Face shields are considered sec-
ondary protection and must be used in
addition to safety goggles to provide
adequate protection.
Dust hazards. Working in a dusty en-
vironment can cause eye injuries and
50 Chemical Engineering www.che.com February 2009
presents additional hazards
to contact lens wearers. Safety
goggles should be worn when
dust is present. Safety goggles
are the only effective type of eye
protection from nuisance dust.
Optical radiation hazards.
Welding, thermal cutting, braz-
ing, laser work and similar
operations create intense con-
centrations of heat, ultraviolet,
infrared, and reflected-light ra-
diation. Some of these activities
can produce optical radiation
intensities greater than those
experienced when looking di-
rectly at the sun. Unprotected
exposure may result in eye in-
juries including retinal burns,
cataracts, and permanent blind-
ness. Many lasers produce invis-
ible ultraviolet, and other forms
of non-ionizing radiation. FIGURE 3. Welding shields are typically equipped
with shaded lenses to protect from optical radiation
The OSHA standard (29 CFR
1010.133) includes a table that lists Institute for Occupational Safety
the minimum shade requirements and Health (NIOSH) issued a Cur-
for eye protection during industrial rent Intelligence Bulletin (CIB) on
processes that generate optical radia- the use of contact lenses in chemical
tion. The selection of eye protection for environments. In this CIB, NIOSH
lasers should depend on the lasers in recommends that workers be permit-
use and the operating conditions, and ted to wear contact lenses when han-
should be consistent with the laser dling hazardous chemicals provided
manufacturers specifications. that certain safety guidelines are fol-
PPE and prescription lenses. If lowed and that contact lenses are not
employees wear reading glasses with banned by regulation or contraindi-
basic magnification, ANSI-approved cated by medical or industrial hygiene
safety glasses and goggles with read- recommendations. OSHA recommends
ing magnifiers in the lenses are an op- against contact-lens use when work-
tion. Its also possible to buy prescrip- ing with certain chemicals, including
tion safety glasses and goggles that acrylonitrile, methylene chloride, 1,2
are compliant with the ANSI stan- dibromo-3-chloropropane, ethylene
dard. ANSI-compliant goggles and oxide and methylene dianiline.
safety glasses are also available to fit The NIOSH recommendations are
over prescription glasses; but, they specifically intended for work with chem-
can be uncomfortable and less likely ical hazards, and do not address hazards
to be worn. from heat, radiation, or high-dust or
Wearing contact lenses under some high-particulate environments [5].
circumstances provides workers with Usable life of PPE. All PPE has a
a greater choice of eye and face pro- usable lifespan. Most manufacturers
tection and better visual acuity. How- recommend how frequently eye-and-
ever, the risk of injury to the eye from face protection equipment should be
chemical exposure is unknown for con- replaced. If the PPE becomes dam-
tact lens wearers compared with non- aged or distorted, it must be replaced
wearers working with chemicals. It is to work properly. As far as the usable
important to note that contact lenses life of a product, a good rule of thumb
are not eye protective devices, and is if your safety glasses are more than
wearing them does not reduce the re- five years old, they may not be compli-
quirement for eye and face protection. ant with the latest ANSI standards, so
In June 2005, the U.S. National you should get a new pair.
 Summary of NIOSH Recommended Safety Guidelines
for Contact Lens Use in Chemical Environments
1. Conduct an eye-injury hazard
evaluation* in the workplace
that includes an assessment of
the following:
Chemical exposures 3. Establish a written policy
Contact lens use
 Appropriate eye and face quirements for wearing contact
protection for contact lens lenses, including the eye and
wearers
2. Provide suitable eye and face
* NIOSH recommends that eye injury
hazard evaluations be conducted by
competent, qualified persons such as
certified industrial hygienists, certified
safety professionals, or toxicologists.
Information from the hazard evaluation
should be provided to the examining oc-
cupational health nurse or occupational
medicine physician.
Gateway Safety
Employee training
BLS reports that workers injured
while not wearing protective eyewear
often say they believed it was not re-
quired by the situation. Even though
the vast majority of employers furnish
eye protection to employees at no cost,
about 40% of workers report they re-
ceive no information on where and
what kind of eyewear should be used.
OSHA requires that all employees
who use PPE for protection against
workplace hazards be trained in when
and how to use their PPE. This train-
ing must include the following:
1. A summary of the occupations, work
areas or tasks where eye and face
protection is necessary.
2. What types of eye and face protec-
tion are necessary for each occupa-
tion, work area or task.
3. How the eye and face protection is
put on, taken off, adjusted and worn
to be effective.
4. The limitations of the eye and face
protection used.
5. The manufacturers recommenda-
tions on the proper care, mainte-
nance, useful life and disposal of eye
and face protection devices.
protection for all workers ex- about any defined areas where
posed to eye injury hazards, contact lenses are restricted.
regardless of contact lens 6. Identify to supervisors all con-
wear. tact lens wearers working in
chemical environments to en-
documenting general safety re- sure that the proper hazard as-
sessment is completed and the
proper eye protection and first
face protection required and aid equipment are available.
any contact lens wear restric- 7. Train medical and first aid per- contact lens wear on a case-
tions by work location or task. sonnel in the removal of contact
4. Comply with current OSHA lenses and have the appropri-
regulations on contact lens ate equipment available.
wear and eye and face 8. In the event of a chemical
protection. exposure, begin eye irriga-
5. Notify workers and visitors tion immediately and remove
Source: Contact Lens Use in a Chemical Environment, Current Intelligence Bulletin 59, NIOSH Publication No. 2005-139: June 2005
FIGURE 4. Safety goggles form a
protective seal around the eyes
Training must be con-
ducted prior to an em-
ployees assignment
to the work area or
task where the haz-
ards may be present, and
at the time when the eye and
face protection is issued. There
is no annual requirement for re-
training; the onus of ensuring that
employees understand when and how
to use their eye and face protection
is on the employer. OSHA does state
that when employees are found wear-
ing PPE improperly or not at all, they
need to be retrained.
Emergency care
OSHA mandates that if a hazard ex-
ists, a means for flushing the eye must
be provided. The eye-and-face protec-
tion program should include consider-
ations for emergency eyewash stations
that provide a full 15-min flush. There
is an ANSI standard (Z358) for eye-
wash stations. In the field, eyewash
bottles can be on hand for interim as-
sistance until an eyewash station can
be reached. There are also portable
eyewash stations available that can
dispense water for 15 minutes.
Resources
OSHA has an excellent Eye and
Face Protection eTool on its website
(http://www.osha.gov/SLTC/etools/
eyeandface/ppe/selection.html) that
can assist with hazard assessments,
provides information about selecting
PPE and summarizes specific OSHA
Chemical Engineering www.che.com February 2009
contact lenses as soon as prac-
tical. Do not delay irrigation
while waiting for contact lens
removal.
9. Instruct workers who wear con-
tact lenses to remove the lenses
at the first signs of eye redness
or irritation.
10. Evaluate restrictions on
by-case basis, taking into
account the visual require-
ments of individual workers
wearing contact lenses as
recommended by a qualified
ophthalmologist or optometrist.
requirements. The OSHA guidelines
in Ref. 4 also include a useful PPE se-
lection guide.
The American Industrial Hygiene
Association website (www.aiha.org) of-
fers more information on occupational
health and safety topics including a
list of industrial hygiene consultants
who specialize in safety issues.
Edited by Dorothy Lozowski
References
1. Bureau of Labor Statistics, Case and De-
mographic Characteristics for Work-related
Injuries and Illnesses Involving Days Away
From Work (2006), http://stats.bls.gov/iif/os-
hcdnew.htm
2. Occupational Safety & Health Administra-
tion Inspection Data, http://www.osha.gov/
pls/imis/industryprofile.html
3. Occupational Safety & Health Adminis-
tration, Employer Payment for Personal
Protective Equipment; Final Rule FR
72:64341-64430, Table XV-2 l, Vol. 72, No.
220, Nov. 15, 2007.
4. Occupational Safety & Health Administra-
tion, Compliance Guidelines for Hazard
Assessment and Personal Protective Equip-
ment Selection, 29 CFR 1910, Subpart I, Ap-
pendix B .
5. Contact Lens Use in a Chemical Environ-
ment, Current Intelligence Bulletin 59,
NIOSH Publication No. 2005-139, June 2005.
Author
Victor DAmato is the direc-
tor for Atrium Environmental,
Health and Safety Services,
LLC (11495 Sunset Hills Road
Suite 210, Reston, VA 20190;
Phone: 7036899482; Email:
vdamato@atriumehs.com). He
has been providing environ-
mental, safety and industrial
hygiene support to manufac-
turing, research and develop-
ment, utilities, telecommunica-
tions, aerospace, shipyard, facilities management
and defense industries for more than 20 years.
He has a B.S. in Environmental Health from Old
Dominion University, and a M.S. in Engineering
Management from The George Washington Uni-
versity. He is an American Board of Industrial
Hygiene Certified Industrial Hygienist (CIH) and
a Board of Certified Safety Professionals Certi-
fied Safety Professional (CSP). He is a member
of the American Industrial Hygiene Association
(AIHA) and has served as president of the AIHA
Potomac Section.
51
 Solids
Environmental
Processing
Manager
Managing Compliant
MSDSs and Labels
Compliant MSDS programs
depend on inventory management,
regulatory knowledge and consistent authoring
Jytte Syska and Tamie Webber
3E Company
C
hemical regulatory compli-
ance is growing in complexity
as product sourcing and for-
mulations proliferate and or-
ganizations expand into new markets
on a global scale. Staying current with
the requirements of international reg-
ulatory initiatives such as the Glob-
ally Harmonized System of Classifica-
tion and Labeling of Chemicals (GHS)
and the Registration, Evaluation and
Authorization of Chemicals (REACH)
adds to the challenge (CE, March 15,
2008, p. 3847, and CE, June 15, 2008,
p. 2630).
A key compliance requirement of
these initiatives is the proper man-
agement of Material Safety Data
Sheets (MSDSs) and labels. Both GHS
and REACH significantly impact the
corporate strategy for MSDS man-
agement, as the regulations require
affected companies to update these
documents to maintain compliance.
Comprehensive inventory manage-
ment, thorough regulatory reporting
and consistent MSDS and label au-
thoring are three critical components
of a REACH- and GHS-compliant
MSDS management program.
MSDS management
Accurate chemical inventory manage-
ment is the cornerstone of any MSDS
management program. Regulatory
compliance mandates that companies
obtain and maintain MSDSs for the
raw materials, as well as for the full
composition of mixtures. For REACH
compliance, this requirement helps
companies establish the category into
which individual substances fall: man-
ufactured within the European
Union (E.U.), imported into the
E.U., or purchased from a supplier
within the E.U. For REACH,
companies also must identify
each substances tonnage and
current classification.
Comprehensive regulatory re-
porting is also crucial to achieving
GHS compliance. In addition to
consistently tracking regula-
tions, companies must track
modifications made to ingre-
dients, products, and quanti-
ties and inventories.
While authoring accurate and
compliant MSDSs has always
been a mission-critical obliga-
tion, the advent of REACH
and GHS brings new chal-
lenges to this already highly
demanding task. For higher-
volume hazardous substances,
REACH requires companies to
attach an exposure scenario
to the MSDS. The exposure
scenario describes how the
chemical can be used in a safe
manner, with minimal risk, for the
intended use. The exposure sce-
nario can cover either each
individual substance in the
chemical product or describe
the mixture as a whole.
The first registration dead-
line under REACH passed on
December 1, 2010. Import or
manufacture of high-volume
chemical substances and of
CMR- (carcinogenic, muta-
genic and toxic-to-reproduc-
tion) classified substances is no
Chemical Engineering www.che.com february 2011
Figure 1.
The advent of GHS and R
EACH regulations brings new
challenges to the already demand-
ing and mission-critical task of
authoring compliant MSDSs
longer possible unless the im-
porter or manufacturer has reg-
istered under REACH. Many
of the substances that were
registered for the Decem-
ber 1, 2010 deadline have
exposure scenarios. The
registrant must attach the ap-
propriate exposure scenarios to
their MSDSs, and the company
using these substances must
then include the informa-
tion from the exposure sce-
narios when they prepare
and supply the MSDSs to their
customers. The European Chemi-
cal Agency (ECHA) has recom-
mended a four-section expo-
sure-scenario format, but
many different formats are
being used. Work is ongoing
to harmonize the information in
the exposure scenarios and to
establish a common XML for-
mat for data exchange.
As a step toward improved
compliance, companies should
obtain the necessary overview of
the current regulations and detailed
understanding of how they relate
to their chemicals and their
supply chain. Manufacturers,
users and distributors should
be aware not only of their own
obligations but also of those to
downstream customers and to em-
ployees. In addition, corporate
product-stewardship staff
must understand when these
regulations are implemented
by country in order to know
how to properly plan.
43
 Environmental Manager
GHS
GHS is the United Nations (UN) sys-
tem for harmonization of classifica-
tion and labeling of chemicals. GHS
has already been officially adopted by
certain countries in regions around
the globe. Countries can choose to
implement GHS in its entirety or
implement only certain parts of the
regulation, otherwise known as the
building block approach.
The true complexity of the stan-
dard is realized when reviewing its
implementation status in a sample of
countries around the world. For rep-
resentative comparison purposes, we
have provided a brief overview of the
status of GHS in the following areas:
the U.S., Japan and the E.U.
The U.S. On September 30, 2009, the
U.S. Occupational Safety and Health
Admin. (OSHA; Washington, D.C.;
www.osha.gov) released its proposal
to modify the current Hazard Commu-
nication Standard (HCS) to conform
with the GHS.
The proposed OSHA revisions in-
clude both philosophical and tactical
changes to hazard communications,
which have far-reaching implications
for MSDSs and the authoring, pub-
lishing, distribution and management
of labels. The proposal also includes
revised criteria for the classification
of hazardous chemicals, as well as
changes to definitions and terms used
in the standards, and new training
requirements for employees. When
the final rule is promul-
gated, companies will
face many challenges,
including re-evaluat-
ing how their sub-
stances and mixtures
are classified, re-issu-
ing MSDSs and labels,
Figure 2.
Service providers
can help companies
stay ahead of the
regulatory compli-
ance curve
44 Chemical Engineering www.che.com february 2011
and training staff as appropriate.
After having accepted public com-
ments until the end of 2009, OSHA
has estimated that a minimum of 18
months will pass from the notice of
proposed rulemaking before the final
rule is promulgated. Companies will
have three years from promulgation
to come into compliance with the final
rule and two years to implement train-
ing requirements on the final regula-
tion. Once the rule is final, companies
can begin implementation. However,
in accordance with a letter of interpre-
tation from OSHA on October 6, 2009,
companies can begin following the
E.U. GHS-label format for their OSHA
labels as long as the labels also comply
with the current HCS.
In addition, 26 U.S. states and terri-
tories have their own OSHA-approved
plans. The 26 are: Alaska, Arizona,
California, Connecticut, Hawaii, In-
diana, Iowa, Kentucky, Maryland,
Michigan, Minnesota, Nevada, New
Jersey, New York, New Mexico, North
Carolina, Oregon, Puerto Rico, South
Carolina, Tennessee, Utah, Vermont,
Virginia, Virgin Islands, Washington,
Wyoming. They will have six months
from promulgation to adopt compa-
rable provisions of the final standard.
In the meantime, each individual
state plan will remain in effect until it
adopts the required revisions.
Provided the legal process pro-
ceeds as planned, companies would
be allowed to issue MSDSs and labels
for the U.S. market according to the
changed rule, based on GHS classifi-
cation, no earlier than July 2011, and
all existing MSDSs and labels would
have to be updated in accordance with
the new requirements no later than
July 2014. In comparison, MSDSs for
the European market had to be in
line with the E.U. GHS implementa-
tion no later than December 2010 for
pure substances and by June 2015 for
chemical mixtures.
The label requirements will involve
changes in the printing process and
most likely will require many compa-
nies to invest in new label printers.
Labels on products that are shipped
outside the U.S. must have the chemi-
cal hazard pictograms with a red
frame, signal words and the required
hazard and precautionary statements.
It is still being determined whether or
not empty frames will be accepted on
the label and whether or not labels for
products sold and used exclusively in
the U.S. can have the frame printed
in black.
Japan. In Japan, the Industrial Safety
& Health Law (ISHL; December 2006)
currently details requirements for
MSDSs and labels. The ISHL provides
a list of substances that are subject to
MSDS and label requirements, and
also provides a classification results
list, which details the classification
of about 1,500 regulated chemicals.
The Japanese Industrial Standard
(JIS), which specifies the standards
used for industrial activities in Japan,
also has requirements for labels (JIS
Z 7251:2006) and MSDSs (JIS Z
7250:2005). Recently JIS Z 7252:2009
was published for the standard of GHS
classifications for health and environ-
mental hazards.
E.U. The European Union is currently
in a transitional period with regard to
GHS. The European Parliament final-
ized and issued the Regulation (EC No.
1272/2008 on classification, labeling and
packaging of substances and mixtures
the CLP Regulation) in late 2008. An
amendment bringing the CLP in line
with the third revision of the GHS was
drafted in 2010.
The CLP Regulation provides the
following transitional periods for clas-
sifying and labeling hazardous sub-
stances and mixtures:
Substances must be classified, la-
beled and packaged according to CLP
from December 1, 2010 forward
Mixtures can be classified, labeled
and packaged according to the Prep-
arations Directive until June 1, 2015
Supplemental labeling information,
in line with the old labeling systems
requirements, is still applicable
under the CLP
The transitional periods for re-label-
ing and re-packaging of substances
and mixtures placed in the supply
chain before December 1, 2010 are
postponed until December 1, 2012 and
June 1, 2017 respectively.
REACH
The E.U.s REACH regulation went
into effect on June 1, 2007 some-
times simplifying, but also complicat-
ing, the compliance work for compa-
nies manufacturing in, importing into
or exporting from European countries.
As a result of the new legislation, all
companies manufacturing, importing,
distributing or using chemical sub-
stances (on their own, in mixtures or
in articles) in Europe, are required to
closely examine their chemical inven-
tory for substances within the scope of
the regulation to ensure compliance.
Key components of REACH man-
date the following:
Registration of manufactured/im-
ported chemical substances
Increased information and commu-
nication throughout the entire sup-
ply chain
Evaluation of some registered sub-
stances
Authorization for use of substances
of very high concern
Restriction of the use of certain sub-
stances for specific applications
The scope of REACH is vast, and
there are several issues that could
impact a companys ability to meet
its obligations under REACH, includ-
ing changes to the content of MSDSs.
Under REACH, additional information
from manufacturers is required and
the information must be aligned with
that associated with the registration of
the substance. When an exposure sce-
nario has been required as part of the
REACH registration of a substance,
every safety data sheet for a product
containing this substance must have
the exposure scenario attached.
Choosing the right tools
As many companies realize, establish-
ing GHS and REACH compliance is
difficult. However, there are service
providers who can help companies stay
ahead of the curve with solutions for
identifying and managing the increas-
ingly complex and changing global
chemical-regulatory obligations asso-
ciated with GHS and REACH. These
vendors can help address the new
requirements and associated chemi-
cal regulatory information needed for
compliance management as they arise.
For example, there are several com-
mercially available content tools that
can be used to feed chemical regula-
tory data into corporate EH&S (en-
vironmental, health and safety) and
MSDS authoring systems. These in-
tegrated data tools provide efficient
change management and regular up-
dates as regulations change and new
ones are released. These tools can
help ensure correct, consistent GHS
classification and labeling according
to the U.N. purple book and also take
into consideration the national devia-
tions and other national regulations.
Document templates for producing
MSDS, labels and other hazard com-
munication documents and multilin-
gual phrase libraries can complete the
content suite for efficient production
of accurate documents.
Companies also may choose to employ
outsourced services for their GHS clas-
sification of substances and mixtures as
a separate service or as part of outsourc-
ing MSDS and label authoring.
In-house MSDS authoring staff may
want to adopt an authoring platform
that will help generate hazard com-
munication documents to meet GHS-
related, international-regulatory com-
pliance and business requirements.
Such a platform should provide full
support of hazard communication,
classification and labeling require-
ments to generate globally compliant
MSDS and label documents as well as
business- or user-definable documents,
such as technical data sheets, product
data sheets, hazard summaries and
product stewardship summaries. The
system should use algorithms to ac-
commodate the requirements outlined
in the GHS. This consists of the clas-
sification of substances and mixtures
according to their health, environmen-
tal and physical hazards and hazard-
ous communication requirements for
labeling and MSDS.
For inbound vendor MSDS man-
agement, users will want to search,
print, view and Email vendor and
raw material safety data sheets in a
company-specific database via a Web-
browser interface. As GHS require-
References
1. The U.S. Chemical Safety and Hazard Investi-
gation Board (CSB), webpage, (www.csb.gov).
2. Center for Chemical Process Safety. Guide-
lines for Hazard Evaluation Procedures
With Worked Examples (2nd Ed.), AIChE,
1992.
3. American Equipment, Inc., Process Safety
Management Guidelines for Compliance,
OSHA document 3133, 1994 Appendix., copy-
right 1999 2005; www.exforklifts.com.
Chemical Engineering www.che.com february 2011
ments increase the burden of vendor
MSDS management, a robust MSDS
management system can help reduce
the time and resources necessary for
effective compliance management.
Compliance help is available
Creating, analyzing and managing
globally compliant MSDSs, product
label content and hazardous materials
transportation documents that reflect
both GHS and REACH requirements
can be extremely challenging. Strong
processes, applications, systems and
service providers are needed to sup-
port the various aspects of compliance
management. Companies that are im-
pacted by GHS or REACH, or both, can
seek assistance from providers who
are well versed in data and other con-
tent and information as they relate to
global EHS regulations and who thor-
oughly understand the global regula-
tory environment. These providers will
be able to assist in implementing GHS
and REACH compliance activities into
the organization, and can help facili-
tate compliance with these increas-
ingly complex and changing global,
chemical regulatory obligations.
Edited by Scott Jenkins
Authors
Jytte Syska is managing di-
rector of 3E Company Europe
ApS (Lyngbyvej 20, 3, DK
2100, Copenhagen, Denmark.
Phone: +1-760-602-8700; E-
mail: jsyska@3ecompany.com)
and president of Ariel opera-
tions. Syska has more than 20
years of experience as an envi-
ronmental, Health and Safety
(EHS) regulatory expert.
Prior to her present duties,
she spent several years as 3Es vice president of
international operations. Syska has also held the
position of director of occupational health and
toxicology for the prestigious Danish Toxicology
Center (DTC). Syska holds an M.S. degree in
civil engineering from the Technical University
of Denmark. She is fluent in Danish and English,
and reads and understands German, French,
Norwegian and Swedish.
Tamie Webber is director of
product management at 3E
Company and is responsible
for product management ini-
tiatives, product planning
and strategies and product
development for 3Es product
line, including MSDS dis-
tribution and management,
emergency response, regula-
tory reporting, training and
waste services. She joined 3E
after several years of marketing and product
management experience in the chemical distri-
bution and high-tech industries. She has worked
for Great Western Chemical and Intel Corp. She
holds a B.S. degree in finance and marketing
from the University of Oregon.
45
 Solids
Environmental
Processing
Manager
Effective
Plant Safety
Management
Three critical junctures
and seven critical steps
for plant safety are outlined
Nasim Hassan
HSN Associates of Delaware
A
lthough plant safety should
be a top priority every day,
there are certain times when
extra vigilance is needed to
make sure that safety practices are
being maintained. This article out-
lines three periods deemed to be criti-
cal for plant safety and specific items
to pay attention to in these periods.
Three critical periods
The following three phases in a pro-
duction plants life are potentially crit-
ical in regards to safety practices:
Transition. The first period deemed
to be critical is the transition phase
when a company acquires a production
plant, petroleum refinery or any other
manufacturing facility. During this pe-
riod many experienced people leave the
company while new management is
learning the details of plant operations.
Restructuring. The second period is
a phase of restructuring, when due to
slow economic conditions the demand
for a product has slowed down. The
company needs to save operating and
maintenance costs. During this period
some sections or the whole plant may
be shut down.
Routine. A third period that comes to
mind is when the plant operation be-
comes routine and inexperienced peo-
ple are allowed to take over and run the
facility. The experienced staff relaxes
and becomes less vigilant in reviewing
plant safety on a regular basis.
50 Chemical Engineering www.che.com January 2011
Seven essential elements
During the periods mentioned above,
it is particularly important for manag-
ers to include the following elements
in their management of plant safety.
1. Review of technology. Whenever
a plant is acquired, a thorough review
of the acquired technologies must be
made. Typically, companies feel con-
fident in acquiring facilities when
they are operating similar plants or
manufacturing similar products. For
example, a petroleum refining com-
pany taking over another refinery
has good knowledge of refinery opera-
tions. However there are several dif-
ferent process technologies available
for the same operations or reactions.
For example, there are many different
alkylation processes in operation to
produce motor and aviation fuels.
Similarly in petrochemical produc-
tion, such as for vinyl chloride, there
are many routes for the same reactions
and intermediate products. Some tech-
nologies employ fixed beds for oxychlo-
rination while others use fluidized bed
reactors. A review of process technolo-
gies will reveal the differences in the
technology and operating procedures.
Once the differences have been identi-
fied, then all plant operators must be
trained in all new technologies.
2. Plant operating procedures. Ex-
perience shows that the same plant
can be started in a number of ways.
All companies develop operating pro-
cedures to start up and shut down a
plant. New procedures are certainly
needed when the unit operations are
different in plants where a different
technology is employed. I have seen
a wide variety of processes for eth-
ylene, polyesters, vinyl chloride and
polyvinyl chloride (PVC). There are
similar examples of various processes
in petroleum refining. The operating
procedures also differ depending on
the complexity of the plant. Operat-
ing procedures should be updated on
a regular basis.
3. Review of safety interlock and
shutdown systems and procedures.
The safety interlocks and shutdown
systems are developed after compre-
hensive hazards and operability stud-
ies. Many are introduced after actual
plant experience. These interlock sys-
tems must be reviewed and upgraded
if required.
All major companies control and
operate their plants based on their
in-house development. A thorough
review will certainly improve the ex-
isting systems due to the combination
of expertise from merged or acquired
companies. I have seen many times
that safety interlocks are bypassed
because new plant operators did not
understand the rationale behind the
safety interlocks.
4. Review of corrosion control sys-
tems. Control of corrosion in plant
equipment is intricately linked with
 Revise #1 - 2/4/10
Plant Safety Recommendations
W
hile plant safety should always be given top priority, safety audits are particularly
recommended during the transition phases of mergers and acquisitions. Safety
audits also become important when experienced engineering staff is reduced
due to restructuring or retirement. In fact, experienced engineers should conduct training
programs for plant operators and engineers before they retire. This small investment will
help reduce potential operating problems and industrial accidents.
plant safety. There are many kinds of
corrosion protection systems employed
in chemical and petrochemical indus-
tries. These systems are developed
after years of plant experience. This
is an area that is often ignored when
chemical plants or petroleum refiner-
ies are sold to new owners.
I have seen plants with near-perfect
safety records lose that record within
months of acquisition by new own-
ers. Looking deeply into the reasons, I
found that plant engineers and opera-
tors did not understand the function of
various corrosion protection systems.
In a vinyl chloride plant, for example,
the overhead condenser for an ethyl-
ene dichloride distillation column was
protected by the circulation of dilute
caustic solution. After a new owner
took over the facility, the circulation
of caustic was stopped because the in-
experienced engineers did not under-
stand its role in preventing corrosion
in the condenser and overhead piping.
5. Training of plant engineers and
operators. This area also gets side-
lined, particularly in old facilities. Ex-
perienced engineers take it for granted
that young engineers will learn with
hands-on experience. They can cer-
tainly learn the plant operation, but
may not understand why the system
was designed in a certain way.
There should be an ongoing training
program for new engineers and even
experienced engineers if new tech-
nology is introduced in a plant. The
review of various reactions and side
reactions, optimum operating condi-
tions, limits and safety precautions
should be essential for every person.
In older facilities, there is a wealth
of information available but only ex-
perienced people know how to access
it. Training materials should be easily
accessible to all new operators.
6. Review of past incidents and in-
vestigative reports. My experience
6:00 PM
shows that the same problems occur
over and over again in the same fa-
cility. People are alert for a few years
after an accident. Then safety reports
get filed away and forgotten. A new
company taking over old facilities is
busy with big ticket items. Plant safety
records and reports are reviewed at a
much later stage.
I recommend a thorough review of a
plants past safety history, followed by
a plant safety audit that looks at the
particular areas where past accidents
have occurred.
7. Implementation of EPA (U.S.
Environmental Protection Agency)
and OSHA (U.S. Occupational
Safety and Health Admin.) Stan-
dards. This should be a priority in
any plant, and needs to be given top
priority during the plant shutdown
period. I have seen vent scrubber sys-
tems shut down because a plant is
not routinely sending any hazardous
vapors to the vent scrubber. However,
there may be some storage tanks or
vessels containing hazardous materi-
als. These materials are vented to the
scrubbers in overpressure conditions.
If the vent scrubbers are shut down,
then severe environmental and health
related incidents can occur. 
Edited by Dorothy Lozowski
Author
Nasim Hassan is a con-
sultant (HSN Associates
of Delaware; 7 Signal Hill
Drive, Hockessin, DE, 19707;
Phone: 302-239-9272; Email:
nhassa@yahoo.com; Website:
www.hsnrcs.com) in chemical
technology licensing, plant
safety, process analysis, pro-
cess simulation and process
design. He has over 35 years
experience in the design and
operation of chemical and petrochemical plants.
Hassan has worked for BE&K Engineering (a
KBR subsidiary), ICI Americas and Stauffer
Chemical Co. as senior staff engineer and proj-
ect manager. He has designed and started up a
number of chemical and petrochemical plants
for licensees in Africa, Asia, Europe and the U.S.
Hassan has an M.S.Ch.E. from the Institute of
Chemical Engineering, Punjab University, Paki-
stan and is a member of AIChE.
Chemical Engineering www.che.com January 2011
Centrifuge & Drying
Technologies
Inverting Filter Centrifuge
Cutting edge centrifuge technology for
filtration, washing and drying of solid/liquid
suspensions
t Widest range of applications - hardest to
easiest filtering products can be handled
t No residual heel for exact repeatable
batches and no loss of product
t PAC technology allows drying of the
product inside of the centrifuge
t Thin cake filtration operation allows for
improved quality and production rates
t Full containment eliminates operator
exposure
t Effective automated CIP
Kilo-Lab Conical
Vacuum Dryer-Mixer
Advanced technology
for real Kilo size
drying research
and development
t Utilizes interchangeable agitator systems
either orbiting arm screw or central shaft
t Flexible small scale volume of 150ml to
1500ml
t Plastic view through vessel available
t Designed for real laboratory requirements
of size, with full instrument & data
recording
t Direct scale up to production or pilot size
units
Horizontal & Vertical
Centrifuges
t Size ranges from 200mm to 1800mm
t Wide range of standard & custom designs
t Laboratory size equipment
Lab Testing Available
Rental & Lease Machines Available
www.heinkelusa.com
Tel: 856-467-3399
Circle 11 on p. 58 or go to adlinks.che.com/35061-11
51
 Feature Report
Engineering Practice
Remote Thermal Sensing
By making it easy to
detect heat anomalies,
thermal cameras and
infrared thermometers
support preventive and
predictive maintenance
Roger Mavrides
General Tools & Instruments
Suzanne Shelley
Precision Prose, Inc.
R
ising temperatures and rapid
or excessive heat buildup are
useful markers for determin-
ing the operational health
of many types of machinery and
components that are used in a wide
array of industrial and manufactur-
ing settings. The types of mechani-
cal and electrical systems for which
temperature increases often signal
problems include (but are not lim-
ited to): rotating machinery, such as
motors, turbines, compressors, and
their bearings, couplings and gear-
boxes; other types of process equip-
ment, such as pumps, valves, heat
exchangers, steam traps, heaters,
conveyors belts, rollers, furnaces
and more; steam and electrical
heat-tracing systems; insulation on
pipes and vessels; refractory lining
systems for high-temperature sys-
tems and much more.
Because equipment malfunctions
and abnormal or fault conditions in
mechanical and electrical systems
are often forewarned by a rise in
temperature, the ability to gather
and analyze temperature data in
realtime can often help operators to
both pinpoint existing performance-
related issues and identify the
onset of incipient problems. These
problems can be caused by an array
of issues, including wear, imbal-
48 Chemical Engineering www.che.com January 2014
Figure 1. Rising temperature is often
an indicator of operational problems in
many types of machinery. The ability to
gather and analyze temperature data in
realtime using non-contact options can
help operators to pinpoint issues and
act accordingly
ance, misalignments, insufficient
cleaning or lubrication, friction and
electrical problems.
The traditional approach to tem-
perature monitoring in industrial
settings relies on direct-contact
temperature sensors, such as ther-
mocouples and resistance tempera-
ture detectors (RTDs). While these
devices are certainly proven and ac-
curate, they are not appropriate for
use with certain types of equipment
components or in some types of in-
dustrial settings.
By contrast, remote or non-con-
tact temperature-measurement
devices, such as infrared thermal-
imaging cameras and infrared
thermometers (IRTs) allow use-
ful temperature data to be easily
gathered from remote locations
and thus offer a useful alternative
to direct-contact temperature sen-
sors. The ability to safely carry out
temperature sensing from a dis-
tance, using either a thermal cam-
era or an IRT, is particularly use-
ful for machinery components and
systems that may be hard to reach,
inaccessible, or potentially hazard-
Figure 2. Problems such as over-
heated bearings can be diagnosed with
a thermal imaging camera, which pro-
vides an alternative to direct-contact
temperature sensors, especially for
components that may be hard to reach,
inaccessible or potentially hazardous
ous. In this way, these temperature-
monitoring devices help to enable
realtime temperature measure-
ment while ensuring worker safety
(Figures 1 and 2).
Technology options
The two types of remote thermal-
sensing options infrared ther-
mal-imaging cameras and IRTs
are widely used for tempera-
ture assessment in industrial and
manufacturing facilities. In addi-
tion, two different types of IRTs are
available conventional IRTs, and
so-called scanning IRTs. Used sepa-
rately or together, these devices can
help users across numerous indus-
try sectors to quickly and easily
assess the thermal condition of ma-
chinery, process systems, pipelines
and more.
Conventional IRTs are best
suited for applications that re-
quire accurate spot-temperature
readings, while scanning IRTs and
thermal cameras are useful for ap-
plications for which knowing the
absolute temperature of a surface
is less important than knowing the

Figure 3. Thermal-imaging cameras
measure the surface temperature of the
objects or areas being analyzed in terms
of the amount of infrared (IR) energy
that is emitted, transmitted or reflected
by the object. The data can then be ren-
dered as still or video images to help
operators interpret the result
temperature of a particular surface
relative to other surfaces around it.
The primary advantage of a ther-
mal camera is its ability to display
the thermal spectrum of an entire
area, as seen in Figure 3.
Infrared thermal-imaging cam-
eras. Portable thermal-imaging
cameras are easy to use, and typically
come with a pistol-grip design, as
seen in Figure 4. They use infrared-
imaging techniques to measure the
surface temperatures of the objects
or areas being analyzed and can ren-
der the data in the form of two-di-
mensional images or videos images
to illustrate the data. Specifically,
these specialized cameras measure
surface temperature in terms of the
amount of infrared (IR) energy that
is emitted, transmitted and reflected
by the object or area being analyzed.
The temperature data are displayed
as an IR heat spectrum, using a
range of colors that are correlated to
specific temperature ranges. Todays
thermal cameras have great sensi-
tivity, and provide measurement ac-
curacies of up to 2F.
Thermal cameras produce a
thermal signature or thermogram,
which is a two-dimensional vi-
sual display that depicts the rela-
tive temperature variations across
the objects surface. These images
allow operators to quickly pinpoint
problem areas, since temperature
excursions, such as areas of heat
loss or heat gain relative to the
surroundings, are easily displayed
through color variations in the
rendered image.
Such visual displays of relative
temperature variations across the
surface of the objects gives opera-
tors and technicians unprecedented
insight into the health of equip-
ment and systems, and help to ad-
dress emerging problems efficiently.
Thermograms are especially useful
when thermal imaging is used as
part of regular inspections, because
they allow engineers to quickly rec-
ognize changes that may signal an
emerging problem. Thermal images
that are captured and analyzed over
time for the same component (for
instance, a given motor or pump)
can help users to identify locations
of incipient malfunction or progres-
sive wear or deterioration.
Creating such a record of heat
buildup due to deteriorating con-
ditions allows operators or tech-
nicians to dispatch the most ap-
propriate intervention in a timely
manner. These interventions in-
clude detailed inspection, trouble-
shooting and diagnostic efforts, and
strategic maintenance and repair
activities. Because thermal imag-
ing is carried out at a distance, it
enables the capture of thermal data
from components that are in remote
or hazardous areas, thereby ensur-
ing worker safety.
Infrared thermometers. IRTs are
portable, non-contact devices
again, typically with a user-friendly,
pistol-grip design. IRTs use a spe-
cial lens to focus the thermal radia-
tion that is being emitted by the ob-
ject (in the form of IR energy) onto
an IR sensor. The embedded soft-
ware correlates those IR readings
to the temperature of the object
using information about the mate-
rials emissivity. Like thermal im-
Chemical Engineering www.che.com January 2014
Figure 4. De-
signed with an
easy-to-use pistol
grip, thermal-im-
aging cameras are
used to diagnose
hot spots in ma-
chinery systems
aging cameras, IRTs offer an ideal
way for operators to determine the
temperature of hot or cold surfaces
remotely, which is especially useful
for inaccessible or hard-to-reach ob-
jects or areas.
As noted earlier, two types of
IRTs are available conventional
spot IRTs and so-called scanning
IRTs. Scanning IRTs allow users
to scan an entire area or system
and quickly identify those sections
where there is a significant tempera-
ture differential between the actual
temperature of that section and a
pre-set temperature setpoint value
that the user has programmed into
the device.
Using a conventional IRT with
an appropriate distance-to-spot
(D:S) ratio one that allows the
measurement area to be entirely
focused within the object being
measured plant personnel can
determine the temperature of an
object at a single spot. A built-in
laser-beam sighting source helps
the user to focus the device on
the target precisely, to ensure
measurement accuracy.
While conventional IRTs are use-
ful for remotely gathering point-
source data about the absolute tem-
perature of a given spot, scanning
IRTs are useful for applications
where it is not necessarily impor-
tant to determine the absolute tem-
perature of a surface, but it is useful
to determine the relative tempera-
ture of a surface or area compared
to its surroundings.
Todays scanning IRTs are not
only very affordable (typically
under a hundred dollars, compared
to more than a thousand for thermal
cameras), they are also extremely
easy to use. With a point-and-shoot
design, the user first establishes a
49
 Engineering Practice
baseline temperature that is ap-
propriate for the application, after
which an acceptable bandwidth or
tolerance range is set for exam-
ple, 10 degrees, although tighter
tolerances of 1 or 5 degrees
are possible.
When the trigger on a pistol-grip
scanning IRT is squeezed and the
device is moved slowly across the
target area, the device uses a com-
bination of sound (in the form of
slower versus faster beeping sounds)
and colored lights (for example, red
for above range, green for within
range, and blue for below range)
to alert the user to any location
where the temperature falls outside
of the user-specified threshold val-
ues that define the setpoint range.
The IRT will also provide an abso-
lute temperature reading for that
particular spot.
While scanning IRTs do not pro-
duce a thermal image, they do pro-
vide a quick, easy, and relatively in-
expensive way for facility personnel
to assess specific mechanical assets
and identify those problem areas
that may require closer inspection.
A valuable investment
While individual thermal cameras
typically cost more than a thousand
dollars, they can be a strategic in-
vestment and will easily pay for
themselves over a short amount
of time if their use prevents a
catastrophic failure.
As noted, the use of remote, IR-
based thermal sensing to inspect,
troubleshoot, diagnose and rectify
problems with specific equipment
in realtime can help facility opera-
tors to improve the efficiency and
effectiveness of both component-
specific and plant-wide operation
and maintenance (Figure 5). Such
improvements provide a number of
opportunities for long-term savings
and payback. For instance:
The costs associated with un-
planned downtime in industrial
facilities, such as manufactur-
ing plants and chemical process
facilities can be greatly reduced
when thermal imaging is used to
improve preventive and predic-
tive maintenance tasks
50 Chemical Engineering www.che.com January 2014
Proactive troubleshooting
and diagnostic interven-
tion can help to decrease
the likelihood of expensive
or catastrophic equipment
failures especially im-
portant when considering
mission-critical assets
The ability to plan and ex-
ecute more-strategic re-
pairs helps to cut material
and labor costs and extend
Figure 5. When operators are able to monitor
equipment life, thereby the thermal profile of equipment and systems
helping to reduce both oper- using a remote monitoring technique, they can
ating and capital budgets plan for the most appropriate and timely inter-
Increased uptime allows vention to address a deteriorating condition
the facility to maximize
its throughput capacity, product nents ideal state under normal
yields and profitability. The abil- working conditions. Going forward,
ity to carry out strategic main- any departure from normal temper-
tenance activities helps to not atures that appear in the thermal
only save money, but improve images produced for the unit would
plant and personnel safety and signify trouble spots that require
environmental performance closer inspection.
Proper training and certification
Best practices are extremely beneficial when using
There are several ways to make IRTs and thermal-imaging cam-
best use of thermal-imaging data. eras. While these devices are typi-
Trending opportunities can be used cally simple to operate and provide
to the engineers advantage. For data and images that are easy to
instance, the thermal signature interpret, both rely on sophisticated
from a given component, such as a technology. As such, to ensure the
particular pump that may be sus- most accurate results, users should
pected of having a problem, can be gain a good working knowledge of
compared to the thermal signature the capabilities and limitations of
of similar pumps in the facility. This these tools through proper training
will help to evaluate its condition and certification. Without proper
relative to other equipment with training, the accuracy of the result-
comparable operation. ing thermal data and images could
In addition, the thermal images be compromised. Following vendor-
generated for a given mechani- recommended operating guidelines
cal asset (say, a particular motor), and proven industry best practices
can be strategically captured and is a must for data confidence.
cataloged over time in specific in- Today, a variety of third-party
tervals. This record can provide groups offer training and certifica-
timely indications of deteriorating tion in the proper use of thermal-
conditions and help the operator to imaging cameras, including the
make reasonable predictions about American Society for Nondestruc-
the rate of future deterioration, tive Testing (Columbus, Ohio; www.
so that the required action be asnt.org), the Academy of Infrared
it maintenance, repair or replace- Training (Bellingham, Wash.; www.
ment can be carried out, at the infraredtraining.net), the Infraspec-
appropriate time, in the most cost- tion Institute (Burlington, N.J.;
effective manner. www.infraspection.com), The Snell
When practical, it is a good idea Group (Barre, Vt.; www.thesnell-
to carry out baseline thermal-imag- group.com) and others.
ing inspections on new components, Choosing the right thermal imag-
to establish baseline or reference ing camera for the environment has
images that represent the compo- an effect on the equipments ability
to best collect accurate data. Ther-
mal imaging cameras are widely
used to carry out energy-efficiency
studies and audits in residential,
commercial and business settings,
by helping to identify areas through
which heated or cooled air is escap-
ing from a building. Industrial fa-
cilities can also use these devices
to carry out energy audits to iden-
tify further opportunities to reduce
operating costs.
When this is done, users should
note that regional climate varia-
tions can impact what time of year
the use of a thermal imaging cam-
era will be most effective. For in-
stance, in northern climates, the
use of thermal imaging to carry out
energy audits at a facility tend to
be most accurate when carried out
during winter months (when the
temperature differential between
indoor and outdoor temperatures
is the greatest) and temperature
differentials resulting from the un-
wanted ingress or escape of heated
or cooled air will be most easily
identified by the thermal camera
or scanning IRT assessment. By
contrast, in southern climates, the
summer months tend to guaran-
tee the biggest temperature dif-
ferential between ambient outdoor
temperatures and air-conditioned
settings.
This potential seasonal limita-
tion can be somewhat overcome
by selecting a higher-resolution
thermal camera. Today, a variety of
thermal cameras with a range of
prices and image-resolution capa-
bilities are available. Relatively
low-end models have sensors with
a resolution of 6060 pixels. Mid-
range units have a resolution on the
order of 160120 pixels, and high-
end thermal cameras offer a resolu-
tion of 360280 pixels.
While lower-resolution thermal
cameras may be available at a
lower cost, the desire for cost sav-
ings alone may cost the buyer in
the long run by limiting the num-
ber of months over which the cam-
era can be reliably utilized to carry
out energy audits and other types of
thermal inspections. In general, the
higher the resolution of the thermal
camera, the more reliably it can
depict a thermal difference when
carrying out a thermal assessment
even during those times of the
year when the temperature differ-
ential between indoor and outdoor
temperatures is relatively narrow.
By investing in a higher-resolu-
tion camera, users will be assured
of greater sensitivity and easier
thermal assessments no matter
what climate or time of year. This
helps to ensure more accurate re-
sults and faster payback for the
camera itself.
Another important factor to con-

The Chemical Engineering bookstore offers a

variety of industry topics you will come to rely on.

Environmental Management: Air-Pollution Control

Environmental Management: Wastewater and Groundwater Treatment
Fluid Handling
Gas-Solid and Liquid-Solid Separation
Liquid-Liquid and Gas-Liquid Separation
Managing Bulk Solids
17792

For a complete list of products, visit the Chemical

Engineering bookstore now. http://store.che.com/product/book

Chemical Engineering www.che.com January 2014 51

 Engineering Practice
sider when using thermal cameras
and IRTs is the relative emissivity
values of the materials being sur-
veyed. Emissivity is a measure of
an objects ability to absorb or re-
flect radiation in the infrared range
of the electromagnetic spectrum.
As a simplifying assumption, many
thermal cameras have fixed emis-
sivity values that are associated
with certain commonly encountered
materials programmed into the
control software.
However, todays more-sophis-
ticated thermal-imaging cameras
allow the user to make adjustments
to the emissivity settings, to more
accurately characterize the actual
emissivity values of the materi-
als being analyzed. Such values
can often be found in published
books and reference articles. This
flexibility can help users to im-
prove the accuracy of the resulting
thermal images.
Certain surfaces, such as highly
reflective metals, have very low
emissivity values and cannot be
measured accurately using IR-
based temperature-measurement
techniques. To overcome this prob-
lem and improve the utility and ac-
curacy of IR-based thermal cameras
and thermometers, industrial oper-
ators often paint the target surfaces
or cover them with electrical tape.
This raises the emissivity values of
the components and improves the
accuracy of the IR-based thermal
imaging techniques.
When it comes to evaluating dif-
ferent IRT models, an important
concept to consider is the distance-
to-spot (D:S) ratio (Figure 6). This
characteristic of the device provides
a measure of the optical resolution
that a particular unit can provide.
Every IRT model has a stated D:S
ratio, which determines the dis-
tance for which the device will pro-
vide the most accurate temperature
reading as well as the diameter
of the imaging area. For instance,
most standard IRTs have a D:S
ratio of 8:1. This indicates that an
IRT 8 in. away from the object can
accurately measure the tempera-
ture at a spot that is 1 in. in diam-
eter. Similarly, an IRT that is 48 in.
52 Chemical Engineering www.che.com January 2014
areas. D:S ratio is an important feature to
away from a target will measure the
temperature within a circle that is
6 in. in diameter.
Units with a D:S ratio of up to
100:1 are also available. In general,
the higher the D:S ratio of the de-
vice, the smaller the zone for tem-
perature capture, because less of
the surrounding area is involved in
the measurement. IRTs with higher
D:S ratios also provide for more ac-
curate readings to be gathered from
greater distances.
Closing thoughts
Remote, non-destructive, ther-
mal-sensing techniques, using IR
thermal cameras and scanning or
conventional IRTs, provide useful
alternatives to direct-contact tem-
perature devices based on RTDs and
thermocouples. They provide useful
information on temperature excur-
sions that are often the precursor
to operational problems, allowing
users to plan the most strategic pre-
dictive and preventive maintenance
activities and to carry out the most
Authors
Roger Mavrides was for-
merly vice president of en-
gineering and product devel-
opment for General Tools &
Instruments (80 White St.,
New York, N.Y. 10013; Phone:
1-800-697-8665 x222; Email:
gentools@generaltools.com).
He holds a Certificate of Elec-
tronic Maintenance (CEM)
from Wentworth Institute of
Technology, a B.S. in electri-
cal engineering technology from Northeastern
University and an M.B.A. from Anna Maria
College. Before joining General Tools, Mavrides
was engineering and product manager, test
and measurement, for FLIR Systems (Nashua,
N.H.), sales and product manager for Nidec/
Power General (Canton, Mass.), and senior de-
sign engineer and project manager for Vishay/
BLH Electronics (Norwood, Mass.). Mavrides
holds three patents, is a Level 1 Thermographer,
and was a team leader during the development
of FLIRs MeterLink communication protocol.
He also developed a wireless alternating-cur-
rent circuit identifier that won a Hong Kong
Electronic Industries Association (HKEIA) In-
novation and Technology Grant Award at the
2009 HK Electronics Fair, and developed an
electrically safe video borescope that won the
Bronze HKEIA Innovation & Technology Grand
Award at the 2011 HK Electronics Fair.
Figure 6. Infrared thermometers let
users determine the temperature of a
specific point on hot or cold surfaces
remotely, which is especially useful for
inaccessible or hard-to-reach objects or
consider when choosing an IRT
cost-effective repairs and upgrades.
Such efforts can help to optimize
the productivity and reliability of
the mechanical assets, maximize
the uptime of the facility, and mini-
mize downtime-related losses and
expenses and reduce the risk of
catastrophic equipment failures.
And, remote IR-based temperature
monitoring lets facility personnel
carry out such surveillance with-
out shutting down the machines,
interrupting the process or put-
ting themselves in harms way. This
maintains equipment reliability,
the facilitys desired productivity
levels, as well as personnel safety,
and thereby helps to protect the
facilitys bottom line.
Edited by Mary Page Bailey
Acknowledgements
The authors would like to thank
John Javetski, Adrian Gomez,
Kevin Basso and Peter Harper of
General Tools & Instruments for
their assistance during the develop-
ment of this article.
Suzanne Shelley is the prin-
cipal/owner of Precision Prose,
Inc. (65 West 96th St., Suite
21F, New York, N.Y. 10025;
Phone: 917-975-2778; Email:
SuzanneAShelley@yahoo.com).
In that capacity, she provides
freelance technical writing,
ghostwriting and editing ser-
vices (specializing in science,
engineering, technology and
business) to magazines and
corporate clients. Prior to launching her consultancy
in 2005, Shelley spent 17 years as a full-time editor
at Chemical Engineering magazine, serving as the
magazines managing editor for her last 5 years on
staff. As a freelance writer and editor, Shelley serves
as a regular freelance contributing editor at Chemi-
cal Engineering and Pharmaceutical Commerce
magazines, and as a periodic freelance contribut-
ing editor at Chemical Engineering Progress (CEP;
AIChE). From 20052009, she also served as a regu-
lar contributing editor to Turbomachinery Interna-
tional magazine. Shelley also provides freelance
writing, ghostwriting and technical editing services
to a wide variety of operating and service compa-
nies, consultancies, advertising agencies and trade
associations throughout the global chemical process
industries. She holds a B.S. in geology from Colgate
University (Hamilton, N.Y.) and a M.S. in Geology
from the University of South Carolina (Columbia,
S.C.), and worked as a deepwater exploration geolo-
gist in the Gulf of Mexico for Amoco Production Co.
(New Orleans) in the late 1980s.
 Solids
Environmental
Processing
Manager

Finding the Right Gloves

To Fit the Application
There is a wide range of gloves available for hand protection
on the job. Matching gloves to their chemical-resistance
properties is one criterion for selection
Nelson Schlatter
Ansell

W
orkers in the chemical pro-
cess industries (CPI) would
certainly benefit from the
perfect glove that would
be thick enough for thermal insula-
tion, thin enough to promote dexterity
and tough enough to protect workers
from cuts and abrasion. The glove
would also protect against all known
chemicals and be offered at a minimal
price. This perfect glove, however, does FIGURE 1.
not exist. Processors and handlers, Nitrile gloves
therefore, must provide workers with are low cost and have
the most suitable, available gloves for excellent physical
properties, providing
the application, considering individual protection against
circumstances and how the gloves will oils and grease,
be used. Considerations to help choose xylene, perchloro-
proper gloves for a given application, ethylene and
aliphatic
with particular focus on chemical re- solutions
sistance, are presented here.

Factors to Consider
A person who selects gloves needs to
know much more than just the name
of the chemical to be handled. If an
employee, for example, is working
with nitric acid, then PVC (polyvi-
nyl chloride) and neoprene are the
preferred glove materials. Several
questions, however, must still be an-
swered. Is the worker cleaning up
spills? If so, the person will require
highly chemical-resistant gloves with
good storage stability because the
gloves may be kept in a spill-control
cabinet for several years until they
are urgently needed.
Is the person handling sealed bot-
tles, which will require minimal pro-
tection unless the bottle breaks? Could
a spill be caused by breaking a bottle?
If so, the worker will need protection
from cuts as well as nitric acid. What
52 Chemical Engineering www.che.com July 2008
is the probable length of exposure?
The longer workers are exposed to a
certain chemical, the greater the level
of protection required.
The chemical concentration must
also be considered. Is the compound
present only as a trace impurity? Is
the employee working with a pure
chemical or a dilute solution? If the
chemical is diluted, what solvent was
used for dilution?
Companies sometimes over-specify
gloves, which can lead to unecessary
expense. Once for example, a request
from a chemical company indicated
the need for gloves that would resist
high concentrations of toluene. When
questioned about the application, it
was revealed that workers would be
exposed to groundwater with a toluene
concentration of less than 100 parts
per million. While this is indeed a high
concentration for toluene in ground-
water, in this case, gloves that protect
against water would be more appropri-
ate and cost-effective than gloves that
protect against pure toluene.
Glove types
Gloves are generally categorized ac-
cording to the materials used and
whether they are supported or un-
supported (see definitions below). The
following glove types are available in
a range of lengths, thicknesses and
finishes, while some styles are offered
with special modifications.
Unsupported vs. supported
Unsupported gloves are made of
materials such as natural or syn-
thetic latex, nitrile and neoprene and
 Chemical-Resistance Information for Various Glove Materials
Chemical Laminate Nitrile Neoprene Polyvinyl Polyvinyl Natural Butyl Viton
Alcohol Chloride Latex
PVA PVC
Acetone >480 NR E P NR E >480 P
Ammonium hydroxide E >480 E NR E E >480 >480
Butyl glycol ether >480 E E E P E >480 >480
Ethyl acetate >480 NR F G NR G E NR
Ethylene glycol >480 E E F E E >480 >480
Gasoline (hi-test) E E NR G P NR F >480
Hydrochloric acid,
>480 E >480 NR E E >480 >480
concentrated
Hydrofluoric acid, 48% >480 E >480 NR G E >480 180
Methanol >480 E E NR G E >480 F
Methylene chloride E NR NR G NR NR G E
N-Methyl-2-Pyrrolidone >480 NR NR NR NR E >480 NR
Mineral spirits, rule 66 >480 E E E F NR 60 >480
Perchloroethylene >480 E NR E NR NR P E
Sulfuric acid, concentrated >480 NR F NR G NR >480 >480
Xylene >480 G NR E NR NR P E
NR=not recommended; P=poor; F=fair; G=good; E=excellent; >480 Permeation is also color coded: red indicates breakthrough times of only a
indicates breakthrough times that are greater than 480 minutes (8 few minutes; yellow indicates breakthrough times of up to two hours; and
h) and correlates well with extremely good degradation resistance blue indicates longer breakthrough times

achieve their glove shape by dipping
hand forms directly into the glove
compound without a supporting liner
or fabric. This type of glove generally
provides the dexterity and tactile
sensitivity required for many chemi-
cal applications.
Unsupported gloves offer a broad
spectrum of chemical resistance,
based on the material used. Nitrile
gloves, for instance, are excellent for
many chemical processing, oil refin-
ing, food processing and petrochemi-
cal applications. Gloves made with a
neoprene and natural-rubber-latex
blend provide the protection workers
need in food processing facilities and
certain chemical- and pesticide-manu-
facturing plants.
Supported gloves are made by dip-
ping a hand form wearing a knit-
ted- or woven-cloth liner into a glove
compound such as nitrile. The liner
supports the compound and adds
strength to the glove. Some supported
styles have continuous coatings to en-
sure protection from chemicals. Cotton
and polyester may be used in various
combinations as a yarn for woven or
knitted products, coated with vari-
ous compounds including natural
latex, neoprene and PVC to protect
against petrochemicals, oils, acids, al-
cohols and solvents.
Supported gloves with non-con-
tinuous coatings are better suited
for applications that require comfort
and grip. Many supported gloves also
offer cut, snag, puncture and abra-
sion resistance.
Fabric gloves
This category includes general pur-
pose gloves made with polyester,
nylon and cotton; cut-resistant gloves
constructed with Kevlar, Dyneema
and steel; stretch gloves made with
small percentages of natural latex and
Lycra yarns added to other fibers; and
special purpose gloves with materi-
als, such as thermal foam or vibration
foam. Special purpose gloves include
extra clean gloves and sterile gloves.
General purpose gloves. Nylon,
polyester and cotton gloves are com-
fortable and protect against snags,
punctures, cuts and abrasion. They do
not, however, protect against chemi-
cals and liquids, which is why chemi-
cally resistant gloves may need to be
worn as liners in some applications.
Nylon is an alternative fabric that
may be used in cleanrooms, automo-
tive paint rooms and inspection sta-
tions where there is concern about
lint contamination.
Cut resistant gloves. Gloves made
with Kevlar typically provide the best
cut protection based on price, and
gloves made with Spectra or Dyneema
ultra-high-molecular-weight polyeth-
ylene typically offer the best cut pro-
tection based on weight. Steel assures
the best cut protection based on bulk.
None of these materials, however,
will protect the hands from liquids.
Liquid-protective cut-resistant gloves
need to be coated with a compound,
such as nitrile.
Stretch gloves. Most of todays fabric
gloves are made from knitted fabric
Chemical Engineering www.che.com July 2008
rather than woven fabric, because knits
have a natural tendency to stretch and
fit closely. Natural rubber latex is used
to make the least expensive additive
elastic yarn for making tighter-fitting
gloves and cuffs. Workers, who need a
tighter fit and know or suspect they
have a latex allergy, should consider
switching to gloves made with a syn-
thetic alternative, such as spandex.
Special purpose gloves
Specialty gloves include those made
with thermal foam for protection from
heat or cold, vibration-absorbing foam
for workers handling power tools and
other vibrating machinery, or products
with radiation-absorbing additives
such as lead or bismuth. This cat-
egory also includes extra-clean gloves
for electronics cleanrooms and sterile
gloves for pharmaceutical facilities.
Chemical Barriers
Descriptions of glove materials that
may be used with various chemicals
follow, with advantages and disadvan-
tages listed. In addition, the Table above
summarizes some chemical-resistance
information. Gloves are listed below in
order, by degree of specialized use.
Natural rubber (latex)
Natural-rubber latex gloves (Figure 2)
are generally unsupported and avail-
able in many styles, including clean-
room and sterile styles. These gloves
provide excellent protection from bases,
alcohols, and dilute water solutions of
many chemicals, with fair protection
53
 Environmental Manager
against aldehydes and ketones.
Advantages: Low in cost, good physi-
cal properties, good cut protection in
heavy duty styles1, excellent dexterity
Disadvantages: Poor protection
against oils, greases and organic com-
pounds; risk of protein allergies. Some
manufacturers use shortcuts that re-
sult in a poor quality product
Polyvinyl chloride (PVC)
Gloves made with PVC are generally
available in heavy supported or light-
weight disposable styles, and protect
against strong acids, strong bases,
salt solutions and some heavy organic
chemicals. Many PVC or vinyl gloves
offer good abrasion and cut resistance,
although some styles may be suscep-
tible to cuts.
Advantages: Low in cost, fair physi-
cal properties, minimal risk of aller-
gic reactions
Disadvantages: Organic solvents can
wash plasticizers out, leaving holes
in the glove polymer on the molecular
level that may allow rapid chemical
permeation; gloves from some manu-
facturers are poor quality
Nitrile (Buna, NBR)
Nitrile gloves (Figure 1) are gener-
ally available as disposable, medium-
weight unsupported, or lightweight
supported styles. They protect against
oils and greases (including animal
fats), xylene, perchloroethylene and
aliphatic solvents. They also protect
against most agricultural pesticide
formulations, chemicals and biologi-
cal components used in weapons, and
other chemicals.
Advantages: Low in cost, excellent
physical properties, good dexterity; ex-
cellent resistance to snags, punctures,
abrasions and cuts
Disadvantages: Poor protection
against many ketones, some aro-
matic chemicals, and medium-polar
compounds
Neoprene
Neoprene is available in disposable,
1. Products that provide cut resistance and cut
protection do not completely prevent or elimi-
nate the potential for cuts or punctures, and
are not intended or tested to provide protection
against powered blades or other sharp or rotat-
ing equipment. Users are encouraged to always
use caution and care when handling sharp ma-
terials.
54 Chemical Engineering www.che.com July 2008
medium-weight unsupported,
medium-weight supported and
heavy supported styles. Neo-
prene protects against a broad
range of oils, oxidizing acids
(nitric and sulfuric), polar aro-
matics (phenol and aniline),
glycol ethers, oils, greases and
many other chemicals. Other
types of gloves may, however,
offer better protection against
some of these chemicals.
Advantages: Medium cost,
FIGURE 2. Natural-rubber latex gloves are avail-
medium physical properties, able in many styles and provide excellent protec-
medium but broad-ranging tion from bases, alcohols and dilute water solu-
chemical resistance tions of many chemicals
Disadvantages: Less resis-
tant to snags, punctures, abrasions and Disadvantages: Poor resistance to
cuts than nitrile or natural rubber certain solvents, including ketones,
esters and amines; poor physical prop-
Butyl rubber erties, extremely expensive
Butyl rubber is used only in medium-
weight unsupported gloves (Figure 3). Sealed-film (laminate) gloves
Advantages: Dexterity and out- Laminate is one of the most chemi-
standing resistance to moderately cally resistant materials available
polar organic compounds, such as and protects against almost any-
aniline and phenol, glycol ethers, thing, including most chemicals and
ketones and aldehydes biological compounds used in weap-
Disadvantages: Poor protection ons. Gloves made with this material
against non-polar solvents, including are excellent for hazmat applications.
hydrocarbons, chloro- and fluoro-car- Laminate gloves are often used as
bons; expensive liners, which takes advantage of their
thinness and is often the best way to
Polyvinyl alcohol (PVA) address their disadvantages.
PVA is used for medium-weight sup- Advantages: Moderate cost; thin;
ported gloves that provide a high level outstanding resistance to almost all
of resistance to many organic chemi- organic compounds
cals, such as aliphatics, aromatics, Disadvantages: No grip finish, poor
chlorinated solvents, fluorocarbons physical properties (very low resis-
and most ketones (except acetone), es- tance to physical damage), not as
ters and ethers. form-fitting as dipped gloves
Advantages: Very rugged and highly
chemical-resistant; good physical Chemical
properties with resistance to snags, Resistance Tests
punctures, abrasions and cuts Chemical-resistant gloves are typi-
Disadvantages: Will quickly break cally tested for degradation, perme-
down when exposed to water and ation-breakthrough times, and final
light alcohols, less flexible than many permeation rates.
other types of chemically resistant Degradation is a deleterious change
gloves, expensive in the physical properties of a glove
due to the effects of a chemical. It is
Viton commonly evaluated by measuring
This compound is used primarily in weight or dimension changes upon
medium-weight unsupported gloves to exposure. Currently, no widely used
protect against aromatics, chlorinated standard degradation test exists for
solvents, aliphatics and alcohols. gloves, although several groups have
Advantages: Good dexterity, out- tried to converge on one. The problem
standing resistance to many organic is that during actual use, only the out-
compounds side of a glove is exposed to chemicals,
 wearers hands to higher temper- show how much of a chemical perme-
atures than those used in testing, ates the glove barrier during continu-
which will make chemicals per- ous exposure. While lower ratings are
meate more rapidly. Gloves will be better, there is still no allowable skin
flexed and squeezed rather than contact threshold limit value (TLV) for
held in place in a test cell, which any chemical. Most people who select
will also make chemicals perme- gloves tend to pay more attention to
ate more rapidly. the breakthrough time.
FIGURE 3. Unsupported butyl
Gloves, however, are typically
gloves offer dexterity and out- worn to protect against possible ConclusionS
standing resistance to moder- splashes not continuous liquid Since the perfect glove does not exist
ately polar organic compounds contact. Chemicals, therefore, will to protect workers in every chemical
permeate through more slowly application, gloves should be selected
and it is difficult to test the outside of during actual use than in a standard that provide the appropriate level of
a multi-layer glove without having the permeation test. Breakthrough times protection for the specific chemical
results distorted by the properties of and permeation rates from F 739, handled. Questions should be con-
the inner layers. therefore, should not be considered as sidered regarding the chemical con-
Permeation is the process by which absolute constants. The data is use- centration and length of exposure,
a chemical moves into and through ful only to compare gloves and obtain dexterity, tactile sensitivity and cut
a chemically resistant glove film by a general indication of how well they protection required.
adsorption on the outside, diffusion can be expected to perform. Many glove manufacturers offer
through, and then desorption on the hand protection products that pro-
inside. Measurements commonly tabu- ASTM F 739-07 tect workers from a variety of chemi-
lated include breakthrough time (how This is the newest version of ASTM cals. In general, natural rubber latex
long after exposure the permeation F 739 and uses 27C as the standard is the least expensive alternative for
valve was opened) and permeation temperature since gloves are gener- tight-fitting gloves that resist bases,
rate (how wide it was opened). ally warmer than room temperature acids, alcohols and the diluted aque-
Both degradation and permeation during use. Much older data and Eu- ous solutions of most chemicals.
result from a chemical being absorbed ropean data were obtained at room Unsupported gloves offer a broad
and diffused into the film. The amount temperature (21C). Europeans de- spectrum of chemical resistance,
of chemical absorbed and the rate of cided that a consistent established depending on the material used.
transport, however, are independent method is more important than a Supported styles are often used for
variables, which means that neither somewhat more realistic method, general purpose, chemical- or cut-
property can be predicted based on which is why they continue to use resistant requirements.
the test results for the other property. 21C as the test temperature. When  Edited by Dorothy Lozowski
A simple degradation test, therefore, selecting gloves, it is important to
cannot be used to extrapolate the re- compare test data that were obtained Notes
sults expected from a more complex at the same temperature. 1. Kevlar is a registered trademark of Dupont.
Dyneema is a registered trademark of Royal
and time-consuming permeation test. DSM, N.V. Lycra is a registered trademark of
ASTM F 1383 Invista.
2. Neither this article nor any other statement
ASTM F 739 This test can be used to simulate inter- made herein by or on behalf of Ansell should
This is the permeation test most com- mittent contact applications, although be construed as a warranty of merchant-
ability or that any Ansell product is fit for
monly used and represents the origi- the test focuses on permeation only. a particular purpose. Ansell assumes no re-
nal permeation standard, although a During F 1383 tests, permeation test sponsibility for the suitability or adequacy of
an end users selection of gloves or clothing
substantial amount of data has been cells are filled and emptied repeat- for a specific indication. Upon request, Ansell
accumulated since it was first issued. edly on a schedule developed to match will provide a sample of material to aid you
in making your own selection to meet your
European-standards writers adopted the intended end use. Air or nitrogen own individual safety requirements.
ASTM F 739 with minor modifica- is blown through the chemical com-
tions as EN 388. During testing, per- partment during the entire empty Author
meation test cells are filled once and cycle, with evaporation away from and Nelson Schlatter is a technical applications
remain filled. The results typically diffusion into the sample occurring chemist for Ansell Occupational Healthcare
(1300 Walnut St., Coshocton, OH 43812; Phone:
provide the time delay until the chem- simultaneously. The ASTM F1383 In- 740-623-3591; Email: nschlatter@ansell.com;
Fax: 740-623-3556). He has worked with Ansell
ical breaks through, and the flowrate termittent-Contact Permeation Test, for 30 years on the chemical interaction and pro-
through the material, which increases therefore, may provide a more realis- cessing of textiles, rubber and solvents. He pres-
ently answers technical inquiries and provides
to a final constant value. tic permeation measurement of real- recommendations regarding the proper use of
Method F 739 does not realistically world breakthrough times. Ansell gloves and clothing. Schlatter holds a B.S.
degree from the University of Delaware. He is
simulate most end uses for chemically Breakthrough ratings indicate how an active member of ASTM Committee F-23 on
resistant gloves. During actual use, long a glove may be safely worn after a Protective Clothing and belongs to International
Safety Equipment Organization Committees on
gloves are likely to be warmed by the splash occurs. Final permeation rates hand protection and clothing.

Chemical Engineering www.che.com July 2008 55

 Solids Processing
Environmental Manager
Dust Control in the
Chemical Processing
Industries
Walter S. Kessler
Chilworth Technology, Inc.
D
ust in the chemical pro-
cessing industries (CPI) is
present in numerous op-
erations, including powder
processing; the transport of mate-
rials on belt or rotary-screw con-
veyors; grinding materials in giant
shredders and pulverizers; machin-
ing, sawing, grinding or sanding op-
erations; dumping bags of materials
into reactors; handling coating ma-
terials; as well as the processing of
pharmaceuticals and foods. An in-
tegral industry-safety issue arises
when we consider the handling
and suppression of the dust and
particulate matter associated with
these processes, namely prevent-
ing the formation of combustible
clouds that can create a fire hazard
or even trigger an explosion. Such
a catastrophic dust explosion in a
manufacturing facility is depicted
in Figure 1. Additionally, protect-
Figure 1.
The lack of proper
dust-control tech-
niques can result in
catastrophic fires
and explosions in
many types of
facilities
ing workers from the dangers of
potentially dusty environments,
including contact irritations and in-
halation exposure, is an important
factor in determining the meth-
ods and extent to which the par-
ticulate matter or dust needs to be
suppressed or eliminated.
Regulations
Dust hazards are so problematic
to employees that the U.S. Occupa-
tional Safety and Health Adminis-
tration (OSHA; Washington D.C.;
www.osha.gov) has implemented
rules for personal protective equip-
ment (PPE) specifically dust
masks and respirators, under stan-
dard 29CFR 1910.134 to protect
workers [1]. This standard states:
In the control of those occupational
diseases caused by breathing air
contaminated with harmful dusts,
fogs, fumes, mists, gases, smokes,
Chemical Engineering www.che.com July 2014
The prevention of dust
hazards in the CPI is
integral to process-safety
management wide-
reaching mitigation
schemes must be
implemented
sprays or vapors, the primary objec-
tive shall be to prevent atmospheric
contamination. This shall be accom-
plished as far as feasible by accepted
engineering control measures (for
example, enclosure or confinement
of the operation, general and local
ventilation, and substitution of less
toxic materials). When effective en-
gineering controls are not feasible,
or while they are being instituted,
appropriate respirators shall be
used pursuant to this section. It
goes on to state: A respirator shall
be provided to each employee when
such equipment is necessary to pro-
tect the health of such employee.
The employer shall provide the res-
pirators, which are applicable and
suitable for the purpose intended.
The employer shall be responsible
for the establishment and mainte-
nance of a respiratory protection
program. The program shall cover
each employee required by this sec-
tion to use a respirator.
Additionally, the National Fire
Protection Association (NFPA;
Quincy, Mass.; www.nfpa.org) has
stepped in and set consensus stan-
dards, based upon good engineering
practices, to control and prevent
combustible-dust related hazards.
NFPA 654 (Standard for the Pre-
vention of Fire and Dust Explosions
from the Manufacturing, Process-
ing and Handling of Combustible
Particulate Solids) applies to all
phases of the manufacture, process-
ing, blending, pneumatic conveying,
repackaging and handling of com-
bustible particulate solids or hybrid
mixtures, regardless of concentra-
tion or particle size, where the ma-
terials present a fire or explosion
hazard [2]. In addition to this stan-
dard, the NFPA has also set stan-
59
 Environmental Manager
dards for specific industries and
processes, such as: NFPA 61 (Stan-
dard for the Prevention of Fires and
Dust Explosions in Agricultural and
Food Products Facilities), NFPA 484
(Standard for Combustible Met-
als) and NFPA 664 (Standard for
the Prevention of Fires and Explo-
sions in Wood Processing and Work-
ing Facilities). These standards
hold, as a top priority, the action
of minimizing and controlling dust
and particulate matter in process-
ing industries from both a fire- and
explosion-hazard safety standpoint.
In addition to the human-health
impact, combustible airborne dust
and particulates in the right envi-
ronment where confinement, dis-
persion, concentration, oxidants and
ignition are present or potentially
present can create explosions.
Combustible dust, according to both
OSHA (CPL 03-00-008) and NFPA
654, is defined as a particulate solid
that presents a fire or deflagration
hazard when suspended in air or
some other oxidizing medium over a
range of concentrations, regardless
of particle size or shape. The most
common industrial sources of com-
bustible dust include: food (such as
candy, sugar, spice, starch, flour and
feed), grain, tobacco, plastics, wood,
paper, pulp, rubber, textiles, pesti-
cides, pharmaceuticals, dyes, coal,
metals (such as aluminum, chro-
mium, iron, magnesium and zinc),
and fossil-fuel power generation [3].
When these dusts are ignited, they
can produce a fireball 810 times
larger than the original volume of
the cloud in the absence of confine-
ment. Conversely, in a confined en-
vironment, explosive pressures can
increase to as high as 810 times
the original pressure.
Control measures
Every dust has unique physical and
chemical characteristics that impact
its level of hazard. Physical char-
acteristics include size, shape and
moisture content, among others.
Chemical characteristics include
flammability or combustibility, ex-
plosibility, susceptibility to thermal
degradation and instability, suscep-
tibility to ignition, and chemical re-
60 Chemical Engineering www.che.com July 2014
Figure 2. Even dust-mitigating equip-
ment can be the source of explosion
hazards if not correctly installed and
maintained
activity. It should be noted that the
physical characteristics of the dust
also effect the chemical character-
istics. For example, as particle size
and moisture content decrease, the
maximum explosion potential and
maximum rate of pressure rise per
unit time increases and the mini-
mum ignition energy (MIE) gener-
ally decreases [4]. The dust-explo-
sion hazardous classes (ST) range
from 0 (no explosion potential) to
3 (very strong explosion potential).
Testing of representative dust sam-
ples is the best method to classify
dust materials. Typically, the cost
of testing is recovered in reduced
engineering-control costs.
To control dust and particulate
matter, vacuum systems with vari-
ous forms of filtrations and particle-
removal devices have been designed
and installed in industry. The engi-
neering and design behind some of
these systems can be very detailed
and intricate. The key to their de-
sign is to move the dust and air at
a velocity fast enough to keep the
dust or particulate matter sus-
pended and moving inside of trans-
port ductwork until they reach the
filter mechanism or cyclone separa-
tor. At this point, they are removed
from the air to an acceptable level
before the air is released back into
the facility or outside environment.
These mechanical air-purifying
systems are intended for remov-
ing dust and particulate matter. An
important aspect to their function,
besides proper design, is proper
maintenance specifically chang-
Figure 3. Extra care must be taken any
time personnel have direct contact with poten-
tially combustible dusts or solids, for example,
when pouring a bag of reactant into a vessel
ing filters and keeping the system
clean. This ensures that dust and
particulate matter do not build up in
the equipment and cause a hazard-
ous fire and explosion scenario; this
is the reason they were installed in
the first place. In addition to proper
design, the installation of explosion-
protection (such as venting or sup-
pression systems), explosion-isola-
tion, and spark-detection systems
may be required, depending upon
the properties of the dust being col-
lected. One might think this is ob-
vious; however, systems installed to
mitigate dust and particulate mat-
ter have historically been involved
in fire and explosion incidents, as
seen in the dust-collector explosion
in Figure 2.
In some cases, dust is removed
through a wet-scrubbing arrange-
ment where the dust or air stream
comes into intimate contact with
water, and the dust is removed from
the air stream. The dust may be
dissolved in the water or may cre-
ate a slurry. In either case, this type
of collection requires further treat-
ment of the water solution or sludge
for environmental purposes.
In addition to vacuum-bag or
filter systems, water and various
polymer-additive systems have also
been used to control dust, mainly in
coal, mining and dirt-road applica-
tions. These systems typically con-
sist of a water-only, water-polymer
solution or oil-based liquid that is
laid down or sprayed onto a surface
by a nozzle-manifold or firehose-
like system.
Avoiding incidents
Due to the number of incidents re-
lated to combustible dust there
have been 281 major events reported
from 19802005, which have killed
119 workers, injured another 718
workers, and destroyed many indus-
trial facilities [3,5] OSHA reissued
its CPL 03-00-008 Combustible Dust
National Emphasis Program (NEP)
on March 11, 2008. Although OSHA
does not yet have its own standard
pertaining to combustible dust, the
agency cites combustible-dust haz-
ards, including fire deflagration, ex-
plosion and related hazards under
the General Duty Act of 1970. This
act states that an employer shall
furnish to each of his employees em-
ployment and a place of employment
which are free from recognized haz-
ards that are causing or are likely to
cause death or serious physical harm
to his or her employees [6], and relies
on NFPA standards for the recogni-
tion of such hazards. This includes
the statement that employers must
furnish each employee with a place
of employment that is free from rec-
ognized hazards that are causing, or
are likely to cause, death or serious
physical harm. In 20072009, OSHA
conducted approximately 1,100 com-
bustible-dust inspections and issued
over 4,900 citations for both combus-
tible-dust hazards and other safety
violations. From these violations, the
following list summarizes some of
the General Duty Violations issued
by the OSHA inspectors [7]:
Dust collectors were located
inside buildings that lacked
proper explosion-protection sys-
tems, including for venting and
suppression purposes
The rooms with excessive dust ac-
cumulations were not equipped
with explosion-relief venting dis-
tributed over the exterior walls
and roofs of the buildings
The horizontal surfaces, including
those on beams, ledges and screw
conveyors at elevated surfaces,
were not minimized to prevent ac-
cumulation of dust
Equipment, such as grinders,
shakers, mixers and ductwork,
were not maintained to minimize
escape of dust into the surround-
Figure 4. The dust explosion penta-
gon represents the conditions that can
lead to a catastrophic event
ing work area. This becomes espe-
cially problematic when employ-
ees may have direct contact with
powder or dusty materials, such
as when pouring bags into reac-
tors or vessels (Figure 3). Also,
the employer did not prevent the
escape of dust from the packaging
equipment, creating a dust cloud
in the work area
Interior surfaces where dust ac-
cumulations could occur were not
designed or constructed to facili-
tate cleaning or to minimize com-
bustible-dust accumulations. Reg-
ular cleaning frequencies were not
established for walls, floors and
horizontal surfaces, such as ducts,
pipes, hoods, ledges and beams
Compressed air was periodically
used to clean up the combustible-
dust accumulation in the presence
of ignition sources
Explosion vents on dust collec-
tors and bucket elevators were
directed into work areas and not
vented to a safe, outside loca-
tion away from platforms, means
of egress or other potentially
occupied areas
Process hazard analysis (PHA)
was not conducted to determine
whether the process hazards ne-
cessitated the installation of ap-
proved devices, such as explosion-
protection systems, interlocked
rotary valves, deflagration vents
and flame-front diverters
The employer did not provide
adequate maintenance and de-
sign of dust-collector systems,
which created insufficient air as-
pirations, low duct velocities and
blocked ducts
Even with OSHA implementing the
NEP program, combustible-dust
related incidents continue to be a
major industrial problem in the U.S.
and globally. There were over 500
Chemical Engineering www.che.com July 2014
combustible-dust related incidents
reported in 2011 in the U.S. alone
[8]. That number does not include
the events associated with grain el-
evators or coal-fired power plants, or
smaller flash fires that were quickly
extinguished, as well as other near-
misses that failed to be reported.
From the above list, some good
rules of thumb and generally ac-
cepted good engineering prac-
tices for facilities that contain
dust and particulate matter from
production are [9]:
Implement appropriate engineer-
ing designs and controls to mini-
mize the presence of dust and to
prevent the dust explosion penta-
gon (Figure 4) from occurring
Perform manufacturer-recom-
mended maintenance on all equip-
ment to ensure that it is function-
ing as designed
Implement good housekeeping
practices, including using surfaces
that minimize dust accumulation,
periodic inspections for hidden
areas of dust accumulation and
controlling sources that could
cause dust to become airborne
to maintain a clean, dust-free
working environment
Follow other measures, including:
bonding of equipment to ground
to control static electricity; con-
trolling smoking and sources of
open flames and sparks; manag-
ing friction and other sources for
mechanical sparks
Provide hazard recognition train-
ing for employees per OSHA
3371-08 2009
Establish overall safe work prac-
tices, such as proper electrical-
area classifications, physical bar-
riers for the hazard, cleaning that
does not generate dust clouds
and locating relief valves away
from dust
Report and record any and all
incidents or near-misses, as
they are a valuable tool and re-
source for preventing larger, more
serious incidents
With all these recommendations,
regulations, reported incidents
and near-misses, facilities should
strongly consider use of compe-
tent personnel and knowledge-
61
 Environmental Manager
able individuals to provide testing,
hazard assessments, risk analysis
studies and engineered solutions
to minimize or control danger-
ous dusts and particulate matter
in manufacturing.
To achieve process-safety excel-
lence, employees must be genuinely
proficient and competent in their
requisite technical disciplines, and
appropriate levels of knowledge
must be embedded in key positions
References
1. OSHA Standard 1910: Personal Protective
Equipment, 19982011.
2. NFPA 654: Standard for the Prevention of
Fire and Dust Explosions from the Manufac-
turing, Processing and Handling of Combus-
tible Particulate Solids, 2013.
3. OSHA, Combustible Dust: an Explosion Haz-
ard, OSHA website, https://www.osha.gov/
dsg/combustibledust.
4. Amyotte, P., An Introduction to Dust Ex-
plosions: Understanding the Myths and Re-
alities of Dust Explosions for a Safer Work-
place, Elsevier, 2013.
The Chemical Engineering bookstore
offers a variety of industry topics
you will come to rely on.
Environmental Management:
Air-Pollution Control
Environmental Management: Wastewater
and Groundwater Treatment
Fluid Handling
Gas-Solid and Liquid-Solid Separation
Liquid-Liquid and Gas-Liquid Separation
Managing Bulk Solids
For a complete list of products, visit the
Chemical Engineering bookstore now.
http://store.che.com/product/book
throughout an organization with
a mechanism for longevity. Or, put
more simply, organizations must
have the right people, with the
right skills, implementing appropri-
ately designed process-safety pro-
grams, motivated by the right orga-
nizational culture, in the right way.
Results are far-reaching and broad,
affecting finances, the environment
and most of all, workforce safety.
Properly managing dust protects
5. U.S. Chemical Safety and Hazard Investiga-
tion Board, Investigation Report: Combusti-
ble Dust Hazard Study, Report No. 2006-H-1,
November 2006.
6. OSHA General Duty Act of 1970, Sec. 5
Duties, 1970.
7. OSHA Status Report on Combustible Dust
NEP, October 2009.
8. National Fire Incident Reporting System.
(NFIRS) Combustible Dust Policy Institute,
2011 Combustible Dust Related Incidents
Fact Sheet, April 2013.
9. OSHA, Combustible Dust in Industry: Pre-
venting and Mitigating the Effects of Fire
and Explosions, SHIB 07-31-2005.
17792
Circle 3 on p. 68 or go to adlinks.che.com/50978-03
Chemical Engineering www.che.com July 2014
the workforce, facility and the envi-
ronment, while maintaining stake-
holder confidence and encouraging
compliance with all legislation and
avoiding regulatory intervention.
Eliminating or minimizing dust is
a small investment compared to the
costs of the harm that can occur if
not appropriately addressed. 
Edited by Mary Page Bailey
Author
Walter S. Kessler is a se-
nior process engineer and
safety specialist at Chilworth
Technology, Inc., a DEKRA
Company (113 Campus Drive,
Princeton, N.J. 08540; Phone:
609-799-4449; Email: walter.
kessler@dekra.com). He has
20 years of experience in pro-
cess design, development and
improvement, process control,
logic-controller programming,
process-hazard analysis, layers-of-protection
analysis (LOPA), safety-integrity levels (SIL)
and process safety management in petroleum
refineries, gas and chemical plants, as well as
pharmaceutical, manufacturing and wastewater
facilities. Kessler holds a B.S.Ch.E. from Penn-
sylvania State University and is currently ob-
taining a Ph.D. at the University of Houston.
63
Feature Report

The Integral Role of Waste

Management in Capital
Projects
To avoid unnecessary efforts and mitigate risks, all phases throughout the design and
implementation of large capital projects should incorporate comprehensive planning
for waste management

P
Frederick V. roper waste-manage-
Jones and ment planning and exe-
Susan Rankin
cution during all phases
ERM
of a large capital project
in the chemical process indus-
tries (CPI) can contribute greatly
In Brief to a projects success. How well
Company philosophy waste management is integrated
Capital project into a project is often an indicator
structure of the projects overall risk man-
agement.
project
This article provides guidance
infrastructure
on how to integrate waste-man-
planning
agement planning and execution
managing phase into the entire lifecycle of capital
transitions projects to reduce risk, manage
FEED-level activities costs and maintain schedules.
This lifecycle approach takes the
construction phase process from project scoping and
operations phase early design through startup, op-
erations and decommissioning.
decommissioning
The article also identifies specific
closing thoughts approaches and mechanisms that
can be integrated into a compa-
nys existing project-development Figure 1. Scrap metal, along with scrap wood, spent solvents, used filters
and execution tools, and provides and used oils are among the high-volume waste streams that must be
approaches to move waste-man- handled during the construction phase of a capital project
agement programs forward within
and between all stages of a capital project. lays, higher costs and increased liability for
An area of waste-management planning the project owner. Construction wastes are
found most frequently to be challenging is usually high-volume streams, such as used
integrating and managing the ever-expand- oils, used filters, used tires, solvents, scrap
ing contractor activities into the operators metal and wood, waste cement and so on
waste-management strategy and philoso- (Figure 1). Although many of these waste
phy. Contractor-management approaches, streams are not hazardous, the high volume
such as auditing programs and third-party can be a problem for disposal, especially if
contract requirements, are often not ad- local handling capacities are exceeded. Par-
dressed until the project is well into design ticularly in greenfield areas outside the U.S.,
and construction, resulting in confusion, de- implementing proper waste-handling proce-

Chemical Engineering www.chemengonline.com july 2015 43

 TABLE 1. Early-Phase Considerations For Waste Management in capital projects companys program during interna-
Waste-management objec- Determine the most important objectives and strategies for project waste tional expansion is to ensure that
tives, philosophy and strategy management, addressing priorities and considering tradeoffs. There is no its overall operational excellence/
one right way to manage wastes; a specific vision that team members can operational integrity management
plan for will ensure consistency across project phases and disciplines.
(OE/OIM) system is robust enough
Permit acquisition and The plan should include anticipated permit requirements, required data,
execution plan reports, timing, fees, reporting and training.
to handle international consider-
ations and capital projects. Waste is
Waste inventory Identify anticipated waste for all project phases (early works, construc-
tion and operation) and activities (camp, infrastructure construction and only one aspect of this overall OE/
facility construction). The inventory will be a living document and will be OIM system, but if the OE/OIM sys-
expanded in its level of detail as the project progresses. tem is not structured well, then all
Waste-management infra- Identify and evaluate local and regional waste-management facilities for of the management programs will
structure review all anticipated waste streams and management methods. Consider all be hampered, including waste man-
possibilities at a high level.
agement. Once these foundational
Transportation infrastructure Identify and evaluate local and regional transportation options for all
review anticipated waste streams and management methods. Consider all pos-
issues have been addressed, the
sibilities at a high level. focus can be shifted to the specific
Requirements inventory Develop an inventory and comparison between regulatory, company and capital project in question.
lender project requirements. Ensure all possible requirements are inven-
toried and considered with a high level of rigor. Any possibly applicable Capital project structure
requirement should be included, with the understanding that if it is later
determined to not be applicable, it can be removed. See below for
Successfully planning for and inte-
additional notes. grating waste management into a
Risk analysis Identify and determine the magnitude of waste-management risks, and capital project requires both planned
develop options for resolving major risks. How can past risk-review find- and free-flowing communication be-
ings be utilized and expanded to this phase? Are resolutions of risk tween project sub-teams that are
still applicable? simultaneously executing different
Preliminary waste-manage- Integrate all of the above information. The P-WMP provides a guiding project work scopes. As with all as-
ment plan (P-WMP) document for all subsequent project activities. More detailed activity and
phase-specific waste-management plans will be developed based on the
pects of a project, the extent to which
project P-WMP. sub-teams are siloed increases the
Bridging document Define design-basis contract requirements for early works to support the risk that waste-management plan-
front-end engineering design (FEED) contractor designs for waste facilities ning will be inadequately integrated
when appropriate. into project development. Specific
Additional notes on the requirements inventory: Any possibly applicable requirement should be included, with approaches that can encourage
the understanding that if it is later determined to not be applicable it can be removed. This is key, as it is not more disciplined integration earlier
uncommon for a regulation or lender requirement originally thought to not be applicable to later be found
to be applicable. For example, the domino effect of more stringent discharge requirements than originally in the project-development process
thought can have a tremendous impact on project design once the design has been initiated and may are detailed in the sections below.
result in project delays. The resulting facility may end up being suboptimal, as a facility design that mitigates Establish a collaborative project
late new information will often differ from a facility design whose basis originally included the applicable lim-
its. Have a third party review the inventory and confirm that all possible requirements are considered.
culture. Companies have come to
recognize the importance of fully in-
dures is critical, because low waste- the first step to a sound waste-man- tegrating personnel from all project
handling capacity, poor, potentially agement approach is often to review scopes, including facilities develop-
high-risk transportation routes and these corporate guidelines and ex- ment, design and construction, as
local regulations can often prove pand them if necessary to include well as auxiliary operations, such
to be problematic. issues that will impact a capital-proj- as housing, wastewater-treatment,
ect program. Some issues of con- solid-waste management, spill re-
Company philosophy cern may include the following: sponse, pipeline operations and
Most companies have developed a Expanded and more restrictive so on. Personnel involved with all
basic waste-management philoso- country regulations (for example, project phases (including person-
phy and strategy based on good those related to wastewater- nel from operations) should also be
practices and home-country stan- discharge standards, hazard- integrated into project development
dards and regulations, as well as ous-waste definitions and final and execution. However, environ-
internal operating and management handling, treatment and disposal mental planning functions are often
systems. References 17 give more requirements) not integrated into project planning.
details on specific waste-manage- Local-country third-party contrac- Environmental, social and health-
ment standards and regulations. tor requirements related to waste impact assessment (ESHIA) prepa-
Based on the companys inter- management ration is typically conducted by an
national presence, these internal environmental team (in many in-
Infrastructure limitations
guidelines are often the first stage stances, a third-party contractor).
Engineering, procurement and
and foundation of a capital-project At this point, the core environmen-
construction (EPC) contractor
waste-management program. For tal team is in a non-collaborative
companies that are expanding into programs role, such as gathering information
international areas for the first time, Another important aspect of a from other team members instead
44 Chemical Engineering www.chemengonline.com july 2015
of having an active role in the plan-
ning and execution of the project. If
ESHIA preparation is a companys
sole mechanism for waste-man-
agement planning during the early
project phases, and if the person-
nel preparing the ESHIA are not fully
integrated into the project, the com-
mon outcome is that waste-man-
agement planning will be inadequate
and untimely.
Include waste management in
the project development, design
and execution. All project planning
evolves from the initial projects phi-
losophy, objectives, plans and basis
of design (BOD). Ensuring waste
management is included in the origi-
nal project philosophy and BOD will
help to ensure its timely consider-
ation in the subsequent plans for
environmental management, con-
tractor management, construction
management and other programs
that follow.
Consider waste management as
necessary infrastructure. Civil en-
gineers were the original waste-man-
agement engineers. Over the last 30
years, waste management has shifted
to being an environmental discipline,
and is no longer specifically covered
by a core engineering discipline dur-
ing project development. As a result,
planning for waste-management
infrastructure is often conducted
separately from other infrastructure
planning and often lags behind
other infrastructure planning. En-
suring that the waste-management
infrastructure scope is covered by
the projects overall infrastructure
development will ensure waste-
management planning is in step
with the overall project development.
Waste-management planning should
take place in parallel with plans for
power distribution, roads and water-
supply infrastructure.
Project infrastructure planning
Large capital projects in remote loca-
tions or countries with poor infrastruc-
ture typically require an infrastructure
to be established in order to support
the project. The infrastructure is usu-
ally established in phases, with mul-
tiple and sequential developmental
projects required to establish basic
infrastructure before full construc-
Chemical Engineering www.chemengonline.com
TABLE 2. FEED-Phase Considerations For Waste Management in
capital projects
Expand strategy and Expand the strategy to include lifecycle waste generation.
objectives
Waste inventory development Develop a more detailed waste-generation inventory and an analysis
of waste streams for all phases (early works, construction, operations
and closure). Evaluate what risks are related to the volume and types of
wastes generated.
Waste-management infra- Identify optimal waste-management infrastructure installation timing and
structure expansions scaling. With increased personnel comes an increased demand for camp
facilities, resulting in more domestic and industrial waste streams. Vol-
umes are initially low. However, if not managed correctly and accounted
for in the initial camp construction, they can cause problems. Sewage
overflows and solid-waste disposal problems are typical examples.
Risk analysis Conduct a more detailed risk analysis that covers all aspects of construc-
tion and operation. Hazard identification (HAZID) analyses are the industry
standard; however, environmental considerations are sometimes limited
in these HAZID analyses. Make sure waste-management is included. Ad-
ditional considerations regarding risk analysis are addressed below.
Waste-management facility Conduct a more detailed audit of candidate waste-management facilities
and transportation audits and transportation options to determine the viability, capacity and compli-
ance status of each option. The initial baseline review and preliminary
audit should be conducted during the early new-venture operations so
that operations personnel have a basis for waste disposal. The scope is
expanded to include infrastructure and construction waste streams during
this audit.
Baseline studies Conduct detailed studies to provide data relevant to FEED design and to
ESHIA development, including waste-management facility design.
Baseline and other site activ- The baseline studies themselves will generate wastes that must be han-
ity wastes dled as part of the project. These wastes need to be planned for as part of,
or separate from, early works.
Integrated waste-manage- Update each of the waste documents discussed above with more specific
ment plan (I-WMP) information from FEED contractors, personnel involved in government
engagement and other sources. The resulting I-WMP will be used for the
updated and final capital project basis of design (BOD). The I-WMP must
also address how to tie multiple contractor waste streams into an overall
project waste program. While addressed to some extent during the FEED
phase, this will become a major issue during initial construction activities
within the construction phase.
Additional notes regarding risk analysis: The risk analysis should consider changes made to initial design,
construction and execution concepts as the BOD is finalized. The expanded risk analysis should include
scenarios for the lifecycle of the project. Representatives from the FEED and construction contractors, as
well as baseline and ESHIA contractors should be involved. Regulatory compliance must also be considered
as part of the risk analysis.
tion of the capital project itself can opment and execution. Instead, a
be initiated. Waste-management company may recognize that a solu-
planning in such situations is quite tion or combination of solutions may
challenging, as the infrastructure be costly but necessary in the short
projects, in many instances, have to term. In the meantime, it can develop
be advanced faster than the capital more economical, localized facili-
project itself. ties to support later project phases.
It must be understood that waste- Writing off a legally and logistically
management methods will change viable management method in the
as infrastructure develops. Waste- short term often results in stockpiled
management methods likely need to wastes. These stockpiled wastes will
be unique for different project phases frequently require continued invest-
and a specific waste stream may ment to maintain stability while a
have more than one solution at differ- permanent solution is developed or
ent points in time. As a result, blan- awaits government approval. Often,
ket management requirements (for the stabilization method increases
instance, the refusal to ship waste the waste volume, and the low-cost
out of a region because it is cost pro- solution that was originally envisioned
hibitive, or that waste cannot be sent ends up not being viable.
to a certain facility because the facil- Also crucial to develop is a philoso-
itys throughput will not be adequate) phy that dictates waste-management
can be detrimental to project devel- infrastructure development be nested
july 2015 45
 TABLE 3. Construction-Phase Considerations For Waste Management in
will be part of the permanent-camp
capital projects project and will not be used until the
Contract development Ensure early alignment among personnel who are developing contract
operations personnel arrive onsite.
packages, to ensure that the contracts clearly communicate waste-man- Depending on the size and complex-
agement roles and expectations. If one contractor is expected to manage ity of the project, multiple sites may
wastes for other contractors, this needs to be clearly communicated. be developed, each with varying lev-
Consolidated waste-manage- Develop a consolidated waste-management plan shortly after the EPC els of waste-management facilities.
ment plan contract award. All contractors and company or operator personnel
should be involved in the plans preparation. The plan must address waste
Construction may also continue well
streams that will change custody, how custody exchanges will be man- after operations startup, resulting
aged, specifications the waste streams must meet in order for another in overlapping needs and require-
contractor to take custody of the waste streams, timing and frequency ments that will need to be consid-
available for custody transfer and how the receiving party will manage the
waste thereafter. ered in the design, as well as the
management of facilities.
Waste-management market- Develop a plan to clarify how involved the EPC contractor will be in de-
development plan veloping waste-management and recycling markets. The operator will Operations camp. While a project
need to engage with the EPC contractor to ensure that the EPC contractor landfill may have been in operation
meets company expectations, and understands the operators long- and during early project construction,
short-term market-development plan.
new permanent waste-management
Contractor waste-manage- Include requirements for all contractors to prepare their own waste- facilities will often need to be brought
ment plans management program for their activities, which shall be aligned with the
operators consolidated waste-management plan (discussed above). online roughly when the capital-
Waste-management facility If the project is in a remote area with little or no waste-management
project facility starts up. Typically,
consolidation infrastructure, the EPC contractor typically constructs project waste- dedicated waste equipment, such
management facilities shortly after mobilizing. It is at this time that waste as incinerators and domestic waste-
management becomes more consolidated. Typically, incinerators used for treatment facilities, are brought on-
early works are taken out of service or combined with the EPC waste fa-
cility. Waste is often redirected to a central project facility. Sewage is also line with operational startup. How-
typically redirected from the previously established camps and is taken ever, construction facilities may be
to the construction or permanent wastewater-treatment plant. The point kept on location for future expansion
at which these waste streams are consolidated requires engineering and
logistics planning.
work, resulting in multiple waste fa-
cilities. Remote camps may also be
within and integral to the overall to mobilize and establish a construc- constructed away from the primary
capital project. The greatest chal- tion camp at the site. An early camp facility, such as those associated
lenge with waste-management in- will often be required, along with with pipeline booster stations and
frastructure planning is the chicken- associated waste and wastewater- coastal loading and offloading facili-
and-egg scenario finalizing the treatment facilities. The early-works ties. These camps may or may not
waste-management infrastructure contractor should consider building have separate waste facilities, but
BOD, when the capital project facility out from earlier camps, depend- will surely have waste generation
BOD scope and front-end engineer- ing on conditions and location. The that will have to be managed.
ing design (FEED) are not yet final- waste-management infrastructure Integrated waste-management
ized. Ensuring that the waste-man- requirements may then be a simple planning that considers each of
agement infrastructure work is fully expansion of these camp structures. the above phases as possibilities
integrated within the capital-project Alternatively, a new camp may be re- will be required, in order to ensure
design will provide the foundation for quired. From a waste-management that the waste-management in-
the waste-management infrastruc- perspective, this would typically frastructure is sized appropriately
ture and ensure that the projects ad- still include a package wastewater- and brought online at the right
vance on time and within the capital treatment plant, one or more incin- time. The following planning items
project budgets. erators and perhaps other solid- require consideration:
The company must also work waste-management facilities, such 1. It is imperative that all aspects of
with and demonstrate great com- as storage facilities, initial treatment wastes from the construction and
petence to stakeholders at plan- facilities and possibly the beginning operations phase be considered
ning and executing the early project of a limited landfill. Waste recycling when sizing waste-management
phases, in order to ensure that the opportunities should be consid- facilities. If not integrated into the
project stays on schedule, and that ered for certain waste streams, permanent camp, early-works
stakeholders continue to be aligned such as scrap wood, scrap metal camps may need to be demol-
with the project. Planning project and used oil. ished or abandoned. The con-
development phases requires proj- Full construction camp and ac- struction and demolition (C&D)
ect team integration for all functions tivities. The construction contractor waste from these camps removal
as detailed below. will construct and initiate the use of will need to be captured by the
Early works and camp. The early- some permanent waste-manage- construction or permanent-camp
works scope consists of the infra- ment facilities, such as a landfill and waste-management planning, or
structure required to enable the cap- a waste-storage and transfer station. handled by a third-party contractor
ital-project construction contractor Other waste-management facilities separately from operations. These
46 Chemical Engineering www.chemengonline.com july 2015
 Adroscoggin Valley Regional Refuse Disposal District
TABLE 4. Operations-Phase Considerations For Waste Management in
capital projects
Contractor-demobilization Plan for a large influx of waste from the construction camps demo-
planning bilization. These wastes can overwhelm the operators facilities if this
occurrence is not planned for. For demobilization waste streams, timing,
volume, type and treatment must be considered with careful attention.
Industrial wastewater consid- Ensure measures, such as injection wells or bio-treatment, are considered
erations early on during design and risk assessment. The increased levels of low-
volume, high-toxicity industrial wastewater streams that are difficult to
handle and treat may cause regulatory problems.
Operations waste-manage- If applicable, ensure that maintenance and routine waste streams are Figure 2. A master-planned landfill is extremely
ment during startup managed properly at remote locations. This is particularly the case for helpful in aiding waste-management efforts, and
wastes that will be generated only during startup. Equipment often comes landfill facility design should take into account
with lubricating oil or other materials designed specifically to maintain waste streams from all project phases
equipment integrity during shipping, and these typically must be drained,
removed and replaced with a different material for operations.
must include such requirements
into the invitation to tender (ITT)
Recycling market stability Onsite and offsite recycling programs can be expanded in many cases
now that routine volumes are generated. Recyclable material can be uti- package for each contractor,
lized onsite more efficiently during future construction projects, and offsite with clear understandings of how
recyclers can count on a steady input stream. waste will be managed onsite and
by whom. Before final pricing of
added waste streams need to be take care of its own wastes based any phase, the contractors whose
incorporated into the design of the on company restrictions? Deter- onsite presence will overlap should
waste facilities (in particular the mine how this will be managed have an opportunity to develop a
landfill). Figure 2 shows an exam- 6. Domestic wastewater-treatment- coordinated transition plan that
ple of a master-planned landfill that plant discharges for the construc- will address all areas of transition
can incorporate industrial waste tion and operations camps need including waste management.
streams. Another item to consider to be modeled in conjunction with The plan development must also
is whether any of the early-works industrial discharges. In many include waste generators that
waste-management facilities will cases, the initial ESHIA covers only are not specifically involved in the
be or should be used by the con- the first few project phases. Lon- transition, so that all parties under-
struction contractor ger-term modeling and planning stand their roles, expectations and
2. All project-development camps for varying discharge scenarios impacts
(all camps except for the facility is required in order to understand 2. What are the interdependencies
operating camp) will need to be and mitigate for this scenario, even between the extent to which the
decommissioned and abandoned if the ESHIA does not yet consider onsite contractor completes its
at some point in time. A camp clo- the intermediate and final build-out scope and the ability of the next
sure plan is required for each camp plans. Careful consideration of how contractor to mobilize onsite?
to establish the expected camp discharge locations may shift as Careful consideration of the onsite
end-of-service timing and closure the project develops is required for contractors schedule will show
vision. Plans for these wastes and this modeling the point at which construction is
facilities must be accounted for sufficiently complete for the next
3. In general, the construction con- Managing phase transitions contractor to mobilize and take
tractor will be advised to arrange Transitioning between the early- over waste-management tasks.
for suppliers to remove any un- works, full-construction and op- This vision of how these will likely
used materials, shipping contain- eration phases is typically a time of look needs to be clearly outlined
ers and packaging. Deviations overlapping contracts, with many and frozen for bidding. Also im-
from this expectation need to be contractors concurrently onsite. The portant to include is a transition
understood and planned for overlap, particularly in the case of risk review of waste management
4. In some cases, unused construc- full construction and operation, may to make sure everyone is aware
tion materials, shipping contain- take place over several years. Mini- and knows what to do
ers and packaging, and other mizing project disruption and prop- Some of the activity-specific nuts
items may be used for community erly preparing contractors ultimately and bolts to consider for each phase
engagement and improvement. results in lower project costs, as of a capital project are detailed in the
Advance planning with legal, can- clear expectations reduce execution following sections.
didate contractors, and stake- risk. Some considerations to help
holder-engagement personnel is reduce cost and risk are as follows: Early phase
required in order to plan appropri- 1. Contractors coming onsite must During new venture reviews, several
ately for these wastes have reviewed and incorporated key activities, along with planning,
5. Once operations start up, will the the companys waste philosophy need to be conducted prior to be-
operator manage residual and fu- and BOD with regard to waste ginning field operations. These in-
ture construction project wastes or into their scope of work and work clude understanding both the regu-
will the contractor be required to expectations. Thus, the company latory and social conditions for the

Chemical Engineering www.chemengonline.com july 2015 47

 TABLE 5. Decommissioning-Phase Considerations For Waste Management in streams are still common ones (for
capital projects instance, used oils). The principal
Facility optimization A philosophy that abandonment is part of the facility optimization and waste problem now becomes one
philosophy operation needs to be followed, and the operating organization should be of logistics, movement and handling.
framed with this in mind.
Table 3 addresses some items that
Camp decommissioning and All project-development camps (all camps except for the facility's final
abandonment operating camp) will need to be decommissioned and abandoned, as
must be accounted for during the
discussed earlier. The decommissioning costs are typically borne by the construction phase.
contractor that developed the camp if they still occupy it, or the operator, if
the operator is occupying the camp. Operations phase
Facility rationalization and A facility rationalization plan is required to provide guidance in optimizing Operation of the facility usually results
partial abandonment facilities as production starts to decline.
in demobilizing the construction facil-
Strategic abandonment Financially, asset abandonment has been found to be best done in tandem ities and consolidating residual waste
planning with operations. Production sharing contracts, partnership contracts, tax
incentives and financial reporting typically incentivize a company to aban- streams into the permanent waste fa-
don as they go, instead of abandoning at the end of an assets life. cility during closeout. Some residual
Asset-retirement waste- An asset-retirement plan that includes waste-management planning is construction may continue with a re-
management plan required, based on the facilities rationalization and optimization plan. What duced construction camp. The camp
scrap yards, and other facilities are required, where, when and with what and associated equipment may re-
throughput or capacity? How can infrastructure log jams be minimized
by optimizing the distribution of facilities to be retired, and aligning plans main onsite with the anticipation of
with other capital projects? expansion needs or may be removed
completely, depending on operator
environment and infrastructure of tractor will manage a large part of strategy. The volume of waste will
the area. How difficult will it be to the baseline work from an engineer- be reduced considerably; however,
obtain permission to operate? What ing standpoint, including evaluating with the startup of operations, other
will be the cost in time and equip- soil types, water levels, weather and waste streams will come online that
ment to support project activities? land data and so on. The company may be more hazardous (process
These are key questions that need will manage other studies, such wastewater streams, solvents, tank
to be answered in the initial evalu- as the full-project ESHIA work and bottoms and so on) that may require
ation of a new venture. Can these increased regulatory and social ef- additional third-party contractors and
risks be reduced to manageable forts, to make sure the project possibly expanded recycling and or
levels or will the costs be too high? moves forward. Table 2 provides treatment facilities. Table 4 explains
In some cases, a local office in some considerations for FEED-level of the waste-management concerns
the country will be arranged for waste management. that arise once the capital project
workers to start the process of in- The initial activity involves devel- becomes operational.
troduction of the company to the oping an ITT package for contractor
local government and get a realtime bids for pre-FEED and FEED items Decommissioning
understanding of the issues that and for the development of a BOD Planning for capital-project decom-
can arise. Although often of limited document. Environmental consid- missioning (including asset retire-
size and scope, this office complex erations, including waste manage- ment) has historically been done to
will typically be the first activity the ment, need to be included in the meet facility operating requirements,
company will have related to waste development of these bid packages often in response to a sudden need
management within the country of and BOD documents. for the facility to ratchet down its
operation, and can provide insight operating parameters to accommo-
into future problems. From a waste Construction phase date production decreases. The in-
perspective, Table 1 describes some During the construction phase, the dustry has recently begun to initiate
of the actions that should be taken. largest expansion of personnel usu- asset retirement in a more planned
ally occurs, and the volume of waste and proactive fashion. As a result,
FEED-level activities generated is often at its maximum. wastes related to decommissioning
FEED activities relate to the BOD The types of waste that must be and asset retirement require more
and the beginning engineering of handled will expand to include re- advanced planning and earlier man-
the project facilities and pipelines, as maining waste streams from the agement, while the waste-genera-
well as baseline studies as needed, completion of work on early-phase tion rate peaks at a lower rate. Table
and a full-project ESHIA program if infrastructures, and now full-blown 5 lists some considerations related
required. Accordingly, the project construction waste streams, as to decommissioning.
often has an increased demand well as domestic waste streams.
for rooms and lodging, increased Large volumes of used oils, lubri- Closing thoughts
travel in and out of the country cants, waste filters and used tires, Waste management is generally not
and various contractors arriving for as well as scrap wood, plastic and viewed as a make or break item
both short and long assignments concrete are generated. Although for capital projects. However, if not
to collect information and conduct the volume of hazardous waste in- managed correctly, it can be a major
research. Typically, the FEED con- creases, the main types of waste frustration and distraction to the
48 Chemical Engineering www.chemengonline.com july 2015
team leads, resulting in lost time and
efforts to correct problems that arise.
Waste management can also create
a loss in value to the project, and in
many instances can be a very costly
liability, if not managed correctly.
Risk assessment, and transitioning
lessons learned from one phase to
another can reduce this liability. A
well-structured and communicated
waste-management philosophy can
go a long way toward reducing risk
and costs.
Understanding local laws and re-
quirements, and coordinating con-
tractors through detailed BOD and
ITT packages can also be a risk re-
ducer. Understanding the local area
infrastructure abilities and planning
well in advance for treatment and
disposal options can, in the long run,
save time and expenses when deal-
ing with waste generation.
Finally, demobilization is a key fac-
tor in a projects overall waste strat-
egy. A large volume of waste, with
large volumes of hazardous waste,
can result from demobilization of fa-
cilities once construction is finished.
Making sure these waste streams
are accounted for in the design of
the waste program can reduce liabil-
ity and overall risk once the project
is completed. Phone: (832) 730-4381; Email:
Edited by Mary Page Bailey
References
1. American Petroleum Institute (API), Environmental
Guidance Document: Waste Management in Explora-
tion and Production Operations, Publication No. API
E5, 2nd Ed., Exploration and Production Department,
American Petroleum Institute, Washington, D.C.,
Feb. 1997.
2. IPIECA, Petroleum Refining Water/Wastewater Use
and Management, Operations Best Practice Series,
2010.
3. Environmental Protection Agency (EPA), Final Stan-
dards Promulgated for Petroleum Refining Waste,
EPA 530-F-98-014, 1998.
4. API, Manual of Disposal of Refinery Wastes, Volume
on Solid Waste Refinery Department, 1980.
5. API, Manual of Disposal of Refinery Waste Volumes
on Liquid Wastes, Publ. 931 C18, 1975.
6. API, The Generation and Management of Wastes
and Secondary Materials in the Petroleum Refining
Industry, API Pub No. 4530, 1991.
construction, commissioning and operations oversight.
7. International Finance Corp. (IFC), Environmental,
Health and Safety Guidelines for Petroleum Refining,
World Bank Guidelines, 2007.
Authors
Frederick V. Jones is a technical
director with Environmental Re-
sources Management (ERM; 840
West Sam Houston Parkway North,
Suite 600, Houston, TX, 77024;
fred.jones@erm.com). Jones
works with companies to develop
HSE management systems, capital
projects, regulatory compliance,
ESHIA, waste- and water-management programs, and
environmental support of projects and production oper-
ations to achieve lower corporate exposure and liability,
reduce costs and achieve regulatory compliance. Jones
has 35 years of experience in oil and gas drilling, pro-
duction and project development and operations.
Susan Rankin is a senior consul-
tant with ERM (same address as
above; Phone: (832) 209-8827;
Email: susan.rankin@erm.com).
Rankin has over 20 years of expe-
rience in the environmental indus-
try, both as consultant and as cli-
ent in industry and government.
Rankin has extensive experience
with industrial and municipal
waste management, including: initial project scoping;
waste-stream identification; characterization and inven-
tory development; facility siting and permitting; project
development and FEED; and as project engineer during

MONITOR VISCOSITY SIMPLY

SENSE MIXER MOTOR HORSEPOWER
WITH UNIVERSAL POWER CELL
EASY INSTALLATION PROFILING A PROCESS
No holes in tanks or pipes Power changes reect viscosity changes
Away from sensitive processes Good batches will t the normal prole for
that product
VERSATILE 24
POWER DECREASE
One size adjusts to motors, from 22
SHOWS BATCH
small up to 150hp 20
IS DONE

Works on 3 phase, xed or variable 18

frequency, DC and single phase power POWER

16
SENSOR
SENSITIVE 14

10 times more sensitive than 12

just sensing amps 10 DRY MIX ADD LIQUID

HIGH SPEED LOW SPEED MIXER
CONVENIENT OUTPUTS 8 MOTOR
BEGIN HIGH
For meters, controllers, computers 6
SPEED MIX
4-20 milliamps 0-10 volts 4

0
BATCH 1 BATCH 2 BATCH 3

CALL NOW FOR YOUR FREE 30-DAY TRIAL 888-600-3247

WWW.LOADCONTROLS.COM

Circle 13 on p. 62 or go to adlinks.chemengonline.com/56199-13

Chemical Engineering www.chemengonline.com july 2015 49

 Why Bad Things Happen Part 1

to Good Steam Equipment

Accounting for an entire steam-trap population is crucial to avoiding safety incidents and sub-
optimal production high-priority consideration must be given to steam-system management

H
James R. Risko istorically, steam systems
TLV Corp. have provided the most
effective source of read-
ily conveyable heat to
In Brief industrial process applications, in-
What can go wrong cluding those in the chemical pro-
in a steam system? cess industries (CPI), and there is
no similar low-cost substitute that
Analyzing a steam-
can replace steam. Without steam,
trap population
industrial production would be dra-
management matically curtailed, and the low-cost
prioritization manufactured products that are
Evaluating the made from steams heat or power-
numbers generation assistance would not
exist. Without steam, our quality of
system goals life, economies and society in gen-
creating a new eral would suffer.
paradigm While many CPI workers may ap-
preciate that steam systems are a
necessity, the same individuals may
have also experienced negative
steam-related incidents throughout
their careers, making them harbor
unfavorable thoughts. Specifically,
these events may have resulted in
safety issues, equipment failures
and unscheduled shutdowns of a
unit or a full production line. Safety
events are extremely challenging
and sorrowful issues if someone Figure 1. Effective steam traps keep heat in the system to optimize
production rates and heat quality, and they also discharge condensate to
is injured, and shutdowns can be provide for system safety and reliability. Major problems can occur if con-
disruptive to the entire workforce. densate is not readily drained from the system
It is not surprising, then, that peo-
ple lack an enthusiastic attitude when it ment. Such destruction may render critical
comes to steam. process equipment, such as turbines, flares
or heat exchangers, unusable. Additionally,
What can go wrong in a steam system? high return-system backpressure caused by
What types of issues can arise in a steam steam leakage or blowthrough from bypass
system, and can listing and classifying these steam might restrict production quantity or
items help to determine an executable pre- quality through heat-exchange equipment.
vention path or risk-mitigation procedure? Of Backpressure can also cause the heat-ex-
course, it is relatively easy to identify the most change application to be put on bypass or
common maladies seen in a steam system. to waste condensate by routing to ground
There can be waterhammer, erosion damage level. While some steam-system failures are
and steam leaks in utility systems or equip- common, the challenge then is to identify the

50 Chemical Engineering www.chemengonline.com march 2015

 12,000
10,000
8,000
Number of CDLs
6,000
4,000
2,000
Target monthly replacements
0 replacements
1 2 3 4 5 6 7 8 9 10 11 12
sources of these incidents. This leads to the
larger question, Why do bad things happen
to good steam equipment?
The technical answer lies in understanding
that the cause for a large percentage of failures
might be due to the steam system not being
maintained to the as-built or design speci-
fication. The original designers analyzed the
plant requirements and determined the most
suitable design, according to their expertise
and standards. That original design included
the correct number of steam traps and the as-
semblage of piping and components that help
to drain condensate from the system. Com-
monly referred to as a condensate discharge
location (CDL), the total assembly is required
to remove condensate and effectively maintain
the design performance of the steam system.
Once built, the plant is handed over to the end
user for operation and maintenance with
the expectation of sustaining the initial design.
Unfortunately, that is when operational bud-
gets, personnel turnover and, in some cases,
inexperience with steam systems may play a
causal role in negative events.
Analyzing a steam-trap population
Steam traps (Figure 1) are ubiquitous in
steam systems, and when operating effec-
tively, they can efficiently retain heat in the
system. Consider the analysis of a hypothet-
ical plants steam-trap population shown in
Figure 2. Based on this plants state of the
population summary, it is seen that the orig-
inal design (and as-built) condition included
11,000 CDLs. Over time, the plants manage-
ment decided to decommission and remove
1,000 steam traps from service, leaving only
10,000 in-service CDLs. In this example, no
plant personnel could find documentation
to support why 1,000 steam traps were de-
Chemical Engineering
Figure 2. A typical steam
Design CDLs plant has less in-service
condensate discharge loca-
In-service CDLs tions (CDLs) than the original
design. As total failures are
reduced through monthly
Good
replacements, the number
of good CDLs increases. If
actual replacements are less
than known total failures, car-
ryover failures result, thereby
reducing available good CDLs
to effectively drain the system
in the subsequent year. For
more details, see the calcula-
tion box on p. 26
Total failures actual
Total failures target Carryover failures
Actual monthly
Month
commissioned, because no part of the plant
had been shut down.
When the in-service quantity of CDLs is
lower than the design calls for, then it may
be that the reduction was due to a misun-
derstanding of the importance in maintain-
ing the design total, or possibly from general
neglect of the trap population. If the discrep-
ancy is not explainable, then the systems
drainage is restricted from the originally re-
quired capability. Here, the target for total in-
service CDLs should be increased to equal
the original design total of 11,000 CDLs.
When routinely administered, steam-
trap surveys conducted by plant personnel Figure 3. Accurate and
regularly sustained diagnosis
(Figure 3) provide invaluable information for of steam traps operating
evaluating the health of the steam system. conditions is essential to
As it turns out, in this hypothetical example, determining the populations
several years had passed since the last trap current health (state of the
population). Once failures
survey was completed. So, when the current are identified, the informa-
survey was finished, it was found that there tion is valuable for allocating
were 3,000 trap failures (both hot leaking resources to restore all con-
densate discharge locations
failures and cold-blocked low-temperature to an as designed operating
failures occurred); and 7,000 traps were condition for safety and reli-
considered to be in good condition. This sit- able performance
www.chemengonline.com march 2015 51
 Calculation examples
Estimating annual new failures
(Reported failures Carryover failures) / Years be-
tween surveys
Example: (3,000 reported failures 1,000 carryover
failures) / 2 years = 1,000 annual new failures
Average trap life
(Total in-service population) / (Annual new failures)
Example: 10,000 traps / 1,000 traps failed in a year =
10 year trap life expectancy
uation creates a significant operational and
maintenance dilemma for the site.
If the goal is to have a zero reset sce-
nario, in which all failed traps must be re-
paired, and the cost to repair the average
failure is $600, then $1.8 million is needed
from the years budget to accomplish the
target. To achieve this goal, 250 steam traps
must be replaced each month, which repre-
sents two or three maintenance crews work-
ing full time for a year, assuming that every
trap is accessible and can be isolated for
repair or replacement. It is certainly a monu-
mental task that may have been caused by
an improper course of action a number of
years before, when trap replacements were
not completed in sufficient quantity to keep
pace with the annual failure rate (amount of
failures per year) of the population.
Consider that the average annual failure
rate of a steam trap population in a mature
plant can be estimated from historical re-
cords, provided that there are at least two
survey events within a period of 45 years.
Simply subtract the carryover failures (failed
traps recorded from a prior survey for which
no action was taken to repair) from the re-
ported failures that were recorded in the
survey. The remainder is the quantity of new
failures. The number of new failures can be
divided by the number of years since the prior
survey to provide an estimate of the average
annual new failures of a steam trap popula-
tion. Several related useful calculations are
shown in the box above.
In this hypothetical plant history, if the
management of the trap population had
been continuous and sustainable at an an-
nual failure rate of 1,000 (10%), then the
plant should only have to support the repair
of 84 traps per month, not 250 traps. How-
ever, in this case, the trap population was al-
lowed to deteriorate to 3,000 failures (30%
failure state), and this situation places a tre-
mendous burden on resources. Worse still, if
not corrected, the failures can be expected
52 Chemical Engineering
Annual Failure Rate
(Annual new failures) / (Total in-service population)
Example: 1,000 traps / 10,000 traps = 10% annual
failure rate
State of the Population
(Failures) / (Total in-service population)
(Good traps) / (Total in-service population)
Example: 3,000 failures / 10,000 traps = 30% state
of failure
7,000 good traps / 10,000 traps = 70% good state
to grow until catastrophic events occur.
One takeaway from this example is that for
every 1,000 traps that must be repaired in
a year, there is a requirement for the plan-
ning and repair resources to correct 84 traps
monthly without fail. That replacement re-
quirement equates to repairing 45 CDLs
per day, with any lesser amount creating a
gap at year-end; the difference here equaling
the next years carryover failures.
Steam traps must be replenished or re-
paired in order to maintain a sustainable
operation. Not taking the highest-priority
corrective action with regard to steam trap
failures is somewhat comparable to reusing
a teabag even though it no longer dispenses
flavor although there is a teabag in ser-
vice, it no longer provides a useful purpose.
Similarly, if a steam trap has failed par-
ticularly via a cold failure then the CDL is
no longer serving its intended purpose, and
must be repaired.
Is there any question about what outcome
should be expected if all of the steam traps
and related CDLs in a system were simply iso-
lated by valving, thereby completely remov-
ing their drainage capability? There would be
no way to automatically remove condensate
from the system, creating a highly dangerous
situation. What about if just 30% were shut
off, or 50%? It is akin to gambling with the
safety of the plant, as the potential forma-
tion of condensate slugs in the pipeline can
lead to unstable, hazardous conditions. It is
extremely distressing to consider a site that
is not replacing trap failures due to a bud-
get constraint, because the timely repair of
traps allows the system to operate at the
intended conditions. This must be seen as
an inflexible demand. Timely action should
be mandatory, not optional, to optimize the
operation of a steam system.
Management prioritization
In a properly drained and maintained steam
system, it is critical that the steam flowing
www.chemengonline.com march 2015
 1,600
* 20 lb/h average leak
1,400
1,200
Number of steam leaks
1,000
800
600
400
200
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12
within the system remains at near-
saturated quality and that avoidable
backpressure in the return header is
reduced, in order to diminish the likeli-
hood of waterhammer, erosion, corro-
sion and plating. Superheated steam
or steam with near-saturation quality
normally cannot cause hammer or high
erosion at normal velocities because
there is not enough condensate to be
propelled downstream. Hammer and
erosion occur when liquid pools in the
system are thrust at high speeds, but
when a system is properly drained, the
damaging component is missing. Once
again, it is evident that steam-trap fail-
ure rates are reduced in successfully
maintained steam systems, enabling
efficient condensate removal from
the system.
Maintaining CDLs to manufacturer
specifications helps to eliminate steam
leakage (hot failures) and blocked dis-
charge conditions (cold failures). When
CDL failures both hot and cold
are minimized, insulation is maintained
and boilers are not pushed beyond their
specified limits, the steam system can
be optimized with regards to conden-
sate drainage, as well as the ability to
sustain steam quality when transported
throughout the system.
It sounds simple enough maintain
the insulation, keep the boiler within its
limits and ensure that the steam trap
population provides for quality con-
densate drainage. It certainly sounds
straightforward in concept. If it were
so readily attainable, then why isnt this
goal accomplished more often?
Chemical Engineering
Number of steam leaks 35,000 Figure 4. Often, high priority is given to
fixing the leakage failures or hot failures
Cumulative steam loss to reduce steam loss and increase profit.
Commonly, the repair of hot failures with
30,000
high-value loss is given first priority in
instances other than very critical applica-
25,000 tions. A representative example of the
* 15 lb/h average leak correlation of yearly repairs to reduction
in steam loss is shown here
20,000
Steam loss (lb/h)
* 10 lb/h average leak
15,000
10,000
5,000
Month
Evaluating the numbers
In many plants, the typical program to
manage the steam trap population may
be controlled by budget constraints, by
changing responsibilities or by a lack of
priority. Although plant operations can
be critically affected, the budget may be
controlled by maintenance that requires
close collaboration and coordination
between departments. However, re-
gardless of the circumstance or cause,
it sometimes seems that once a sys-
tem has reached a manageable level
after years of cooperation and dedica-
tion, then the operational problems are
minimized and some portion of the trap
maintenance budget is reassigned to a
different project not related to trap repair.
Then, the portion of the trap population
considered to be good suffers by see-
ing an increase in carryover failures, and
often no survey action takes place for
an extended time. All the while, new fail-
ures are accumulating, thereby reducing
the safety, reliability and performance
of the system. This situation is unfor-
tunate, because the system had finally
reached a relatively sustainable condi-
tion. Subsequently, after several years of
inaction, the plants trap population may
deteriorate so much that a significant
negative incident occurs. At this point,
the follow-up (some might say knee-jerk)
reaction may be to fix the steam system
very hastily. This is an all-too-frequent
scenario that can lead to the previously
presented example where 3,000 traps
failed at a single site.
Once failures have been identified,
then the focus is often placed on fixing
www.chemengonline.com march 2015 53
 Figure 5. For zero reset of 3,000
all 3,000 failures, monthly
replacement of 250 CDLs is
required. If actual replacem-
nents monthly are reduced, 2,500
the gap results in year-end
carryover failures that reduce
the effective good in-service
CDLs in the next year 2,000
Total failures target
Number of failures
1,500
1,000
500
Target monthly replacements
Actual replacements
0
0 1 2 3 4 5 6 7 8 9 10 11 12
the hot failures, because these repairs can
be readily justified by simple energy-cost
analysis. Figure 4 illustrates a typical close
relationship between hot-failure traps and re-
duced steam loss. Especially in times of very
high energy prices, incredible emphasis is
placed on reducing the cost of operations by
fixing leaking steam traps. A progress chart,
similar to the one shown in Figure 4, can be
generated and tracked. Thus, if the project
plan is to eliminate 3,000 failures from the
hypothetical plant, then the corresponding
requirement is to repair 250 CDLs per month,
otherwise there will be a carryover failures
gap (see Figure 5 for additional analysis).
The carryover failures represent the real
world; rarely do plants correct all or even
nearly all of the failures. The result is that
a significant number of CDLs are not re-
stored to proper drainage operation, and it
is not uncommon to carry a sizable number
of failures over to the following year. Once
carryover failures are accepted in a plants
operations, the steam system becomes
destined for sub-optimal and potentially
unreliable operation.
What follows when 1,000 carryover failures
are extended into the following year? Instead
of correcting all 3,000 failed steam traps (a
zero reset mentality), suppose that the plant
management allocated a budget for repair
of only 2,000 failures instead. Perhaps this
thought process stems from the expecta-
tion that with 2,000 failures corrected, the
next annual period will only require budget
for 1,000 failures. However, that is not an
accurate scenario, because those 1,000
CDLs represent carryover failures only, and
the plant also must consider new failures. In
54
Total failures actual
Carryover
failures
Month
a plant with an average steam-trap lifecycle
of ten years, the actual failures could be es-
timated as 2,000, consisting of the 1,000
carryover failures just identified and 1,000
new failures. While the diagnosis of each
trap in the population is performed at regular
intervals (usually annually or semi-annually),
new failures are constantly occurring, as il-
lustrated in Figure 6.
If carryover failures are included as part of
the repair strategy, a significant number of
CDLs will be operating improperly, thereby
increasing the chances of a debilitating in-
cident occurring within the plant. For this
reason, it is not recommended to adopt a
work process that allows carryover failures
and focuses only on those traps that have
already been fixed. Instead, a paradigm shift
is required.
System goals
Applying maintenance action to individual
steam traps is a path action, but not an
overarching system goal. Risk lies in the
false sense of security that is given when
the goal is simply to repair a given number
of steam traps, and the real goal of achiev-
ing an optimized steam system is neglected.
So, if the annual target is established to fix
2,000 steam traps and only 1,000 traps are
repaired, there might be some explanation of
mitigating events that explains the lapse. In
such a situation, the carryover failures might
be considered acceptable under the circum-
stances, and a new goal is assigned for the
next year. However, this is a dangerous situ-
ation, because while the potential for dam-
age is not visible, it is prevalent in the unad-
dressed sections of the steam system, and
Chemical Engineering www.chemengonline.com march 2015
 12,000 Figure 6. There is always an
Design CDLs identification-repair response-
In-service CDLs time lag between diagnosis
10,000 and maintenance action.
Target Although a site may elect to
Zero reset perform sufficient repairs to
state
8,000 reach a target state of the CDL
population at year-end, new
Number of CDLs

Actual
failures occuring during the
6,000 repair period always lower the
actual year-end state. It is just
one of many justifications for
4,000 adopting a zero reset
mentality for maintenance-
response planning
2,000

New failures
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Month

the downstream recipients of that steam. recognized until the next survey.
The potential for peril can increase in sever- The result is that the best state deterio-
ity over time. rates between the prior and new surveys,
Without a full understanding of the long- with the theoretical best sustainable con-
term impact on a steam system, there can dition being realized at the beginning date
exist a false impression that a system can of the next survey. This theoretical point
be well managed, even if there is allowance occurs midway between the prior survey
for carryover failures. Figure 7 provides ad- and the next survey date if annual sur-
ditional insight into a longer-term view. As it veys are conducted, and repair is immediate
turns out, the best possible theoretical state and linear, then the best good state occurs
of the population occurs only after a zero at midyear.
reset condition is experienced from the prior If the original goal of the designers is
survey report, and when accumulated re- recalled, it was to have 11,000 fully func-
pairs equal new accumulated failures. After tional CDLs at the site. For whatever rea-
the midway point, zero reset has been theo- sons, the plants management decided to
retically reached, and new replacements decommission and remove 1,000 installa-
cease, as there are no known failures that tions, leaving a population of only 10,000
remain to be repaired. However, unidentified in-service CDLs (a 9% reduction from the
new failures are still occurring and will not be design). Even if there is perfect harmony
12,000 Figure 7. Sites sometimes
Design CDLs decide upon a fixed number
of annual repairs instead
of striving for zero reset,
In-service CDLs thinking they can achieve a
10,000
* * nearly 100% good in-service
CDL population state over
Best
several years. However, due
Target * Best
sustainable to new failures during the
8,000 repair cycle, the theoretical
Actual
best sustainable condition
is reached midway through a
Number of CDLs

perfectly correlated, linear in-

6,000
spection and repair term, after
which the number of CDLs in
good condition will decline
* Adjustment for
new failures

4,000

Accumulated repairs
2,000
New accumulated failures

0
0 6 12 18 24 30 36
Month

Chemical Engineering www.chemengonline.com march 2015 55

 Figure 8. The number of Design
in-service CDLs is already a
reduction from the number of In-service
CDLs in the original design, Best
which is not desirable unless Best Sustainable
Minimum
clearly justified; and even with
Caution
a zero reset focus and per-
fectly harmonized inspection Warning Cold failures
and repair, the best sustain- Hot failures
Number of CDLs Total failures
able condition is controlled Danger
Danger
by the number of annual Warning
failures. Establishing critical Caution
Minimum
threshold values provides Best sustainable
clear direction for sustainable Best
steam-system performance Design
In-service
3,000
Total failures
2,000
1,000
0
Hot failures
0 1 2 3 4 5 6 7 8 9 10 11 12
Month

with replacement equipment to achieve a new failures are considered, it is possible to

best sustainable state, that already rep- have only 7,000 or 8,000 correctly function-
resents an 18% drop in drainage capability ing CDLs; which in the case of the former,
from the design. At this point, it is also im- represents just 63% of the original design.
portant to consider how much redundancy Starting the next year with only 63% of the
the professional engineering firm included original design total considered to be in good
in the original design. Understanding the condition is dangerous, especially if there is
effect of reducing the number of in-service an additional lapse in repair action. It can be
good CDLs from the design, it can be seen expected that another 1,000 traps would fail
that a key step is to increase the quantity of without any repair of previously failed traps.
in-service CDLs until it reaches the design In such an instance, the portion of the popu-
total, less any trap stations clearly suitable lation functioning correctly for system drain-
for decommissioning. age could be reduced to 6,000 traps only
Furthermore, once carryover failures and 54% of the original design.

Figure 9. A paradigm shift

directs full attention to the
performance-based method,
rather than focusing mainly Best sustainable good
on hot failures. Knowing the
number of good CDLs rela-
tive to key threshold levels
provides a clear indication of Minimum good
the steam systems health and
performance expectations Caution

Warning
Danger
Good
Good
Cold failures

Cold failures
failures
Cold

Good
Hot failures

failures

failures
Hot

Hot

1 2 3 1 2 3
Year Year
Failure-based method Performance-based method

56 Chemical Engineering www.chemengonline.com march 2015

Creating a new paradigm repaired CDL. In some instances at least,
This is where the necessity of a paradigm the engagement and dedication to a steam
shift comes to the forefront. The goal is not system by owner personnel can be higher
to repair failed traps, but rather to maintain than with contractors, so a regular and on-
a minimum threshold of CDLs in functional going audit process is recommended to help
condition. That quantity of good CDLs should obtain sustained high-quality work on such
always refer to the original design total, not critical responsibilities.
to the quantity that are currently in service. With consideration for the site employees
Instead of the focus on hot failures, steam who manage the steam system because
loss or accumulated repairs, a site should responsibilities and personnel change
shift attention to sustaining an acceptable guidelines must be established. In order to
good threshold value for the state
of the population with specific,
strict dates to start the survey every
period. The established survey start
date becomes sacrosanct and is
held steadfast, regardless of daily
interferences. The survey takes
such a high priority simply because Introducing Corzan HP, creating piping systems with better

a steam system is indispensable to

the production success of a safely
operating plant. It should not be a
CORROSION RESISTANCE
second- or third-priority focus, but AND HIGH-TEMPERATURE PERFORMANCE
the key focus, thus helping to en-
sure an optimized steam system. for transmission of CHEMICAL FLUIDS
Another important factor that must
be addressed is outsourcing, which
all made with LUBRIZOL TECHNOLOGY
TO GIVE YOU MORE INSIDE

is the allocation of responsibilities to .

third-party entities as contractors
to perform certain work. In order
to achieve a sustainable program
under an outsourcing scenario, there
are several crucial requirements:
certified test personnel; validated
equipment for making accurate
condition judgments; and standard-
ized application drawings for cor-
rect installation of repaired equip-
ment. In some plants, testing and
maintenance may be conducted by
Corzan HP is a high-pressure, high-temperature, high-impact-strength
the same third-party entity. Regard-
piping system engineered with Lubrizols specially formulated chlorinated
less, whether a singular third-party polyvinyl chloride (CPVC) compounds that meet the ASTM D1784
or multiple entities are involved with certification for 24448 cell class. Its the first pipe that meets
activities that can affect the state of ASTM F441 material classification 4120-06, with a pressure rating
the population, it is important that 25% higher than standard CPVC at 180F (82C).

the capability for each process is Choose Corzan HP for Chemical Processing, Semiconductor & Electronics,
Follow us
confirmed. For example, an incred- on Twitter Water Treatment, Industrial Manufacturing, Mining and Power
ibly capable contractor for piping @LZ_CPVC Generation applications.

repair may be insufficiently trained
for testing activities, and must be
suitably educated and certified for
these tasks. Most critically, there
should exist a regular audit process
to check and confirm the accuracy
of the surveyors qualification cer-
tificates, the actual judgments per-
formed by those surveyors, and in
See how the details on the inside make all the difference on the outside.
Call a piping systems consultant at 1.216.447.7397
or visit corzancpvc.com to learn more.
2014 The Lubrizol Corporation, all rights reserved. All marks are the property of The Lubrizol Corporation.
The Lubrizol Corporation is a Berkshire Hathaway company.
GC 140681

the case of maintenance contrac-

tors, the correct installation of each
Circle 24 on p. 82 or go to adlinks.che.com/56195-24
Chemical Engineering www.chemengonline.com march 2015 57
 implement clear ongoing parameters for safe
and reliable operation, a plant should deter-
mine required threshold target and notifica-
tion values that can be used as the primary
focus for the team responsible for maintain-
ing system performance. Figure 8 shows the
different key threshold levels of good CDLs
as: minimum, caution, warning and
danger. Note that best sustainable is a
theoretical condition for which it is possible
to approach this level, and minimum is the
Circle 42 on p. 82 or go to adlinks.che.com/56195-42
58 Chemical Engineering
determined threshold below which safe and
reliable operation of the steam system can
be adversely affected. Other tiers provide
information to operations and maintenance
personnel on the reasonable expectations
of increased risk, should appropriate action
not be executed.
If a plant wants to have a secondary focus
on cold failures to reduce problems caused
by condensate in the system, or to repair
hot failures to reduce backpressure and
recover profits lost to unnecessary
energy production, that is certainly a
fine approach to fixing failed CDLs.
However, when the big picture is to
achieve safe and reliable plant oper-
ations, the primary focus, as shown
in Figure 9, must be to measure the
number of good CDLs that are drain-
ing the system. Plants should estab-
lish the target of keeping the good
CDLs between the minimum good
and best sustainable good state
threshold values.
When the goal is to maintain a
state of the population at or above
minimum good levels for safety
and reliability considerations, the
target is straightforward. Now,
the repair of failures becomes
just a path, not a goal. With a
clear message to personnel, the
steam-system drainage can be
optimized and sustained for best
plant performance.
Edited by Mary Page Bailey
For suggested additional reading on steam sys-
tems, see the online version of this article
at www.chemengonline.com.
Author
James R. Risko is the president
of TLV Corp. (13901 South Lakes
Dr., Charlotte, N.C. 29873;
Phone: (704) 597-9070; Email:
risko@tlvengineering.com). The
author of more than 30 articles
related to steam and condensate
systems, Risko is active in both
the standards and technical-
writing activities of the Fluid
Controls Institute (FCI), and has previously served as
the organizations chairman, standards chair, and
chair of the Secondary Pressure and Steam Trap sec-
tions. He has earned three energy-management cer-
tifications, from the Association of Energy Engineers,
North Carolina State Univ., and the Institute of Energy
Professionals. He holds an M.B.A from Wilkes Univer-
sity, and two B.S. degrees, in mathematics/education
and business administration/accounting, from Kutz-
town University. He is also licensed as a Certified
Energy Manager (CEM) and a Professional
Energy Manager (PEM).
www.chemengonline.com march 2015
 Engineering Practice

Rotating Machinery: What You Should Know

About Operational Problems
Follow this guidance to improve the operation, safety and reliability of
rotating machinery in chemical process plants

Amin Almasi
Consultant

T
roubleshooting problems in
rotating machinery (such as
compressors, pumps, steam
turbines, gas turbines, turbo-
expanders and more) can present
difficult challenges during day-to-day
operations in facilities throughout the
chemical process industries (CPI).
Reliable operation is an important
factor in keeping plants running, par-
ticularly as plants push to operate at
higher capacities. Lost profits due to
persistent machinery problems can
never be recovered.
Rotating machinery often pres-
ents the most difficult operational
problems in a CPI plant. This article FIGURE 1. Shown here is a steam turbine dismantled for repair. This is the driver of the centrifugal com-
provides a practical, in-depth review pressor
of the art and science of machinery
troubleshooting and problem solv- A systematic approach is the key for proach provides greater value, by
ing. Five case studies are presented solving any machinery operational enabling earlier machinery trouble-
to cover a variety of relevant rotat- problem. Comprehensive machinery shooting through improved early
ing-machinery systems, such as assessments should be carried out detection of unusual deviations. This
compressors, pumps, gas turbines using advanced data-collection tools can provide an earlier indication of
and steam turbines. All of the case and methods (Figure 1). For instance, problems that may be developing.
studies presented here were devel- the use of automated machine- In modern operations, expert op-
oped from actual field experience. data-collection software which erators often opt to manage critical
This article offers a broad working automatically collects a wealth of in- machine components using time-
knowledge of troubleshooting princi- formation specific to the machinery based data trends. Useful trends can
ples and practice, to help operators component coupled with sophis- be developed when the behavior of
gain insight into different techniques ticated modeling techniques, can individual performance variables,
and various methods for machinery provide great insight to support trou- such as pressure, temperature, flow,
problem solving. bleshooting efforts related to com- lubrication oil characteristics and
plex rotating machinery. Such insight more, are plotted over time on a
A systematic approach can help operators to diagnose the graph. These trends can then be an-
Proper CPI plant operation requires actual root causes of underperform- alyzed in conjunction with additional
sustained attention to the following: ing machinery and premature ma- information, such as that derived
Identifying opportunities to im- chinery failures. from alarm screens and plant con-
prove machinery reliability and When CPI operators rely too trol-system graphics. By monitoring
safety heavily on alarms and trips, there is trends, operators can anticipate po-
Developing solutions for chroni- limited time to take remedial action tential problems and propose smart
cally problematic rotating ma- in response to a machinery issue. solutions before any major upset
chines Instead, a proactive monitoring ap- occurs. Trends in relevant data can

74 Chemical Engineering www.chemengonline.com march 2015

support operators and machinery
engineers, allowing them to stay
aware of situations as they begin
to arise, diagnose the root cause
of problems and resolve abnormal
situations. The ability to gather and
analyze trend data can also help op-
erators to compare past and present
behavior, thereby supporting efforts
to anticipate future changes.
The importance of monitoring
In one situation, a CPI company avoided an estimated $10 million in property damage and
business-interruption loss in a year as a result of the early detection of high vibration levels
in its rotating machinery. The high-vibration monitors indicated the deteriorating conditions
(a developing problem) within a large, critical compressor train. The developing problem
resulted in the loss of half of the bolts, a circumferential crack in the coupling spacer (about
80 mm long) and a radial crack emanating from the initial break at the time of shutdown.
If the fault had progressed and all of the bolts had sheared before operator interven-
tion, the compressor would have been unrestrained with potentially disastrous con-
sequences. This clearly illustrates why, particularly in large CPI plants, it is vital to monitor
critical components on a continuous basis.

Keep an eye on key variables
Vibration monitoring is an important
indicator that can help operators to
identify machinery problems at an
early stage. The location of vibration
sensors is vital, to avoid introducing
errors during the measurement. In
general, the best location for vibra-
tion sensors in rotating machinery
is at the bearing or bearing housing
(Figure 2). Vibration measurements
at the machinery casing, in addition
to bearing vibration, can provide
useful data for identifying and solv-
ing machinery problems.
However, monitoring vibration
alone is not sufficient to gauge the
health of rotating machinery espe-
cially when large, critical machines
are concerned. For instance, in a re-
cent case, it was observed that one
electric motors temperature and
current increased when the machine
was overloaded well before any
vibration was detected. This under-
scores the importance of monitoring
other parameters (such as tempera-
ture and current), in addition to vibra-
tion, to allow for early fault diagnosis
of electric machines.
When vibration analysis is used
alone, it may be difficult to estab-
lish the real cause of the vibration.
In general, it is better to monitor
both vibration and overall machin-
ery performance (such as delivered
pressure and flow in pumps or com-
pressors) to obtain a more reliable
fault diagnosis. And, it is important
to realize that the vibration produced
by the machinery is not a fault on its
own but rather a symptom (and
early warning sign) of a developing
failure condition.
Temperature is another critical
variable, and the ability to moni-
tor it accurately is of paramount
importance in CPI plants. Today, a
wide range of temperature sensors
and measurement systems exists,
including thermocouples, resis-
tance thermal detectors (RTDs; also
known as resistance thermometers),
thermistors, filled thermal systems,
pyrometers, infrared thermography
techniques and glass thermometers.
All are widely used to provide
process and machinery temperature
measurements.
The measurement of flowrate pro-
vides another useful indicator for the
health of many processes and ma-
chinery systems. Traditionally, tech-
niques based on differential pres-
sure, turbine and vortex shedding
have been used to measure fluid
flowrate. However, these methods
might not be suitable if an accurate
measurement is required for gases,
since errors can be introduced in the
measurement due the compressible
nature of gases. To use these meth-
ods with gas flows, it would be nec-
essary to compensate for the gas
density fluctuation in order to reduce
the errors.
Modern Coriolis mass flowme-
ters are a more appropriate choice
for measuring gas flowrate, which is
usually independent of temperature,
pressure, density and composition.
This meter uses the Coriolis effect to
measure the amount of mass mov-
ing through the element. In a simple
form, the fluid to be measured runs
through a U-shaped tube that is
caused to vibrate in a perpendicu-
lar direction to the flow. Fluid forces
running through the tube interact
with the vibration, causing it to twist.
The greater the angle of the twist,
the higher the flow. These flowme-
ters have no rotating parts. Modern
Coriolis flowmeters are highly accu-
rate (providing accuracy to around
0.5% of the mass flowrate).
For relatively low-speed rotating
machines, vibration analysis may not
be suitable to detect degradation
or operational problems. However,
acoustic-emission sensors, strate-
gically located on the machine, can
give operators early indication of de-
terioration in the machines operating
condition.
Control valves are vital compo-
nents in every process and in many
types of equipment packages where
fluid is handled and regulated. How-
ever, the most common malfunction
with control valves is internal leak-
age, which cannot be detected eas-
ily. Internal leakage can result from
one of several factors, including an
eroded valve plugs or seats, or insuf-
ficient seat load.
Five case studies
Case study 1: Insects in an air-
compressor inlet filter. In a CPI
plant in a remote area, insects
caused problems for the inlet-air fil-
tration unit of a large process-type
air compressor. The accumulation of
insects on the filters led to high dif-
ferential pressure in the unit, leading
to periodic compressor shutdowns.
Multiple cases of surge and surge-
related trips occurred because of
this issue. Regarding a problem of
persistent insects impacting the air-
compressor, two solutions should be
considered:
1. A multi-sided insect screen. For
example, three-, four- or five-
sided insect screens can be con-
structed (for instance, in a cube or
similar configuration) in the front
of the air filter, with a surface area
roughly three to four times that of
the front face of the air filter inlet.
The rationale is that insects will
gather on the larger-surface area
of the screen (far from actual in-
take) and thus experience less
intake air velocity, allowing them
to fly away, reducing their oppor-
tunity to trap and plug the air filter.
When a screen is built with three
times the area of the air intake,
the air velocity across the insect
screen would be around 33% of
the original velocity. When the

Chemical Engineering www.chemengonline.com march 2015 75


FIGURE 2. Tilting pad thrust bearings are widely used in relatively high-speed rotating machinery. These
bearings are vulerable and can be damaged easily; shown here is one that was removed for repair
screen is built at four times the in-
take area, the air velocity across
the screen would be just 0.25%
of the original velocity.
2. Specially designed pulse-clean
filters. Another possibility is to
install pulse-clean filters that
apply a periodic reverse pulse of
air to dislodge the accumulated
dirt and insects from the filter
inlet at a pre-disposed differen-
tial pressure. This will require a
specialized filter assembly with
a compressed-air supply and
associated equipment and con-
trols. However, this solution is not
usually recommended since this
is a complex, expensive and risky
option.
In this case study, the first pro-
posed solution was used and a
four-sided insect screen was built.
For such an installation, the mesh
size should be selected properly to
stop the smallest anticipated insects
from penetrating, and it should be
reasonably rigid to allow for cleaning
with a soft brush. If the mesh has a
dense or tight weave (say, with holes
around 1-mm dia.) to ensure that
small insects cannot pass through,
then it may be necessary to com-
pensate with increased surface area
(so that air intake is not impeded and
pressure drop is not increased). For
this case, a window screen mesh
(with holes roughly 1-3-mm dia.)
was selected. Routine inspections
by technicians have been required
to clean the screen. The structure of
the insect screen was made rugged
enough to resist wind and ensure
longevity of the asset. This insect
screen provides an added bene-
76
fit, by helping to keep rain and dirt
from being driven into the filter inlet
by wind.
Case study 2: Discrepancies in
performance maps. During the
commissioning stage of a CPI plant,
it was noted that there were two dif-
ferent sets of performance maps in
the manual for a particular, critical
centrifugal compressor one from
the compressor manufacturer and
another from the anti-surge-system
sub-vendor. The compressor had
been delivered three years earlier
but it was not installed and commis-
sioned upon delivery because of a
three-year delay in the project and a
stop in the plant construction. Some
of the engineers who had worked
on this machine and associated unit
had left their jobs and there was no
proper documentation to clarify how
this discrepancy originated.
At this facility, there were concerns
about surge. During the initial period
of operation, the compressor was
tripped several times because of
surge or other operational issues re-
lated to the existence of two different
performance maps and the allow-
able operating range. Any attempt to
identify which performance map and
surge line were the most appropri-
ate for use failed because of person-
nel changes at both the compressor
vendor and the anti-surge- system
sub-vendor during the 3-yr period.
In general, turbocompressors
should be surge tested once they
are installed, to develop accurate,
as-built surge lines. For this machine,
it was decided to perform the surge
test at the site to identify the surge
line and the correct performance
Chemical Engineering
map. This test was completed by
hooking proper surge-detection
hardware into the compressors vi-
bration-monitoring system. At each
operating speed, the compressor
discharge valve was slowly closed.
Prior to the surge, the surge-de-
tection system (using vibration sen-
sors as part of the compressor vibra-
tion-monitoring system) detected an
increase in low-frequency vibration,
which indicated the onset of surge.
This was used to define the surge
line. The compressor was not actu-
ally surged, just taken to the point
of onset of stall. Once the as-built
surge line was identified using this
technique, it was seen to be different
from what both the vendor and the
sub-vendor indicated in their respec-
tive performance maps.
Often, it is common to obtain a
site-tested, as-built surge line that
differs from the vendor-calculated
surge line, or even the shop-tested
surge line, because of different com-
pressor piping and arrangement in
the vendor shop test compared to
the final site installation. The site-
tested, as-built surge line should be
considered the most reliable data for
any operator.
Case study 3: Re-rating a steam
turbine. The steam turbine driver
of a large pump train in a CPI unit
created a bottleneck that prevented
the unit from achieving a desired 5%
increase in capacity. The operation
team asked to run the steam tur-
bine using 49-barg steam instead
of 46-barg steam to achieve higher
power generation from the turbine
and higher pump capacity. Simula-
tions showed that using 49-barg
steam, it was possible to generate
higher power, and this modeling also
showed that the pump train could
handle higher power and generate
more flow within the pump curve. In
fact, this investigation showed that
the new pump operating point was a
bit closer to the best efficiency point
(BEP) than the old operating point. In
general, significant engineering as-
sessments are required in order to
re-rate a steam turbine.
The steam turbine casing in this
case was originally designed and
rated to 46 barg. In order to increase
the pressure rating, the design of
the casing was checked by both
the vendors engineering team and
www.chemengonline.com march 2015
an independent consultant. It was mulated dirt in the air-compressor in order to reduce drag and improve
determined that it could safely and section of a gas turbine can increase the heat rate and power output. Inlet
reliably contain steam at 49 barg. fuel consumption, and lead to more air filters play a significant role in the
The casing was re-rated, after frequent maintenance outages and a degree of fouling that gas turbines
a proper hydro-testing, as per decreased hot-section life. experience. With the original, older-
code requirements. At a given CPI plant, gas turbines style air filters, the operators of this
Using the same rotor, the pressure were used to both generate power plant were faced with a tradeoff: In-
drop per stage was increased. The and drive the gas compressor. The stalling higher-efficiency filters that
existing impellers were checked to original air filters were used in these remove more of the airborne par-
ensure that they were strong enough gas turbine air-inlet systems, which ticles creates a higher initial pressure
and the rotor did not require chang- mandated turbine shutdown and of- drop and typically have a shorter life
ing. Simulations and investigations fline water washing every six months. span compared to lower-efficiency
showed that diametric changes and The dirty air reduced the efficiency filters. The filter pressure drop de-
changes to internal static compo- of the gas turbine fleet between creases the density of the air going
nents within the casing were not cleanings. For instance, the gas tur- to the gas turbine, which reduces
required. The steam turbine was bine trains exhibited efficiency drop the gas turbines output power.
re-rated for 49 barg. The steam-tur- of as much as 2.5% on average in Added pressure drop across the air
bine-driven pump train has enjoyed the three-month period before each filters increases the amount of en-
smooth and trouble-free operation offline water washing. The sched- ergy needed to draw air into the gas
since re-rating after the changes uled offline water washing every six turbine, which increases the gas tur-
were made. months on each gas turbine of the bines heat rate.
Case study 4: Inlet filter for gas CPI plant (just for cleaning), in addi- In general, there are tradeoffs be-
turbines. In general, gas turbines tion to other outages for inspection tween efficiency, cost, filter life and
ingest different materials depend- and maintenance, was wasteful and pressure drop that all need to be
ing on the location and the plant non-productive. The operation team considered when evaluating which
primarily dirt and dust, but also ice, asked for a solution. air filters may be best for a particu-
rain, snow and salt. The cleanliness Getting maximum performance lar operation. The small particles that
of a gas turbine is the key factor in its and reliability out of a gas turbine cause most of gas turbine fouling
efficiency, reliability and safety. Accu- requires keeping the blades clean are less than 3 micrometers (m;

Febru
a
2015 ry
02

www
He .che
Fundam

Two-P at Exchange
men

Written for engineers, by engineers

gonlin
art Fea
entals

rs e.co
ture Re : m
oF

Advan port
HigH-s

in 3-D ces
Hear

Printin
dis

Focus g
perser

o
Housen Industrial
More and more, business in the Chemical Process Industries
s

keepin
Hea

g
t exc

(CPI) is not local, its global. To keep up with this rapidly

Hanger

Facts
s

a
Dimen t Your Finge
sionless rtip

Valv
Numb s:
ers evolving marketplace, you need a magazine that covers it all,
Extrem es for
not just one country or region, not just one vertical market,
Fund
e Serv
ice

High- amentals o but the entire CPI. With editorial offices around the world,
Shear f Chemical Engineering is well-positioned to keep abreast of all
Dispe
page
40

rsers the latest innovations in the equipment, technology, materials,

Vol.
122 No
.2

and services used by process plants worldwide. No other

februa
ry 201

publication even comes close.

5

To subscribe or learn more about membership, please visit

www.chemengonline.com/subscribe
www.chemengonline.com
25430

Chemical Engineering www.chemengonline.com march 2015 77

 sometimes less than 0.5 m) in size,
below the standard capabilities of
old-fashioned filters. Thus, empha-
sis should be placed on comparing
the differences in filter efficiency for
the removal of particles smaller than
3 m (as well as considering filter
efficiency on particles smaller than
0.5 m) since those particle sizes
generally contribute more to
air-compressor fouling rather than
larger particles.
Switching to better, higher-effi-
ciency air-inlet filters is a good way
to keep air-compressors cleaner,
but they can offer some drawbacks.
The most appropriate air filter should
be selected with respect to the ef-
ficiency, the filter life and pressure
There are tradeoffs between efficiency, cost, filter life
and pressure drop that need to be considered when
evaluating which air filters may be best for a particular
operation.
drop. Today, new filter designs can
keep the gas turbine clean with the
same pressure drop and a better fil-
ter life compared to older-style filters.
Todays modern filters have a struc-
ture and layers that allow for the ef-
fective capture of even the smallest
particles (below 3 m and some-
times below 0.5 m) to keep the
gas-turbine intake air clean without
significant filter plugging.
Advanced hydrophobic high-ef-
ficiency particulate air (HEPA) filters
have the same pressure drop as the
older filters. However, the key to their
improved filter performance is a novel
multi-layer media construction that
includes a hydrophobic membrane
layer. Usually the first filter layer cap-
tures most of the airborne dirt. The
middle layer or layers are most
often made from expanded polytet-
rafluoroethylene (PTFE) membranes
or similar, and they stop both sub-
micron particles and water that may
penetrate the first layer. Other layers
are support layers added to provide
strength and burst pressure resis-
tance. These support layers give the
filter more than twice the required
burst strength of the old-fashioned
filters. This extra burst strength pre-
vents filters from breaking before
78
they reach their expected lifespan.
Modern hydrophobic HEPA filters
were installed for all gas turbines
in this CPI plant. They completely
eliminated the need to shutdown the
gas-turbine trains for offline water
washing. The operation has been
satisfactory since the filter change.
Case study 5: Problems in tur-
bines steam-admission valves.
In this case, problems in the steam-
admission valves used to supply the
steam turbine reduced the required
steam flow into the steam turbine,
which impacted the performance of a
critical centrifugal compressor driven
by the turbine, and this negatively
impacted the plants production rate.
There were four poppet valves in the
steam chest of this steam turbine.
These should be staged to open at
different load requirements.
Problems and issues for steam-
admission valves have been re-
ported extensively for steam turbines
in different CPI plants. For any steam
turbine governor or admission-valve
issues, the following initial investiga-
tions are recommended:
1. Study the changes in perfor-
mance of the steam-admission
valve system and governor sys-
tem
2. Evaluate the actual performance
of steam turbine versus the de-
sign specification
3. Investigate the possible causes
for discrepancies
Usually one or more of the follow-
ing issues are reported for steam-
admission valves:
Insufficient nozzle-opening size
Blockage inside the admission
valve system
Incorrect calibration of the admis-
sion valve system
Problem with internal setup of the
steam-admission valve system
Gradual buildup of deposition or
degradation products
Based on the steam-turbine man-
ual for this unit, the governor should
Chemical Engineering www.chemengonline.com
only be opened to around 78% for
the first normal operating condition
case. The investigation showed the
position indicator on the governor
was at 90% for this operating case
(around 12% discrepancy). The initial
suggestion was that the governor or
steam-admission valves might not
be correctly calibrated. The physical
movement of the admission valves
has not changed since commission-
ing, which indicated that the prob-
lem was not related to the gradual
build-up of deposition or to degrada-
tion. A highly nonlinear relationship
was reported at low valve-opening
positions (around 020% of the rated
flow). Below 20% of the governor
position, there was very little steam
flow through the steam turbine. This
indicated that there was a problem
with the first poppet valve.
There was almost zero increase
in the steam flow once the governor
position got higher than 91% of the
rated flow. This was evident from
the sharp drop-off at the end of the
stroke-versus-steam-flow curve of
the admission valve. There was also
a problem with the last poppet valve.
In other words, the lack of response
during the first 20% of governor
opening indicated a problem with
the first poppet valve. The lack of re-
sponse to the final 9% of governor
opening indicated a problem with
the last poppet valve. The steam-
admission valve was overhauled for
quick replacements of the first and
last valves. The steam turbine opera-
tion has been satisfactory after this
valve change. n
Edited by Suzanne Shelley
Author
Amin Almasi is a rotating-equip-
ment consultant in Australia
(Email: amin.almasi@ymail.com).
He previously worked at Worley
Parsons Services Pty Ltd. (Bris-
bane, Australia), Technicas Reuni-
das (Madrid, Spain) and Fluor
Corp. (various offices). He holds a
chartered professional engineer
license from Engineers Australia
(MIEAust CPEng Mechanical), a chartered engineer
certificate from IMechE (CEng MIMechE), RPEQ (regis-
tered professional engineer in Queensland) and he also
holds M.S. and B.S. degrees in mechanical engineering.
He specializes in rotating machines, including centrifu-
gal, screw and reciprocating compressors, gas and
steam turbines, pumps, condition monitoring and reli-
ability. Almasi is an active member of Engineers Austra-
lia, IMechE, ASME, Vibration Institute, SPE, IEEE, and
IDGTE. He has authored more than 60 papers and arti-
cles dealing with rotating machines.
march 2015

Solids
Environmental
Processing
Manager
Chemical
Protective Clothing
ISO 16602 offers a much-needed,
common global language for expressing
protective clothing performance
Susan Lovasic, DuPont
C
hemical protective garments
are available in a wide vari-
ety of fabrics and designs that
provide varying levels of pro-
tection for chemical process industries
(CPI) workers, who may be exposed to
chemicals in the form of vapors, liq-
uids and particles. While other forms
of personal protective equipment
such as safety glasses and hard hats
are highly regulated in the U.S. and
Canada, the performance of chemical
protective clothing is not. An interna-
tional standard for classifying chem-
ical-protective-clothing performance,
however, is available in ISO 16602. The
ISO 16602 Standard (Protective cloth-
ing for protection against chemicals
Classification, labelling and per-
formance requirements) provides an
objective system to appropriately test,
classify and label chemical-protective-
apparel products. This article shows
how ISO 16602 can be used as a tool to
help with the selection of appropriate
chemical protective clothing by provid-
ing an objective means of defining the
performance of chemical protective ap-
parel in situations involving potential
exposure to chemical hazards.
Since, in North America, chemical
protective clothing is not assessed by
a comprehensive standard that quan-
tifies the performance of garments,
safety and health professionals select-
ing chemical protective clothing can
only rely upon manufacturers claims
to choose the appropriate garments.
Unfortunately, the performance of
chemical protective clothing can vary
by manufacturer and there is no es-
tablished means to compare the perfor-
mance of the various garments on the
market. There is also no standard lan-
DuPont
guage in place that describes either the
overall performance of the protective
ensemble or the performance of the
individual components against types
of different chemical hazards. While
some standards do exist for hazardous
materials (Hazmat) emergencies, little
regulation exists regarding routine in-
dustrial workplace applications. This
void in recognized standards for every-
day, industrial-use chemical protective
clothing can lead to confusion and un-
certainty for the safety professionals
who must select the appropriate ap-
parel for a given hazard or application.
Established by the International Or-
ganization for Standardization (ISO),
the requirements outlined in ISO
16602 provide a common language for
the performance of chemical protec-
tive clothing. While ISO 16602 was in-
troduced fairly recently (in 2007), it is
already recognized and accepted inter-
nationally, and is poised to become the
global language for expressing chemi-
cal-protective-clothing performance.
By offering a common, global clas-
sification system and a standard pre-
sentation of the performance data,
ISO 16602 can simplify the selection of
chemical protective apparel by safety
and occupational health professionals.
Standardizing the protective cloth-
ing performance data allows objective
evaluation and comparison of the vari-
ous garments available in the market-
place, rather than relying solely upon
vendors claims. Although ISO 16602
permits a range of performance levels
for a series of key properties, it also
establishes a minimum level of perfor-
mance for each major type of hazard.
ISO 16602 defines performance re-
quirements based on results obtained
Chemical Engineering www.che.com March 2011
from existing test-method standards.
Many companies are multinational
and do not have access to the same
chemical protective apparel in each
location even though they may be
engaged in the same work tasks and
therefore have the same needs for
protective clothing. The ISO 16602
description of chemical-protective-
clothing performance is vital to these
businesses as it provides a common
description that can be used to se-
lect the appropriate, locally available
products. In addition, ISO 16602s ob-
jective performance levels allow safety
professionals a means of identifying
and selecting higher performing gar-
ments if a specific selection of chemi-
cal protective clothing fails to perform
as intended in the work environment
for which it was chosen.
Clothing selection
The first, and most critical, step to
selecting appropriate chemical pro-
tective clothing is the completion of
an accurate hazard assessment. By
identifying the task, the surround-
ings, the chemical(s) and any hazards
beyond chemical exposure, a hazard
assessment provides a basis for un-
derstanding what chemical protective
clothing and other personal protective
equipment (PPE) are necessary to pro-
tect the workers operating in specific
situations. The environment, chemi-
cal hazard and work activity must be
considered in order to be able to se-
lect the most appropriate protective
clothing including the fabric, seam
type and garment design most suit-
able. Without adequate protection
from chemical protective apparel and
proper work practices, the likelihood
51
 Environmental Manager
of a worker being exposed to poten-
tially harmful chemicals increases.
Conversely, overprotecting a worker
with unnecessary clothing can be bur-
densome and result in heat stress, re-
duced field of vision, restricted mobil-
ity and increased physical exertion for
the wearer, as well as possibly higher
purchasing costs for the employer.
Every work situation is unique,
and the ISO 16602 Standard does not
consider all specific hazards that may
be present in the work environment.
Safety and occupational health profes-
sionals need to consider hazards that
are specific to their work conditions
and then look to ISO 16602 to find the
minimum requirements for chemical
protective garments for that type of
situation. ISO 16602 focuses exclu-
sively on typical chemical hazards
and protective clothing requirements
for such hazards so the work situation
may also require additional forms of
PPE, such as footwear, gloves, face pro-
tection, fall protection and respirators.
When evaluating protective clothing
performance under ISO 16602, gloves,
footwear and respirators must be in-
cluded in the whole-garment ensem-
ble testing. It is important to validate
that any additional PPE components
will work properly with the chemical
protective clothing selected for use.
Types of clothing
The ISO 16602 standard is based upon
a series of types and classes. ISO 16602
designates minimum performance lev-
els for six types of chemical hazards.
The type is the overarching system,
with chemical protective clothing fit-
ting into one of the six different types.
The garment type designation is based
upon the physical state of the hazard,
for example vapors, liquids, aerosols
or particles (Figure 1).
When setting the requirements for
each type in ISO 16602, the entire gar-
ment is tested as well as the individual
components. During the whole-garment
tests, a human subject wears the test
garment and accompanying PPE, such
as gloves, boots and respirator. The
human subject is exposed to non-haz-
ardous test chemicals while in an en-
closed chamber performing a series of
movements meant to simulate actual
work activities. This whole-garment
52 Chemical Engineering www.che.com March 2011
testing is used to validate the barrier
performance of the entire ensemble
against a specific type of chemical
threat (gas, liquid or particle). Whole-
garment testing is conducted for each
ISO 16602 type, and the exposure con-
ditions for the tests vary according to
the hazard defined by that type. Note
again that the test chemicals are non-
hazardous and are used to essentially
determine how much of a similar phase
chemical will leak into the suit. The
whole-garment tests do not evaluate
the chemical permeation properties of
the garments. This is assessed in the
class testing portion of ISO 16602.
For example, the Type 3 whole-gar-
ment test is conducted by having the
test subject march in place with ex-
aggerated arm-pumping movements,
while slowly turning, in front of a high-
pressure, high-volume liquid spray
nozzle. This liquid is tinted with a dark
color to be visible if it penetrates the
test garment. Once the high-pressure
liquid exposure portion of the test is
completed, the outer test garment is
carefully removed and an inner in-
dicator garment is checked for evi-
dence of liquid penetration, which is
evident by staining from the dark liq-
uid. Beyond the whole-garment tests,
additional tests are conducted on the
garments fabric(s) and components to
qualify the class performance level for
the garment against specific chemical
challenges and the physical demands
of the task and work surroundings.
ISO 16602 defines different class lev-
els of performance for each type of pro-
tection as described below.
Classes of clothing
Within each of the six garment types,
there are also requirements directed
at the mechanical, barrier and basic
flammability properties of the fabrics
and components used to make chemical
protective clothing. Laboratory tests
are used to determine the mechanical
durability, the barrier against specific
chemical hazards, and ease of ignition
of the garment materials. The results
of these tests will fall into a unique
performance class. Each type within
ISO 16602 specifies a combination
of barrier and durability tests levels,
establishing a minimum performance
class for each of the tests to meet the
specific type requirements. A higher
class rating denotes a higher level of
performance for that property.
The flammability requirements
outlined in ISO 16602 establish a
minimum performance level of flame
spread once the material is ignited; it
does not qualify an ensemble as suit-
able for protection against heat and
flame hazards. Specific evaluation of
chemical protective clothing for heat
and flame protection is not in the pur-
view of ISO 16602. In North America,
relevant standards specifically related
to protective clothing for use near fire
and electric-arc hazards are NFPA
2112 (Standard on Flame-Resistant
Garments for Protection of Industrial
Personnel Against Flash Fire) and
NFPA 70E (Standard for Electrical
Safety in the Workplace).
The class tests are as important as
the whole-garment tests described ear-
lier. Both are integral to evaluating the
overall integrity and expected perfor-
mance of chemical protective clothing.
This integrated rating system aids the
selection of the most appropriate chem-
ical protective apparel for specific work
tasks. Testing the whole garment and
the individual components separately
ensures the sum of the parts is able to
provide the appropriate protection.
Applying ISO 16602
While understanding ISO 16602 is
critical for safety and occupational
health professionals, it must also be
applied consistently and correctly in
the workplace. Consider the task of
opening a flange in a process pipe con-
taining a hazardous liquid under pres-
sure. Even though the pipe should be
depressurized and drained as much as
possible before the flange is removed,
this activity would likely require an
ISO 16602 Type 3 garment for protec-
tion against the possibility that pres-
surized liquid exposure might occur.
The Type 3 whole-garment jet test
would have demonstrated the liquid-
penetration resistance of the ensem-
ble and the performance of the connec-
tions between the garment and other
items of PPE, such as gloves, boots
and respirator. The chemical barrier
performance of the suit material and
components should have been deter-
mined with permeation tests using
 Six types of CPC outlined in ISO 16602
Type 5
Type 4
Type 6
Protection
against
Protection
liquid
Limited against
airborne, aerosols
protection
against solid
particulate
liquid mist
chemicals
Figure 1. Do not assume that the terms gas-tight and liquid-tight refer to abso-
lute barrier performance in chemical protective clothing (CPC). They can oversimplify
the level of barrier provided by such garments. Instead, these terms should be con-
sidered as vapor-protective and liquid-protective, respectively
the specific chemical that is in the pro-
cess piping. The mechanical durability
of the suit against the rigors of the job
would have been assessed with the
mechanical-strength laboratory tests
outlined in ISO 16602. The hazard as-
sessment and experience of the safety
and occupational health professional
would lead to the selection of a gar-
ment with a specific performance class
in each of the barrier and mechanical
tests, plus the type requirement.
Another example can be found in
work situations where exposure to
hazardous particles might be likely
to occur, such as sanding lead-based
paints or mold-abatement work. For
protection against airborne hazardous
particles, a Type 5 ensemble would be
the obvious choice, barring any other
hazard or complications. Most likely,
the garment would be a one-piece cov-
erall with an attached hood, but noth-
ing in ISO 16602 precludes a multiple-
piece ensemble, such as a separate
jacket, hood and pants. ISO 16602
does not specify the design of the en-
semble; it specifies the level of perfor-
mance. As with the previous case, the
nature of the work task will determine
the class of mechanical durability re-
quired for the work activity and envi-
ronment. Note that the use of a res-
pirator might also be required when
working near airborne particles.
Labeling
In addition to delineating performance
standards for chemical protective
clothing, ISO 16602 also sets mini-
mum requirements for the labeling
and documentation for such garments.
Chemical-protective-clothing labels
complying with ISO 16602 must be
permanently attached to the garment
and should include the manufacturers
Type 1
Type 2
Type 3 Gas-tight
Protection
Non
gas-tight
protection
against
chemicals
The Chemical
Engineering
against protection
and vapors
pressurized and toxic
liquid
particles
bookstore offers
chemicals
a variety of
Type indication doesnt tell you everything
about the performance of the CPC. It only
tells that it meets the minimum require-
industry topics
ments of a particular type certification
you will come
to rely on.
contact information, the garment style
and model number, the ISO 16602
type, the date of manufacture, size
and care instructions. With each gar-
ment, the manufacturer should pro-
vide complete instructions that include
safety considerations and limitations,
user instructions and all inspection,
warranty, and maintenance informa-
tion. Since ISO 16602 does not require
third-party certification to claim com-
pliance, the safety and occupational
health professional should request and
t
&OWJSPONFOUBM

examine validation reports provided by
the garment manufacturer that sup-
.BOBHFNFOU


port all performance claims made for "JS1PMMVUJPO
$POUSPM
the chemical protective clothing. Use of
t
&OWJSPONFOUBM

ISO 16602 by garment manufacturers
.BOBHFNFOU

can provide objective information that
8BTUFXBUFS
BOE

enables safety and occupational health
professionals to select the most-appro-
(SPVOEXBUFS

priate chemical-protective clothing for 5SFBUNFOU
their employees.
t
'MVJE
)BOEMJOH
Edited by Dorothy Lozowski
r
(BT4PMJE
BOE


Author
-JRVJE4PMJE

Susan Lovasic is a senior
research associate for Du- 4FQBSBUJPO
Pont Protection Technologies
(5401 Jefferson Davis High-
way, Richmond, Va., 23234; r
-JRVJE-JRVJE
BOE


Phone: 804-383-5060; Email: (BT-JRVJE
4FQBSBUJPO
susan.l.lovasic@usa.dupont.
com). Since 1996, her re-
search efforts have focused r
.BOBHJOH
#VML
4PMJET
on protective apparel applica-
tions to help protect wearers 17792
against chemical, thermal
and fire hazards. In her role, Lovasic has con-
ducted extensive research to assess the critical
For a complete list
properties of protective apparel ranging from
assessment of comfort and durability properties of products, visit the
of protective garments to instrumented, thermal
Chemical Engineering
mannequin fire and electric-arc flash testing.
She has worked for DuPont since 1984 in both bookstore now. http://store.
research and marketing capacities across a vari-
che.com/product/book
ety of businesses. Lovasic holds four U.S. patents
associated with materials used for fabrics and
clothing for electric arc, flame and fire protection.
She is a member of several professional societies
including the AIChE, National Fire Protection
Association (NFPA), American Society for Test-
ing and Materials (ASTM), American Associa-
tion of Textile Chemists and Colorists (AATCC),
and the American Society of Safety Engineers
(ASSE). Lovasic received her B.S.Ch.E. from
Pennsylvania State University.
Chemical Engineering www.che.com March 2011 53
 Propylene Production
via Metathesis
By Intratec Solutions

P ropylene is typically considered a co-product in steam

crackers and fluid-catalytic-cracking (FCC) processes,
which are primarily driven by ethylene and motor gaso-
line production, respectively.
Recently, most U.S. ethylene producers have shifted to
FIGURE 1.

Fresh ethylene

Fresh butene
1
2
3 3'
For
disposal

lighter feedstocks due to the availability of low-cost natural FU

gas and natural-gas liquids (NGLs) from shale. This shift
drastically decreases propylene byproduct output. Since
propylene demand is growing faster than that of ethylene,
and is expected to continue increasing, a gap between pro- (1) Reactor feed treater 4
FU
pylene supply and demand is arising. This brings about the
possibility of establishing chemical processes for on-purpose (2) Reactor feed heater
Lights to fuel
propylene production, and one of the processes depends on (3) Metathesis reactor
metathesis chemistry. RF CW
(4) Regeneration gas heater
The metathesis process (5) Deethylenizer column PG
Metathesis, also known as disproportionation, is a reversible
(6) Depropylenizer column propylene
reaction between ethylene and butenes in which carbon-
carbon double bonds are broken and then rearranged to form 5 6
propylene. Both ethylene and 2-butenes are mainly supplied
from steam-cracker units, but can also be obtained from FCC FU: fuel
units. The metathesis process depicted in Figure 1 is similar to CW: cooling water
ST ST
the OCT (olefins conversion technology) process developed by RF: refrigeration fluid
Lummus Technology (part of Chicago Bridge & Iron Co. N.V.; ST: steam
the Hague, The Netherlands; www.cbi.com) and makes use Heavies to fuel
of a tungsten oxide catalyst, along with a magnesium oxide
co-catalyst. The OCT process typically achieves a propylene yield of lion, the lowest among the regions compared. Germany presented a
about 90%. higher capital investment $295 million but the lowest operating
The process is divided into two main areas: purification and reaction; cost, at about $1,320/ton, compared to $1,480/ton in the U.S. In
and separation. Brazil, both the initial capital investment for a metathesis plant and the
Purification and reaction section. Fresh and recycled ethylene and operating costs were the highest.
butene are mixed and fed to treatment equipment that removes potential
catalyst poisons, such as oxygenates and sulfur. After treatment, the Global perspective
stream is vaporized and heated to the reaction temperature, between Despite the higher operating costs, metathesis plants in the U.S. have
280 and 320C. The metathesis reaction occurs in a fixed-bed catalytic exhibited the highest EBITDA (earnings before interest, taxes, deprecia-
reactor, and butene is commonly employed in excess to minimize even- tion and amortization) margins, since the propylene product prices in
tual side reactions that produce mostly five- to eight-carbon olefins. Coke, North America were higher than those in Europe.
another byproduct of the reaction, is deposited on the catalyst throughout The higher EBITDA, coupled with the lowest total capital investment,
the process. Each reactor can run for about 30 days before requiring makes the U.S. the most promising region for new metathesis projects.
regeneration, where coke is burned off in a controlled atmosphere. However, the current number of metathesis units worldwide shows that
Separation section. Propylene purification is carried out in two col- such units can also be profitable outside the U.S., as shown in the world
umns. The stream leaving the reactor is cooled and sent to the deethylen- map (Figure 2).
izer column, the overhead stream is recycled back to the reactor and Edited by Scott Jenkins
the bottom stream is fed to the depropylenizer column, which produces
polymer-grade propylene in the overhead, as well as a heavies product Editors Note: The content for this column is supplied by Intratec Solutions LLC
stream (four-carbon compounds and greater) that is also recycled. (Houston; www.intratec.us) and edited by Chemical Engineering. The analyses and
models presented herein are prepared on the basis of publicly available and non-
Economic performance confidential information. The information and analysis are the opinions of Intratec
and do not represent the point of view of any third parties. More information about
An economic evaluation of the process was conducted for three distinct the methodology for preparing this type of analysis can be found, along with terms
locations the U.S. Gulf Coast region, Germany and Brazil and of use, at www.intratec.us/che.
is based on data from the third quarter of
2011. The following assumptions were taken FIGURE 2.
into consideration: Each mark in the
A 350,000 ton/yr metathesis unit erected map corresponds
inside a petrochemical complex (all equip- to an existing
ment is represented in the simplified metathesis plant.
The nominal
flowsheet in Figure 1) capacity of each
Storage cost of the main product is equal plant follows the
to 20 days of operation, and storage cost legend below:
for feedstocks was not considered
The excess fuel gas generated in the pro-
cess is considered to be sold to a nearby Up to 149,000 ton/yr
chemical plant at natural gas prices
From 150,000 to 199,000 ton/yr
The estimated capital investment (including From 200,000 to 249,000 ton/yr
total fixed investment, other capital expenses
and working capital) and for such a plant At least 250,000 ton/yr
on the U.S. Gulf Coast is about $245 mil-
 Feature Report
Engineering Practice
Preventing Fires and
Explosions in Pilot Plants
Richard Palluzi
ExxonMobil
Research and Engineering Co
A
mong pilot plants, those in petro-
chemical or petroleum-refining
service pose special risks of fire
and/or explosion. The key strate-
gies to minimize those risks are these:
properly conducted preconstruction
safety reviews; leaktight design and
construction; close attention to area
electrical classification; well-designed
ventilation; well-planned storage;
properly conducted maintenance; and
strictly followed procedures for the
control of change.
Intelligent application of those seven
strategies can be aided by a closer look
at just why petrochemical and refin-
ing pilot facilities are so risky.
They routinely handle significant
volumes of flammable materials,
many at elevated temperatures
and pressures. While most pilot
plants deal with only small volumes
of these materials at a given time,
the overall quantity of flammable
and combustible materials used and
stored in most pilot plant areas is
far from trivial
In comparison with full-scale pro-
cess plants, pilot plants are usually
characterized by a high density of
operating equipment, so the risk to
personnel is significant in the event
of even a small fire or explosion
Pilot plants often involve novel op-
erations, whose nature and hazards
are imperfectly known
Pilot plants frequently switch from
one process operation to another, and
sometimes the scale from one opera-
tion to the next differs by orders of
magnitude, to a greater extent than
is true for a full-scale plant
The processes studied in pilot plants
are typically still evolving and being
modified. Similarly, the equipment,
while admittedly small in size, is
often home-made, one of a kind, and
52 CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006
The issues and complications are different from
and in many ways more challenging than
those for full-scale plants
less tested and/or less robust than
conventional process equipment.
In order to ensure a safe operating en-
vironment, a comprehensive approach
is required. No single activity can en-
sure safety; there is no magic bullet.
Instead, many complementary and
overlapping approaches are needed.
And in addition to the seven strategies
listed at the beginning of this article
and elaborated in the rest of it, behav-
ioral based safety (BBS) efforts and
overall company safety programs and
procedures are also critical; but they lie
outside the scope of this discussion.
Preconstruction safety review
Sometimes called a pre-design or pre-
approval safety review, a preconstruc-
tion safety review aims to ensure that
any proposed design or modification
is safe before a final commitment to
carry it out is made. This review may
often be preceded by other safety re-
views (such as conceptual ones) for
relatively hazardous operations, but
the preconstruction safety review re-
mains the most comprehensive, de-
tailed and critical review.
Properly conducted, it is a long and
detailed exercise, which typically takes
4 to 12 hours to complete. To be effec-
tive, a preconstruction safety review
should be conducted by an experienced
group of trained individuals familiar
with pilot plant design, equipment and
operations, on a predetermined basis.
Each of those three attributes is impor-
tant:
Trained implies that the reviewers
must understand and be able to ef-
fectively apply risk assessment and
hazard analysis methodology
Familiar means they understand
pilot plant design and operations,
are fully aware of how a pilot plant
differs from a full-scale plant, and
are able to address the research-spe-
cific needs knowledgeably
The predetermined basis reflects the
fact that the review requires struc-
ture and organization, both of which
need to be determined and set before
the start.
The review should scrutinize all facets
of the proposed operation, including
startup, shutdown, material charg-
ing, sampling, waste disposal, routine
operation and all conceivable process
upsets. It should examine each com-
ponent of the proposed design, among
them the layout, detailed piping, wir-
ing, and safety-shutdown logic and
implementation.
Every component, from valves to fit-
tings, controllers to indicators, pumps
to analyzers and mixers to flanges,
should be reviewed and examined in
detail for potential hazards. For each
potential hazard uncovered, the team
then needs to assess the risk associ-
ated, and decide whether the proposed
design and/or operation mitigates the
risk to an acceptable level or, instead,
requires additional appropriate cor-
rective measures.
The preconstruction-safety-review
committee should also make sure to
address compliance with all standards
legal, external and companywide
and recommend or approve variances
or deviations where they seem war-
ranted or necessary. The committee
then approves the project advancing
to the construction phase.
Keep in mind that the purpose of
the preconstruction safety review is
to identify areas where the risk is un-
 TABLE 1. PART OF A TYPICAL PRECONSTRUCTION-SAFETY-REVIEW SUMMARY SHEET
Concern Perceived Risk Mitigative Measure(s) Comment Acceptable/
Unacceptable
1. Loss of Cooling Reactor overheats a) Flow Switch FS-02 alarms on Confirm that Rupture Disk Acceptable
Water Pump and leads to runaway flowrates less than 1 gal/min, RD-3 is sized for maximum
PU-07 reaction alerting operators to turn on possible release rate in the
Backup Pump PU-08 event of a runaway reac-
b) High-temperature Alarm T06 tion
opens Reactor Vent Valve V17
2. Loss of nitrogen Build up of flammable a) All electrical components in Add flow switch to nitrogen Acceptable
head sweep components in reac- reactor head space are Class to confirm sweep is turned
tor head space 1 Division 2 Group C & D rated on. If not on, interlock to
prevent heating
3. Addition of Operator exposure a) Operators will wear appro- Develop method to load Unacceptable
solid feed priate personnel-protection without operator pouring, or
equipment (PPE) provide suitable negative
pressure enclosure to retain
any dust

TABLE 2. A TYPICAL PILOT-PLANT RISK MATRIX

Consequence Probability
Frequent Probable Rare Remote Improbable Impossible
Catastrophic Highest risk No risk
Serious
Minor
Negligible Lowest risk
Scenario: Operator pours bag of solid feed while wearing respirator and standing on portable rolling ladder. Bag slips and
operator knocks off respirator while grabbing bag, breathing in dust.
Consequence: Operator exposed to dust for several minutes until he/she can secure bag and reset respirator. Due to toxicity
of dust, this is assessed as Serious.
Probability: Likely to occur several times a year based on experience with other units handling less toxic solid feeds. As-
sessed as Probable.
This would not be considered an acceptable risk.

acceptable and needs to be mitigated.
While the committee conducting the
review may chose to recommend cor-
rective actions when these are obvious
and straightforward, it should not at-
tempt to design the mitigative mea-
sures in the review. Rarely are such
spur-of-the-moment corrective designs
as effective and economical as those
created by the project team later. (Re-
call the old advice about a camel being
a horse designed by a committee.)
A portion of a typical review sum-
mary is shown in Table 1. And a typi-
cal matrix for the associated hazard
analysis appears in Table 2.
Poor practices with preconstruction
safety reviews include the following:
Holding the review before the de-
sign of the pilot plant is complete.
This timing forces the review team
to guess at the details of the final
design, and is always guaranteed to
produce potential problems. At best,
the review ends up wasting time by
working some of the design, or hav-
ing to reconvene later to review the
missing parts; at worst, the review as-
sumes what the final design will look
like, often with unfortunate results.
Instead, a well scheduled review is
held as late as possible in the design
effort, generally just before construc-
tion and materials procurement
Holding the review after construction
has started (or in extreme cases
after it has ended!). This delay
always leads to a tendency to accept
work already completed, or to tolerate
equipment already ordered. The engi-
neer should always keep in mind that
a given design is either acceptably
safe or unacceptably risky based on
an analysis of the hazard (if any), the
potential consequence and the per-
ceived probability, and not on whether
that design is already being installed
and would be costly to modify
Lacking adequate reviewer exper-
tise in a particular area of concern,
such as vessel design, safety shut-
down, or process chemistry. With
this lack, accurate evaluation of the
risks can be difficult, if not impos-
sible. The reviewers either have to
make their best guess as to the haz-
ards, or rely solely on the laboratory
personnels assessment. Review with
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006
inadequate personnel is as likely
to generate a major overkill in the
safety requirements as it is to miss
a key hazard, because the review-
ers recognize their lack of skill and
seek to compensate for it by play-
ing it safe. This state of affairs can
result in requirements that greatly
increase the cost and complexity of
the design and decrease its operabil-
ity, in many cases with little to no
commensurate increase in safety
Inadequate or incomplete analysis,
usually due either to a lack of time
or to a desire to rush the always-
tedious review process. Failure to
allow enough time leads to a rushed
process. Conversely, scheduling one
marathon session almost always en-
sures less rigor as the day drags to a
close. The hours needed to properly
conduct a preconstruction safety
review of a large, complex or par-
ticularly hazardous pilot unit may
best be addressed over several days
or, even more effectively, in several
smaller and more bearable sessions.
Anyone who has ever been trapped
in the eleventh hour of a 12-to-14-
53
 TABLE 3. TYPICAL MAINTENANCE INTERVALS FOR SAFETY
Periodic reinspection and testing of:
Engineering Practice Spring-loaded pressure regulators 12-48 months
Spring-loaded relief valves 6-12 months
Conservation vents 12-36 months
Excess-flow devices 6-12 months
hour review knows that even the
most dedicated individuals start to Unit grounding integrity 12-24 months
think more about their families than Unit vent system 6-12 months
about the hazards. Ground fault interrupter (GFI) operation 1-3 months
Emergency shutdown systems 3-6 months
Leaktight design, Public address systems 1-4 weeks
sound construction Smoke alarms 1-6 months
All pilot plant designs, particularly
Flame detectors 1-3 months
those in petrochemical and petroleum-
Fire extinguishers 6-12 months
refining operations, should provide
thorough containment of flammable Halon or equivalent automatic fire-extinguishing systems 6-12 months
and/or toxic fluids. High levels of con- Inspection, and repair if required, of:
tainment are especially prudent when Mechanical seals 1-6 months
the systems will serve multiple re- Packing glands 1-2 weeks
searchers, because the lack of discrete, Bearings 3-12 months
identified ownership often creates spe- Steam traps 6-12 months
cial problems. (For instance, the ap-
Filters Each run; 2 months
parently higher-than-expected usage
Insulation 6-12 months
of a hazardous component is almost
always attributed to research by an- Unit lighting 3-12 months
other group, rather than to a possible Indicator lights 1-3 months
leak.) And if ignition sources cannot Audible alarms 1-4 weeks
be eliminated or mitigated, extra at- Pressure gauges 12-36 months
tention everywhere else in the design Leak testing of:
is prudent, to lower the probability of Fittings
even a small leak or other release.
Gaskets
Specific components and equip-
Valves
ment- or performance-based specifi-
cations should be employed. Typical Welds
Each run; as required
measures include: Heat-exchange tubes
Requiring sealless valves, pumps Piping and tubing
and mixers Vessels
Requiring welded lines and compo- Pumps
nents whenever practical Compressors
Using heavier-duty components and
Analytical equipment
more-expensive materials of con-
struction Hydrostatic testing of vessels and welded components 1-5 years
Specifying known and tested com- Flushing As required
ponents with a history of successful Checking:
leakfree operation Drain traps filled 1-6 months
Testing unknown or novel compo- Hood ventilation rates 6-12 months
nents or equipment before making
a final decision to incorporate them Purged-enclosure integrity 6-18 months
into the design Pressurized-control room integrity 6-18 months
Once constructed, all pilot plants need Area ventilation rates 1-2 years
to be leak-tested initially, and then at Area ventilation patterns 1-3 years
frequent intervals, the interval for a Emergency equipment storage 3-12 months
given situation depending on the par-
ticular fluid(s) and the nature of the present in the levels necessary). Op- (such as batch reactors) is common in
pilot operation. Leak-test procedures erating personnel should, in fact, be pilot plant work, routine leak-testing of
that are specific to pilot plants can be trained in leak-test procedures and the entire pilot plant is very rare after
found in the literature (for example, should routinely use them. Further- initial construction. Sad to say, virtu-
see Palluzi, Testing for Leaks in Pilot more, the use of the procedures should ally all pilot units, even those well de-
Plants, Chemical Engineering, No- be documented and tracked to ensure signed and constructed and using the
vember 9, 1987, pp. 8185) and they that adequate routine leak testing is proper-quality components, develop
should be adhered to, rather than just completed on whatever frequency or numerous leaks over time. When not
relying on the operators untrained cycle the hazard suggests. located early, these leaks can become
common sense and good judgment While routine leak-testing of fre- more numerous and, therefore, more
(both being valuable, but not always quently disassembled components dangerous.
54 CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006
 TABLE 4. TYPICAL PILOT PLANT CONTROL OF CHANGE REVIEW LEVELS
Change
Change in kind
Change of specific model but pur-
pose and operating conditions re-
main unchanged
Change of purpose but operating
conditions remain unchanged
Change of operating conditions
within pilot plant design parameters
Change of operating conditions out-
side pilot plant design parameters
Process change
Safety system change
Piping or wiring change
Whenever a pilot plant is built or re-
vamped, ensure that the construction
is carried out soundly. All too many
pilot plants are built either by re-
search personnel who are not as famil-
iar with good construction techniques,
or by contractors with little experi-
ence in pilot-scale equipment. Either
alternative is likely to lead to unsound
construction decisions; for example, a
failure to weld a long line that, during
service, will be frequently heated and
cooled, and is not only covered with in-
sulation but also situated in a location
that is difficult for leak-testing. Even
explicit safety issues can arise; for in-
stance, when a section of piping that
must be removed frequently for clean-
ing or charging is fabricated from com-
mon pipe fittings and in such a way
that it is difficult to remove and even
more difficult to routinely reseal.
Proper construction quality also
implies continual and ongoing quality
control, including uniform construc-
tion procedures, appropriate site-wide
training, proper component selection,
proper installation, and ongoing con-
struction supervision. There should
be spot tests of completed assemblies
or components during construction,
the gauging of compression fittings,
and the X-raying of welds and/or the
hydrostatic testing of welded compo-
nents. Equally important is, simply,
insisting that knowledgeable and ex-
perienced pilot plant personnel visit
the construction site frequently. They
Review Level
None if failure was non-hazardous
and routine (such as a clogged filter)
First-line supervisor if failure was non-
hazardous but unexpected
Safety department if failure was haz-
ardous
First-line supervisor or engineering
department
Engineering department or safety de-
partment
Safety department
Safety department and engineering
department
Safety department, and usually an-
other preconstruction review
Safety department, and usually an-
other preconstruction review
Safety department, and usually an-
other preconstruction review
are often able to spot bad practices or
inadequate methods well before these
become major problems.
Of course, problems identified dur-
ing construction, start up or operations
should not be solved and then forgot-
ten. Rather, they should lead to re-
vised internal company specifications/
standards, modified future designs,
improved training and/or appropriate
communications. The amount of hard-
won and costly experience that is lost
each day through not communicating
the lessons learned is substantial in
most pilot plant organizations.
Electrical classification
Proper area electrical classification,
made initially and then maintained
diligently, is a key factor in prevent-
ing fires and explosions, even though it
somewhat raises the cost and difficulty
of pilot-plant construction. Many codes
and standards nominally address this
area. Among them are these:
NFPA-497A Classification of Class
I Hazardous Locations for Electrical
Installations in Chemical Plants
NFPA-30 Flammable and Combus-
tible Liquids Code
NFPA-70 The National Electric
Code
NFPA-496 Purged and Pressurized
Enclosures for Electrical Equip-
ment
API-500A Classification of Locations
for Electrical Installations In Petro-
leum Refineries
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006
Comment
Only if component is identical in all respects.
A switch to a different manufacturer or a dif-
ferent model should require review
Cause of failure should be examined; deter-
mine if it could have led to a hazard
Review all specifications in detail and ad-
dress any differences
Confirm new purpose does not produce new
concerns or modify original assumptions
Checkout of entire safety system (includ-
ing all unmodified parts) is mandatory after
change
API-540 Electrical Installations In
Petroleum Refineries
Sadly, none of these are specifically
intended to address pilot plant con-
cerns, and all of them fail to address
issues that are major for pilot-plant-
scale work. Thus, additional hazard
analysis, consultation with experi-
enced pilot plant safety personnel and
alternate methods will almost always
be needed with respect to electricity-
related safety.
All petrochemical and petroleum-
refining sites should generate and
maintain detailed site standards that
show the area electrical classification
of all pilot plant areas, site-wide. In
general, at least Class I Division 2
construction is prudent for all pilot
plant areas where units are handling
any appreciable amounts of flam-
mable materials. For some research
situations, alternative or special de-
signs may be required in order to meet
the intent of the area electrical clas-
sification. Examples include purging
in accordance with NFPA-496, small
positive-air-flow enclosures similar to
control rooms per NFPA-496, ventila-
tion barriers, equipment separation,
barricades, and special conditions and
equipment to prevent ignition. All
such measures require additional haz-
ard analysis and assessment of their
own, and they usually result in addi-
tional design and construction effort.
As emphasized in next section,
proper ventilation is essential for
55
 FIGURE 1. DUCT DILUTION EXAMPLE FIGURE 2. POTENTIAL ISSUES WITH PILOT PLANT
In order to protect a pilot plant, a combustible gas sensor WALK-IN HOODS
is placed in the exhaust plenum downstream of all the ex-
haust points.

Pilot Plant Area

Combustible Exhaust
gas sensor



Pilot Plant Area: 100 ft x 25 ft = 2,500 ft2
Pilot Plant Area Volume: 2,500 ft2 x 10 ft ceiling = 25,000 ft3

Exhaust Rate: 3 cfm/ft2 or floor area or 7,500 cfm
Combustible gas sensor set for 10% or LEL of hydrogen gas or 0.4%
At 0.4% of 7,500 cfm, there has to be 30 cfm of hydrogen being ex-
hausted (or a standard gas cylinder every 7 minutes) for the detec-
tor to go into alarm at 10% of the LEL.

Assuming one leak, the area around the leak is probably in the
explosive limit for an area of at least 3 ft around the leak (and prob-
ably more like 5-10 ft with the jetting effect).

pilot-plant safety. However, reliance
on a high ventilation rate alone as an
alternative to proper area electrical
classification is risky it can lead
to unanticipated fires and explosions,
because the concentration of flam-
mable material always has to descend
through the explosive range before the
ventilation-induced dilution brings the
concentration below the lower explo-
sive limit. Even when the boundaries
of this explosive range are relatively
narrow, the sparks or other sources of
ignition from nearby electrical devices
(in the absence of proper area electri-
cal classification) can cause a disaster.
An electrically heat-traced line, for
example, virtually always includes
heater elements that are too close to
allow dilution before contact.
In summary, with careful design,
proper area electrical classification
minimizes the risk of explosions and
reduces the risk of fire (usually to a
minor hazard at worst) while accom-
modating almost all research needs
with minimal adverse operational and
economic impact.
Ventilation
All pilot plant and laboratory areas
should be provided with exhaust ven-
tilation. The optimal rate depends on
the operations involved and the tox-
icity of the chemicals handled but,
in general, rates of at least 3 cfm per
square foot of floor area are prudent.
The ventilation rate alone is not suf-
ficient; an appropriate ventilation pat-
tern (directional sweep) is also neces-
sary. For instance, take care to ensure
that the flow is not disrupted by large
openings, such as overhead doors or re-
movable panels, or by newly installed
equipment that takes up a lot of space.
For pilot plant areas, both high- and
low-point ventilation are required, to
capture and exhaust light and heavy
gases alike.
Smoke tests are often needed, to
confirm that the design is ventilating
properly. This is especially true after
bulky new equipment is installed, or
in difficult-to-assess situations.
Most pilot-plant and laboratory areas
should also be continuously monitored
via fixed sensors for combustible and/
or toxic gases. This monitoring should
alert personnel to the onset of a poten-
tial hazard and shut off the hazardous
feed at the source until the problem is
located and corrected.
Typically, monitoring takes place in-
side the pilot plant area but outside of
any hoods or ventilated enclosures; in
other words, it takes place where the
operators are likely to be. Monitoring
inside the ventilation duct should be
avoided. Sensors placed there are dif-
ficult to maintain and are subject to
problems with dirt and dust. More sig-
nificantly, they can be prone either to
numerous false trips or to a failure to
detect anything short of a catastrophic
release, due to dilution (see Figure 1 for
an example related to this point).
In addition to general ventilation,
local exhaust should be provided
where needed, such as locations that
involve sampling, vessel filling, ves-
sel draining, product collection, areas
of relatively high leak potential (such
as seals), equipment that needs to be
removed or opened frequently (such
as reactors, filters, or feed cases) and
equipment requiring frequent mainte-
nance. Ventilation employing fixed en-
closures and flexible arms/hoses alike
should be used to capture any emis-
sions or leakage at the source before it
becomes a hazard.
Equipment that handles very toxic
materials may need to be placed
wholly or partially in hoods or simi-
lar ventilated enclosures that meet
ASHRAE 110 and AIHA ventilation
requirements.* This requirement in-
volves significant design work, both
for the hood itself and to ensure that
the pilot plant layout is not keeping
the hood from functioning properly.
Figure 2 shows potential problems
related to hoods. Operator training
should address the safe use of large
hoods. And an ongoing maintenance
program is required to ensure that
hoods continue to function properly.
Storage, maintenance
To safely store hazardous materials
in pilot plants, full compliance with
NFPA-30 as well as local building/fire
codes is required; and in some cases
involving highly flammable or explo-
sive materials, it is prudent to go even
beyond those code requirements.
Gas cylinders should not be placed
inside pilot plant areas; instead, they
should be stored outside or in sepa-
rate cylinder storage areas. Solvents
should be stored in flammable storage
cabinets whenever possible; and not
just if required by code. Try to keep
only small amounts of hazardous ma-
terials on hand; the quantities that
operators or engineers declare to be
necessary for operations should be pe-
riodically challenged and reassessed.
Routine maintenance of all safety
*ASHRAE is the American Soc. of Heating,
Refrigerating and Air-Conditioning Engineers;
AIHA is the American Industrial Hygiene Assn.

56 CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006

equipment and systems is an inte- needs. A procedure should be in place tions are reviewed before the work is
gral part of keeping pilot plants safe. to mandate reviews for all changes started. Craft workers and supervisors
All safety systems must be checked and modifications no matter how should be trained not to commence a
on a frequency stipulated during the trivial. The mandated level of review job if the required approvals have not
plant design. In addition to the usual can vary with identified hazard and been obtained. And safety-system ac-
OSHA-mandated (in the U.S.) work assessed risk, as shown for example in cess for any modification or disabling
permit for maintenance, lock-out/tag- Table 4. Keep in mind that procedural should be tightly controlled as well as
out programs and confined-space and operational controls alike are closely monitored.
programs, consider setting up site- needed, to ensure that all modifica- Edited by Nicholas P. Chopey
wide programs to automatically and
routinely maintain safety systems.
Common examples include programs
to ensure that all relief devices are
properly selected, sized, installed, and
automatically scheduled for routine
maintenance. All gas monitors should
be routinely calibrated at the interval
specified by the manufacturer, typi-
cally monthly to quarterly. No opera-
tor action should be required to initi-
ate these tracking, maintenance and
inspection operations; they should
be handled automatically. Examples
of other pilot plant maintenance op-
erations, with common intervals, are
shown in Table 3.

Control of changes
A strong control-of-change procedure
is also a key to preventing fires and
explosions in pilot plants. Even more
than full-scale process plants, pilot Objection #1: We Can Do Our Own Pressure Relief Audit
plants routinely involve numerous
and frequent modifications to accom-
modate the ever-changing research Are You Asking Too Much of Your Engineers?
Crude oil and refined products prices are up and your process engineers
are focused on capital projects that maximize returns at your operating
Author
Richard Palluzi is a Distin-
facilities. Taking them from essential tasks to perform a relief system
guished Engineering Associ- design audit is not a good use of your most valuable resources.
ate at ExxonMobil Research
& Engineering Co. (1545
Route 22 East, Annandale, NJ With thousands of audits under our belt, Berwanger is the leader in
08801; Phone: 908-730-2323),
responsible for the design, verifying and documenting relief systems. Our patented, equipment-based
construction and support of approach and Pressure Protection Manager software anchor our Best
pilot plants and laboratories
for the firms research site Practice, which is widely used by the worlds largest operating companies
in Clinton, N.J., and con- and recognized by regulatory agencies. And, our fixed-price (no change
sulting on pilot-plant issues
throughout ExxonMobil worldwide. The author orders) implementations reduce risk and drive innovation and flexibility.
of 2 books, 30 articles and 40 presentations on
all phases of pilot plant and laboratory safety
and operations, he is a member of AIChE, ISA,
the American Soc. of Safety Engineers and the For more information, call toll free
National Fire Protection Assn., and was recently Download a free whitepaper
1-800-658-8809
elected to the last-named organizations commit-
tee on NFPA-45 Fire Protection for Laboratories Code, Law and Safe Operation
Using Chemicals. He teaches three pilot-plant or email us at
training courses for AIChE, and is a past chair www.berwanger.com/reliefsystems/ce05
PressureReliefinfo@berwanger.com
of the AIChE Pilot Plant Committee, as well as
of the Clinton site's Safe Operations Team. Also,
he currently chairs ExxonMobil's Pilot Plant
and Laboratory Safety Standards Committee.
Palluzi has consulted for both the U.S. Dept. of UPSTREAM PRESSURE RELIEF ENGINEERING INFORMATION MECHANICAL PHA / LOPA / SIL
Energy and Dept. of Defense on research-related SERVICES SYSTEMS MANAGEMENT MANAGEMENT INTEGRITY
issues. He holds Bachelor of Engineering (1974)
and Master of Engineering (1976) degrees from
Stevens Institute of Technology in chemical en- 2006 Berwanger Incorporated. All rights reserved. Pressure Protection Manager is a registered trademark of Berwanger Incorporated. berwanger.com
gineering.
Circle 32 on p. 58 or go to adlinks.che.com/5829-32
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006 57
 Feature Report

Fire & Gas in Safety Systems

Switchgear "MBSN
MJTUT

MCC BOE
BMBSN
EJTQMBZ

Fire detection
Notification
r
4NPLF BQQMJBODFT
LOGIC
r
)FBU CFBDPOT
SOLVER
r
'MBNF TPVOEFST

r
.BOVBM
DBMM
Fire and gas
point
DPOUSPMMFS
Fire
TVQQSFTTJPO
or
Gas detection TZTUFN
r
5PYJD 'JSF
BMBSN

Integrating fire-and-gas r
'MBNNBCMF
r
"DPVTUJD

QBOFM
Safety
TZTUFN
detectors and mitigation
MFBLBHF
r
.BOVBM

TIVUEPXO
TZTUFN
systems into overall process
DBMM
QPJOU

safety control can help ensure Figure 1. A typical F&G safety system comprises detection, logic
control, and alarm and mitigation functions. The logic solver is the

fast responses to emergencies central control unit of the overall F&G detection and control system.
The controller receives alarm and status or analog signals from field
monitoring devices required for fire and gas detection. The controller
Scott Hillman handles the required actions to initiate alarms and mitigate the hazard
Honeywell

B
y continuously monitoring for
abnormal situations, such as a
fire, and combustible or toxic
gas releases, fire and gas (F&G)
detection and mitigation systems are
the linchpins to maintaining overall
safety and operation of industrial fa-
cilities. The premise of these systems
is to protect processes and the envi-
ronment by providing early warning
of incidents and enabling actions that
prevent them from escalating. Also,
implementing an integrated F&G
strategy based on the latest automa-
tion technology, the chemical process
industries (CPI) not only meet pro-
tection requirements, but also have
potential to enhance business success.
F&G safety awareness
In December 2005, an explosion at the
Buncefield fuel terminal in Hertford-
shire, U.K. ignited millions of gallons
40 Chemical Engineering www.che.com May 2009
of fuel. Water and land quality were
adversely affected across southern
England, resulting in a clean-up ef-
fort that cost hundreds of millions of
pounds. Such incidents illustrate the
potential threats to the worlds oil-
and-gas supply chain and other vital
process industry resources. They also
serve as a critical reminder that, due
to the very nature of the business and
the products associated with it, the
CPI is inherently fraught with risks.
These risks range from raw mate-
rial and intermediate toxicity and
reactivity, to energy release from
chemical reactions, high tempera-
tures and high pressures. With all
too frequent news reports of cata-
strophic incidents occurring at man-
ufacturing facilities across the globe,
safety has never been a more scru-
tinized issue than it is today. Com-
pounding the already high visibility
of this issue is the state of the global
economy; manufacturers simply can-
not afford the monetary losses com-
monly associated with plant safety
incidents.
Amid this backdrop, integrated
process control and safety systems
have emerged over the last few years
as effective solutions for providing
safer working conditions in plants. In
providing operators better visibility
across process and safety, CPI compa-
nies are realizing that integrated sys-
tems can reduce costs by minimizing
equipment damage and incidents that
impact people and the environment.
At the same time, the reflection on
corporate image is positive.
Elements and integration
International standards organize
safety implementation under a series
of six protection layers:
 Plant evaluation system/
Plant security center,
(normally not in F&G scope)
Ethernet
Digital
video server
Digital
video cameras
Addressable fire detectors,
modules, manual call points
Non-process manned buildings
Figure 2. A good F&G system combines state-of-the-art fire and gas detectors, conventional and analog addressable fire
panels, clean-agent and inert-gas fire suppression systems, and an SIL 3 certified fire-and-gas logic solver into a consistently
designed and executed solution. An integrated system provides common tools, operating interface and networking, resulting in a
common platform with independent systems
Inherently safe plant design
Process control systems that are
safe and secure
Automatic safety, security and
mitigation systems
Work procedures
Alarm systems
Mechanical protection systems
Acting as the prevention safety layer,
the safety shutdown system takes
automatic and independent action to
prevent a hazardous incident from oc-
curring and to protect personnel and
plant equipment against serious harm.
Still, in reality, incidents unfortunately
escalate far beyond this layer.
This problem is the main reason
that F&G systems should remain at
the core of any integrated solution.
Used for automating emergency ac-
tions with high-integrity safety and
control solutions, F&G systems act as
the mitigation safety layer that takes
action to lessen the consequences of
hazardous events. They are also criti-
cal to resuming full production when
recovering from these incidents.
A good F&G system combines state-
of-the-art detectors, conventional and
analog addressable fire panels, clean-
agent and inert-gas fire suppression
systems, and a safety integrity level
(SIL) certified logic solver (Figure 2).
F&G operator
station
Plant wide
historian
Serial interface
Battery
backup
Addressable
fire alarm panel
F&G detectors, manual call points
The logic solver acts as the central
control unit (Figure 1), which receives
alarm and status or analog signals
from field monitoring devices such as
detectors or manual pull stations, ini-
tiates alarms and mitigates the haz-
ard. Correct and proven connection
of detectors to plant safety systems
is an important factor in reliable per-
formance of the F&G system and for
establishing the desired SIL.
In the past, a proprietary F&G sys-
tem was standalone or a hardwired
mimic overview panel that was linked
to a control system. With these older
systems, fire-control measures had to
be manually activated, which was far
from an ideal practice. Today, however,
F&G detection systems are gener-
ally programmable electronic systems
(PES) with high safety availability and
mitigation effectiveness. As modern
F&G systems are tightly integrated
with the overall process safety strategy,
mitigation is either enacted through
an emergency shutdown (ESD) system
or directly by the F&G system itself.
Using these new techniques and add-
ing intelligence to instruments to re-
duce the number of spurious alarms
has greatly improved detection rates.
Thanks to these advances in F&G
detectors, the systems can detect early
Chemical Engineering www.che.com May 2009
DCS/SIS operator
station
F&G mimic
panel
Fault-tolerant
ethernet
DCS
F&G logic solver ESD safety
safety manager manager
F&G
Sounders, beacons mimic panel
Process unit areas
warnings of explosive and health haz-
ards, including combustible and toxic
gas releases, thermal radiation from
fires and minute traces of smoke
in sensitive equipment enclosures.
They also provide audible and visual
alarm indications to ensure that op-
erators and personnel are informed
of hazardous situations. With these
improved detection capabilities, F&G
systems automatically initiate execu-
tive actions, such as deluge systems
and evacuation procedures. This
minimizes the escalation of safety in-
cidents and protects personnel, prop-
erty and the environment.
Further value can be realized an
F&G solution is plugged into an inte-
grated system that provides common
tools, operating interface and network-
ing, resulting in independent systems
that are tied together on a common
platform. A recommended approach
is integration at the controller level,
which provides: plant-wide safety in-
strumented system (SIS) point data;
diagnostics and system information;
alarms and events, operator displays;
and sequence of event information to
any station. This minimizes manual
intervention and drastic plant shut-
downs, reduces hardware costs, and
allows plants to recover more quickly
41
 Feature Report
and easily from process upsets and ab-
normal situations.
Modern safety systems, when seam-
lessly integrated with the plant au-
tomation system through a secure
communication network, will transfer
alarm signals, fault and trip signals
and system diagnostics. Information
from all related systems can be man-
aged from the same location, enabling
that additional layer to monitor the
status and operability of the total
F&G detection and control system.
Overall SIS technology integrates
safety measures dispersed throughout
a plant to reduce risk to employees
and assets, increase process availabil-
ity, and improve regulatory compli-
ance. SIS solutions can be integrated
with F&G detectors for increased pro-
tection, and unified with third-party
systems to reduce validation and ac-
ceptance testing costs.
Combined with the new generation
of F&G solutions, an integrated sys-
tem provides alerts of abnormal situ-
ations in a fast, accurate and struc-
tured way, giving personnel time to
decide upon the correct course of ac-
tion. These solutions, which include
new integration capabilities with
process-simulation tools, F&G detec-
tors and control communication pro-
tocols, enable safety engineers to de-
sign large integrated and distributed
plant-wide safety strategies. With in-
novative simulation solutions, safety
engineers can easily test the impact
of safety strategies on the overall
plant design and operations before
implementation. This reduces overall
risk and the impact of system modifi-
cations and ultimately increases prof-
itability by bringing new plants into
full production much faster.
In addition, new field-device con-
figuration tools allow plant personnel
to automatically configure intelligent
safety devices and integrate them
into the control system database.
Facilities subsequently save money
by using a single tool to manage all
equipment assets.
Toward integrated systems
As more plant owners move toward
highly integrated plant-production
systems, appropriate integration with
multiple interdependent yet interre-
42 Chemical Engineering www.che.com May 2009
Figure 3. Industrial operations benefit from a holistic approach to safety that
supports a secure process control network to the perimeter of the plant to protect
people, assets and profitability. A layered safety strategy encompasses process and
system technology and the people who interact with that technology to help
plants achieve their safety objectives
lated layers of protection (Figure 3)
plays an important role in improving
safety and efficiency.
The F&G system should have com-
munications integration with the
plant distributed control system
(DCS) in order to have F&G graphics
and alarms displayed to the operator.
However, there also should be inde-
pendent displays, such as independent
human-machine interfaces (HMIs),
for plant operators to respond to F&G
excursions when the DCS HMIs are
not available. The plant F&G system,
with a fire system for occupied build-
ings, should also be integrated with
the plant evacuation and site security
center for efficient plant-evacuation
procedures. This enables plant man-
agers to keep better tabs on personnel
and efficiently coordinate with first
responders during emergencies.
As part of an overall plant-safety
strategy, end users need a unified
platform for emergency shutdown and
F&G detection. A single window for
operators and a common tool for engi-
neering and maintenance drives down
operational risk and costs.
Integrated control and safety sys-
tems (ICSS) provide multiple benefits
to process plants. They help opera-
tors minimize intervention and shut-
downs, and recover more easily from
process upsets. They also allow facili-
ties to reduce hardware and installa-
tion costs, and ensure easier system
configuration with preconfigured
function block selections.
Plants implementing an ICSS plat-
form for F&G, ESD and DCS systems
can significantly lower their opera-
tion and maintenance costs, and in
Secure process control
Asset management and protection
Abnormal situation management
Effective operating environment
Boundary management
Emergency shutdown system
Physical protection
Emergency response
many cases, reduce overall wall-to-
wall project costs by 25%. Seamless
integration with the ESD and DCS
through a common network proto-
col provides a safe landing in case of
emergencies and eliminates the need
for additional equipment or engineer-
ing. Integration of fire detection and
security systems for offsites and utili-
ties with the plant automation infra-
structure further improves operator
efficiency, through single-window ac-
cess for alarm visualization, diagnos-
tics, and events and historians.
These truly integrated safety sys-
tems deliver several benefits, includ-
ing the following.
Integrated operational interface
Integrated peer control
Integrated diagnostics
Integrated postmortem analysis
Integrated F&G system
Integrated power supplies
Integrated modifications
Integrated simulation and
optimization
Operational integration provides
a seamless interface to the process
under control, and at the same time,
maintains safe separation. From an
operational perspective, it makes no
difference where the application is
running. All required information is
available to the operator. This allows
applications ranging from rotating
equipment and compressor-protective
systems, to emergency shutdown sys-
tems and large plant-wide F&G ap-
plications to be monitored from any
operator console.
Additionally, industrial operations
benefit from a holistic approach to
safety that supports everything from a
 Review
potential F&G incidents as
part of risk assessment
Define
role required of F&G
system in risk reduction
Select
appropriate system type
Set
performance standard
Develop
specification test, mainten-
ance, inspection procedure
Implement
performance standard
Figure 4. An effective project
strategy starts with an assessment
of future or existing fire and gas per-
formance according to functional safety
standards. Based on this assessment,
end users have a detailed roadmap for
installing new equipment or updating
obsolete infrastructure to an optimal
level of safety. The main automation con-
tractor can help to identify F&G hazard
points and possible risks, and develop
basic design packages and related ac-
ceptance test criteria to meet safety
requirements
secure process control network (PCN)
to the perimeter of the plant. This lay-
ered safety strategy encompasses pro-
cess and system technology and the
people who interact with that tech-
nology to help plants achieve their
safety objectives.
A layered safety strategy unifies
all plant protection layers (including
basic control, prevention and mitiga-
tion, as outlined in IEC 61511 stan-
dard) required for achieving optimum
functional safety. It also provides the
required functional safety with a high
SIL. This includes superior visualiza-
tion and logging facilities enabling op-
timal operator response and accurate
evaluations. By integrating basic con-
trol, prevention and mitigation com-
ponents, a CPI company can vastly
reduce its overall project costs and
ongoing maintenance expenses.
At the core of a layered safety strat-
egy is process design the embodi-
ment of the business, safety and pro-
duction considerations necessary for
effective operations. At the next layer,
the layered approach implements
Integrated safety in oil and gas
L
ike other process industry operations, oil and gas terminals present difficult challenges
for automation and safety technology. Tank farms, storage areas, and loading and un-
loading operations all require F&G and safety systems to protect personnel, assets and
the environment. The consequences of incidents at oil and gas terminals, as illustrated at the
Buncefield fuel terminal, can be enormous (Figure 5).
In oil and gas terminal applications, operators not only need overfill protection but also an
integrated fire and life safety system allowing proactive response to alarms and events and
a single realtime view to any potential threat. Industrial plants have procedures and safety
systems designed to bring operations to a safe state in the event of equipment malfunctions
and other operational problems. In the case of a significant security incident, an integrated
system can activate these same procedures and systems. An integrated system also leads to
less expensive implementation and maintenance, since all components work together.
An integrated fire-and-life safety solution for terminals typically includes high-high tank
alarms and loading system interlocks (compliant with IEC 61511) that prevent overflow
through shutdown of pumps. Likewise, it incorporates the latest-technology gas and vapor
detectors, integrated with the overfill shutdown system, to detect incidents before they esca-
late. The system will mitigate safety incidents by ensuring that all personnel are informed of
hazardous situations in a clear and unambiguous manner, and providing fast and efficient
response to associated risks.
An effective terminal automation safety and security solution will not only include overfill
protection, but will also integrate security access control, personnel mustering systems and
video monitoring to reduce the possibility of unauthorized access or intrusion. Integration of
technologies such as security biometrics and wireless mesh networks enhance the operation
and the lower cost of implementation of these systems.
tools and procedures for managing safety lifecycle as outlined in safety
abnormal situations and reducing standards such as IEC 61511. An in-
incidents. When an abnormal situa- tegrated main automation contractor
tion occurs, alarm management, early (I-MAC) can help identify F&G haz-
event detection, and abnormal situa- ard points and possible risks, and then
tion management (ASM)-designed develop basic design packages and re-
displays ensure operators have the lated acceptance test criteria to meet
information available in the context safety requirements. This results in
they need it. This enables faster reac- optimized risk reduction and opera-
tion to hazardous situations. tional performance, better compliance
Next, properly designed emergency- with safety standards, and increased
shutdown systems and automated lifecycle sustainability.
procedures can move a plant to a safe Supplier assistance can extend
state in the event that an incident es- to implementing SIS solutions; live
calates beyond the inner sphere of pro- hot cutover, implementation and ex-
tection. Should an incident occur, F&G ecution of revamps; and installation,
detection solutions, coupled with rapid commissioning and safety validation.
location of individuals and a carefully To sustain the end users F&G sys-
designed emergency response proce- tem performance, leading automa-
dure, will help contain the impact. tion contractors also provide lifecycle
Finally, a layered approach to safety support services that include periodic
protects the perimeter of the plant proof testing; system maintenance;
using physical security that safe- training programs on safety, code and
guards access to structures and moni- standard compliance; and spare parts
tors traffic approaching the facility. management.
When designing a truly integrated
system, an effective project strategy Challenges and the road ahead
starts with an assessment of future Industrial standards such as IEC
or existing F&G performance accord- 61511 (ANSI/ISA S84.01 in the U.S.)
ing to functional safety standards play a large role in developing, imple-
(Figure 4). Based on this assessment, menting and installing F&G systems.
end users have a detailed roadmap for The overall safety lifecycle model de-
installing new equipment or updating scribed in the IEC standard lists all of
obsolete infrastructure to an optimal the necessary project activities, from
level of safety. This process begins the concept (definition) phase to the
with a hazard and risk assessment decommissioning phase, necessary to
(such as HAZOP) and then contin- ensure the functional safety of equip-
ues through the various steps of the ment under control (EUC). These ac-
Chemical Engineering www.che.com May 2009 43
 Feature Report

tivities can be divided over a wide

range of categories, such as proce-
dures, documentation, testing and
validation, planning, hardware
and software development, and
risk assessment.
The IEC 61511 standard con-
cerns the determination and devel-
opment of risk-reduction measures
(RRMs) required as the outcome of
an EUC risk assessment. The basic
principle of risk assessment is that
all potential risks to the EUC are
identified and analyzed. This in-
cludes calculating the probability
of each potential EUC hazard and
determining the risk reduction Figure 5. Like other process industry op-
erations, oil & gas terminals present difficult
measures required to achieve an challenges for automation and safety tech-
acceptable SIL. This is defined by nology. Tank farms, storage areas, and load-
the following expression. ing/unloading operations all require fire and
Risk = Probability x Consequences gas and safety systems to protect personnel,
The prescribed reduction mea- assets and the environment. The conse-
quences of incidents at oil and gas terminals
sures either decrease the risk can be enormous
probability (for example, ESD sys-
tems), or mitigate their consequences engineers to design and build stand-
(for example, F&G systems). The risk alone safety applications and distrib-
of EUC hazards can be reduced by a uted plant-wide safety topologies.
combination of several RRMs, where Plants must find ways to improve
each measure handles one part of the F&G system effectiveness through
total required risk reduction factor optimization of F&G detector cover-
(RRF). age, system safety availability and
Recently, there have been discus- mitigation effectiveness. At the same
sions regarding whether F&G detec- time, they must reduce the cost of
tion systems should contribute to risk ownership for safety equipment.
reduction, or be considered only as a Many facilities are also dealing with
protection for the installation. The the cost of upgrading and refurbish-
ISA technical report TR84.00.07, pro- ing existing, non-integrated F&G
viding guidance on the evaluation of systems. Compounding this issue is
fire-and-gas system effectiveness, is the fact that plants must meet their
currently in draft review. safety needs while ensuring opera-
Additionally, a growing number of tional and business readiness at proj-
regulations have drawn more atten- ect startup. Faced with this reality,
tion to process safety and the role they seek the lowest risk and high-
played by F&G systems. Manufactur- est value protection from their safety
ing plants must cope with business system and F&G technology.
challenges ranging from increased ac- Edited by Kate Torzewski
cident, incident and insurance costs, to
compliance with strict standards and
codes, such as NFPA, API and OSHA Author
in the U.S., and BS EN and SEVESO Scott Hillman is the global
solutions director for Honey-
II in Europe. Also, issues related to wells New Construction Solu-
corporate image and environmental tions (2500 West Union Hills
Drive, Phoenix, AZ 85027
stewardship have growing implica- Phone: 602-313-4138; Fax:
tions in the global market. 602-313-4040; Email: scott.
hillman@honeywell.com). In
To help address these issues, indus- his 20 years with Honeywell,
trial plants need effective solutions Hillman has held project en-
gineering, training, consult-
for improving a wide range of process- ing and a variety of product
manager positions. He also is a Certified Safety
safeguarding practices. This requires a Functional Safety Expert (CFSE) and a member
control-system architecture that allows of the CFSE Governance Board.
Circle 32 on p. 70 or go to adlinks.che.com/23014-32
44 Chemical Engineering www.che.com May 2009
 Feature Report
Engineering Practice
Anti-surge Valves for
Dynamic Compressors
Follow this guidance to
improve the selection
and sizing of anti-surge
valves for axial and
centrifugal compressors
Amin Almasi
WorleyParsons Services Pty. Ltd.
O
ne of the most significant limi-
tations on the operation of dy-
namic compressors (axial and
centrifugal compressors) is
the low-flow limit known as surge.
A sudden valve closing, such as for
emergency shutdowns or for a suc-
tion/discharge valve, presents sig-
nificant challenges to anti-surge
valve sizing and selection. Anti-surge
valve requirements depend to a large
degree on the compressor type, the
compression system arrangement,
and particularly the volume of the
discharge piping downstream of the
compressor. Important aspects of
anti-surge valve sizing and selection
for plants in the chemical process in-
dustries (CPI), as well as examples,
are discussed in this article.
The basics
At low flows, the performance of a
dynamic compressor particularly
centrifugal and axial compressors
is limited by the occurrence of aero-
dynamic flow instabilities that could
lead to a catastrophic failure of the
compressor due to mechanical or ther-
mal loads. The surge is an unstable
flow situation that occurs when the
flow is too low. When a dynamic com-
pressor reaches its surge limit, the
flow pattern through the compressor
collapses, and a sudden backward flow
of gas occurs from the discharge to the
suction side of the compressor.
This surge phenomenon can cause a
reversal of the thrust loads (large dy-
namic/transient forces on the rotor),
which can result in damages to bear-
ings (particularly thrust bearings),
seals and other parts of the compres-
sor. For these reasons a surge and
particularly high-energy surges
should be avoided. Recent innovations
in control technology, dynamic simula-
tions, rotating machinery knowledge
and control-valve design (the anti-
surge valve) have made it possible to
supply anti-surge systems that are
capable of coping with rapid flow fluc-
tuations and process gas variations.
A dynamic compressor, also known
as a turbo-compressor, is generally
provided with an anti-surge valve (Fig-
ure 1) in order to keep the compres-
sor in a stable operating range. This
is achieved by assuring a suction flow
that is higher than the corresponding
flow at the surge. An anti-surge valve
is placed in the anti-surge circuit (a
recycle circuit).
The main purpose of this valve is to
avoid a surge and surge-related dam-
ages, particularly irreparable dam-
ages (Figure 2). The anti-surge system
and the controller, which comprise
Chemical Engineering www.che.com November 2012
Figure 1. This
picture shows an
example of an anti-
surge valve
both hardware and software, should
be based on a design algorithm that
operates very quickly when opening
the anti-surge (recycle) valve to avoid
the surge. However, this anti-surge
system should be designed and pro-
grammed so as to not make the circuit
(or the compression unit) unstable.
Centrifugal and axial compressors
experience a surge whenever they
are suddenly tripped. Controlling the
head and total energy of the surge
event, particularly a surge at the trip,
is the main objective of dynamic simu-
lations and anti-surge valve sizing
and selection. Meaningful gains can
be made by a better understanding of
the interactions between the compres-
sor, the anti-surge valve(s) and the
facility piping and layout (including
coolers, scrubbers, check valves, vent
valves and more). Key parameters for
anti-surge valve selection and sizing
include the following: fast response,
optimum size, high integrity, accuracy
and noise abatement.
The anti-surge valve
The size of an anti-surge valve influ-
ences the amount of fluid that can be
moved from the discharge side of the
compressor to the suction side. Larger
valves allow more flow. The speed
with which an anti-surge valve opens
43
 Engineering Practice
is also important in rapidly reducing
the discharge pressure (the head and
stored energy at the compressor dis-
charge). One important trade-off in
relation to anti-surge valves is that
larger valves exhibit slower opening
rates. There are certainly cases when
a faster, smaller valve is better than a
larger, slower valve, but in all cases,
the anti-surge valve should be large
enough to handle the flow supplied by
the compressor at a sufficiently low
differential pressure (head).
Another important factor is control-
lability. Larger valves are more dif-
ficult to control. The faster the valve
can be opened, the more flow can pass
through it and the more effective it is
in surge prevention. There are limits
to the valve opening speed, dictated
by the need to control intermediate
positions of the valve. There are some
practical limits to the power of the ac-
tuator. The situation may be improved
by using a valve that is boosted to
open, combining a high opening speed
for the surge avoidance with the capa-
bility to avoid oscillations.
During a surge event, the flow in-
creases rapidly at first and then de-
creases more slowly as the pressure
difference across the compressor
and consequently the anti-surge
valve, too decreases. With a rela-
tively large anti-surge valve, the en-
ergy level at which the surge occurs
is much lower compared to a smaller
anti-surge valve. An anti-surge valve
with a relatively large capacity (com-
pared to the compressors volumetric
flowrate), but with the same opening
time (compared to relatively small
anti-surge valves), will allow the
shutdown surge to be avoided (or the
shutdown surge will occur at a lower
head condition). On the other hand, a
relatively small anti-surge valve could
cause a more energetic and more po-
tentially damaging shutdown.
The ability to quantify these ef-
fects and select a proper valve size
by dynamic modeling can be very
useful. Because of this, accurate
simulations are encouraged in order
to select an anti-surge valve with an
optimum size.
The effect of valve flow coefficient.
In a case study for a centrifugal com-
pressor in a CPI plant, two anti-surge
44 Chemical Engineering www.che.com November 2012
valves with the same opening time
(around one second) were evaluated
and compared. The small anti-surge
valve resulted in a drop in flowrate
without much change in the pressure
head. The shutdown surge occurred at
a moderately high discharge pressure.
The large anti-surge valve with a
valve flow coefficient (Cv) about two
times that of the small valve re-
sulted in a major reduction in the
head. The shutdown surge occurred
at a discharge pressure just above the
suction pressure. The surge event in
this case was not even noticeable. The
flow reversed through the compressor
at a low head point. This case study
shows the surge-preventing effects of
an anti-surge valve with a higher Cv
compared to a smaller valve with a
lower Cv.
Additional requirements for anti-
surge valves include the following:
1. The reduction of stroke time (in
response to control signal step
changes)
2. A stable response
3. Minimized overshooting during
the valve adjustment steps
The requirements for an anti-surge
valve should always be optimized. Too
fast a response can result in excessive
overshoot and poor accuracy. How-
ever, too slow a response could result
in sluggish opening of the anti-surge
valve and possible inadequate pro-
tection of a compressor from a surge.
Most information required for the siz-
ing of anti-surge valves is available
on the compressor map (the compres-
sor curves). Well-known compressor
manufacturers and anti-surge system
suppliers have their own standards
for the safety margins on anti-surge
valve capacity. This figure is usually
based on the compression system dy-
namics, the valve response and the
system characteristics.
The anti-surge valve should be ca-
pable of passing 100% of the surge
flowrate at around 50% of the valve
opening. In other words, the anti-
surge valves Cv is selected to be ap-
proximately two times that of the re-
quired Cv, based on the compressor
surge flow on the highest compressor
speed curve (on the compressor map).
Valve noise. When the anti-surge
valve is opened by the anti-surge sys-
tem, the valve dissipates a large por-
tion of operating power of the compres-
sor. A portion of the power dissipated
by the valve is converted into acoustic
energy, which becomes valve noise.
Apart from the noise, considerations
should also be given to the fluid veloc-
ity in the valve outlet. This should be
kept within some limits to avoid pipe
vibrations. Typically, CPI plants use
an anti-surge-valve noise limit range
(before the external attenuation) from
85 to 100 dB(A). The anticipated noise
level (before the external attenuation)
should not exceed 105 dB(A) with fluid
velocities below Mach 0.3.
Anti-surge valves should be de-
signed to increase the total system
efficiency over a wide range of the op-
eration. The inline and symmetrical
flow path eliminates indirect flows
and unnecessary changes in the flow
directions through the valve. Axial-
flow anti-surge valves are most often
specified. This design is rugged and
offers high performance. It is highly
reliable and requires little mainte-
nance. The breaker vanes are often
used in the downstream section of the
valve body (downstream of the valves
flow-control internals), which cut and
streamline any flow turbulence. This
can result in significant reductions in
the noise and the turbulence, elimi-
nating excessive vibrations in the
valve and the associated piping, sup-
ports and structures.
The anti-surge valve should be ca-
pable of changing its position in a
short period of time. For reasons of
reliability, a spring-loaded actuator is
desirable. The typical control instru-
mentation and accessories provided
on an anti-surge valve can include a
valve positioner, a volume booster and
a solenoid. In other words, usually a
solenoid valve with a well-designed
booster device (that accelerates the
positioner operation) is used.
The correct size and configuration of
the required actuators, instrumenta-
tion and accessories could guarantee
an anti-surge-valve response time of
less than 2 s (to fully open). An over-
shoot on intermediate changes should
be kept to an absolute minimum.
Control range. The anti-surge valve
should have a wide control range. The
whole valve system should operate

very quickly in case of an emergency
(particularly an emergency-trip surge
event). A very high operating range-
ability is usually required. Typically, a
rangeability of 150:1 is specified. The
rangeability parameter is defined as
the ratio between the rated Cv for a
completely open valve and the mini-
mum Cv that the valve can control.
A high rangeability value means suc-
cessful control even with a high differ-
ential pressure and a low flow.
To increase the rangeability, special
trims are used where the area distri-
bution is made according to the valve
characteristic curve. For example, in
some valve designs, until 40% of the
stroke, the Cv is typically limited to
1525% of its maximum value to ob-
tain superior controllability. Too often,
a special trim with multiple jumps is
specified for an anti-surge valve.
Service reliability. Good service re-
liability even under very severe condi-
tions (for example, compression ratios
lower than 0.3, a discharge pressure
ranging up to 600 bars or more, and
other difficult situations) is required.
In order to obtain these requirements,
the following considerations should
be respected:
Anti-surge valves in various trim
styles are usually fitted with pres-
sure balanced pistons. The required
thrust should be virtually indepen-
dent of the differential pressure
across the valve
Bushings should be anti-seize and
self-lubricated. The internal slid-
ing area requires high-quality and
special design and materials (usu-
ally chromium-plated internals are
specified)
Proper packings should be used.
Typically, a charged polytetrafluo-
Figure 2.
Axial com-
pressors are
very vulnera-
ble to damage
from surges.
This photo de-
picts the type
of blade dam-
age that can
be expected
as a result of
a surge event
in an axial
compressor
roethylene (PTFE) with a special
design is specified. For low molecu-
lar-weight gases (such as hydrogen)
high-quality graphite packing is
commonly specified
The trim materials should be cho-
sen for good corrosion-proofing and
erosion-proofing. For usual services,
a proper stainless-steel alloy is typi-
cally specified. For a special appli-
cation, a special, sintered tungsten
carbide could be a good selection.
The trim design should allow for ex-
cellent sealing
For high-pressure anti-surge valves,
additional requirements should be
considered. For example, the welded
seat is replaced by a high-quality
cage seat for superior sealing, better
centering and easier maintenance
and operation
Leakage. Leaks across an anti-surge
valve can influence the efficiency of a
compressor system. Even under the
most severe working conditions, high-
quality anti-surge valves should main-
tain a tight shut-off (within practical
limits) over the full pressure range.
With this feature, the compressor ef-
ficiency is maintained at the highest
levels during the normal operating
mode. In a case study for a medium-
size high-pressure centrifugal com-
pressor (8-in. ASME Class 900 recycle
control valve), with an ASME Class
IV leakage rate and a differential
pressure of around 70 bars, the valve
leaked approximately 150 Nm/h of a
compressed gas.
The leakage feature should be inde-
pendent of the actuation method. New
designs do not need a higher torque as
was required in older designs. In mod-
ern designs, the ASME Class IV seal-
ing class is usually achieved by the
Chemical Engineering www.che.com November 2012
position and not by the torque. Rapid
changes in the differential pressure
across the anti-surge valve should
have no effect on the stability of the
valve position. A fast response is usu-
ally achieved with properly sized ac-
tuators in contrast to those fitted on
conventional control valves. These
advanced actuators are usually very
efficient and compact. They cannot
be bulky or massive. Modern designs
contain a minimum number of moving
parts and a short valve stroke. These
advanced designs permit fast actions
and reliable operation.
Various studies, different simula-
tions and extensive operational ex-
periences pertain to a typical one
anti-surge valve, one compressor
casing arrangement. More com-
plex systems of cascaded valves (or
valves around multiple compressors)
require a more detailed analysis and
sophisticated provisions.
Valve and size selection. An anti-
surge valve should be sized to meet
two diverse objectives: steady-state
operation and transient cases.
During steady-state recycling, the
required capacity of the anti-surge
valve can be directly derived from
the compressor map (the compressor
curve). Typically, smaller valves ex-
hibit smoother control. During tran-
sient conditions, however, the required
valve capacity increases. To avoid a
surge during an emergency shutdown
(or similar transient events, such as
a sudden suction or discharge valve
closing, or others), a bigger valve will
give better performance. To fulfill both
above-mentioned requirements, an
anti-surge valve with an equal per-
centage characteristic (defined below)
is recommended. Two types of anti-
surge valves are generally used: globe
valves and ball valves (noise-attenu-
ating ball valves).
The equal percentage characteris-
tic spreads the first half (50%) of the
valves fully open capacity over the
first two thirds (66%) of the valves
travel for a globe valve, and about one
third (33%) of the valves fully open
capacity over the first two thirds of
the valves travel for a ball valve. An
anti-surge valve should be described
by its maximum capacity (Cv), and by
its capacity as a function of the valve
45
 Engineering Practice
travel, and the opening behavior. The
globe valves Cv approximately var-
ies with the square of the percentage
travel. The ball valves capacity varies
roughly with the cube of the percent-
age travel. A ball valve will have more
capacity to depressurize the discharge
volume compared to a globe valve of
the same size.
In a case study for a 6-in. (150 mm)
anti-surge valve, the Cv of a ball valve
was more than 2.5 times of the Cv
for a same-sized (6-in.) globe valve.
At two thirds of valve travel, the ball
valve flowrate was more than 50%
higher than that for the same-sized
globe valve. This additional flow ca-
pacity would sometimes, theoretically,
make the ball valve a better choice in
an anti-surge application. This is par-
ticularly true where there is only a
single anti-surge valve per stage (for
example, there is no hot-gas bypass)
and the discharge volumes are large
(especially when the anti-surge loop
is taken from downstream of the dis-
charge cooler). However, this is just a
theoretical and textbook idea.
From a practical standpoint, the
globe valve behavior is more predict-
able. The globe valve is more comfort-
able to control and manage. It is rela-
tively less nonlinear. The ball valve
usually offers a highly nonlinear be-
havior (Cv varies with the cube of the
percentage of travel).
The anti-surge valve should be
matched to the compressor system.
Both globe and ball valves are used in
modern anti-surge systems (depend-
ing on the application). Practically, a
globe valve is more common in anti-
surge systems.
In some compressor installations,
it is necessary to have both a nor-
mal anti-surge valve (the recycle line
taken from downstream of the after-
cooler or the anti-surge line that in-
cludes a cooler) and a hot-gas bypass
valve, which can be opened to rapidly
transfer the compressed gas back to
the suction side of the compressor
without cooling (usually for a short
period of time). A hot-gas-bypass al-
lows the pressure head across a com-
pressor to be rapidly reduced during
a trip. Various options for anti-surge
valve size, opening rate, recycle loop
arrangement and operation sequence
46 Chemical Engineering www.che.com November 2012
should be evaluated. Parametric stud-
ies of the anti-surge valve size and
opening time can yield optimum sizes
for an anti-surge valve (and a hot-gas-
bypass valve, if required).
Overheating. A large portion of the
mechanical energy produced by a com-
pressor is converted into heat in the
discharged gas. In an uncooled recy-
cle-loop system, this heat is recycled
into the compressor suction end, and
then more heat energy is added to the
recycling gas. In a typical natural-gas
centrifugal compressor at a 100% re-
cycle without recycle cooling, the tem-
perature of the gas increases about
1C per second. Eventually, this could
lead to overheating at the compressor
discharge side.
In theory, the recycling can lead to
heating whenever the compressed gas
cannot be cooled to the initial operat-
ing temperature at the compressor
suction side. This is the case both for a
hot-gas bypass and a partially cooled
bypass (if the cooled bypass is not
designed properly, for example, when
sufficient cooling capacity is not pro-
vided). The overheating could be de-
fined as the gas temperature within
the compressor exceeding 150190C,
depending on the process. The same
problem may also occur if there is a
long period of time between the start
of rotation (the compressor startup)
and overcoming the pressure down-
stream of the discharge check valve,
when some kind of hot-gas bypass is
used for the startup.
Control and testing
An anti-surge-system configuration
usually allows an operator to open
the anti-surge valve further than the
valve position that is defined by the
anti-surge controller (based on the al-
gorithm) in manual-mode operation.
However, the operator cannot close
the anti-surge valve any further than
the surge controller permits. The con-
troller can override the manual mode,
if necessary, to open the anti-surge
valve. The fail safe philosophy that
is commonly employed is that the
anti-surge valve should be an air-to-
close valve (fails open). The control
is used to force the anti-surge valve
fully open whenever the compressor
is shutdown. This helps to prevent a
surge on a compressor shutdown, and
also holds the anti-surge valve in a
fully open position for a startup.
Extreme care has to be taken when
testing an anti-surge system during
operation. The compressor should
not accidentally be pushed into a
surge. The controller should take
over if something causes the oper-
ating point to head toward a surge.
Such features should be available,
enabled and working correctly. A sud-
den movement of the anti-surge valve
could have an adverse impact on the
remaining gas system (the upstream
and the downstream of the com-
pressor). All of these control actions
should be evaluated properly.
It could be useful to assess how the
anti-surge valve behaves in response
to a small 12% change in the com-
mand signal. Bad signs are hunting,
stiction (the static friction that needs
to be overcome when the anti-surge
valve-position should be changed), hys-
teresis and delayed lifting, all of which
could make the control more difficult.
The anti-surge step test should be
performed with the controller in the
manual mode and with the valve off
its seat, preferably somewhere be-
tween 1585% open. The compressor
will be on partial-recycle through-
out the test (the remaining process
should be able to cope). The first
valve step could probably be in the
open direction so as to move the com-
pressor further from the surge. If the
response is unknown, a small valve
movement (even typically 12%)
should be selected.
One important consideration when
operating in the recycle mode is that
the gas could thin out due to loss of
the heavier components. The surge
line of the lower molecular-weight
gas may be rather different from the
normal gas. Some amounts of new gas
should always be fed through the com-
pressor. Another useful test is to in-
troduce a disturbance to the compres-
sor, such as cutting back the suction
throttle valve slightly or changing the
driver speed (if it is a variable-speed
compressor train). It should be veri-
fied that the anti-surge valve changes
position without any delay and also
without a significant undershoot of
the setpoint.
Examples 17-4-PH) trim and a special graphite of the unit. The calculated Cv for the
Example 1. The first example is pre- packing. Since the compressor train nitrogen operation (the surge flow
sented for a two-process-stage centrifu- inertia is low (it is a steam-turbine, at the high-speed curve) is around
gal compressor. The two process stages direct-driven compressor train), the 260. In other words, a surge could be
are low pressure (LP) and high pres- anti-surge loop is taken from the im- expected in case of an unscheduled
sure (HP). The compressor is a vari- mediate downstream of the compres- rapid trip during the operation with
able-speed synchronous electric-motor- sor (upstream of the cooler). the nitrogen, if a small anti-surge
driven unit (using gears) to compress Based on experience, for a compres- valve is selected.
natural gas from around 10 barg to ap- sor configuration like this, a relatively Edited by Dorothy Lozowski
proximately 100 barg. Each compres- low Cv (around or lower than 1.7) is
sor stage has an anti-surge valve. sometimes specified by vendors. Some Author
Amin Almasi is a lead
The calculated Cv for the LP and vendors used Cv factors (the anti-surge rotating equipment engi-
HP compressor casings (at the surge valve Cv divided by the calculated Cv at neer at WorleyParsons Ser-
vices Pty Ltd. in Brisbane,
flow of the high-speed curve) are 215 the high-speed surge flow) even as low Australia (amin.almasi@
and 81, respectively. The selected as 1.55 or 1.6 to keep the anti-surge worleyparsons.com). He pre-
viously worked in Technicas
anti-surge valves Cv values for the valve size small. This is not recom- Reunidas (Madrid) and Fluor
(various offices). He holds a
LP and HP stages are 385 and 169, mended, however, since a sufficiently chartered professional engi-
respectively. In other words, the Cv large anti-surge valve is necessary to neers license from Engineers
Australia (MIEAust CPEng-
factors (the anti-surge valve Cv / the handle emergency rapid shutdown sit- Mechanical), and a chartered engineer certificate
calculated Cv at the high-speed surge uations and unscheduled trips at un- from IMechE (CEng MIMechE), RPEQ (Regis-
tered Professional Engineer in Queensland). He
flow) are approximately 1.79 and 2.09 common (off-design) operating cases. also holds M.S. and B.S. degrees in mechanical
for the LP and HP loops, respectively. In other words, 1.7 is the minimum engineering. He specializes in rotating machines
including centrifugal, screw and reciprocating
The selected anti-surge valves are acceptable factor. This hydrogen com- compressors, gas and steam turbines, pumps,
condition monitoring and reliability. He has au-
globe valves. The opening time is pressor train is expected to compress thored more than 45 papers and articles dealing
2 s and the closing time is 5 s. The nitrogen during the initial startup with rotating machines.
valves were purchased with a NACE
(National Association of Corrosion
Engineers)-compliant carbon-steel
body, a stainless-steel (grade 17-4-PH)
trim and special PTFE packing. Grade
17-4-PH stainless steel is a precipita-
tion hardened (a chromium-copper
precipitation hardened) stainless
steel, employed for applications re-
quiring high-strength and a high level
of corrosion resistance.
Example 2. The second example is
presented for a petroleum-refinery
type hydrogen-service centrifugal
compressor. It is a steam-turbine-
driven, hydrogen-recycle centrifugal
compressor for a hydrocracker unit to
compress the hydrogen from around
140 to 190 barg. The calculated Cv for
the surge point at the high-speed curve
(on the compressor map) is 179. The
recommended range for the anti-surge
Cv is 1.72.1 times the calculated Cv
for the maximum-speed surge flow.
Applying the 1.7 factor (the minimum
factor) results in an anti-surge valve
Cv above 305.
The selected anti-surge valve is an
8-in. globe valve, 1500# class, RTJ (ring
type joint) facing, with a Cv of 308,
an opening time of 2 s and a closing
time of 6 s. The anti-surge valve was
selected with a carbon-steel body (a
NACE grade), a stainless-steel (grade
Circle 4 on p. 60 or go to adlinks.che.com/40273-04
Chemical Engineering www.che.com November 2012 47
 Solids
Environmental
Processing
Manager
Preventing
Dust Explosions
Risk management programs are critical
for safe handling and processing of combustible dust
as well as for OSHA regulatory compliance
Geof Brazier and Mitch Rooker
BS&B Safety Systems, LLC
H
ousehold items, such as
breakfast cereal, flour, corn
starch and sugar are so com-
mon that many were left
surprised after the 2008 combustible-
dust explosion at a sugar refinery in
Georgia. Heard in much of the early
media coverage was the question How
could sugar explode?
Plant explosions are, thankfully, not
routine occurrences. However, when
they do occur, it is the unusual event
often a combination of abnormal
events that is typically the trigger.
This article outlines some of the guide-
lines that are available to help prevent
dust explosions. It also offers a list of
factors to consider when choosing ex-
plosion vent technology to minimize
combustible-dust-explosion hazards.
The facts are that in the U.S. and
elsewhere, the chemical process in-
dustries (CPI) are working to mini-
mize the risk of combustible dust ex-
plosions by following a revised suite of
National Fire Protection Association
(NFPA) standards, that are recog-
nized by the U.S. Occupational Safety
and Health Administration (OSHA).
A documented combustible-dust, risk-
management program that is regu-
larly updated, forms the cornerstone of
a plants owner/operator responsibility
to provide a safe workplace. Identifica-
tion of areas of risk, along with imple-
mentation of prevention and protec-
tion strategies, achieves compliance
with these safety standards.
OSHAs Combustible Dust National
Emphasis Program (NEP) and the NF-
PAs current standards define best en-
gineering practices and are designed to
protect both personnel and plant from
combustible dust explosions. These
standards are typically adopted by state
fire marshals, insurance companies,
consultants and their legal representa-
tives and are referenced by the OSHA
NEP. OSHA can and does issue safety
citations in relation to combustible dust
risks. Its citations are based on the Gen-
eral Duty Clause whereby the owner/
operator of a facility shall provide a safe
workplace, as well as by reference to
specific NFPA standards including 61,
68, 69, 86, 484, 499, 654, and 664.
OSHAs NEP
OSHA introduced the Combustible
Dust National Emphasis Program in
October 2007 issued by the Directorate
of Enforcement Programs and revised
it on March 11, 2008 with directive
number CPL 03-00-008*. This docu-
ment describes OSHAs policies and
procedures regarding inspection of fa-
cilities that handle combustible dust
as well as their expectations regarding
owner/operator compliance with U.S.
safety standards. The scope is national
and administered by national, regional
and area offices of OSHA or the offi-
cials of participating state programs.
Its focus includes the following three
technical areas of risk management:
1) Dust control
2) Ignition control
3) Injury and damage control
Dust control. To summarize, dust
control is achieved by the implemen-
tation of appropriate combinations
* Available for review on www.bsbipd.com
Chemical Engineering www.che.com October 2009
Figure 1. This photo
shows the effects of
pneumatic transport of
powder to a silo through
a hose not designed
to conduct away static
electricity. The friction
caused by static electric-
ity has created enough
heat to cause a hole to
form.Code compliance
with OSHA requires con-
tinuous monitoring of
potential ignition hazards,
including the control
of static electricity and
process equipment hot
spots, similar to the pic-
tured example
of housekeeping and dust collection-
and-filtration measures. These mea-
sures are combined with inspection to
ensure that they remain effective. A
layer of dust that is only 1/32-in. thick
is a concern when it covers 5% or more
of a workplace floor area that has
dust-laden structural members, such
as joists and I-beams. Successful dust
control reduces fugitive emissions of
combustible dusts, which in turn cuts
the dust explosion risk.
Quantification of a combustible
dust risk requires that the charac-
teristics of the material be identified
by test. Key parameters include the
maximum explosion pressure (Pmax),
the deflagration index (Kst) and the
minimum ignition energy (MIE) of the
dust. Prevention and protection strat-
egies must also be based on knowledge
of process conditions, such as dust con-
centration, airflow velocity, operating
pressure, temperature and humidity.
The higher the Kst value, the faster
the rate of pressure rise due to com-
bustion. Kst values above 600 are
considered extremely explosive. Most
common plant dusts have Kst values
between 100 to 150. Even if a plant
owner/operator knows the characteris-
tics of a particular dust, a sample must
be tested under NFPA 68. Particle size
can profoundly affect explosive proper-
ties, as the finer the dust, the higher
the Kst value. For example, sugar has
a recorded Kst value of about 138. Of
more concern with regard to the po-
tential damage to process equipment
is the fact that a fine-sugar-dust explo-
sion will generate a pressure in excess
of 100 psi within an enclosed volume in
less than 100 milliseconds. Other key
49
 Environmental Manager
parameters include particle size, with
smaller particles equating to more
rapid combustion and easier ignition.
The reason is that more surface area
is freely accessible by the surrounding
air to support combustion.
Ignition control. Understanding and
monitoring for potential ignition haz-
ards at a plant provides a first layer
of protection to the facility. Essential
measures include proper grounding
and bonding of equipment and duct-
ing, appropriate wiring of electrical
equipment, the control of static elec-
tricity (see Avoiding Static Sparks in
Hazardous Atmospheres, CE, June
2009, pp. 4449) and monitoring of
process equipment hot spots (Figure
1). Preventive maintenance programs
ensure that such design safety mea-
sures remain effective.
Review of dust-explosion loss history
proves that it is important to consider
normal and potential abnormal cir-
cumstances while evaluating processes
within a facility. Many times a dust
explosion is the result of an abnormal
event, such as when an automated pro-
cess fails and it is replaced by a tem-
porary, manual activity. Other times
it could be a consequence of a change
in product packaging. For example, an
operator at a manual bag-emptying
station receives a shrink wrapped pal-
let of bags. Instinctively, he walks to-
ward the pallet, across a concrete floor,
while wearing appropriately specified
conducting shoes. The surrounding
metal equipment is grounded. The op-
erator stands on shrink-wrap-packag-
ing material (an insulator) as it is un-
wrapped, tears open a bag of material
and empties it into an adjacent hopper.
As the dry powder flows, static charge
builds on the now insulated operator
until there is sufficient electrical po-
tential to release a spark having suffi-
cient energy to ignite the dust present
in the emptying station.
Injury and damage control. The
OSHA NEP anticipates the deploy-
ment of specialty technology to miti-
gate the effects of a dust explosion.
Referenced techniques fall into the
following four areas:
Prevention: Detection of sparks and
embers traveling through a dust
collection system can lead to their
quenching before reaching an area
50 Chemical Engineering www.che.com October 2009
Figure 2. Isolation in-
troduces a barrier to flame
propagation that can prevent
a primary dust explosion from
amplifying into a secondary fire
and explosion in an intercon-
nected piece of equipment. This
example shows chemical barrier
isolation on a 36-in.-dia. duct
of high dust-explosion risk, such as
a filtration unit
Isolation: Introducing a barrier to
flame propagation can prevent a
primary dust explosion in one piece
of equipment from amplifying into
a typically more-severe secondary
event in a connected item of equip-
ment. Solutions include chemical
isolation (particularly well-suited
to large or irregular-shaped duct-
ing) and mechanical isolation barri-
ers (Figure 2), such as pinch valves,
knife gate valves and rotary airlocks
Venting: Pressure relief is provided
to process equipment and to building
structures by releasing the products
of combustion to the atmosphere in
a safe trajectory. Flameless venting
provides for over-pressure protec-
tion without the release of flame
or particulates to the atmosphere
when following the requirements of
NFPA68-2007
Suppression: Explosion effects can
be minimized by injecting a flame
quenching agent into process equip-
ment to arrest the combustion pro-
cess explosion suppression equip-
ment responds rapidly to prevent
the full development of a dust explo-
sion, preventing a destructive over-
pressure from developing (Figure 3)
Figure 3. In
this illustration,
the explosion
suppression sys-
tem extinguishes
the fireball as it
occurs, rapidly
injecting a heat-
quenching agent
into the tanks and
duct to extinguish
the fireball. The
outside world
does not hear or
see the fireball.
The readings from
the sensor and
monitor are the
only indication
that an explosion
has occurred
In addition, an emergency action plan
with properly maintained exit routes
ensures the right safety response for
protection of personnel.
Understanding the risks
The course of action to minimize com-
bustible-dust-explosion hazards be-
gins, preferably, in the planning stages
of a process by identifying the risks,
determining the explosive reactivity
of the combustible process dust, and
implementing a combination of both
prevention and protection measures.
Economic considerations favor the
use of explosion vent technology in
terms of cost of equipment and instal-
lation. However, many factors must be
considered before choosing explosion
vents. Consider the following applica-
tion factors:
Can the flame ball that is ejected
from an open vent be accepted? It
will usually extend 30 to about 100 ft in
length and about half this in diameter.
An 1,800-ft3 vessel venting from a single
position will produce a flame ball over
100 ft in length and 50 ft in diameter.
Simple free venting must be directed to
a safe location where personnel will not
be present and other equipment cannot
be damaged (Figure 4).
If venting equipment is installed

Figure 4. The photo shows explosion
venting of a dust collector.A vented
flameball is a mass of dust and combus-
tion gases over 2,000F that may extend
30100 ft from the point of exit. Simple
free venting must be to a safe location
where personnel will not be present and
other equipment cannot be damaged
indoors, can a vent duct to a safe
outdoor location be provided?
Vent ducts will always increase the
required vent area, and their use may
not allow smaller process volumes at
higher Kst values to be protected by
venting at all.
Can the required vent area be ac-
commodated? As well as requiring the
space for vent installation, can the reac-
tion forces during venting be sustained
and, for tall equipment, can a near-
thrust, neutral vent arrangement (top-
sidewall-mounted vents) be achieved to
prevent collapse during relief?
If there are process inlets and out-
lets to the protected equipment,
are these protected to prevent
propagation of the dust explosion
to other equipment or work areas?
Standards are now very clear in re-
quiring isolation of vented equipment
to prevent secondary explosions. Sec-
ondary explosion risks are typically
much greater in their potential for
damage and destruction.
Can the clean up of a vented explo-
sion be accepted? Depending upon
the design basis adopted, a vented
explosion may require replacement of
capital equipment components that
have become damaged by the pressure
wave, resulting in loss of production
while delivery is awaited. Most vented
equipment is designed with a pres-
sure relief area that prevents failure
of the equipment structure. A higher
vent area is always required to protect
equipment within its design pressure.
What will the neighbors think?
A vented explosion is a spectacular
event that will draw considerable
community attention.
What if the process material is
toxic or hazardous? A vented re-
For custom fabrication,
Mueller is the obvious choice.
lease must simply be avoided for cer-
tain materials.
As illustrated by the previous series
of questions posed for a vented-dust-
explosion application, each process
needs to be considered both alone and
as a component of a production facil-
ity to ensure that the right explosion
protection and prevention technology
is implemented. There are always op-
tions for dust-explosion-risk manage-
ment. The technical solutions adopted
must take proper account of the practi-
cal needs of each process and the con-
sequences of safety system operation.
Some final rules of thumb include:
plan for the abnormal; and maintain
a strict management of change policy
that will catch the potential conse-
quences of product material changes,
hardware changes, and procedural
changes. Dust-explosion-risk manage-
ment requires periodic detailed review,
even at the best-protected facility. 
Edited by Dorothy Lozowski

Specialty isnt just our name,

Authors
Geof Brazier serves as presi-
dent of Industrial Protection
Devices, LLC and as the di-
rector of product and market
development for BS&B Safety
Systems, LLC (7455 East
46th St., Tulsa, OK 74145;
Phone: 918622-5950; Email:
gbrazier@bsbsystems.com).
He holds more than 20 U.S.
patents in the areas of pres-
sure-relief and industrial
wireless devices. Brazier also collaborates with
the Minnesota Workplace Safety Consultation or-
ganization in providing seminars about OSHAs
Combustible Dust NEP. He was educated at the
University of Bristol in England as a physicist.
Mitch Rooker is a senior
development engineer at
BS&B Safety Systems LLC
and along with his co-author,
serves as a member of the
NFPA Committee on Explo-
sion Protection Systems. He
has more than 30 years of ex-
perience in research, design
and development of industrial
products and many patents in
the area of overpressure pro-
tection. He holds a degree in Mechanical Engi-
neering from the University of Oklahoma.
its what we do. For over 50
years, Mueller Steam Specialty
has been the preferred source
for the widest variety of special
service valves and strainers
in the industry. This capability
is further enhanced with the
addition of in-house fabrica-
tion. Using state-of-the-art cut-
ting, fabricating, and welding
equipment, our ASME-certified
craftsmen manufacture to our
customers specific needs.
Quality is assured by in-house,
dedicated Design Engineering
professionals.
Designed, Engineered and
Manufactured in the USA
muellersteam.com A Watts Water Technologies Company
Circle 31 on p. 66 or go to adlinks.che.com/23019-31
Chemical Engineering www.che.com October 2009 51
 Feature Report
The Next Step Change
in Process Safety
Leveraging the convergence of operational and information
technologies can aid in minimizing risk
Andrew Soignier
Ventyx
Angela Summers
SIS-Tech
Mike K. Williams
Modern Automation Consulting
Services, LLC
F
rom plastics and synthetics to
fertilizers and fuel, the gen-
eral public takes for granted
so much of what the chemical
process industries (CPI) produces,
and the risks associated with CPI
production facilities are not typi-
cally considered. That is, until the
next headline-creating industrial
catastrophe occurs. The human toll
of such disasters plays out across
ruined lives, devastated communi-
ties and obliterated opportunities.
The effects can stretch out for years
in the form of chronic health condi-
tions, diminished earning capacities
and contaminated environments.
However, those connected to the
CPI do not forget the risks, nor can
they afford to. Catastrophic incidents,
large and small, have increased the
focus on process safety management
in production facilities around the
world. Although enormous progress
has been made in preventing them,
incidents continue to occur. In fact,
while safety incidents have been de-
clining in number since 2008, those
that do happen have been increasing
in severity, according to the Ameri-
can Institute of Chemical Engineers
(AIChE; www.aiche.org).
Today, the industry is on a cusp,
seeking the next breakthrough im-
provement in process safety man-
agement one that can empower
Prevention barriers
Hazard
Hazard Major incident
Hazard or other
consequence
Figure 1. The Swiss Cheese model involves the implementation of various safe-
guards and barriers to prevent the propagation of a hazardous situation however,
gaps exist in this setup and must be addressed
facility operators by notifying them cedures; poor communication causes
when risk increases, and enabling insufficient worker-to-worker han-
them to respond before abnormal doffs; stress and excessive fatigue
operation escalates uncontrollably impair decision-making and con-
and causes injuries, monetary loss tribute to procedural missteps; and
or environmental damage. This ar- poor human-machine interfacing
ticle examines how key elements, can lead to operator confusion and
such as automation and informa- delayed responses.
tion-management systems, as well Many companies minimize the
as operating discipline, can begin to contribution of human error in in-
converge to make this proactive ap- cident initiation through the imple-
proach a reality. mentation of safeguards or barri-
ers based on the so-called Swiss
Cascading causes of incidents Cheese model of accident causation
Numerous studies and analyses have used in risk management (Figure
concluded that the majority of safety 1). The purpose of this model is to
incidents result from human error. reduce the likelihood of an incident
Operating discipline is typically at occurring, or to reduce the impact
fault, since these errors spring pri- if an incident does occur. However,
marily from defects and deficiencies once these safeguards and barriers
in following operating or mainte- are introduced into operations, rig-
nance procedures, and in applying orous procedural controls are then
necessary administrative controls necessary to ensure their integrity.
to ensure competency, effective com- Otherwise, inevitable human errors
munications, performance measure- and equipment degradation reduce
ment and change management. For the effectiveness of the model.
example, work instructions that are Shifting tasks from manual to
incomplete, inaccessible or illegible automated operation reduces the
lead to inconsistent execution of pro- potential for direct human errors in
Chemical Engineering www.chemengonline.com october 2014 53
 Feature Report
initiating an incident, but in order
for the automated functions to be
effective, asset-integrity systems
covering the operation and mainte-
nance of these safety-critical pieces
of equipment become necessary.
Whether manual or automated, a
consistent and accurate execution
of safety-critical tasks requires op-
erating discipline and the ability to
monitor for changes in risk. Incident
investigations reveal that deteriora-
tion of the barriers and safeguards
often start long before the accident
occurs, and that no systems existed
to detect and report their loss. There
are three main contributors to this
deterioration of barriers and safe-
guards: the passage of time, covert
risks and complacency.
The passing of time. Just because
a safety incident has not occurred
for some time does not mean that
all is well. If assets are poorly main-
tained and operating processes not
regularly checked for safety effec-
tiveness, they will eventually stop
providing the level of risk reduc-
tion they were originally designed
to provide.
Covert risk. Risk has a propensity
to emerge from the least-expected
places. The inability to visualize
where the risks are and the source
of the next incident is an open door
to disaster.
Complacency. Statements like
This is how we always do it and
Dont fix it if it isnt broken are
heard frequently. As months and
years pass without incidents, it
is all too easy to become compla-
cent, especially when it is not read-
ily apparent where hazards can
stealthily develop. That is when
poor habits can infiltrate processes,
and running overtime becomes the
new normal.
Overall, the Swiss Cheese model
alone is too porous and static to
achieve the next safety perfor-
mance breakthrough. Instead, the
confluence of human error, operat-
ing data and information-manage-
ment systems must be addressed.
Process-safety challenges must be
viewed as a holistic system-inte-
gration problem in order to make
meaningful progress.
54 Chemical Engineering www.chemengonline.com october 2014
Data Sources
Planning
Scheduling
Cost control
Logbooks
Rounds
Da
Procedures
ta
Inspections
Maintenace Dat
a
DCS
PLC Data
SIS
SCADA ta
Da
HAZOP In
nt
nteg
Integrated P
LOPA
SIL/SIF
SRS
Figure 2. Various sources of information can come together to provide a realtime
visualization of operational risks
Holistic systems integration
Non-integrated process-safety pro-
grams may address some of the low
hanging fruit, but they do not pre-
vent everything, as is evident from
the prevalence of catastrophic in-
cidents. The integration of manual
tasks with automated procedures
and smart interface design aids in
avoiding deficiencies in procedure
implementation and communica-
tion, as well as minimizing the po-
tential for human error. The key
pieces already exist in the opera-
tional-technology (OT) and informa-
tion-technology (IT) realms and now
must be pulled into conjunction to
comprehensively address the grow-
ing complexity of the human factors
that contribute to safety incidents.
For example, on the operational
technology side, companies can
implement state-based unit-control
schemas to address a range of nor-
mal and abnormal operating condi-
tions. They can also employ auto-
mation systems to detect abnormal
events and take preemptive actions
to stop incident propagation. Addi-
tionally, they can implement abnor-
mal situation management (ASM)
graphical standards to optimize op-
erator navigation during abnormal
events and alarm management to
promptly focus attention on safety-
critical issues.
On the IT side, companies can
integrate standard operating pro-
cedures (SOPs) with automated
delivery of the latest revision of
the required tasks, electronic time-
Analytics Visualization
Operational risk
Process safety
management
IInformation
In
Information
I
Info
Procedures
Operational efficiency
Actions
stamped signatures, prescribed re-
cords and quality tolerances. Any
procedure can be further supported
with safety rules and automated in-
terlocking functions, to ensure that
manual tasks are sequenced prop-
erly, that the proper individuals are
notified for certain tasks, and that
detailed findings are consistently
recorded for monitoring purposes.
Further, workflow applications can
be deployed to aggregate the results
to the desired level of granularity
to facilitate adequate information
sharing and reduce the chances of
human error.
Most promising of all is the oppor-
tunity to integrate realtime analysis
into the overall picture to identify
operational risk before it translates
into incidents, and to drive the most
effective risk-mitigation schemes.
Figure 2 illustrates how data can be
leveraged through proper integrated
procedures to help visualize risk.
Static versus dynamic risk
Risk is not static; things are al-
ways changing. Safeguards can de-
velop faults, or they can be down for
maintenance. New and different ac-
tivities may be taking place during
installations. Organizations, people,
resources and logistics can easily
and quickly shift. In short, nothing
should be taken for granted. Rather,
organizations must bolster safe-
guards and barriers in a dynamic
fashion. To do this, they can moni-
tor leading indicators of increased
vulnerability to incidents on a day-
 Feature Report

tionable intelligence. Technologies

Figure 3. Cumulative risk is shown for an entire operating unit, drawing attention
to all areas that are prone to incident-causing situations
to-day basis. A daily review of cumu-
lative risk enables organizations to
digest new operations information.
Where are breaks in the plan that
will defer planned work? What de-
fective equipment has been found
that cannot be repaired immedi-
ately? Which employee is unexpect-
edly off the job today? The accumu-
lated information can be used to fuel
risk assessments, both operational
and safety-critical, leading to deci-
sions on whether to shut down or to
take compensating measures.
While this provides a reliable
basis for operational decision-mak-
ing and control, and ensures that
levels of cumulative risk remain
tolerable, the key step is to leverage
the convergence of OT and IT and
react ahead of incidents through
realtime risk monitoring, analysis
and advising. Figure 3 shows a visu-
alized representation of cumulative
risk for an operating unit.
facilitys wealth of realtime opera-
tional data, such as instrumenta-
tion and control data, with technol-
ogy for dynamic risk analysis. When
done comprehensively, this can re-
sult in a realtime dynamic risk ad-
visory capacity for monitoring and
immediately alerting personnel to
risks as they change and develop,
and for providing an optimal course
of action to maintain the integrity
of the facility.
Accessing and amalgamating all
of the required data is challenging,
because data can reside everywhere:
from operations, maintenance and
automation systems, log books, oper-
ator rounds, mechanical inspections,
lock-out or tag-out applications and
databases. Additionally, information
is being created and changed con-
stantly. Data needs to be accessed
from disparate systems, validated,
transformed, integrated and contex-
tualized in order for it to drive ac-
do exist to complete the heavy-lift-
ing integration tasks. Thus, analyt-
ics applications can then connect the
dots, comparing asset operation and
maintenance performance against a
safety basis, and alerting personnel
to any deviations.
Further, the proactive interface
can be strengthened even more
through the use of visualization
technologies in addition to analyt-
ics. Analytic technology can calcu-
late the change in risk dynamically,
(for instance, weighing the conse-
quences of an uncompleted operator
round or a non-implemented proof
test on a critical device) and update
a facilitys risk matrix in realtime
with the revised impact. Visualiza-
tion technology can then help focus
employees attention on the change
in risk using geospatial represen-
tation and color-coded graphics
(Figure 4) for impacted facilities,
units and equipment.
Putting intelligence into action
The next step is to automatically di-
rect appropriate corrective actions
based on this realtime intelligence.
In an example scenario, an operator
scans a screen that maps an entire
facility and displays the dynamic
risk levels associated with each of
the facilitys unit operations at that
moment. A specific unit displays an
elevated level risk and the operator
investigates to discover a deviation
from the design specification that is
now raising the likelihood of an inci-

Proactive analysis Likelihood of hazardous event

Traditional approaches to process
safety management have leaned 1 2 3 4 5 6 7 8
heavily on post-mortem analysis, A
investigating what went wrong
B
and lessons to be learned. Although
analysis after the fact is useful, C
organizations must start asking
How can we be more proactive in
D 1
addressing operational risk? in 2 3 2
E 2
order to truly slow the pace of inci-
dent occurrence and decrease inci- F 3 1 1

dents severity. The answer lies in G

3
the integration of operational and
3 3
information technologies. H
The clearest way to be more pro- Figure 4. The capability to visually classify all the various risks for a facility allows
active is to integrate a production for quick responses by personnel

Chemical Engineering www.chemengonline.com october 2014 55

 Feature Report
dent, and perhaps one of higher se-
verity, than was originally planned
into the design of the unit. In effect,
the operator is viewing a window of
a units behavior as operating ver-
sus as designed.
Examining further, the operator
consults a safety rulebook to iden-
tify the cause of the increasing risk
and to see what tasks are necessary
to resolve the problem, ensure con-
tinued safe operation and restore
operational integrity. The operator
can then schedule those tasks ap-
propriately, and monitor their com-
pletion, while taking the production
unit or some of its equipment offline
if required by procedure.
Multilevel utilization of data
The same collected operational
data, along with its attendant data-
integration infrastructure, that
drives realtime analytics and im-
proved situational awareness can
also be leveraged in other ways to
increase operational risk aware-
ness, power predictive insights and
enable proactivity.
By definition, realtime data have
a limited shelf life and quickly be-
come historical data. As todays re-
altime data become tomorrows his-
torical data, they can be compiled
to inform periodic critical reviews
with asset managers and technical
authorities, monthly governance re-
views with asset leadership teams
and quarterly reporting to executive
committees. Engineers are able to
analyze the accumulated data to en-
hance designs and processes, while
managers and executives can utilize
the data to improve process safety
and enterprise asset-management
strategies. Meanwhile, analyzing re-
altime data in conjunction with his-
torical data can help organizations
more effectively identify and track
trends, not just within one facility
but across many, and help drive suc-
cessful preemptive actions.
However, time-based information
alone is not enough to present a
complete picture of operational risk.
The attendant data-integration in-
frastructure mentioned above is
required to provide the context (for
example, the unit processing state)
56 Chemical Engineering www.chemengonline.com october 2014
Daily review of cumulative
risk to understand the
impact of new information
Break-ins to the plan, which will
defer planned work
Defective devices or
equipment are found
and cannot be
immediately repaired
Unplanned changes
to organization
capability, such as
vacancies, sickness
or absence
Daily discussion between
operational integrity
manager and operations
superintendent
Safety-critical and operational risk
assessment raised as required with task-
analysis support
Figure 5. Day-to-day information aggregation can help in scheduling periodic re-
views regarding process safety and asset management
in which the data are collected and
presented. This context is a necessity
for comparing best-practice histori-
cal data with realtime operational
performance. Within this process-
state context, events (planned or
abnormal) can be collated, analyzed
and acted upon quickly.
Procedural, state-based process-
control standards, such as ISA 106,
have been established to provide
the appropriate data context and
process-control vehicle for realtime
event mitigation of an abnormal
situation. Without providing such
context keys, it is very difficult to
provide the multilevel datasets re-
quired to enable the analytics engine
to work. However, with state-based
process automation standards in
place, basic process control is fully
integrated with safety shutdown
systems, providing a key piece of
enablement technology for rapid,
closed-loop response to an abnor-
mal situation. State-based control
strategies also provide the event
context to continuous, streaming in-
formation, which is vital to analysis
and reporting.
Overall, interactions with inte-
grated data, aggregated at the ap-
propriate level, improves the ability
to benchmark performance, plan al-
ternative courses of action and miti-
Bi-weekly critical review
with asset managers and
technical authorities
Monthly
governance
review with asset
leadership team
Quarterly reporting
to group executive
committee
gate abnormal situations effectively.
Figure 5 illustrates some ways that
aggregated data can be used to
schedule reviews and discussions
about safety-critical tasks.
Employing cloud and mobile
The previously discussed emerging
paradigm for process safety man-
agement is proactive and driven by
data from a myriad of sources. Now,
it is time to discuss how to best de-
liver the data to the appropriate lo-
cation, using three technologies that
have proven to be game-changing in
many industries: mobile (display),
cloud (database) and industrial Eth-
ernet (IE) communications.
Mobile technology has been prob-
lematic from a control standpoint
for the CPI. There is a natural pro-
pensity for individuals to use mobile
technologies to move things along
outside of normal workflows. Some-
times this works to great effect;
sometimes it does not. Nonetheless,
there is no arguing the value that
the correct application of mobile
technology can have in maximizing
asset health and process safety.
Areas where mobile is particularly
valuable include field-data capture,
event logging, operator rounds,
safety inspections and audits and
communicating operations and
 Feature Report

maintenance instructions, among
others. The larger the facility, and
the more dispersed the resources
and assets are, the more valuable
mobile capabilities become. Mobile
technology is also economical, elimi-
nates paperwork, enhances compli-
ance and because it is so perva-
sive in everyday life the learning
curve is far from steep.
The advantages of cloud deploy-
ments are clear as well. Cloud-based
process-safety management enables
organizations to quickly roll out
new process-safety capabilities, and
can enhance collaboration between
individuals and departments. The
cloud can also reduce the costs of ac-
cessing and integrating data, help
eliminate information silos and en-
able cost-effective enterprise-wide
visibility. Importantly, the cloud is
also pivotal for transferring data in
a controlled and secure fashion.
Another key enabler is IE. Sup-
ported by wireless-mesh network-
ing and secure communications
protocols, IE enables a flexible, re-
sponsive, end-to-end networking ar-
chitecture that provides connectiv-
ity, collaboration and integration. As
a result, data can be delivered from
the shop floor to the cloud, and then
from the cloud to a mobile device
efficiently and securely.
Operational integrity windows
The new imperative for CPI com-
panies is knowing with certainty
the source of the next incident, and
being able to avoid it or at least min-
imize its effects. Regulatory agen-
cies, shareholders and employees
demand this. Doing this will require
the ability to:
Reduce operational risk by un-
locking data for realtime insights
and enterprise-level visibility
Examine the big picture
through enterprise-level bench-
marking and trend analyses to
guide safety strategies and keep
incidents at bay
Trust the data by leveraging vali-
dation, whether it is mobile data, or
data created automatically by in-
strumentation and systems
Receive instantaneous alerts on de-
veloping issues through visualization
and alarm or alert technologies on top
of realtime analytic applications
Use mobile and cloud technolo-
gies to deploy and access data
from the point of work or the
process edge
Implement realtime detection
and automated response to ab-
normal conditions, be they partial
process impairment or emergency
shutdown events
Have a single window into op-
erational integrity powered by dy-
namic risk analysis and leverag-
ing existing risk data, automation
systems, maintenance and opera-
tions procedures and business sys-
tems for planning, scheduling and
cost control
When all these abilities fall into
place, a company can capitalize in
new ways on existing infrastructure

PREMIUM
EXPLOSION
Ensure your PROTECTION
www.rembe.com

Indoor / Outdoor Venting

piping integrity With Flame Absorber

www.rembe.us
and Dust Retainer

Q-Box

Q-Rohr-3

In todays operating environment, its more

100% inspection of
important than ever that the piping within
internal and external
your Mechanical Integrity Program complies
with standards such as API-570 and API-574.
pipe surfaces
Inspection results
BULK FLOW
Quest Integrity offers a comprehensive solution
tailored to comply with CONTROL
API-570 and API-574
REMBE - QRQBCL2-4C-RSCRIN

for Powder,
for piping circuits using our proprietary,
LifeQuest Fitness- Granulate,
ultrasonic-based intelligent pigging technology for-Service results Grain, Flour
combined with LifeQuest Fitness-for-Service and Dust
All rights reserved

tailored to comply with

software. API-579

Ensure your piping integrity by identifying

degradation before loss of containment occurs. WE DO IT BETTER
contact us for Europe for North America
www.QuestIntegrity.com/CE REMBE GMBH SAFETY + CONTROL REMBE, INC.
59929 Brilon/Germany Charlotte, NC 28217
+1 253 893 7070 A TEAM Industrial Services Company
T + 49 (0) 29 61 - 74 05 - 0 T +1 704.716.7022
F + 49 (0) 29 61 - 5 07 14 F +1 704.716.7025
info@rembe.de info@rembe.us

Circle 29 on p. 72 or go to adlinks.che.com/50981-29 Circle 30 on p. 72 or go to adlinks.che.com/50981-30

Chemical Engineering www.chemengonline.com october 2014 57
 Feature Report
Authors
Andrew Soignier is serving
as vice president of Oil, Gas &
Petrochemical Solutions for
Ventyx, an ABB company (800
Town & Country Blvd. Suite
500, Houston, TX 77024;
Phone: 225-751-3348; Email:
andrew.soignier@ventyx.abb.
com). In this role, he directly
oversees Ventyx Level 3 solu-
tions in the areas of opera-
tions, safety and asset health.
Soignier joined Ventyx in 2007 and has held sev-
eral positions in executive account management,
and as the sales director of Oil, Gas & Petrochem-
ical Solutions. Prior to joining Ventyx, he spent
11 years in the process automation industry, with
a focus on electrical, process safety, critical con-
trols and rotating equipment. He holds a B.S. in
electrical engineering and an executive MBA
from Louisiana State University.
and benefit from a breakthrough
in process safety performance. A
process-safety paradigm shift is
achieved through integration of
realtime information, best operat-
ing practices and closed-loop con-
trol. Mobile procedure assistants
can help avoid human error. How-
ever, in todays complex processes,
human response time may not be
PROTECT PUMPS
DRY RUNNING CAVITATION BEARING FAILURE OVERLOAD
MONITOR PUMP POWER
Best Sensitivity
Digital Display
TWO ADJUSTABLE SET POINTS
Relay Outputs
Adjustable Delay Timers
4-20 MILLIAMP ANALOG OUTPUT
WHY MONITOR POWER INSTEAD OF JUST AMPS?
POWER
AMPS
Power is Linear-Equal Sensitivity
at Both Low and High Loads
NO LOAD FULL LOAD
WWW.LOADCONTROLS.COM
Circle 18 on p. 72 or go to adlinks.che.com/50981-18
58 Chemical Engineering www.chemengonline.com october 2014
Angela Summers is presi-
dent of SIS-TECH (12621
Featherwood Drive, Suite
120, Houston, TX 77034;
Phone: 281-922-8324; Web-
site: www.sis-tech.com). She
has more than 25 years of ex-
perience in safety controls,
alarms and interlocks, pro-
cess engineering and environ-
mental engineering. She is an
active participant in stan-
dards committees and has written over 60 pa-
pers, technical reports and book chapters. She is
currently editing the Center for Chemical Pro-
cess Safety (CCPS) book Guidelines for Safe
Automation of Chemical Processes. Summers
received a B.S.Ch.E. from Mississippi State Uni-
versity and a Ph.D. in chemical engineering from
the University of Alabama.
quick enough. This is where pre-
programmed safety systems pro-
vide the final layer of protection to
return a process to a safe state.
Unfortunately, safety systems
are conservatively designed to fail
safe. In some cases, this mode of ac-
tion may be premature, resulting in
false process trips, causing undesir-
able loss of productivity. Advances
COMPACT EASY MOUNTING
Only 3.25" x 6.25" x 2"
Starter Door Panel
Raceway Wall
UNIQUE RANGE FINDER SENSOR
Works on Wide-range of Motors
Simplies Installation
PUMP POWER
VALVE CLOSING
PUMPING
No Sensitivity
For Low Loads
NO LOAD FULL LOAD
CALL NOW FOR YOUR FREE 30-DAY TRIAL 888-600-3247
Mike K. Williams is a consul-
tant with Modern Automation
Consulting Services, LLC
(6204 Woodview Pass, Mid-
land, MI 48642; Email:
mike.k.williams1952@gmail.
com). With over 39 years of ex-
perience in the CPI, Williams
provides work-process guid-
ance in the automation and
operation of batch and contin-
uous facilities. Prior to start-
ing external consulting, Williams worked for The
Dow Chemical Company in the Specialty and Ad-
vanced Materials divisions, where he consulted
on and provided analysis and investment plan-
ning for a number of new and rehabilitation capi-
tal projects. He has a B.S.Ch.E. from the Univer-
sity of Michigan, Ann Arbor and is certified in
Lean-Six Sigma methodologies.
in state-based control can respond
in advance of the last line of defense
and address the abnormal event,
when it is less than critical, before
shutting down the entire process.
This example of man-machine-
method integration provides the op-
timum response to ensure high lev-
els of both safety and productivity.
Edited by Mary Page Bailey
VALVE OPENING NO FLUID
 Solids
Environmental
Processing
Manager
Selecting Laboratory
Exhaust Systems
With upfront thought, exhaust systems can ensure
personnel safety and improve building efficiency
Paul A. Tetley
Met-Pro Corp.
A
s components of chemical re-
search facilities, fume-hood
exhaust systems for labora-
tory workstations are often
taken for granted. Yet the choice of
exhaust system, when designing or
specifying a laboratory workstation,
can have a tremendous impact on the
safety of the laboratory buildings oc-
cupants as well as those in adjacent
buildings. The choice of a fume-hood
exhaust system can also affect a build-
ings longterm operating costs.
A better upfront understanding of
the various types of exhaust systems
and their operation can eliminate po-
tentially significant safety problems
in the future while optimizing the effi-
ciency of the building and the reliabil-
ity of the system.
Exhaust system basics
The main purpose of a fume-hood ex-
haust system in a laboratory is to re-
move contaminated air resulting from
chemical fumes, reactions and other
laboratory activities. The system must
provide maximum protection while
using the least possible amount of air.
It must control the flow of two types
of air: workstation exhaust, and the
clean air that replaces it.
When designing new laboratory facil-
ities, or upgrading existing ones, safety
with regard to re-entrainment of
workstation exhaust into the building
or adjacent buildings and indoor air
quality to protect building occupants,
are the most important considerations
for a fume-hood exhaust system. In
addition, the systems ability to com-
ply with appropriate pollution-control
standards (including elimination of
unpleasant odors in the discharge
stream) is also a key consideration.
Other important considerations in-
clude: system efficiency, initial and oper-
ating costs and roofline aesthetics (Fig-
ure 1). For example, quiet, vibration-free
performance can be important to protect
delicate laboratory instruments.
Before selecting a laboratory work-
station fume-hood exhaust system,
the location of the workstation should
be carefully thought out. Its location
can significantly affect efficiency and
safety. For example, occupants in the
laboratory can create air turbulence
simply by walking past the worksta-
tion, causing contaminants to be pulled
outside of its air space. For a work-
space designed with an air diffuser di-
rectly above the fume hood, significant
air turbulence can cause contaminants
to escape into the laboratory.
In addition, airflow into the labora-
tory will affect the performance of a
fume-hood exhaust system. Since the
laboratory should maintain negative
air pressure with respect to the out-
side, the doors, windows, air vents and
other openings should remain closed
during normal operation to ensure that
any contaminants will be exhausted
through the fume hood, rather than
by other laboratory openings.
The effectiveness of a laboratory
fume-hood exhaust system can be
gauged in terms of safety provided.
The principle design parameter is to
allow safe removal of fumes away from
the operator and from the occupants of
the laboratory. The velocity of the air
entering the hood from the room is the
key factor in achieving this safety. If
the velocity is too low, the fumes can
escape the hood and enter the room. If
the velocity is too high, air turbulence
Chemical Engineering www.che.com September 2010
Figure 1. Roofline aesthetics are
among a host of considerations when
selecting a fume-hood exhaust system
for a laboratory facility
at the sash (damper) can occur, which
may push fumes into the room. Airflow
velocities that optimize safety can vary
by manufacturer, but typically are in
the range of 100 ft/min.
Exhaust flow from a fume hood should
be greater than the supply air to create
outside-in air movement and thereby
contain airborne contaminants.
Workstation fume-hood exhaust sys-
tems can be either constant-air-volume
(CAV) or variable-air-volume (VAV) de-
signs. A CAV fume hood exhausts air at
a constant rate, with an airflow velocity
that increases as the sash is lowered. A
VAV laboratory fume hood attempts to
eliminate the variable velocity by vary-
ing the air volume exhausted through
the hood. In doing so, it can reduce the
total supply of exhaust air, and con-
tribute to reduced operating costs. VAV
fume hoods use a closed-loop exhaust
system to monitor the amount of air
being exhausted in order to maintain
the required average face velocity. The
control system also helps monitor the
presence of contaminants. In addition,
many VAV hoods use visual and audi-
ble alarms and gages to warn of inad-
equate air velocity or exhaust system
malfunctions.
Exhaust system fans
Laboratories can choose from a number
of fume-hood exhaust-system designs
based on the requirements of the facil-
ity. Exhaust systems rely on hidden,
but essential subsystems, including ex-
haust fans and ventilation systems.
Fume-hood exhaust systems typi-
cally employ a network of air ducts
(plenums) leading to a roof-mounted
exhaust system that is driven by one
of three fan technologies: axial, cen-
43
 Environmental Manager
trifugal, and mixed-flow impeller. The
fans in each of these categories can
be direct-drive designs, with the fan
wheel mounted directly on the motor
shaft, or belt-driven, with the fan wheel
mounted on a separate drive shaft and
driven by power from the motor trans-
ferred through one or more belts.
Axial fan systems Axial fan blades
draw air into fan housings and dis-
charge air in the same axial direction
ideal for moving large volumes of air,
rated at cubic feet per minute (ft3/min)
at low pressures, and also rated in static
pressure (SP). Fan blades used in axial
fans are typically plastic, aluminum,
stainless steel or fiberglass. Different
types of axial fans used for industrial
applications include panel fans, which
are wall-mounted to draw air into or
exhaust air out of a room, and tube
axial fans, which are connected to a se-
ries of ductwork conduits for exhausting
contaminated or hot air from a location.
Centrifugal fan systems Centrifugal
fans have blade wheels mounted inside
a housing. The blades can be forward-
curved, backward-curved or straight.
Exhaust air is pulled into the inlet of
the blower housing and deflected by the
fan blades into a spiral pattern. This
pattern of airflow is then focused by
the scroll housing and directed into a
single outlet airstream. The airstream
leaves the fans housing at a right
angle to the axis of the housing assem-
bly. A variation on the traditional, cen-
trifugal scroll-type fan housing is the
tubular centrifugal fan, which uses a
cylindrical housing rather than a scroll
housing and cutoff assembly.
A centrifugal fan is designed to work
at fairly high revolutions per minute
and with a tall, roof-mounted exhaust
stack in order to discharge the exhaust
air at sufficient height above a facility
to prevent re-entrainment of exhaust
air (and its possible noxious fumes or
unpleasant odors) into the building
and surrounding facilities. Because of
the size and height of the stack, it re-
quires expensive and complex mount-
ing hardware for stability and to mini-
mize vibration. This includes elbows,
flex connectors, spring vibration isola-
tors and supporting guy wires.
Mixed-flow impeller systems Mixed-
flow impeller systems generally pro-
vide greater airflow than centrifugal
44 Chemical Engineering www.che.com September 2010
fans, and greater air pressure than
axial fans. These systems eliminate re-
entrainment of exhaust air and odors
into a facility and surrounding build-
ings with improved indoor air quality
(IAQ). This technology typically incor-
porates low-profile, roof-mounted ex-
haust systems with lower-speed fans
as compared to the other technolo-
gies. To minimize noise levels, these
systems employ low-vibration, direct-
drive motors to draw in the exhaust
air from a laboratory fume hood.
A noticeable difference between ex-
haust systems with centrifugal fans
and those with mixed-flow impellers
is the height of the exhaust stacks:
mixed-flow impeller systems present a
lower profile and smaller footprint on
the roof (Figure 2). This characteristic
has obvious value in accommodating
space and weight restrictions, as well
as in its aesthetic qualities (more and
more communities are passing ordi-
nances limiting overall building height,
including rooftop appurtenances).
As the name implies, a mixed-flow
impeller system combines the best at-
tributes of both axial and centrifugal
air movement technologies. Impellers
in this approach have high efficiency,
consume less energy and are typically
quieter. In addition, true mixed-flow
technology is stable throughout the
impeller performance curve. Therefore,
these impellers do not have a stall or un-
stable performance region, as is found
on axial and centrifugal fan curves in
regions of higher pressure. The concern
is that systems operating at or near
this unstable performance region can
give rise to significant performance,
sound and vibration problems.
Mixed-flow impellers blend the ex-
haust air from a laboratory fume hood
with outside, ambient air to reduce the
concentration of contaminated particu-
lates, as measured in parts per million
(ppm) or milligrams per cubic meter
(mg/m3) to a level that can be safely
and legally exhausted to the atmo-
sphere. These fans draw exhaust air
FIGURE 2. A difference
between exhaust system
types is the height of
rooftop stacks, a charac-
teristic with value in ac-
commodating space and
weight restrictions
into a network of ducts that transport
it to the outside of the facility, usually
to the roof, where it is mixed with am-
bient air. In some cases, particularly
where local ordinances dictate rooftop
appearances, mixed-flow impeller sys-
tems may be mounted in an enclosed
facility (this is usually done for aes-
thetic reasons).
Using base-mounted bypass damp-
ers, a mixed-flow impeller can draw in
twice as much fresh air as exhaust air,
resulting in a much lower concentra-
tion of contaminants and odor-causing
elements to be treated from the labo-
ratory fume hood.
Release to atmosphere
Because a mixed-flow-impeller ex-
haust system can generate relatively
high escape velocities, even at fairly
low fan speeds, additional fresh air is
entrained into the exhaust mixture
when it is expelled from the rooftop
stack by means of a nozzle/windband
section. This mechanism creates a pro-
tective envelope around the discharged
exhaust-air mixture and leaves the
surrounding air stationary to improve
entrainment. The exit velocities of
these escape gases can exceed 6,300
ft/min and can send an exhaust plume
more than 120 ft into the air, even into
steady crosswinds. Of course, a variety
of factors can determine the effective-
ness of the system, including building
architecture, surrounding terrain and
even weather conditions.
As an example, a mixed-flow impel-
ler moving an exhaust and clean air
mixture at 80,000 ft3/min will typi-
cally generate 6,300 ft/min exit veloc-
ity. According to wind tunnel research,
additional air is entrained into an air
plume through aspiration, even at exit
velocities of 3,000 ft/min (the mini-
mum velocity specified by the ANSI
Z9.5 standard for clearing chemical
contaminants in laboratory ventila-
tion systems).
A belt-driven, centrifugal-fan ex-
haust system and duct network can
also mix contaminated and clean air
similar to a mixed-flow impeller, with
useful dilution effects when released
into the atmosphere. But the centrif-
ugal fan approach requires a rooftop
stack with sufficient height, some-
times as high as 40 ft or more. For the
same dilution and operating efficiency,
a mixed-flow impeller system will typ-
ically rise only about 15 ft above the
roofline. Because of the lower height,
the mixed-flow impeller exhaust sys-
tem would not require the extra stabi-
lizing hardware needed with centrifu-
gal fan exhaust systems.
Tall exhaust stacks are not the only
structures associated with centrifugal
fan systems. Often, centrifugal fan
systems also need a rooftop penthouse,
mainly to protect service personnel.
Because centrifugal fan systems rely
on belt-drive motors in contrast to
the more reliable direct-drive motors
of mixed-flow impellers periodic
maintenance visits are required to ad-
just belt tension or to replace pulleys
or failed belts. The penthouse enclo-
sure maintains safe, workable condi-
tions for maintenance personnel.
In terms of reliability, laboratory
workstation fume-hood exhaust sys-
tems based on mixed-flow impellers
are relatively simple and maintenance-
free compared to the belt-driven fans
and blowers of typical centrifugal and
axial fan systems. The belts alone in
a belt-drive system, which can stretch,
loosen and break over time, require
regular maintenance and repair, re-
sulting in shorter expected operating
lifetimes. For example, the minimum
expected lifetimes (L10) of bearings
in belt-driven systems are typically
40,000 h, while the expected operat-
ing lifetime of direct-drive motor bear-
ings can be at least 200,000 h for some
mixed-flow impeller systems.
A quieter solution
Noise management is also an important
consideration for most laboratories and
areas in some cases requiring firms
to perform noise studies at their prop-
erty lines to determine the noise levels
of a facilitys various systems, includ-
ing the laboratory exhaust system. The
lower speeds and inherently quieter
design configuration of the direct-drive
motors in mixed-flow-impeller exhaust
systems, compared to axial and cen-
trifugal fan systems, usually translate
into lower noise and vibration levels. In
addition, for applications where noise
must be tightly controlled, mixed-flow
impeller systems can be designed with
integrated noise attenuators that do
not increase the overall height of the
roof exhaust fan.
Axial and centrifugal fan exhaust
systems can also be equipped with
noise-reduction attachments, such as
a discharge silencer, added onto the
exhaust stack. But this adds to the al-
ready tall stack height, and requires
additional mounting hardware to mini-
mize vibration induced in the building.
Even at their lower motor speeds, there
is little sacrifice in performance with
a mixed-flow impeller. It can provide
greater airflow than a centrifugal fan,
and greater air pressure than an axial
fan, at lower rotational speeds.
Compared to belt-driven centrifugal
and axial fans, direct-drive mixed-flow
impeller fans are energy-efficient and
cost-effective choices for fume hoods,
with about 25% greater energy effi-
ciency than belt-driven centrifugal fan
systems. Improperly aligned shafts,
pulleys, pillow blocks and motors in a
belt-driven system can rob it of power,
and can introduce reliability issues.
Also, the outlet nozzle of a mixed-flow
impeller system can be designed for
high exhaust-gas exit velocities even
at lower fan speeds and corresponding
reduced power requirements.
For further energy efficiency, it may
be necessary to specify a variable-speed
drive system. The use of variable-speed
drives can contribute to lower energy
costs, running at slower fan speeds at
night when laboratories are unoccu-
pied. In addition, if one fan fails in a
multiple-fan exhaust system, the re-
maining fans increase speed to com-
pensate. This is critical in many bio-
safety level (BSL) laboratories, where
redundancy is mandatory.
In addition, energy savings are also
possible by recovering some of the heat
normally lost through exhausted air.
Mixed-flow impellers can be equipped
with unique heat-recovery modules
that extract warmed air from the ex-
haust flow and transfer the heat energy
to the intake of a buildings ventilation
system, significantly reducing heating
Chemical Engineering www.che.com September 2010
fuel costs for the building (for every 1F
in added heat, there is a corresponding
3% reduction in heating fuel costs).
Ductless laboratory fume hoods
The three major laboratory fume-hood
exhaust systems discussed so far as-
sume the use of ductwork networks to
vent contaminated air from the labo-
ratory. In recent years, ductless labo-
ratory fume hoods have emerged as
lower-profile alternatives. In ductless
designs, the hood incorporates a carbon-
bed or other type of filter to remove and
control contaminants. The filter media
must be matched to the types of chemi-
cals used in a laboratory. General-pur-
pose filters will adsorb a percentage of
most chemicals and gases, while par-
ticulate matter filters may be needed
for some applications. In some cases, a
chemical impregnate may be needed in
the filter to alter the target chemical to
a less hazardous material.
A ductless, laboratory fume hood has
no external connections other than its
power cord, making it extremely por-
table and useful for facilities that re-
quire more than occasional redesign.
The power cord connects to electronic
systems that monitor airflow and filter
conditions for operator safety, as well
as provide visual and audible alarms
when needed. A ductless approach is
not a substitute for all ducted systems,
and should never be used in any chemi-
cal laboratories involving unknown
results (for which the filter may not
work). Admittedly, ductless laboratory
fume hoods sacrifice efficiency for por-
tability, especially compared to a cen-
trifugal or mixed-flow impeller system,
and they require careful monitoring of
the carbon filters condition so that re-
placement can occur once the filter has
become saturated. 
Edited by Scott Jenkins
Author
Paul A. Tetley is execu-
tive vice president of Met-
Pro Corp. (160 Cassell Rd.,
Harleysville, PA 19438;
Phone: 215-723-4700; Email:
p t e t l e y @ s t r o b i c a i r. c o m ) .
Prior to his current position,
Tetley was vice president and
general manager at Strobic
Air Corp., a Met-Pro subsid-
iary. Since joining Strobic Air
in 1989 as engineering pro-
duction manager, he has designed or invented
many innovative Tri-Stack fan systems, an
acoustical silencer nozzle, and a unique multi-
fan plenum system.
45

Solids Processing
Environmental Manager
Vacuum Systems:
Recommendations
For Safe Operation
Follow this guidance to ensure that steam ejector
systems, mechanical vacuum pumps and integrated
vacuum systems are designed, operated and
maintained to ensure process safety
Stanley S. Grossel
Process Safety & Design Consultant
T
his article is the first in a two-
part series, in which the types
of process vacuum equipment
(steam ejector systems and
mechanical vacuum pumps) are dis-
cussed.1 Emphasis is given on how
they are constructed and operate, and
how to design associated parts of the
systems (such as foundations, suction
piping, discharge piping, and so on), to
ensure safe operation.
Many unit operations including
distillation, evaporation, drying, crys-
tallization, filtration, and others are
often conducted under vacuum condi-
tions. To achieve the desired vacuum,
the following three process- vacuum
systems can be used:
Steam ejector systems
Mechanical vacuum pumps
Integrated vacuum systems (which
combine steam ejectors and me-
chanical vacuum pumps)
Available capacities and operating
ranges for vacuum pumps and vacuum
pumping systems that are most often
used in process operations throughout
the chemical process industries (CPI)
are listed in the Table on p. 60 [1].
This article reviews the hazards as-
sociated with steam ejector systems
and mechanical vacuum pumps, and
recommends design and operating
practices that can be taken to prevent
or mitigate these dangers.
1. Editors note: This is the first half of a two-
part article. Part 2 is scheduled to appear in the
October 2012 issue.
Suction
Condensate
Steam ejector systems
Steam ejector systems are gener-
ally categorized into one of four basic
types: single-stage, multi-stage non-
condensing, multi-stage condensing,
and multi-stage with both condensing
and non-condensing stages.
For many years, steam ejector sys-
tems dominated process applications
thanks to their simplicity, low pur-
chase cost, good reliability and their
ability to be designed for very large ca-
pacities (in excess of 1,000,000 acfm)
and be operated at very low pressures
in the micron range (for instance, six-
stage units are routinely designed for
suction pressures in the range of 310
microns). They also are available in a
variety of materials of construction.
Steam ejector systems are ideal for
wet-vacuum and highly corrosive ap-
plications. As a result, they are still
being used instead of mechanical
vacuum pumps in applications that
require a reliable vacuum system that
can tolerate corrosive chemicals, liq-
uid slugs and solids carryover.
Safety considerations. In gen-
eral, there are very few safety prob-
lems associated with the operation
of steam ejector vacuum systems.
However, one safety issue associated
with their design is the potential for
back-streaming of steam into the pro-
cess equipment.
For instance, steam ejectors nor-
mally use a steam bleed to control
the suction pressure. If the process
Chemical Engineering www.che.com september 2012
PI
Steam
supply
Cooling water
T Condensate
supply
PI
X
Cooling water
return
T
YZ intercondenser
Y
Z
Aftercondenser
TI
To seal pot
TI
To seal pot
Gate valve PI Pressure indicator
Four-way valve TI Temperature indicator
T Steam trap Piccolo connection
FIGURE 1. Shown here is a schematic
of a three-stage steam ejector system
with inter-condensers (Reprinted with
permission from Ref. 1)
contains a water-reactive chemical
(such as acetic anhydride), a steam
bleed is not acceptable. If the ejector
should backfire, this will push steam
back into the process and could result
in a possibly violent reaction, which
could overpressure the system.
To avoid this problem, a nitrogen
bleed can be used to control the suc-
tion pressure. If the jet should back
fire, nitrogen will prevent the steam
from reaching the process. However,
using nitrogen to control an existing
multi-stage ejector may not be techni-
cally feasible. If nitrogen is to be used
to control a multi-stage jet, the user
must indicate this when writing the
performance and equipment specifica-
tions for the steam ejector.
Using nitrogen to control the suc-
tion pressure of the first stage that
which is closest to the process will
affect the design of subsequent stages,
and using nitrogen to control suction
pressure can significantly affect the
design of the inter-condensers. If the
back flow of steam or water into the
process equipment from the ejector
system can cause a serious safety or
quality problem, then a liquid separa-
tor (a so-called knockout pot) may be
installed between the ejector system
and the process equipment.
Another possible safety problem
may result from the loss of the inter-
condenser cooling medium (usually
water), which could lead to overpres-
sure of the system.
59
 Environmental Manager

In instances where suction gases, Table 1. Capacity and operating range for vacuum pumps
or materials entrained in the suction and vacuum pumping systems commonly used in process
gases, might freeze inside the ejector, applications
ejectors with external heating jackets Type Blind or Lower limit for Single unit
can be specified. base pres- process appli- capacity
Instrumentation. It is good engi- sure cations range, ft3/min
neering practice to provide adequate Single ejectors
instrumentation to monitor and con- One stage 50 torr 75 torr 101,000,000
trol the operation of the steam ejector Two stages 4 torr 10 torr
system. The following instrumenta- Three stages 800 micron* 1.5 torr
tion items are recommended for moni- Four stages 100 micron 250 micron
toring and controlling the vacuum Five stages 10 micron 50 micron
system, and to support troubleshoot-
Six stages 1 micron 3 micron
ing efforts:
Liquid-ring pumps
Control system for controlling the
60F water-sealed:
steam bleed to maintain the desired
vacuum level One stage 50 torr 75 torr 318,000
Pressure gage on the main steam- Two stages 20 torr 40 torr
supply line Oil-sealed 4 torr 10 torr
Steam pressure gage on each Air ejector first stage 2 torr 10 torr
ejector Rotary-piston pumps
Suction pressure gage on the inlet to One stage 5 micron 100 micron 3800
each ejector Two stages 0.001 micron 10 micron
Vacuum gage to monitor absolute Rotary-vane pumps
pressure at the process vessel Operated as a dry compres- 20 torr 50 torr 206,000
A flowmeter to monitor the water sor
flow to each inter-condenser Oil-sealed, rough-vacuum 0.5 torr 20 torr 50800
Temperature gages on the inlet and pump
outlet water lines to and from each Oil-sealed, high-vacuum
shell-and-tube inter-condenser pump:
Temperature indicator in the baro- One stage 5 micron 100 micron 3150
metric leg from each barometric Two stages 0.001 micron 10 micron
condenser Rotary-lobe blowers:
One stage 100 torr** 300 torr 3030,000
Mechanical vacuum pumps Two stages 10 torr** 60 torr
Design options. Mechanical vacuum
Integrated pumping systems
pumps usually are classified as either
Ejector-liquid ring pump 1 micron 3 micron 100100,000
wet pumps or dry pumps. The fol-
lowing lists show which mechanical Rotary-blower, liquid-ring 1.0 torr 5 torr 10010,000
pump
vacuum pumps are wet and which are
Rotary-blower, rotary-piston 0.1 micron 0.10 torr*** 10030,000
dry [2]: pump
Wet pumps:
Rotary-blower, rotary-vane 20 micron 200 10030,000
Oil-sealed rotary vane (single- and pump micron****
dual-stage) * 1.0 micron = 0.001 torr
Oil-sealed rotary piston **Based on intercooled design that uses gas admitted to a trapped dis-
Liquid ring charge pocket to cool the blower
Dry pumps: ***Based on using a two-stage, rotary-piston pump as the backing pump
****Based on a two-stage, rough-vacuum, rotary-vane design that exhibits a
Rotary claw
base pressure of approximately 0.5 torr
Rotary lobe (Roots)
Source: Ref. 1
Rotary scroll
Rotary screw
Rotary vane tween rotary components and hous- During operation, an oil-sealed
Rotary piston ing to reduce gas slippage pump compresses the gas to a pres-
Diaphragm To transfer the heat of gas compres- sure that is slightly higher than atmo-
Oil-sealed vacuum pumps. Both sion away from wear surfaces spheric to expel it through the exhaust
rotary-vane and rotary-piston To lubricate the rotary internals valve. At this pressure, and at an op-
vacuum pumps rely on oil for the To flush away moisture erating temperature of 70100C, ac-
following functions: To inhibit corrosion of internal tive gases readily attack pump bear-
To seal the internal clearances be- parts ings, seal materials and the oil. The
60 Chemical Engineering www.che.com september 2012
 Separator
baffle with
Seal screen
0.0001-in.
clearance Inlet
Outlet
Discharge
valve
Inlet
Slide
Rotor Vanes pin
Stator
FIGURE 2. Shown here is a schematic of the internals of a rotary-vane vacuum
pump (left) and a rotary-piston vacuum pump (right) (Reprinted with permission
from Ref. 1)
selection of an appropriate oil for the
application is critical for good pump
maintenance.
This dependency on the oil for inter-
nal lubrication is a limiting factor in
the use of these types of pumps in the
CPI. The integrity of the oil must be
maintained to avoid internal damage
that could cause contaminant buildup.
Particulates in the oil must be filtered
out ahead of the vacuum pump, and
water or solvent vapors must either be
removed ahead of the pump by using
pre-condensers, or prevented from
condensing within the pump by the
use of gas ballasting (air stripping) or
oil distillation.
For oil-sealed, rotary-vane vacuum
pumps, a variety of oils are available,
each with very different properties.
The sealing oil used in such pumps
that are destined for chemical service
has several additional requirements.
For instance, it must:
Be inert to chemical attack by
acids, bases, halogens and halogen-
induced aging)
Be resistant to strong oxidants such
as oxygen, chlorine and fluorine
Possess good solvent power for prob-
lematic, mostly organic constituents
of pumped media, such as oligomers,
polymers, resins, crystalline decom-
position or reaction products
Both oil-sealed, rotary-vane and pis-
ton vacuum pumps need periodic oil
changes. The chief purpose of chang-
ing of the oil is to remove particles that
either originate from aging of the oil
or particles that are generated by the
process (such as dust, decomposition
products, and other contaminants).
Discharge
Oil level
view port
Piston
Automatic
decanting
Cam
arrangement
When either an oil-sealed rotary-
vane or rotary-piston vacuum pump
is improperly installed a hazardous
condition can result. The following
installation practices will minimize
potential problems with these types of
oil-sealed vacuum pumps [1]:
Foundations. The foundation design
will depend on the type and size of the
oil-sealed pump. If the pump is inher-
ently balanced, it can be mounted on
any floor that is able to support the
weight of the pump. However, pumps
that are partially balanced, should
be installed with vibration mounts
or matting between the base and the
floor. If the pump is not balanced, the
base should be securely fastened to a
concrete foundation with anchor bolts
and grouting. The pump manufacturer
should be consulted when designing
the pump foundation. It is always good
engineering practice to install flexible,
bellows-type connectors in the suction
and discharge piping; these can help
to isolate the rest of the system from
pump vibration.
Suction piping. Certain precautions
must be taken when installing the suc-
tion piping for an oil-sealed vacuum
pump. Care must be taken to prevent
any backstreaming of pump oil into
the process vessel and to prevent pro-
cess liquids and solids from reaching
the pump. Backstreaming of oil is es-
pecially significant with larger pumps.
Baffles and traps are frequently in-
stalled in the suction line to prevent
this from happening.
Cold traps, knockout pots, wet scrub-
bers, and bag filters are some of the
devices used to protect the pump from
Chemical Engineering www.che.com september 2012
liquids and particulates in the suc-
tion stream. The degree of protection
necessary will depend on the process
application. If the upstream knock-
out pot is not continuously drained, a
high-level sensor should be installed
to automatically stop the pump or
close an isolation valve to prevent any
overflow from the knockout pot from
reaching the pump.
Oil-sealed vacuum pumps must be
protected from flooding if the oil flow
does not automatically stop when the
pump is shut down. To prevent flood-
ing, a valve can be installed in the
suction line to admit air or nitrogen
to break the vacuum when the pump
stops. This prevents the siphoning of
oil from the reservoir when the oil is
not being discharged from the pump
by the normal pumping action. A zero-
speed switch or solenoid valve is used
to actuate the air or nitrogen bleed
and close the isolation valve. Closing
the isolation valve allows the process
to remain under vacuum and allows
the vacuum in the pump to be broken
more quickly.
Flooding may be avoided with some
oil-sealed pumps by sealing the dis-
charge line using a barometric leg
or a discharge check valve. This al-
lows the system to pull a vacuum on
the oil reservoir so that oil siphoning
stops. A check valve or knockout pot
must be installed in the suction line to
prevent reverse flow from carrying oil
into the process vessel. A knockout pot
is recommended for critical applica-
tions where carryover of oil into the
process vessel might result in a fire
or explosion.
Discharge piping. Condensable vapors
that pass through an oil-sealed pump
tend to condense in the discharge line.
The discharge piping should slope
away from the pump, and a drip-leg
or knockout pot should be installed
in the discharge line to prevent the
backflow of condensate into the pump.
Backpressure on the pump should
be minimized.
Serious mechanical damage can re-
sult if the pump is operated against
a closed discharge valve. Discharge
valves are therefore not recommended
unless the pump must discharge into
a common exhaust header. If a dis-
charge valve is required, an interlock
61
 Environmental Manager
(that ensures that the exhaust valve is
open before the pump can be started)
is recommended; in fact, this is a con-
dition of the warranty for some vac-
uum-pump manufacturers.
A fine mist of oil drops is emitted
with the exhaust gases that are dis-
charged from an oil-sealed pump. Ex-
haust filters are normally installed
to eliminate this oil mist. The most
popular design has removable filter
elements and can be mounted any-
where in the discharge line. Units are
available that can remove up to 99.9%
of the exhausted oil mist. Oil mist fil-
ters should not be allowed to become
blocked, as hazardous over-pressures
could occur. A differential pressure
sensor and alarm can be installed on
oil mist filters to monitor their opera-
tion and detect potential blockages.
If the oil mist is contaminated with
organic compounds from the process
gas, especially if they are toxic, then
an activated carbon filter should be in-
stalled after the oil mist filter. The ac-
tivated carbon filter will remove these
contaminants by adsorption.
Pump cooling. Oil-sealed pumps must
be cooled to remove the absorbed heat
of compression. Both water- and air-
cooled models are available. Air-cooled
models should be installed in an area
that will allow an ample flow of fresh
air to reach the pump. The jackets on
water-cooled pumps usually contain
small clearances that can plug with
solids. A strainer or filter should be
installed in the cooling water supply
line to remove any solids. A regulating
valve should be installed in the cool-
ing-water supply line to control the
pump operating temperature. A high-
temperature cutout switch is some-
times installed to protect the vacuum
pump from overheating.
Movement of oil and flammable
vapor droplets from oil-sealed vacuum
pumps has been known to generate a
static charge, which could lead to a fire
or explosion [3]. To avoid or minimize
this potential hazard, it is critical to
inert the exhaust stream with nitro-
gen to eliminate oxygen or reduce it to
below the limiting oxygen concentra-
tion (LOC).
Also, in many cases, it is prudent to
use nitrogen to dilute the gas stream
upstream of the pump inlet to move it
62 Chemical Engineering www.che.com september 2012
out of the flammable range. It is criti-
cal that the nitrogen supply source be
free of oxygen. There have been a few
instances where the nitrogen supply
was compromised, allowing in air that
can lead to an explosion [3]. For pyro-
phoric gases, such as silane, special
care must be taken to prevent air or
oxygen from entering the system.
Liquid-ring vacuum pumps
Liquid-ring vacuum pumps are widely
used in the CPI. They are available
as single-stage (one or two impellers
in parallel) or two-stage (two impel-
lers in series) units. This type of pump
does not require internal lubrication
of the impellers, since they do not con-
tact the housing. The sealant liquid,
used for both sealing and cooling, can
be virtually any liquid that is compat-
ible with the process gas. For instance,
water, low-viscosity oils, glycols and
many process solvents such as tolu-
ene, xylene, methanol, ethanol, pro-
panol, butanol and ethylbenzene have
been used as sealants in liquid-ring
pumps. Other liquids have also been
used, such as ethylene glycol and pro-
pylene glycol, and even sulfuric acid.
The motor horsepower must be cor-
rected for the sealant viscosity and
density (see Reference 1, p. 273). These
sealants can be recirculated in a full
recovery system that includes a gas-
liquid separator tank and a heat ex-
changer to cool the sealant liquid. Even
higher vapor-pressure sealants can
be used if a low-temperature coolant
(such as glycol solution) is used in the
heat exchanger to reduce the sealant
liquid temperature. This recovery sys-
tem allows process materials to be col-
lected in the pump and either returned
to the process or collected for disposal,
thereby minimizing contamination of
other liquids or the environment.
The most advantageous property of
liquid-ring vacuum pumps is the fact
that they are nearly isothermal. The
liquid in the pump provides an ample
heat sink for the removal of the heat
of compression. This means that the
temperature rise in the casing is very
small. It is a function of the charac-
teristics of the sealant fluid used, its
temperature, flowrate, and of course,
the process gas conditions. The result-
ing low internal temperatures provide
improved safety, especially when com-
pressing explosive gases and fluids.
Generally, internal temperatures
in liquid-ring vacuum pumps are the
lowest of most mechanical vacuum
pumps. Hence, gases exit at almost
the same temperature, or in some
cases, at a lower temperature than
they enter the vacuum pump. Further,
since the gases handled are in direct
contact with large amounts of seal-
ant liquid, even if touching of inter-
nal parts (the rotor with the casing)
occurs, the resulting sparks are not
hot enough or of sufficient duration
to cause explosions or ignition of even
the most hazardous gases. This pro-
vides a significant safety advantage,
especially when handling easily ignit-
able gas streams. The occurrence of an
explosion within liquid-ring vacuum
pumps handling flammable vapors is
very unlikely as the sealant liquid in
the pump casing acts as a hydraulic
flame arrester, quenching any flame
that may be generated and preventing
it from propagating.
However, one of the major drawbacks
of liquid-ring vacuum pumps is that
they must cope with cavitation when
running at low inlet pressure. If cavi-
tation is allowed to continue over long
periods of time, serious damage can be
done to the pump. The amount of cavi-
tation can be affected by the sealant
liquid, sealant temperature, impeller
speed, blade angle and inlet pressure.
For a given pump and sealant liquid,
cavitation can normally be suppressed
by bleeding air (or nitrogen) into the
pump inlet to raise its total pressure
above the vapor pressure of the seal-
ant at operating temperatures.
All liquid-ring pump units have aux-
iliary equipment such as a vapor-liq-
uid separator, a sealant recirculation
pump, a sealant cooler, and various
instrumentation items, relief devices
and valving.
Some advantages and disadvan-
tages of liquid-ring pumps are sum-
marized below:
Advantages:
Has a simple design that uses only
one rotating assembly
Can be fabricated from any castable
material
Creates minimal noise and
vibration

1 In this sector liquid moves outward,
draws gas from inlet ports
into rotor chambers
Rotating
liquid
compressant
3 In this sector
compressed gas
escapes at discharge ports
Experiences little increase in the
temperature of the discharged gas
Experiences no damage from liquid
or small particles in the process gas
Enables simple maintenance and
rebuilding
Uses a relatively slow rotational
speed (1,800 rpm or less), which
maximizes operational life
Can use any type of liquid for the
sealant fluid in situations where
co-mingling of the sealant with the
process vapor is permissible
Has no lubricating liquid in the pro-
cess equipment to be contaminated
Can accommodate both condensable
and non-condensable gases while
operating as both a vacuum pump
and condenser
Disadvantages:
Mixing of the condensed gas/vapor
with the sealing liquid, which re-
quires external equipment to sepa-
rate them
The risk of cavitation requires a
portion of the process load to be
noncondensable under operating
conditions
High power consumption requires
large motors to form and maintain
the liquid ring
Achievable vacuum is limited by the
vapor pressure of the sealant liquid
at the operating temperature
The power consumption of a liquid-
ring pump is normally higher than
that of other types of mechanical
vacuum pumps
The proper installation of a liquid-ring
vacuum pump is critical to its opera-
tion and maintenance. The following
recommendations should be followed
to ensure proper installation [1, 4]:
Foundations and pump checking.
2 In this sector liquid moves inward,
compresses gas in rotor chambers
Inlet
port
Rotor
Discharge
ports
FIGURE 3.
This figure
shows a cross-
sectional view
(above) and
general view of
assembly (right)
of a liquid-ring
vacuum pump
(Reprinted with
permission from
Ref. 1)
Since liquid-ring pumps do not usu-
ally create vibration problems, special
foundations are not required. Pumps
that are about 50 hp or above are best
placed on a concrete pad. Smaller units
may be mounted on existing floors and
skids. On pumps of all sizes, the base
should be leveled and the coupling
alignment should be checked during
installation. On belt-driven models,
the sheaves should be aligned and the
proper v-belt tensions set according to
the manufacturers instructions. After
the motor has been wired, the pump
rotation must be checked to ensure
that the polarity is not reversed. Do
not test the polarity without liquid in
the pump.
Suction piping. A check valve or a
valve that closes automatically when
the motor is shut down should be in-
stalled in the suction line to prevent
dumping of the liquid from the pump
back into the upstream process ves-
sel in the event of pump failure. This
check valve will also serve to prevent
any backflow of air into the upstream
process vessel.
Where protection against back-
streaming air is critical due to process
requirements, the vent and drain on
Chemical Engineering www.che.com september 2012
Inlet
connections
Body
Liquid
Inlet
port
Discharge
connections
the separator should be barometrically
sealed. This can be accomplished by
discharging to a seal pot or by install-
ing a vertical run in the suction line to
serve as a barometric leg between the
pump and the process vessel. If the
suction line is used as a barometric
leg, the vacuum pump discharge must
be sealed beneath the liquid level in
the separator.
A vacuum relief device should be
installed in the suction line on pumps
that are not automatically controlled.
Otherwise, the pump may cavitate dur-
ing operation at low suction pressures
or when it is dead-headed by mistake.
The vacuum relief valve should be con-
nected to a supply of inert gas (usually
nitrogen) if admitting air to the sys-
tem is unacceptable.
If inlet spray nozzles are used to
condense incoming vapors, the nozzles
should be located in the suction piping
as close as possible to the pump. An
enlarged pipe section or a small vessel
may be needed to provide contacting
space between the spray liquid and in-
coming vapors.
Discharge piping. The pump and
discharge piping must be installed to
minimize backpressure on the pump.
63
 Environmental Manager
Both the compressed gases and the
sealing liquid are discharged through
the pump discharge connection, and
the discharge piping must be sized
for two-phase flow. Piping of the same
diameter as the discharge connection
will usually be adequate, but a larger
line may be required if the distance
to the vapor-liquid separator is more
than a few feet. The discharge pip-
ing should never rise more than 23
ft above the pump before the gas and
liquid are separated. The discharge
piping should be designed so that it is
free to drain after a shutdown.
Pump draining. The pump casing
should be drained to the shaft level
before a liquid-ring pump is started.
Starting the pump when it is com-
pletely filled with liquid will overload
the motor and place undue mechanical
stresses on the shaft and rotor. Shaft
failure is not uncommon on large liq-
uid-ring pumps operated in this way.
A solenoid valve installed in the seal-
ant supply line to the pump is the best
way to stop the flow of sealant when the
pump is shut down. It will not be neces-
sary to drain any residual liquid from
the casing. If sealant is recirculated
from the vapor-liquid separator, the
sealant will automatically drain to the
shaft level if the separator is designed
with an overflow nozzle at the pump
shaft level. The sealant may be manu-
ally drained to the shaft level through
a hole drilled in the pump casing or by
a drain in the sealant supply line.
If the pump is located outdoors or
in unheated buildings, it should be
completely drained for freeze protec-
tion during prolonged shutdowns.
Similarly, the pump must never be re-
started dry; it must be partially filled
with sealant liquid. This may neces-
sitate a bypass around the solenoid
valve in the sealant supply line.
Sealant piping. There are three pos-
sible arrangements for sealant flow
to liquid ring pumps: Once-through
(no recovery), closed loop (recirculated
sealant) and partial recirculation sys-
tem (partial recovery).
Once-through arrangement. A once-
through installation does not neces-
sarily imply that none of the sealant
is recovered; rather, the term simply
refers to the absence of any return
piping from the vapor-liquid separa-
64 Chemical Engineering www.che.com september 2012
tor drain to the pump-sealant supply
connection. This is a common arrange-
ment where conservation or con-
tamination of the seal liquid is not a
concern.
The line supplying sealant to the
pump should contain a shut-off valve
followed by a strainer with a blow-
down valve to remove particulates. An
automatic solenoid valve may be in-
stalled in the supply line, interlocked
with the pump motors operation, to
prevent flow of sealant to the pump
in case the motor stops running, (in
this case, if the motor stops, the valve
closes to prevent the pump casing
from filling with liquid).
If a solenoid valve is used, a bypass
should be installed to allow some seal-
ant to be introduced before starting
(when the pump has been completely
drained). The sealant flow can be reg-
ulated manually or automatically.
The manual method regulates the
flow by a globe valve (upstream of
an orifice), which uses pressure drop
across the orifice to set the desired
pressure to the pump. The orifice is
usually chosen so that the proper flow
to the pump is set by maintaining a
pressure drop of 5 to 10 psi across the
orifice. Automatic control is achieved
by using a flow controller to maintain
a constant pressure drop if the sealant
Edwards
Vacuum
FIGURE 4. This is a cutaway
view of a rotary-claw type dry
vacuum pump, showing all
components of the pump
supply pressure tends to fluctuate.
Closed loop (total recirculated seal-
ant) arrangement. In this arrange-
ment the system is designed to oper-
ate as a closed loop with no addition
of fresh sealant. Only the vapors that
condense in the pump need to be re-
moved from the system. In actual
practice, however, some fresh sealant
is often continually added to control
the concentration of low boilers or
other contaminants in the sealant or
to make up for evaporative losses from
the system.
A heat exchanger is installed in this
system to remove the heat of compres-
sion and condensation from the seal-
ant before it is reintroduced into the
pump. The heat exchanger is normally
water cooled, but any suitable cool-
ant can be used. Often, the system is
provided with a sealant recirculation
pump. This allows more flexibility in
the design and operation of the vac-
uum system. The sealant cooler can be
designed for a higher pressure drop,
which results in better heat transfer
and reduced fouling. A recirculation
pump must operate at suction pres-
sures that are too high to siphon
an adequate flow of sealant from the
separator and in operations involving
frequent or prolonged evacuation. A
recirculation pump is often used for
systems in which several vacuum
pumps share a common separator.
The fresh sealant makeup line is in-
stalled in much the same way as for
a once-through installation. In addi-
tion, a check valve should be installed
in the makeup line to prevent reverse
flow if makeup supply pressure is lost
while the vacuum pump is operating.
The recirculation line should con-
tain a strainer and one or more tem-
perature gages. The strainer removes
solids that would otherwise accumu-
late in the sealant. Temperature gages
should be installed on both sides of
sealant coolers to monitor heat ex-
changer performance.
Partial recirculation (partial recov-
ery) arrangement. A partial recircu-
lation system has no sealant cooler.
Fresh solvent makeup is used as the
heat sink. A portion of the total seal-
ant flow is supplied as cold makeup
and the rest is recirculated from the
separator. The fresh and recirculated
streams mix and come to some inter-
mediate temperature before entering
the vacuum pump. The desired tem-
perature is controlled by varying the
ratio between the two streams until a
satisfactory temperature is reached. A
partial recirculation arrangement can
reduce fresh sealant requirements
by 50% or more compared with once-
through systems, depending on the
temperature of the fresh stream.
The recommendations about piping
design and filters given in the section
on closed-loop arrangements apply
here as well.
Potential hazards and options to
prevent or mitigate them. Listed
below are some potential hazards of
liquid-ring vacuum pumps and mea-
sures that can be used to prevent or
mitigate them [5]:
The escape of flammable or toxic
vapors from failed mechanical seals
can create external hazardous con-
ditions. Liquid-ring vacuum pumps
handling non-hazardous vapors
usually are provided with single me-
chanical seals. If these pumps han-
dle flammable or toxic vapors, then
double mechanical seals should be
installed with seal-welded connec-
tions to eliminate leak paths
Improper installation, mounting,
and alignment of pumps: This can
cause damaging vibration, which
can lead to potential seal or other
mechanical failures and leakage to
the environment of internal fluids
(sealant, gas or both)
Improper installation of piping
can result in external loads on the
casing leading to pump failure, or
again, leakage of internal fluids to
the atmosphere
Liquid-ring pumps use seal liq-
uids that are in direct contact with
the process gases, and if the gases
pumped are flammable or toxic, the
discharged gases and sealants will
potentially also be hazardous
Since seal liquids are in direct con-
tact with the process gases, liquid
ring pumps act as mixers in a sense,
and any potential reactions between
the seal liquid and the process gas
must be considered
Gases and seal liquids are dis-
charged together and proper separa-
tion must be done if the gases are to
go to subsequent processes. If gases
or vapors condense or dissolve in the
seal liquid, consideration must be
given to selecting the proper materi-
als of construction for the discharge
system to resist attack by any of
the gases present in the process gas
stream to the pump, or introduced
and/or formed in the seal liquid
Seal liquids, if recirculated, will
become saturated to equilibrium
states with the process gases being
pumped. As a result, the overflows
from a liquid-ring pump system
must be treated as a hazardous liq-
uid if the gases are flammable or
toxic. Since, in many cases the seal
liquid is reused in the system, the
concentrations of contaminants can
increase with time
Since compression in a liquid-ring
vacuum pump is essentially isother-
mal, this means that any vapors or
gases exiting will be saturated with
the seal liquid and that any poten-
tially hazardous vapors or gases
may condense and be discharged as
liquids to the discharge system
Proper materials of construction
must be selected if liquid-ring
pumps use seal liquids that are cor-
rosive (such as sulfuric acid) or if the
process gases contain corrosive com-
ponents
Chemical Engineering www.che.com september 2012
Liquid-ring vacuum pumps are not
generally built for high pressures
and use non-live sealed gasket sys-
tems for the most part. This means
that if pressure can rise in hazard-
ous systems quickly, the liquid ring
system is often the lowest pressure
component, and must be protected
from the rise or particular design
variations made, to ensure that they
are resistant to maximum pressures
that may occur
Dry vacuum pumps
Dry vacuum pumps have been used
since the 1980s, originally in the semi-
conductor industry, then in the phar-
maceutical industry, and now in other
chemical process applications. They
have several advantages, as follows:
No contamination of the process
caused by backstreaming of sealing
liquids or lubricating oils
No contamination of evacuated gas
with sealing liquid or lubricant
Due to lack of condensation (which
is assured because the pump runs
hot), the pump can be fabricated of
standard, inexpensive cast iron
They have a rugged rotor design
which is constructed of sturdy cast
iron, or ductile iron, without any
flimsy rotating components
Noncontact design: Timing gears
are oil lubricated in a sealed-off end
chamber to synchronize the rotors
for proper phasing and noncontact-
ing operation
High operational speed: Operation
at high speeds reduces the ratio of
gas slip to displacement, increases
net pumping capacity and reduces
ultimate pressure. To accomplish
this, the rotors are well balanced
They can be designed with multiple
staging
The pump can discharge to the at-
mosphere
They can be used for handling cor-
rosive gases
They can be used for handling flam-
mable gases as the pumps are de-
signed for containment of an inter-
nal explosion for this service
However, they also have the following
drawbacks:
They cannot handle particulates,
nor slugs of liquid
They may require a silencer
65
 Environmental Manager

They may discharge gases at high References Author

temperatures, in some cases as high
as 350400C. Newer designs allow
running at lower temperatures and
have precise temperature control
Some models are difficult to repair
or rebuild
They may require a gas purge for
cooling or to protect the bearings
and seals from the process gas
Due to the high operating tempera-
ture, some process materials such as
monomers, may have a tendency to
polymerize
An excellent overview of dry vacuum
pumps is presented by Ryans and
Bays [6]. Two types of vacuum pumps
that are used in high and ultra-high
vacuum applications are the turbo-
molecular pump and diffusion pump.
However, they are not discussed in
this article.
Edited by Suzanne Shelley
1. Ryans, J.L., and Roper, D.L., Process Vacuum
System Design and Operation, McGraw-Hill
Book Co., New York, N.Y.1986.
2. Chew, A.D., Mechanical Vacuum Pumps,
BOC Edwards paper.
3. Private communication (email) of July 8, 2011
from Mark Nichols of Edwards Limited.
4. Aglitz, A., and others, Installing liquid ring
vacuum pumps, Chem. Eng., November 1996.
5. Private communication (email) of August 11,
2011 from John Campbell of SIHI Pumps
Americas.
6. Ryans, J., and Bays, J., Run clean with dry
vacuum pumps, Chem. Eng. Prog., pp. 3241,
October 2001.
7. Edwards Ltd., Safety Manual-Vacuum
Pump and Vacuum Systems, 2008.
8. NFPA 69, Standard on Explosion Preven-
tion Systems, National Fire Protection As-
sociation, Quincy, Mass., 2008.
Acknowledgements
The author wishes to thank J.L. Ryans
of Eastman Chemical Co. (Kingsport,
Tenn.) for his review of the draft of
this article and for his helpful com-
ments and recommendations. I also
acknowledge the assistance of the fol-
Stanley S. Grossel is president of Process Safety
and Design, Inc. (4 Marble Court, Unit 9, Clifton,
N.J. 07013-2212; Phone: 973-779-3668; Email:
psadi28@aol.com). He provides consulting ser-
vices related to process safety and loss prevention,
storage and handling of bulk solids and powders,
air pollution control, and process design of batch
plants. Prior to his 1993 retirement, Grossel spent
25 years in the Engineering Dept. of Hoffman-La-
Roche, in a variety of roles. He was honored as the
only Senior Engineering Fellow ever to be desig-
nated by Hoffman-LaRoche. Grossel is a fellow of
AIChE, and holds AIChEs Norton Walton/Russell
Miller Award for outstanding contributions to loss
prevention. He has lectured and been published
widely, and has authored major standards and
guidelines. Grossel is also a contributor on process
safety to the 7th Ed. of Perry's Chemical Engi-
neers Handbook. He holds a B.S.Ch.E. from City
College of New York and an M.S.Ch.E. from Drexel
Inst. of Technology, and has completed courses to-
ward a doctoral degree.
lowing people who provided technical
information that was used in writing
this article: Peter N. Lodal (Eastman
Chemical Co.), John Campbell (SIHI
Pumps Americas), Mark Nichols (Ed-
wards Ltd.), and Eric Johnson (Gra-
ham Manufacturing Co.).

September 2012

Literature Review
CHE.COM

Visit CE Lit Online

Featuring Brochures and Catalogs of Products and Services for the CPI at che.com

Precise Density Measurement EquipNet provides asset management ser-

For Caustics and Acids
The Dynatrol Density System is de-
signed to handle temperatures up to 300
while calculating specific gravity, Baum
and percent concentration. Provides ex-
ceptional resistance to corrosion from caus-
tics, acids, ammonia, sulfuric acid, hydro-
chloric acids, and nitric acids. Choose from
Tantalum media contact parts or Hastelloy
C-276, Carpenter 20, and Stainless Steel.
Call 800-231-2062 x19
for a free Catalog today!
Or visit www.DynatrolUSA.com
Circle 291 on p. 72 or go to adlinks.che.com/40271-291
vices to world-class chemical companies
Our offerings include asset evaluation
and appraisal, site reconfiguration con-
sulting, project management, and sales
and marketing. These services yield
a large supply of high-quality process
equipment and unused MRO parts avail-
able for sale at the EquipNet Market-
Place, the largest website of its kind.
Contact: John Cote, Managing Director:
Chemical (781) 821-3482 x2103
jcote@equipnet.com
www.EquipNet.com/chem
Circle 292 on p. 72 or go to adlinks.che.com/40271-292

Chemical Engineering www.che.com JUNE 2012 66

66 Chemical Engineering www.che.com september 2012

Department Editor: Scott Jenkins
W
orkers in the chemical process in-
dustries (CPI) can wear chemical
protective clothing to protect them
from exposure to potentially hazardous
chemicals in the form of vapor, liquid
and solid particles. While chemical pro-
tective clothing has not been regulated
to the extant that safety glasses, hard
hats and some other forms of personal
protective equipment (PPE) have been,
an international standard is available
for classifying the performance of chemi-
cal protective equipment.
ISO 16602
ISO 16602 (Protective clothing for protec-
tion against chemicalsclassification,
labeling and performance requirements)
provides an objective system to appro-
priately test, classify and label chemical-
protective apparel. ISO 16602 designates
minimum performance levels of chemical
protective clothing for six types of chemical
hazards. The garment type designation
is based upon the physical state of the haz-
ard, for example vapors, liquids, aerosols
or particles (Figure 1). The garment types
are as follows:
Type 6: Limited protection against
liquid mist
Type 5: Protection against airborne
solid particulate chemicals
Type 4: Protection against liquid aero-
sols
Type 3: Protection against pressurized
liquid chemicals
Type 2: Non-gas-tight protection
Type 1: Gas-tight (vapor-protective) pro-
tection against chemicals and vapors
and toxic particles
The requirements outlined in ISO 16602
provide a common language for the per-
formance of chemical protective cloth-
ing. ISO 16602 defines performance
requirements based on results obtained
from existing test-method standards.
Although ISO 16602 permits a range
of performance levels for a series of key
properties, it also establishes a minimum
level of performance for each major
type of hazard.
Each workplace environment is
unique, and the ISO 16602 Standard
does not consider all specific hazards
that may be present. Safety and oc-
cupational health professionals should
consider hazards that are specific to
their work conditions and then consult
the ISO 16602 standard to determine
the minimum requirements of chemical
protective garments used in the particu-
lar situation.
ISO 16602 focuses exclusively on
typical chemical hazards and protective
clothing requirements for those hazards,
so each work situation will likely require
additional forms of PPE, possibly includ-
ing footwear, gloves, face protection,
fall protection and respirators.
Six types of CPC outlined in ISO 16602
Type 4
Type 5
Type 6 Protection
against
Protection
liquid
against
Limited aerosols
airborne,
protection
against solid
liquid mist particulate
chemicals
Figure 1. Do not assume that the terms gas-tight and liquid-tight refer to absolute barrier
performance in chemical protective clothing (CPC). They can oversimplify the level of barrier
provided by such garments. Instead, these terms should be considered as vapor-protective
and liquid-protective, respectively
Hazard assessment
A hazard assessment provides a basis for
understanding what chemical protective
clothing and other PPE are necessary to
protect the workers operating in specific
situations. The environment, chemical haz-
ard and work activity must be considered
in order to be able to select the most ap-
propriate protective clothing including
the fabric, seam type and garment design
most suitable. The goal is to select cloth-
ing with adequate protection, but without
overprotecting a worker with unnecessary
clothing, which can result in heat stress,
reduced field of vision, restricted mobility
and increased physical exertion for the
wearer, as well as higher costs for the
employer.
Performance testing
When setting the requirements for each
type in ISO 16602, the entire garment
is tested as well as the individual compo-
nents.
Whole-garment tests. During the whole-
garment tests, a human subject wears
the test garment and accompanying PPE,
such as gloves, boots and respirator. The
subject is exposed to non-hazardous test
chemicals while in an enclosed chamber
performing a series of movements meant to
simulate actual work activities. This whole-
garment testing is used to validate the
barrier performance of the entire ensemble
against a specific type of chemical threat
(gas, liquid or particle). Test chemicals are
used to determine how much of a similar
phase chemical will leak into the suit.
Class tests. Whole-garment tests do not
evaluate the chemical permeation proper-
ties of the garments. This is assessed in
the class testing portion of ISO 16602.
Beyond the whole-garment tests, additional
tests are conducted on the garments
fabric(s) and components to qualify the
class performance level.
Within each of the six garment types,
Chemical
Protective Clothing
Type 1
Type 2
Type 3
Gas-tight
Non protection
gas-tight against
Protection chemicals
against protection
and vapors
pressurized and toxic
liquid
particles
chemicals
Type indication doesnt tell you everything
about the performance of the CPC. It only
tells that it meets the minimum require-
ments of a particular type certification
there are also requirements directed
at the mechanical, barrier and basic
flammability properties of the fabrics
and components used to make chemi-
cal protective clothing. Laboratory tests
are used to determine the mechanical
durability, the barrier against specific
chemical hazards, and ease of ignition
of the garment materials. The results
of these tests will fall into a unique
performance class. Each type within
ISO 16602 specifies a combination of
barrier and durability tests levels, estab-
lishing a minimum performance class
for each of the tests to meet the specific
type requirements. A higher class rating
denotes a higher level of performance
for that property.
The flammability requirements outlined
in ISO 16602 establish a minimum per-
formance level of flame spread once the
material is ignited; it does not qualify
an ensemble as suitable for protection
against heat and flame hazards. Spe-
cific evaluation of chemical protective
clothing for heat and flame protection is
not in the purview of ISO 16602, but is
covered within NFPA standards.
Conclusion
Both testing approaches are integral
to evaluating the overall integrity and
expected performance of chemical protec-
tive clothing. This integrated rating system
aids the selection of the most appropriate
chemical protective apparel for specific
work tasks. Testing the whole garment
and the individual components separately
ensures the sum of the parts is able to
provide the appropriate protection.
Editor's note: This Facts at your Fingertips
column was adapted from Lovasic, S., Chemi-
cal Protective Clothing, Chem. Eng., March
2011, pp. 51-53.

Department Editor: Scott Jenkins
W
hen flammable or combus-
tible atmospheres are present,
uncontrolled discharges of static
electricity can be potentially dangerous
or even catastrophic. A significant por-
tion of industrial explosions and fires are
attributable to static electricity each year.
In theory, controlling static electricity by
grounding potential sources is simple, but
in practice, doing so effectively requires
thorough knowledge of processes and
operations, sound engineering controls,
properly specified safety equipment and
properly trained operational staff. This
column provides information on potential
sources of static electrical discharge.
Static risk
Static electricity discharges are possible
almost continuously in the CPI, because
static electricity is generated whenever
surfaces come into contact and then
separate. In most cases, the charge
generation in industrial processes is
small typically no greater than 1
104 Amp. However, even small charges
can be a problem in hazardous areas
when allowed to accumulate on objects
that are not at ground (earth) potential.
If no ground is present, voltages in
excess of 30 kV can develop. Depend-
ing on the capacitance of the object,
this may result in significant levels of
energy being available for discharge.
If the energy equals or exceeds the
minimum ignition energy (MIE) of the
surrounding flammable atmosphere, the
potential for an explosion and fire exist.
Many commonly used solvents and other
flammable chemicals have MIEs that are
relatively low on the order of 1 mJ or
less (Table 1).
Isolated conductors
Isolated conductors are conductive
objects that are either inherently or ac-
cidentally insulated from earth. During
day-to-day operations at industrial facili-
ties, isolated conductors are probably
the most likely source of static ignition
incidents.
The insulation effectively keeps any
static electricity buildup from safely
discharging, thereby resulting in ac-
cumulation of charge on the object. If
the isolated conductor then comes into
proximity with another object at a lower
potential, energy could be released in
the form of an incendive spark.
Isolated conductors may arise from
metal flanges, fittings or valves in pipe-
work systems; portable drums, containers
or vessels; tanker trucks, railcars and
intermediate bulk containers (IBCs); and
even people.
Many modern industrial paints,
coatings, gaskets, seals and other
non-conductive materials are sufficiently
insulating, so as to to possibly prevent
the proper dissipation of static charge.
Static discharge sources
Static discharges come in several forms,
the most important for CPI interests being
spark and brush discharges. A spark
is a discharge from a charged isolated
conductor to another conductor at lower
potential. A brush discharge occurs from
an electrostatically charged insulator to a
grounded conductor.
Typical possible sources of static elec-
trical discharges include the following:
Spark discharges from any conductive
but not earthed (grounded) bag, bin,
drum, container and so on, from which
the powder is transferred into the reac-
tor
Brush discharges from any non-conduc-
tive bag, bin, drum, container, and so
on, from which the powder is trans-
ferred into the reactor
Spark discharges from any conductive
but not earthed auxiliary device used
in the transfer procedure, including, but
not limited to, shovels, funnels, chutes
and pipes
Spark discharges from the operator, if
he is not adequately earthed
Brush discharges from any non-conduc-
tive auxiliary devices, such as shovels,
funnels, chutes and pipes
Brush discharge from the dust cloud
formed within the reactor during pow-
der transfer
Spark discharges from any conductive
but not earthed fixtures and fittings
within the reactor
Brush discharges from the charged sol-
vent, suspension or emulsion preloaded
in the reactor
Brush discharges from the powder
heap formed on top of the liquid phase
within the reactor
Cone discharges from the powder heap
formed on top of the liquid phase
Liquids flowing through pipelines or
filling into drums and tanks
Persons walking across an insulating
floor
Static discharge prevention
Where recommendations tend to
converge is in the recommendation to al-
ways use conductive or static dissipative
materials, and to ensure effective bond-
ing and grounding. For information on
grounding best practices and examples
of preventing static electrical discharge,
consult National Fire Protection Associa-
tion (NFPA; Quincy, Mass.; www.nfpa.
org) standards 77 and 30.
In this context, the term conductive
would apply to metal materials, such
as stainless or carbon steel, aluminum
and others, and static-dissipative may
indicate rubber or plastics that have
Static Electricity
Discharge and Fire
Prevention
Table 1. Typical MIE values
Material Minimum
(gas/vapor or pow- Ignition Energy
der/dust) (MIE), mJ
Carbon disulfide 0.009
Methanol 0.14
Xylene 0.20
Toluene 0.24
Propane 0.25
Ethyl acetate 0.46
Zirconium 5.00
Epoxy resin 9.00
Aluminum 10.00
Sugar 30.00
Wheat flour 50.00
Note 1: Minimum ignition energy (MIE) is defined
as the minimum energy that can ignite a mixture
of a specified flammable material with air or oxy-
gen, measured by a standard procedure.
Note 2: MIE values are provided for guidance
only specific MIE data for any material
should be verified. Source: NFPA, IchemE
been formulated with some added semi-
conductive additives. Bonding means
linking these objects together by means
of a suitably strong conductor (wire), and
grounding refers to a true ground/
earth connection that is applied to one
or more of the bonded objects.
When one or both of these techniques
is applied, and while a low resistance
connection between the objects and
ground is maintained, operators are
able to prevent dangerous levels of static
charge from accumulating.
In the case of fixed installations such
as pipe work, storage tanks and so forth,
this is relatively simple to implement.
However, these preventive measures
are more difficult to implement with
mobile/portable objects, such as drums,
IBCs and tankers. In these instances,
purpose-designed temporary ground-
ing and bonding devices must be used,
with strict procedures to ensure that they
are always in place prior to starting the
process. For instance, specific types of
clamps and devices for grounding and
bonding portable or mobile plant equip-
ment, drums and containers are recom-
mended in NFPA 77, and such ground-
ing clamps and devices generally should
employ sharp contact points. These
contact points should be made of a wear-
resistant material, have positive spring
pressure, and be universally adaptable to
a wide range of plant objects.
References
1. Tyers, G., Avoiding Static Sparks in Hazardous
Atmospheres, Chem. Eng., June 2009, pp.
4449.
2. Glor, M., Preventing Explosions During the
Transfer of Solids into Flammable Solvents,
Chem. Eng., October 2007, pp. 8895.
Editors note. This edition of "Facts at your Finger-
tips" was adapted from information in the two ar-
ticles referenced above.
 Liquefied-
Industrial-Gas Safety
Department Editor: Scott Jenkins

M
any industrial gases are transported, without self-contained Table 1. Properties of
stored and transferred in their liquid breathing equipment. common industrial gases
form, and they present significant Flammability. In the case Substance Chemical Boiling Density, Latent heat of
health and safety risks. This one-page refer- of cryogenic gases that are symbol point, C kg/m3 vaporization,
ence highlights the best safety practices for flammable, including hydro- kJ/kg
preventing injury in personnel that work gen, methane, and acety- Air -- 194.3C 1.29 205.0
with cryogenic fluids. lene, ignition sources must Nitrogen N2 195.8 C 1.25 199.1
High-purity industrial gases, including be prohibited. Flammable Helium He 268.9C 0.18 20.28
nitrogen, oxygen, helium, argon and other materials must be stored
Oxygen O2 183.0 C 1.43 213.0
noble gases, are commonly generated by away from liquid oxygen
Argon Ar 185.9C 1.78 162.3
air-separation units (ASUs), which use a areas, since it can support
cryogenic distillation process. Liquefied and accelerate combustion. Acetylene C2H2 83.3C 1.17 614.0
industrial gases, especially liquid nitrogen, Be aware that liquefied inert Hydrogen H2 252.8C 0.09 446.0
are often delivered to end-users by cryo- gases and extremely cold Ethane C2H6 88.6C 1.36 489.4
genic tankers into onsite vacuum-insulated surfaces can condense oxy- Methane CH4 161.5C 0.72 509.9
storage vessels for on-demand use as a gen from the atmosphere, Neon Ne 246.0C 0.90 86.3
liquid or a gas. Among the myriad uses of causing oxygen entrapment Xenon Xe 108.2C 5.85 96.3
industrial gases, several involve their use in unsuspected areas. Carbon CO2 78.5C 1.25 571.3
in liquid form. These include: research and dioxide
development; cooling of process equip- Safety practices Ammonia NH3 33.3C 0.77 1,371.0
ment (reactors, crystallizers, storage tanks Tasks with the potential for
Krypton Kr 153.4C 3.71 107.5
and so on); lyophilization (freeze-drying); cryogenic exposure should
recovery of volatile organic compounds only be conducted by
(VOCs), food freezing, cryogenic milling of competent staff that has received appropri- jewelry should not be worn, as metals can
solids, cryogenic preservation of biological ate training on specific cryogenic-related become frozen to the skin.
materials, plastic and rubber deflashing health and safety risks, standard operating Handling. The following are tips for safe
and grinding, and others. procedures, and actions to be taken in the handling of cryogenic liquids.
event of an exposure incident. Do not overfill containers
Exposure hazards Personal protective equipment (PPE). Pour slowly to minimize boiling and
The extremely low temperatures and high Cryogenic exposure is often attributable to splashing
expansion rates (from a phase change) that inadequate PPE. Where the potential for Avoid the path of boil-off gases
characterize liquefied gases present health cryogenic exposure exists, all personnel Ensure that cryogenic fluids are stored in
and safety hazards. should use the following PPE: appropriately insulated containers, which
Cryogenic burns and frostbite hazards. Eye protection, with a full-face shield minimize the loss of product due to boil-off
Cryogenic liquids and cold gas that come over safety glasses, should be used when Containers of cryogenic liquid should be
in contact with human skin can cause seri- transferring liquids to an open container. allowed to vent. Where a special vented
ous damage to living tissue, including cold Gloves should either be designed to stopper or venting tube is used, as on
burns and frostbite. Damage can occur with prevent cryogens from flowing into the some small portable containers, the vent
exposures of longer than a few seconds, glove or be loose fitting so the glove can should be checked regularly to ensure
especially to delicate tissues, such as the be easily be shaken off after accidental it has not plugged with ice formed from
eyes. In the result of a large spill, exposure contact. Cotton or nylon gloves covered condensedwater vapor from the air
to a cold atmosphere for an extended time with disposable nitrile or vinyl gloves The materials used in cryogenic systems
places personnel at risk for hypothermia. are recommended for work that requires must have the appropriate physical prop-
High-pressure gas. Storing cryogenic fluids delicate handling of cold items in close erties to qualify them for use at extremely
present hazards from high-pressure gas, proximity to liquid nitrogen, but is not low temperatures. Acceptable materials
since the liquefied gases are usually stored designed to allow immersion of hands include: aluminum; 304 and 316 stainless
at or near their boiling points (see Table). and fingers into the liquid. Cotton gloves steels; copper; brass; fiberglass
Large expansion ratios on vaporization can wick the cold liquid, so should not Pressure relief. The follow are tips related to
can cause a buildup of large pressures. be used without the nitrile outer glove. pressure relief.
The evaporation rate depends on the fluid, Lab coats, disposable coveralls or long- All system vents must be directed away
storage-container design and environmental sleeved shirts and pants that provide from personnel or designated work areas.
conditions, but all storage containers should complete coverage of skin not otherwise Venting fluids (liquids or cold gases)
allow for the evaporation of the liquid. protected by PPE and/or attire, should be should not impinge on any part of the
Displacement of oxygen/asphyxiation. worn. body. Ensure that pressure relief devices
Because of the large expansion when a Contact. Cryogenic liquids and uninsulated are checked regularly and maintained.
cryogenic fluid evaporates, they are capa- cryogenic equipment and pipes should Pressure relief devices must be sized for
ble of causing asphyxiation by displacing generally not not be directly touched. Tongs maximum backpressure
breathable air. Natural ventilation should can be used to withdraw objects immersed Ensure all safety valves and vent valves
be ensured, and transporting and using in a cryogenic liquid. All parts of the body are unobstructed and functioning properly.
cryogenic fluids in enclosed spaces should must be protected from uninsulated pipes Check the safety vents on liquid nitrogen
be avoided. In normal dry air, oxygen or vessels containing cryogenic liquids, tanks at least twice a week.
makes up 20.95% by volume. No person- because the extremely cold metal may stick Ensure that an oxygen alarm is present in
nel should enter an area where the oxygen fast to the skin and result in torn flesh when the work area when appropriate.
concentration has dropped to below 19.5% the skin is withdrawn. Watches, rings and Spills. If cryogenic material is spilled and
 Environmental Manager
Things You Need to Know Before Using an
Explosion-Protection Technique
Understanding the different classification methods is necessary to better select the
explosion-protection techniques that will be used
Class I, Division 2
(from 5 ft to 10 ft radius)
Outdoors
Gasoline
storage
10 ft (3m) tank
without
floating
roof
Class 1, division 2
FIGURE 1. Shown here is a typical example of a Class I hazardous area uti-
lizing division methods of area classification
Robert Schosker
Pepperl+Fuchs
E
xplosion protection is essen-
tial for many companies, and
those companies have deci-
sion makers. But before any
decisions can be made, there are
some important factors one must
consider. These factors include what
is most efficient and economical, as
well as knowing the basics of explo-
sion protection; so the decision mak-
ers are headed in the right direction.
We will highlight many of the different
things to know, but first, lets step
back in time and take a look at the
background of explosion protection.
Backdrop
After World War II, the increased
use of petroleum and its derivatives
brought the construction of a great
number of plants for extraction, refin-
ing and transformation of the chemi-
cal substances needed for techno-
logical and industrial development.
The treatment of dangerous sub-
50
Class I, Division 1
(5 ft around vent)
Zone 2
Vent
10 ft (3m)
Circular dike
around tank
FIGURE 2. The example hazardous area shown in Figure 1 is here classified ac-
cording to the zones
stances, where there exists the
risk of explosion or fire that can be
caused by an electrical spark or hot
surface, requires specifically defined
instrumentation located in a hazard-
ous location. It also requires that
the interfacing signals coming from
a hazardous location are unable to
create the necessary conditions to
ignite and propagate an explosion.
This risk of explosion or fire has
been the limiting factor when using
electrical instrumentation because
energy levels were such that the
energy limitation to the hazardous
location was difficult, if not impossi-
ble, to obtain. For this reason, those
parts of the process that were con-
sidered risky were controlled with
pneumatic instrumentation.
Moving forward
Now lets move forward 70 years,
where almost everything you can
think of can be found at the touch of a
finger. From pneumatics to quad core
processors, information gathering
has definitely changed, but the same
CHEMICAL ENGINEERING
Zone 1 Vent
Zone 0
Gasoline
storage
tank
without
floating
roof
principles for working or gathering
information out of a hazardous area
remain the same. Its just that today
we have multiple options. In order to
exercise those options, we must first
determine if the danger of an explo-
sion exists and how severe it may be.
What is a hazardous area?
Hazardous areas are most frequently
found in places where there is a pos-
sibility of an emission of flammable
gas or dust. A hazardous area can
occur in normal operation, in the
event of a fault, or due to wear and
tear of seals or other components.
Now the risk of an ignition of an
air/gas mixture in this hazardous
area depends on the probability of
the simultaneous presence of the
following two conditions:
Formation of flammable or explo-
sive vapors, liquids or gases, or
combustible dusts or fibers with
atmosphere or accumulation of
explosive or flammable material
Presence of an energy source (elec-
trical spark, arc or surface temper-
WWW.CHEMENGONLINE.COM JULY 2015
 TABLE 1. DEFINING AREAS FOR DIVISIONS
Class Type of Material
Class I Locations containing flammable
gases, flammable liquid-produced va-
pors, or combustible liquid-produced
vapors
Class II Locations containing combustible
dusts
Class III Locations containing fibers and
flyings
ature) that is capable of igniting the
explosive atmosphere present
Determining hazardous areas in
a plant is normally performed by
experts from various disciplines. It
may be necessary for chemists, pro-
cess technologists, and mechani-
cal engineers to cooperate with an
explosion-protection expert in order
to evaluate all hazards. The possible
presence of a potentially explosive
atmosphere as well as its properties
and the duration of its occurrence
must be established.
Also understanding terms such as
minimum ignition energy (MIE), upper
and lower explosive limit (UEL/LEL),
flash point, and ignition temperature
in the evaluation of your hazardous
area will also provide a clearer direc-
tion on how severe a hazardous area
might be.
In any situation involving an ex-
plosive material, the risk of ignition
must be taken into account. In addi-
tion to the nominal rating of materi-
als under consideration, parameters
related to the process involved are
especially important in the evalua-
tion. For example, the risk of explo-
sion may be caused by the evapora-
tion of a liquid or by the presence of
liquid sprayed under high pressure.
It is also important to know which
atmospheric conditions are present
normally and abnormally. The range
of concentration between the ex-
plosion limits generally increases as
the pressure and temperature of the
mixture increases.
Divisions and zones
Once it has been determined that a
hazardous area exists, it now needs
to be classified. While the physical
principles of explosion protection
are the same worldwide and are
not differentiated, there are two dif-
ferent and distinct models to define
your hazardous area divisions and
zones both of which are accepted
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
TABLE 2. THE BREAKDOWN OF CLASSES INTO SUBGROUPS
Class Subgroup Atmospheres
Class I Group A Atmospheres containing acetylene
Group B Atmospheres containing hydrogen and flammable process gases with
more than 30 vol.% H2, or gases or vapors posing a similar risk level,
such as butadiene and ethylene oxide
Group C Atmospheres such as ether, ethylene or gases or vapors posing a similar
risk
Group D Atmospheres such as acetone, ammonia, benzene, butane, cyclopropane,
ethanol, gasoline, hexane, methanol, methane, natural gas, naphtha, pro-
pane or gases or vapors posing a similar threat
Class II Group E Atmospheres containing combustible metal dusts, including aluminum,
magnesium, and their commercial alloys, or other combustible dusts
whose particle size, abrasiveness and conductivity present similar haz-
ards in the use of electronic equipment
Group F Atmospheres containing combustible carbonaceous dusts, including
carbon black, charcoal, coal or coke dusts that have more than 8% total
entrapped volatiles, or dusts that have been sensitized by other materials
so that they present an explosion hazard
Group G Atmospheres containing combustible dusts not included in Group E or
Group F, including flour, grain, wood, plastic and chemicals
TABLE 3. THE DIVISION METHOD
Division Class I Class II Class III
(gases and vapors) (flammable dust or powder) (flammable fibers or sus-
pended particles)
In accordance with NEC In accordance with NEC
500.5 and CEC J18-004 500.6 and CEC 18-008 In accordance with NEC
500.5 and CEC 18-010
Division 1 Areas containing dangerous Areas containing dangerous Areas containing dangerous
concentrations of flam- concentrations of flammable concentrations of flammable
mable gases, vapors or mist dusts continuously or oc- fibers or suspended particles
continuously or occasionally casionally under normal continuously or occasion-
under normal operating operating conditions ally under normal operating
conditions conditions
Division 2 Areas probably not contain- Areas probably not contain- Areas probably not contain-
ing dangerous concentra- ing dangerous concentra- ing dangerous concentra-
tions of flammable gases, tions of flammable dusts tions of flammable fibers or
vapors or mist under normal under normal operating suspended particles under
operating conditions conditions normal operating conditions
and utilized worldwide. may exist due to flammable gases,
In rather simple terms, we can dif- vapors, or liquids within a plant or
ferentiate between the International working environment. These are ex-
Electrotechnical Commission (IEC; plained by defining the type of haz-
Geneva, Switzerland) (zones) and ardous material present, severity of
the North American (division) proce- the hazard, and probability of the
dures. The differences lie in the cat- hazard. It may also depend on the
egorization of hazardous areas, the likelihood of the hazard, risk of an
design of apparatus, and the installa- explosion, and the boundaries of the
tion technology of electrical systems. hazardous location.
The categorization of these areas is This is usually determined by a
carried out in North America in ac- HAZOP (hazard and operability)
cordance with the National Electrical study and documented on a set
Code (NEC) NFPA 70, article 500. of electrical plot plans on record in
The European Zone practice is de- every plant.
scribed in IEC/EN 60079-10. For divisions, the type of material
So how does each work? First lets is given by a class designation, as
start at the basics, and then well shown in Table 1. These can be bro-
cover each individually. ken down further into sub-groups,
as shown in Table 2.
Defining the area Once we have determined the haz-
Hazardous location or area clas- ardous material we are working with,
sification methods specify the dan- the probability of an explosion and
ger of fire or explosion hazards that boundaries must also be taken in to
JULY 2015 51
 Hazardous atmosphere
Interstice
Length of
junction
FIGURE 3. Explosion-proof protection is based on
the explosion-containment concept, whereby the
enclosure is built to resist the excess pressure
created by an internal explosion
consideration. The division method
is divided into two areas: Division 1
and Division 2 (Table 3). These were
created in 1947 when the NEC first
recognized that different levels of risk
exist in hazardous locations. Figure 1
shows a typical example of a Class
I hazardous area utilizing Division
methods of area classification.
In comparison to the division-
based area classification, which is
prevalent throughout North America,
the zone-based architecture prevails
in the rest of the world.
Zones are similar in nature to divi-
sions where type of hazardous ma-
terial present, severity of the hazard,
and probability of the hazard and
boundaries must be determined.
Zones are in accordance with IEC/
EN 60079-10, which states that any
area in which there is a probability of
a flammable gas or dispersed dust
must be classified into one of the
areas shown in Table 4.
Similar to the division method of
area classification, zones can be
TABLE 4. DEFINING AREAS BY ZONES
Zone Type of material
Zone 0 An area in which an explosive air/gas
mixture is continuously present or
present for long periods of time
Zone 1 An area in which an explosive air/gas
mixture is likely to occur in normal
operation
Zone 2 An area in which an explosive air/gas
mixture is unlikely to occur; but if it
does, only for short periods of time
Zone 20 An area in which a combustible dust
cloud is part of the air permanently,
over long periods of time or frequently
Zone 21 An area in which a combustible dust
cloud in air is likely to occur in normal
operation
Zone 22 An area in which a combustible dust
cloud in air may occur briefly or dur-
ing abnormal operation
52
Hazardous atmosphere
Q atmosphere
P
FIGURE 4. In purging or pressurization protec-
tion, a dangerous air/gas mixtures is not allowd
to penetrate the enclosure containing the electri-
cal parts that can generate sparks or dangerous
temperatures
better rationalized by looking at the
example shown in Figure 2.
With a slightly different approach,
IEC 600079-0 requires apparatus to
be subdivided into two groups, as
shown in Table 5.
The groups indicate the types
of danger for which the apparatus
has been designed. Group I is in-
tended for mines. Group II concerns
above-ground industries (electrical
apparatus for hazardous areas with
potentially explosive gas (dust) at-
mosphere except firedamp hazard-
ous mining areas) and is subdivided
into II G (gases) and II D (dusts).
Similar to divisions, the zones
offer a sub material classification
as well. Table 6 shows how this
approach compares to the North
American equivalent.
Finally, when classifying your haz-
ardous area, whether it be division
or zones, you must also classify the
maximum surface temperature that
can go in to the hazardous area. The
maximum surface temperature must
be below the minimum ignition tem-
perature of the gas/dust present.
In North America as in Europe, six
temperature classes are differenti-
TABLE 5. APPARATUS GROUPS PER IEC
600079-0
Group Apparatus
Group I Apparatus to be used in mines where
the danger is represented by methane
gas and coal dust
Group II Apparatus to be used in surface
industries where the danger is rep-
resented by gas and vapor that has
been subdivided into three groups:
A, B and C. These subdivisions are
based on the maximum experimental
safe gap (MESG) for an explosion-
proof enclosure or the minimum igni-
tion current (MIC) for intrinsically safe
electrical apparatus
CHEMICAL ENGINEERING
Hazardous
Io L R
U
C
S Uo
FIGURE 5. Intrinsic safety is based on the prin-
ciple of preventing an effective source of ignition
ated, T1 to T6. The classes T2, T3
and T4 are divided into further sub-
classes, as indicated in Table 7.
In Europe, the apparatus are
certified on the basis of design
and construction characteristics.
From a practical point of view, the
two systems are equivalent, even
if there are minor differences, but
before you run out and choose the
most convenient method for you, it
is important that you consult your
local authority having jurisdiction to
learn what method is allowed or, in
fact, preferred.
The initial steps to determine
whether a hazardous area exists and
classify that area may seem rudimen-
tary to some, but they are important
as they now open up the multiple
methods of protection, which may
or may not be allowed, depending
on whether you classified your area
by divisions or zones.
Protection methods
There are three basic methods of
protection explosion containment,
segregation and prevention.
Explosion containment. This is the
only method that allows the explosion
to occur, but confines it to a well-de-
fined area, thus avoiding the propaga-
tion to the surrounding atmosphere.
Flameproof and explosion-proof en-
closures are based on this method.
Segregation. This method attempts
to physically separate or isolate the
electrical parts or hot surfaces from
the explosive mixture. This method
includes various techniques, such
as pressurization, encapsulation,
and so on.
Prevention. Prevention limits the
energy, both electrical and thermal,
to safe levels under both normal op-
eration and fault conditions. Intrinsic
safety is the most representative
technique of this method.
WWW.CHEMENGONLINE.COM JULY 2015
 TABLE 6. SUB MATERIAL CLASSIFICATION FOR ZONES
Material Apparatus clasification
Europe (*IEC)
Methane Group I (mining)
Acetylene Group IIC
Hydrogen Group IIC
Ethylene Group IIB
Propane Group IIA
Conductive dust (metal) Group IIIC*
Non-conductive dust Group IIIB*
(carbon)
Cereal/flour Group IIIB*
Fibers/suspended Group IIIA*
particles
My application requirements
Now the questions really start racing
in: Which should I use? Which one
offers the best protection? What if
all of my equipment is not low pow-
ered? My plant is already using a
technique; can I use another pro-
tection method? Can they co-exist?
Who makes that decision? Why
should I use one method over the
other? Can I use two methods at
the same time? So many questions,
all of which are very important, and
with a little understanding of your
process, they will guide you to best
method(s) to use.
Hazardous-area protection method
selection depends on three impor-
tant factors: (1) area classification, (2)
the application and (3) the cost of the
protection method solution.
Area. Area classification depends on
the type of hazardous substances
used, operating temperature, and
explosion risk due to how often the
dangerous substance is present in
the atmosphere and the boundary
of the substance from various parts
of the process. Area classification
is determined by either the division
method or zone method.
Application. Application character-
istics also affect which protection
method is used. For example, some
methods are more appropriate for
large equipment protection, while
others are more appropriate for high-
power applications.
Cost. Cost is also an important fac-
tor for many engineers. For example,
if their application requires Division
2 protection, they may not want to
purchase more expensive equipment
rated for Division 1. For that reason,
it is important to understand the in-
terplay of all three factors classi-
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
TABLE 7. TEMPERATURE CLASSES
Apparatus classification Ignition energy Tmax, C Tmax, F T Class in N.A.*
450 842 T1
North America
300 572 T2
Class I, Group D
280 536 T2A
Class I, Group A > 20 J
260 500 T2B
Class I, Group B > 20 J
230 446 T2C
Class I, Group C > 60 J
215 419 T2D
Class I, Group D > 180 J
200 392 T3
Class II, Group E
180 356 T3A
Class II, Group F
165 329 T3B
Class II, Group G 160 320 T3C
Class III 135 275 T4
120 248 T4A
100 212 T5
fication, application, and cost in
85 185 T6
helping users find the ideal solution
to match their needs. *N.A. = North America
In addition to considering the nor-
mal functioning of the apparatus, area classification, there are many
eventual malfunctioning of the appa- other options when an area is classi-
ratus due to faulty components must fied using zones, but for now we will
be a consideration. And finally, all concentrate on the above as they
those conditions that can acciden- are most commonly used.
tally occur, such as a short circuit, XP. The explosion-proof protection
open circuit, grounding and errone- method is the only one based on the
ous wiring of the connecting cables, explosion-containment concept. In
must be evaluated. The choice of a this case, the energy source is permit-
specific protection method depends ted to come in contact with the dan-
on the degree of safety needed for gerous air/gas mixture. Consequently,
the type of hazardous location con- the explosion is allowed to take place,
sidered in such a way as to have the but it must remain confined in an en-
lowest probable degree of an even- closure built to resist the excess pres-
tual simultaneous presence of an sure created by an internal explosion,
adequate energy source and a dan- thus impeding the propagation to the
gerous concentration level of an air/ surrounding atmosphere.
gas mixture. The theory supporting this method
None of the protection methods is that the resultant gas jet coming
can provide absolute certainty of from the enclosure is cooled rap-
preventing an explosion. Statistically, idly through the enclosures heat
the probabilities are so low that not conduction and the expansion and
even one incident of an explosion dilution of the hot gas in the colder
has been verified when a standard- external atmosphere. This is only
ized protection method has been possible if the enclosure openings
properly installed and maintained. or interstices have sufficiently small
The first precaution is to avoid dimensions (Figure 3).
placing electrical apparatus in haz- In North America, a flameproof en-
ardous locations. When designing closure (in accordance with IEC) is, as
a plant or factory, this factor needs a rule, equated with the "flameproof"
to be considered. Only when there is designation. In both considerations,
no alternative should this application the housing must be designed for
be allowed. a x1.5 explosion overpressure. The
North American version Explosion
Choosing the best method proof (XP) must withstand a maxi-
After carefully considering the above, mum explosion overpressure of x4.
we can look at three more popular Furthermore, in North America,
methods of protection, XP (explo- the installation regulations (NEC 500)
sion proof/flameproof), purging and specify the use of metal conduit for
pressurization, and intrinsic safety. the field wiring installation. It is also
Although these are the most com- assumed here that the air-gas mixture
monly used methods in the division can also be present within the con-
JULY 2015 53
 duit system. Therefore, the resulting
explosion pressures must be taken
into consideration. The conduit con-
nections must be constructed ac-
cording to specification and sealed
(that is, lead seals) with appropriate
casting compound. The housing is
not constructed gas-tight. Of course,
large openings are not permitted on
the enclosure, but small ones are in-
evitable at any junction point. Some
of these gaps may serve as pressure
relief points. Escaping hot gases are
cooled to the extent that they cannot
ignite the potentially explosive atmo-
sphere outside the housing. Ignition
is prevented if the minimum temper-
ature and minimum ignition energy
of the surrounding potentially explo-
sive atmosphere is not reached. For
this reason, the maximum opening
allowed for a particular type of joint
depends on the nature of the explo-
sive mixture and width of the adjoin-
ing surfaces (joint length).
The classification of a flameproof en-
closure is based on the gas group and
the maximum surface temperature
which must be lower than the ignition
temperature of the gas present.
Purging or pressurization. Purg-
ing or pressurization is a protection
method based on the segregation
concept. This method does not allow
the dangerous air/gas mixture to pen-
etrate the enclosure containing electri-
cal parts that can generate sparks or
dangerous temperatures. A protective
gas air or inert gas is contained
inside the enclosure with a pressure
slightly greater than the one of the ex-
ternal atmosphere (Figure 4).
The internal overpressure remains
constant with or without a continu-
ous flow of the protective gas. The
enclosure must have a certain de-
gree of tightness; however, there
are no particular mechanical require-
ments because the pressure sup-
ported is not very high.
To avoid pressure loss, the pro-
tective gas supply must be able to
compensate during operation for
enclosure leakage and access by
personnel where allowed (the use of
two interlocked doors is the classical
solution). Because it is possible for
the explosive atmosphere to remain
inside the enclosure after the pres-
surization system has been turned
off, it is necessary to expel the re-
maining gas by circulating a certain
quantity of protective gas before re-
54
starting the electrical equipment.
The classification of the electrical
apparatus must be based on the
maximum external surface tempera-
ture of the enclosure, or the maxi-
mum surface temperature of the in-
ternal circuits that are protected with
another protection method and that
remain powered even when the pro-
tective gas supply is interrupted.
The purging or pressurization
technique is not dependent upon the
classification of the gas. Rather, the
enclosure is maintained at a pressure
higher than the dangerous external
atmosphere, preventing the flamma-
ble mixture from coming in contact
with the electrical components and
hot surfaces inside.
In the U.S., the term pressuriza-
tion is limited to Class II applica-
tions. This is the technique of supply-
ing an enclosure with clean air or an
inert gas, with or without continuous
flow, at sufficient pressure to prevent
the entrance of combustible dusts.
Internationally, the term pressuriza-
tion refers to a purging technique
for Zones 1 and 2.
The divisional model of the purging
protection method is based on the
reduction of the classification inside
the enclosure to a lower level. The
following three types of protection (X,
Y, and Z) are identified in relation to
the hazardous-location classification
and the nature of the apparatus.
Type X: reduces the inside of the
enclosure from Division 1 to a non-
hazardous state that requires an
automatic shutdown of the system
in case of pressure loss
Type Y: reduces the inside of the
enclosure from Division 1 to Divi-
sion 2
Type Z: reduces the inside of the
enclosure from Division 1 to a non-
hazardous state, requiring alarm
signals only
Intrinsic safety. Finally, intrinsic
safety is based on the principle of
preventing an effective source of ig-
nition. The electrical energy is kept
below the minimum ignition energy
required for each hazardous area
(Figure 5).
The intrinsic safety level of an elec-
trical circuit is achieved by limiting
current, voltage, power and tem-
perature; therefore, intrinsic safety is
limited to circuits that have relatively
low levels of power. Of critical impor-
tance are the stored amounts of en-
CHEMICAL ENGINEERING
ergy in circuits in the form of capaci-
tance and inductance. These energy
storage elements must be limited
based on the voltage and current
levels present in a particular circuit
or make-break component.
In normal operation and in the
event of a fault, no sparks or thermal
effects may occur that could lead to
the ignition of a potentially explosive
atmosphere. Intrinsically safe circuits
may therefore be connected and
disconnected by experts during op-
eration (even when live), as they are
guaranteed to be safe in the event of
a short circuit or disconnection.
Intrinsic safety is the only ignition-
protection class that allows connec-
tors to be opened and intrinsically
safe apparatus to be removed and
replaced by an equivalent device in
a hazardous area. Because of the
level of freedom this brings, intrinsic
safety has become one of the most
important methods of protection in
the industrial automation industry.
Final remarks
Each method offers its own advan-
tages and disadvantages, and in most
cases no one method will be or can
be the only method used in a process
plant. Generally, this mixed system
does not present installation difficulty
if each of the protection methods is
appropriately used and is in compli-
ance with the respective standards.
No matter how you classify your
plant or which method of protection
you chose, it is always important
to remember that the method you
choose today may not necessarily
be the appropriate choice tomorrow.
Evaluate, choose and protect not
only to keep your plant safe, but to
keep your personnel safer. n
Edited by Gerald Ondrey
Author
Robert Schosker is the product
manager/team lead for intrinsic
safety (IS), remote I/O, HART, sig-
nal conditioners, power supplies
and surge protection at
Pepperl+Fuchs Inc. (1600 Enter-
prise Parkway, Twinsburg, OH
44087; Phone: 330-425-3555;
Fax: 330-425-4607; email:
rschosker@us.pepperl-fuchs.
com). Since joining the company in 1995, Schosker has
been focused on technology and product-related sup-
port, and is involved in a wide range of activities and
roles including certifications, sales, and marketing. He
has been the key lead in many IS and HART projects
resulting in the development of new products for intrin-
sic safety and HART infrastructure. Schosker holds a
B.S.E.E. from the University of Akron.
WWW.CHEMENGONLINE.COM JULY 2015
 Feature Report
Advanced Control
Methods for Combustion
Advanced control techniques can
raise efficiency and lower pollutant
emissions in industrial combustion.
The capabilities and adoption of several
methods are discussed
Pal Szentennai
Budapest University of Technology and Economics
Maximilian Lackner
Vienna University of Technology
F
uel combustion in engines,
power plants, boilers, furnaces
and other equipment provides
energy for transportation, heat-
ing, electricity and goods manufac-
turing. Optimizing efficiency while
lowering pollutant emissions is the
main goal of industrial combustion,
and reaching those goals, coupled
with climate-change issues, have
triggered a great deal of research in
this field. One approach to improv-
ing the efficiency and emissions
performance of a combustor is to
apply advanced control techniques
for both new and existing installa-
tions. This article presents several
methods of advanced control for
combustion, ranging from new diag-
nostic techniques to enhanced con-
trol schemes. Key benefits are cost
savings through higher efficiency,
environmental benefits through
lower levels of pollutants, and in-
creased safety.
Classic versus advanced control
Combustion proccesses account for
85% of global primary energy pro-
duction [12], from electricity and
heat generation to propulsion in
terrestrial, marine and aerial trans-
portation. Measures to increase ef-
ficiency encompass the combustion
34 CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014
FIGURE 1. The concept and components of closed-loop
control are shown here
process itself, as well as fuel trans-
portation and storage, and the use of
its outcome (for instance, in cogene-
ration, waste heat is being utilized).
Concerning pollutants, measures to
limit or reduce their impact can be
taken pre-combustion, during com-
bustion and post-combustion. Prom-
inent examples of these strategies
include sulfur-free diesel fuel, air or
fuel staging and selective catalytic
reduction (SCR) of NOx.
Energy-efficiency and emissions
strategies depend on the ability to
control combustion processes. Clas-
sic combustion controls are based on
temperature (for example, adjust-
ment of the air-to-fuel ratio by peak
exhaust gas temperature), flame
emissions (such as the detection of
misfiring in stationary engines by
ultraviolet light detection), mea-
surement of in-cylinder pressure in
engines and measurement of CO,
CO2 or O2 in the exhaust gases.
Advanced combustion-control
strategies generally represent
extensions and improvements to
classical control methods. An ad-
vanced control system has a closed
feedback loop (open-loop control
systems do not have feedback). In
combustion control, a sensor deliv-
ers data to a decision maker, which
sends a signal to an actuator for re-
action (Figure 1). Closed-loop con-
trol can steer a process better than
one using an open-loop system [3
4]. This article focuses on advanced
sensors and decision-maker algo-
rithms, because these control-sys-
tem elements have seen explosive
development in the past two de-
cades, whereas valves and actua-
tors have changed comparatively
little over that time. For more in-
formation on the fundamentals of
process control, see Refs. 56.
Advanced combustion control can
be applied to virtually all types of
combustors, including the following:
gas turbines, furnaces, boilers and
reciprocating internal-combustion
engines, to name a few. Often, it is
possible to retrofit existing instal-
lations so that significant economic
and environmental benefits can be
achieved [7].
Modern combustion options
Based on the huge variety of com-
bustion applications, modern com-
bustion processes differ signifi-
cantly from each other. Distinctions
can be made among combustion
processes of gaseous, liquid and
solid fuels. The two major modes of
combustion are those with the fuel

FIGURE 2. Oxygen concentration can be measured by tunable diode-laser absorp-
tion spectroscopy [10]
and oxidizer premixed and those
where they are not premixed. There
are also differences between lami-
nar and turbulent combustion. For
technical applications, turbulent
combustion dominates, because it
speeds up the process. However,
non-premixed combustion is some-
times preferred for safety reasons.
The following combustion processes
can be termed advanced:
Lean combustion (to achieve low-
NOx emissions)

Carbon capture and storage
(sequestration of CO2 from
combustion)
Chemical looping combustion
Oxyfuel combustion
Co-combustion of fossil fuels with
biomass

Combustion of alternative fossil
fuels, such as methane hydrate
HCCI (homogeneous charge com-
pression ignition) [8]
Advanced combustion is also dis-
cussed in Refs. 2 and 9.
Advanced sensors
Combustion processes are difficult
to probe, since high temperatures
are involved. Often, conditions are
transient, high pressures are pres-
ent and multiple interferences, such
as soot particles and background
radiation, disturb measurements.
Parameters to be measured include
temperature, pressure and species
concentrations. Generally, one can
distinguish in-situ measurements
(which probe the combustion pro-
cesses directly at the spot) and
ex-situ measurements, which draw
a sample that is analyzed in a con-
trolled manner outside of the pro-
cess (for example, in a gas chroma-
tography instrument). The criteria
for selecting suitable measurement
methods are time resolution, sensi-
tivity, type of measurement (such
as point-measurement or two-di-
mensional, spatially resolved or in-
tegrated signal) and cost.
Optical, and particularly laser-
based methods, lend themselves
to in-situ measurements. A major
advantage of optical techniques is
that they do not perturb the probed
system, hence avoiding measure-
ment errors from sampling. Light
can interact with matter in three
basic modes: absorption, emission
and scattering.
Advanced light-absorption sen-
sors. The technique of tunable
diode-laser absorption spectros-
copy (TDLS) can measure species
concentrations and temperatures
of simple gases, such as CO, CO2,
O2, NH3 and CH4, quantitatively
by selective light absorption in the
infrared spectral region by the tar-
get molecules. By tuning the laser
wavelength around the respective
absorption feature, users can com-
pensate for non-specific effects,
such as beam steering, background
radiation and partial beam blockage
by soot particles. A typical setup in
the stack of a combustion process is
depicted in Figure 2. It consists of a
transmitter and a receiver unit.
In an exhaust stack, this type of
sensor can measure NH3, for ex-
ample, so it can minimize the use of
urea supplied to the NOx removal
unit further upstream. Other exam-
ples are given for a municipal waste
incineration plant (Figure 3). In this
example, oxygen and temperature
are measured in the combustion
chamber. The measurement range
for O2 is 015% (accuracy 0.2%),
and that for the temperature 750
1,100 C (accuracy 30C). The re- The term decision maker refers
sponse time is 13 s. The signal can
be used to control the feed of fresh
air. This is important because the
composition of the fuel varies a great
deal and the measurement can help
achieve optimal combustion.
CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014
On the exhaust side, emissions
control based on TDLS measures
NH3 and HCl to control fluegas
treatment for optimum cost and
lowest emissions. Cost savings can
be achieved by avoiding excessive
chemicals dosing and lowering cor-
rosion. TDLS has become a mature
technology in advanced combustion
control. A related approach, infrared
absorption tomography for active
combustion control, is described in
Ref. 12.
Advanced sensors based on light
emission. A technique based on
light emission that combustion re-
searchers frequently use is laser-
induced fluorescence spectroscopy
(LIF). It can be deployed to obtain
two-dimensional images (planar-
LIF, P-LIF) of species concentra-
tions, including radical species such
as OH. However, the setup is com-
plex, so this technique has not yet
found industrial use for combus-
tion-control purposes. The test of a
closed-loop equivalence ratio con-
trol of premixed combustors using
a chemiluminescence signal is de-
scribed in Ref. 13.
Advanced sensors based on
light scattering. Laboratory-
based methods that rely on scat-
tering are Raman scattering and
Cars (coherent anti-Stokes Raman
scattering) spectroscopy. Like LIF,
due to their complex experimental
setup and difficult data evaluation
and interpretation, they are not
yet used for industrial combustion
control applications.
Advanced sensors based on other
signals. In Ref. 14, an advanced
closed loop combustion sensor based
on ionization is described. Ref. 15
uses pressure signals, and Ref. 16
image-based controls for a closed-
loop setup. A review of sensors for
combustion control is provided else-
where [17].
Advanced decision makers
to the computer hardware and soft-
ware necessary to run the control
algorithms that govern the adjust-
ment of the combustion system pa-
rameters. The hardware basis of the
central, decision-making part of ad-
35
 Feature Report
vanced combustion-control systems
will always be chosen in accordance
with the expectations of the specific
application. These expectations can
vary widely among applications,
even in cases where the combus-
tion itself is basically similar. Gas
turbines are applied, for example,
in both stationary power plants and
in jet aircraft engines. The expecta-
tions of the former case allow rack-
mounted hardware elements in an
extra chamber, while special micro-
computers (embedded systems)
of minimal sizes and weights are
required in the aviation industry.
Rack-mounted hardware and em-
bedded systems can be considered
as the two ends of the spectrum of
the hardware that is generally used
for decision makers of combustion
control, and selecting among them
will be mostly determined by the
application area, rather than by the
combustion process itself.
Although the hardware platforms
of the control decision-maker can
be very different, a common re-
quirement is superior reliability.
The same is also true for the soft-
ware side of the decision makers.
In order to satisfy the need for high
reliability with an acceptable cost,
supplier companies deploy many
modern tools of quality control and
standardization on several levels.
Another consequence of the need
for high reliability may be some de-
gree of conservatism with regard to
new ideas for improving reliability.
This phenomenon can be observed
in many other similar situations
in which reliable operation is an
absolute priority. Several industry
branches, including oil-and-gas,
chemicals and food, are character-
ized by the presence of high pres-
sures and temperatures, and the
involvement of hazardous materials
and expensive assets and raw mate-
rials. This necessitates a high level
of security. Although for industrial
combustion control this conserva-
tism may be somewhat exaggerated,
numerous statistical analyses [18
20] do suggest a definite lag in the
use of advanced control technologies
for combustion processes instead of
traditional ones, compared to the
36 CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014
FIGURE 3. TDLS measurements are used in a municipal waste incinerator. The dust
load is 510 g per standard cubic meter, the path length of the measurement is 38 m.
Fast measurements with large temperature variations pose the main challenge [11]
use of advanced controllers in other
types of industrial processes (see
box, p. 37).
Built-in software tools realizing
some advanced control algorithms
as part of decision makers will be
provided by most suppliers of in-
dustrial control systems. Although
very often, ranges and functional-
ities of these tools are still rather
limited. In these cases, an option
for realizing them is an external,
general purpose (maybe PC-based)
computer, where the need for clear
interfaces between traditional and
advanced control systems will be
set up and realized. This approach
can be followed not only in the case
of new control systems for combus-
tions systems, but also while up-
grading existing plants.
All advanced methods offered
by the control theory to be real-
ized as software elements of deci-
sion makers cannot be discussed
here, of course. The literature is
voluminous, but some textbooks are
available [2529]. Some existing
directions and results of modern
control theory may have crucial sig-
nificance in other application areas
like robotics and flight stabilization
[3032]. However, a brief overview
is included here of those methods
that (1) are mature enough, and
(2) show significant potential for
use in industrial combustion con-
trol applications. Most of them are
model-based procedures, a fact
that revalues all activities on mod-
eling combustion systems. Both
theoretical and empirical model-
ing directions have their roles and
importance, and a rapid increase
of semi-empirical modeling can be
observed nowadays [33].
Model predictive control. Almost
all distributed control system (DCS)
vendors offer model predictive con-
trol (MPC) as an advanced exten-
sion to their standard control algo-
rithms [15], and this appears to be
the most frequently used advanced
control method currently. The rea-
sons for its relatively wide use come
from its relative simplicity and from
its inherent properties that fit well
with the general requirements of
industrial combustion control.
The history of MPC has resulted in
several variants, the most important
of which is DMC (digital matrix con-
trol). It can be considered as a subset
to MPC, characterized by simpler al-
gorithms requiring less online com-
putational demand, and the lack of
some services of MPC, like handling
constraints inherently.
This is actually one of the most
important advantages of MPC in
many real industrial applications.
While theoretical control may disre-
gard constraints, real applications
definitely cannot. Examples in com-
bustion processes are actuators (for
instance, fuel-supply valves), which
are evidently characterized by their
limited operating ranges. Further-
 ADOPTION OF ADVANCED CONTROL TECHNOLOGIES

A
dvanced control techniques offer opportunities for effectively
handling combustion systems that have strong nonlinearities
or internal couplings, and that therefore cannot be satisfac-
torily handled by the commonly applied traditional PID (propor-
tional, integral derivative) controller, which is characterized by
linear and scalar behaviors. Published data are available on the
shares of advanced control methods for several years up to 2008,
which clearly show that the application of advanced control meth-
ods for power plants and industry boilers are far below those in
the chemical, oil-and-gas and petroleum refining industries, for
example. In 2008, the power industry spent the most on DCS (dis-
tributed control system) upgrades and replacements, while it spent
the least on advanced process control (see Table 1 in Ref. 18).
What reasons could explain these data? Instead of exact an-
swers, some differences will be identified and discussed here. It
may be that combustion processes do not dictate so definitely the
applications of advanced controllers, because some other chemi-
cal processes are characterized by much stronger nonlinearities.
But in reality, combustion processes are also nonlinear, especially
when considering their entire load ranges. It is true that these non-
linearities are less significant, and they predominantly can still be
kept in operation by means of traditional linear tools. However,
industrial practice shows clearly that this approach leads to ef-
ficiency reductions already in the simplest case of a power plant
boiler running below its base load.
TABLE 1. RATIOS OF WORLDWIDE SALES OF APC
RELATIVE TO WORLDWIDE SALES OF DCS*
Industry Refining Oil and Chemical Power
gas
Ratio of APC 0.092 0.061 0.050 0.010
sales over
DCS sales
* Ratios of worldwide sales of APC (advanced process control) relative
to worldwide sales of DCS (distributed control systems) in several in-
dustries in 2008. Adapted from Ref. 18.
In some cases of combustion technologies, the suboptimal efficiency
associated with traditional control methods is no longer acceptable,
or the traditional control methods simply do not work. In these cases,
the application of advanced controllers is dictated already. Many
such examples were reported together with success stories of solving
them [2124]. They include a wide range of combustion processes,
from compensating for unacceptable fluctuations of biomass boilers
with soft sensors, to lambda control of internal combustion engines
by means of a robust predictive controller.
A further question is, of course, what factors block the dissemina-
tion of advanced control methods in those cases in which an obvi-
ous limitation of traditional control may not necessarily be present,
but where advanced control methods would assure significant in-
creases in efficiency and flexibility, or decreases in lifetime con-
sumptions of critical parts.
For collecting the answers to this question, and for formulating
some advice on overcoming most blocking factors, a workshop
took place in May 2012 at the Budapest University of Technology
and Economics. A group of experts in three categories plant
operators, equipment suppliers and researchers came together
from four countries to discuss the issue in an unstructured, inter-
active format. From the workshop, a number of basic statements
emerged. They are summarized here.
An important reason for the low level of advanced control meth-
ods in combustion processes in the energy industry is the presence
of gaps between several actors. A lack of communication exists
among people and institutions working on the same topic. Key
people involved working on industrial combustion projects in the
areas of control, energy, management and financials are often
unable to balance benefits and drawbacks from the perspective of
each of their separate fields. Several platforms for bringing people
together could help the situation, as could the use of a common
language. This common language could be money. However, ex-
pressing certain benefits (such as safety, usability and lifetime re-
duction) in the context of money is not straightforward. Education
could also help in connecting control theory with system theory,
university with industry, and practice with theory.
Safety is another issue that slows down the utilization of advanced
control in combustion processes. Safety is the highest priority,
however, it can be warranted not only by traditional controllers
but also by means of a proper structure in the control system. Tra-
ditional controllers should be used on a higher, supervising level,
and very clear interfaces should be built with the advanced control
elements [21]. Consider that this approach will also assure further
benefits. Additional certification costs for safety-related control
elements can be avoided, as can some of the risk arising from
frequent version changes of the hardware and software platforms
running the advanced control computation methods.
Software elements of advanced decision makers must follow the
same safety-related requirements as their hardware components,
as discussed above. Thats why, also the above discussed consid-
erations in applying them should be followed in this case es-
pecially regarding the structure and clear interfaces joining tradi-
tional and advanced elements. The same idea also suggests that
the range of control tasks to be realized on the basis of advanced
methods must be selected carefully. Positioners of actuators are, for
example, advised to be kept unchanged for simplicity and reliabil-
ity. Also, simple tasks of low significance should not be involved in
the area of tasks to be realized by advanced control methods.

more, MPC is also able to consider
other constraints without direct re-
lationships to the process in ques-
tion. If the control system is well de-
signed, this ability can be perfectly
utilized for limiting temperature
stresses in key structural elements
of a combustion system during load
shifts, for example.
Another advantage of MPC
against the traditional PID is that
is it multidimensional. As a result,
internal cross-couplings of the com-
bustion process can be handled
inherently. There is no need for
figuring out the significance of the
cross-couplings. An example is the
cross-couplings between load con-
trol and lambda control in a boiler.
The basic MPC algorithm itself
is a discrete time on-line optimi-
zation, which requires a dynamic
process model. The more accurate
the model, the better the control
performance that can be expected.
However, model inaccuracies will be
effectively compensated by the con-
troller. Prior to the controller design,
the constraints must be given, along
with two further parameters, the so-
called weight matrices. They form a
very clear representation of the de-
sign criteria for balancing between
two contradictory requirements of
all control tasks accuracy and
quiet actuator movements.
CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014
Intelligent control methods.
With respect to control methods,
the term intelligent suggests
those with origins in artificial intel-
ligence research. Two basic control
methods fall into this category: ar-
tificial neural networks and fuzzy
control. Both are excellent tools for
controller design in cases where the
formulation of control-system rules
on the basis of traditional mathe-
matical formalism is very difficult.
In many situations, the control
rule is available by verbal state-
ments only, which arise from human
thinking. This set of control rules can
be called expert knowledge, and a
very effective way of representing
37
 Feature Report

it is a new direction of mathemati-
cal set theory called fuzzy logic.
Artificial neural networks are very
effective when the learning capa-
bilities and methods of the human
brain need to be applied on the basis
of corresponding input and output
data sets of a system. Although sev-
eral similarities exist between these
two intelligent control methods, and
their combined application is a pop-
ular solution, they will be discussed
further separately.
The central element of a fuzzy
controller is a rule base, containing
rules formulated on an if ... then ...
logic scheme. (Example: if the mix-
ture temperature is very high, then
set coolant valve to totally open.)
The previous fuzzy control element
is called fuzzification, while the
following one is defuzzification.
In the prior one, analog quantities
(such as temperature) will be fuzzi-
fied; that is, ordered into the dis-
crete groups (sets) that correspond
to intermediate, high, and very
high. An important point is that
the borders of these groups (or sets)
are not crisp, like they would be in
classical set theory. This grouping
is much closer to human thinking,
which generally does not draw a
crisp borderline between the sets
distinguishing the age of a person
to sets like youth or middle-aged,
for example. Defuzzification is a
procedure in the opposite direction.
The characteristics of a fuzzy con-
troller make it effective for control-
ler design. The simple realizability
of each element of the fuzzy con-
troller is another advantage. This
simplicity of operation allows the
FIGURE 4. Conventional
control (left) and advanced
control (right) of a power
plant. The latter allows a
higher average boiler operat-
ing temperature, which trans-
lates into efficiency gains [21]
possibility of using it in low-cost ap-
plications. Limitations of fuzzy con-
trollers arise from the basic princi-
ple, which may result in unsmooth
outputs. Clear stability criteria for
those are still lacking.
The basic concept behind artifi-
cial neural networks (ANN) was
inspired by the human nervous
system, the basic element of which
is the neuron. Like its biological
counterpart, an artificial neuron
is a multi-input, single-output ele-
ment, which is organized into net-
works that give rise to multi-input,
multi-output (MIMO) systems. The
strengths between the neuron in-
terconnections can be varied, which
establishes the basis for the ability

References 12. Gouldin, F.C., J.L. Edwards, Infrared absorp-

1. Lackner, Maximilian, rpd B. Palots, Franz
Winter, Combustion: From Basics to Ap-
plications, Wiley-VCH Verlag GmbH & Co.
KGaA, Weinheim, Germany, ISBN: 978-
3527333516. 2013.
2. Lackner, Maximilian, Franz Winter, Avinash
K. Agarwal, Handbook of Combustion, Wi-
ley-VCH Verlag GmbH & Co. KGaA, ISBN:
978-3527324491, 2010.
3. Carlucci, A.P., and others, Advanced closed
loop combustion control of a LTC diesel en-
gine based on in-cylinder pressure signals,
Energy Conversion and Management, 77, pp.
193207, January 2014.
4. Al-Durra, Ahmed, and others, A real-time
pressure estimation algorithm for closed-
loop combustion control, Mechanical Systems
and Signal Processing, 38 (2), pp. 411427,
July 2013.
5. Tan, K.K., Qing-Guo Wang, Chang C. Hang, Tore
Haggl, Advances in PID Control, Springer,
London, ISBN: 9781447112198, 2012.
6. Baukal, Charles E., Industrial Combustion
Pollution and Control, Taylor & Francis,
ISBN: 9780824746940, 2003.
7. Advanced Combustion. http://www.carbon-
trust.com/media/147147/j7971_ctl058_ad-
vanced_combustion_aw.pdf. 2014.
8. Haraldsson, Gran; Tunestl, Per; Johans-
son, Bengt; Hyvnen, Jari, HCCI Closed-Loop
Combustion Control Using Fast Thermal
Management, In SAE Transactions, Journal
of Engines 113(3). pp. 599610, 2004.
9. Syred, Nick, Artem Khalatov, Advanced
Combustion and Aerothermal Technologies:
Environmental Protection and Pollution
Reductions (NATO Science for Peace and
Security Series C: Environmental Security),
Springer, ISBN: 978-1402065132, 2007.
10. Webpage: http://www.analyticjournal.de/
images_firmenregister/siemens_sitrans_sl_
aufbau.jpg, 2008.
11. Webpage: http://www.analyticjournal.de/im-
ages_siemens_user_2004/siemens_haefller_
sum_2004_5_gr.jpg, 2004.
tion tomography for active combustion con-
22. Banaszuk, A. and A. Annaswamy, Control
trol, Chapter 2 in Combustion Processes in
Propulsion, Taylor and Francis Group, pp.
920, 2005.
13. Docquier, Nicolas, and others, Closed-loop
equivalence ratio control of premixed com-
bustors using spectrally resolved chemilumi-
23. Krstic, M. and A. Banaszuk, Multivariable
nescence measurements, Proceedings of the
Combustion Institute, 29 (1), pp. 139145,
2002.
14. Glavmo, Magnus, Peter Spadafora and Rus-
sell Bosch, Closed Loop Start of Combustion
Control Utilizing Ionization Sensing in a
Diesel Engine, International Congress and
Exposition, Detroit, Michigan, SAE paper
1999-01-0549, March 14, 2009.
15. Samad, T. and A. Annaswamy, (Eds.), The
Impact of Control Technology: Overview,
Success Stories, and Research Challenges.
Available at www.ieeecss.org. IEEE Control
Systems Society, 2011.
16. Chen, Junghui, Yu-Hsiang Chang, Yi-Cheng
Cheng, Chen-Kai Hsu, Design of image-
based control loops for industrial combustion
processes, Applied Energy, 94, pp. 1321,
June 2012.
17. Docquier, Nicolas and Sbastien Candel,
Combustion control and sensors: a review,
Progress in Energy and Combustion Science,
28 (2), pp. 107150, 2002.
18. Smuts, J.F. and A. Hussey, Requirements for
Successfully Implementing and Sustaining
Advanced Control Applications. Presented at
the 2011 ISA POWID Symposium, 06/05/11
06/10/11, Concord, NC, presented at the 2011
ISA POWID Symposium, Concord, N.C., 2011.
19. Dittmar, R. and B.-M. Pfeiffer, Modellbasi-
erte prdiktive Regelung in der industriellen
Praxis, Automatisierungstechnik, 54 (12), pp.
590601, 2006.
20. Qin, S.J. and T.A. Badgwell, A survey of in-
dustrial model predictive control technol-
ogy, Control Eng. Pract., 11(7), pp. 733764,
2003.
21. Szentannai, P., Power Plant Applications of
Advanced Control Techniques, Process Engi-
neering, ISBN: 978-3902655110, 2010.
of Combustion Instability, in The Impact of
Control Technology: Overview, Success Sto-
ries, and Research Challenges. Available at
www.ieeecss.org ., IEEE Control Systems So-
ciety, pp. 207208, 2011.
adaptive control of instabilities arising in jet
engines, Control Eng. Pract., vol. 14, no. 7, pp.
833842, 2006.
24. Sardarmehni, T., and others, Robust predic-
tive control of lambda in internal combustion
engines using neural networks, Arch. Civ.
Mech. Eng., 13(4), pp. 432443, 2013.
25. Hangos, K.M., J. Bokor, and G. Szederknyi,
Analysis and Control of Nonlinear Process
Systems, 2004th ed. London: New York:
Springer, 2004.
26. Tatjewski, P., Advanced Control of Indus-
trial Processes: Structures and Algorithms.
Springer, 2007.
27. Astrom, K.J. and D. B. Wittenmark, Adap-
tive Control, 2nd ed., Mineola, N.Y: Dover
Publications, 2008.
28. Astrom K. and B. Wittenmark, Computer-
Controlled Systems: Theory and Design, 3rd
ed., Mineola, N.Y: Dover Publications, 2011.
29. Grne, L. and J. Pannek, Nonlinear Model Pre-
dictive Control: Theory and Algorithms, 2011
edition. London; New York: Springer, 2011.
30. Zelei, A. and G. Stpn, Case studies for com-
puted torque control of constrained underac-
tuated systems, Mech. Eng., 56 (1), pp. 7380,
2012.
31. Magyar, B. and G. Stpn, Time-optimal com-
puted-torque control in contact transitions,
Mech. Eng., 56 (1), pp. 4347, 2012.
32. Siqueira, D., P. Paglione, and F. J. O. Moreira,
Robust fixed structure output feedback flight
control law synthesis and analysis using sin-
gular structured value, Aerosp. Sci. Technol.,
30 (1), pp. 102107, 2013.
33. Havlena, V., J. Findejs, Application of model
predictive control to advanced combustion
control, Control Engineering Practice, 13 (6),
pp. 671680, June 2005.

38 CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014

to learn. Special optimum-seeking
algorithms are used during the
learning phase to find these inter-
connecting strengths (weightings),
so that the network outputs that
result from training inputs would
best fit the target outputs that cor-
respond to the same inputs. If the
output of a combustion process will
be applied as training input to the
ANN, and the process input is ap-
plied as a target output of the ANN,
the artificial neural network will
be trained as a controller. After the
training phase, the ANN will be able
to find the process input to match
the desired process output.
General experience with artificial
neural networks suggests that they
have excellent capability for learn-
ing the corresponding inputoutput
pairs. However, in the case of inputs
that lie somewhat further from the
training-input data, error margins
from the ANN output may be sur-
prisingly large.
Advanced controller tuning. A
possible upgrade of traditional con-
trollers, such as PID controllers, is
the application of advanced meth-
ods for optimizing their parameters,
or possibly changing their configu-
rations. Most of the configurations
on this basis belong to a wider theo-
retical approach called adaptive
control. Most important for com-
bustion control are gain schedul-
ing and multimode control. Both the actual process is not linear, then
concepts require a scheduling vari-
able, the actual value of which es-
sentially characterizes the process
behavior. In the case of combustion
processes, a very good candidate for
this is the load signal. However, the
introduction of further variables
may result in two- or multi-dimen-
sional scheduling variables as well.
In the most frequently applied
case, the one-dimensional schedul-
ing variable will be divided into dis-
crete ranges, within which one set
of tuning parameters or controller
configurations will be applied. If the
parameter set of a controller will
only be changed while shifting from
one section to another one along
the scheduling variable range, the
system is called gain scheduling,
while in other cases, the entire con-
troller will be changed. The latter
case is called multimode control,
which has greater adaptivity but
also higher complexity.
The relative simplicity of these
techniques is a great advantage, al-
though assuring bumpless changes
between the operating ranges re-
quires significant designer effort
and knowledge.
Loop decoupling. Similar to the
previous controller tuning, loop
decoupling is not really an ad-
vanced control method, but rather
an advanced extension to classical
control methods. This technology
can be applied to processes char-
acterized by strong internal cross-
couplings. That is, for processes
in which the independent, scalar
control loops strongly disturb each
other. Based on a process model, a
so-called decoupler can be designed.
A decoupler is a dynamic system
in the DCS, and its outputs will be
directed to the inputs of the actual
process. The aim of this design is
that the virtual system that results
from the decoupler plus the process
itself form a system free of internal
cross-couplings. And this type of
system can already be controlled by
a series of classical controllers. The
limitations of this method are evi-
dent, however. If linear controllers
(like classical PIDs) are intended to
be used for the virtual system, and
the method can get rather compli-
cated.
Benefits of advanced control
The advantages of advanced com-
bustion control are mainly higher
efficiency and lower pollutant lev-
els, although other goals such as
higher safety levels are possible.
Advanced control can be character-
ized by fast response times, and al-
lowing a process to be run within a
narrower window of process param-
eters (Figure 4).
When the deviations around the
setpoint are smaller, it is possible,
as the example shows, to increase
the average operation temperature
of the combustor (in this case a
boiler), and still avoid overheating
the furnace. By raising the operat-
CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014
ing temperature, the overall ther-
mal efficiency can be augmented.
The example discussed earlier
illustrates a special characteristic
of the advanced combustion con-
trol. Namely, that two goals can be
served simultaneously in con-
ventional cases, these are in oppo-
sition to each other. For example,
the introduction of carbon capture
facilities will significantly increase
fuel consumption, and by extension,
operating costs, while an efficiency
boost will decrease both emissions
and fuel cost. And this end can be
achieved through the use of ad-
vanced combustion control, the
investment costs of which are far
lower than those of any other modi-
fications in the process itself.
It is likely that several advanced
combustion-diagnostic methods
currently used exclusively in labo-
ratories will eventually find their
way into industrial applications to
further optimize combustor perfor-
mance in various applications. And
existing combustors stand to bene-
fit as well, beause they can often be
upgraded with advanced combus-
tion control technology [7].
Edited by Scott Jenkins
Author
Pal Szentannai is a professor
at the Budapest University of
Technology and Economics
Department of Energy Engi-
neering (Muegyetem rkp. 9,
H-1111, Budapest, Hungary;
Phone: +36 1 463 1622; Email:
szentennai@energia.bme.hu)
He is also the executive com-
mittee member designated by
Hungary in the IEAFBC (In-
ternational Energy Agency
Fluidized-Bed Conversion) Implementing Agree-
ment. He has authored and edited several books
on the topics of advanced power-plant process
control and fluidized-bed combustion. Szentan-
nai has several years of direct experience in the
energy industry as an engineer working on de-
sign and commissioning of numerous thermal
and control systems.
Maximilian Lackner is a
chemical engineer who lec-
tures at Vienna University of
Technology (Getreidemarkt
9/166, 1060 Vienna, Austria;
Phone: +43 681 81 82 6762;
Email: maximilian.lackner@
tuwien.ac.at) and Johannes
Kepler University Linz. He
has developed in-situ laser
diagnostics for combustion
processes and carried out re-
search on laser ignition. Lackner has founded
five companies. He has written the textbook
Combustion: From Basics to Applications
(Wiley VCH, 2013). Lackner is also editor of the
five-volume reference work Handbook of Com-
bustion (Wiley VCH, 2010).
39
 Engineering Practice
Flare Consolidation Considerations
Consolidating multiple flares can help mitigate emissions and maintenance
Chris Ng and Zubin Kumana
Siemens, Process Safety Consulting
M
any petroleum-refining
and petrochemical fa-
cilities have experienced
tremendous growth in
recent years in response to increas-
ing demand for fuels and chemical
precursors. Due to environmental
regulations and economic consid-
erations, it is frequently more de-
sirable to increase the throughput
on an existing facility rather than to
build a new plant. As shown in Fig-
ure 1, the total crude-oil processing
capacity in the U.S. has increased
steadily as the number of operable
refineries declined over the years
[1,2]. The expansion of these facili-
ties has typically resulted in an array
of independent flare systems built
to meet the requirements of specific
expansion projects.
Flare consolidation is one attrac-
tive solution that is based on a ho-
listic approach to updating the exist-
ing infrastructure to meet growing
demands. Integrating separate flare
systems into a single flare network
can be an economically and envi-
ronmentally viable solution to meet
challenging operational and regula-
tory requirements.
The benefits
Generally speaking, independent
flare systems could be integrated
through the installation of cross-
over piping jumpers at appropriate
location(s) between the main lines
or headers of two existing systems.
The rationale and engineering re-
quirements for these modifications
are described below.
Though many factors can affect
the sizing of flare systems and relief
headers, it is usually a type of utility
failure, such as total or partial power
failure or a loss of cooling water flow,
that dictates the design basis of the
disposal system. In most petroleum
58
refineries, either partial power failure
or cooling water failure is typically the
controlling scenario.
Such events are generally local-
ized to a particular area of the plant.
For instance, if a chemical facility
has three independent cooling-water
systems, the loss of any one system
should not result in releases through-
out the entire facility.
Thus, the availability of a plant-wide
flare system can provide increased
relief capacity for such scenarios to
support the expansion of processing
capacity. Since safe operating limits
can be set by relief- and flare-system
capacities, a well-implemented flare
consolidation may lead to an in-
crease in the plant capacity through
modest upgrades to debottleneck
the flare systems.
Ease of maintenance
With a single large network of mul-
tiple flare systems servicing an entire
facility, some of the system capacity
gained may allow for any one of the
multiple flares to be removed from
service for maintenance without sig-
nificantly impacting the operation of
the facility.
With independent flares, flare
maintenance would require a
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
istock.com/hhakim
planned shutdown of the affected
units; whereas emergency flare
maintenance would require either an
unplanned shutdown or turndown of
a portion of the plant. In both cases,
production would likely be curtailed
and would thereby result in higher
overall costs.
The ability to perform emergency
or scheduled maintenance on the
flare system with minimal impact on
operations could lead to a signifi-
cant increase in overall up-time for
many facilities.
EPA requirements
The U.S. Environmental Protection
Agency (EPA; Washington, D.C.) has
released a National Petroleum Re-
finery Initiative to encourage refiners
to make a commitment to emission
reductions. Since 2000, 109 refiner-
ies in 32 states and comprising over
90% of the total U.S. refining capac-
ity have agreed to comply [3]. As
part of this initiative, a covered facil-
ity may not emit more than 500 lb of
sulfur dioxide in a 24-h period during
non-emergency situations, based
on Emergency Planning and Com-
munity Right-to-Know Act (EPCRA)
Section 304.
One option to minimize emissions
FEBRUARY 2015
is to install a flare-gas recovery unit
(FGRU) for each flare system. The
consolidation of multiple flare sys-
tems into a single flare network could
significantly reduce the cost of com-
pliance as a single FGRU unit could
Number of U.S. operable refineries
be installed to service the entire facil-
ity (or two FGRUs could be installed
to improve availability). A simplified
diagram of flare debottlenecking op-
tions using this approach is shown
in Figure 2. In addition, consumption
rates of various utilities, such as air,
steam or fuel gas to support normal
flare operations, could be reduced
if a cascaded or staggered flare ar-
rangement was adopted.
OSHAs PSM standard
The consolidation of multiple flare
systems into a single integrated
network requires a thorough under-
standing of the design basis for the
flare systems. There is no reasonable
means of combining multiple inde-
pendent systems without a concrete
and well-documented design basis
for the flare systems, which in turn
is part of the process safety infor-
mation requirement per the Process
Safety Management (PSM) Standard
29 CFR 1910.119(d)(3)(i)(D) [4] of the
U.S. Occupational Safety and Health
Admin. (OSHA; Washington, D.C.).
Methodologies
All credible flare loads should be
considered to ensure that the modi-
fied system can safely, economically
and efficiently handle the expected
loads. It is also necessary to under-
stand the current layout of the flare
system to optimize flare header con-
nections. A thorough understanding
of the current system and the pro-
posed changes allows the designer
to confirm that the modified system
can operate safely and to validate
that flare maintenance requirements
can be met.
Identify and document
All potential loads to the flare sys-
tem must be analyzed and docu-
mented. The loads can originate
from a variety of sources (such as
relief valves, control valves, bleeder
valves, emergency shutdown [ESD]
systems and so on) and can enter
the flare system due to a variety of
causes. To ensure that the system
is adequately sized, all emergency
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
Number of U.S. operable refineries
250
230
210
190
170
150
130
110
90
70
50
1987
FIGURE 1. In the U.S., the total crude-oil processing capacity has increased steadily as the number of
operable refienries declined over the years
loads (due to area fires, power fail-
ures, cooling water failures and
major equipment failures) must be
considered, documented, and eval-
uated. To properly size the FGRU
and isolation seal drums for com-
pliance with the EPA requirements,
releases during routine operations
and maintenance must also be ac-
counted for and documented.
Develop a model
Evaluation of the location of the ex-
isting flare headers on a plot plan or
other facility map will allow for a quick
overview of the system connections
and aid in the design of proposed
crossovers. Once the consolidated
flare system has been developed, a
hydraulic model of this flare network
can be developed to verify that it
can handle the expected loads ad-
equately. Three types of modeling
will be necessary: emergency load,
non-emergency load and mainte-
nance load.
Emergency load modeling. A
standard method, such as the one
outlined in American Petroleum In-
stitute (API) Standard 521 [5], is ap-
plied to ensure that the system can
handle the expected emergency
loads. Critical items to verify include
the following:
1. Backpressure limits To ensure
that the relief capacities for indi-
vidual sources will remain adequate
despite the pressures built-up in the
disposal system
2. Header velocities To ensure
n U.S. operable crude oil distillation capacity (million barrels per calendar day)
18.0
17.7
U.S. operable crude oil distillation capacity (MBPD)
17.4
17.1
16.8
16.5
16.2
15.9
15.6
15.3
(Raw data from www.eia.gov, ref. 1 and 2)
15.0
1990 1993 1995 1998 2001 2004 2006 2009 2012 2014
Year
that there are no localized areas of
high velocity in the flare network that
could cause mechanical problems
due to excessive momentum forces
or acoustically induced vibrations
3. Flare equipment sizing To en-
sure satisfactory performance of all
relevant flare equipment, such as
liquid knockout pots (for vapor-liquid
separation), flare seals (including
seal height evaluation), flare tip per-
formance (in terms of flare radiation,
noise, flame stability and so on), va-
porizers and heaters and more
4. Mechanical limits of the header
To ensure that the flare header itself
can adequately handle the expected
operating requirements
5. Liquid removal facilities To en-
sure that the integrated flare loads
will not result in excessive mechani-
cal stress due to undesirable flow
patterns (such as high-velocity liq-
uid slugs)
Non-Emergency load model-
ing. All releases to the flare system
during normal operation or routine
maintenance of the facility should be
accounted for. Critical items to verify
include the following:
1. FGRU To ensure that the normal
flare loads from all sources such
as leaking relief valves or normally
open vent lines plus intermittent re-
leases during maintenance (equip-
ment blowdown), and operations
(backpressure control valves, startup/
shutdown requirements, purging)
can be handled by the FGRU. This
information is used to generate the
FEBRUARY 2015 59

Process unit 1 Process unit 2
New process
Other new units
unit
FIGURE 2. A single flare-gas recovery unit (FGRU) could be installed to service an entire facility, as
shown in this simplified diagram of flare debottlenecking options
sizing parameters for the FGRU (es-
pecially the compressor).
2. Seal drums To ensure that ex-
cessive or insufficient pressure will
not disrupt the liquid seal in the seal
drum and result in unintended re-
lease from a flare stack during nor-
mal operations. Consideration must
also be given to ensure that the seal-
ing fluid does not freeze.
3. Liquid knockout While most
FGRU compressors can handle lim-
ited liquid on the inlet, large liquid
flow to these compressors should
be avoided.
Maintenance load modeling. Eval-
uation of both emergency and non-
emergency loads to ensure that all of
the established limits would not be
exceeded and that the facility could
be operated safely should an indi-
vidual flare (or portion of the header)
be removed from service for mainte-
nance reasons.
The results of this modeling could
lead to the modification of the con-
solidated header design or flare
operation in areas where facility up-
grades are planned. The changes
may include the allocation of larger
diameter headers, adding mechani-
cal supports, modifying relief devices
or flare equipment, adding process
safety systems, implementing oper-
ational changes, training and so on.
The flow distribution and flow pat-
tern of the flare loads throughout
the integrated flare system must
be evaluated, because consolidat-
ing flares could result in areas of in-
creased corrosion or erosion rates,
temperature-induced embrittlement,
mixing of chemically incompatible
fluids, or a host of other operational
60
Flare 1
Other units
Flare 2
FGRU
Flare 3
Flare debottlenecking option(s) shown
as black/dashed
or safety concerns.
Fluids follow the path of least re-
sistance. Sizing a flare header that
is too large could result in overload-
ing one flare stack in a consolidated
system even though the total system
may have sufficient capacity. Addi-
tional modeling, such as Flare Quan-
titative Risk Assessment (QRA) may
be required to account for the time
dependency of flare loads or the sta-
tistical probability of various emer-
gency scenarios.
Other design considerations
While the multiple benefits of flare
consolidation have been described
above, other design considerations
could preclude its implementation. In
particular, segregation of relief streams
could be necessary due to incompat-
ible functional requirements in terms
of operating pressure, temperature,
fluid composition, or physical proper-
ties of the relief streams. Issues such
as corrosion, toxicity, reactivity, phase
change, and flow pattern of the com-
bined relief streams upon flare con-
solidation should be evaluated.
Concluding remarks
While flare consolidation may be
driven by environmental or regulatory
considerations, a properly consoli-
dated flare could provide additional
flare capacity, improve operational
flexibility, and result in operational
savings for the whole facility.
Multiple independent flare sys-
tems are not always practical, and
there may be more than one option
to eliminate the need for flare con-
solidation. However, performing flare
maintenance only during unit shut-
CHEMICAL ENGINEERING
downs or turnarounds, installing a
dedicated FGRU for each flare, and
other options each have their own
limitations or drawbacks.
When implemented properly, a
flare consolidation program could be
an economically attractive solution
that allows facilities to maximize the
utilization of their disposal systems
while maintaining compliance to
regulatory requirements of the EPA
and OSHA. Sound engineering is re-
quired to ensure that environmental
impacts and maintenance concerns
are not mitigated at the expense of
facility safety.
Edited by Gerald Ondrey
References
1.Energy InformationAgency,www.eia.gov/dnav/pet/hist/
leafhandler.ashx?n=pet&s=mocleus2&f=m
2.Energy InformationAgency,www.eia.gov/dnav/pet/hist/
leafhandler.ashx?n=pet&s=8_na_8O0_nus_c&f=a
3. www2.epa.gov/enforcement/petroleum-refinery-
national-case-results
4. Process safety management of highly hazardous
chemicals, Title 29, Part 1910.119, US Code of
Federal Regulations
5. Pressure-relieving and Depressuring Systems, API
Standard 521, 6th edition, January 2014
Acknowledgements
The authors would like to thank our Siemens (and old
Berwanger) colleagues who have contributed invalu-
able insights in pressure relief and flare design over the
years. This article would not have been possible without
their support.
Authors
Chris Ng is a technical advisor
at Siemens, Process Safety Con-
sulting (4615 Southwest Freeway,
Suite 900, Houston, TX 77027;
Phone: +1-713-570-1204; Email:
chris.ng@siemens.com). His ex-
pertise includes process design
and safety, steady-state and dy-
namic modeling of multiphase
flowlines and process unit opera-
tions, process hazard and risk analysis. Ng has executed
upstream and downstream design projects for major
operating companies. He has a B.S.Ch.E. degree from
the University of Western Ontario in Canada, and is a
licensed professional engineer in Texas.
Zubin Kumana is a consultant
at Siemens , Process Safety Con-
sulting (same address as above;
Email: zubin.kumana@siemens.
com). He has a solid understand-
ing of relief device sizing and in-
stallation codes and standards,
including ASME, API, and Design
Institute for Emergency Relief Sys-
tems (DIERS), and has performed
projects for many of the major domestic and interna-
tional oil and gas companies. Kumana holds a B.S.Ch.E.
degree in from the University of Florida, is a certified
project management professional, and is a licensed
professional engineer in Texas.
WWW.CHEMENGONLINE.COM FEBRUARY 2015
 Feature
Cover Story
Report
Burner Inspection and
Maintenance
Burners and their
components can be quite
complex. Establishing
maintenance and
inspection best
practices encourages
longterm operational
reliability
Mohammed H. Al-Hajji
Saudi Arabian Oil Co.
(Saudi Aramco)
B
urners are mechanical devices
that are utilized for mixing
proper quantities of fuel and
air, and also for maintaining
a stable flame inside fired equip-
ment. Burners are critical compo-
nents that must be periodically
maintained to ensure the reliability
of fired equipment in the chemical
process industries (CPI), as well as
many other industries. This article
discusses the internal components
and applications for different types
of burners (gas, oil and combina-
tion), as well as the maintenance
procedures that are required to en-
sure the integrity and reliability of
burners in the CPI.
Gas burners
There are two principal types of
gas-fired burners used in the CPI:
raw gas and pre-mix.
Raw-gas burners are used for
most applications. In these burn-
ers, the fuel gas passes through
orifices in the gas tip and is injected
directly into the combustion zone,
where it mixes with air. A stabi-
40 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
Tip
Steam orifices
Fuel oil orifices
Oil
Steam Steam
FIGURE 1. Fuel atomizers are designed
to provide proper mixing between the oil
and steam in oil burners
lizer cone is located just below the
gas to improve combustion stabil-
ity. These burners are suitable for
mounting in plenum chambers (the
area where air enters the burner),
and can be used with preheated
combustion air.
Pre-mix burners are sometimes
used in specialized applications. In
these burners, the kinetic energy
made available by the expansion
of the fuel gas through the fuel-
gas orifice introduces about half of
the combustion air (called primary
air) into the Venturi mixer. This
mixture then exits through a large
burner tip, where it is mixed with
the balance of the combustion air
(secondary air). This secondary air-
flow enters the burner through the
Part 1
pro
x.
2f
t
Ap
Tip
Mixer
Orifice
FIGURE 2. Pilot burners provide both
ignition and stabilization for the
main flame
outer, secondary air register. Pre-
mix burners require less furnace
draft than raw-gas burners. Pre-
mix burners can produce a wide
range of flame shapes.
Oil burners
Steam is always added to assist in
the atomization of liquid fuels. In
oil burners, the steam and oil are
mixed in specially designed fuel at-
omizers, where the kinetic energy
of the steam jets breaks up the fuel
into small droplets. The resulting
mixture of steam and finely dis-
persed oil is then released into the
airstream through a number of ori-
fices in the burner tip. A typical at-
omizer is shown in Figure 1.
The steam used for atomization
must be perfectly dry. If there is
moisture in the steam, this mois-
ture will flash when it mixes with
the oil, causing erratic oil flow.
Furthermore, the atomizing steam
should be superheated.
Mechanical atomization can be
used when steam is not available.
In this method, the kinetic energy
in the oil itself is used for atomiza-
tion by releasing the oil through
the tip under very high pressure.
Mechanical atomization is usually
used only in very large burners or
with extremely clean fuels, since
the very small orifices required in
smaller burners can become plugged
by small dirt or coke particles
in the fuel.
Combination burners
The principles of gas and oil burn-
ers can be combined into a third
category of burner, the combination
burner. A typical combination gas-
and-oil burner can be used to fire
liquid, gas or a combination of liq-
uid and gas fuels, depending upon
the fuel systems supplied with the
burner. Liquid fuel is fired through
a centrally located oil gun. The
oil and atomizing steam are fed
through separate pipes in the feed
tube to the atomizer and burner tip.
The fuel-steam mixture is injected
through orifices in the burner tip
into the primary airstream, where
combustion begins.
Reducing NOx in burners
Combustion operations frequently
create nitrogen oxides (NOx), spe-
cifically NO and NO2, which must
be highly controlled, as they can
be harmful pollutants. Currently,
three burner configuration meth-
ods exist for reducing nitrogen ox-
ides in burners: staged air, staged
fuel and internal fluegas recircu-
lation combined with staged air
or staged fuel.
Staged air burners. These types
of burners work by introducing
100% of the fuel into the burner
and only part of the combustion air
(primary air), thus creating a sub-
stoichiometric flame. This flame
has a reduced temperature and
therefore inhibits NOx formation.
The flame is completed with the ad-
dition of the secondary air to com-
plete the combustion process. This
process allows for greater control at
lower burner loads and also accom-
modates a wider range of fuels.
Staged fuel burners. This burner
method introduces 100% of the
combustion air into the burner and
splits the fuel supply into primary
and secondary volumes. The primary
fuel mixes with the combustion air
to create a flame. Again, as with
staged air burners, the peak flame
temperature is lower, and NOx for-
mation is reduced. Secondary fuel
is added to complete the combus-
tion process. Staged fuel burners
provide greater NOx reduction, as
the fuel supply has a larger effect
on NOx formation. This method is
more commonly used when a con-
sistent fuel supply is available.
Internal fluegas recirculation
burners. This method combines ei-
ther staged air or staged fuel with
internal fluegas recirculation to
help reduce NOx formation. The
best results are obtained where in-
ternal fluegas recirculation is used
to dilute the fuel gas in a staged
fuel burner, creating a gas with a
low calorific value.
Pilot burner
Pilot burners (or, simply pilots) are
small burners used for ignition of
the main burner flame. In some
systems, a pilot can also act as a
stabilizer for the main flame. Pilots
in process-heater burners usually
operate continuously. Individual pi-
lots should be removable for mainte-
nance, even while the process heater
remains in operation. A typical pilot
burner is shown in Figure 2.
More and more process heaters
require the presence of a permanent
pilot with electrical ignition. These
types of pilots use a low-tension ig-
nition rod and usually combine the
function of the low-tension rod with
ionization detection.
When the ignition transformer is
connected directly to the end of the
pilot, standard low-tension cable
can be used. This eliminates the
need for high-tension cable and, as
a result, longer cable runs can be
used from the pilot to the distrib-
uted control system (DCS).
An electrical current is con-
ducted between the flame and the
flame-retention head. The flame is
ionized, and the ionization part of
the flame rod detects this change
in the current. This electrical cur-
rent change is then relayed by a
lamp to indicate the presence of the
flame (Figure 3). It is crucial to en-
sure that the pilot flame has a very
strong core to enable consistent de-
tection of the flames presence in
the pilot burner.
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
FIGURE 3. Pilot burners use electrical
current to detect the flames presence
Burner components
Burners are very complex devices
not only are there many varieties
of burners utilized in the CPI, each
burner is equipped with a number
of integral components that are
critical to operations.
Burner air registers. Air enters
the burner through the air register.
Airflow can be controlled by adjust-
ing the size of the openings in the
register. For natural-draft burners,
the most common type of air regis-
ter consists of fixed and moveable
concentric cylinders, each with
slots. For forced-draft burners, air
registers are controlled either man-
ually or automatically. Burner reg-
isters may also be a single-bladed or
twin-bladed opposed butterfly type.
This variety is used for greater con-
trol of the combustion air pressure.
Plenum chamber. Sometimes
called a windbox, the plenum
chamber is where all the air for the
burners enters the device. The ple-
num chamber may contain control
vanes, as well as the air registers
for the burners. The plenum cham-
ber can serve several purposes,
including noise and emissions re-
duction, single-source combustion
air supply and total airflow to the
furnace controls.
Burner fuel-gas piping. Burner
fuel-gas piping consists of the man-
ifolds and piping that deliver fuel
to the burner tips. The fuel passes
through one or more openings in
the tips, which act as restriction or-
ifices. Here, the fuel is injected into,
and mixes with, the airstream. Var-
ious types of burner tips are used,
depending upon the type of fuel
and the flame pattern desired.
41
 Cover Story
Burner throat tiles. Located at
the burner edge are the throat tiles
of the burner. Throat tiles help sta-
bilize combustion and shape the
flame. The burner tile is usually
shaped so that one section has a
minimum cross-sectional flow area.
This area, called the throat, acts
as a Venturi component in the air-
stream. The opening in most burner
tiles is circular. In some cases, the
tile and other components are de-
signed to produce a non-circular
flame shape. A rectangular-shaped
tile opening can be used to produce
a flat flame, which is needed in
some furnace arrangements.
Swirlers. The function of swirl in
burners is to increase flow turbu-
lence. Turbulence facilitates the
mixing of fuel and air, and various
elements within the burner can
induce a swirl phenomenon. Some
burners contain tangentially dis-
posed doors in the air registers,
which increase turbulence. Other
burners contain axially disposed
spin vanes to impart swirl in the
combustion air. Other burners may
use adjustable swirl vanes to in-
crease turbulence.
Fired equipment
Burners are utilized in fired equip-
ment in the CPI to combust fuel.
The energy created from combus-
tion is mainly used to convert water
into steam for various processes,
or to heat up a process to a de-
sired temperature. The main fired
equipment that utilize burners are
boilers, heaters and heat-recovery
steam generators (HRSGs).
Fired boilers use forced-draft
burners to combust either oil or
gas to heat water and convert it
to steam. Boiler tubes can con-
tain water (water-tube boilers),
or there may be flames in the
tubes (fire-tube boilers)
Heaters use natural-draft burn-
ers that rely on the draft (nega-
tive pressure) in the furnaces
radiant box to induce the air re-
quired for combustion
HRSGs use duct burners for the
supplementary firing to pro-
duce more steam to satisfy the
operation requirements
42 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
Sulfur recovery units
Sulfur recovery units (SRUs)
consist of many different fired-
equipment elements, so they
merit special mention here. The
fired equipment and burners
found in SRUs are detailed in the
following section.
Air and acid preheater. Multi-
ple burners and continuous pilots
are incorporated into air and acid
preheaters. The pilots and burn-
ers are individually supervised,
each with its own flame monitor.
The pilots are lit via integrated
spark igniters.
Reaction furnace. Typically
equipped with one high-intensity
burner, reaction furnaces are an
important element in an SRU. The
high-intensity burner is a forced-
draft burner that requires a rela-
tively high air-pressure drop. The
burner carries out a thermal con-
version process to convert acid gas
into sulfur. This conversion oper-
ates sub-stoichiometrically, mean-
ing that there is insufficient air to
allow complete combustion of the
acid gas. The burner is usually ig-
nited by a high-energy, direct-spark
ignition system. Optical flame
monitors are used to check for the
presence of the main flame.
Reheater burner. Reheater burn-
ers use a single forced-draft, fuel-
gas or fuel-gas/acid-gas fired burner
to reheat the process stream above
the liquid-sulfur dewpoint prior to
entering the next catalytic conver-
sion stage. Burners operate at near
stoichiometric fuel-to-air ratios.
The burner usually ignites by a
high-energy, direct-spark ignition
system. Optical flame monitors are
used to check for the presence of
the main flame.
Thermal oxidizer. Thermal oxi-
dizers use either natural- or forced-
draft burners to provide the heat
input required to combust the SRU
tailgas in the incinerator chamber
to form sulfur dioxide (SO2). Ther-
mal oxidizers can use either single
or multiple burners with or with-
out continuous pilots. Optical flame
monitors can be ultraviolet (UV) or
infrared (IR), and are used to check
for the presence of the main flame.
FIGURE 4. A burner in a thermal oxi-
dizer in a sulfur recovery unit combusts
the incinerator gas to form SO2
Figure 4 shows a typical SRU ther-
mal-oxidizer burner.
Flare system burners
Flare systems form an important
part of the emergency relief sys-
tems in CPI plants, allowing for
safe dissipation and disposal of
certain gases by combustion. The
flare burners tip can be located
at ground level or elevated on a
support structure. Figure 5 shows
a typical flare burner with an
elevated tip.
There are various types of
flares and burner configurations,
and their applications depend on
the requirements of the process.
An awareness of smoke in the
flare system is especially impor-
tant for combustion operations.
The various type of flares include
the following:
Pipe flares are used where the
gases to be burnt do not create
smoke, or there is no require-
ment for smokeless combustion
Steam-assist flares inject steam
into the flare tip through a
series of nozzles to promote
entrainment of air to im-
prove combustion and reduce
smoke formation
Air-assist flare tips introduce
air directly into the flare tip to
improve combustion and reduce
smoke formation
High-pressure (sonic) flare tips
use the high exit velocity of the
gas to be burnt to entrain air,
improving combustion and re-
ducing smoke formation
Burn pits. The burn pit of a
flare system is equipped with a
burner to safely combust hydro-
 Cover Story
FIGURE 5. A typical flare burner tip
helps to dissipate hydrocarbon gases
via combustion, and can be located at
ground level or on an elevated structure
carbon fluids before they are dis-
charged to the atmosphere. The
burner fires horizontally into a pit
at ground level.
Flare pilots and ignition. In a
flare system, the flare burner is
equipped with pilot burners, which
are used to ignite the flared gases
when needed. The number of pilot
burners is dependent on the size
of the flare. The pilot burners are
installed at the flare tip. The pilot
burners have their own ignition
system. Two main types of igni-
tion systems are used to ignite the
flare pilots: flame-front genera-
tors (FFG), and high-energy igni-
tion systems. Some flares use both
systems as primary and backup
ignition sources. If desired, flame
monitoring can be installed in
the flare pilot.
Burner inspection
Burners are expected to be continu-
ously reliable and efficient enough
to meet the rigorous requirements
of the CPI. Periodic inspection and
maintenance should be conducted
according to the burner manufac-
turers guidelines. Malfunctioning
burners can result in inefficient
combustion or poor flame patterns.
This can lead to localized over-
heating and damage to furnaces or
equipment components, resulting
in increased maintenance costs.
This damage can also cause pre-
mature shutdowns due to failure
of critical equipment components,
such as tubes or refractory lining.
The following is a checklist of the
main burner components that oper-
ations and maintenance personnel
should inspect when developing an
inspection worksheet:
The atomizer of an oil burner
should be inspected for any de-
fects or misalignment
The gas nozzles of a gas burner
should be inspected for any de-
fect or deterioration
The nozzle diameter of an at-
omizer or gas nozzle should
be inspected to ensure that it
is within the burner vendors
specification datasheet
The movement of the mechanical
air registers should be inspected
to ensure proper free movement
The gasket at the atomizer of an
oil burner must be inspected
The burner throat should be in-
spected by measuring its dimen-
sion and ensuring it matches the
vendors drawings
The burner throat must be in-
spected to detect any damage
In addition to inspecting the vari-
ous burner components, there are
several physical dimensions that
must be confirmed in order to en-
sure correct operations. The main
components and allowable devia-
tions from vendors datasheets that
must be measured and verified in
the burners are as follows:
The burner tile diameter must be
within 1/8 in.
The burner tile concentric-
ity (roundness) must be within
1/8 in.
The tip port angles must be
4 deg
The bolting dimensions must be
1/8 in.
The gas tip locations should be
no more than 1/8 in. deviation
in the horizontal direction, and
no more than 1/4 in. deviation
in the vertical direction
Piping and auxiliaries
The integrity of the piping and aux-
iliaries of the burners also requires
attention, and there are certain
tasks that must be performed to
maintain normal operations. Some
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
best practices for maintaining
burner piping are as follows:
Perform leakage tests on the
burners associated piping and
valves by conducting a soap
test or three-in-one gas tester
procedure
Check the pilot pressure regulator
setting to ensure that it is within
the recommended setpoint
Verify the tightness of scanners
and conduit connections to pre-
vent water intrusion
Inspect electrical compo-
nents to ensure that they are
working properly, including
switches, lights, relays and other
related parts
Check that the pilot gas jet is
clear and that the pilot flame
is strong
Most burners are fitted with opti-
cal main flame-detection systems,
which also require maintenance.
The following must be checked to
ensure best performance of flame-
detection systems:
The sight path from the view-
ing head to the flame is aligned
correctly to provide visibility of
the flame
The head mount and sighting
tube are clear from any fouling
The wiring is installed as speci-
fied by the original operation
and maintenance manual
Burner maintenance
A burners performance deteriorates
with operating time due to foul-
ing, plugging and wear on burner
components. Fouling, plugging and
wear reduce the effectiveness of
fuel-air mixing and can affect the
flame and heat flux patterns, result-
ing in lower heater efficiency and
heating capability. Burner parts
requiring frequent maintenance
to avoid serious performance loss
or safety issues include the orifice,
tip, atomizer, tile, flame stabilizer,
register, damper, pilot burner and
detection systems.
Gas tip and orifice. In a burner,
the fuel-gas tips and fuel-gas orifice
have drilled ports that direct the
stream of fuel into the airstream
and combustion zone. These ports
must be kept free of foreign mate-
43
 Cover Story
rial that could decrease the effec-
tive port size. If the ports become
partially or completely plugged,
the quantity and distribution of
fuel entering the burning zone may
vary from the design objective, and
combustion problems may occur.
The type of material plugging the
ports determines how it should be
cleaned. Foreign material must be
carefully removed from the fuel
orifices so as not to affect the ori-
fice dimensions. Fuel tips should
be replaced if any orifices exceed
the specified diameter by more
than one to two twist-drill sizes.
If the material is not easily re-
moved, some cleaning guidelines
should be followed:
Soak the burner part in a solvent
to loosen the deposit
Use a twist drill to remove the
deposit. Never use a power tool
with the twist drill because it is
likely to enlarge the ports
Use a welders file to gently re-
move the scale from the orifices.
If the scale cannot be removed,
the fuel tip or orifice must
be replaced
Use wet steam or hot water if the
source of the foreign material is
an amine compound, because
amines are water-soluble
Shut down the fuel gas valve
and inject steam into the burner
if the amine plugging occurs fre-
quently without removing the
burner. Note that injection clean-
ing may not clear all tip ports
equally
Soak the tips in a hydrocarbon
solvent, followed by cleaning
with a twist drill if the tips are
plugged with polymers
Remove the coking, polymer or
solid deposit by oxidation in a
small, high-temperature furnace.
Many tips can be cleaned at the
same time with this technique
Oil tip and atomizer. Fuel-oil
tips are harder to maintain than
gas tips and require more frequent
cleaning. Tips in light fuel-oil ser-
vice may have only a slight carbon-
like deposit on the surface. To clean
fuel-oil tips, do the following:
Use a wire brush to remove
slight carbon-like deposits. Tips
44 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
in heavy fuel-oil service will usu-
ally have stubborn hard deposits
on the surface and in the ports
Use steam cleaning or a twist
drill, in combination or individu-
ally, to remove harder deposits.
It is not recommended to use a
power drill or tool to clean ports
or oil tip surfaces, as any nick or
notch on the oil tips surface acts
as a site to collect oil and acceler-
ate tip-coking problems
Soak the tip in a solvent, fol-
lowed by blowing steam through
the orifice to remove the remain-
ing oil or solid deposits
Handle the atomizer carefully
when removed from the oil gun
to protect the labyrinth seal.
This seal separates the oil from
the steam by a series of rings on
the atomizer with a tight toler-
ance between the atomizer and
the gun body. If the atomizer is
continuously getting stuck and
cannot be removed from the gun
without breaking the atomizer,
the metallurgy may be at fault.
In this case, it is recommended
to exchange the atomizer metal-
lurgy with a different metallurgy
than that of the oil tip
Burner tiles. The unique profile of
a burner tiles surface is challeng-
ing to reproduce, making tiles diffi-
cult to fix and return to service suc-
cessfully. Also, the refractory of the
tile undergoes phase transforma-
tion while in service, and refractory
repairs usually do not adhere to the
surface for very long periods of time.
Burner tiles should be replaced if
they experience any cracks or de-
formations. Especially of concern
are major cracks, mainly in wall- or
roof-mounted burner tiles, where
the tile pieces are expected to fall.
Also, one should look carefully for
multiple cracks in a section, which
present evidence of crushing due to
restrained expansion, and are rea-
son for replacement. The primary
or oil tile shall be checked as well.
If it is badly pitted or cracked, the
recirculation of gases within the
tile is uneven, and coking can occur
on the tile. This coking can lead to
oil dripping and spilling from the
burner. If this is observed, the tile
must be removed and cleaned.
Flame stabilizer. Different
flame-stabilizer designs are used
to maintain a stable flame in the
combustion zone. For natural-draft
burners, these designs take into ac-
count the diffuse cones, ledges and
tapered tiles of the burner. Poor
conditions or the absence of a sta-
bilizer can result in unsatisfactory
flame shape or an unsafe flame that
can lift off the burner and leave the
zone where combustion is initiated.
Operations suffer when part of the
flame lifts off the burner due to a
damaged stabilizer because irregu-
lar, erratic flame patterns occur,
and new risks are introduced into
the process. A damaged stabilizer
must be replaced.
Air registers and dampers. All
registers and dampers should be
exercised periodically to ensure
that they are functional. If damp-
ers are inoperable, check the actua-
tor and repair if necessary. Damp-
ers are often inaccessible during
operation and repairs must wait
until the equipment has been shut
down. Clean the register of foreign
materials, spilled oil, sand, rust and
insulation that blocks airflow. If the
register is deformed, possibly due
to incorrect operating procedures,
or the operating handle is broken,
repair or replace the part.
Pilot burners. The most common
pilot burners used in the CPI are
small, heat-release pre-mix burn-
ers. The basic parts of these pilot
burners include the pilot tip, gas
mixer, mixing tube and orifice. The
gas orifice is commonly 1/16 in. in
diameter and can be easily plugged
by pipe scale. This element is
cleaned by hand with a twist drill,
as previously described. If inspec-
tion reveals a damaged pilot tip,
the tip should be replaced.
Flare pilot burners. When the
flare system is in service, monitor
the supply pressure and clean the
fuel strainers, knockout pots and
drains that are accessible from
grade. When the flare system is
not in service, it is recommended to
clean the strainer upstream of the
orifice and to inspect the orifice.
Flame-front generators require
 Cover Story

maintenance for valves (cleaning),

gages (recalibration) and the spark
generator (adjusting the spark gap).
These ignition systems should be
regularly tested to ensure correct
operation for unplanned startup.
High-energy ignition systems
should be cleaned, inspected and FROM FEEDSTOCK
replaced, if damaged. Ignition sys- TO END PRODUCT
tems are a critical safety control
system, so periodic functional tests You wont nd a more comprehensive
of these components is highly rec- offering of separations and phase
ommended, as well as training contacting process internals than
for the operators. AMACS. Whether its demothballing,
Pilot burners with flame detec-
debottlenecking or grassroots,
tion. In addition to checking the
gas jet above the burner, check that AMACS can provide the components,
the ionization rod is set at the cor- hardware, know how and field
rect spark-gap distance for ioniza- support to streamline your process.
tion and ignition. Also ensure the From your process inlet to its outlet,
integrity of the ionization rods in- AMACS can provide the engineered
sulation by checking for earth leaks
solutions to meet your desired
between the rod and pilot body
casing by connecting a multimeter performance requirements or build
at the rod and to the pilot flame to meet your specications. In
retention head. short, if it is separations or phase
The inspection and maintenance contacting related contact AMACS!
best practices outlined in this ar-
ticle should provide engineers
with the know-how required to
ensure optimal longterm opera-
tion from all types of burners and
all of their components, in many
different applications in the CPI.

Edited by Mary Page Bailey

References
1. API Standard 535, Burners for Fired Heat-
ers in General Refinery Services
2. API Standard 537, Flare Details for General
RANDOM PACKING STRUCTURED PACKING LIQUID DISTRIBUTORS
Refinery and Petrochemical Service
3. NFPA 85, Boiler and Combustion Systems
Hazards Code

Author
Mohammed H. Al-Hajji is a
fired-equipment engineering
specialist at Saudi Aramco
(WC-1038C Al-Midra Build-
ing, Dhahran, Saudi Arabia;
Phone: +966-3-880-9591; Email:
hajjmh0a@aramco.com.sa)
with more than 25 years of
oil-and-gas processing experi-
ence. His areas of expertise
are boilers, heaters, heat-re- MESH MIST ELIMINATORS INLET DIFFUSERS TRAYS
covery steam generators
(HRSGs), sulfur plants, refractory and insula-
tion materials and water treatment. He is a
member of the Saudi Arabia Section of American
www.amacs.com
Institute of Chemical Engineers (SAS-AIChE),
Saudi Council of Engineers and Toastmasters AMACS 24-7 Emergency Service Available at (281) 716 - 1179
International. He earned a B.S.Ch.E. from the
University of Tulsa in 1989. 2013 Amacs Process Towers Internals. All Rights Reserved.

Circle 4 on p. 180 or go to adlinks.che.com/50982-04

CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014 45
Feature Report
Connecting Operations
Personnel to Reliability Efforts
Seven methodologies are described to help operations staff take greater ownership
of asset performance
T
he definition of ownership to act
as an owner implies certain re-
sponsibilities. Consider the range
of behaviors demonstrated by indi-
viduals when it comes to the automobiles
they own. For some, car ownership leads
to a relentless pursuit of caring for every
aspect of the car, from operation to main-
tenance. For others, it is a daily wish that
their cars will simply start when the ignition
is engaged. The outcome of the diverse be-
haviors along this continuum of ownership
will have a direct impact on the reliability,
longevity and cost of ownership of these
complex machines.
Throughout the chemical process indus-
tries (CPI), owners of complex, costly ma-
chines and systems (assets) tend to act along
this same continuum. In general, CPI opera-
tors typically want to ensure the delivery of
performance levels in terms of three impor-
tant measures reduced lifecycle costs,
improved reliability, and increased longevity
before replacement. However, different indi-
viduals will go about achieving these objec-
tives in different ways. Within any CPI facility,
the quest to ensure reliability is thought to
require three partners personnel from the
operations, maintenance, and engineering
Depts. All three sets of individuals play vital
roles in helping the asset to meet its impor-
tant objectives, via their interactions through-
out the lifecycle of each asset.
The efforts of the engineering department
should be building in reliability since the de-
sign itself has a greater impact on reliability
compared to the efforts by maintenance
and operations depts. combined. For many,
maintenance department efforts are thought
to be the primary element responsible for
the reliability of the installed assets. How-
ever, from our knowledge of various paths
for equipment failure (the majority of which
are random in nature), it turns out that op-
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
erations personnel hold the key to delivering David
optimal business objectives, through their ef- Rosenthal
forts related to the ongoing operation of the Reliability Strategy
assets. and Implementation
To further explain this important con- Consultancy, LLC
cept, we must first understand the bathtub
curves, developed by Nolan and Heap in
the 1960s and 1970s [1], which have driven
maintenance practices in the airline industry
for decades. The authors developed six fail- IN BRIEF
ure curves that demonstrate how the prob-
OPERATIONS
ability of failure is a function of run time for OWNERSHIP
machine components. The major finding
was that 89% of these failure modes occur THE PATH FORWARD
randomly often with little to no warning. DEVELOPING THE VALUE
At the time of these findings, industrys ap- PROPOSITION
proach to maintaining industrial and other
ESTABLISHING METRICS
assets had been to rely heavily on preven-
tive or time-based activities, such as planned CHANGING THE
overhauls. However, given that the majority EQUIPMENT-CONDITION
of failures occur randomly, it is not practi- MINDSET
cal to expect that a time-based approach to TRAINING AND
equipment maintenance will detect or iden- TROUBLESHOOTING
tify all potential failures. While online moni-
INTEGRATING
toring options can provide a close proxy for
MAINTENANCE WORK
realtime surveillance in some instances, we
PROCESSES
cannot place a mechanic at each machine
to constantly monitor its condition on a real- PERFORMING
time basis. The operations department is the EQUIPMENT ROUNDS
only group with enough continual exposure AUTONOMOUS
to the assets on the plant floor to be able to MAINTENANCE
detect the earliest signs of many impending
failures. So why do many CPI facilities still
experience relatively high levels of reactive or
breakdown-related maintenance, and fail to
effectively deploy their operational personnel
to provide close ongoing surveillance of the
assets in the field?
Operations ownership
In recent decades, there has been a transition
in operations department culture. Many retir-
ees lament the bygone era when operators
knew not only how to operate their equipment
MAY 2015 59
 but also understood how to maintain
the function of all of the assets under
their command. Historically, opera-
tors relied on using their four senses
(hearing, seeing, touching and smell-
ing) to keep close track of how the
equipment was operating during their
shift. They would adjust settings, add
oil or grease, unplug and monitor their
equipment and be able to detect small
changes in the equipment condition
and rapidly report their findings to the
maintenance department Some even
performed minor tasks to fix a prob-
lem early so as not to allow it to grow
into a major downtime event.
However, more recently, that cul-
ture has slowly been replaced with
a new mindset about the appropri-
ate division of labor that opera-
tions personnel run the equipment
and maintenance personnel fix the
equipment.
Nonetheless, today, due to the
complexities associated with main-
taining complicated equipment and
support systems, participation by
both of these functions is essential.
In some industries, operating per-
sonnel have become more comfort-
able with elaborate control systems
and control rooms, which took the
operators exposure away from the
equipment in the field. This drove an
additional wedge into the culture of
ownership needed to maintain equip-
ment reliability. Now the question is,
how do we return a state where op-
erations personnel are once again
empowered to be a critical partner
and allowed to take more of an own-
ership role in the quest to maintain
the asset base, as needed?
The path forward
Back in the 1950s and 1960s, man-
agement did not have to present much
of a business case for operations per-
sonnel to perform all of the tasks that
are required to maintain equipment.
Operator rounds and minor mainte-
nance were an accepted part of the
job description. Today, owners must
develop more of a business case to
justify the use of operations person-
nel for such tasks, allowing them to
act like the true owners of their equip-
ment assets.
The path forward involves selling
the benefits of more direct owner-
ship by operations personnel. These
60
TABLE 1. ALIGNMENT OF LEADING AND LAGGING METRICS AND MEETINGS
Frequency Metric Meeting
Daily Number of bad actor failure triggers Daily production meeting
(leading)
Number of critical-equipment failure
triggers (leading)
Number of on-condition exceptions
(leading)
Number of audit exceptions (leading)
Weekly Percent of preventive maintenance (PM) Maintenance-scheduling
work completed (lagging) meetings,
Percent of predictive maintenance route production meetings
completion (leading)
Pump repair-sheet compliance (leading)
Operator-driven reliability exceptions
no response
Number of breakdown work orders
analyzed (leading)
Percent of proactive work completed
(lagging)
Percent of proactive work scheduled
(leading)
Number of machines that missed on-
condition checks (leading)
Monthly or Quarterly Downtime (lagging) Reliability management
Pump MTBF (lagging) meeting;
Percent on-condition work orders created Annual site strategy
(leading) meeting
Excessive work orders (rotating/electrical
and Instrumentation) (lagging)
Mechanical Availability (lagging)
Pump failures (lagging)
Mechanical integrity inspections overdue
(lagging)
Care plans created (leading) and imple-
mented (lagging)
include a more predictable and safer attainable by operations personnel.
work environment for all personnel, Normalization of the abnormal
improved business targets for cost occurs when sub-optimal equip-
savings, higher overall productiv- ment conditions are tacitly accepted
ity (through reduced downtime and by those who operate the equip-
higher asset optimization), and the ment. Left uncorrected, these sub-
development of new skill sets for optimal conditions typically lead to
plant personnel. reactive maintenance cultures, since
The business impact of these the early signs of failure are not ac-
added operations department efforts knowledged and used to drive pro-
should be demonstrated, by tracking active repair. For instance, a valve
metrics that are related to key busi- that has always been hard to close
ness results. However, to be fair, the is often taken for granted, until one
tracking should involve only business day it does not close at all. Once the
results that operations personnel new normal state of equipment
could actually influence directly. Too conditions are established (the valve
often, management tries to translate is replaced), the early detection of
the impact of operations department potential failure modes can be rec-
efforts using metrics that may be ognized.
too strategic such as mechanical The importance of the need to rec-
availability and mean time between ognize early signs of failure should be
failures (MTBF) and thus cannot driven to the floor-level personnel, so
be impacted easily by operations they can quickly recognize failure and
personnel actions. To be success- request repairs when they have the
ful, this path forward also requires smallest impact on overall plant oper-
changing the mindset of company ation. This involves using periodic au-
and plant managers, to establish dits and basic troubleshooting tools,
metrics that really show the perfor- and providing accurate descriptions
mance of operations personnel in the of what equipment requires repair.
care of the equipment, and are thus Operations personnel should be en-
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015
couraged and allowed to return to
performing basic and simple repairs
(so-called autonomous maintenance).
This approach will allow maintenance
personnel to focus on performing
proactive, strategic care tasks that
are designed to move the facility away
from reliance on reactive maintenance.
In general, reactive maintenance
often engenders higher costs, more
downtime, and a workplace that is
less safe overall.
Detailed below are several meth-
odologies that can help drive a more-
effective partnership between own-
ers and the operations department
Each of these is discussed in greater
detail in the sections that follow: (1)
Developing the value proposition; (2)
Establishing metrics; (3) Changing
the equipment-condition mindset;
(4) Training and troubleshooting; (5)
Integrating maintenance work pro-
cesses; (6) Conducting operator
rounds; and (7) Increasing autono-
mous maintenance. No single meth-
odology will secure the role of the
operations personnel as owners of
the assets, but efforts to include op-
erations personnel more completely
in the care of the assets will eventu-
ally deliver the desired results.
1. Developing the value proposi-
tion. In general, a value proposition
is a business or marketing statement
that summarizes why an individual
consumer should buy a product or
use a service. This statement should
convince a potential consumer that
one particular product or service will
add more value or solve a problem
better, compared with similar offer-
ings. The ideal value-proposition
statement is short and appeals to the
customers strongest decision-mak-
ing drivers. It is important to make sure
operations personnel understand
all of the reasons why they should
take a more active role in equipment-
care activities.
Management should first focus on
selling operations personnel on the
potential benefits of joining their col-
leagues in maintenance and engi-
neering depts. in the pursuit of more
reliable operations. Historically, man-
agement has told operations person-
nel that their primary job was to en-
sure that the manufacturing process is
operated within the acceptable range
of key operating variables such as
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
TABLE 2. LIST OF CONDITIONS TO BE RECOGNIZED DURING A RELIABILITY WALKTHROUGH
PRIORITY ONE WORK ORDER
Leaks Repair obvious leaks from pumps, seals, valves and utilities
Address corrosion that indicates thinning and imminent failure
Environmental, health and Address any condition that presents a potential EHS hazard
safety (EHS) related
PRIORITY TWO WORK ORDER
Condition monitoring Any out-of-place heat, odor, noise and vibration
Be sure that oil levels and automatic greasers are readable and have an
appropriate date and level given their age in service
Lubricate any rotating equipment that shows signs of not being recently
oiled or greased
Field instrumentation and Clean or replace dirty non-readable field instruments
electrical conditions
Ensure that instrumentation is indicating process conditions as specified
(flows, transmitters and seal panels)
Address leaks and disconnected lines and wires
Ensure that cabinets and other housings should show adequate pressures
and flows to meet area electrical classifications
Repair electrical fittings, exposed wires, missing fittings and broken con-
duit, as needed
Housekeeping Remediate obvious housekeeping issues
PRIORITY THREE WORK ORDER
Visual factory Replace any missing or damaged oil tags, vessel signs, hazard diamonds
and reading indicators on critical gages
Insulation and scaffolding Repair or replace any insulation that is missing or loose
Address any scaffolding that is in need of repair or removal
Painting Repaint any color-coded pipe that has become aged, per plant standards
Labeling and signage Ensure that correct signage and line labeling is present per plant and unit
standards
PRIORITY FOUR WORK ORDER
Painting Refresh aged paint on vessels and structures, per plant standards
Demolition Remove unused, unrepairable or obsolete equipment
temperature, pressure and so on. achieve many safety, productivity,
Historically, operations personnel and cost goals that are established
have stressed that their first respon- by management. In many facilities,
sibility is to operate the process in a operators are hoping for a predict-
safe manner. They are also tasked able work shift, where the process
with data recording and meeting is running at steady state with little
responsibilities. One can argue that variation and upsets. Processes that
some of the data they are record- are not reliable tend to call for reac-
ing during operator rounds for tive maintenance, which contributes
instance, Is the pump running? to unsafe behaviors. Thus, a useful
are a form of management control value proposition for CPI operators
and may not even be reviewed by may be expressed as follows: For
supervisors. Some operators do not operators of CPI assets who want to
understand what the data tells them work in a safe and sustaining envi-
and they question why such data are ronment in order to provide for their
being recorded at all. Based on this families and loved ones, reliable op-
background, it is no wonder that for erations require a day-to-day focus
many operations personnel, equip- on reporting out-of-range condi-
ment monitoring often takes a back tions, recognizing the early signs of
seat to other responsibilities and equipment failure, troubleshooting
they may not understand its full value loss of function, recording required
in influencing asset reliability. data, and looking out for each
Appropriate monitoring of equip- others safety.
ment and providing basic care does Management needs to reinforce
improve the operating environ- that safety, sustainability and predict-
ment for operators, and will help to ability are the strongest drivers in all
MAY 2015 61
 CPI operations. These words and
concepts should be constantly
reinforced at all levels of supervision,
and should appear on information
boards throughout the facility and be
stressed at daily meetings. A small
portion of the operations group will
already understand the message,
while another portion of the Opera-
tions group will require some evi-
dence to get them involved.
2. Establishing metrics. The adage
What gets measured, gets im-
proved is heard throughout the
business landscape. However, met-
rics can be a double-edged sword
and sometimes individuals and
groups can become bogged down
by paralysis by analysis when ex-
cessive metrics are tracked, but
for no clear purpose. Nonetheless,
tracking of appropriate metrics can
drive behaviors. People adjust their
behaviors based on what aspects
of their performance are being mea-
sured. For instance, if plant person-
nel are evaluated for process uptime
alone, they may not make the best
decisions about how they operate
the equipment.
Tracking the right combination of
metrics can propel an organization
toward desired targets for improve-
ment, while focusing on the wrong
mix can steer people toward contra-
dictory actions and may lead to more
inefficiency in terms of wasted costs
or time. The right mix of metrics in-
cludes both leading metrics, which
measure process activity such as the
amount of practive work scheduled,
and lagging metrics, such as main-
tenance-schedule compliance and
mechanical available, which mea-
sure an outcome.
In general, lagging metrics are
more strategic, and thus manage-
ment tends to put disproportionate
emphasis on them. However, opera-
tors are often not able to meaningfully
impact these metrics. For instance,
in the case of reliability, focusing on
MTBF with maintenance and op-
erations personnel generally draws
blank stares. However, directing their
attention to leading metrics, such as
percent of work orders with work his-
tory, percent of scheduled lubrication
routes completed, and percent of
exception found on equipment mon-
itoring routes, allows them to move
62
TABLE 3. SAMPLE WORKSHEET TO PRODUCE A "FIVE WHY" REPORT
Equipment location
Equipment description
Failed component
Date of event and time
Name (Leading the discussion)
Names (Participating)
1. Failure description What happened?
2. Why did it happen?
3. Why did it happen?
4. Why did it happen?
5. Why did it happen?
Root-cause statement: The impeller failed from cavitation that occurred as a result of changing flow
conditions.
Major corrective action: How can this be prevented? Where else does this apply? Ensure that valves up-
stream of the pump are tested for proper operation.
Actions and responsibilities: Maintenance will change the valve and replace the pump impeller.
Action: Who:
Action Who
the needle on plant operations that
will eventually impact MTBF. Table 1
illustrates this concept further.
Table 1 also shows the timing of
reporting leading and lagging met-
rics. Leading metrics should be
discussed daily to weekly, whereas
lagging metrics should be discussed
weekly, monthly and quarterly, be-
cause the ability to change lagging
metrics generally takes more time.
Operations and maintenance per-
sonnel can become frustrated when
seeing little movement in lagging
metrics, when their focus should re-
ally be on moving the needle with
those metrics they can impact di-
rectly over shorter time horizons.
3. Changing the equipment-
condition mindset. As noted, nor-
malization of the abnormal is the
enemy of reliable operations over the
long run. The acceptance of sub-
optimal existing conditions, such as
loose fittings, small leaks, tough-to-
close valves, and many others rep-
resent the waiting room for failure.
Unfortunately, these conditions be-
come part of the landscape in many
manufacturing facilities, and with the
existence of higher-priority reactive
work, they often never get fixed.
Operations personnel are exposed
to these conditions on every shift.
They often bring attention to these
issues but get little response. When
this pattern persists at a facility, it is
difficult to recruit operators as part-
ners in the pursuit of improved reli-
CHEMICAL ENGINEERING
First floor
Process pump
Impeller and valve
2/4/15 second shift
T. Jones
Various
The pump was making a loud noise.
Cavitation is occurring.
Flow conditions changed.
The inlet to the pump was plugged.
The inlet valve upstream failed.
Date:
Date
ability, because they can point to
many examples that indicate that
management is not willing to fix
items they report. The path forward
is for management to demonstrate
its commitment to remediating these
early signs of failure, as a proven way
to forestall larger problems later.
One useful method to deploy is
the reliability walkthrough. Just as
many manufacturing operations per-
form safety walkthroughs of their
units, another set of audits should
be performed to monitor equipment
condition. Plenty of preparatory work
is required before starting. The first
step is to gain buy-in with produc-
tion management to perform these
audits. Such buy-in can be gained
by showing existing field evidence
of conditions that require repairs,
such as missing conduit covers, bad
valves and missing oil containers.
Recognizing a prevailing lack of at-
tention toward equipment is vital to
encourage a change in attitude. Le-
veraging managements commitment
to improving reliability is another.
Once buy-in is achieved, the main-
tenance and reliability departments
need to set up a standard for what
conditions are considered abnormal,
and make it a priority to fix these
conditions in accordance with the
existing maintenance execution pro-
cess (Table 2).
Next, establish a schedule for
these audits. Participation should in-
clude Production management, op-
WWW.CHEMENGONLINE.COM MAY 2015
erators, maintenance or reliability en-
gineers and maintenance personnel.
One attendee is assigned to be the
scribe to record what is found and
set the priorities for the work.
After the audit, the list is converted
to work-order requests. These au-
dits should be more than just a
fix-it tour. They should represent a
culture-changing event so that over
several months of audits, operations
personnel will begin to better under-
stand what represents acceptable
equipment conditions. They begin to
see managements commitment to
improved reliability and safety. Even-
tually (after a year or so), a variety of
metrics are used to show progress.
Relevant metrics include percent
reactive work, percent mechanical
availability, percent process uptime
and MTBF. Consistent effort will gen-
erally show improvement across all
classes of assets, especially when
participants demonstrate patience,
commitment and consistency. Con-
sistency is important plant per-
sonnel must avoid the temptation
to postpone audits due to other de-
mands or priorities, attendance is-
sues, weather, and downtime.
4. Training and troubleshooting.
Operator training is often restricted
to safety and process-operations-
related areas. For operators to prog-
ress to increased levels of responsi-
bility, they generally focus on improv-
ing their breadth of unit knowledge,
achieving better process under-
standing and developing increased
analytical capabilities. However, often
left out of the training is attaining im-
proved understanding of equipment
operation. Also, understanding the
principles related to pressure, tem-
perature and flow measurement may
not be part of their training matrix.
And yet insufficient training in these
topics can lead to failure in CPI op-
erations. The majority of failures in
the manufacturing environment re-
sult from how the equipment is op-
erated. Examples include improper
pump operation, running equipment
outside of design limits, improper
setup, lack of lubrication and missing
needed adjustments, to name a few.
A lack of understanding of machine
operation is another hurdle facing op-
erations personnel. For operators to
partner effectively with maintenance
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
TABLE 4: OPERATOR ROUND CHECKLIST
Focus on quality, cost, safety, environmental and productivity data
Ensure the data you collect or the tasks you ask your operators to perform add true value to your busi-
ness focus on critical assets
Minimize travel and enable remote monitoring
Include expected ranges and support with visual techniques
Provide feedback systems so personnel know the work they are performing means something
Set up clear responsibility and accountability
Enable data gathering with simple tools
Train people to do inspections and gather data properly
Include troubleshooting steps if variations are found
Manually record data first, automate the recording of data second
Convert manual recording to the use of electronic handheld devices (10% productivity improvement)
and engineering personnel, the first processes. A partnership is built on
steps are to determine the gaps in eliminating boundaries. Many manu-
their understanding of equipment facturing locations restrict operations
operation, and related principles of department access to their comput-
pressure, temperature and flow. Any erized maintenance management
identified gaps should be included in system (CMMS). This barrier pre-
the training matrix that is required for vents operations personnel from ex-
operator responsibility progression. ecuting their role in equipment care,
Trainers may come from in-house potentially creating restrictions in
engineers, maintenance personnel, submitting work requests. At some
training professionals, third-party locations, operations personnel
vendors and even local colleges. must contact a maintenance rep-
During training, the topic of trouble- resentative or a supervisor to sub-
shooting deserves special attention. mit work requests. This added step
The aim is to drive troubleshooting to can restrict which needed work
the floor level, so that problems can is performed.
be solved quickly and avoid involve- In general, operations personnel
ment from the maintenance depart- should be given limited access to
ment Such an approach benefits the CMMS. For example, opera-
both maintenance and operations tors should be able to submit work
efforts. Operators should be required requests at any time. The mainte-
to perform basic troubleshooting nance department gatekeeper on
from the first signs of variance from the CMMS system will ultimately
normal operations. One easy tool to decide the priority of all submitted
use is a Five Why structure (Table work requests. The site can set up a
3). It requires the participant to ques- few logon identifications and termi-
tion each observed result by asking nals for access and provide opera-
Why? five times to drill down on tions personnel with access to view
the events that occurred, in order to work orders and their status.
identify a root cause. The quickest way to frustrate any
Although the Five Why approach initiative by operations personnel who
is limited in its application, it does are willing to participate in the care of
apply to many situations faced by the equipment is to not provide feed-
operations personnel. Oftentimes, back to the suggestions they make.
operations personnel can resolve the Giving operations personnel access
issue themselves before calling their to the CMMS will allow them to view
colleagues in maintenance. Even if and track the status of audit findings
they cannot resolve the issue, the and other submitted work orders and
information derived from the initial suggestions they may submit.
investigation and troubleshooting ef- The maintenance department also
forts will improve the content of work has a role to play in this partnership
order requests, which will help main- when it comes to work-order execu-
tenance personnel to be more effi- tion. For instance, work-order sched-
cient. Operations personnel should ules should be distributed to opera-
also be included as part of more tions personnel so they can prepare
formal root-cause investigations, as for the work to be performed. Priori-
a participant, so they can contribute ties should reflect the current state of
needed information. the operation department priorities.
5. Integrating maintenance work After responding to a work order,
MAY 2015 63
 maintenance personnel should seek
out the specific operations personnel
with the work-order response to ac-
knowledge their submission and en-
sure their satisfaction with the work
and the condition of the area after
the work was performed.
6. Performing equipment rounds.
A widely held belief in industry is
hat Operators can fail the best de-
signed equipment, but they can also
run marginal equipment.
Some equipment-round sheets are
no more than checklists and contain
no standards to help personnel rec-
ognize abnormal operating condi-
tions. Data are typically gathered by
operators with little understanding of
why it is important to the operations
of the process. Periodic equipment
rounds in some cases represent a
form of control designed by super-
vision to keep people moving. Given
this landscape, operators tend to
pencil-whip (give a cursory effort
but not complete) rounds as they
may not understand their purpose.
To make equipment rounds most
effective, operations personnel
should always start with a focus on
the critical assets (at least at the be-
ginning), rather than all assets, to
make best use of the time.
Useful data to be gathered in the
field should indicate the health of
the assets. The use of visual tech-
niques will allow an individual to
understand quickly if many types of
equipment are operating normally.

ONE WORLD.
For example, note the expected
range on a pressure gage and
mark on the operator-rounds sheet

ONE WATER.
whether that gage is operating within
the target range. Ensure that some-
one is reviewing the operator-rounds
sheets (or their equivalent in an elec-

ONE EVENT. tronic database) and that feedback is

given when variances are observed.
Table 4 provides a brief checklist that
can be used during operator rounds
so their efforts align with the recom-
mendations discussed above.
7. Autonomous maintenance.
The early detection of failures ben-
efits chemical process operations

COUNTLESS
through greater uptime, reduced
maintenance costs and a safer
working environment. As operators
are closest to the daily operation of

OPPORTUNITIES. mechanical assets in a CPI facility,

increased operator awareness and
involvement in all reliability efforts is
a key enabler to this early defense
WEFTEC 2015 is the event for water professionals, industry experts, and the
most innovative companies from around the world to gather together for the warning system for impending func-
advancement of water. tional failure. Constant monitoring
by these strategic personnel pro-
STAY DISCOVER ACCESS GLOBAL MAKE vides an opportunity to correct a
COMPETITIVE THE NEWEST BUSINESS VALUABLE variance before it has a chance to
AND RELEVANT IN INNOVATION OPPORTUNITIES CONNECTIONS affect overall plant operations.
YOUR CAREER AND SOLUTIONS WEFTEC is the only WEFTEC hosts Even more advantageous to the
WEFTEC offers the WEFTEC features water show selected more than 22,000 facility is when relatively easy repairs
highest-quality, most the largest water to be a part of the attendees from can be made on the spot (involving
comprehensive quality exhibition in U.S. Commercial around the world such rudimentary tasks as tightening
educational sessions the world with more Service International and all sectors of flanges, replacing packing and so
available today. than 1,000 exhibiting Buyer Program. water quality. on), rather than requesting mainte-
companies.
nance department involvement and
then awaiting their arrival. The time
lost waiting for repairs can increase
the cost of the repair and take main-
88th Annual Water Environment Federation
Technical Exhibition and Conference
tenance personnel away from other
September 26 30, 2015
Register Today! proactive tasks that are needed to
www.WEFTEC.org provide asset care.
McCormick Place, Chicago, Illinois USA
Encouraging autonomous mainte-
nance activities for certain tasks by
Circle 60 on p. 102 or go to adlinks.chemengonline.com/56197-60
64 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MAY 2015
 TABLE 5. OPERATOR-DRIVEN RELIABILITY ACTIVITIES
Level one: Standard operations Level two: Operator-involved main-
Non-contact, tenance Inspection and monitor-
troubleshooting ing (Non-contact)
Startup, operations, change-
over, adjustment, shutdown
Use human senses
Address source of leaks, dirt,
grease other problems
Raise awareness of bad
actors
Change procedures and
carry out training
Carry out routine house-
keeping and cleaning
Change filters
Clear plugged pump suction
operators offers one solution. Unfor-
tunately, industrys history with uti-
lizing operators for these tasks has
not always yielded a success story.
Site-specific work rules, lack of train-
ing, lack of a value proposition, and
perhaps exaggerated concerns for
safety tend to keep operators from
being allowed to handle these impor-
tant tasks.
However, with strong manage-
ment support, these obstacles can
be and have been overcome
at many CPI locations. First, a policy
is needed to indicate which elements
of corrective (autonomous), preven-
tative and predictive care can be
performed by operators, and then to
establish buy-in among all affected
parties. Maintenance personnel can
be assured that operators are not
trying to replace them. Staff support
is needed to convert the appropri-
ate preventive maintenance tasks to
condition-based tasks for operators
to perform. In fact, roughly 30% of
preventive maintenance tasks can
be done by operators. Table 5 sum-
marizes the important steps that are
needed to establish a multistep ap-
proach one that recognizes three
levels of possible operator-driven re-
liability activities.
The selection of which of these
levels to perform will be a function
of site needs, culture and opera-
tor skill level. As indicated in Table
5, operators in level one perform
normal operating tasks along with
non-contact tasks involving the four
senses (hearing, seeing, touching
and smelling). Of course touching
can be done with limitations. In ad-
dition, the Level one group includes
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
Level three: Operator-
performed maintenance
Autonomous maintenance
Walkarounds Tighten bolts, adjust
PM inspections (including check- belts and tighten valve
ing oil and lube levels) packings
Manage lubrication closely Change light bulbs and
Low-tech condition monitoring other indicator lights
(using strobe light, stethoscope, Install flange blinds
vibration pens, IR temperature) Assist in maintenance
Monitor critical process variables Adjust and calibrate
Bearings and motor temperature Replace filters and other
Inspect couplings consumables
Inspect belts and sheaves Fix leaks
Inspect chains and sprockets Perform failure analysis
Check ultrasonic leaks
executing simple tasks, such as
setup, cleanup, adjustment, align-
ment and checking that are required
to ensure proper asset operation.
Level two tasks attempt to use
operators for some condition-based
tasks, including the use of some non-
contact tools to diagnose asset con-
dition. Lubrication is included in this
level and will require the setup of a lu-
brication program (consisting of mini-
mal selection of lubricants, establish-
ing color-coded lubrication locations,
establishing lubrication storage, and
developing checklists that direct ap-
propriate lubrication protocols). Per-
formance of level two tasks does not
require mechanical skills, perhaps just
some rudimentary training.
Level three moves closer to the
definition of autonomous mainte-
nance, with operations personnel
carrying out some basic care tasks,
using a few select tools. What is im-
portant in this level is the inclusion of
the expectation that the operations
personnel will assist with the trou-
bleshooting of equipment failures.
Maintenance personnel will even-
tually come to view the assistance
from operations personnel as a ben-
efit to the overall mission of the facil-
ity; which is to maintain and restore
safe, reliable function. In facilities
that have adopted this approach,
the maintenance department often
remarks that consistency in the way
in which the equipment is operated,
monitored and maintained provides
great benefits.
Closing thoughts
Success in achieving site reliabil-
ity is based upon the triad that is
MAY 2015
formed by a solid partnership be-
tween operations, maintenance,
and engineering personnel, all
performing their respective roles
in applying the right practices
throughout the life of all plant as-
sets. No successful reliability effort
can be accomplished without all
three groups pulling their weight.
Unfortunately, cultural norms within
CPI operations often do not allow
all parties to participate fully.
In particular, operations person-
nel often consider that their role is to
run the equipment while mainte-
nance personnel repair it upon failure.
Nonetheless, operations personnel
have the most direct and consistent
exposure to the equipment over time,
which gives them invaluable access to
detect the early signs of potential fail-
ure. Thus, they can play a critical role
in reducing machine failures. Each of
the seven methodologies discussed
here provides essential elements to
help drive an operational excellence
program, and promote greater coop-
eration among the three groups in the
triad, in order to maximize safety and
process uptimes while lowering oper-
ating costs. n
Edited by Suzanne Shelley
References
1. Nowlan, F. Stanley, and Heap, Howard F., Reliability-
Centered Maintenance, National Technical Information
Service, Report No. AD/A066-579, December 29,
1978.
Author
David Rosenthal is a reliabil-
ity consultant with more than 35
years of experience, and owner of
Reliability Strategy and Implemen-
tation Consultancy, LLC (2914
Ocean Mist Ct., Seabrook, TX,
77586, Phone: 215-620-2185;
Web: www.reliabilitywithoutfailure.
com; Email: davida.rosenthal@
prodigy.net). He provides a wide
range of maintenance and reliability consulting services,
aimed at designing and implementing asset-care strate-
gies to improve uptime and reduce operating costs. He
previously led asset management services for Jacobs
Engineering Group (Houston). Rosenthal spent the ma-
jority of his career with the Rohm and Haas Co., a spe-
cialty chemical manufacturer. During his 29-year career
with Rohm and Haas, Rosenthal held roles related to
maintenance leader, reliability leader, process and proj-
ect engineering, and technical management in various
facilities. In 2012, Rosenthal served as president of
AIChE. He currently leads the Advisory Board to the Soc.
of Maintenance and reliability professionals. Rosenthal
graduated from Drexel University with a B.S.Ch.E., and
holds an M.S.Ch.E. from the University of Texas. He is a
registered professional engineer in Pennsylvania, and a
certified maintenance and reliability professional
(CMRP).
65
 Feature
Cover Story
Report

Improve Rotary Equipment

Reliability with Checklists
Design selection and
commissioning of rotary
equipment can benefit by
following a structured,
checklist-based method
and by promoting end-
user involvement
Sourav Kumar Chatterjee
Hindustan Petroleum Corp.

R
otary equipment, such as cen-
trifugal pumps, often represent
the heart of plants in the chemi-
cal process industries (CPI). A
critical factor in the longterm reliabil-
ity of such equipment is a high-qual-
ity design and commissioning process,
so that initially, the rotary equip-
ment will have the maximum built-in
strength to survive upset operating
conditions, yet not be overdesigned
in a way that leads to higher overall
lifecycle costs. To sucessfully design
and commission plant assets, two
critical objectives come into play: es-
tablishing a culture of cohesive team-
work among the consultant, project
and end-user teams; and following a
structured procedure with the help of
equipment-specific checklists.
The teamwork component of this
approach is intended to systematically
capture on-the-ground realities of
the plant and to realize longterm ben-
efits by incorporating improvements
into the design, construction and com-
missioning processes. To do this, the
active participation of the end-users
is essential throughout the process
(Figure 1). The second component,
following a structured set of detailed
checklists, is intended to optimize the
equipment design and setup at the
outset, to allow maximum reliability.
52 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013
FIGURE 1. Nurturing a culture of cohesive teamwork among consultants, project
teams and end-users is difficult, but can generate benefits for equipment reliability
End-user interactions
The end-users of the plant asset can
provide a great deal of data to the en-
gineering and construction team, and
that information should be captured
in the right perspective, and then for-
malized and structured in a way that
promotes understanding. If that is
achieved, the result is a high degree
of practicality and flexibility in the
way the information is implemented.
During the data collection process be-
tween the engineering and end-user
groups, there is potential for a great
deal of irrelevant and impractical in-
formation to be exchanged. Hence, the
organization and use of the data must
be carried out by a team of subject-
matter experts with visionary wis-
dom, so that the basis of action items
is justified from an economic, as well
as a reliability, point of view.
The mode of interaction among the
end-users and project-management
group for the information transfer
described above is of immense impor-
tance, because the two groups have
different core objectives to achieve.
It is often a difficult task to align the
mindset of both groups and to bring
them under a common umbrella of un-
derstanding, so that the exchange of
knowledge becomes collaborative, fo-
cused and effective, resulting in real-
ization of an integrated common goal.
In other words, a mutually agreed-
upon structured methodology must
be in place for communication and
knowledge management that takes
into account the profile of participat-
ing teams, the nature of the informa-
tion available, and the areas where
that information will be applied.
When this structured methodology is

established, the project workflow will
remain on a progressive track.
This process of value addition by
effectively merging end-users experi-
ence from the design selection stage to
commissioning of the asset is unique
for each project and has to be engi-
neered to allow for modifications to
accommodate the systems and proce-
dures for different organizations. If
the goal of smooth knowledge trans-
fer is achieved, a safe, on-time, cost-
effective execution of the design and
installation process will translate into
a sustained reliable and productive
operation thereafter.
Defining the five Gs
While several methodologies can be
derived for the interaction of end-users
and project management consultants
(PMCs), it is extremely important to
convert the information into execut-
able tasks that keep the inherent val-
ues intact. The most popular and ef-
fective roadmap for structuring these
interactions is to develop checklists
and procedures for the design selec-
tion, pre-commissioning and commis-
sioning activities and to assign respon-
sibility to the teams in the loop. Such
checklists play vital roles in ensuring
that proactive measures are taken for
all asset-reliability issues, operability
issues and maintainability issues. To
the extent that those measures can
be taken, a more trouble-free commis-
sioning is possible, and an enhanced
effective service life of the equipment
will follow. The process flow can be ex-
pressed as the five Gs:
1) Get. Get end-users expertise and
participation
2) Gather. Gather information and
data on reliability, operability,
maintainability and safety issues.
Develop procedures and a check-
list system to incorporate proac-
tive mitigating measures in design,
erection and commissioning of the
new asset or facility, in consulta-
tion with the PMC, licenser or OEM
(original equipment manufacturer)
3) Grow. Grow the new facility by com-
mencing the design, erection and
pre-commissioning in line with the
developed system and procedures
4) Gain. Gain an on-time, safe and
smooth commissioning by closely
TABLE 1. GENERAL MAINTENANCE-REDUCTION CHECKLIST FOR ACHIEVING
BUILT-IN RELIABILITY FOR CENTRIFUGAL PUMPS
Description of check point Adequate Remark
no. Yes No
1 Design provides the suitable materials of construction of major
components, surface sealing materials and hardware to with-
stand the most adverse operating conditions possible in the sys-
tem, including all mechanical and hydraulic forces acting upon it
2 Design provides inherent protection to components from corro-
sive or erosive environments to which it is, or can be, exposed as
a part of the operating process
3 Design provides a correct and realistic input process condition
for the equipment
Design provides a correct and realistic equipment output perfor-
mance within the limit of specified input condition
4 Design provides the most suitable hydraulic balancing for equip-
ment, so as to minimize the forces acting on components
5 Design provides a specified trouble-free operating zone without
affecting the desired performance and component life
6 Design provides correct and robust shaft-sealing system that
can withstand the most adverse operating conditions possible in
the system without failure
7 Design provides an appropriate bearing-support system to
provide maximum stiffness to the rotating mass at minimum fric-
tional loss
8 Design provides proper lubrication system that is most suitable
to protect the bearings at minimum friction loss
9 Design provides the most suitable lubricant for the bearing sys-
tem, considering the load and service conditions
10 Design provides for bearings and seals with wear- or failure-
monitoring capability to permit scheduling of maintenance prior
to actual component failure or component damage
11 Design provides suitable casing design pertinent to service con-
ditions to provide highest built-in reliability
12 Design provides that all running clearances are minimized with
applicable tolerances for thermal expansion
13 Design provides monitoring facilities for crucial performance
and physical dynamic parameters of the equipment, so as to as-
sess the health of the machine
14 Design provides suitable bearing-housing seals to avoid ingress
of contamination and leaking of the lubricating oil
15 Design provides the base frame that enforces adequate rigidity
to dampen the vibration generated in the system and prevent
the transmitted vibration from other sources to enter
16 Design provides an appropriate piping support system that pre-
vents piping stress to the equipment and also to eliminate the
potential of generating unsymmetrical dynamic forces in the
system during operation
17 Design provides suction, discharge and auxiliary piping system
that will not cause excitation of flow-related problems and sizing
shall not be over-designed or under-designed in relation to the
maximum rated flow, NPSH (net positive suction head) required
18 Design provides a coupling assembly between the driver and
driven system that should be rated for a minimum of 1.5 times
the service factor and above the maximum starting torque re-
quired, as per the speedtorque curve of the machine
19 For single-stage rotor, the entire assembled rotor, along with cou-
pling hub mounted, should be balanced in accordance with API
684 or ISO 1940 grade 2.5
20 For multistage rotors, individual impellers should be balanced be-
fore assembly. The assembled rotor should be balanced thereafter
21 Design provides the driver rating for end-of-the-curve operation
22 Design provides adequate external protection from weather and
harmful interference from other equipment and systems
23 Design provides a firm identification tag posted at a suitable
location close to the equipment that mentions rated operating
parameters, allowable physical operating parameters and alarm
limits of critical performance parameters
24 Design provides applicable interlocks, alarm annunciation and
a trip system for crucial parameters to mitigate unplanned cata-
strophic failure of equipment
25 Design provides color-coding of equipment and connected pip-
ing to identify its service media
26 Design provides vortex breakers in the suction and discharge
nozzles to avoid separation
27 Design provides a provision for an oil-mist lubrication system for
antifriction bearings of pumps
CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013 53

Cover Story
following the commissioning pro-
cedure framed earlier. It should be
enriched with workarounds for all
probable hurdles and hitches
5) Give. Give the end-user a trouble-
free plant asset or facility
Checklists for pumps
The series of tables included in this ar-
ticle (Tables 19*) were developed and
have been used as part of a program of
proactive reliability management. The
checklists shown here are specifically
designed for installing and commis-
sioning centrifugal pumps, but they
can also be applied to other pieces of
rotary equipment beyond pumps.
The purpose of the checklists found
in these tables is to provide a sum-
mary of the design review points for
assessing the maintainability of new
or existing rotary equipment for petro-
leum refineries or other CPI facilities.
The checklists specifically focus on
the identification of equipment-design
features, tasks, or procedures that
impact equipment downtime, repair
costs, labor hours and the skill level
requirements for maintenance staff.
The specific objectives for each table
are as follows:
Table 1. General maintenance
reduction
Table 2. Safety and environmental
system design features
Table 3. Design standardization
features
Table 4. Design features for routine
maintenance
Table 5. Design features for trouble-
shooting
Table 6. Design features for repair
and replacement
Table 7. Accessibility for visual in-
spections and monitoring
Table 8. Design for physical
accessibility
Table 9*. General precommissioning
checklist and commissioning proce-
dure for centrifugal pumps
The Description of check point column
in each table covers the essential fea-
tures for completion of items required
for the safe and first-time-right com-
missioning of rotary equipment in the
given category.
*Editor's note: Table 9 can be viewed in the
online version of this article (www.che.com)
54 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013
TABLE 2. SAFETY AND ENVIRONMENTAL SYSTEM-DESIGN FEATURES
FOR CENTRIFUGAL PUMPS
no. Description of check point Adequate Remark
Yes No
1 Design provides non-sparking, non-gulling metallurgy for close-
clearance areas and dynamic components that probably will
impact the static component of the assembly under operating
conditions other than normal. For instance, throat bush, bearing
guard, coupling guard
2 Design ensures that sealing materials for all components are
compatible with safety procedures and can function safely as
required by service conditions
3 Design provides a safe shaft-sealing system for the pertinent ser-
vice conditions, including all possible adverse operating condi-
tions
4 Design provides service-specific alarms and protection systems
for hydrocarbons, toxic media, smoke detection and fire
5 Design provides monitoring, alarm annunciation and protection
systems for all parameters that are crucial for human, equipment
and environment safety
6 Design provides all drain and vent points of the pump casing,
bearing housing and seal-oil system are connected to the des-
ignated closed drain and vent systems with proper isolation and
anti-reverse-flow devices
7 Design provides an applicable in-built protection system for elec-
trical equipment and instruments in line with hazardous area clas-
sification API RP 500, IS 5572, IEC 79-10 :1995, NFPA 69
8 Design provides that service-specific emergency handling for
safety equipment and systems are readily available and stepwise
procedures are displayed close to equipment
9 Design provides a data historian system for critical process and
environment safety information
10 Design provides for automatic actuation of safety-protection sys-
tems for equipment that handles flammable, toxic, auto-igniting
and other hazardous media
11 Design provides equipment-specific safe operating and mainte-
nance procedures
TABLE 3. DESIGN STANDARDIZATION FEATURES FOR CENTRIFUGAL PUMPS
Description of check point Adequate Remark
no.
Yes No
1 Design provides interchangeability of equipment that is in similar
or nearly similar service to all new pieces of equipment that are
under procurement
2 Design provides for interchangeability of equipment that functions
similarly with equipment that already exist in the operating facility
3 Design describes the full scope of standardization and inter-
changeability of components, such as shafts, impellers, wear
parts, mechanical seals, seal spares, bearings, couplings, gaskets,
hardware, valves, connectors and others
4 Design provides vendor-qualification criteria for maintaining stan-
dardization as a nodal decision-making point
5 Design provides for interchangeability of monitoring, control in-
struments, piping support system and so on
Implementing the five Gs the basic specifications and execution
The remaining portion of this article plans, while the licensor would pro-
elaborates on the methodology, and vide the most applicable technology
provides examples of checklists using and operating profiles.
centrifugal pumps as the asset. There Under the guidance of a companys
are three broad phases for the process. management, one or more teams from
Phase 1 Get and gather. Each various disciplines would be formed,
participating group should prepare making sure to involve the relevant
its own objectives to share, depend- knowledgeable persons and having
ing upon the project type. The project- that particular project department as
management department prepares the principal facilitator.
the project details, including the back- In the first phase, the end-users
ground, business objectives, location, feedback and participation comes into
mandate and so on. This would be the play in the form of input to the project
primary input and the basis for select- management consultant on the practi-
ing the consultant, technology licenser cal problems of reliability, operability,
or plant-equipment system or asset in- maintainability and safety, so that the
volved. The consultant would prepare weak points can be addressed at the
 TABLE 4. DESIGN FEATURES FOR ROUTINE MAINTENANCE
Description of check point Adequate Remarks
no. Yes No
1 The equipment and all mechanical adjustment and check points are located in primary maintenance zones
with permanent identification tags
2 The base frame is provided with jack bolts for moving drivers within the slotted zone for hold-down bolts
3 Coupling guard has a window for condition monitoring while in operation
4 The suction and discharge piping is provided with spectacle blinds for positive isolation without disturbing
the piping and casing orientations during dismantling of the rotor assembly
5 Quick-connect connectors are installed on frequently changed lines, for cooling-water seal flushing lines and
cables
6 Routine service and inspection points are not located behind other components or structural members, in
enclosed spaces or in the secondary maintenance zone
7 All components subject to routine maintenance are immediately accessible; for example, bearing housing,
seal-flushing system, cooling-water system, suction strainer, coupling
8 The drain- and vent-line isolation valves are immediately accessible
9 All block valves, control valves, control instruments, spring hangers, monitoring gages, seal-flush system
valves, strainers and transmitters are in immediately accessible locations
10 Procedures for routine maintenance and checklists for each component and system are displayed close to
the maintenance point
11 Procedural and personal safety precautions applicable for each activity of routine maintenance are dis-
played close to the maintenance point
12 A lubrication cart consisting of a tagged lubricant, seal-flush media, tools and tackle is available at the
maintenance point
13 Display of personal protective gear required for routine maintenance and also for unit area entrance is avail-
able at the maintenance point
14 Procedure and facilities for housekeeping and waste disposal by category are available at a designated
place outside maintenance point

TABLE 5. DESIGN FEATURES FOR TROUBLESHOOTING Examination of the possibilities for

OF CENTRIFUGAL PUMPS standardizing assets for the new
Description of check point Adequate Remarks equipment that may have operating
no. Yes No conditions that are close to equip-
1 Details of designed and rated operating specifications, includ- ment from the existing operating
ing media density, flow pressure, temperature range, minimum plant a cost-saving and inven-
flow, load-specific normal power, are available at the site
tory-reduction measure
2 Manufacturer-defined roto-dynamic parameters, such as allow-
able bearing temperature, vibration level and noise level are
Technical review of specifications to
available at the site ensure the inclusion of measures to
3 Equipment-specific normal startup and shutdown procedures mitigate potential problems faced
and check points as per OEM manual are displayed at the site previously in the plants operation
4 Mechanical seal specific allowable operating parameters and Finalization of the vendor-selection
checklists with procedure are available at the site
criteria to ensure reliable equip-
5 General layout facilitates visual inspection of major compo-
nents, connections, couplers, interfaces and potential damage ment and after-sales service. Prefer-
points ence must be given to vendors with
6 All mechanical interfaces are visible from the sides or end of proven track records for similar
the machine equipment and assets, and those
7 Manual test points are located in the primary maintenance who have local service centers
zone for all critical components of equipment and its auxiliaries

Examination of interchangeability
8 Test points are designed to eliminate or minimize the need to
remove components for testing of spare parts for the new and ex-
9 Test points are labeled and located close to the control or dis- isting equipment an inventory-
play with which they are associated reduction measure
10 General design and layout provides for rapid and positive Phase 2 Grow. This is the execution
identification of component malfunction, such as leakages, phase, which comprises erection and
high vibration, temperature rise, deterioration of performance,
in terms of low pressure, loss of flow, cavitation, high power pre-commissioning activities. Here, the
consumption, abnormal noise and so on execution team will work under the su-
11 A guide chart for equipment-specific problems, as per the pervision and guidance of the primary
manual, is available
team, who will carry out quality con-
12 Shop-test data and baseline data during the performance test
at the site are available
trol activities and ensure adherence to
13 Performance curves, data sheets, cross-sectional and assem-
safety rules during erection.
bly drawings for the pump and mechanical seals are available The following represent key objec-
14 Common potential failure modes with threshold values are tives for the project:
available Preservation. Proper preservation of
the equipment that arrives for erec-
design stage. The goal is to achieve adding activities that help meet the tion, as advised by the respective
trouble-free commissioning and long project mandate. A few follow: manufacturers or as per standard pro-
lifecycle with minimum resource uti- Interactive sessions to identify idle cedures.
lization and downtime. assets that can be reused in a project Erection. Erection of equipment in
Phase 1 also includes several value- a cost- and time-saving measure a way that adheres to the applicable
CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013 55
 Cover Story
TABLE 6. DESIGN FEATURES FOR REPAIR AND REPLACEMENT OF CENTRIFUGAL PUMP
Description of check point Adequate Remarks
no. Yes No
1 All pump components are labeled to positively identify part type, mounting location, type of lubricant required,
direction of rotation and other pertinent information
2 The dismantling and assembly procedures for each part and component are provided and available
3 Equipment-specific procedure for isolation and depressurization should be available to ensure safe and swift
repair. Procedure for replacement of parts should also be available
4 A monorail with trolley-mounted pulley block should be available over the pump slab for removing and install-
ing pumps and components without interfering with adjacent equipment, pipelines, cables, cable racks, valves
and so on
5 Sufficient free headroom over the pump and driver is provided for unhindered movement of the pump and
components as they are lifted in the pulley block, out to the dropdown area outside the unit
6 The weightlifting capacity of the pulley block along with the rail for travel should be 1.5 times the weight of the
heaviest single piece of equipment installed within the pulleys range of travel
7 The pulley-block travel rail shall be extended to the designated dropdown place outside the unit
8 For equipment at higher elevations, the deck floor should allow the equipment or components to access the
dropdown location safely
9 All heavy components should have lifting lugs or eye bolts of the required working-load capacity
10 The equipment assembly should be provided with the best possible self-cleaning and self-draining systems for
leakage of fluids, such as lubricants and service media, without spilling or splashing into the adjacent area
outside the equipment frame
11 All interface assembly components should be permanently marked with the correct manner of installation, in-
cluding direction, placement and geometrical orientation
12 Design should not require special tools or jigs unless they are the same as what is included in the scope of
supply by the OEM
13 Spare consumables for overhaul and repair should be available from a local vendor or OEM office to minimize
lead time
14 Equipment should be provided with jack bolts for shifting movable equipment for alignment

engineering procedure, with-

TABLE 7. DESIGN FEATURES FOR VISUAL INSPECTION AND ACCESSIBILITY POINTS
out creating operability and
Description of check point Adequate Remarks
maintainability hindrances. no.
Yes No
Here, the team members
1 All maintenance points should be visually accessible from all sides, or the
from operation, maintenance end of the machine, and should provide line-of-sight inspection capability
and technical departments 2 Design provides for clear and rapid visual identification of parts and safe
have major roles to play in access for checking with key monitoring checklist and checking procedure
pointing out issues and pro- during operation and when idle, so as to assess the health of the equip-
ment as a whole
viding advice for practical
3 Any unsafe points should be clearly identified and provided with protective
solutions to the project and guards and covers. The guard covers shall comply with applicable design
execution groups. and safety standards to achieve maximum built-in reliability
Safety. Safety audits by a 4 Access openings should be large enough to permit visual contact with the
component while the work is being performed on it (OISD 118 compliance)
multidisciplinary team to
identify the potential safety 5 Visual access points on the machine should have adequate space for
inspection and monitoring by the operator without introducing ergonomic
gaps and to track the imple- stress. A minimum of 48 in. of free space above equipment is suggested
mentation of corrective mea- 6 Visual access to openings should not be located underneath the equip-
sures that have been recom- ment body or behind other components that restrict visibility
mended. 7 For less frequently performed maintenance tasks, the maintenance point
may be located behind a protective cover. The component, however, should
Tracking. Tracking of prog- be directly visible when the protective cover is removed
ress with respect to sched- 8 Maintenance and service points should be located no further than 91 cm (36
uled timelines to identify any in.) from the maintenance workers head at the time of inspection
lagging portion and to deter- 9 Monitoring and check points should enable free access and be protected
mine a practical pathway for by a toe guard, hand railing and slip-proof flooring
expeditious progress.
Pre-commissioning. All pre-commis- tivities; measurement and metrics in- features of commissioning equipment
sioning activities of asset equipment dexes; audit recommendations compli- are the following:
systems and preparation for commis- ances; management review; emergency Ensure availability of startup and
sioning ensuring completion of the fol- management procedures. shutdown procedures
lowing: procedures and checklists; safe 
All protective functions and inter-
work practices; integrity and complete- Phase 3 Gain locks should be checked for proper
ness checks; contractor management Once pre-commissioning activities are configuration and functionally tested
modalities; training and performance complete, the door is open for step-wise All systems and subsystems should
assurance; management of change; commissioning of units, as per the pro- be checked for completeness
operational readiness; sequence of ac- cess-flow requirements. The essential Integrity plan and records should be
56 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013
 TABLE 8. DESIGN FOR PHYSICAL ACCESSIBILITY
Description of check point Adequate Remarks
no. Yes No
1 All components are accessible from the side or the end of the machine
2 All drain valves for casing, reservoirs and sumps are accessible from the side or end of the machine
3 All valves for equipment operation (for example, suction, discharge, cooling-water system, seal-flush system,
steam-quench system) are freely accessible and operable. Valves above 65 in. from ground or the level of
the base frame of the equipment should have a firm and protected platform
4 All components weighing 23 kg or more should be removed from the side or end of the machine and
should not have to be lifted up and over the machine frame or other components
5 Access openings should be sufficiently large to permit removal and replacement of all components con-
tained within that area
6 For enclosed equipment, hinged or quick-release access opening covers should be used where practical
with the hinges on the side or bottom so that the door will remain open during maintenance
7 The minimum number of bolts or fasteners should be used on access covers, equipment bay doors or other
protective shielding
8 For components weighing more than 45 kg, access openings and workspaces should be sufficient to permit
the attachment of hoisting and lifting devices
9 Screws, nuts and bolts should be located to allow free access and space for the use of hand tools required
to remove or replace them
10 Access openings should be sufficiently large to permit removal and replacement of all components con-
tained in that area
11 Non-hinged access opening covers should be designed with built-in handles or lifting-device attachment
points
12 Design provisions are made to support components weighing over 23 kg while they are being unbolted or
bolted into place
13 All components can be removed and replaced in a straight line from their place of attachment, so that
components do not have to be maneuvered around or over structural features or components
14 Illumination around the equipment should be sufficient at all access points, so that no accessibility hin-
drance occurs due to invisibility

reviewed. These include construc-
tion and testing records for inter-
connected piping, systems, electrical
circuits and control loops
Ensure adequacy of fire protection
systems, for example, fire-water sys-
tem, gas-leak detection system, fire
extinguishers, and so on
Ensure safe movement and ap-
proach of equipment
Note equipment-specific hazards
Determine deployment of trained
manpower
Ensure use of personal protective
equipment (PPE)
Ensure that equipment installation
procedures are consistent with re-
spective equipment standards
Apart from technical activities, the
most important areas that the proj-
ect management team must strive to
implement are the following:
Preparing a daily activity list and
review of progress
Adhering to safety guidelines
throughout the process
Establishing an effective good-
housekeeping procedure
Maintaining 360-deg communica-
tions with all stakeholders
Holding a toolbox meeting to ap-
praise the respective execution per-
sonnel on the scope of the job, safety
rules, job-specific safety precautions
and PPE
Adhering to the work permit system,
without exception
Documenting and record keeping
for commissioning activities and in-
cidents with absolute transparency
Managing progress updates
Benefits of giving
The benefits for following through on
the giving component of the process
are felt by end-users, and also extend
to the the organization as a whole. The
benefits include the following:
Improved reliability
Increased safety
On-time delivery of quality product
Higher customer confidence
Improvement through learning
Fostering long-lasting mutual re-
spect among teams
Solidifying ownership concept among
the post-commissioning team
Fostering collaborative culture
Increased plant familiarity and
hands-on training of end-user per-
sonnel (competency development)
Optimize inventory management
through standardization and inter-
changeability of asset-equipment
system spare parts
In addition, another benefit of end-user
participation in precommissioning and
commissioning is that, because end
users hold good rapport with internal
and external local resource providers,
they can guide the executing contrac-
tor on the availability of quality fab-
ricators in cases where urgent site
modifications or hardware changes
are required. This can save a great
deal of time and money.
Closing comment
The process described here has become
an inherent part of any project man-
agement because it nurtures a win-
win bond between the project group
and end-users group deriving sub-
stantial benefit for the organization.
Without such a structured approach,
projects could draw displeasure from
the end users as well as from manage-
ment for any design, operability and
maintainability issues in the newly
commissioned plant. Futhermore, end-
users could struggle to sustain the op-
eration with a problematic system, re-
sulting in a state of unreliability. With
this system in place, projects are likely
to have reduced post-commissioning
problems and more reliable plants.
Edited by Scott Jenkins
Author
Sourav Kumar Chatterjee is
chief manager of rotary equip-
ment at Mumbai Refinery of
Hindustan Petroleum Corp.
ltd. India (Petroleum House 17,
Jamshedji Tata Road, Mumbai
400020; E-mail: skchatterjee@
hpcl.co.in; Phone: +022 228-
63900). Chatterjee is a char-
tered mechanical engineer
and qualified boiler proficiency
engineer with a diploma in
electrical engineering. He has more than 23 years
experience in thermal power plant operation and
maintenance, and design maintenance operation
of refinery rotary equipment. He specializes in root
cause failure analysis and equipment reliability. He
had presented many papers at national and inter-
national conferences, and has published papers in
various journals and magazines.

CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013 57

 TABLE 9. PRE-COMMISSIONING AND COMMISSIONING CHECKLIST FOR CENTRIFUGAL PUMPS
Description of check point Adequate Remarks
no. Yes No
1 Pump is installed at site as per plot plan and piping and instrumentation diagram (P&ID), as well as-
foundation and location drawings
2 Pump Identification tag with service and key operating parameters is displayed at the site
3 Record of foundation-concrete quality test is available
4 Record for base-frame-leveling check is available
5 Process piping is as per isometric drawing with rated service gasket, supports and hardware
6 Hydro test of pressure parts and piping carried out satisfactorily, and record is available
7 All lines are cleaned by air-blow and cardboard-blasting test or by target-plate-verification test satis-
factorily of equipment requirement, and the record is available
8 All cooling water piping for equipment, sealing system, lube oil system (wherever applicable) are in-
stalled with valves, monitoring devices and instrumentations as per P&ID and system requirement for
reliable operation
9 Mechanical seal flushing plan installed as per sealing system drawing and verified for correct orien-
tation of piping connections, proper installation and routing of vent lines with all applicable monitor-
ing and control devices in line, as per P&ID and safety requirements wherever applicable
10 Steam-warmup lines and quench lines are connected, as per P&ID and general-arrangement draw-
ing of the pump, sealing system and all applicable inline monitoring and control devices
11 Seal shop-test report, materials of construction of components, rotor-balancing report, shop-test-run
reports of pump are available
12 Suction and discharge pipe-flange alignment with pump nozzle flanges checked for no parallel or
angular misalignment with specific gap for gasket only
14 Lubrication system installed and ready for operation (oil cup, breather, oil ring and so on)
15 All vent and drain connections are installed and routed to designated place as per P&ID
16 All process and auxiliary piping circuit integrity checked and established
17 All control and instrumentation circuit integrity checked, established and recorded
18 All field instruments (pressure gages, flowmeters, temperature indicators, dial thermometers) are cali-
brated and installed with date of calibration noted
19 All pressure-relief devices are installed after testing for rated pressure with values and date tag attached
20 For hot service pumps, piping insulation is complete
21 Driver (motor) no-load run test done and found to be satisfactory, along with the direction of rotation
22 All shop test certificates of driver motor are available. IR value of winding, cable checked and found okay
23 Protective relay setting done and checked
24 Driver (turbine) steam inlet-line target plate blowing done and found satisfactory
24 Driver (turbine) no-load run test done and overspeed trip system checked and found okay
25 Driver (turbine) exhaust system connected to designated header
26 Driver and driven equipment coupled and non-sparking coupling guard installed
27 Suction strainer checked and installed, and lube-oil system charged
28 Priming of pump in closed discharge condition done
29 Free rotation of pump checked and found to be okay
30 Recirculation system lined up and ready for use
31 Leak off line and balancing line system for multistage pumps lined up and ready for use
32 Start pump in closed discharge (crack-open discharge for multistage pump) condition and check
direction of rotation, shutoff pressure and amperage (motor) and shutoff pressure rpm steam flow
and pressure (turbine) at rated speed
33 Slowly opening recycle valve does not draw current (as there is no system backpressure, a large
opening will cause overloading of the driver)
34 Keep close monitoring on drop in suction pressure (shall not be more than 3%), discharge pressure,
vibration, noise, bearing temperature rise, leakages and establish minimum flow requirement of
pump. In case of any abnormality, stop pump
35 Mechanical run test (MRT) for minimum one hour at a flow above minimum flow required by pump
keeping close monitoring and record of flow-, suction- and discharge pressure, differential tempera-
ture rise, noise, vibration level, bearing-temperature rise, seal-flush flow, temperature rise, lube-oil con-
dition, and so on. Check for any leakage from system. In case the differential temperature (between
suction and discharge) of the media reaches 10C, stop pump
36 Commissioning of equipment is recommended within one month of MRT
37 The equipment should be blind-flanged, and the bearing housing should be filled with lubricating oil
up to the oil cup level. For HC oil service pumps, preferably casing should be filled with lube oil. Pump
shall be rotated by 180 deg once in three days during the period of idling
39 Once MRT is completed, check for commissioning shall start
40 Completeness of equipment upstream and downstream systems in all respects, including mechani-
cal completion. Control and instrumentation system completeness, electrical system completeness,
all punch points pre-commissioning safety audit are complete and recorded after physical
verification

58 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013

 TABLE 9. (Continued) PRE-COMMISSION CHECKLIST AND COMMISSIONING OF CENTRIFUGAL PUMPS
Description of check point Adequate Remarks
no. Yes No
41 All alarm and trip logics are set and functional check done
42 Display of all process and dynamic parameters of equipment are available at local control
panel and control room panel as well
43 Complete and documentation and records for the following activities should be available.
Equipment specific operating procedures and checklists
Safe work practices
Integrity and completeness checked
Contractor management modalities and defined roles with limits
Training and Performance assessment tips to field crews
Operational Readiness
Sequence of activities
Measurement and metrics indexes
Audit recommendations compliances
Management review
Emergency management procedure
44 The following activities are complete and documents are available before commissioning
Availability of startup and shutdown procedures
All protective functions and interlocks are checked for proper configuration and functional
tests
All systems and subsystems to be checked for completeness
Review integrity plan and records; that is, construction and testing records for interconnected
piping, systems, electrical circuits and control loops is done and found OK
Adequacy of fire protection systems (fire water system, gas leak detection system, fire extin-
guishers and so on) is checked and found OK
Adequacy for safe movement and approach is available
Awareness of field crews on of equipment specific hazards checked and found satisfactory.
Deployment of trained manpower is ensured
Use of personal protective equipment instruction is displayed at entrance of units.
45 Commissioning of equipment done in presence of equipment manufacturer, operation mainte-
nance. Technical and project members as per sequence of operation and set procedure
46 All process and equipment dynamic parameters are monitored continuously at local/CC panel
and found within normal limit
47 All control systems and alarm annunciation systems are functioning normally
48 Ready for GTR
49 Documentation mandatory documents for new project are available
Operation and maintenance manual
Process and equipment specification sheet
Performance curves., GA drawing
(Equipment, sealing system etc.) Cross sectional drawing (all major equipments)
Pre-commissioning completeness documents (Process equipment safety)
Electrical system lay out drawing
Instrumentation lay out drawing
Piping system & fire protection system lay out drawing
List of commissioning and two years spares.
Spares Interchangeability chart.
Base line performance data
Part list, special tools tackles list and P&ID (as built) Shop test data, MOC test data

CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013 59

 An Exercise In
Fire Protection
Asking the right questions about fire protection will prevent minor incidents
from escalating
E
Fred Ashmore very year the process industries
International Fire suffer hundreds of fires, some seri-
Investigators and ous and most minor. Fire protec-
Consultants tion matched to the risk is what
makes the difference between the fire that
torpedoes your business and the one that
doesn't even make the local newspaper.
For most process engineers, fire protec-
tion is something that specialists deal with.
But wrong judgments or decisions make
the difference between a controlled incident
and one that escalates. Because specialists
tend to plan for the worst-case scenario, the
plant may overdo fire-protection systems.
Although this sounds like a good idea, it
means investment and maintenance effort
that could be better directed.
So it is important for the non-specialist to
work with fire-protection designers and ask
the right questions. The fire protection of a
process plant should be based on a realistic
assessment of fire cases.
Fire-risk assessment is a hot topic today
and there is a mass of guidance, systems,
publications and consultants eager to help
(see box, page 78). Performing a fire-risk as-
sessment comes down to asking the ques-
tions:
What have we got?
What can go wrong?
What should we do about it?
Are we really doing what we should?
The fourth question is especially important
for the operating plant.
If an assessment exercise is carried out
properly, there should be a better under-
standing of fire hazards and precautions in
the plant, and why they are in place. This is
desirable because fire-protection systems
are non-productive and often lie idle for
years. Their maintenance priority can eas-
ily slip unless all concerned understand the
systems.
2 CHEMICAL ENGINEERING
Equally important: Writing down the rea-
sons for choosing certain fire protection is a
record for those who come afterwards. It is
easy to forget the reasons behind important
safety systems, causing accidents to be re-
peated.
What have we got?
The list of fire hazards in a process plant is
nearly always topped by flammable liquids.
Flammable gases are also common, but the
fire potential can often be mitigated by dis-
posing of the inventory by flaring.
Process plants dealing with materials other
than flammable liquids may have problems,
too. For example, paper plants feature high
in the list of major loss by fire, often involv-
ing storage, but also including incidents in-
volving paper machines. Food plants handle
combustible materials, often with potential
for dust explosion.
Other areas may have hazards that seem
less likely to be spectacular but could be
equally damaging to the business. For exam-
ple, a fire in a warehouse full of agrochemi-
cals can lead to runoff of contaminated-fire-
fighting water or a toxic-smoke plume drifting
off site. Or the loss of a thermal-fluid heater
could stop production for months. Although
preventing such incidents is the priority, one
needs to understand what fires could occur
and be able to deal with the consequences.
A useful approach is to prepare an inven-
tory of fire hazards for the area or site under
concern, and to list possible fire scenarios.
With this, you can talk seriously with fire-pro-
tection specialists to determine what protec-
tion is essential and what is optional.
A fire-hazards inventory should specify the
combustibles, define their fire properties,
and assess how they might be involved in
a fire. The hazards of process materials are
defined in the material safety data sheet of
commercial products.
WWW.CHEMENGONLINE.COM MARCH 1998

Pinpointing the fire hazards of solids can
be more difficult compared with liquids, be-
cause the form affects how they respond to
ignition sources. For instance, lump sugar is
not readily ignited, but a cloud of icing sugar
dust can cause a severe explosion.
When a flammable gas or vapor is in-
volved in a confined area, there is an explo-
sion hazard. Explosion and fire hazards are
linked, so be sure to note these in screening
fire problems.
As a simple example, consider the small
batch plant shown in Figure 1 with its inven-
tory of materials with fire hazards. The plant
is an open process area with enclosed con-
trol room and offices. For the process area,
there will be releases in various forms, and
the fire scenarios arise following ignition. For
this set of material's hazards, the following
scenarios are realistic:
A flammable liquid fire, either from a
pressurized release or from flowing liquids.
The total inventory of flammable and com-
bustible liquids is 19 m.t., including the heat
transfer fluid. This fire could be: a pool fire;
a jet fire involving sprayed liquids or vapor;
a flash fire of a vapor cloud; or a running-
liquids fire
Possible explosion if there is a big re-
lease of hot solvent
Fire at the heat transfer fluid heater. It
could involve the whole inventory of the
system (5 m.t.)
Fire involving the bag storage of chemi-
cal C. Check whether it's likely to melt and
spread to other storage
Fire involving ordinary combustibles or
maintenance items in storage area
Control room or office fire
What can go wrong?
With the inventory of combustibles and
notes on how they are arranged, the next
step is to determine what can go wrong. For
this type of exercise it is wise to take a team
approach to make sure all the major prob-
lems are caught. The team must include
people who understand the engineering,
operations and safety issues.
From the list of fire hazards, the team de-
velops a list of fire scenarios. Possible igni-
tion sources are not considered initially be-
cause the concern is with consequences,
not probabilities.
A possible explosion the second item
above should be assessed as part of the
overall safety analysis of the plant. Calcula-
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
tions may already be available for the maxi-
mum overpressures that could develop from
an explosion. For the purposes of assess-
ing fire protection needs, note explosions
as something that could disable protective
systems. An explosion is more immediately
hazardous to life than is fire, because it can
happen too fast to permit either reaction or
flight. Fire protection does not control explo-
sion hazards.
For a fire scenario with process fluids, as-
sess how long it could last if there is no re-
sponse. Both jet and pool fires can impose a
heat flux of around 200 kW/m2to structures
or equipment that are engulfed. This could
raise the temperature of a major steel mem-
ber to failure point in a few minutes. It can
also weaken process vessels so that they fail
under internal pressure, even with pressure
relief.
The duration of a pool fire depends on a
number of factors, including the inventory
of flammable solvents, burning rates, leak
rates, and drainage. It may be helpful to use
one of the many computer packages for
process hazard analysis which are available.
For this example, suppose that you as-
sess the maximum area of a pool fire as 50
m2. If the solvent has a pool-regression rate
of 5 mm/min this is a typical value for or-
ganic solvents and hydrocarbons the total
burning rate would be 0.25 m superscrip 3/
min. With a 12-m superscrip 3 inventory, the
pool fire could last 48 min.
In contrast to a pool fire, jet fires might
only last a few minutes if there is an effec-
tive blowdown system. However, even with
an overpressure of one or two barsg, there
can still be a substantial rate of release, so a
jet release could last longer than the 15-min
nominal blowdown time under API 521.
Other types of plant can be analyzed in
the same way, one scenario at a time. Pro-
cesses with complex chemistry and batch
plants with the potential for exotherm require
more detail and may have to consider ad-
ditional types of fire scenario. In addition to
simple leaks, a runaway reaction could lead
to major failure of a reactor or pipework. And
changes in process conditions from unfors-
eeable upsets could lead to major change in
fire potential. A systematic analysis will help
identify all fire cases.
Check the engineering
Before fire protection is considered, the
scenarios need to be checked against en-
gineering safeguards and shut-down sys-
tems. Do these safeguards place a limit on
MARCH 1998 3

the credible size of a fire? Looking again
at the set of incidents, check the pre-
cautions that limit fire size and escala-
tion. Also note any information gathered
on explosion hazards.
Considering a process plant with sep-
arate units, the key questions relate to
the largest credible release and the en-
gineering measures which may limit it or
help dispose of it. In quantitative risk as-
sessment, the analyst will usually start by
looking at isolatable inventories between
remotely operated isolation valves. This
should include all inventories that could
credibly be released, such as liquids,
flash gas and flammable gas.
How the isolation valves are to be
closed is one factor to be considered. If
it will be done manually, there must be a
realistic look at who will do it, how, and
under what circumstances. If it is auto-
mated, the level of reliability needs to be
checked. In any case, it is always wise
to allow for failure of at least one isola-
tion valve.
As a base case, it is reasonable to
allow for operation of flare and blow-
down systems. But it is equally reason-
able to consider a case when the flare or
blowdown system has failed itself. You
may decide not to adopt special fire-
protection measures, but it would be
wise to note these possibilities of failure
and how they could affect an incident.
Likewise, drainage systems should
dispose of spills, but oily drains may not
be sized to take large volumes of fire
fighting water. If there is a credible re-
lease involving tons of flammable liquids,
the drains may back up and float a burn-
ing spill to other areas.
Some notes on the engineering safe-
guards for the plant in Figure 1 are:
Fire detection is provided. Process
shutdown on confirmed fire stops
pumps, minimizes spray fire hazard
Process shutdowns and automatic
blowdown are provided for major up-
sets including leakage
Isolation by actuated process valves
should be discounted, because the
valves are not fire-safe design
Safety analysis calculates explosion
overpressure of 300 mbar (4 lb/in2) in
plant, 150 mbar (2 lb/in2) at the control
room. The most severe case could
cause failure of process pipework
The heat-transfer package has
instrumentation for hostile heater fire
leading to shutdown and isolation of
4 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
the main inventory. Discount spreading
to other areas because of the location
and drainage
The plant drainage system has a
design capacity of 120 m superscrip
3/h. Layout and catchment should
localize the area of a liquid spill to 150
m2. The possibility of overflow with
firefighting water exists
Layout means no credible spread
to process areas from bag storage,
stores or control room
The standard and capability of the fire-
fighting response must also be estab-
lished. If the plant has a professional fire
team that will be at an incident in one or
two minutes, you may well decide that
no active fire protection systems are
needed. However, if the team is made
up of operators, effective firefighting re-
sponse could take several minutes, al-
lowing a fire to escalate.
Make sure you know what the fire
team is expected to do, how they are
equipped and how they would respond
to the incidents in the plant. The list of
credible fire incidents must be shared
with whoever leads the fire response.
Non-process areas should not be
overlooked when preparing a fire haz-
ards inventory. Plant services, control
and electrical systems, storage areas
and support services may not feature on
the profit centers, but they can still suffer
fires that hit profits.
Plant services such as power gen-
eration and fuel systems are suspects.
Don't forget that a pressurized spray of
lubrication oil can behave like a flame-
thrower. Check the total inventory that
could be involved. Heat transfer sys-
tems often use combustible fluids at
high temperatures and many of these
can and do burn.
Control areas and electrical distribu-
tion are tricky to assess. Electrical pro-
tection does fail, cables and densely
packed wiring do burn. How do you
assess the largest credible incident if
no one does anything and the protec-
tion doesn't work? In most such areas
the likelihood of fire is low but the effects
can be rather serious.
Warehouse-storage areas are often
discounted as a fire risk on the basis of
ignition-control measures. However, the
rate of growth of a fire in rack storage
is quite startling. In tests of fire involving
storage racks, it may take as little as 30
s for a fire to spread from the base of a
NOVEMBER 2015
20-ft rack to the top. By then it is gener-
ating temperatures of about 1,000 de-
grees C. You should reckon that manual
firefighting is unlikely to be effective even
with efficient fire detection and a crack
fire team. It is important to assess the
overall effect if a warehouse takes fire
and burns out, especially when toxic
materials are involved.
What is the likelihood?
On the level of assessment discussed
here, it can be helpful to rank the likeli-
hood of the various fires identified for a
unit or plant. A group exercise is again
a good way to do this. The team that
agrees a set of fire scenarios should also
agree on possible frequency of credible
incidents.
A convenient scale is shown in Table
1. A more formal assessment of the like-
lihood of fire in a process plant is a major
exercise, and one is referred to informa-
tion given elsewhere .
Quantitative fire risk analysis is gener-
ally carried out at a level where the de-
tails of fire protection are not examined
and major assumptions are made at
every stage, from failure of plant through
ignition probability to protection effec-
tiveness. It does not help much in as-
sessing the needs of an individual plant,
though it may be useful in setting priori-
ties between plants.
What do we do about it?
Armed with a list of credible fire sce-
narios and an assessment of relative
frequency, it is now possible to consider
fire-protection options. Above the level
of first-aid fire fighting, fire protection
can be divided into three categories:
Passive fire-protection measures
Active fire-protection systems
Systems for the fire team.
Passive fire protection: Passive pro-
tection covers two main areas, fire in-
sulation for structures including vessel
supports and fire insulation for equip-
ment. One needs to examine what is
already in place, how long it will stand
up to the potential fires you identify, how
long the credible fires could burn, with
pessimistic assumptions.
The main options in passive protection
systems are lightweight concrete and
fiber-type insulation. Board systems and
applied coatings, such as epoxy-based
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
intumescent resins, are also used.
Concrete is the traditional fire protec-
tion for structures. General advice for a
large process plant is to fire protect criti-
cal structures up to 10 m above grade.
Deciding which structures to protect de-
pends on the analysis of credible fires.
Typically, it would include primary steel-
work in process units with the potential
for pool-fire, and pipe tracks.
Board systems for passive protection
are used for structural members and
protective enclosures, for example on
cable systems. Fiber-based protection
systems are similar to process insula-
tion, though the cladding may need to
be to a higher specification if it is to re-
sist process fire exposure. Coatings can
be expensive, but some types will pro-
tect against corrosion.
Some operators have had horrible
experiences with corrosion behind pas-
sive protection of process equipment. A
system which prevents corrosion, such
as an applied coating or one which fa-
cilitates inspection, is desirable.
In some plants, process thermal insu-
lation can double as passive fire protec-
tion. A small increase in specification at
the design stage can yield benefits in
fire safety. Stainless-steel cladding and
banding, mineral-fiber insulation rather
than glass wool, improved details of
joints can allow you to rely on thermal
insulation for passive fire protection.
Active systems: Where active fire pro-
tection in process units is found to be
necessary, the standard approach is to
use waterspray deluge, usually based
on NFPA 15. The purpose of most del-
uge systems is to control heat input to
vulnerable equipment or structures.
NFPA 15 lists the following functions:
Extinguishment of fire
Control of burning
Exposure protection
Prevention of fire
Waterspray can extinguish certain
types of fire, but is primarily used to con-
trol burning and limit the severity and for
exposure protection. These in turn serve
to limit further failures that could lead to
additional releases.
The classical waterspray system uses
open nozzles spraying water from about
0.5-1.5m off the protected equipment
or structures. Pipework is arranged in a
NOVEMBER 2015 5
 tree or grid. Each system served by a
master control valve.
A deluge system will generally be con-
trolled by an automatic detection sys-
tem that trips the deluge when there is
a confirmed fire in the protected area.
Tradition is to use pilot head detection
of fusible elements on a pressurized air
line that is linked back to the deluge
control valve. This approach is not ideal
because of potentially long response
times. The fusable element is normally
a sprinkler that is designed to respond
to hot air rising from a fire and collect-
ing under a ceiling. In the open, hot air
won't collect, so the heads take longer
to respond. That means they require a
large fire to operate at all. In addition,
pilot heads mean another system of
small-bore pipework wrapped round the
process plant. In general a flame detec-
tion system is much more sensitive.
Equipment for fire team use: No mat-
ter what fire protection is chosen, fire
teams will respond to a fire. Even if the
fire is under control, the fire fighters will
have to carry out the final stages of ex-
tinguishing it.
Every process plant will have hydrants
in the area, and some will have a good
deal more to support the fire team, de-
pending on whether they are expected
to achieve control as well as final extin-
guishment. This in turn depends on the
overall approach to plant fire protection.
In addition to the portable monitors
that most fire teams have available, there
may be a need for fixed monitors and,
possibly, remote-controlled monitors. All
these are tools of the fire fighter's trade,
and the fire team needs to be closely in-
volved in selecting the equipment, and
in deciding where to locate it. Leaving
such decisions to a design engineer can
lead to dissatisfaction, to put it mildly.
Your analysis may well lead you to reject
active fire-protection systems and decide
that passive protection allied to fire team
response is enough. The fire team requires
the right equipment for a credible fire, and
an adequately-sized water supply.
Calculating the water demand
The main parameter of deluge protec-
tion is waterspray density. Table 3 sum-
marizes the principal cases from NFPA
15 that are likely in a plant.
The water demand for any deluge
system has to include an allowance for
what is called hydraulic imbalance, the
6 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
effect of friction losses in the pipework
and elevation changes. Because of this,
the nozzles near the supply end will see
higher pressure than those at remote
points. A typical value for a unit without
large elevation changes is 115%.
What would this mean for the example
plant discussed earlier? Suppose there
was a decision to provide area water-
spray protection for the process area to
control the potential pool fire, plus vessel
protection for six vessels. Assume that
the main steelwork has been given pas-
sive fire protection. As shown in Table
4, the water demand for this exercise is
6,800 L/min.
An additional allowance for manual fire
fighting with hose streams and monitors
must be added. A minimum value would
be around 1,000 L/min, though a more
prudent allowance at this plant might be
twice that since there is a significant in-
ventory of flammable liquids. The supply
pressure will typically need to be 7 bars
subscript g or more at the deluge valve.
There are other concerns over del-
uge systems, apart from the size of the
water supply requirement, the complex-
ity and the cost of the system. Because
the pipework is open to the atmosphere,
internal corrosion can be a major prob-
lem unless a higher specification mate-
rial is chosen. Modifications to the plant
usually also require modifications to the
deluge system to maintain its protection.
Overall, deluge protection is not a choice
to be made without careful thought.
There are other options in active pro-
tection. It is certainly worth considering
the use of a foam system, which can
extinguish a flammable-liquid pool fire
rather than just limiting damage by ex-
posure protection. The rates of applica-
tion are lower than with waterspray sys-
tems, so choosing a foam system may
be especially relevant in situations where
the plant's firewater supply is somewhat
limited.
Modern foam equipment is extremely
effective, provided the fire matches the
design basis of the equipment. How-
ever, foam systems cannot provide ex-
posure protection in all types of fire. For
example, a severe jet fire, or a fire trav-
eling down a structure, is outside their
capability.
In enclosed areas, automatic sprin-
klers are often a good option. In most
applications, automatic sprinkler design
is based on operation over a limited area
NOVEMBER 2015
of about 300 m2.
However, for indoor flammable-liquid
processes, the water demand is often
based on sprinkler operation over the
entire enclosed area, because of the
high heat released. Predicted water
demand might well end up nearly the
same as for a deluge system covering
the same indoor area.
The advantages of using sprinklers
indoors are that a separate detection
system is not needed, thus lowering
the cost. Second, a problem with the
control valve won't lead to flooding,
whereas with a deluge system, any ac-
tuation, including false actuation, means
the whole system is discharging water.
Third, although the water demand has
to be reckoned on the basis of most of
the sprinklers operating, a small fire may
well be controlled by just a few, mean-
ing less clean-up. Fourth, pipework
corrosion is less of a problem since a
sprinkler system is sealed off from the
atmosphere.
Many process plants have some spe-
cial risks, involving process areas with
high-value or critical equipment. Exam-
ples may include major gas compres-
sors, gas-turbine-driven equipment,
and control electronics.
If the high value equipment is in an
enclosed area, gaseous systems using
Halons or carbon dioxide used to be the
choice. Now, however, Halons are no
longer an option, as will be discussed
below. Water sprinklers or a local deluge
system can be satisfactory for certain
special risks, such as under a turbine to
protected lubrication oil systems. More
recently, water mist systems present an
additional option for special risks.
Water-mist systems are based on de-
sign concepts similar to those of auto-
matic sprinkler and water-spray deluge
systems. The water supply is connected
through a control valve to nozzles on
distribution pipework. Systems are
available with either closed nozzles, op-
erated by a temperature-sensitive bulb
or fusible link, as in sprinkler heads, or
open nozzles, as are used in water-
spray deluge systems.
Water-mist systems use smaller drop-
lets than water sprinklers or waterspray
units. A typical droplet-size distribution
is 20-200 m, though some systems
provide a proportion of larger droplets
to give the spray more momentum. The
fine mist is efficient at quenching flame
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
and in fire tests, the progress looks more
like that of a gas-extinguishing system,
except that the agent is cheaper.
Water supply requirements for water-
mist systems are much lower than those
for waterspray or sprinkler protection.
Total water requirements for extinguish-
ing may be a few hundred liters. Be-
cause of this, skid-mounted supplies
are practical especially for remote sys-
tems. Alternatively, the supply can be
from a pressurized, clean-water system
of relatively low capacity. Typical pres-
sure requirements are between 7 and
15 bars, depending on the vendor.
Because these systems produce a
fine spray, they are vulnerable to wind
and unsuitable for outdoors protec-
tion. Water-mist protection of special
risks depends on correct selection and
placement of nozzles to achieve correct
distribution and at present the supplier
should do such designs.
The throw of the nozzles is limited,
and the spray will not penetrate a strong
fire plume beyond a few meters. A wa-
termist protection system can be used
to protect a small room, but probably
would not suit most indoor process
areas. Typical applications include the
protection of gas turbines, diesel gen-
erators, cable tunnels and some flam-
mable liquids.
Halon replacement agents
Because of their ozone-depletion po-
tential (ODP), Halon fire-extinguishing
agents such as Halon 1301, are effec-
tively a thing of the past. This is a pity
for fire-protection engineers, because
Halon 1301 systems for enclosed risks
were efficient and relatively easy to en-
gineer. But the need remains for clean,
fast-acting fire protection, which can be
used in manned areas. CO2is a practi-
cal option with a long pedigree, but life-
safety problems limit its applicability and
add engineering requirements.
A number of alternative agents have
been developed to replace Halons.
The two main groups are halogenated
agents and inert-gas agents. Haloge-
nated agents extinguish by both chemi-
cal and physical action on the flame, the
precise mechanisms depending on the
agent. Accepted agents combine ex-
tinguishing capability with low toxicity
and low ODP, but some have significant
global-warming potential and could per-
sist in the atmosphere.
NOVEMBER 2015 7
 Inert-gas agents work by reducing
oxygen concentration in the protected
space. The available agents are basi-
cally argon or argon-nitrogen mixtures.
One of them adds some carbon dioxide
to the mix as a precaution to improve
oxygen uptake for anyone exposed to
an extinguishing concentration. These
agents have zero ODP since their com-
ponents are atmosphere gases.
Most of the agents require similar hard-
ware to that which was used in Halon 1301
systems, with a skid storing the agent under
nitrogen pressure, actuated release valves,
distribution pipework and nozzles. To main-
tain the concentration of agent, the control
system must isolate air-handling systems
and close doors. A room protected with
a gas agent must be reasonably leak-free.
Room-integrity testing should be part of the
system commissioning.
The commercially available clean-
agent systems should be just as effec-
tive for fire protection as Halon 1301 in
terms of safety and damage reduction.
Possible applications for chemical fire
protection will again be control and elec-
trical areas, and enclosed risks.
A final look at the case study
Let's look again at the small process
plant considered in Figure 1. The likely
decisions with this particular plant will
depend chiefly on whether the fire team
can respond fast enough to handle the
largest credible incident. A possible
combination might be:
Passive protection for structures in
the process area
Area coverage by flame detectors,
with voting to avoid false alarms
Monitor coverage to supplement
the hydrants round the unit
Deluge protection for the process
vessels and pumps
Fire detection systems covering
storage areas and control room and
electronic equipment
8 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM
The actual decisions must match the
actual plant, based on a reasoned as-
sessment of fire risks and likelihood,
and a realistic view of the worst cases
and what they might do for plant vi-
ability. One of the basic assumptions in
system design, and also in fire risk as-
sessment, is that we only have to deal
with one major incident at a time. If a fire
occurs and spreads to involve multiple
units or areas, it is likely to overtax your
resources.
So, the decisions on protection at a
major complex will be different from
those for a small, isolated pharmaceuti-
cal plant. In a refinery, active fire-protec-
tion systems are unlikely to be an op-
tion because of the size of units. In an
offshore complex, the need to prevent
escalation may mean that active fire pro-
tection is deemed essential. The sce-
nario-based methodology outlined here
is appropriate to any of these cases, but
may lead to diverse conclusions.
Edited by Gerald Ondrey
References
1. Dinenno, P.J., ed. ``SFPE Fire Pro-
tection Handbook,'' 2nd Ed. National
Fire Protection Assn.,'' Quincy, Mass.,
1995.
2. Lees, F.P., Loss Prevention in the
Process Industries, 2nd Ed., Butter-
worth, London, 1997.
Fred Ashmore has worked in safety and fire-
related fields for 20 years. He is a consultant
with International Fire Investigators and
Consultants (220 Ayr Road, Glasgow,
G77 6DR U.K.; Phone: 44-141-639-6611;
Fax: 44-141-639-1297). Ashmore is a
member of the Institute of Fire Safety and
the Soc. of Fire-Protection Engineers, and
is a Fellow of the Safety and Reliability
Soc. He has a B.A. from Cambridge and
a Ph.D. in chemistry from London.
NOVEMBER 2015