Professional Documents
Culture Documents
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
Facts at Your Fingertips: Static Electricity Discharge and Fire Prevention.......................................... 158
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
presented in terms that are useful for SIL ability of an SIF. Once a safety instrumented
afety instrumented systems (SIS) assignment. Instead, the literature usu- function is designed, reliability calculations
are in the spotlight these days. ally contains some variation of this state- can show if the SIF is reliable enough to reduce
More companies have come to rec- ment: Each organization is responsible the risk to a tolerable level.
ognize the importance of recently for establishing its own risk-tolerance Safety integrity levels range from 1 to 4, but
published consensus standards, and re- criteria. Some organizations have corpo- it should be noted that SIL 4 is not recognized
cent fatal accidents tied to the failure of rate risk-tolerance criteria that speak to in all SIS standards. Instead, there is an expec-
instrumented safeguards have brought the total corporate risk, but many have tation that the process will be re-designed to
reduce the risk. An SIS can also include SIFs
further attention. The result is that facili- no risk-tolerance criteria at all. It is still
that are not SIL rated. Various users refer to
ties all over the world are looking at their a rare organization that has established them as N/R (not rated) or SIL 0 handy
safety interlocks and concluding that risk-tolerance criteria that are actually terms, albeit not defined in any standards.
what they really need is a full-blown SIS. useful at the level of a process unit, where
As some work to upgrade their SISs, the SIS resides.
they find plenty of guidance available for Some companies define these terms, but
the early steps process hazard analysis Recommended approach may use frequency intervals that are not
(PHA), consequence analysis, and layers A typical approach to SIL assignment, uniformly spaced. A typical set might be:
of protection analysis (LOPA) and also but certainly not the only one, is to use Event occurs once a year
guidance for the later steps component a risk matrix. Risk is a product of conse- Event occurs once every 10 years (~1
selection, failure probability and SIL quence and likelihood. As shown in Table order of magnitude)
(safety integrity level) verification calcu- 2, consequence and likelihood categories, Event occurs once every 50 years (~
lations, and system-design guidelines. A like the SILs themselves, are separated order of magnitude)
disconnect comes when end users have de- by orders of magnitude. The challenge for Event occurs once every 150 years (~
veloped their lists of safety instrumented each company then, is to define the conse- order of magnitude)
functions (SIFs) and are asked to assign quences that are considered serious and Not likely to occur
an SIL to each one (Table 1). the benchmark frequency (f). The same holds true for the severity in the
Detailed design of an SIS cannot begin Many companies already use some type consequence categories; a typical (nonlin-
until each SIF has been assigned an SIL. of matrix, usually to prioritize recommen- ear) arrangement might be:
This is the point in the process where dations from HazOps (hazard and opera- Loss of life; damage over $1 million
many are tempted to throw up their bility reviews) or other PHAs. An example Lost-time injury; damage over
hands in despair and want to arbitrarily of this is shown in Table 3, where a rat- $500,000
assign everything as SIL 3. This is not the ing of 1 would mean something should be Medical treatment; damage less than
answer. At the very least, it fails to comply done immediately, and a 10 would mean $500,000
with the letter and the spirit of the stan- that no action is required. Minor injury; near miss; poor quality
dards, but worse, it misallocates resources Typical likelihood categories include No injury, impact on process
to overdesign when these resources could terms such as frequent, occasional, sel-
be better used to reduce risk elsewhere. dom, remote and unlikely but what do Converting an existing matrix
The disconnect occurs because SIL as- they mean? A company may leave it to the To convert an existing matrix to an SIL-
signment depends on a comparison of individual participants in the PHA to de- assignment matrix, the first step is to ad-
two values: determined risk, about which cide for themselves. just the likelihood categories so they are
Table 1. Safety Integrity Levels Table 2. Risk Matrix
Safety Probability of Risk Reduction Consequences (per event)
Integrity Failure on Demand Factor (RRF) < 0.01 x > 0.01 x > 0.1 x > 1 x > 10 x
Level (PFDAVG) Serious Serious Serious Serious Serious
SIL 4 10-4 > PFD > 10-5 10,000 < RRF 100,000 f > High N/R SIL 1 SIL 2 SIL 3 SIL 4
Likelihood
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
Likelihood
one order of magnitude apart. After that, 3 9 8 7 6 3
consequence categories should be adjusted 4 10 9 8 7 4 'JSTUBJET
so they are also one order of magnitude 5 10 10 9 8 5
apart. The intervals could be 1.5, 15 and /FBSNJTTFT
Table 4. Relative cost of
150 years, or 5, 50 and 500 years, as long consequences
as they are an order of magnitude apart. FIGURE 1. The Accident Triangle or Safety
Fatalities 1.0x Pyramid depicts the simple observations of
The next step is to define the conse- Serious injuries 0.1x H.W. Heinrich [1]
quence categories. The universally used Injuries (reportables) 0.01x 'BUBMJUJF
scale and absolutely required by the First aids (non-reportables) 0.001x NBO
*OUPMFSBCMFSJTL
SIS standards is of consequences to per-
NBOZFBS ZFBS "VTUS
sonnel. But other important consequence more severe consequences. This led to the XPSLFS
QVCMJD
/4
scales, like community, societal or envi- now familiar, safety pyramid (Figure 1). )POH,
ronmental consequences, may also show While the premise of the safety pyra- "-"31
up in one of these matrices. Just keep in mid has recently been called into ques- PSUPMFSBCMFSJTL /FUIFSMB
mind that as an SIF for reducing the risk tion, Heinrichs original observation is SFHJPO
NBOZFBS ZFBS 6O
of a specific hazard is evaluated, the conse- still valid: near misses, first aids, and XPSLFS
QVCMJD
,JOHE
quences of that hazard must be evaluated fatalities are each separated by about an
on each scale. The event will have the same order of magnitude (Table 4).
frequency, regardless of the types of conse- Many have published cost estimates as- /FHMJHJCMFSJTL
quences considered, and the consequence sociated with increasing levels of injury,
FIGURE 2. The ALARP principle, developed
that results in the highest SIL rules. with costs ranging from a few hundred dol- in the U.K., states that risks should be reduced
While some organizations want to con- lars for a first-aid injury to a million dol- to a level that is as low as reasonably practi-
sider site and operability consequences of lars for a fatality. What the various scales cable
a hazard in their analyses, there are seri- have in common is that severity of conse-
ous disadvantages to keep in mind. One, it quences, when reduced to dollars, seems to one operating a process with risks in the
puts a dollar scale on the same matrix as be separated by orders of magnitude. tolerable risk region must demonstrate
environmental, health and safety (EHS) Once likelihood and consequence cat- that they have achieved the lowest risk
consequences, and it is not a big leap for egories that are separated by orders of possible, taking into consideration cost
a plaintiffs attorney to equate the EHS magnitude have been established, an SIL- versus risk reduction.
consequences to the dollar amounts. Also, assignment matrix can be developed. An
functions that strictly protect against eco- example of a blank table is given in Table Government mandates
nomic loss dont need to be burdened with 5. All that remains is to associate an SIL Some countries have mandated upper and
the documentation requirements of the with each cell in the matrix. lower levels for tolerable risk. Most coun-
SIS. Organizations are usually better off tries that have set these levels use the
to leave the economically driven functions What is a tolerable risk? average level of risk faced by the entire
in the basic process control system and SILs represent the residual risk between workplace population, regardless of oc-
avoid giving them SIL ratings. what is left after all other layers of protec- cupation, as the upper limit for tolerable
tion have been applied to a hazard, and risk. In many developed countries, that
Consequence categories the tolerable risk. But what is the toler- is the equivalent of one fatality per year
If classifying likelihoods into categories able risk? Consider this: Whether or not per 1,000 workers. This upper bound-
that are separated by orders of magni- your company has established a tolerable ary represents the level of fatality rate
tude is hard, its nothing compared to risk criterion for your facility, what do you already tolerated in industries such as
separating consequences into categories. believe it should be? How infrequently underground mining or deep-sea fishing.
Once a benchmark frequency is defined would a fatality occur for you to be able The lower boundaries are generally two
as a number, dividing or multiplying it to describe your facility as safe? Once a or three orders of magnitude lower.
by ten is straightforward. Development of year? Once every three years? Once every The Netherlands took a similar ap-
consequence categories, however, requires 5,000 years? Once every million years? proach, but then declared that workplace
agreement on what consequence is ten The ALARP principle (Figure 2) was de- risk should be 1% of general risk, because
times worse than another. veloped by the Health and Safety Execu- it is not voluntary risk. This results in lev-
In the early twentieth century, H.W. tive of the U.K. It states that risks should els that are 100 times lower than those
Heinrich [1] proposed the idea that for be reduced to a level that is As low as chosen by most countries that set levels.
each fatality there were many incidents reasonably practicable. The ALARP prin- These levels, as shown in Figure 3, are
with less serious consequences, and that ciple first divides risk into three areas: sometimes called bright lines, especially
for each incident with less serious con- Intolerable Risk at the high end, Negli- when discussed in the U.S., where there
sequences, there were even more near gible Risk at the low end, and the Toler- has been no success in setting tolerable
misses. From that simple observation able Risk that falls between the two. Risk risk levels. The argument was made that
grew the notion that if we could be more in the middle region can be tolerated as a strict bright line approach to decision
alert to near misses and reduce them, we long as all cost-effective measures to re- making is vulnerable to misapplication
would necessarily reduce the incidence of duce risk have been put into place. Any- since it cannot explicitly reflect uncer-
70 Chemical Engineering www.che.com September 2007
'BUBMJUJFT
NBOZS
"VTUSBMJB $PNQBOZ
/48
*
)POH,POH $PNQBOZ
/FUIFSMBOET
**
$PNQBOZ Bioengineering
***
6OJUFE
,JOHEPN
4NBMM
DPNQBOJFT
Inversina
FIGURE 3. Government mandates defining
tolerable risk are often referred to as bright
FIGURE 4. CPI benchmarks show that
large, multinational companies tend to set
the gentle way
lines. The color coding here refers back to
the 3 levels in the ALARP principle as shown
levels consistent with government man-
dates. Smaller companies tend to operate in of mixing.
in Figure 2. The U.S. does not set tolerable wider ranges and implicitly at higher levels
risk levels, nor offer guidelines of risk
The Inversina mixes solids or liquids
Table 6: Voluntary risks thoroughly and efficiently. The process
Smoking 9.7 x 10-3/yr or 9,700 fatalities per year per million smokers
is clean, because mixing takes place in
closed containers that can be quickly
Automobile accident 1.4 x 10-4/yr or 140 fatalities per year per million people
interchanged. The Inversina mixes a
Lightning strike 1.5 x 10-7/yr or 15 fatalities per 100 years per million people
diverse range of components rapidly
and in an extremely gentle way.
tainty about risks, variation of suscep- ing out over the next 100 years. A small
tibility within a population, community company with a single plant will look at Segregation does not occur, even
preferences and values, or economic con- a single event that occurs once every 250 after extended mixing times, by virtue
siderations all of which are legitimate years and may very well decide that once of the eversion phenomenon ( Paul
components of any credible risk manage- every 250 years might as well be never. Schatz principle ) .
ment process. In other words, in the U.S., A large company, with 50 similar plants Applications for the Inversina: analyti-
tolerable risk remains very much subject around the world would look at that same cal labs, metal finishing shops, powder
to the political process. risk as once every 250 years for each plant, metallurgy and nuclear industry,
but once every 5 years for the corporation, manufacture of batteries, cement,
CPI benchmarks and make a very different decision about ceramics, cosmetics, dental products,
Almost all major companies in the chemi- whether the risk is tolerable (Figure 4).
diamond tools, dyes and pigments,
cal process industries (CPI) have guide-
electrical and electronic devices,
lines for tolerable risk, but they are Voluntary and natural risks
generally considered highly confidential. The levels mandated by governments, explosives and pyrotechnics, foods,
No company is going to announce to the or self-imposed by companies, can best homeopathic products, household
public, for example, that they would tol- be understood in terms of some of the products, medicines and pharmaceu-
erate up to one fatality per year per ten risks to which people voluntarily expose ticals, plastics, printing inks and many
thousand workers. themselves. That is not to say that risks other products. The Bioengineering
Some companies, especially smaller like these should ever be compared to Inversina is available with capacities
ones, do not have explicit guidelines. How- process risks in a public setting, which of 2, 20, 50, 100 and 300 L .
ever, the risk they tolerate can be inferred would likely inflame the audience. But,
from the risk reduction measures they for ones own understanding, it is helpful
have in place. Smaller companies with- to know that the upper level for ALARP is
out explicit guidelines generally tolerate typically set at 1 x 103 per year, or almost
greater levels of risk than their larger ten times lower than the risk to which
counterparts and this isnt inappropriate. smokers expose themselves. If a chemical Bioengineering, Inc.
A small company, struggling to establish process had the same risk as smoking, it Waltham, MA 02451, USA
itself or to make payroll, is going to look at would be considered an intolerable risk. Bioengineering AG
the amount of risk that it can and must On the other hand, dying from a light- 8636 Wald, Switzerland
tolerate to get through the next year very ning strike would generally be considered info@bioengineering.ch
differently from a company that is look- a negligible risk. If you consider your own www.bioengineering.ch
Circle 52 on p. 92 or go to
adlinks.che.com/6899-52
Likelihood
2. Occasional N/R N/R SIL 1 SIL 2 SIL 3
chances are that you have not invested in (1 event/15 years)
lightning protection at your home. Even 3/4. Seldom/Remote N/R N/R N/R SIL 1 SIL 2
in your plant, the lightning protection is (1 event/150 years)
for electrical equipment, not personnel. 5. Unlikely N/R N/R N/R N/R SIL 1
The risk of a fatality in a car accident is
within the ALARP region. Of course driv- Further assume that at this facility, safety and confined-space entry issues,
ing has the potential for fatalities, but the the workforce includes 250 workers who and transportation-related fatalities.
risk is tolerable because the convenience would be exposed to process risks: For Example A, assume that half of the
or benefit is worth it. If it wasnt, society (1 year/250 man-years) (1 fatality/100 risk comes from process safety risks or
would insist on lower speed limits and years) = 1 fatality/25,000 man-years process hazards. If total tolerable risk is
more safety features in cars. = 4 x 105 fatalities/man-year 4 x 105 fatalities/man-year, then process
= total tolerable risk safety risk is half of that, or 2 x 105 fa-
Example A: 1 fatality/100 years In this case the total tolerable risk is one talities/man-year.
Lets get back to SIL assignment. While fatality per 25,000 man-years, or 4 x 105 Although weve assumed that no more
each of us has a different level of personal fatalities per man-year, which falls in the than half of the risk comes from the pro-
tolerable risk, when asked how small the middle of the tolerable risk region. cess hazards, we need to consider all of
average frequency for a fatality would the process hazards that could lead to a
need to be for a plant to be considered Process risk is only part of risk fatality. Unless there is only one poten-
safe, it is not at all uncommon for people Note that the tolerable risk established tially fatal hazard, not all of the process
to pick 100 years. This is longer than any above is for all fatalities in the plant. How- safety risk can be allocated to a single
one persons career but short enough to be ever, not all fatalities result from process hazard. One approach is to assume work-
meaningful in terms of the life of a facil- safety risks, that is from process hazards. ers are each exposed to a certain number
ity. So consider a facility with a tolerable There are still the slips, trips and falls, of potentially fatal hazards, say five, and
fatality rate of one per 100 years. the housekeeping issues, the electrical then distribute the risk budget over
XXXDIFDPN
5IFXFCTJUFGPS$1*1SPGFTTJPOBMT
XXXDIFDPNPGGFST
M"DDFTTUPUIFNPTUSFDFOUJTTVFPG$&
M$&"SDIJWFToFWFSZBSUJDMFQVCMJTIFE
JO$&GSPNUIFQBTUZFBST
M$&/FXTMFUUFST
M$&8FCJOBST
M*OEVTUSZFWFOUJOGPSNBUJPO
M"OENVDINPSF
'PSNPSFJOGPSNBUJPODPOUBDU+PIO4USB[[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 !-
i`}iV
5. Unlikely N/R N/R SIL 1 SIL 2 SIL 3
those five hazards in establishing SIL se- tially fatal hazards. In this case, however,
lection criteria. Other hazards, although senior management has declared that risk
not expected to be fatal, can then use the needs to be reduced to the level consid-
same criteria. If total process safety risk ered negligible on the ALARP diagram
is 2 x 105 fatalities/man-year, then: (Figure 2). That is, less than one fatality
N/R: 4 x 106 fatalities/man-year per million man-years.
SIL 1: 4 x 105 fatalities/man-year The portion of total risk to an indi-
SIL 2: 4 x 104 fatalities/man-year vidual attributable to process safety risk
SIL 3: 4 x 103 fatalities/man-year would again be half, or 5 x 107 fatalities
SIL 4 or re-design: > 4 x 102 fatalities/ per man-year. Once again, with each in-
man-year dividual worker exposed to about five
potentially fatal hazards, the threshold
Putting a stake in the ground for avoiding an SIL-rated SIF is 1 x 107
With the tolerable risk criteria settled, fatalities per man-year.
then the next step is to assign an SIL to Likewise, SIL 1 is less than 106, SIL 2 is
one of the cells in the matrix. For this ex- less than 105, and SIL 3 is less than 104.
ample, lets look at cell 22 in Table 5. In the CPI, anything larger than 104 would
We want to consider the risk at the mid- need to be redesigned to reduce the hazard
dle of the cell. Since the scale is logarith- before a safety instrumented function for
mic, the middle of the cell on the likelihood risk reduction was even considered.
scale is the log (geometric) mean of 1.5 Looking at the middle of the same cell,
and 15 years, or 4.7 years. The log mean of we still end up with a risk of 2.7 x 104
0.1 fatalities per event and 1 fatality per fatalities per man-year. The plant is the
event is 0.32 fatalities per event. same, so the risk is the same. Whats dif-
Doing the math, the center of our tar- ferent is the tolerable risk. In the case of
get cell has a risk of 2.7 x 104 fatalities Plant B, the tolerable risk is set at a much
per man-year. Going back to our tolerable lower level, and so in this case, the risk
risk criteria, that is less than 4 x 104, so is not quite low enough to meet the SIL
it translates to SIL 2. 3 rating:
Once we can assign SIL 2 to that cell, i>{`iV>`ii>iLii
completing the rest of the matrix is easy (1 event/4.7 years) (0.32 fatalities per }iVVivi`iii
(Table 7). The cell to the left is one order event) (1 yr/250 man-years) = 2.7 x 104 >``VviL>i`Vvi
of magnitude less risk, so it is SIL 1. Ev- fatalities/man-year ViV>>`>>ViV>`/
}>`} iiiVi ii> v
erything to the left of that is not rated.
SIL rating = SIL 4 or redesign >`i>LiVi>
Similarly, the cell below is one order of
vi>ViiViL>ii
magnitude less risk, so it is also SIL 1. As in the first example, the rest of the `}`Vvi}i
Everything below that cell is not rated. matrix is easily populated. Because of the >>`>iv>Vi
Filling in the rest of the matrix follows difference in tolerable risk, the SIL as-
the same logic. signments have shifted toward the lower Please visit us at the Chem Show. Booth #431
When were finished, we have a risk left-hand corner of Table 8.
matrix that can be used for SIL assign-
ment for all SIFs, whether or not fatal Variations to consider
consequences are anticipated. A variation to consider, depending on risk
Interestingly enough, when the risk philosophy, is what part of a cell to bench-
matrix is finished, it spells out the crite- mark to. In our examples, we took the cen-
ria for assigning an SIL to an SIF without ter of a cell. If the upper right corner is
,"9 ,L
ever explicitly acknowledging a tolerable used instead, the result will be an SIL as- ,i}>>i
risk. The tolerable risk is only implied. signment that is higher by one level. This x7iL>`ii>
is clearly a more conservative approach. /i{xn
Example B: 1 x 106 fatalities/man-yr Another variation to consider is the
vJV`>``i
Now consider Plant B. It is very similar to number of categories for both likelihood
Plant A, with the same number of exposed 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
a larger matrix, the team doing risk as- only need to be estimated to the nearest less under-specification with intolerably
sessment will probably spend more time order of magnitude, most SIL assign- high risk, and much less over-specifica-
debating categories than is warranted. By ments can be made without using de- tion with resulting misallocation of lim-
the same token, if the matrix is smaller tailed and expensive quantitative risk- ited safety resources.
than 3 x 3, it is unlikely to have enough analysis techniques. It is usually easy to Edited by Dorothy Lozowski
detail to be worthwhile. get agreement, based solely on the expe- Reference
In general, it is a good idea to extend rience of the risk-assessment team, that 1. Heinrich, H.W., Industrial Accident Prevention: A Sci-
the matrix far enough to the left to get a an event is likely to occur between 10 and entific Approach, 4th ed., McGraw-Hill, N. Y., 1959.
column of all non-rated SIFs, or else all the 100 years, rather than spend the time de- Author
SIFs, regardless of risk, will end up being termining whether it will be once every Mike Schmidt is a principal
consultant with Emerson Process
SIL rated. By the same token, there is no 37 years or once every 54 years. Management at the Refining and
reason to extend the matrix farther to the Since it is usually easy to get agree- Chemical Industry Center in St.
Louis (641 Lambert Pointe Drive,
left than a column that is all N/R, or far- ment in the middle regions of consequence Hazelwood, MO 63042; Phone:
314-872-6069; Fax: 314-872-8686;
ther to the right than is all redesign. 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
Business results achieved situations, such as hazards that are either other PHAs, consequence analy-
When a company establishes a risk expected to be very, very rare, or to have sis, facilitating LOPAs, reviewing
and preparing SRSs (SIS safety requirements speci-
matrix for SIL assignment, the matrix very, very large consequences. And even in fications), and performing SIL calculations. Schmidt
receives a fairly broad distribution. To 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
the extent that the company is uncom- 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
fortable publicizing its tolerable risk the SIL assignment of a particular SIF. CPI since 1977. The majority of his career has been in
criteria, the risk matrix approach to Finally, risk matrix is easy to apply, operating companies, including Union Carbide, Rohm
SIL assignment does not require an ex- 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-
plicit statement of tolerable risk. 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
Because likelihood and consequence project to project. This will result in much mittee on overfill protection.
MUELLER
years of proven technology with
hands-on experience to improve
your bottom line.
Your Systems Specialists!
www.jenike.com
2007 Paul Mueller Company 385-3c
a higher level
Manufacturer / waste generator
C
hemicals in production and Figure 1. Managing post-use chemicals is a vital part of chemical lifecycle man-
laboratory environments agement. Recycling waste chemicals is an option that can contribute greatly to
alike have three distinct sustainability goals
lifecycle phases: the first is
procurement and inventory storage; local teams and staff who have other challenge, when it comes to chemical
the second is use in a manufacturing primary responsibilities. Regardless lifecycle management, is to find op-
process or research program; and the of size, laboratory and production erationally efficient and cost-effec-
third is post-use, including onsite han- managers must periodically review tive ways to reduce the overall waste
dling, removal and beneficial reuse, their policies, conduct training and stream through process optimization,
recycle or disposal. audit their chemical inventories and efficient material use, tolling or sell-
A number of considerations includ- waste materials. Requirements, pro- ing recyclable materials to a second-
ing regulations, safety, material uti- cesses and regulations change, and ary market.
lization efficiency, costs and sustain- it is managements responsibility to
ability govern all of the steps in this remain in compliance, while meeting Purchasing and storage
lifecycle. The challenge for laboratory everyday production demands. Effective chemical management
and production managers is to balance Environmental sustainability con- starts before the chemicals arrive at
these sometimes competing forces to tinues to increase in importance and the site. Laboratory and operations
create an effective process that meets is embraced as a corporate goal at personnel need to work carefully with
the organizations overall needs. many manufacturing organizations. purchasing staff to order the correct
Proper chemical handling during This takes chemical management be- amounts, minimize order errors and
processing is typically emphasized in yond the traditional regulatory and ensure that proper receiving and
production facilities, but how to han- safety considerations. The chemical storage provisions are in place. For
dle chemicals post-use is often given process industries (CPI) have become operations personnel, it is a balance
less attention. While this article ex- even greater stewards of the materials between anticipating customer orders
amines the full chemical lifecycle, it throughout their entire lifecycles as and manufacturing volume in order
focuses on this post-use phase. they seek ways to further reuse or re- to eliminate production interruptions,
cycle materials that were historically minimize storage and other inventory
Sustainable lifecycles disposed of as waste (Figure 1). costs, and to avoid having to dispose
Every laboratory and chemical pro- Sustainability requires the mini- of outdated materials. Purchasing
cessor must follow environmental, mization of consumption and waste staff can consolidate orders to lower
health and safety (EH&S) proce- in order to reduce environmental im- costs, but they must have an under-
dures and regulations. This is often pact through process optimization and standing of the chemical properties
a challenge, especially for smaller reuse or recycling whenever possible. and intended uses in order to ensure
operations, since implementation fre- This can require additional record- that materials are consumed before
quently rests on individual managers, keeping and administrative steps. The they begin to degrade and negatively
44 Chemical Engineering www.che.com December 2010
affect chemical reactions or final prod- label chemicals in secondary containers with the current date when full, and
uct properties. Overall, the CPI have at their stations. Unlabeled or improp- removed within three days to the main
done a good job utilizing continuous erly labeled chemicals violate hazard- storage area (MSA).
improvement techniques to manage ous communication (Hazcom) regula- The MSA is governed by additional
their supply chains. tions. Informal practices also lead to rules. One of the most important re-
Once received, the materials must the potential for abandoned chemicals lates to allowed accumulation time
be stored in accordance with local, as researchers move on to other jobs or once materials enter the MSA. Small
state and provincial regulations, or projects. Unidentified chemicals may quantity generators (SQGs) organi-
the International Code Councils require testing, special handling and zations that generate more than 100
(ICC) International Building and In- expensive disposal if they cannot be kg but less than 1,000 kg of hazard-
ternational Fire Codes, and handled verified by the researcher. ous waste per month can accumu-
in accordance with company EH&S Consistency and training, sup- late materials for 180 days from the
procedures. In the U.S., handling ported by accurate, up-to-date la- start date.1
must follow the Occupational Health beling and recordkeeping, are key Large quantity generators (LQGs)
& Safety Admin.s (OSHA) Code of ingredients to a successful chemical those that generate 1,000 kg or
Federal Regulations (Title 29 CFR management program. more of hazardous waste per month,
1910.1200), which covers Hazard or more than 1 kg per month of acutely
Communication Standards (HCS), Post-use hazardous waste are subject to a
the federal Risk Management Pro- Handling of solvents, catalysts and 90-day accumulation rule.2
gram for certain highly toxic chemi- other chemicals can become much Regardless of size, the MSAs must
cal compounds regulated under the more problematic once they have be secured against unauthorized entry;
Clean Air Act, and the Chemical Hy- been used in a production or labora- hazardous waste and container con-
giene Standard for laboratories (Title tory process. They are no longer in tents labeling rules must be followed;
29 CFR 1910.1450). their original containers and are often containers must be inspected weekly
combined with other substances. In with reports kept on file; and pre-
Production and laboratory use the U.S. for example, OSHA, the En- paredness and prevention equipment
As materials move into the second vironmental Protection Agency (EPA), (for example an emergency phone,
phase of the lifecycle use in a pro- Dept. of Transportation (DOT) and, in alarm, fire suppression and spill pre-
duction environment or laboratory some cases, Drug Enforcement Admin. vention equipment) is required.
the potential for handling errors in- (DEA) regulations also come into play. There are additional regulations
creases. Although industrial users and A few important fundamental facts and exceptions to the rules noted
researchers are concerned with, and must be kept in mind at all times: In above. For instance, peroxide form-
trained in, safe handling procedures the U.S., all chemical storage areas ers, such as ethers and dioxane, are
for hazardous materials, they are less fall under Resource Conservation and generally managed under a peroxide-
likely to focus on storage and removal. Recovery Act (RCRA) regulations and former program that tracks the ma-
This leads to some potential problems are subject to inspection by regulators. terial based on the manufacture and
in chemical lifecycle management. To comply with that, a limited number retention dates from the day it arrives
Chemicals are generally well man- of conveniently positioned and prop- onsite until the day it is consumed or
aged in production environments erly organized satellite accumulation disposed of. SAA storage times and
since production procedures are well- areas (SAAs), under the control of the 90/180-day MSA rules are all subordi-
defined and engineered. Failure most EH&S or chemical hygiene manager, nate to the peroxide-former deadlines.
often results from human error or should be set up near production or So, if the earliest peroxide-former re-
system breakdowns. Although the research areas. tention date is 45 days in the future,
results can be catastrophic, they are There are a number of RCRA rules that is the containers deadline. Even
largely preventable through adher- for SAAs. For instance, containers the waste disposal company must
ence to established EH&S procedures, must be in good condition without track peroxide formers and follow ex-
effective maintenance practices and rust, dents, or cracks; they cannot be pedited destruction schedules based
frequent training. stored near drains or other structures on the retention date.
Research environments are less that could pose an environmental Managing chemical storage areas.
systematized, which presents greater risk; hazardous waste containers can- A minimum number of properly
opportunity for errors. Researchers not exceed 55 gal; acutely hazardous trained staff should be assigned to
often work independently on projects waste cannot exceed one quart; SAAs transport materials from the SAAs to
and they use smaller quantities of a must be inspected weekly. There are the MSA and manage the materials in
greater number of chemicals. This additional guidelines including re- all of the storage areas. Some opera-
presents an exponentially higher quirements for clear and visible labels tions are large enough to have a dedi-
number of possible chemical reactions stating hazardous waste and listing
1. http://www.epa.gov/waste/hazard/generation/
and storage issues. the container contents, as well as the sqg/index.htm
Working alone or on small teams, three-day rule that requires that the 2. http://www.epa.gov/solidwaste/hazard/
researchers often neglect to properly containers are immediately marked generation/lqg.htm
A
n added benefit of recycling and reusing chemicals can be reducing the volume
Environmental Manager 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:
cated full-time staff, while for others it Aluminum (fume or dust) The TRI listing for the following three
is a part-time function (that requires Aluminum oxide (fibrous forms) chemicals is based on the chemical activity
Ammonia (anhydrous and aqueous) rather than the form of the chemical:
the same annual refresher training).
Asbestos (friable) 1. Dioxin and dioxin-like compounds
Properly organizing the handling Hydrochloric acid (acid aerosols) (only if they are manufactured at the fa-
and in-house movement of chemicals Nitrate compounds (water dissociable; cility; or are processed or otherwise used
removes the responsibility from re- aqueous solution) when present as contaminants in a chemi-
searchers and substantially reduces Phosphorus (yellow or white) cal but only if they were created during the
training requirements. Sulfuric acid (acid aerosols) manufacture of that chemical)
Another benefit of using dedicated Vanadium (except when contained in an 2. Isopropyl alcohol (only if it is being
staff is that it puts a third party in po- alloy) manufactured by the strong acid process)
sition to regularly inspect warehouse, Zinc (fume or dust) 3. Saccharin (only if it is being manu-
production and laboratory areas to factured)
ensure that chemicals are properly 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/
stored and handled in accordance with ry2009rfi121709.pdf). See Table II-3, p. 143 for the full alphabetical list of chemicals and qualifiers
regulatory and company policies. Here
are some best practices that we have
seen in the field and that we practice Perhaps, most importantly, they may Using the same outsourcer to handle
at our own sites and for customers: not stand up to regulatory scrutiny if materials in-house as well as for re-
Ensure that proper procedures the need arises. moval and disposal eliminates inter-
(such as labeling and tracking) are Specialized software and online organizational hand-offs and can fur-
being followed at each production chemical-waste-management systems ther reduce costs.
and laboratory location, and that handle all of these issues and also Communications between waste
waste chemicals are stored only at address many other important cri- generators and chemical disposal
approved sites teria, such as chemical movements, companies. Before a waste disposal
Conduct periodic safety training for 90/180-day storage-time limits, chemi- company can remove hazardous
research and production personnel cal expiration dates and reporting. chemicals from a generator, all paper-
Establish collection schedules to ef- More sophisticated software and on- work must be in order, the materials
ficiently move materials from SAAs line tracking systems assign cost cen- must be properly labeled and pack-
to the MSA in a timely way ter allocations, and facilitate move- aged and ready to go. It is important
Ensure that SAA containers are in ment, inventorying and packing, as that chemical waste is properly char-
good condition, properly closed and well as documenting the transfer of acterized in order to avoid unneces-
placed within secondary containers, the material to a certified waste dis- sary expense or, in a worst case, addi-
and that incompatible materials are posal company. tional transshipping to another waste
properly segregated disposal site. Confusion creates lost
Catalog, store and consolidate MSA Waste disposal companies time and increased costs for chemical
materials in bulk containers and Given the complexity of effective characterization tests and EPA and
pack smaller quantities, elbow bot- waste-chemical management, many DOT documentation.
tles, and so on, for transportation companies find that it is more ef- Communications with the waste
Identify unknowns before shipping ficient and cost effective to turn to disposal company are key. Direct links
Follow up on the disposal or recov- waste disposal companies to handle between the generator and the waste
ery of chemicals for company and their chemical waste throughout the disposal company are efficient since
regulatory reporting entire post-use lifecycle phase. Service any materials or paperwork that are
Software. Most large companies companies manage the entire process not in compliance will be identified
utilize chemical inventory manage- including onsite movements, storage, early, waste profiles and DOT mani-
ment systems, while many laborato- packing and removal. In that way, fests can be generated, and placards
ries still use spreadsheets to manage generators do not have to invest as assigned before the disposal company
their inventories. Spreadsheets can be heavily in RCRA training and can rely arrives on site.
helpful, but they are not the answer on the contractor to apply its exper- A direct link to the waste disposal
for corporate-wide or comprehensive tise, provide staff and backup staff as company can also report contain-
lifecycle-management programs. They necessary, monitor regulatory federal, ers that are ready for packing and/or
are generally home-grown standalone state and local developments, and im- pickup, enabling the removal company
applications without built-in report- plement best practices. to cost-effectively schedule pickups.
ing mechanisms. They do not extend This allows plant and laboratory Automated procedures also help com-
beyond the department or integrate management to properly oversee the panies stay in compliance with RCRA
with corporate systems. They do not function without adding headcount. It 90/180-day rules or special handling
tie into industry resources and docu- also enables staff to focus on their core requirements for peroxide formers
mentation, such as material safety responsibilities and not be burdened and other acute hazards.
data sheet (MSDS) databases, that are with additional training or oversight Regardless of the degree of auto-
necessary to build chemical profiles. of chemical management functions. mation and communications, under-
46 Chemical Engineering www.che.com December 2010
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-)
Hearing Protection
For CPI Workers &NQMPZFFBVEJPHSBN
)FBSJOHMFWFMJOE#
3JHIU
a hearing-conservation
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 and the hearing of an additional nine at a chemical plant to reach 100 dB or
chemical process industries million is at risk from other agents, higher nearly 30 times the OSHA
(CPI) are many, with falls, including ototoxic chemicals, such as Action Level1.
respiratory hazards, and solvents and heavy metals [1]. Indeed, Growing evidence shows the CPI to
explosion risks confronted daily. One according to the U.S. Centers for Dis- be particularly vulnerable to hearing
hazard, however, causes no pain nor ease Control and Prevention (Atlanta, loss. Exposures to certain compounds
visible scars. Its effects accumulate Ga.), NIHL is the number one hidden (those containing lead, toluene, n-butyl
with each overexposure and can go disability [2] and the most common alcohol and carbon monoxide), alone,
unnoticed, often for decades. If statis- occupational illness in North America. have been linked to increased hearing
tics are any indication, it is the most It is also on the rise globally, with the loss. It is now known that these ex-
common permanent and preventable U.N. World Health Organization (Ge- posures also combine synergistically
occupational injury at the plant. This neva, Switzerland) calling it the most with noise to cause increased hearing
is by no means a silent hazard in common, permanent and preventable loss. Individual susceptibilities appear
fact, this hazard is noise. occupational injury in the world. quite varied, and the offending noise
What makes noise-induced hearing levels are difficult to generalize. Hear-
loss (NIHL) one of the most insidious Hearing hazards ing professionals recommend increas-
occupational hazards is that there are Noise hazards in the CPI are typically ing the frequency of audiometric test-
no outward signs of damage. Loud continuous (as opposed to intermit- ing for noise-exposed employees who
noise affects nerve cells deep inside tent or fluctuating), and can be caused are also exposed to significant levels
the ear, and once those nerve cells are by processing units, transfer pumps, of ototoxic solvents, heavy metals and
gone, they do not come back. NIHL is high-speed mixers, grinders and pul- other chemical compounds.
permanent. There is no pill, surgery, verizers, blowers, dryers and other
nor treatment to restore hearing that equipment. hearing regulations
has been lost to noise exposure. The U.S. Occupational Safety and In the U.S., OSHAs Occupational
Despite years of regulation, the prob- Health Administration (OSHA; Wash- Noise Exposure Standard and Hearing
lem of NIHL is growing. According to ington, D.C.) defines 85 dBA (decibels Conservation Amendment (29 CFR
the U.S. National Institute for Occu- A-weighted) as the noise level at which 1910.95) outlines a rigorous, worker-
pational Safety and Health (NIOSH; precautionary measures must be im- oriented program to mitigate risks of
Washington, D.C.), approximately plemented, or the OSHA Action Level.
30-million workers in the U.S. are ex- It is not uncommon, however, for noise 1. Noise is measured using a logarithmic decibel
(dB) scale. A 3 dB increase represents a doubling
posed to hazardous noise on the job, levels in close proximity to equipment of the sound energy
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.
Circle 65 on p. 78 or go to adlinks.che.com/6893-65
38 Chemical Engineering www.che.com March 2007
A Safe Process
throughout process
design lowers the risk of
a hazardous event
Phil Leckner
CH2M Hill Lockwood Greene
A
ccidents do happen. While not
everything can be predicted, 'FFE %JTUJMMBUJPO -1TUFBN
addressing safety concerns
throughout the design of a pro- 8BUFS
cess can help to prevent accidents $POEFOTFS
from occurring. Designing with safety
)1TUFBN
in mind can also help to minimize po- 7FOU
$POEFOTBUF 3FCPJMFS
tentially serious consequences that
would result if an accident did occur.
On April 12, 2004, toxic allyl alcohol 3FDFJWFS
#PUUPNT 1VNQ
and allyl chloride were released from
a reactor at a facility in Dalton, Ga.
The consequences included injuries &YBNQMFPGBCMPDLGMPXEJBHSBN %JTUJMMBUF
and chemical contamination to people
Figure 1. Block flow diagrams give a
and property in the surrounding area. simple overview of the major steps in a process
According to their report [1], the U.S.
Chemical Safety and Hazard Inves- Dalton, Ga. and Texas City, Tex. This cally used in the CPI. Equipment is
tigation Board (CSB) concluded that methodology applies to process de- usually identified and shown with an
better process design, engineering, signs throughout the chemical process alphanumeric designation. The PFD
and hazard analysis would likely have industries (CPI), which include not includes major and some minor pro-
prevented the 2004 runaway chemi- just chemical production, but also, for cess streams as well as utility streams,
cal reaction and toxic vapor cloud re- example, wastewater-treatment fa- such as steam, condensate and cooling
lease... cilities, pharmaceutical and food-and- media. This diagram can also be used
On March 23, 2005, an explosion at beverage plants. While following these to show process safety requirements,
a refinery in Texas City, Tex. killed 15 guidelines will cost time and money, such as proposed locations for relief
workers and injured 180 others when the practice can be a very inexpensive valves. Often, a heat-and-material
flammable liquid and vapor overfilled way to help prevent the much more balance and major control loops are
a blowdown drum during the startup costly consequences of not providing included.
of the refinerys isomerization unit [1]. the safest design possible. The PFD should be used as the basis
All of the fatalities and many of the for generating the more detailed pip-
injuries occurred in and around trail- Basic Engineering ing-and-instrument diagram (P&ID).
ers that had been positioned near the Process flow diagrams Prior to P&ID preparation, the PFD is
isomerization unit to support mainte- As a process engineer, there are two reviewed by the design team and is-
nance activities on adjacent refinery types of flow diagrams that interest sued for design (IFD), with a revision
units. The CSB report on this incident me: the block flow diagram (BFD) and number of zero.
[1] recommended that new guidelines the process flow diagram (PFD). At a minimum, the basic design
be developed for the placement of The BFD (Figure 1) presents an package should include the BFD and
these and similar temporary struc- overall picture of the process, show- the PFD. However, I believe that an-
tures around hazardous areas. ing only major process steps. These other document, which takes the PFD
Having a procedure in place to pro- steps are shown as black boxes with to a new level, better promotes a safe
mote safe process design can help simple descriptions. Equipment can process design. This document, which
ensure that safety concerns are con- be depicted singly or grouped together I call the process definition drawing
sidered at appropriate phases in the as a system. (PDD), is not ordinarily a part of the
design. The methodology presented The PFD (Figure 2), meanwhile, basic design package, but is a great
here, as outlined in Table 1, may help depicts major and minor equipment tool for the process design engineer.
prevent accidents such as those in with specific symbols that are typi- The PDD includes operating-and-
30 Chemical Engineering www.che.com December 2006
'FFE "
%8(/P 1*$
5*$
'-
'*$ -*$
R 5
1
1 5
)14 /
'*$ 5
YMCIS
)& #PUUPNT
-*$ %8(OP
5
)1$
Figure 2. Process flow dia-
%JTUJMMBUF grams give more detail than
%8(/P BFDs, and include at least
major process streams and
equipment items. Symbols that
1 are typical for the CPI are used
design conditions for each equipment ment that provides a wealth of in- cerns in the design safety review.
item, each control-valve station and formation including the proper han- Using the most up-to-date version
all relief devices. The benefit of this dling of a substance, special storage of the PSR, all features associated
document is that the process engineer requirements (such as keep out of the with safety, environmental and layout
can quickly recognize inconsistent or sun) and required personal-protection issues are reviewed, including the fol-
conflicting operating-and-design con- equipment (PPE), such as breathing lowing:
ditions. It helps the engineer think apparatuses. Defining the hazardous location
about the process in terms of how The MSDS also provides basic phys- classification (HLC), or the electri-
it will operate and what needs to be ical-property and toxicity data, expo- cal area classification as it is some-
done to make it safe. sure limits and flammability ranges. times called The HLC is used to
Detailed engineering work is also It may describe what to do in case of determine electrical-design criteria,
aided by the PDD, since much of the spills. The manufacturer or supplier such as equipment that may require
information included on this docu- of the raw materials and various web- explosion-proof motors. The HLC
ment will be used in the generation of sites on the Internet are sources for boundaries should be shown on ap-
process-equipment, instrumentation MSDSs. If the facility is producing a propriate documents such as the
and safety-device duty specifications. finished product, then the plant owner PFD or equipment-layout drawings
The PDD is a living document and will have to develop an MSDS for (if available). One source of defini-
will change as the process design pro- that product and make the document tions for HLCs can be obtained from
gresses. As such, it is often sketched available to potential users. Table 2 Ref. [3].
using rudimentary drawing software lists information typically found in an Locating major pressure-relief de-
rather than CAD, thereby making it MSDS. vices, such as relief valves and rup-
simple to construct and maintain. Once the documentation is gath- ture disks, explosion panels and
ered, the PSR is assembled into a flame arrestors These devices
Preliminary safety review formal report and issued to all perti- should be indicated on the PFD
After the PFDs are created and ap- nent members of the design team for and, in more detail on the PDD. The
proved for design, a preliminary safety comment. The report is given to proj- destination of the vent from these
review (PSR) is undertaken. ect management for distribution as a particular safety devices needs to
The PSR is the gathering of docu- revision 0 issue. Note that the PSR be considered. The applicability and
mentation with an emphasis on pro- is a living document and is subject to use of alternate safety systems, such
cess safety. Included are items such as change as the design progresses. The as safety-instrumented systems
the project scope definition, the pro- PSR document is issued as soon as (SIS) is also evaluated at this time.
cess design basis, the process descrip- feasibly possible because it will form (For more, see Part 2 of this report,
tion and a material safety data sheet the foundation for the balance of the p.34)
(MSDS) for each substance used. An safety review of the project. Evaluating the layout with respect
extensive list of documentation that to minimizing hazards As noted
may be included in the PSR can be Design safety review earlier in the Texas City, Tex. ac-
found in Ref. [2]. Once the PSR is issued, the process cident, personnel placement is a
The MSDS is an important docu- can be reviewed for major safety con- very important consideration. For
Chemical Engineering www.che.com December 2006 31
Table 1.
equipment placement, the American review, all documents and results are A methodology to promote
Petroleum Institute (API; www.api. collated and issued. This compilation a safe process design
org) recommends that all equipment is included as part of the front-end de- I. Basic process engineering
with a ground area of 2,500 to 5,000 sign package and the design proceeds
ft2 should be considered part of the to the next phase of the project, the Create the process flow diagram
same fire-relief scenario [4]. Deter- detailed process engineering. Perform a formal review of the pro-
mine if larger-sized equipment can cess flow diagram
be moved outside the common fire Detailed Engineering Conduct a preliminary safety re-
zone to reduce overall relieving ca- Piping and instrument diagram view
pacity. Also be cognizant of where At the start of the detailed process- Perform a design safety review
chemicals are stored in proximity to engineering phase of design, all out-
each other. You may not want acids standing issues and especially those II. Detailed process engineering
stored near bases, for example. brought up during the design safety Create the piping and instrumenta-
Locating flares in safe areas, taking review are examined and addressed. tion diagram
into account their radiation effects The P&IDs, PFDs and the PDDs are Conduct a formal P&ID review
at ground level. updated as required. The P&IDs are
Perform the process hazard analy-
Evaluating the need for and place- then reviewed in a formal setting as sis on the process
ment of, fire and gas detectors (for a team, which should include the pro-
Perform the process hazard analy-
more, see p.18). cess, mechanical, and piping and in- sis on the control system
With the major pressure-relief de- strumentation engineers.
vices located and the operating and The review will expose any last min- III. Implement a management of
design conditions fully defined, it is ute safety and design issues that must change procedure
appropriate to evaluate modifications be addressed before the more detailed,
that might make the system safer. and required, process hazard analysis Facilities that do not fall within 29 CFR
For example, determine if there are (PHA) is undertaken. Changes are 1910.119 would still benefit greatly
safety advantages to changing stor- documented by again updating the from the PHA if not for the safety as-
age requirements from concentrated P&IDs and PFDs. I suggest that these pects of the process, then for process
solutions to more dilute solutions, or documents be issued with a separate operability. I cant stress this enough:
vise-versa. Evaluate the advantages revision number established exclu- not being required to do a PHA by law
and disadvantages of breaking unit sively for the PHA. should not exclude you from doing one
operations into smaller, more discrete to ensure that your process is indeed
pieces to make process equipment, The process hazard analysis safe in design and operation.
such as heat exchangers and reactors, The PHA evaluates the design in The team make-up. The PHA team
smaller. terms of both safety and operability. should include a facilitator and a scribe
Preliminary hazard analysis. Fi- The analysis should be performed on (who serves the function of record-
nally, some type of preliminary hazard the process, as well as instrumenta- ing secretary), the design firms area
analysis (PrHA) should be undertaken tion and control systems, such as the process engineer, the plant process
before moving into the detailed pro- digital control system (DCS). engineer and representatives from
cess engineering phase of the project. PHAs are mandatory for all plants the plants safety-and-environmental
During a PrHA, team members vi- that fall within the scope of OSHA 29 and operations-and-maintenance de-
sualize ways in which a process design CFR 1910.119 [5]. This scope applies partments. If vendor packages are in-
can malfunction or be operated incor- to plants that meet the following two volved, a vendor representative is also
rectly. The PrHA can take one of many criteria: advised as a team member. An exten-
forms, such as a pre-HAZOP, a what 1. Those whose processes involve one sive list of possible participants can be
if, failure mode-and-effects analysis or more of certain chemicals (listed found in Ref. [2]. The PHA team make-
(FMEA) or FMEA check list. The vari- in appendix A of the regulation), and up and the extent of their efforts will
ous types of hazard analyses that are in quantities at or above the thresh- vary based on the nature and complex-
acceptable to the U.S. Occupational old given ity of the process design.
Safety and Health Administration 2. Those whose processes involve flam- The facilitator should be chosen
(OSHA) are outlined in Refs. [2] and mable liquid or gas onsite in one lo- with the following criteria in mind:
[5]. The PrHA is performed on the de- cation in a quantity of 10,000 lb or The person should be knowledgeable
tailed PFD that is issued for design. more with the exception of: in the type of PHA to be performed
The documents collected during the a. Hydrocarbon fuels used solely for The person should not be intimately
PSR and the PDD provide reference workplace consumption as a fuel involved in the process design (thats
materials. Note that a preliminary or b. Flammable liquids stored in at- what the process engineers are for)
IFD issue of the P&ID may have been mospheric tanks and kept below The person does not necessarily need
developed by this time and if so, would their normal boiling point with- to be very knowledgeable about the
be part of the PrHA. out the aid of chillers or refrigera- particular process to be reviewed
At the conclusion of the design safety tion The facilitators function is to guide
32 Chemical Engineering www.che.com December 2006
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 explanation of the fire risk is offered Besides classical process plants in
requires the understand- and approaches that can minimize ig- the chemical, petrochemical and plas-
ing and application of the nition potential and property damage tic industries, two other applications
fundamental tenets of safe are addressed. have recently come into the focus by
design, construction, and operation utilizing large volumes of DP:DPO
of each facility. In that context, many Introduction HTFs (see box entitled Large volume
years of practical experience with sys- Basics of system requirements applications): electrical energy pro-
tems that use very high-temperature Process plants require heating and duction by concentrating solar power
organic heat-transfer media has led usually cooling capabilities. A typical (CSP) and converting natural gas into
to much knowledge regarding de- HTF system is composed of an energy synthetic oils in a process called gas to
sign features to adopt and those that source, such as fired heaters or waste liquids (GTL).
should not be repeated. heat recovery systems; pumps to force As with all HTF systems, the de-
For the media itself, the severe the fluid flow; an expansion tank to sign must accommodate the volume
stresses of extreme operating temper- absorb the volume expansion of the expansion of the heating fluid, which
atures involved can sometimes nar- fluid; and last but not least, a heat is typically provided by an expansion
row the number of candidate fluids to consumer. High-temperature heat- or surge tank of sufficient volume and
one: the eutectic mixture of diphenyl transfer systems are usually closed headspace. For plants using DP:DPO
and diphenyl oxide (DP:DPO; Figure systems, and hence, a release of fluid near its maximum temperature of
1). This article shares best practices can typically only occur in case of ac- 400C (750F), the volume expansion
that have been developed over the cidents or malfunctions. Heat transfer from ambient temperature is greater
years to ensure the safest possible use media used in such systems are usu- than 30% [1]. The expansion vessel is
of this heat transfer fluid (HTF). Ef- ally water-and-steam, water based typically pressurized to keep the HTF
fective methods to identify and deter- fluids, mineral oils or synthetic heat- in liquid phase and is equipped with
mine the order of magnitude of leaks transfer fluids specially designed for a vent system to permit the exit of
are discussed. Meanwhile, a review of this purpose. Based on their physical nitrogen, degradation products, and
the concepts necessary for effectively and chemical properties, such fluids a portion of DP:DPO into the ullage
maintaining a leak-tight system is imply certain handling risks in addi- vent-collection system. The extensive
presented. And, for occasions in which tion to the general risk of operating an network of HTF handling piping, in-
unexpected leaks do occur, a thorough HTF system. struments, and vessels, combined with
26 Chemical Engineering www.che.com December 2010
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.
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
Figure 1. Extreme operating temperatures some- is a cracking and conversion of the hydrocarbons into the desired final
times narrow the number of candidate fluids to one: products [2]. Due to high temperature requirements of some of the process
the eutectic mixture of diphenyl and diphenyl oxide steps, DP:DPO heat transfer fluid is utilized for process heat control.
(DP:DPO)
elevated vapor pressures, can increase remain in service today, while most has maintaining adequate sealing. Some
the potential for leaks from systems been replaced over time. Experience in users, however, have reported that
if adequate design and maintenance the mode of failure finds that the hose some of the rotary joints failed more,
measures are not also incorporated. develops small cracks through which and without warning.
the DP:DPO fluid seeps slowly at first. Whether from flexible hose or rotary
Incident history of leaks, fires The leakage can be visually observed joints, the close proximity of these de-
Examination of the leak history from as a darkened area on the outside of vices to nearby ignition sources can
DP:DPO handling systems indicates the hose assembly. Prompt mainte- lead to the ignition, or even autoigni-
that the primary sources are flanged nance can permit the piping section tion of a released cloud of HTF mist if
connections, flexible connectors or to be shut down and isolated for hose surface temperatures exceed 1,100F
rotary joints, and pump seals. Fire- replacement prior to a rupture. When (593C). For safety of personnel, re-
resistant gaskets used in DP:DPO proper experience and care is applied sponse typically involves remote isola-
service can require substantial com- during hose installation, the hose can tion of the piping circuit to allow the
pressive force on the gasket face. In be kept free of the torsion and mis- fire to burn itself out in remote or non-
cases where insufficient flexibility is alignment that reduce its service life. congested areas. Alternate measures
provided in piping networks (Figure It is important to follow hose manu- may be appropriate to protect against
2), the resulting force applied to the facturer recommendations on inspec- peripheral damage in more congested
flange pair reduces compression on a tion and replacement frequency, as the areas. Extinguishing the fire prior to
portion of the gasket, leading to leak- hose has smaller wall thickness than stopping the HTF release may result
age. Such leaks can sometimes be ad- rigid pipe. in later, repeated ignition. Operating
dressed by tightening sufficiently, but Rotary joints have the advantage of practices must incorporate the fre-
only as a temporary repair. Permanent using greater wall thickness for a more quent monitoring of the integrity of
repair will require gasket replacement robust joint, yet have the flexibility to field piping systems through regular
and improved piping flexibility. support the daily cycling demands of inspections and/or remote video sur-
Stainless-steel flexible hoses were the mirrored rows of heat collection veillance, or both.
first installed on the early solar en- elements (HCEs). Designs can incorpo- Pumps in DP:DPO service can have
ergy generating systems (SEGS) in rate high-temperature, metal-to-metal single or double mechanical seals.
the California Mojave Desert in the seals and can include provision for Some of the pumped liquid may be
1980s. Some of the original flex hose injectable, graphite-based packing for required to adequately lubricate seal-
Chemical Engineering www.che.com December 2010 27
Cover Story
Fo
can also vaporize the HTF resulting
rce
Reduced
in no lubrication and ultimately cause n compression Fo
rc
io
mechanical damage to seal face mate- u lat of gasket n e
Ins io
ct
rials. This mechanism can create par- fle
De
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 Fixed restraint
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 Figure 2. Poor piping flexibility can cause leaks
lower relative-normal boiling point
than other high temperature HTFs, Table 1. DP:DPO Key Fire-Related Properties
permitting its more rapid evaporation
from soaked insulation before it can Characteristic Method C F Vol. % in air
be heated to its autoignition tempera- Flash point (open cup) ASTM D-92 124 255 ---
ture (AIT) by the oxidation exotherm
within porous insulation. Cellular Fire point ASTM D-92 127 260 ---
glass insulation has proven most ef- Flash point (closed cup) ASTM D-93 110 230 ---
fective in interrupting the mechanism
Autoignition temperature ASTM E-659 601 1,114 ---
of HTF insulation fires among most
organic HTF chemistries. Lower combustion limit --- 149 300 0.8
260 500 0.5
Consequences of leaks, fires Upper combustion limit --- 149 300 3.3
There are no positive consequences 260 500 6.2
of leaks and fires, except perhaps the
heightened awareness of design or
construction flaws, which leads to im- ments, possibly including incineration Fluid Chemistry and
provement. The immediate negative or bioremediation. Hot fluid released Inherent Properties
consequences include risks to human and collected should not be reused, as Formulation
life and health from released fluid it will have become oxidized and per- DP:DPO heat transfer fluids are a eu-
ignition and fire fighting response, haps contaminated. Equipment wet- tectic mixture of diphenyl and diphenyl
or inhalation of partially combusted ted with HTF condensate should be oxide. These two components exhibit
hydrocarbons. Personnel should be cleaned for proper housekeeping, and the highest thermal stability available
trained and equipped to respond by to remove residue that could poten- among organic heat-transfer media.
remotely isolating systems, and stay- tially fuel a future fire. Their eutectic ratio ensures the lowest
ing out of the path of smoke, vapors In areas where leakage is of high- freeze point possible, thereby creating
and liquid runoff. In CSP plants, the est risk (such as flex hose or rotary the widest working range of tempera-
amount of fluid contained in a 100-m joints, pump seals, valves or instru- ture. As diphenyl and diphenyl oxide
mirrored row with 70-mm O.D. heat- ment manifolds, and so on) efforts exhibit tremendous thermal stability,
collection elements will be limited should be made to limit the amount the weakness of the specific fluid formu-
to nearly 0.33 m3 (87 gal), plus the of instrumentation and wiring poten- lation would then lie in its impurities.
volume contained in piping up to the tially exposed to fire. Doing so will Therefore, it is most desirable to have
isolation valve(s). Much of the HTF minimize the downtime required to very high assay of the two primary con-
released from piping containment return the equipment into operation. stituents, and limit the percentage of
will flash into the vapor phase, and Any equipment, piping, instrumenta- impurities possible. Doing so protects
then quickly condense into a mist tion, insulation or controls that have thermal stability of the fluid and also
cloud. The balance will rapidly cool been exposed to fire or DP:DPO spray ensures that the physical properties of
in contact with equipment and earth, should be inspected by qualified engi- the HTF will have minimal deviation
remaining in liquid phase. Contami- neers and technicians for reliable op- from its key components.
nated soils will require disposal or erating integrity, prior to returning it Low chlorides content (< 10 ppm)
treatment per regulatory require- into service. ensures long life of stainless-steel
28 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
Figure 3. Pipe flanges are potential sources of leaks. Design considerations are potential. It is important to note that
important here any significant extension of operating
temperatures above those currently
system components without exces- mechanical boundaries, tremendous used in DP:DPO HTF systems would
sive risk of stress-corrosion cracking. pressure can result, possibly leading require more exotic and costly alloys
There are SEGS systems, for example, to release of the product through the for piping and equipment, as well as
that have been utilizing low-chloride- weakest constraint, such as flanged the development of satisfactory, new
content heat-transfer fluid since the connections, valve stems, pump seals, heat-transfer fluids. Useful reference
1980s with no reported cracking is- and so on. When thawing a frozen sec- standards and codes are included in
sues. Low chlorides content should tion of piping or equipment, it is very the References section [714].
therefore be a requirement of DP:DPO important to accommodate the expan- With the HTF being significantly
fluids used in plants to help maintain sion in volume into unobstructed pip- above its normal boiling point during
longterm mechanical integrity. ing or equipment. operation, adequate overpressure pro-
tection in the form of pressure relief
Tendency toward leakage Key fire-related properties valves (PRVs) and rupture disks must
Operators of high temperature, or- Key properties of interest in gauging be provided with adequate capacity
ganic heat-transfer fluids should be the potential for fire with organic heat for relief between selected points of
aware of the potential for leakage of transfer fluid are provided for DP:DPO isolation. With proper design consid-
DP:DPO from components that may in Table 1. erations, the PRVs can be installed in
be leak-tight when pressure tested Minimum ignition energy (MIE) series with rupture disks to minimize
with water. It has been noted that, is a measure of an amount of energy fugitive emissions from the system.
the physical and solvent properties below which an explosive mixture in General good practices for HTF system
of organic fluids allow the fluids to air will not be affected, and above design are typically available from the
penetrate ordinary valve and pump which the mixture can be affected. HTF manufacturer, including informa-
packing and bleed through porous ma- Measured values for heat transfer tion on system components, filtration
terials including some cast irons [4]. media are not readily found in pub- and expansion tank design.
For piping flanges, fire-resistant gas- lished literature. As T. H. Pratt ex-
kets are preferred, which can require plains, the MIE concept applies only Piping, flanges and gaskets
higher compression to adequately seal to capacitive spark discharge and Seamless carbon steel has been demon-
as compared to softer gasket materi- that, based upon numerous factors, strated as an appropriate piping mate-
als. To help minimize leak potential one must recognize that MIEs vary rial for use with organic HTFs in plants
and keep the DP:DPO confined within with most everything [5]. Taking this up to their maximum bulk operating
the system, engineers and safety pro- into account, NFPA 77-2007 states temperatures. Threaded fittings in pip-
fessionals have developed key prac- that the MIE for most saturated hy- ing are to be avoided in preference to all-
tices as outlined later in the section on drocarbon vapors is near 0.25 mJ, and welded construction. Graphite-based,
preventing leakage. that mists can have MIE values one paste thread sealants have demon-
When DP:DPO eutectics freeze, or two orders of magnitude higher [6]. strated marginal success with threaded
the material contracts in volume by This value can have considerable de- connections that cannot be avoided,
over 6%. If the product then melts be- viation from the MIE experienced in such as instrument connections, pump-
tween frozen plugs of product or other actual field environments. casing drain plugs, and so on.
Chemical Engineering www.che.com December 2010 29
Cover Story
With any metallurgy employed, the high temperature service can have the failure, and not just the conse-
linear expansion and contraction of double mechanical seals, or can be of quences of the failure
the piping must be accommodated by a sealless design. The pumped fluid is 3. Learn, document, and practice the
use of expansion loops or flexible con- used to lubricate and cool bearings in recommended preventive mainte-
nection members, and pipe supports either type. Excessive temperatures nance of the equipment
should be generously spaced to pre- at mechanical seal faces (typically
vent sagging. When thin-walled expan- tungsten carbide, stellite, or silicon Leak Detection
sion joints are selected, manufactur- carbide) can vaporize the HTF result- Human senses
ers recommendations for inspection ing in no lubrication and mechanical DP:DPO fluid can be operated in liq-
and replacement frequency should be damage to seal face materials. This uid or vapor phases. When there is
strictly followed. Piping should be pro- mechanism can create particles that leakage, even above its normal boiling
vided with shoes, which can slide un- can erode the seal face and create a point, the vapor emitted can quickly
restrained axially on the pipe supports separation of seal faces resulting in condense to form a visible, near-white
with thermal cycles. Where flanges are leakage. Therefore, stuffing box and mist cloud. The cloud quickly increases
necessary, Class 300 and Class 600 seal gland cooling is important to its transparency as it dissipates. For
ring-joint flanges or raised-face flanges maintain lower temperatures and also small leaks it may be possible to ob-
are commonly used. While the number improved lubricity of the HTF. A seal serve liquid droplets present at the
of flanged pairs should be minimized flush can be effective in minimizing source of the leak, such as valve stems,
to reduce the number of potential excessive seal face wear and particle pump seals, flanges, and so on, as well
leak points and fugitive emissions, accumulation on flex metal bellows by as on the ground. As mentioned ear-
designers should consider maintain- removing harmful deposits with fil- lier, a small leak from a flex hose can
ability and install such connections to tered fluid [15]. appear as a darkened, wetted area.
all major pieces of equipment and in When cool ambient conditions permit,
other key locations for ease of servic- Valves frozen DP:DPO may be observed as
ing equipment (Figure 3). Valves in DP:DPO service may in- white crystalline solid. Visual detec-
When used with raised-face flanges, clude forged- or cast-steel, or stain- tion capability for leaks can be greatly
gaskets should have a metal ring for less-steel bodies, balls, plugs and enhanced by use of remote video-sur-
blow-out resistance and graphite-filler disks. 13-Chrome (Cr) stems and veillance monitoring.
within 316 stainless-steel, spiral wind- 13-Cr or stellited seats should be The DP:DPO eutectic mixture
ings for fire resistance. A number of considered. Fire-resistant graphite has been measured to have an odor
manufacturers offer such gaskets, and packing should be specified in select threshold of 9 parts per billion (ppb) in
some designs provide more-exotic-alloy configurations available. Bellows seal air, making detection by odor possible
spiral windings for improved memory designs can provide physical barri- without exceeding established airborne
and longer use-life for cycling tem- ers for reduced emissions and leaks. exposure limits for either component.
perature services. A key necessity for With frequent updates of valve fea- In indoor areas, the airborne HTF may
flanged connections is to ensure that ture availability, the designer should not dissipate readily, preventing easy
adequate and uniform sealing compres- always consult with the valve manu- tracing back to the leak point. In out-
sion (seating stress) is provided on the facturer to consider the latest inno- door environments, the odor is more
gasket faces. Data on compression force vations available. Soft-seat materials closely localized around the leak point
of gasket surfaces are available from should be avoided since they can burn due to the more rapid dissipation of
gasket manufacturers, which will then out in case of fire, potentially adding the components in air. While odor may
determine proper bolt/stud/nut selec- to the complexity of the HTF release. be a sensitive detection technique, the
tion and torque requirements. Small valves with welded end-con- use of instrumentation is a preferred
For non-circular gaskets, such as nections should be considered to re- detection-and-measurement method
in some pump casings and valve bon- duce potential leak points, and larger with superior quantifying capability
nets, foil-inserted graphite has been valves should be considered with for leak tracing.
successfully used. Where larger gas- flanged end-connections. Connecting
kets are required for heat exchanger piping should have adequate flexibil- Specialized instrumentation
heads with partitions, and so on, foil- ity to minimize torque induced from Instruments using the principal of
inserted graphite, tang or corrugated thermal expansion and contraction of photoionization are well suited for use
metal-inserted graphite can be used. piping applied to flanges, which may in CSP and process plants. Typically,
With any gasket type selected, proper develop leaks. these instruments can readily detect
installation techniques prescribed by DP:DPO concentrations in air as low
the manufacturer should be followed. Maintenance practices as the ppb range. For realtime per-
Simply put, three keys for adequate sonnel exposure monitoring, the ppb
Pumps and pump seals maintenance practice are as follows: sensitivity is appropriate to quantify
Pumps in HTF service are gener- 1. Respond to identified leaks promptly exposure at levels below established
ally of cast steel for withstanding the with repairs occupational exposure limits for the
stresses of thermal shock. Pumps in 2. Repairs should address the cause of fluids components. For maintenance
30 Chemical Engineering www.che.com December 2010
Water fog mist suppression system
Fog nozzles
EN
HE
each of these successfully to avoid and incorporating a culture of
YG
their fiery combination? effective preventive mainte-
AT
OX
Oxygen. Within process vessels, pro- nance and inspections for the
vide blanketing with dry, inert gas. Ni- systems proper operational
trogen is typically used, and should be integrity. For the unexpected
a minimum of 99% purity. leaks, designs incorporating
Ignition sources. HCEs of CSP remote isolation capability can FUEL
plants are the previously identified help minimize DP:DPO release
Figure 5. Oxygen, an ignition source and fuel
point of AIT concern, as the surfaces and personnel exposure during (DP:DPO) are the three necessary components of
can reach temperatures above the these events. the fire triangle for an HTF fire scenario
DP:DPO AIT, particularly when leak-
age reduces the flowrate of the HTF Fire fighting response cialized training and certification
through the HCE. To date there has For small or incipient-stage fires with prior to being authorized to respond.
been no known provision made to avert a limited and minimal source of fuel Onsite fire brigades can address fires
the potential drifting of a DP:DPO (DP:DPO), the use of a handheld fire much more rapidly than offsite fire de-
cloud across the HCE. Upon the first extinguisher can be an effective and partments, and can help to minimize
detection of leakage from piping com- appropriate selection to extinguish potential damage, system downtime
ponents in the proximity of potential a fire. Approach to the fire should and repairs.
ignition sources, prompt maintenance be from upwind to avoid breathing As mentioned earlier, fires involv-
should be scheduled for repair or re- partially combusted materials, with ing larger releases of HTF in non-con-
placement. Potential for open flames standard sweeping discharge of dry- gested areas of a facility that are be-
(such as smoking, welding and cut- chemical or carbon-dioxide media. yond the capability of incipient-stage
ting torches) can be addressed by Consider placement of 2030 lb (914 response techniques are perhaps best
strict compliance with adequate ad- kg) dry-chemical fire extinguishers in managed by remote isolation of the
ministrative controls. Electrical area easily accessible locations at each level DP:DPO source, and allowing the fire
classification design should be prop- of structure where the fire potential to burn out without risk to employees.
erly addressed so that it conforms to exists in HTF handling areas [12]. Runoff of liquid should be directed
applicable code requirements. This Personnel must be properly trained in away from areas that might result in
should establish a minimum radius the operation of selected extinguishers costly damage, extensive downtime
around each potential DP:DPO leak and techniques for approaching and and repairs. Water monitor stations
point, and should be determined ac- extinguishing HTF fires. equipped with water fog nozzles may
cording to specified standards for the Larger dry-chemical units are avail- provide needed protection and cooling
requirements of electrical enclosures, able on trailers and can be quickly of equipment in certain areas, which
switches, motors and so on. transported to distant locations within should be assessed in planning for a
Fuel (DP:DPO). Use of HTFs at very the expanse of a facility. It is recom- fire-water management scheme.
high temperatures often requires an mended that personnel intended to Indoor areas may require sprinkler
organic heat-transfer fluid of high- operate such equipment receive spe- protection when potential leak points
References
1. Therminol VP-1 Heat Transfer Fluid Vapor 7. NFPA Standard 30, Flammable and Com- 13. National Board Inspection Code, National
Phase / Liquid Phase Heat Transfer Fluid, bustible Liquids Code, NFPA, 2008. Board of Boiler and Pressure Vessel Inspec-
Pub. No 7239115C, Solutia Inc. tors, Columbus, OH, 2007.
8. ASME B31.1 Power Piping, American
2. Apanel, George J., Process Economics Pro- Soc. of Mechanical Engineers, New York, NY, 14. API 510, Pressure Vessel Inspection Code,
gram Report 247, Gas To Liquids Update, 2004. Am. Petroleum Institute, Washington, D.C.,
SRI Consulting, 2002. 2006.
9. ASME B31.3 Chemical Plant and Petro-
3. Systems Design Data, pp. 4.3 4.4, Pub. leum Refinery Piping, ASME, New York, 15. Systems Design Data, Pub. No. 7239193
No. 9239193, ver. C., Solutia Inc. NY, 2004. ver. C, Therminol Heat Transfer Fluids, So-
4. Organic Heat Transfer Fluids and Equip- lutia Inc.
ment, Ind. Risk Insurers, IM.7.1.5, p. 1, 10. ASME Code Sec. VIII Div. I Pressure
Vessels, ASME, New York, NY, 2004. 16. Vincent, G. C., Nelson, R. C., Russell, W. W.,
June 1, 1975. Hydrocarbon Mist Explosions Part II, Pre-
5. Pratt, Thomas H., Electrostatic Ignitions of 11. ASME Code Sec. VIII Div. II Pressure vention by Water Fog, Loss Prevention, (Vol.
Fires and Explosions, Wiley-AIChE, Center Vessels-Alternative Rules, ASME, New 10), American Institute of Chemical Engi-
for Process Safety, July 15, 1997. York, NY, 2004. neers, 1976.
6. NFPA 77, Recommended Practice on Static 12. Loss Prevention Data Sheets 7-99, Organic 17. Britton, L. G. , Spontaneous Fire in Insula-
Electricity, National Fire Protection Assoc., Heat Transfer Fluids, Factory Mutual, tion Plant/Operation Progress (Vol. 10, No.
2007. 2009. 1), pp. 33, Jan., 1991.
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.
W
Figure 1. Many self-testing ground-
e are all familiar with the bustible liquids. One employee and ing clamps also monitor cable condi-
effects of static electricity in one firefighter were injured. The main tion back to the designed ground point,
everyday life the sharp warehouse structure was destroyed, and will alert the operator if the cable
spark we experience as we the operators business was signifi- has worked itself loose or is broken
approach a metal file cabinet or the cantly interrupted, and surrounding
crackle we feel as we remove a garment businesses were evacuated. tance of the object, this may result in
made of synthetic material. While these significant levels of energy being avail-
effects are largely harmless, in the Best practices able for discharge well above the
chemical process industries (CPI), Anyone responsible for, or working minimum ignition energy (MIE) of the
whenever flammable or combustible in, potentially explosive atmospheres surrounding flammable atmosphere.
atmospheres are present, uncontrolled (Class I, II or III) is aware of the fact Typical MIE values (Table 1) vary ac-
discharges of static electricity can be that all potential sources of ignition, cording to whether the flammable at-
potentially disastrous. including naked flames, hot surfaces mosphere comprises vapor, dust or gas,
The National Fire Protection Assn. and electrical sparking must be elimi- but many commonly used solvents and
(NFPA; Quincy, Mass.; www.nfpa.org) nated. Today, a variety of safety codes other flammable chemicals have MIEs
tracks the causes of all industrial and guidelines (discussed below) spec- that are relatively low on the order
fires and explosions reported to U.S. ify safe working practices in hazardous of 1 mJ or less. If the isolated conduc-
fire departments. Roughly 280 inci- areas, and provide specific guidance tor then comes into proximity with an-
dents/year in the U.S. are attributed on how to control static electricity. other object at a lower potential, much
to static electricity, and these result The risk of static electricity is ever- of this energy could be released in the
in significant damage to property, lost present, as it is generated continu- form of an incendive spark.
production, injuries and even fatali- ously, whenever surfaces come into Of course, in order for there to be
ties. According to the U.K. Institute contact and then separate. Examples static ignition of the flammable atmo-
of Chemical Engineers, roughly 350 include liquids flowing through pipe- sphere, there would need to also be a
incidents per year are caused by static lines or filling into drums and tanks, suitable concentration of fuel (vapor,
electricity. And this figure is probably powder dropping down a chute, and dust or gas) in air, but for the purposes
conservative, as more such incidents even a person walking across an insu- of safe plant design, the very fact that
likely occur but go unreported, or do lating floor. there is an identified flammable atmo-
not have an electrostatic cause sub- The extent of charge generation in sphere should suggest that this is pos-
stantiated during the investigation. industrial processes is usually very sible or likely.
low, typically no greater than 1 Despite these potential hazards, the
Case in point 104 Amp. However, these generated danger associated with static electric-
One major industrial fire at a tote- charges become a problem in hazard- ity can be effectively mitigated by fol-
filling operation at a chemical distri- ous areas when allowed to accumu- lowing national and industry codes
bution facility in Des Moines, Iowa, late on objects that are not at ground of practice, technical standards, and
was investigated in 2008 by the U.S. (earth) potential. Unfortunately, many working practices such as NFPA
Chemical Safety Board (CSB; Wash- of today's paints, coatings, gaskets, 77 (Recommended Practice on Static
ington, D.C.; csb.gov), and the cause seals and other non-conductive mate- Electricity), and NFPA 30 (Flammable
was determined to be static electricity. rials are sufficiently insulating so and Combustible Liquids Code) that
(the full report is available at csb.gov). they actually prevent the proper dis- specify proper grounding and bonding
CSB concluded that a combination of sipation of static charge. techniques. Such techniques work on
technical and procedural factors led to Without proper engineering and the assumption that if a) conductive
an uncontrolled static discharge, which worker training, charges can quickly or static-dissipative plant, equipment
created a fire that spread rapidly to a build up to a very high potential, with and materials are used, and b) these
wood-framed warehouse, igniting a voltages in excess of 30 kV not being materials are properly bonded and
large volume of flammable and com- uncommon. Depending on the capaci- grounded, it will be impossible for
44 Chemical Engineering www.che.com JUNE 2009
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 metal flanges, fittings or valves in In all situations it is also impor-
to accumulate and result in an uncon- pipework systems; portable drums, tant to periodically test the control
trolled discharge. containers or vessels; tanker trucks, measures used, to check the condition
NFPA 77 and NFPA 30 are ex- rail cars and intermediate bulk con- of the clamp, contact and cable, and
tremely useful, in that they provide tainers (IBCs); and even people. Dur- to verify the all-important connec-
practical examples of common op- ing day-to-day operations at industrial tion back to the grounding point (bus
erations, and detail effective ways facilities, isolated conductors are prob- bar). Instruments like Ohm testers or
to eliminate, control or mitigate the ably the most likely source of static ig- multi-meters are useful for perform-
problem. Where today's recommenda- nition incidents. ing such verifications, but, of course,
tions tend to converge is in the recom- Fortunately, the problems associ- these will need to be approved as in-
mendation to always use conductive or ated with isolated conductors can be trinsically safe instruments if you are
static dissipative materials, and to en- remedied by effective grounding and working live in a hazardous area.
sure effective bonding and grounding. bonding. In the case of fixed installa- According to the recommendations
In this context, the term conductive tions such as pipe work, storage tanks in the latest edition of NFPA 77, the
would apply to metal materials, such and so forth, this is relatively simple maximum resistance to ground in
as stainless or carbon steel, alumi- to implement. However, these preven- conductive (metal) bonding systems
num and others, and static-dissipa- tive measures are more difficult to im- should be tested to 10 Ohms or less.
tive may indicate rubber or plastics plement with mobile/portable objects, With all-metal systems and tight con-
that have been formulated with some such as drums, IBCs and tankers. In nections, it should be easy to achieve
added semi-conductive additives. these instances, purpose-designed this level of resistance to ground.
Bonding means linking these ob- temporary grounding and bonding de- While higher resistance levels may
jects together by means of a suitably vices must be used, with strict proce- be theoretically low enough for static
strong conductor (wire), and ground- dures to ensure that they are always charge removal from conductive ob-
ing refers to a true ground/earth in place prior to starting the process. jects, resistance levels higher than
connection that is applied to one or For instance, specific types of two-digit figures in all-metal systems
more of the bonded objects. When one clamps and devices for grounding usually indicate problems, such as cor-
or both of these techniques is applied, and bonding portable or mobile plant rosion, loose or broken connections or
and while a low resistance connection equipment, drums and containers are a buildup of coating or insulating ma-
between the objects and ground is recommended in NFPA 77, and such terials somewhere in the system. Any
maintained, operators are able to pre- grounding clamps and devices gen- such readings should be immediately
vent dangerous levels of static charge erally should employ sharp contact investigated and rectified.
from accumulating. points. These contact points should The presence of semi-conductive or
be made of a wear-resistant material, so-called static-dissipative materials
Hidden dangers have positive spring pressure, and be may also exhibit higher resistance
In any type of flammable atmosphere universally adaptable to a wide range levels, owing to their varying charac-
there may be hidden dangers lurking of plant objects. teristics (which in some cases may be
in the form of so-called isolated conduc- In some cases, these grounding as high as 1 108 Ohms).
tors. These are conductive objects that clamps and devices may have addi-
are either inherently or accidentally tional third-party notified-body test- Additional considerations
insulated from earth. The insulation ing such as Factory Mutual (FM) or Even when the appropriate static
effectively keeps any static electric- the European ATEX Directive (Figure safety equipment has been specified,
ity buildup from safely discharging, 2). If these units are properly specified there are some further concerns that
thereby resulting in accumulation of and correctly used, in most cases the must be addressed by all those respon-
charge on the object. user can be reasonably sure of effec- sible for operations within hazardous
Isolated conductors may arise from tive static control. areas. In operational terms, attaching
Chemical Engineering www.che.com JUNE 2009 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- to be safely dissipated to ground, but times be necessary to also employ other
ous Atmospheres (Cont. from p. 46) sufficiently high to afford some pro- protection measures like static charge
tection against electric shock. If SD neutralization, inert gas blanketing (to
tential may develop (especially in dry, footwear is specified, it can be tested remove the flammable atmosphere),
low-humidity atmospheres). for ongoing effectiveness using test and to look at process related ways to
If the person now approaches a con- stations that test the ground leakage minimize charge generation and maxi-
ductive material at either ground or a from the individual, via the footwear, to mize charge dissipation rates.
lower electrical potential, a spark may ground. These devices help prevent in- Controlling static electricity in haz-
be promoted. It is generally thought correct shoes from being used in static- ardous areas may be simple in theory,
that sparks from people can reach en- sensitive environments (Figure 4). but in practice requires a thorough
ergies as high as 30 mJ sufficient In general, proper mitigation of knowledge of processes and operations,
to ignite almost all flammable vapors static electricity risk in hazardous sound engineering controls, properly
and gases, and many sensitive com- areas demands a holistic approach to specified safety equipment and prop-
bustible powder (dust) atmospheres. plant, process and personnel safety. erly trained operational staff.
In order to control this, it is neces- Any multi-part control program is Edited by Suzanne Shelley
sary to ground the employees working only as good as the weakest link in the
in sensitive hazardous areas. This can chain. As the speed and scale of mod- Author
take the form of ground wrist-straps ern manufacturing and distribution Graham Tyers is president
of Newson Gale Inc. (www.
for sedentary workers, or specialized techniques increase, and the range of newson-gale.com), a manufac-
turer of static control equip-
static-dissipative (SD) safety shoes for materials used and processed grows, ment for hazardous areas (460
personnel who need to move around to this basic approach to safety will be- Faraday Ave., Unit B, Suite 1,
Jackson, NJ 08527). He has
carry out their tasks. In the case of the come even more important. worked in the specialist field
latter, it is also necessary for the floor More detailed information on this of electrostatic control and
hazardous atmospheres for
materials to be sufficiently conduct- subject may be found in industry nearly 20 years. His industry
ing to allow a ground leakage path safety guidelines, such as NFPA 77, background is in monitoring
and control systems, and general electrical equip-
to safely dissipate any static charge as and OSHA Regulations, Section 1910, ment for potentially explosive atmospheres. Tyers
is a member of the NFPA, ESDA, and ISA. He reg-
each foot contacts the ground. which provide safety managers with ularly conducts training and presents to industry
In the U.S., SD footwear is manu- practical guidance for specific situa- and safety groups on the subject of static electric-
ity and hazardous areas. In addition to his tech-
factured according to ASTM F2413-05, tions. It is also important to remember nical subject background, he holds a Diploma in
which specifies maximum and mini- that proper grounding and bonding may Business Management (Open University, United
Kingdom). As part of the management team of
mum levels for ground leakage via the not be enough on their own to ensure Newson Gale in the U.K., Graham was awarded
shoe. According to ASTM, it should be completely safe operations. Depending the Queens Award for Enterprise in 2005. He now
heads up Newson Gales North American Opera-
sufficiently low to allow static charges on the materials involved it may some- tions, based in New Jersey.
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- sourcing contractors and by in-house strates in solution and as a slurry
cidents do occur in the chemical personnel. When developing safety Consider the potential and impact of
process industries (CPI). These documentation, it is important to keep unwanted vapor-phase reactions
incidents have, in part, led to in- in mind that it must comply with com- Develop a chemical-interaction ma-
creased attention to reactive-chemical pany policies and procedures as well as trix for materials present in the
issues by industry, government and country and local regulations. reaction mixture, classify the reac-
other stakeholders. We know that good tivity, and communicate this infor-
safety-management systems take the Process safety checklist mation to operational personnel
reactivity of chemicals as well as the The following items should be consid- Plant considerations:
energetics of both desired and adverse ered in relation to process safety. Conduct a basic energy balance to
reactions into consideration. Preliminary hazard assessment: consider the heats during various
This article provides a safety check- Determine the thermal stability of additions, heat generated during the
list that can be used as a guide for the all reaction components within the chemical reaction, and the heat re-
design of a new process. Alternatively, minimum and maximum process moval capability of the plant reactor
it can be used to identify information temperatures attainable under a system. Remember to include reac-
gaps when existing processes undergo worst-case scenario tor agitation as a source of energy
periodic review. Identify unwanted interaction be- (~2250 Btu/h/hp)
tween reagents and solvents Consider the impact of possible de-
Runaway reactions Identify potential reaction contam- viations from the intended reactant
It is known that in many cases, ther- inants that may have an inhibitory charges and operating conditions
mal runaway reactions occur due to or catalytic effect on the desired Identify all heat sources connected
the following factors: reaction to a reaction vessel and assume
1. Lack of understanding of the pro- Quantification of desired reactions: the maximum possible worst-case
cess chemistry and energy for the Determine the heat of reaction and off- scenario
desired reaction gas rates for the desired and quench Determine the effect of the lowest
2. Inadequate heat transfer capacity reactions, including the heat resulting possible temperature to which the
at the plant level from accumulation of reagents or slow reactor heat-transfer fluid could cool
3. Insufficient control and inadequate forming intermediates the reaction mixture
plant-safety back-up systems, in- Determine the maximum adiabatic Consider the impact of temperature
cluding emergency venting temperature for the reaction, and gradients and other issues, such as
4. Ineffective plant operational proce- determine the basis of safety rela- increased viscosity, freezing at re-
dures and inadequate training tive to the estimated boiling point of actor walls, fouling, and so on, in
The reader is encouraged to add the reaction mixture plant-scale equipment
plant-specific items to this list, as Understand the relative rates of all General chemistry and engineer-
needed. A hazard analysis can then chemical reactions ing design concepts:
be used to identify appropriate con- Quantification of adverse reactions: Design reactions that occur fairly
trols and safeguards to reduce risk. It Assess the thermal stability of the rapidly
is important to note that as a process reaction mixture over a wide tem- If possible, avoid batch reactions
undergoes changes, the safety informa- perature range in which all the potential chemical
tion also needs to be updated. The final When optimizing the robustness of energy is present at the onset of the
process-safety package should be at the process, consider other reaction reaction
such a level that it can be used for tech- variables, such as pH, concentra- Use semi-batch processes for exo-
nology-transfer purposes at the R&D tion, conversion rate, off-gas rate, thermic reactions in which the batch
or commercial-production stage by out- stability of starting and product sub- temperature and any off-gassing can
Chemical Engineering www.che.com April 2008 61
Environmental Manager
be maintained through controlled tion hazards in materials, as well as ized adiabatic calorimeter that uses
addition of the reagent the desired and adverse reactions. We temperature, pressure, and rate data
For highly exothermic reactions, use estimation techniques, differential to allow for sizing emergency vents).
avoid using temperature control scanning calorimetry (DSC), Carius The references given below were used
of the reaction mixture as the only Tube, and reaction calorimetry as to develop this article and are an ex-
means for limiting the reaction rate needed. Identify any adverse or ther- cellent source of information. n
When scaling up a reaction, account mal runaway reactions and, if needed, Edited by Dorothy Lozowski
for the impact of vessel size on heat characterize them using adiabatic
generation and heat removal: The calorimetry, such as ARC (accelerat- Author
Richard Kwasny is the as-
volume of the reaction mixture in- ing rate calorimetry) or an Adiabatic sociate director of the Pro-
creases by the cube of the vessel Dewar Calorimeter. If required, the cess Safety Laboratory at
Chilworth Technology Inc.
radius but the heat-transfer area in- vent size can be determined using (250 Plainsboro Road, Build-
creases by the square of the radius Design Institute for Emergency Relief ing 7, Plainsboro, N.J. 08536;
Email: rkwasny@chilworth.
A comprehensive hazard evalua- Systems (DIERS, an AIChE industry com; Phone: 6097994449),
tion should be conducted using appro- alliance) methodology with data gen- where he has responsibility
for the quality, safety and
priate estimation and experimental erated using an adiabatic dewar or productivity of the thermal
hazard laboratory. His exper-
techniques to identify potential reac- Vent Sizing Package (VSP; a special- tise encompasses all areas of chemical reaction
hazards and flammability of dusts, vapors and
gases. Kwasny completed his Ph.D. at London
References 3. Barton, J. and Rogers, R., Chemical Reaction
Hazards, 2nd ed., Gulf Publishing, 1997.
South Bank University. His research developed,
in part, an assessment strategy to allow for the
1. Hendershot, Dennis C., A Checklist for In- safe scale up of reduction reactions. Before his
herently Safer Chemical Reaction Process 4. Brethericks Handbook of Reactive Chemical current position, Kwasny headed the Process
Design and Operation, Center for Chemical Hazards, 6th ed., Butterworth Heinemann, Safety & Hazard Identification group for Scher-
Process Safety International Conference and 1999. ing-Ploughs Chemical and Physical Services
Workshop on Risk and Reliability, 2002. 5. Guidelines for Chemical Reaction Hazard Dept. Previously, he managed the Dow Chemical
2. Kwasny, Richard S., Hazard Assessment Evaluation and Application to Process De- Reactive Chemicals testing function for Canada.
Strategies for Reduction Reactions, South- sign, Center for Chemical Process Safety of Kwasny is a member of the ASTM Committee
Chem_Eng1-1_177x123_M_April08
bank University, London, 1999. 07.12.2007 11:33
the AIChE, 1995.Uhr Seite 1 E27 on hazard potential of chemicals.
1.1
Lidded Dr
ums
th bungs
/ hopper
cans
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 Color B C
Candela
Boris Viner
Humatrack SafeLight Coverage Configuration
T
he use of appropriate hazardous- A better approach
area lighting in the chemical An appropriately de- F G
process industries (CPI) is a key signed industrial light-
Longevity Integrity
factor when it comes to improv- ing program requires
ing overall plant safety and worker a keen understanding
productivity, and supporting the best of both the human fac- Figure 1. When devising the best lighting scenarios
for CPI facilities, all of these factors which influence
maintenance practices. The poten- tors that dictate how a the true impact that the selected lighting options will
tially hazardous conditions associ- person processes visual have in the workplace must be considered
ated with many industrial settings cues and lighting, and
place high demands on the lighting the nature of the lighting scenario This has potentially serious implica-
fixtures, and it is essential that path- itself. Today, sufficient data are avail- tions in industries where employees
ways, walkways and stairways in able to substantiate that proper light- may be exposed to hazardous chemi-
multi-level platforms or facilities be ing is a key contributing factor to over- cals or fire, because workers who may
properly illuminated. all plant safety, and that well-lighted be inadvertently exposed to chemical
This article discusses the impor- facilities tend to have better safety compounds in the workplace must be
tance of using the most effective and productivity records compared to able to get to the emergency eyewash
lighting practices in hazardous envi- poorly lit ones. or safety shower as quickly as possible
ronments, and presents a conceptu- Unfortunately, while chemical to wash the affected areas.
ally new lighting approach dubbed plants, petroleum refineries and other Because the first few seconds after
safety-centered lighting that seeks types of CPI facilities often expend sig- exposure to a hazardous substance
to help facility operators choose and nificant time and effort to incorporate are critical to avoid more severe injury
install the best hazardous-area light- improved operating practices that are or permanent scarring, eyewashes
ing throughout the facility by combin- intended to keep their workers safe, and safety showers must not only be
ing knowledge of human physiology opportunities to improve the type and located within reach of employees,
with the latest advances in lighting nature of the lighting throughout the but they must be easily identified and
technologies, area-illumination tech- facility are often overlooked. properly illuminated and accessible
niques and installation practices. One of the reasons for this dilemma during both daytime or nighttime con-
Todays lighting fixtures based is that conventional lighting fixtures ditions, so that affected workers can
on light-emitting-diode (LED) tech- have limitations in terms of how the reach them without hesitation.
nology offer significant improve- light fixures function and how the re- As discussed below, LED-based
ments over conventional options sulting light is perceived by the end lighting can deliver an immediate
for hazardous-area lighting such as user. For instance, incandescent or and valuable benefit over prevailing
incandescent, fluorescent or high- fluorescent lamps are commonly used incandescent or fluorescent fixtures
pressure sodium (HPS) lamps. As to identify the locations of emergency during the precious seconds needed to
discussed below, LED fixtures can safety systems, such as industrial locate safety showers and eyewashes,
provide direct improvements in showers and eyewash stations. How- because LED-based fixtures are bet-
overall safety, while reducing main- ever, these fixtures provide practically ter able to focus the light at an op-
tenance and energy requirements, no useful signaling during the day be- timal, targeted color range, making
as well. cause the lighting blends into sunlight. the safety shower or eyewash loca-
Chemical Engineering www.che.com April 2010 53
100 680
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
As a result, one should really deter-
1.5
6.5-7.0 mine the effective lumens of the HPS
1 6.0-6.5 light source at scotopic (night) levels.
5.5-6.0 To do this, the lamp power at each
5.0-5.5
0.5 4.5-5.0
wavelength would need to be multi-
4.0-4.5 plied by the scotopic eye sensitivity
0 3.5-4.0 at each wavelength, and the values
3.0-3.5 summed. The effective lumens value,
2.5-3.0
-0.5 therefore, will be different from the
2.0-2.5
1.5-2.0 conventional lumens value (which, as
-1 1.0-1.5 noted, is considered only under photo-
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 pic or daytime conditions).
0.0-0.5
Because a relatively small portion of
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 overall energy output of the HPS
The color legend represents the intensity (candelas) of a cross-sectional photometric distribution lamp occurs at wavelengths shorter
curve.Isofootcandle chart, 8-ft mounting than the peak, the effective lumens
available for scotopic (night) condi-
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
tions is greatly reduced. This shows
artificial, industrial white lighting on straight lines and avoid spillage, thereby contrib- that sodium lamps (being a source of
uting to optimum illumination, especially in multi-level platforms yellow light) produce relatively little
useful light (effective lumens) for
determination of lamp lumens involves ing level calculations could ignore nighttime conditions, and therefore
knowing both the spectral power dis- light source color, relying only on a their effectiveness under low light lev-
tribution (SPD) of the lamp and the vi- certain lamp lumen rating (usually els when it is essential to have ad-
sual response of the eye. Interestingly, as provided by the manufacturer). In equate industrial lighting is drasti-
it is not simply defined as a form of these instances, calculated values of cally reduced.
energy in the same way as other forms candelas, lumens, or lux (lumens per By contrast, for a white light source,
of radiation; rather, it is defined as en- square meter) are not dependent upon such as that produced by an LED-
ergy as evaluated by the human eye (a whether the light source is white, blu- based fixture, the energy output lies in
reflection of the visual effect which ish white, yellow or pink in nature. the high-sensitivity region of the eye
differs by individual that is created However, published research in- for low light levels (Figure 2).
by that light energy source). dicates the importance of lamp SPD. As a result, the effective lumens in-
Color. When it comes to how the eye It is only recently that the lighting crease as the light level is reduced at
responds to color, vision scientists industry has begun to acknowledge night and the eye shifts to a greater
have known for most of the twentieth the magnitude of the color effects in- blue/green peak sensitivity.
century that lighting conditions play volved, and to show serious interest in As shown in Figure 2, the peak sen-
a big part. While its been widely ac- considering the true properties of the sitivity of the human eye for night vi-
cepted that the cones in the eye handle light. With this in mind, some light- sion is a wavelength of 507 nanome-
day vision and the rods are designed ing experts have started to use the ters (nm), and the peak sensitivity of
for night vision, up until now light- term effective lumens to define the the human eye for daytime vision is a
ing manufacturers have continued modified lumen output of a lamp, tak- wavelength of 555 nm. The region be-
to rely on light meters to measure a ing into account the shifting color sen- tween these two peaks is referred to
lamps lumen output that is calibrated sitivity of the eye at low light levels. as mesopic viewing conditions. Even
by examining the eyes sensitivity to For example, until recently the under mesopic viewing conditions,
only cone-activated (photopic) vision lumen has been defined as the amount yellow light sources have reduced ef-
completely ignoring the effect of of light as perceived by the eye under fectiveness, while LED sources have
rod-activated (scotopic) vision. Unfor- photopic (day) conditions. Based on superior overall effectiveness.
tunately this approach represents a this definition, HPS lamps tend to To provide a mechanism to deter-
gross oversimplification of how human have high lumen ratings. mine the true quality of light, the
vision works, and signals a missed op- However, due to the yellow color author proposes the concept of a ratio
portunity in terms of optimizing in- of the source light produced by HPS called the Quality of Light Factor
dustrial lighting options. lamps, at least half of the light cant (QLF) to determine a rating for the
This is perhaps one of the most fas- be effectively utilized by human eyes illumination provided by a fixture that
cinating and important subjects to be at night. So, it is not as important that reflects actual lumens utilized by the
addressed for efforts to improve in- the sodium lamp produces a high out- human eye. The author proposes that
dustrial lighting in recent years. In put of energy, but rather that its en- the value of QLF can be simply defined
the past, it has been widely accepted ergy peak is near the maximum pho- as a mininum ratio value for com-
that under virtually all circumstances topic sensitivity wavelength of the eye parative purposes. For instance, if an
where artificial light is used, light- (see Figure 2, yellow region). HPS lamp is rated with a QLF equal
Chemical Engineering www.che.com April 2010 55
Dialight
Engineering Practice
to one, then the QLF of the LED lamp duces the need for maintenance
would need to be rated as a minimum personnel to climb structures to
of 2, as the strength of the scotopic eye maintain or repair light fixtures
sensitivity of LED light would be a or replace bulbs. With this in mind,
number of times greater than the sco- the use of state-of-the-art light fix-
topic eye sensitivity of HPS source. tures, such as those based on LED
It is beyond the scope of this article technology, which tends to have
to recommend actual QLF values that longer service life compared to
should be used for different sources other types of conventional lamp
under different conditions. Nonethe- sources (incandescent, fluorescent
less, as a general rule of thumb, the and HPS), typically translates
strength of the scotopic eye sensitivity into improved safety and reduced
of an LED is a number of times higher maintenance and lifecycle costs.
than that of an HPS source. Thus, LED-based lights also tend to have
when purchasing industrial lighting reduced energy requirements com-
systems, the QLF of an LED-based pared to conventional lamp types,
system should be rated much higher and this helps to provide a rapid
than the QLF of a conventional system, return-on-investment. Figure 5. Todays LED-based industrial
such as one based on HPS lamps. Integrity. The integrity of the cho- lighting fixtures typically include cooling
fins to help disseminate heat buildup during
When it comes to safety shower and sen lighting solution relates to a operation
eyewash locations, lights must be visi- fixtures optimized mechanical and
ble under both daytime and nighttime electrical design. For instance, LED options. This is important in appli-
conditions. The color chart provided fixtures provide greater integrity cations such as gas-recovery units
in Figure 2 shows that the best type thanks to several modernizing ad- where numerous railings straight
of lights for signaling safety showers vances that are now widely available, walkways and multi-level structures
and eyewashes are those that provide such as universal input voltage capa- need to be properly illuminated. Un-
a wavelength in the mesopic range (as bilities and much cooler temperature like conventional lights, some of to-
described above). As shown in Figure ratings. Both of these aspects contrib- days patented LED fixtures contain
3, an LED-based light fixture is per- ute to greater integrity, multi-applica- several reflectors to further optimize
fectly visible during daytime condi- tion capabilities and improved safety. light distribution. The ability to ef-
tions. LED fixtures are also particularly fectively focus the light on straight
Candela. Candela is a measure of the useful as a robust light source be- lines where it is needed helps to
intensity of a light source. It relates cause they have no filaments, which reduce spillage of light beyond the
to the light focusing capabilities makes them particularly well-suited target focus area to just a few feet.
of a given light fixture and is thus a for areas that may be exposed to LED fixtures also require lower
function of the light fixtures optical high vibration. wattage than conventional lighting
design. Additionally, in an LED system, compared to incandescent, fluores-
Coverage. Coverage is a measure of lumen maintenance is directly propor- cent or HPS lighting options
the amount of effective light that is tional to thermal management. LED The functional advantages described
spread out for the intended area-illu- systems are typically designed with above also enable LED lights to be
mination application. Adequate cover- cooling fins to compensate for heat spaced at a greater distance than
age is achieved through optical design buildup during use (Figure 5). what would be required using con-
of the LED fixture, to ensure optimum ventional light fixtures. As a result,
light distribution for industrial white Recommended practices the use of LED fixtures can help
lighting (Figure 4). Given the discussion presented above, users to reduce the number of light-
Configuration. The configuration the author offers the following rec- ing fixtures required to illuminate
of a balanced lighting layout refers ommendations for optimizing light the same area by as much as 50%
to the exact positioning of the fix- source efficiency using LED-based compared to the use of conventional
tures (Figure 6). Proper configuration fixtures for hazardous-area illumina- lighting options. The ability to use
within the industrial space is essen- tion. Although the example discussed fewer lights reduces capital costs,
tial, and it is achieved by proper spac- below focuses on a flare-gas-recovery installation time and costs, mainte-
ing between the stanchion mounts of plant, the following recommendations nance costs and overall energy re-
industrial lights. The goal is to ensure can be considered by most chemical quirements
optimum area illumination that usu- plants, petrochemical units and other Unlike conventional lights, todays
ally contributes to safety. CPI plants. LED fixtures can easily be installed
Longevity. Longevity is a measure LED technology provides better using multiple mounting options
of how long-lasting a light source is, directional control of light than is (that is, they can be wall-mounted or
and it is a direct contributing factor possible using conventional incan- stanchion-mounted and can be used
to overall facility safety, since it re- descent, fluorescent or HPS lighting as side-by-side modules (Figure 6).
56 Chemical Engineering www.che.com April 2010
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 safety (Figure 7). Understanding the Author
Boris Viner is president of
dozen types of lighting fixtures to complex properties of the human vi- Humatrack SafeLight (P.O,
be used at different locations, so the sual system provides us with the an- Box 107, Avon, CT 06001;
Phone: 860-673-3498; Email:
ability to standardize on such multi- swers we need to develop better and boris.viner@humatrack.com).
purpose LED fixtures can drasti- safer lighting fixtures that will have He has worked in the light-
ing industry for more than
cally reduce the amount of inven- significant impact on overall facility 25 years. He holds a Masters
tory the facility must manage while safety throughout the CPI. degree in electrical engineer-
ing and a Masters degree
still providing the desired lighting The lighting profile discussed here in management Rensselaer
Polytechnic Inst. (RPI; Troy,
at different locations throughout which considers the complex inter- N.Y.). Viner has presented numerous papers at
the facility actions between competing lighting technical conferences and engineering seminars,
and is a hazardous-area-lighting consultant
Appropriate lighting plays a critical options and the human eye pro- to industrial lighting company Dialight Corp.
role in supporting overall industrial vides a new methodology for evaluat- (www.dialight.com).
Feature Report
Engineering Practice 6
Designing
5
4
3
Safer
2
1
Process Plants
Several often-overlooked strategies 1. Process design
2. Basic controls, process alarms and
M
systems (SIS) or ESD
any individuals and organi- energy transfer to achieve chemical 5. Physical protection (relief devices)
zations have made impor- transformation 6. Physical protection (dikes)
tant contributions to the cre- Some manufacturing processes pro- 7. Plant emergency response
ation of inherently safer (IS) duce benign products but require 8. Community emergency response
products, processes and process plants hazardous chemical intermediates
[13]. A brief survey of successful case in their manufacture FIGURE 1. Shown here are some
typical layers of protection that can be
histories shows that most reported For these reasons, rigorous process employed in a modern process plant [4].
applications relied on only a few of and product safety practices must be At the core is an inherently safe process
the core IS principles. This paper em- used throughout the lifecycle of pro- design. Moving outward from the core,
phasizes the opportunities presented cess plants and must be applied to the proposed options move through the
by three particular and often-over- their associated raw materials and spectrum from inherent to passive to
active to procedural or administrative
looked possibilities for inherently products. In recent years, this has controls, which are considered to be
safer processes. led to major efforts in green chemis- progressively less reliable
The methods proposed here ensure try and engineering to develop prod-
integration of IS methods beginning ucts, manufacturing processes, and objectives of a number of creative indi-
with process conception and continu- plants that are safer for both people viduals and organizations since Kletzs
ing through process plant engineering and the environment. path finding proposal, with many no-
design. Particular emphasis is given Before green chemistry and engi- table successes.
to matching the IS principles with the neering achieved prominence, there Complete coverage of the entire prod-
state of the project. For example, sub- were pioneering insights in the de- uct/process/plant lifecycle is needed to
stitution is best applied during prod- sign of safer process plants. Early ap- assure optimum health, safety and en-
uct and process research, while limita- proaches to safer processes often em- vironmental performance of a chemi-
tion of effects is most effective during ployed additional instrumentation and cal enterprise.
plot plan layout and equipment ar- procedures. These measures were often This article focuses on how to en-
rangement. helpful and necessary, but instrumen- sure maximum incorporation of IS
The chemical process industries tation and operators can fail, especially processes into the creation of a pro-
(CPI) face the challenge of working when faced with complexity. cess plant by beginning at the product
with processes and products that Trevor Kletz [1] recognized that and process research stages and con-
present many hazards, such as What you dont have cant leak, when cluding with the detailed design. No
the following: he first proposed the concept of the effort is made to address the applica-
The manufacture of fuels uses and inherently safer chemical processes tion of inherently safer principles be-
produces products that burn with in 1977. His approach placed an em- yond plant design, although these are
significant energy release phasis on the inherent nature of the also important.
Certain basic chemicals, such as process. Since then, important related
mineral acids and halogens are toxic concepts such as product design for Layers of protection
and/or corrosive safety and safer products, process and The classical onion diagram (Figure
Many manufacturing processes ei- plant lifecycles have also advanced. 1) illustrates the safety layers that
ther release or require significant Creation of IS processes has been the technical professionals throughout
44 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
FIGURE 2. The Flixborough tragedy ushered in a new era in process safety [6] inspired Trevor Kletz to propose his
inherently safer design concept. The
the CPI use to prevent process plant Basic concepts Flixborough plant carried out liquid-
incidents. This diagram helps to ex- Inherently safer process concepts are phase oxidation of large inventories of
plain the following four basic process summarized below [1]: hot cyclohexane in large pressurized
risk-management strategies: Inher- Substitution vessels. When containment was lost, a
ent, passive, active, and procedural or Minimization or intensification large flammable vapor cloud formed,
administrative Moderation or attenuation ignited, and exploded with devastating
Inherent safety is at the core of the Simplification effect (Figure 2, from Mannan [6]).
onion the process design. A process Limitation of (hazardous) effects The traditional cyclohexane-oxida-
that cannot have a major fire, explo- Avoiding knock-on effects tion process to produce a mixture of cy-
sion or toxic release is inherently safer Making incorrect assembly impossible clohexanone and cyclohexanol (K/A oil
than one that could if one or more lay- Make status clear or ketone/alcohol oil) was operated at
ers of protection were to fail. Tolerance of error low conversion rates (typically 35%)
Passive safety layers represent the Ease of control to avoid formation of unwanted byprod-
addition of such safety features as a Administrative controls or proce- ucts. The K/A oil was subsequently con-
dike or a blast wall. Because passive dures verted into adipic acid and caprolactam
layers of protection require no active In 2007, the Center for Chemical for the production of nylon.
intervention by a human or by a ma- Process Safety (CCPS) of the Ameri- Oxidation of cyclohexane with air
chine, they are deemed more reliable can Institute of Chemical Engineers instead of oxygen is common practice
than active layers of protection or (AIChE) concluded that these eleven to reduce risks of transition from a
procedural layers of protection. None- basic concepts could be reduced to the partial oxidation reaction to an un-
theless, the ability to make an explo- following four principles [2]: controlled deflagration in bubbles
sion impossible when possible is Minimize or in the vapor space in the reactor.
clearly better than trying to mitigate Substitute Low conversions and reaction
the effects of a potential explosion by Moderate rates led to large inventories of liquid
adding a blast wall. Moderate and simplify cyclohexane.
Active layers of protection repre- This more concise set of principles During systematic research on the
sent such features as the basic process makes IS practices simpler to under- flammability and deflagration haz-
control system, a safety-instrumented stand and easier to apply. The excel- ards of cyclohexane, air and oxygen
system, and mechanical interlocks. lent new CCPS book (2009) goes on to mixtures, Chen [5] discovered that the
Procedural or administrative distinguish between first-order and addition of a small amount of water
safety layers are generally considered second-order IS: which is inert and does not par-
to be the least reliable and include op- First-order IS efforts change the ticipate in the reaction helped to
erating procedures and operator inter- chemistry of a process inert the otherwise flammable vapors.
vention. Depending on the site-specific Second-order IS efforts change the Cyclohexane and water are known to
hazard, procedural or administrative process variables form minimum-boiling azeotropes.
controls may be entirely appropriate. As can be seen by a survey of the pro- The increase in the vapor pressure of
In general, the preferred ranking cess safety literature, most published the cyclohexane/water liquid results
of methods to control process risks is work has applied one or more of the from the increased vapor pressure of
shown below: first four concepts of the eleven cited the water. The water vapor inerts the
by Kletz and Amyotte [1] For this vapor mixture by lowering the upper
Inherent > passive > active > proce-
reason, this article emphasizes three flammable limit of the vapor [5].
dural or administrative
other promising concepts. Chens work suggests that it will be
Chemical Engineering www.che.com April 2011 45
Qheat generated
E
B
Engineering Practice Qout
Qout Heat removed
3
Qout
Btu/h
D
safe and practical to use pure oxygen other oxidations include
1
for cyclohexane oxidation. Benefits in- chlorination and bromi- 2
clude both IS operation and improved nation reactions, for ex- C
productivity. They also suggest that ample. And there may be A
this approach could be extended to other examples of vapor-
safer processes for partial oxidation of liquid reactions, such as
Reactor temperature
other liquid hydrocarbons using pure hydrogenation reactions,
oxygen. where addition of a new FIGURE 3. Heat-generation (Q
heat generated) and heat-
Chens approach is a first-order IS chemical could improve removal (QOut) rates as a function of reactor temperature
process innovation because it changes the safety of the process. for three different heat-removal designs [9]. Heat gen-
the chemistry of the gas phase in a Addition of an ad- eration is equal to heat removal at points A, C, D, E, and
gas-liquid reaction and prevents the ditional compound to a B, so steady state operation is possible. However, the
reactor is not stable at point D without the addition of
unwanted side reaction of combustion reaction mixture to min- controls or a modification of the design
from occurring in the gas phase. imize hazardous reac-
Although reference [5] did not claim tions may add complexity to the puri- in the process variables and is a form
to have demonstrated a new IS con- fication process, but it may be justified of Moderate, a second-order inherently
cept, Chens work is different from the by the increased safety. safer design.
classical definition of the Substitute Chens [5] paper on cyclohexane CCPS [2] also cites the work of
principle because the same reactants, oxidation illustrates transformation Luyben and Hendershot [8] that high-
chemical reactions, and products are or hybridization, in which the basic lights how minimization or intensifi-
involved. If the name Substitute were chemistry is maintained, but the ad- cation in a reaction system that is in-
broadened to names such as Change dition of another chemical component tended to improve process safety may
in Chemistry or Hybridize, then it transforms a potentially hazardous re- lead to less robust processes with the
could be lumped in with the many suc- action process into a much safer one. opposite effect.
cessful applications that are possible 2. Create a robust process to sta- I propose here that Stabilize or En-
when using the Substitute concept. bilize or ensure dynamic stability. sure Dynamic Stability be added to
Chens innovation permits rapid cy- Not all process designs are inherently the list of IS concepts to be sure that
clohexane oxidation at lower tempera- stable, and if the process design is to be it is not overlooked in the quest for in-
tures and pressures, and could thus safe, the process engineer must ensure herently safer processes.
be said to be an example of the inher- dynamic stability as well as ensuring Application of some of the other IS
ently safer principle Moderate. How- that the steady-state mass and energy principles can adversely affect the dy-
ever, Chens approach enables more balances are achieved. A number of namic stability of a process. For exam-
moderate conditions by narrowing the processes exist that have narrow safe- ple, reduced liquid inventories (Mini-
flammability limits through the addi- operating limits but have been made mize) in a distillation train make the
tion of a new component, water. It is stable by the addition of control sys- process inherently safer from one per-
thus an example of supplementation tems. Dynamic stability and control spective because the smaller process
or hybridization. of chemical processes has been exten- inventory decreases the consequences
Although not proposed by Chen [5] sively studied [7]. of loss of containment. However, the
himself, his work suggests that there Designing the process to be more smaller inventory also shortens the
may be many other opportunities inherently stable to process upsets response time of the distillation sys-
for transformation or hybridization with and without control systems is tem to process upsets, increasing the
of other potentially hazardous reac- clearly inherently safer, although this risk that the basic control system will
tions to make them inherently safer. principle is not addressed in most dis- not be able to restore the distillation
Although water would be high on any- cussions of IS. The IS principle Ease of system to the desired operating condi-
ones list as a potentially transform- Control has usually been interpreted tions and avoid a potentially unsafe
ing additive, it probably will not help to mean a process with a control sys- operating condition and/or an un-
many potentially hazardous reactions. tem that the operator can understand scheduled process shutdown [2].
However, there are many other chemi- clearly and manage effectively. Chemical reactors carrying out
cals that may be inert to the reaction CCPS briefly mentions the advan- exothermic chemical reactions are
and thus also be capable of inerting the tages of designing processes that are perhaps the best known examples of
vapor phase involved in an otherwise inherently more stable or robust [2]: processes that can be dynamically
reactive liquid-vapor reaction. For in- It is inherently safer to develop unstable. Harriott [9] provides the il-
stance, there are many examples of processes with wide operating limits lustration of an irreversible first-order
azeotropic mixtures in the literature that are less sensitive to variations in chemical reaction being conducted in
and there are many compounds that the operating parameters...Sometimes a continuous-flow, stirred-tank reactor
could prove inert to oxidation reac- this type of process is referred to as a (CSTR). Figure 3 shows the heat-gen-
tions (such as, certain halocarbons). forgiving or robust process. eration rate by the chemical reaction
Applications are not limited to Designing a robust process increases as a function of reactor temperature.
partial oxidation with air or oxygen; inherent safety by imposing a change Heat-generation rates are low at low
46 Chemical Engineering www.che.com April 2011
Tools for Inherently Safer Process Plant Design
P rocess hazards reviews Dynamic process simulation O ccupied building evaluation
Chemical interaction matrices Inherent safety analysis and design
Dow Fire and Explosion Index and Area electrical classification
Chemical Exposure Index P eriodic design reviews during Safety integrity level assessments
Fire, explosion and toxic-release product and process research, and safety instrumented systems
consequence modeling and risk development and design Human factors reviews
assessments R
eviews of plant siting, plot plan, Ergonomics reviews
Layer of protection analysis equipment arrangement and 3-D Safety case development
Spacing tables for units and for computer models The design process itself
process equipment
temperatures, but as temperature more than it increases the rate of heat of both the uncontrolled process and
increases, the reaction rate increases removal by the reactor cooling system, the controlled process to ensure a ro-
rapidly because of the exponential so the reactor temperature increases bust process.
dependence of the reaction rate co- more instead of cooling back to the de- 3. Limit hazardous effects during
efficient on temperature. At higher sired operating point. conceptual and detailed engineer-
reactor temperatures, the shrinking This further increase in reactor tem- ing. David Clark published a seminal
concentration of reactant (due to con- perature then leads to an even larger paper [10] on the limitation of effects
version to product) reduces the reac- rate of heat generation rate by the when siting and designing process
tion rate and partially overcomes the reactor and additional heating of the plants. He reminds us that there is a
still-increasing reaction-rate coeffi- reactor. Without any effective control strong, non-linear decrease of fire, ex-
cient. The heat-generation rate even- actions, the reactor temperature will plosion, and toxic effects with separa-
tually reaches a constant maximum tend to increase to point E in Figure 3 tion distance. Comparatively small de-
value when the reaction has reached before it stabilizes. creases in separation distance have a
complete conversion. Similarly, in Figure 3 a decrease in major effect, while larger increases in
Figure 3 also shows three different reactor temperature at point D could separation offer diminishing returns.
straight lines for the heat-removal eventually lead to the reactor temper- Methods, such as the Dow Fire and
rate from the reactor for three differ- ature and conversion dropping back to Explosion Index [11] and the Dow
ent reactor-cooling-system designs. To point C. Chemical Exposure Index [12, 13], pro-
achieve a steady-state energy balance, Clearly, of the three reactor cooling- vide quantitative screening estimates
the rate of heat generation (Qheat gen- system designs represented by the of the hazards from various parts of a
erated) by the chemical reaction must three straight lines in Figure 3, the chemical process. Other indices have
equal the rate of heat removal (Qout) reactor cooling system represented by been developed and evaluated to per-
by the reactor cooling system. That line CDE is the least desirable from form a similar objective to the Dow in-
energy balance occurs when the heat a dynamic-stability perspective. Ad- dices [1, 2, 14]. These screening tools
generation curve intersects the heat dition of an effective control system can identify those parts of a process
removal curve (where Qheat generated might be able to provide dynamic sta- where increased separation distances
= Qout). In Figure 3, the three differ- bility but at the cost of installation are needed to limit potential escala-
ent heat-removal-rate lines intersect and maintenance of the control sys- tion of an incident.
the reactor heat generation rate curve tem and at the cost of residual risk if In one typical plant design, a 10%
at five points. At four of these points the control system fails. increase in separation distances for
(A, B, C, E), the steady-state energy Another example of potential all units increases total plant invest-
balance solution is stable. At each of sources of process instability results ment cost by only 3%. Similarly, dou-
these points, if there is an increase in from efforts to improve energy effi- bling the separation distance for a
temperature, the rate of heat removal ciencies in distillation trains through hazardous unit representing 10% of
increases more rapidly than the rate heat integration. In these cases, the the investment cost of the plant would
of heat generation by the reaction and feed to a column may be preheated cost only 3% more. Because of the non-
the reactor temperature tends to re- by the bottoms product of a second linear effect of separation distance,
turn to the desired operating point. downstream column. This may in- doubling the separation distance for a
Similarly, if the temperature drops crease the risk of process upsets due hazardous unit could reduce explosion
slightly at one of these four operating to increased interactions between the overpressures on the adjacent units
conditions, the rate of heat removal two columns. by a factor of four or more.
decreases more than the rate of heat While avoidance of add-on controls The strong decrease in hazardous
generation by the reactor and the tem- has always been a goal of inherently effects with modest increases in sepa-
perature trends back up to the desired safer design, achievement of that goal ration distances will often more than
operating condition. has seldom mentioned the concepts justify increased capital cost.
In contrast, point D in Figure 3 is an of Ensure dynamic stability or Stabi- Spacing also offers important ben-
inherently unstable operating condi- lize as tools of the process engineer. It efits in crane and other maintenance
tion even though the steady state rate should be considered when consider- access, ergonomic advantages and
of heat generation by the reactor equals ing other means to assure inherently decreased risk of incident escalation.
the rate of heat removal by the reactor safer processes during process design. Future plant expansions or process
cooling system. At point D, an increase The process engineer should work improvements are also facilitated, al-
in reactor temperature increases the closely with the control systems engi- though expansions that decrease spac-
rate of heat generation by the reactor neer to address the dynamic stability ing may increase hazardous effects.
Chemical Engineering www.che.com April 2011 47
Engineering Practice
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 All project changes that are
used for quality control in a made in the laboratory should be
production environment and first evaluated for potential haz-
those in research and devel- ards. Having a process in place to
opment, are an integral part of the evaluate and manage change is
chemical process industries (CPI). extremely helpful. Inviting oth-
Everyday demands of business can ers to get involved as you work
easily overtake safety when it comes through the process is smart, as
to setting priorities in the laboratory. well. Your supervisors, advisors,
Deadlines must be met. Samples co-workers and environmental,
must be analyzed. Questions must health and safety (EH&S) ex-
be answered and quickly. perts can put their knowledge and gate the risks are key components
But even for the most experienced experience to work by helping to in managing process change. This
chemists and laboratory workers, identify potential hazards and of- ensures that others know about
safety must remain at the forefront. fering ways to mitigate those risks. the changes, as well as their poten-
This is especially true when new tial hazards and impact. The more
chemicals, employees or equipment 2. Identify potential hazards everyone in the laboratory under-
are introduced. Changes such as You cant manage what you dont stands, the safer the work environ-
these are part of life in the labora- know or understand. Using a ment in the laboratory becomes.
tory, yet change represents risk. And change-management process, you
as such, it is an element that must can identify potential hazards and 3. Rely on a second set of eyes
be thoroughly incorporated into mitigation solutions associated Your colleagues can often see what
every CPI laboratorys safety plan. with change. Reviewing predefined you cant. The truth is that we be-
Read on for nine best practices triggers or typical changes that come so familiar with our own work
from identifying hazards to com- are known to create hazards can that we often overlook the obvious.
municating effectively to ensure help as you go through laboratory Inviting a co-worker or supervisor
that your laboratory is as safe as changes step-by-step. The triggers to be your second set of eyes during
possible amid inevitable and ongo- quickly highlight changes that need a change review can help identify
ing change. further consideration. and address potential issues.
A Hazard Assessment Trigger When sharing your situation with
1. Tackle change upfront Grid and Safe Operations Card a reviewer, dont just sit in an office
As the old saying goes, nothing is (SOC) are two tools that can be and talk about your changes with
constant but change. And thats used to identify potential hazards the other person. Go where the ac-
doubly true for work in the labora- [1]. SOCs allow you to evaluate haz- tion is, like the laboratory hood or
tory. In fact, change defines chemis- ards as you plan your laboratory bench. Tell and show your review-
try. So how do we plan for it? project, thereby helping to ensure ers exactly what you will be doing
Take time upfront to carefully as- that any potential hazards are fully (Figure 1). With everything set up
sess the impact that every change identified and assessed early on. and in sight, you and your review-
will have on safety, health and en- They also document appropriate ers may see things that can be
vironmental issues. Laboratory operating ranges, conditions and modified to further enhance safety.
operations designed for safety also emergency response in the event of Reviewers should ask open-ended
generate better, more reproducible an accident. questions to best understand the
results, without placing people or Documenting potential hazards work to be completed and potential
equipment at risk. and evaluating how you will miti- worst-case scenarios, as well as to
54 Chemical Engineering www.che.com August 2014
Safety Inspections: The Kindness Factor
I
t is amazing what you can find when you take time to look, ask Scott Geller says that actively caring means practicing system-
and connect with those around you. During times of change, atic and purposeful acts of kindness to keep other people safe
which can introduce extra stressors into the environment, this and healthy.
element of connection becomes extremely important in ensuring When you and your co-workers care about safety and each
that communication is clear and effective. This is true throughout other you will want to learn how to improve. You will be ac-
laboratory work, but is notably important during inspections. tively involved. So create an environment in your workplace where
Inspections help identify and mitigate safety discrepancies, and people are free to ask questions and know how to escalate issues
they can be formal or informal, depending on what is needed. Yet that become a problem. Make sure that everyone in the laboratory
inspections are more than checking off items on a checklist to see knows that no matter what, safety must be planned into the experi-
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 Finally, after taking the time for inspections and improvements,
what is needed to maximize laboratory safety. remember to recognize success. As improvements are made, or if
In fact, one of the most important components of inspections is a minimum number of safety discrepancies are found, let everyone
the interaction and dialogue that will take place between you, know. Be loud and proud about your successes in incorporating
your co-workers and supervisor. That is why it is a must to make new chemicals, materials, equipment and new people into your
a personal connection during the inspection. Psychologist E. work environment.
understand the layers of protection cess, you would want to assess often must be used. From fume
and safety measures in place to mit- potential worker exposure to any hoods and glassware to gas cylin-
igate risks. nanomaterials that will be used by ders, nearly every piece of equip-
Youll find that both you and your considering tasks that may poten- ment in a laboratory can pose a
project will benefit when you involve tially expose you or fellow labora- hazard when used inappropriately.
a variety of people with different tory workers to nanomaterials. You Be sure to think about and plan for
expertise and experience early on. would consider the dustiness of the how youll handle tasks like equip-
When considering who to approach material, how it will be used, how ment set-up, maintenance and
for this role, seek to match your re- much will be used, and the duration cleaning. The hazards associated
viewers with the type of change that and frequency of the task. with these types of support tasks
youre planning. For instance, if you Once the preplanning is finished, (such as ergonomics, decontamina-
are installing new equipment, get move on to safety equipment con- tion, energy control and more) are
people who are familiar with that siderations. Will a fume hood be often underestimated.
type of equipment. If you are using needed? Is your standard personal Not only must you make sure that
new chemistries, get people who protective equipment (PPE) enough you are completely up to speed on
are knowledgeable about those. Use to keep you safe? Continuing with the safest way to operate the equip-
their knowledge to your advantage. the nanomaterials example, you ment especially if its been some
An added bonus to the review is would want to ensure that you keep time since youve received training
that this type of cross-functional the materials off your skin and out but you also must ensure that
involvement creates networking of your lungs and eyes. As such, ven- others, including new employees,
opportunities. You will get to know tilation systems would be needed to are up to speed as well. In the rush
people throughout your department capture and remove airborne nano- to find results for an experiment,
and in other functions, as well. materials before they are inhaled shortcuts may seem appealing, but
by workers. Depending on the scale proper care always must be taken
4. Understand the risks of the operations, ventilated enclo- even more so when change is afoot.
When new chemicals enter the sures, local exhaust ventilation or If you are uncertain about how to
laboratory, the risks increase. This other types of ventilation could be use a piece of equipment, seek out
situation calls for special planning needed. Go through this planning training. For quick refreshers, re-
to account for how you will plan process for any new material you sources are available on a variety of
your experiment including what introduce, devoting extra time to topics [2].
equipment youll use, how you man- evaluate your safety when dealing
age your waste and decontamina- with pyrophoric materials, hazard- 6. Take care when working alone
tion considerations. ous chemicals, cryogenic liquids and On a busy day at work on a new
To provide an example, lets say reactive chemicals. project, you may be tempted to stay
that you are beginning new labora- When considering disposal of new late to make additional progress. If
tory work that involves nanomate- chemicals or materials, always be you find yourself in this situation,
rials. Your preplanning research sure that different categories of waste you must first look at the work
would tell you that these ultra- are stored in different containers youll be doing and evaluate its haz-
small materials could be absorbed and are clearly separated, and that ard potential before deciding that it
by the body faster than other ma- incompatible chemicals are stored is safe to work alone.
terials, thereby resulting in greater separately. The laboratory is not the If what youre doing is classi-
exposure. Also, they may travel to place for unexpected reactions. fied as a high-hazard operation,
internal organs that were not previ- its never safe to work alone, even
ously accessible to the larger-scale 5. Know your equipment if special precautions are in place.
particles of the same material. When a new material is introduced Some examples of high-hazard op-
As part of the preplanning pro- to the laboratory, new equipment erations include work with process
Chemical Engineering www.che.com August 2014 55
Environmental Manager
FIGURE 1.
plants, any equipment that com-
Pre-startup
presses, and work that involves safety reviews
high quantities of flammable or are an important
extremely hazardous materials. tool to help en-
These kinds of operations require sure laboratory
safety
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, 8. Communicate clearly who actively engage with their
handing samples for routine analy- Communication is critical to ef- teams help deliver outstanding re-
sis, minor maintenance, general fective laboratory safety in any sults and improved performance.
laboratory work and office work. situation, but especially when any They are able to build, strengthen
These operations dont present any change is introduced to your labora- and engage their teams. This is cer-
major risk to safety and are gener- tory. While this may seem obvious, tainly true for safety performance.
ally safe, as long as you have a plan poor communication is the reason Even if you are not a laboratory
in case of a problem. A mechanism for far too many laboratory acci- leader, you have a big role in pro-
like a lone operator alarm or emer- dents. Thats why it is absolutely es- moting leadership engagement.
gency radio should be in place. sential that you communicate with First, help your safety leader in
everyone who could be impacted by identifying key systems needed,
7. Practice good housekeeping any change. such as programs, practices, proce-
In the laboratory, everything you Your plans must always be shared dures and training. As your leader
do, or dont do, makes a difference with anyone who could be impacted identifies key behaviors required to
in the outcome of your experiments by them. Be diligent about this. achieve the desired goals, be open
and your safety. That includes keep- Discuss the experiments you will and willing to adjust, use and cham-
ing the laboratory clean. Whether be conducting, the potential haz- pion those behaviors.
youre introducing change into the ards, worst-case scenarios and any
laboratory or not, youre much more planned emergency response. As Change and safety are constant
likely to notice when something is project owners, be receptive to the Your personal safety, as well as the
amiss when everything is in order. advice and safety improvements sug- safety of your co-workers, depends
Use a checklist to perform in- gested by colleagues. As reviewers, on your commitment to manag-
spections on at least a monthly interact and share your expertise ing change wisely. Remember that
basis. Ensure your space is free in a way that influences safety best change is the main reason to stop
of tripping hazards, unused ma- practices. The engagement should and evaluate potential hazards.
terials, excess paper and clutter. be a positive interaction where risks Whether youre changing chemi-
Check to make sure that the covers are identified and solutions to miti- cals, equipment or people, you can
for drains and trench drains are in gate these risks are identified. manage it successfully and safely.
place. Also, pay special attention The aforementioned SOC is a Edited by Dorothy Lozowski
to surfaces including benches, great tool for helping to plan for
floors and operating equipment and communicate potential haz- References
to see that they are free of hazard- ards connected to process change. 1. Free downloads of these tools are available
under Resources at the Dow Lab Safety
ous dust accumulation. SOCs establish accurate commu- Academy, http://safety.dow.com).
Part of your checklist should nication between you and every- 2. The Dow Lab Safety Academy (http://safety.
focus on safety equipment: Is ev- one else in the laboratory and dow.com) offers short tutorial videos on a va-
riety of topics.
erything in place and function- that is one of the most important
ing as it should be? Be especially things you can do to ensure a safe Author
vigilant when it comes to personal working environment. Lori Seiler is the associate
protective equipment, respirators, director for environmental
health & safety for Dow Re-
eyewash, safety shower, fire ex- 9. Make change part of the culture search and Development at
The Dow Chemical Company
tinguishers and telephone. Once In order to ensure that change (2030 Dow Center, Midland,
your laboratory inspection is done, management is always part of the MI 48674; Email: lseiler@dow.
com). Seiler is a certified in-
openly communicate any problems plan, a strong safety culture must dustrial hygienist (CIH) and
you see, especially unsafe prac- permeate the laboratory. And while holds a Masters degree from
the University of Massachu-
tices and procedures. (See side- everyone is essential to the safety setts at Amherst. She has
bar. p.55: Safety Inspections: The culture, it starts with leaders. been closely involved in Dows university labora-
tory safety initiative and the creation of the Dow
Kindness Factor.) Experience shows that leaders Lab Safety Academy (http://safety.dow.com).
In Brief
CPI project lifecycle
Process Hazard
Identification
T
When to use a given
he chemical process industries (CPI) vamps and decommissioning.
method
handle a wide variety of materials, In each of the plant life phases, different
Safe-design options many of which are hazardous by na- choices are made by engineers that have a
addressing hazards ture (for example, flammable, toxic direct impact on the overall risks in the fa-
early or reactive), or are processed at hazardous cility; however, the highest opportunities for
conditions (such as high pressures or tem- cost-effective risk reduction are present in
final remarks
peratures). The risks associated with CPI the earlier phases of the project. In contrast,
facilities not only extend to the plant person- the cost of implementing changes in the later
nel and assets, but can potentially affect the stages of the project increases dramatically.
surrounding population and environment Hence, it is important for the design team to
sometimes with consequences having re- identify risks, and implement effective design
gional or international scale, as in the case of solutions as early as possible.
toxic vapor or liquid releases. This article covers some of the typical de-
It is for this reason that process safety is cisions that the project design team has to
recognized as a key element throughout the make over the course of a project, with ex-
entire life of the plant, and several industry and amples of how the incorporation of process
professional associations and government safety throughout the entire design process
authorities have issued norms, standards and can significantly reduce the risk introduced
regulations with regards to this subject. by a new CPI facility, while also avoiding po-
Process safety, as defined by the Center tential cost-overruns, or unacceptable risk
for Chemical Process Safety (CCPS), is a scenarios at later stages.
discipline that focuses on the prevention and
mitigation of fires, explosions and acciden- CPI project lifecycle
tal chemical releases at process facilities. A project for a new chemical process facility
Excludes classic worker health and safety usually involves different phases, which are
issues involving working surfaces, ladders, outlined here:
protective equipment and so on. [1] Process A screening or visualization phase. In this
safety involves the entire plant lifecycle: from phase, the business need for the plant is
visualization and concept, through basic assessed. Typical choices at this stage in-
and detailed engineering design, construc- volve defining plant throughput, processing
tion, commissioning, startup, operations, re- technology, main blocks and plant location
50 Chemical Engineering 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.
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-
s
Compliant with FDA and cGMP
guidelines
PROTECT PUMPS
DRY RUNNING CAVITATION BEARING FAILURE OVERLOAD
PUMPING
POWER
AMPS
Circle 18 on p. 74 or go to adlinks.chemengonline.com/56200-18
Some studies have more impact HAZOP can serve to detect the need www.processsystems.sandvik.com
in some phases than in others. For for additional safeguards that were
Circle 26 on p. 74 or go to adlinks.chemengonline.com/56200-26
FeatuCompilati
o identify hazards, and address them
neering from Universidad de Buenos Aires (Argentina,
1996). She has published articles and delivered world-
2014 using the best engineering practices wide seminars focused in the use of simulation tools for
the process industry.
in safe-design from early on in the
Alberto Baumeister is the
project lifecycle, along with prop- CEO and co-owner of the Ecotek
erly timed and executed PHAs, can group of companies (same address
be very valuable in avoiding costly as above; Email: abaumeister@
ecotekgrp.com). He has experience
changes during the EPC phase, or in corporate management, project
even worse: potential damages to management, and senior process
persons and the environment. n consulting in engineering projects
for the chemical, petrochemical,
Edited by Gerald Ondrey petroleum-refining, oil-and-gas,
25139
electrical power-generation and agro-industrial indus-
tries. He has a specialization in environmental engineer-
References ing (gas effluents treatment) from the Universidad Miguel
1. Center for Chemical Process Safety (CCPS), Guidelines de Cervantes (Spain, 2013), a masters diploma in water
Find this and other related for Investigating Chemical Process Incidents, 2nd edi- treatment management from Universidad de Len (Spain,
tion, CCPS, AIChE, New York, N.Y., 2003. 2011), a specialization in management for engineers at
reference material at Instituto de Estudios Superiores de Administracin (Ven-
2. CCPS, Guidelines for Engineering Design for Process
store.chemengonline.com Safety, 2nd ed., CCPS, AIChE, New York, N.Y., 2012. ezuela, 1990), and a degree in chemical engineering
from Universidad Metropolitana (Venezuela, 1987). He
3. AACE International Recommended Practice No. 18R- was professor of the Chemical Engineering School at
97, Cost Estimate Classification System As Applied Universidad Metropolitana between 1995 and 2007, and
in Engineering, Procurement, and Construction for the has written a number of technical publications.
Process Industries.
58 Chemical Engineering www.chemengonline.com august 2015
Environmental Manager
a higher level
Manufacturer / waste generator
C
hemicals in production and Figure 1. Managing post-use chemicals is a vital part of chemical lifecycle man-
laboratory environments agement. Recycling waste chemicals is an option that can contribute greatly to
alike have three distinct sustainability goals
lifecycle phases: the first is
procurement and inventory storage; local teams and staff who have other challenge, when it comes to chemical
the second is use in a manufacturing primary responsibilities. Regardless lifecycle management, is to find op-
process or research program; and the of size, laboratory and production erationally efficient and cost-effec-
third is post-use, including onsite han- managers must periodically review tive ways to reduce the overall waste
dling, removal and beneficial reuse, their policies, conduct training and stream through process optimization,
recycle or disposal. audit their chemical inventories and efficient material use, tolling or sell-
A number of considerations includ- waste materials. Requirements, pro- ing recyclable materials to a second-
ing regulations, safety, material uti- cesses and regulations change, and ary market.
lization efficiency, costs and sustain- it is managements responsibility to
ability govern all of the steps in this remain in compliance, while meeting Purchasing and storage
lifecycle. The challenge for laboratory everyday production demands. Effective chemical management
and production managers is to balance Environmental sustainability con- starts before the chemicals arrive at
these sometimes competing forces to tinues to increase in importance and the site. Laboratory and operations
create an effective process that meets is embraced as a corporate goal at personnel need to work carefully with
the organizations overall needs. many manufacturing organizations. purchasing staff to order the correct
Proper chemical handling during This takes chemical management be- amounts, minimize order errors and
processing is typically emphasized in yond the traditional regulatory and ensure that proper receiving and
production facilities, but how to han- safety considerations. The chemical storage provisions are in place. For
dle chemicals post-use is often given process industries (CPI) have become operations personnel, it is a balance
less attention. While this article ex- even greater stewards of the materials between anticipating customer orders
amines the full chemical lifecycle, it throughout their entire lifecycles as and manufacturing volume in order
focuses on this post-use phase. they seek ways to further reuse or re- to eliminate production interruptions,
cycle materials that were historically minimize storage and other inventory
Sustainable lifecycles disposed of as waste (Figure 1). costs, and to avoid having to dispose
Every laboratory and chemical pro- Sustainability requires the mini- of outdated materials. Purchasing
cessor must follow environmental, mization of consumption and waste staff can consolidate orders to lower
health and safety (EH&S) proce- in order to reduce environmental im- costs, but they must have an under-
dures and regulations. This is often pact through process optimization and standing of the chemical properties
a challenge, especially for smaller reuse or recycling whenever possible. and intended uses in order to ensure
operations, since implementation fre- This can require additional record- that materials are consumed before
quently rests on individual managers, keeping and administrative steps. The they begin to degrade and negatively
44 Chemical Engineering www.che.com December 2010
affect chemical reactions or final prod- label chemicals in secondary containers with the current date when full, and
uct properties. Overall, the CPI have at their stations. Unlabeled or improp- removed within three days to the main
done a good job utilizing continuous erly labeled chemicals violate hazard- storage area (MSA).
improvement techniques to manage ous communication (Hazcom) regula- The MSA is governed by additional
their supply chains. tions. Informal practices also lead to rules. One of the most important re-
Once received, the materials must the potential for abandoned chemicals lates to allowed accumulation time
be stored in accordance with local, as researchers move on to other jobs or once materials enter the MSA. Small
state and provincial regulations, or projects. Unidentified chemicals may quantity generators (SQGs) organi-
the International Code Councils require testing, special handling and zations that generate more than 100
(ICC) International Building and In- expensive disposal if they cannot be kg but less than 1,000 kg of hazard-
ternational Fire Codes, and handled verified by the researcher. ous waste per month can accumu-
in accordance with company EH&S Consistency and training, sup- late materials for 180 days from the
procedures. In the U.S., handling ported by accurate, up-to-date la- start date.1
must follow the Occupational Health beling and recordkeeping, are key Large quantity generators (LQGs)
& Safety Admin.s (OSHA) Code of ingredients to a successful chemical those that generate 1,000 kg or
Federal Regulations (Title 29 CFR management program. more of hazardous waste per month,
1910.1200), which covers Hazard or more than 1 kg per month of acutely
Communication Standards (HCS), Post-use hazardous waste are subject to a
the federal Risk Management Pro- Handling of solvents, catalysts and 90-day accumulation rule.2
gram for certain highly toxic chemi- other chemicals can become much Regardless of size, the MSAs must
cal compounds regulated under the more problematic once they have be secured against unauthorized entry;
Clean Air Act, and the Chemical Hy- been used in a production or labora- hazardous waste and container con-
giene Standard for laboratories (Title tory process. They are no longer in tents labeling rules must be followed;
29 CFR 1910.1450). their original containers and are often containers must be inspected weekly
combined with other substances. In with reports kept on file; and pre-
Production and laboratory use the U.S. for example, OSHA, the En- paredness and prevention equipment
As materials move into the second vironmental Protection Agency (EPA), (for example an emergency phone,
phase of the lifecycle use in a pro- Dept. of Transportation (DOT) and, in alarm, fire suppression and spill pre-
duction environment or laboratory some cases, Drug Enforcement Admin. vention equipment) is required.
the potential for handling errors in- (DEA) regulations also come into play. There are additional regulations
creases. Although industrial users and A few important fundamental facts and exceptions to the rules noted
researchers are concerned with, and must be kept in mind at all times: In above. For instance, peroxide form-
trained in, safe handling procedures the U.S., all chemical storage areas ers, such as ethers and dioxane, are
for hazardous materials, they are less fall under Resource Conservation and generally managed under a peroxide-
likely to focus on storage and removal. Recovery Act (RCRA) regulations and former program that tracks the ma-
This leads to some potential problems are subject to inspection by regulators. terial based on the manufacture and
in chemical lifecycle management. To comply with that, a limited number retention dates from the day it arrives
Chemicals are generally well man- of conveniently positioned and prop- onsite until the day it is consumed or
aged in production environments erly organized satellite accumulation disposed of. SAA storage times and
since production procedures are well- areas (SAAs), under the control of the 90/180-day MSA rules are all subordi-
defined and engineered. Failure most EH&S or chemical hygiene manager, nate to the peroxide-former deadlines.
often results from human error or should be set up near production or So, if the earliest peroxide-former re-
system breakdowns. Although the research areas. tention date is 45 days in the future,
results can be catastrophic, they are There are a number of RCRA rules that is the containers deadline. Even
largely preventable through adher- for SAAs. For instance, containers the waste disposal company must
ence to established EH&S procedures, must be in good condition without track peroxide formers and follow ex-
effective maintenance practices and rust, dents, or cracks; they cannot be pedited destruction schedules based
frequent training. stored near drains or other structures on the retention date.
Research environments are less that could pose an environmental Managing chemical storage areas.
systematized, which presents greater risk; hazardous waste containers can- A minimum number of properly
opportunity for errors. Researchers not exceed 55 gal; acutely hazardous trained staff should be assigned to
often work independently on projects waste cannot exceed one quart; SAAs transport materials from the SAAs to
and they use smaller quantities of a must be inspected weekly. There are the MSA and manage the materials in
greater number of chemicals. This additional guidelines including re- all of the storage areas. Some opera-
presents an exponentially higher quirements for clear and visible labels tions are large enough to have a dedi-
number of possible chemical reactions stating hazardous waste and listing
1. http://www.epa.gov/waste/hazard/generation/
and storage issues. the container contents, as well as the sqg/index.htm
Working alone or on small teams, three-day rule that requires that the 2. http://www.epa.gov/solidwaste/hazard/
researchers often neglect to properly containers are immediately marked generation/lqg.htm
A
n added benefit of recycling and reusing chemicals can be reducing the volume
Environmental Manager 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:
cated full-time staff, while for others it Aluminum (fume or dust) The TRI listing for the following three
is a part-time function (that requires Aluminum oxide (fibrous forms) chemicals is based on the chemical activity
Ammonia (anhydrous and aqueous) rather than the form of the chemical:
the same annual refresher training).
Asbestos (friable) 1. Dioxin and dioxin-like compounds
Properly organizing the handling Hydrochloric acid (acid aerosols) (only if they are manufactured at the fa-
and in-house movement of chemicals Nitrate compounds (water dissociable; cility; or are processed or otherwise used
removes the responsibility from re- aqueous solution) when present as contaminants in a chemi-
searchers and substantially reduces Phosphorus (yellow or white) cal but only if they were created during the
training requirements. Sulfuric acid (acid aerosols) manufacture of that chemical)
Another benefit of using dedicated Vanadium (except when contained in an 2. Isopropyl alcohol (only if it is being
staff is that it puts a third party in po- alloy) manufactured by the strong acid process)
sition to regularly inspect warehouse, Zinc (fume or dust) 3. Saccharin (only if it is being manu-
production and laboratory areas to factured)
ensure that chemicals are properly 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/
stored and handled in accordance with ry2009rfi121709.pdf). See Table II-3, p. 143 for the full alphabetical list of chemicals and qualifiers
regulatory and company policies. Here
are some best practices that we have
seen in the field and that we practice Perhaps, most importantly, they may Using the same outsourcer to handle
at our own sites and for customers: not stand up to regulatory scrutiny if materials in-house as well as for re-
Ensure that proper procedures the need arises. moval and disposal eliminates inter-
(such as labeling and tracking) are Specialized software and online organizational hand-offs and can fur-
being followed at each production chemical-waste-management systems ther reduce costs.
and laboratory location, and that handle all of these issues and also Communications between waste
waste chemicals are stored only at address many other important cri- generators and chemical disposal
approved sites teria, such as chemical movements, companies. Before a waste disposal
Conduct periodic safety training for 90/180-day storage-time limits, chemi- company can remove hazardous
research and production personnel cal expiration dates and reporting. chemicals from a generator, all paper-
Establish collection schedules to ef- More sophisticated software and on- work must be in order, the materials
ficiently move materials from SAAs line tracking systems assign cost cen- must be properly labeled and pack-
to the MSA in a timely way ter allocations, and facilitate move- aged and ready to go. It is important
Ensure that SAA containers are in ment, inventorying and packing, as that chemical waste is properly char-
good condition, properly closed and well as documenting the transfer of acterized in order to avoid unneces-
placed within secondary containers, the material to a certified waste dis- sary expense or, in a worst case, addi-
and that incompatible materials are posal company. tional transshipping to another waste
properly segregated disposal site. Confusion creates lost
Catalog, store and consolidate MSA Waste disposal companies time and increased costs for chemical
materials in bulk containers and Given the complexity of effective characterization tests and EPA and
pack smaller quantities, elbow bot- waste-chemical management, many DOT documentation.
tles, and so on, for transportation companies find that it is more ef- Communications with the waste
Identify unknowns before shipping ficient and cost effective to turn to disposal company are key. Direct links
Follow up on the disposal or recov- waste disposal companies to handle between the generator and the waste
ery of chemicals for company and their chemical waste throughout the disposal company are efficient since
regulatory reporting entire post-use lifecycle phase. Service any materials or paperwork that are
Software. Most large companies companies manage the entire process not in compliance will be identified
utilize chemical inventory manage- including onsite movements, storage, early, waste profiles and DOT mani-
ment systems, while many laborato- packing and removal. In that way, fests can be generated, and placards
ries still use spreadsheets to manage generators do not have to invest as assigned before the disposal company
their inventories. Spreadsheets can be heavily in RCRA training and can rely arrives on site.
helpful, but they are not the answer on the contractor to apply its exper- A direct link to the waste disposal
for corporate-wide or comprehensive tise, provide staff and backup staff as company can also report contain-
lifecycle-management programs. They necessary, monitor regulatory federal, ers that are ready for packing and/or
are generally home-grown standalone state and local developments, and im- pickup, enabling the removal company
applications without built-in report- plement best practices. to cost-effectively schedule pickups.
ing mechanisms. They do not extend This allows plant and laboratory Automated procedures also help com-
beyond the department or integrate management to properly oversee the panies stay in compliance with RCRA
with corporate systems. They do not function without adding headcount. It 90/180-day rules or special handling
tie into industry resources and docu- also enables staff to focus on their core requirements for peroxide formers
mentation, such as material safety responsibilities and not be burdened and other acute hazards.
data sheet (MSDS) databases, that are with additional training or oversight Regardless of the degree of auto-
necessary to build chemical profiles. of chemical management functions. mation and communications, under-
46 Chemical Engineering www.che.com December 2010
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-)
Clearing the
Figure 1. When
engineering controls
are not enough to
reduce or eliminate
airborne hazards,
Air About
respirators are
needed for protection
Respiratory
Protection
Learn the basics about selection
and regulatory compliance for these
potentially life-saving devices
Dennis Capizzi years to produce noticeable harm. the main federal agency charged with
MSA Safety Where toxic substances are pres- the enforcement of safety and health
ent in the workplace and engineering legislation in the U.S. Much of this leg-
A
irborne respiratory hazards are a controls (such as enclosing or confin- islation is written in the form of stan-
very real threat in the chemical ing the contaminant-producing opera- dards. OSHA standards are rules de-
process industries (CPI). Exist- tion, exhausting the contaminant, or signed to provide a work environment
ing in a variety of forms includ- substituting with less toxic materials) free from known dangers or hazards.
ing gases, vapors, dusts, mists, fumes, are inadequate to reduce or eliminate They require the use of certain safe
smoke, sprays and fog, such hazards can them, it is time to turn to respirators practices and equipment, and assign
cause illnesses including cancer and (Figure 1). employers the responsibility of moni-
lung impairment, or even death. Ex- toring hazards and keeping records of
amples of respiratory hazards include What is a respirator? workplace injuries and illnesses. Com-
combustion byproducts; toxic fumes or A respirator is a protective device that pliance with OSHAs standards is man-
dust created by metal melting; ozone covers the nose and mouth or the en- datory and is stringently enforced.
and nitrogen oxides from processes tire face or head to guard the wearer In 2011, Respiratory Protection,
involving an electric discharge in air; against hazardous atmospheres. Res- General Industry (29 CFR 1910.134)
and dust particles released by grinding pirators may be either tight-fitting or was OSHAs fourth most frequently
(notably dry grinding such as blast- loose-fitting. There are two types of cited standard violation. Standard 29
ing) and crushing applications, and the tight-fitting respirators: half masks CFR 1910.134 covers the respiratory
transport, sieving, mixing or screening cover the wearers mouth and nose, protection division of the personal
of any dry material. while full facepieces cover from the protective-equipment category. Essen-
The specific hazardous gases pres- hairline to below the chin (Figure tially, it details the requirements for
ent in a workplace will, of course, 2). Loose-fitting respirators include compliance in workplaces where respi-
vary according to the processes of the hoods and helmets that cover the ratory hazards are a potential threat
facility, but commonly include chlo- head completely. (Figure 3). The aforementioned statis-
rine, chlorine dioxide, ammonia, ni- tic reveals that too many workers are
trogen dioxide, nitric oxide and vola- The OSHA standard not employing appropriate protection
tile organic compounds (VOCs). Some The appropriate use of respiratory pro- from respiratory hazards on the job.
hazardous gases, such as carbon tection is not only important for worker According to OSHA, an estimated 5
monoxide, act rapidly and can cause health and safety in the U.S., it is million workers are required to wear
unconsciousness or death within min- the law. The U.S. Occupational Health respirators in 1.3 million workplaces
utes, while other toxic gases can take and Safety Administration (OSHA) is throughout the U.S.
28 Chemical Engineering www.che.com December 2012
protection in the workplace, including in-house medical, safety-engineering
the following: or fire-prevention departments should
Respiratory protection program re- administer the program in liaison with
quirements the program administrator. In smaller
Standard operating procedures for plants that do not have specialists, an
programs upper-level superintendent, foreman
Respirator selection, limitation and or qualified person must serve as the
use program administrator.
Training Employee training. Any respira-
Fit testing tor program should stress thorough
Maintenance, inspection, storage training of all respirator users. Em-
and disposal ployees must be aware that a respi-
Breathing air and oxygen-deficient rator does not eliminate the hazard.
atmospheres If the respirator fails, the user will be
Recordkeeping overexposed to dangerous substances.
To reduce the possibility of failure, the
Establishing a program respirator must fit properly and be
FIgure 2. This is an example of a The fundamental goal of any respira- maintained in a clean and serviceable
full-facepiece respirator that covers the tory protection program is to control condition. Employers and employees
wearers face from the hairline to below
occupational diseases and injuries must understand the respirators pur-
the chin
caused by breathing air contaminated pose and limitations. Users must not
with harmful dusts, fogs, fumes, mists, alter or remove the respirator even
The role of ANSI gases, smoke, sprays and vapors. The for a short time, regardless of its po-
In addition to OSHA, another driv- defense against these contaminants is tential discomfort.
ing force in the regulatory arena is simple: keep them out of the workers Regular evaluation. Employers
the American National Standards In- breathing air. must evaluate the effectiveness of
stitute (ANSI). As a voice of the U.S. Engineering and administrative their companys respirator program
standards and conformity assessment controls, which have the potential to regularly and modify the written op-
system, ANSIs mission is to enhance completely eliminate the hazard, must erating procedure as necessary to
both the global competitiveness of always be implemented first. These reflect the evaluation results. Work-
U.S. business and quality of life by include actions such as enclosing or place conditions that affect respira-
promoting and facilitating voluntary confining the contaminant-producing tory hazards and respirator use may
consensus standards and conformity operation, exhausting the contami- change over time. This would include
assessment systems, and safeguard- nant, or substituting the contami- new work processes or techniques,
ing their integrity, in an effort to pro- nant with a less-toxic material. Work the use of new or different materials
tect workers. practices such as limiting employee or chemicals, changes in the amount
ANSI thus provides a neutral exposure should also be considered. of a respiratory hazard present in the
forum for the development of policies If these controls do not eliminate the workplace and changes in the types
on standards issues and serves as a hazards, then employers must provide of respirators being used. The written
watchdog for standards development appropriate respiratory protection for program must therefore be updated as
and conformity assessment programs every employee who might be exposed necessary to reflect and address these
and processes [2]. ANSI/AIHA Z88.2 to them. changes. Employees should always no-
Standard Practices for Respiratory Whenever OSHA standards or em- tify their employer if a change in the
Protection is a voluntary consensus ployers require respirator use, there workplace occurs that conflicts with,
standard drafted by the ANSI Z88 must be a complete respiratory protec- or may not be covered by, the exist-
Committee on Respiratory Protection tion program in place. Employers must ing respirator training or established
[3]. This standard is significant, as it have written operating procedures to workplace policies or procedures.
addresses the issue of assigned protec- ensure that employees use the respi-
tion factors (APFs). APFs are the main rators safely and properly. Users must Choosing the correct respirator
factor used by many employers when be familiar with these procedures as Selecting the right equipment is a
they select respirators for worker pro- well as the types of respirators avail- three-step process:
tection from airborne hazards. The able and their limitations. 1. Define the hazard and determine its
ANSI standard recommends protec- A qualified program administra- extent.
tive factors that are very similar to tor. First of all, a qualified program 2. Evaluate user factors that affect res-
those of OSHA, except for a more con- administrator must be responsible for pirator performance and reliability.
servative APF for filtering facepieces the program. This person must know 3. Select an appropriate respirator that
(disposable respirators). The standard enough about respirators to supervise is certified by the National Institute
also addresses other key issues re- the program properly. Larger plants for Occupational Safety and Health
garding the proper use of respiratory or companies with industrial-hygiene, (NIOSH) [1]. Note that equipment
Chemical Engineering www.che.com December 2012 29
Written respirator program
sec (c)
Cover Story
Hazard assessment
sec (d)
A
dditional information on protecting yourself from airborne and what is needed for a respirator to provide protection
hazards can be obtained from the U.S. Occupational Safety www.osha.gov/SLTC/etools/respiratory/index.html OSHAs
and Health Administration (OSHA), the National Institute for eTool, Respiratory Protection (1998), which provides employ-
Occupational Safety (NIOSH), and the American National Stan- ers with instruction on the proper selection of respiratory pro-
dards Institute (ANSI). Safety product manufacturers websites can tection and the development of change schedules for gas/
also be a helpful resource. Here are a few useful website links: vapor cartridges
www.osha.gov/law-regs.html For OSHAs standards www.osha.gov/Publications/OSHA3079/osha3079.html
www.cdc.gov/niosh The home page for the and Healths OSHA guidebook, Respiratory Protection, publication no.
National Personal Protective Technology Laboratory, which con- 3079, (2002). This booklet clarifies key provisions of the respi-
tains a variety of information and links related to respiratory rator protection standard, which was revised in January 1998.
protection and other personal protective equipment It also provides specific guidance on respirator selection, fit test-
www.ansi.org The home page for ANSI ing, hazard evaluation and medical evaluation
www.osha.gov/dte/library/materials_library.html OSHA 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
Directorate of Training and Educations Training and Refer- -06-27-2007.html OSHA quick card, Respirators Quick
ence Materials Library. Provides outreach training materials Card, publication no. 3280 (2005), which provides employ-
(such as slide presentations) for OSHAs respiratory protection ees with a brief description of the different types of respira-
standard, as well as links to additional resources about respi- tors available
ratory protection w ww.cdc.gov/niosh/npptl/topics/respirators/disp_part/
webapps.msanet.com/ResponseGuide/ Website from au- RespSource.html NIOSH Web page, NIOSH Respirator
thors company that contains two interactive tools: Chemical Trusted-Source Information Page, which provides information
Database & Respirator Selection and Cartridge Service Life Cal- to understand the types of respirators, how to identify approved
culator, as well as a complimentary download entitled Manag- models and outlets for purchase, a listing of all NIOSH-approved
ing a Respiratory Protection Program, a guide to simplifying and U.S. Food and Drug Admin. (FDA) -cleared surgical N95
compliance with 29 CFR Part 1910.134 respirators, a listing of recently revoked respirator approvals
w ww.osha.gov/dts/shib/respiratory_protection_bulletin and relevant user notices. It also contains information on how
_2011.html OSHA bulletin, General Respiratory Protection to implement the use of respirators in the workplace and use
Guidance for Employers and Workers (2011), which provides them appropriately, including commonly asked questions and
basic information to employers and workers who may find answers (fact sheets), respirator myths, the science of respirator
themselves using respiratory protection for the first time. It also function and performance, and respiratory protective devices
provides information on what respirators are, how they work, not approved by NIOSH
that every employee can select an ac- introduction of a harmless, but strong, ing the tightness of the connections.
ceptable respirator that fits properly. smell or irritating substance into the SCBAs should be inspected at least
Corrective eyeglasses worn by breathing zone around the respirator monthly, and air and oxygen cylinders
employees can also present a prob- being worn. If no odor or irritation is should be fully charged according to
lem when fitting respirators. Special detected by the wearer, then this indi- the manufacturers instructions. This
mountings are available to hold cor- cates a proper fit. inspection should include a check of
rective lenses inside full facepieces. Quantitative fit testing offers more regulator and warning devices to en-
A qualified individual must fit the accurate, detailed information on res- sure proper functioning. Employers
facepiece and lenses to provide good pirator fit. While the wearer performs must keep a record of inspection dates
vision, comfort and proper sealing. exercises that could induce facepiece and results.
Tight-fitting respirators cannot pro- leakage, a fit-testing instrument nu- Chemical cartridges and gas mask
vide proper protection without a tight merically measures the amount of canisters should be replaced as neces-
seal between the facepiece and the leakage into the respirator. This test- sary to provide complete protection,
wearers face. Consequently, beards ing can be done either by generating following the manufacturers recom-
and other facial hair, the absence of a test aerosol as a test atmosphere, mendations. Additionally, mechanical
normally worn dentures, facial defor- using ambient aerosol as a test agent, filters should be replaced as necessary
mities, jewelry or head gear that proj- or using controlled negative pressure to avoid high breathing resistance.
ects under the facepiece seal can also to measure any leakage. Detailed in- Only an experienced person is per-
seriously affect the fit of a facepiece. structions for performing both quali- mitted to make repairs, using parts
To ensure proper respiratory protec- tative and quantitative fit testing are specifically designed for the respira-
tion, check the facepiece each time you contained in Appendix A of OSHA 29 tor. This person must consult the man-
wear the respirator. You can do this by CFR 1910.134. ufacturers instructions for any repair,
performing either a positive-pressure and no attempt should be made to re-
or negative-pressure user-seal check. Proper inspection and care pair or replace components or make
Detailed instructions for performing It is important to inspect all respi- adjustments or repairs beyond the
these checks are in Appendix B-1 of rators for wear and tear before and manufacturers recommendations.
OSHA 29 CFR 1910.134. after each use, giving special atten- Employers must ensure that res-
Qualitative versus quantitative tion to rubber or plastic parts that pirators are cleaned and disinfected
fit testing. Fit testing is required for can deteriorate or lose pliability. All as often as necessary to keep them
tight-fitting facepiece respirators. You parts including the facepiece, head- sanitary. In addition, the employer
can test the effectiveness of the fit of band, valves, connecting tube, fittings must have emergency-use respirators
the facepiece two ways: qualitatively and cartridges, canisters or filters cleaned and disinfected immediately
and quantitatively. must be in good condition. A respi- after each use.
Qualitative fit testing involves the rator inspection must include check- Respirators should be washed in
Chemical Engineering www.che.com December 2012 31
Cover Story
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 FIGURE 1 Fire-water pump assem-
blies are typically skid-mounted to ease
installation and operation. Shown here
Follow this guidance to improve the selection, design is an example of a skid-mounted, diesel-
engine-driven, fire-water pump package
and operation of pumps handling water for fire-
ent guidelines have been used to
fighting and related systems estimate the ideal water volume
and flow requirements for CPI fire
Amin Almasi around 2035% of the insurance- cases, depending on plant specifics,
Consultant deficiency rating points for a CPI applicable codes, regulations and
plant are related to inadequate fire- fire-fighting methods. As a result,
F
ire-water pumps are criti- water pumping systems. On aver- these specifications will vary based
cal machines that save lives age, 510% of all fire-water pumping on whether the plant is using fire
and prevent chemical pro- systems in CPI plants have failed to control, fire suppression, exposure
cess industries (CPI) facili- provide satisfactory services at the cooling and so on.
ties from damage more so than time required (as evidenced during Fire-water demands are usually
any other plant components. Fire- actual fire cases or drill-type exer- calculated based on the maximum
water pumps are nearly always cen- cises). Thus, it is critical to design, rate of water that will be required for
trifugal pumps with capacity from implement, operate and maintain a worst-case scenario typically a
around 203,000 m3/h. Specific re- these critical systems beyond the potential scenario involving a large,
quirements for fire-water pumps minimum requirements set forth in single-fire incident. The most remote
are briefly noted in fire codes (such published fire codes. unit(s) from the fire-water pumps, or
as NFPA 20 [1]), but this may not the largest unit(s), are typically exam-
be sufficient for specifying high- Fire-water system ined to identify the worse possible fire
performance fire-water pumps in Fire-protection efforts are categorized scenario(s). The potential scenario of
a way that ensures good reliability as passive or active. The primary a vast fire in the largest unit should
and operation as well as optimum passive measures for fire protection be used to define the capacity of fire-
price. This article provides practi- include efforts to ensure sufficient water system. The most-remote fire
cal notes on fire-water pumps to ex- clearances, install protective barri- unit(s) should be used to define maxi-
pand upon the information that can ers, limit and protect fuel sources, mum rated pressure of the fire-water
be found in existing fire codes. and other steps designed to reduce pump.
Various styles and configurations fire risks (such as the use of less CPI fire cases can be very differ-
of fire-water pumps are available hazardous materials, processes and ent, considering the different types
at different prices. In addition to equipment). By comparison, active of materials handled and the types
proper selection, fire-water pumps fire-protection systems are designed of operations carried out at differ-
must be properly integrated into to detect and apply fire-protection ent facilities. Today, computerized
the overall fire-water system, as an measures, which usually rely on some simulations play a critical role in
integral part of the CPI facility. effort to actually extinguish the fire. identifying and modeling potential
Figure 1 shows an example of Commonly used system components fire scenarios, validating fire-fight-
a skid-mounted, diesel-engine- include fire hydrants, monitors, hose ing methodologies, and estimating
driven fire-water pump package. reels, water-spray systems, deluge- the required water capacity.
The performance and reliability of type fire-protection systems or water- When evaluating a potential
a fire-water pumping system is an exposure cooling systems. The fire-wa- fire-fighting scenario, additional
important issue, and details of the ter pump plays an important role in pressure (a safety margin to the
fire-water pumping system are usu- most active fire-protection systems. calculated head) should be added
ally a part of risk studies, HAZOP to maintain the fire-water pres-
and inspection activities. System design and sizing sure in all remote units and critical
Fire-water pumps are important There have been different sets of fire-fighting systems; this is nec-
to different stakeholders, including rules to define the required flow essary to ensure that a fire-water
clients, investors in the CPI plant, and head of fire-water pumps. In stream with adequate pressure can
and insurance providers. Usually other words, a variety of differ- be maintained to support all appli-
Chemical Engineering www.chemengonline.com DECEMBER 2014 53
Environmental Manager
duced-flow problem) and resulting ing details. However, in this au- be considered for pumps above the
inefficient operation (overheat). In thors view, NFPA 20 should be con- 350-kW range.
case of reduced flow, the differential sidered as a minimum requirement
temperature (the discharge temper- for a fire-water pump. For critical Fire-water pump arrangement
ature minus the suction tempera- fire pumps, the well-known API- The location for fire-water pumps
ture) would rise and could indicate 610 pump standard [2] is addition- should be selected carefully to mini-
such a malfunction. ally applied. mize various risks and potential
Because of small leakage and hazard situations. Explosions or
small consumption of fire water, the API-610 fire-water pumps high-hazard fires are major con-
pressure in a fire-water network The API-610 pump standard is used cerns, which can disable fire-water
could decrease slightly. Fire-water to ensure the reliable operation of pumps. Ideally, there should be
pump systems are usually designed high-performance pumps, mainly in 4080 m of clearance between fire-
in a way that spare pumps should the oil-and-gas, petroleum refinery water pumps and a hydrocarbon
be started if fire-water pressure and petrochemical sectors. The API- or chemical process unit or storage
dropped below a certain level. How- 610 is usually considered to be the area. This limit should also be re-
ever, a slight pressure drop should minimum specification for pumps spected for some utility areas, such
not lead always to the startup of a that handle hazardous, flammable, as a power-generation units, gas-
large fire pump, as this could result toxic and explosive liquids since compression units, oxygen-genera-
in many unnecessary on-off oper- any reliability issue associated with tion units, and similar.
ating cycles of the main fire water these pumps could result in a poten- The possibility of an unconfined
pump (Figure 2). tial disaster. API-610 pumps are also vapor-cloud explosion is one of the
On the other hand, small pres- very popular in applications with main concerns, as this could disrupt
sure changes resulting from varia- extreme temperatures, including utilities, damage major support fa-
tions in fire-water consumption pumps for both high-temperature cilities, and damage the fire-water
during a fire incident can result in service (such as boiler-feed-water, or pumping system. Generally, there
an unstable operation of the main BFW, pumps) and low-temperature is a great possibility of the electri-
fire-water pump(s). For instance, applications (for example, pumps cal network or the steam-distribu-
this may lead to unnecessary fast used in liquid petroleum gas (LPG) tion system failing in the event of
changing of the operating point of a liquefied oxygen, and liquefied natu- a major explosion or extensive fire
large pump, which can result in per- ral gas (LNG) service). For critical event. This underscores the critical
formance and reliability issues. (high-risk) CPI units, fire-water role of independent, diesel-engine-
Smaller-capacity pumps (known as pumps are usually specified to com- driven fire-water pumps. Fire-wa-
jockey pumps) are usually employed ply with the API-610, to ensure that ter diesel engines should generally
in conjunction with the main pump(s) they are able to achieve a high re- comply with NFPA 37 [3].
to maintain a relatively constant fire- liability level the same as other Regarding the diesel fuel-tank ca-
water pressure. Jockey pumps usu- pumps in the unit. pacity, typically, a 12-h duration is
ally initiate operation after a rela- Engineers often struggle with specified as the minimum require-
tively small pressure drop (say 0.51 whether or not to use API-610- ment. However, some critical CPI
bar) in a fire-water system. compliant fire-water pumps for plants require 24-h-capacity fuel
Main fire-water pumps are typi- a CPI plant. This decision would tanks for each fire-water pump-die-
cally electrically driven and the depend on the application, pump sel engine.
spare (backup or reserve) fire-water head, power rating, capacity, pump Meanwhile, each diesel engine
pumps are typically driven by die- speed and expected reliability. The should be provided with independent
sel engine. A commonly used ar- main variable is CPI service (the auxiliaries and accessories, includ-
rangement for critical CPI facilities CPI plant and expected reliability). ing a dedicated fuel system and fuel
is to install six fire-water pumps, For instance, for critical units han- tank. The startup of the engine is
including two electric-motor-driven dling flammable liquids and gases, commonly managed by a battery sys-
pumps, two diesel-engine-driven API-610-compliant fire-water pumps tem (with two independent barriers).
pumps and two jockey pumps. Fire- are often preferred. The failure of a diesel engine is
water pumps are nearly always pro- For a fire-water pump with dif- usually the result of a problem with
vided on a prefabricated skid. This ferential pressure more than 20 one of the auxiliary systems. Major
packaging concept can help to ease bar, API-610 is usually specified. reasons for such a failure include
the alignment and installation is- The pump power rating is a bit fuel-system issues, a lubrication-
sues and ensure high reliability. tricky, since there are many non- system problem, a starting issue,
As noted earlier, the fire code API fire-water pumps available a wiring problem or component fa-
NFPA 20 is dedicated to fire-water (with successful references) that tigue. Only clean, high-quality die-
pumps. It specifies proper require- are intended for high power ranges sel fuel should be used, and special
ments for pump tests, pump per- in a wide array of industry applica- attention is required for the lubri-
formance curves, pump accessories tions. As a general rule of thumb cation oil selection and supply.
and auxiliaries, and some packag- for many CPI plants, API-610 can Proper overhauls and repairs
Chemical Engineering www.chemengonline.com DECEMBER 2014 55
Environmental Manager
are required, just like for any other dition to the conventional fire-water tightening) is to install dial indica-
properly designed combustion en- pumps to supply seawater (for tors (or other types of indicators)
gines. Experience has shown that CPI plants located at the coastal re- that monitor movements in critical
the diesel-engine-driven, fire-water gions) or other sources of untreated parts of the machinery train. Usu-
pump is the most reliable option raw water (such as untreated water ally, two dial indicators are used to
currently available for severe loss from a lake or water wells), to observe movement in each machin-
incidents in a CPI plant. The reli- quickly supply additional capacity ery component (such as the driver,
ability and availability of micro- to the plants fire-water network. the fire-water pump and the gear
turbines (small gas turbines in the For a seawater-based emergency unit, if any) compared to the base-
50400 kW range) could be higher fire-water pumping system, the plate or foundation (the main pur-
than diesel engines, but their ef- pumps are usually submerged in pose is to identify improper support
ficiencies are relatively lower (in seawater. For some locations, the of machinery, called soft-foot).
terms of lower operating duration seawater level may fluctuate from Two dial indicators can be used
with the same amount of fuel). Cur- 7 m to +16 m. Considering that to monitor critical bearing housing
rently they are not popular for fire- there is often a long distance from movements in the fire-water pump
water pump systems. the sea to the CPI facility, these fire- (usually in x and y directions). Ac-
Fire-water pumps are typically water pumps should be designed ceptable movements should be
arranged for both manual and au- to produce a relatively high head. below 0.04 mm (40 micrometers) to
tomatic startup. Automatic startup These special fire-water pumps are ensure a proper piping-pump con-
is expected to happen rapidly, in a usually multi-impeller pumps. Prop- nection. A similar limit should be
very reliable manner, once a fire erly designed highly reliable pumps applied to movements in all critical
event has been detected. Fire-water are always specified for such com- pump train locations (such as bear-
pumps are usually stopped manu- plex service. Electric-motor-driven ing housing, coupling, machinery
ally at the pumps local control submersible pumps, or sometimes support and others), and suction and
panel. In other words, operator in- hydraulic-driven pumps are used discharge nozzle flanges (in terms
tervention is usually used to turn for these special applications. These of limiting deformations in all direc-
off the pump, once the situation has are usually down-hole, vertical tur- tions). For special fire-water pumps,
stabilized and the fire is out. bine-type pumps. depending on the machinery design,
A suitable enclosure (or building) In some cases, local regulations speed, power rating and applica-
should be provided for fire-water or plant specifications require three tions, a limit higher or lower than
pumps. Sufficient reinforcement dedicated fire-water pumps (as the the above-mentioned (0.04 mm)
should be considered for the fire- minimum) for special CPI applica- may be specified. n
pump enclosure. This is very im- tions (such as CPI plants that han- Edited by Suzanne Shelley
portant. For example, in the case dle highly explosive or highly flam-
of a major earthquake, fire-water mable materials). In such cases, one References
pumping systems need to be fully of the main concerns is personnel 1 NFPA 20, Standard for the Installation of Sta-
tionary Pumps for Fire Protection, National
operational to respond to fire events safety during a major fire case and Fire Protection Assn. (NFPA), 2013.
resulting from the earthquake. An provisions must be made to ensure 2 API 610, Centrifugal Pumps, American Petro-
open-sided shelter is not recom- a safe personnel evacuation. leum Inst. (API), 2009.
3 NFPA 37, Standard for the Installation and
mended. And, fire-water pumps Use of Stationary Combustion Engines and
should be located at a higher eleva- Installation and commissioning Gas Turbines, National Fire Protection Assn.
(NFPA), 2014.
tion than the majority of the CPI The piping installation and connec-
facility and upwind of it. tion to a fire-water pump can lead
Author
To provide another layer of protec- to relatively high loads on the pump Amin Almasi is a rotating-
tion (in order to avoid common fail- nozzles and to the pump trains sen- equipment consultant in Aus-
tralia (Email: amin.almasi@
ures), the main fire-water pumps, sitive components, such as bearings, ymail.com). He previously
and any other reserve or support- coupling and rotating assemblies. worked at Worley Parsons Ser-
vices Pty Ltd. (Brisbane, Aus-
ing fire-water pumps should not be The fire-water pumps are often tralia), Technicas Reunidas
located immediately next to each left on standby, therefore any high (Madrid, Spain) and Fluor
Corp. (various offices). He
other. Locating fire-water pumps at nozzle loads or misalignment might holds a chartered professional
two separate locations can improve be left unchecked. This can poten- engineer license from Engi-
neers Australia (MIEAust
both the fire-water system reliabil- tially wreck the pump in the first CPEng Mechanical), a chartered engineer cer-
tificate from IMechE (CEng MIMechE), RPEQ
ity and overall fire-water network hours of operation in the event of (registered professional engineer in Queensland)
hydraulic behavior. fire. Periodic checks are important. and he also holds M.S. and B.S. degrees in me-
chanical engineering. He specializes in rotating
A well-known method to monitor machines including centrifugal, screw and recip-
Special fire-water pumps the movements and deformations of rocating compressors, gas and steam turbines,
pumps, condition monitoring and reliability. Al-
For critical CPI plants, additional critical machinery components dur- masi is an active member of Engineers Australia,
emergency (or reserve) firewater ing the piping connection (piping IMechE, ASME, Vibration Institute, SPE, IEEE,
and IDGTE. He has authored more than 60 pa-
pumps should be provided in ad- flange and machinery flange bolt pers and articles dealing with rotating machines.
Eye-and-Face Personal
Protective
Equipment
Gateway Safety
E
the chemical process industries (CPI).
ye injuries of all types occur relatively inexpensive compared to The general requirements of this stan-
at a rate of more than 2,000 the potential expense of an injury. But, dard include that employers select
per day in the U.S., and half no dollar figure can adequately reflect PPE for their employees based on the
of those occur in the work- the personal toll these accidents take hazards present in the workplace, and
place. In 2006, the U.S. Bureau of on injured workers. Prevent Blind- that employers train employees on the
Labor Statistics (BLS) reported over ness America estimates that 1020% proper use and limitations of the PPE
52,000 workplace injuries to the eyes of work-related eye injuries result in selected for use. In November 2007,
and face, and approximately 70% of temporary or permanent vision loss. OSHA updated the standard to spe-
those were eye injuries [1]. The need This article outlines the basics of cifically require employers to pay for
to protect the face and eyes in the eye-and-face protection programs for the PPE required under the standard,
workplace is clear. plant operations, including key com- with some limited exceptions [3].
Implementing an eye-and-face pliance requirements and practical The OSHA Eye and Face Protection
protection program is important for information for personal protective Standard (29 CFR 1910.133) was estab-
several reasons. One is compliance. equipment (PPE) selection. lished specifically to protect employees
Between October 2006 and Septem- from eye and face hazards in the work-
ber 2007, more than 450 citations re- The causes of injuries place. To comply with this standard,
sulted in more than $215,000 in fines Most impacts to the face come from employers must ensure the following:
and penalties to manufacturers for flying objects, usually particulate 1. Employees who may be exposed to
failure to comply with the U.S. Occu- matter. When a particle hits the face it eye or face hazards from flying parti-
pational Safety & Health Administra- might cause a cut or scratch, and the cles, molten metal, liquid chemicals,
tion (OSHA) standard for eye and face skin will repair itself over time given acids, caustic gases, vapors or poten-
protection [2]. proper medical attention. If the par- tially injurious light radiation use
Another reason is the cost of eye ticle hits the eye, it causes direct dam- appropriate eye or face protection.
and face injuries. The financial cost age, and it is far more difficult for the 2. Employees who may be exposed to
of these injuries is enormous and can eye to repair itself. a hazard from flying objects use
range from $300 to $3,000 per case. BLS reports that of the nearly 36,000 eye protection that provides side
More than $300 million per year is lost injuries to workers eyes in 2006, nearly protection.
in production time, medical expenses, 70% resulted from contact with objects 3. Employees who require prescrip-
and workers compensation according like flying particles, falling objects or tion lenses while engaged in op-
to OSHA. The return on investment sparks striking the eye. Contact with erations that involve eye haz-
for implementing an eye-and-face chemicals was responsible for the re- ards wear eye protection that
safety program is huge as its cost is maining 30% of eye injuries. incorporates the prescription in
48 Chemical Engineering www.che.com February 2009
Gateway Safety table 1. Hazard Assessment
Hazard type Examples of hazard Common related tasks
Impact Flying objects, such as Chipping, grinding, machining, masonry
large chips, fragments, work, wood working, sawing, drilling, chisel-
particles, sand and dirt ing, powered fastening, riveting and sanding
Heat Anything emitting ex- Furnace operations, pouring, casting, hot
treme heat dipping and welding
Chemicals Splash, fumes, vapors Acid and chemical handling, degreasing,
and irritating mists plating and working with blood
Dust Harmful dust Woodworking, buffing and generally dusty
conditions
Optical Radiant energy, glare Welding, torch-cutting, brazing, soldering and
radiation and intense light laser work
Source: U.S. Dept. of Labor, Occupational Safety & Health Administration (OSHA), Eye and
FIGURE 2. Face Face Protection eTool, http://www.osha.gov/SLTC/etools/eyeandface/ppe/selection.html
shields protect the
entire face, but do not 2. Identify the sources of the haz- may have been manufactured to
protect the eyes from ards. Talk to employees as they meet earlier editions of ANSI Z87.1.
impact hazards when
worn alone. Safety glasses know best about their trades, OSHA requires that eye and face
or goggles should also be worn work areas and processes. PPE purchased after July 1994 be
Consider all possible hazards, manufactured to meet the 1989 edi-
even those that seem minor or tion of ANSI Z87.1.
its design, or that they wear eye unlikely. Some examples of eye and Eye and face PPE that meets the cri-
protection that can be worn over face hazards and typical sources are teria specified in ANSI Z87.1 must be
the prescription lenses without summarized in Table 1. marked with a manufacturers unique
disturbing the proper position of 3. Organize the data collected. De- marking and the Z87 marking. Spe-
the prescription lenses or the pro- velop a table or spreadsheet listing cial purpose lenses and shaded lenses
tective lenses. the occupations, work areas, and must also be marked indicating the
4. Eye-and-face personal protective processes you identify, the hazards purpose and shade. Eye and face PPE
equipment is distinctly marked so identified in each category, and the must be selected based on the hazards
that the manufacturer is clearly sources of the hazards. identified in the hazard assessment.
identified. 4. Analyze your data, considering the Some examples of the types of protec-
5. Employees who may be exposed probability and severity of each tion available for hazards are summa-
to injurious light radiation during hazard. Keep in mind that mul- rized in the next sections.
work process, such as welding, ther- tiple hazards and sources may be Ultimately, the eye and face pro-
mal cutting, brazing, torch solder- present. This will help in deciding tection must be comfortable to wear
ing, and using lasers, use eye protec- which PPE is appropriate. and not interfere with the employees
tion equipped with filter lenses that 5. Certify your assessment. OSHA re- work. Poor-fitting PPE will not afford
have a shade number appropriate quires that every employer verify its the necessary protection, may be un-
for the work being performed. assessment, identifying the areas comfortable to wear, and may not get
assessed, who performed the assess- used. Since eye-and-face protection
implementation ment, and the dates it was performed. devices can come in different sizes,
Implementing a practical, effective 6. Review and update your assessment care should be taken to ensure that
PPE program includes the following on a routine basis, or if changes in the the right size is selected.
straightforward steps, which are de- workplace or processes are planned. Impact hazards. Safety spectacles
scribed in more detail below. (Figure 1) are intended to shield the
1. Assess the workplace for hazards. PPE selection wearers eyes from impact hazards,
2. Select PPE based on these hazards. The American National Standards such as flying fragments, objects,
3. Train workers on proper PPE use. Institute (ANSI) standard for eye and large chips, and particles. Workers
4. Determine how prompt emergency face protection (Z87.1) sets forth cri- are required to use eye safety spec-
care will be provided in the event of teria related to the description, gen- tacles with side shields when there is
an accident. eral requirements, testing, marking, a hazard from flying objects. Non side-
selection, care and use of eye and face shield spectacles are not acceptable
Hazard assessment PPE in the workplace. The 2003 edi- eye protection for impact hazards. The
OSHA provides guidance on conduct- tion specifies requirements for safety selection of safety glasses is almost in-
ing hazard assessments, which are glasses, safety goggles, face shields finite as a host of manufacturers have
summarized as follows [4]. and full-face and hooded respirators. stepped up to meet employee demand.
1. The first step in the hazard assess- This ANSI standard is revised and Safety goggles are also intended to
ment for eye and face protection is updated routinely to reflect current shield the wearers eyes from impact
to walk through your facility and technologies and the current state of hazards. Goggles fit the face immedi-
identify the potential hazards that practice. While most PPE manufac- ately surrounding the eyes and form
can injure the eyes or face. Hazards turers have adopted the 2003 edition a protective seal around the eyes. This
should be assessed by occupation or of the standard in new products, eye prevents objects from entering under
trade, work area and work activity. and face PPE currently being used or around the goggles.
Chemical Engineering www.che.com February 2009 49
Environmental Manager
Managing Compliant
MSDSs and Labels
Compliant MSDS programs
depend on inventory management, Figure 1.
The advent of GHS and R
regulatory knowledge and consistent authoring EACH regulations brings new
challenges to the already demand-
ing and mission-critical task of
Jytte Syska and Tamie Webber authoring compliant MSDSs
3E Company
C
hemical regulatory compli- ufactured within the European longer possible unless the im-
ance is growing in complexity Union (E.U.), imported into the porter or manufacturer has reg-
as product sourcing and for- E.U., or purchased from a supplier istered under REACH. Many
mulations proliferate and or- within the E.U. For REACH, of the substances that were
ganizations expand into new markets companies also must identify registered for the Decem-
on a global scale. Staying current with each substances tonnage and ber 1, 2010 deadline have
the requirements of international reg- current classification. exposure scenarios. The
ulatory initiatives such as the Glob- Comprehensive regulatory re- registrant must attach the ap-
ally Harmonized System of Classifica- porting is also crucial to achieving propriate exposure scenarios to
tion and Labeling of Chemicals (GHS) GHS compliance. In addition to their MSDSs, and the company
and the Registration, Evaluation and consistently tracking regula- using these substances must
Authorization of Chemicals (REACH) tions, companies must track then include the informa-
adds to the challenge (CE, March 15, modifications made to ingre- tion from the exposure sce-
2008, p. 3847, and CE, June 15, 2008, dients, products, and quanti- narios when they prepare
p. 2630). ties and inventories. and supply the MSDSs to their
A key compliance requirement of While authoring accurate and customers. The European Chemi-
these initiatives is the proper man- compliant MSDSs has always cal Agency (ECHA) has recom-
agement of Material Safety Data been a mission-critical obliga- mended a four-section expo-
Sheets (MSDSs) and labels. Both GHS tion, the advent of REACH sure-scenario format, but
and REACH significantly impact the and GHS brings new chal- many different formats are
corporate strategy for MSDS man- lenges to this already highly being used. Work is ongoing
agement, as the regulations require demanding task. For higher- to harmonize the information in
affected companies to update these volume hazardous substances, the exposure scenarios and to
documents to maintain compliance. REACH requires companies to establish a common XML for-
Comprehensive inventory manage- attach an exposure scenario mat for data exchange.
ment, thorough regulatory reporting to the MSDS. The exposure As a step toward improved
and consistent MSDS and label au- scenario describes how the compliance, companies should
thoring are three critical components chemical can be used in a safe obtain the necessary overview of
of a REACH- and GHS-compliant manner, with minimal risk, for the the current regulations and detailed
MSDS management program. intended use. The exposure sce- understanding of how they relate
nario can cover either each to their chemicals and their
MSDS management individual substance in the supply chain. Manufacturers,
Accurate chemical inventory manage- chemical product or describe users and distributors should
ment is the cornerstone of any MSDS the mixture as a whole. be aware not only of their own
management program. Regulatory The first registration dead- obligations but also of those to
compliance mandates that companies line under REACH passed on downstream customers and to em-
obtain and maintain MSDSs for the December 1, 2010. Import or ployees. In addition, corporate
raw materials, as well as for the full manufacture of high-volume product-stewardship staff
composition of mixtures. For REACH chemical substances and of must understand when these
compliance, this requirement helps CMR- (carcinogenic, muta- regulations are implemented
companies establish the category into genic and toxic-to-reproduc- by country in order to know
which individual substances fall: man- tion) classified substances is no how to properly plan.
Chemical Engineering www.che.com february 2011 43
Environmental 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 Seven essential elements cedures to start up and shut down a
be a top priority every day, During the periods mentioned above, plant. New procedures are certainly
there are certain times when it is particularly important for manag- needed when the unit operations are
extra vigilance is needed to ers to include the following elements different in plants where a different
make sure that safety practices are in their management of plant safety. technology is employed. I have seen
being maintained. This article out- 1. Review of technology. Whenever a wide variety of processes for eth-
lines three periods deemed to be criti- a plant is acquired, a thorough review ylene, polyesters, vinyl chloride and
cal for plant safety and specific items of the acquired technologies must be polyvinyl chloride (PVC). There are
to pay attention to in these periods. made. Typically, companies feel con- similar examples of various processes
fident in acquiring facilities when in petroleum refining. The operating
Three critical periods they are operating similar plants or procedures also differ depending on
The following three phases in a pro- manufacturing similar products. For the complexity of the plant. Operat-
duction plants life are potentially crit- example, a petroleum refining com- ing procedures should be updated on
ical in regards to safety practices: pany taking over another refinery a regular basis.
Transition. The first period deemed has good knowledge of refinery opera- 3. Review of safety interlock and
to be critical is the transition phase tions. However there are several dif- shutdown systems and procedures.
when a company acquires a production ferent process technologies available The safety interlocks and shutdown
plant, petroleum refinery or any other for the same operations or reactions. systems are developed after compre-
manufacturing facility. During this pe- For example, there are many different hensive hazards and operability stud-
riod many experienced people leave the alkylation processes in operation to ies. Many are introduced after actual
company while new management is produce motor and aviation fuels. plant experience. These interlock sys-
learning the details of plant operations. Similarly in petrochemical produc- tems must be reviewed and upgraded
Restructuring. The second period is tion, such as for vinyl chloride, there if required.
a phase of restructuring, when due to are many routes for the same reactions All major companies control and
slow economic conditions the demand and intermediate products. Some tech- operate their plants based on their
for a product has slowed down. The nologies employ fixed beds for oxychlo- in-house development. A thorough
company needs to save operating and rination while others use fluidized bed review will certainly improve the ex-
maintenance costs. During this period reactors. A review of process technolo- isting systems due to the combination
some sections or the whole plant may gies will reveal the differences in the of expertise from merged or acquired
be shut down. technology and operating procedures. companies. I have seen many times
Routine. A third period that comes to Once the differences have been identi- that safety interlocks are bypassed
mind is when the plant operation be- fied, then all plant operators must be because new plant operators did not
comes routine and inexperienced peo- trained in all new technologies. understand the rationale behind the
ple are allowed to take over and run the 2. Plant operating procedures. Ex- safety interlocks.
facility. The experienced staff relaxes perience shows that the same plant 4. Review of corrosion control sys-
and becomes less vigilant in reviewing can be started in a number of ways. tems. Control of corrosion in plant
plant safety on a regular basis. All companies develop operating pro- equipment is intricately linked with
50 Chemical Engineering www.che.com January 2011
Revise #1 - 2/4/10
6:00 PM
Centrifuge & Drying
Technologies
Plant Safety Recommendations Inverting Filter Centrifuge
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 shows that the same problems occur
corrosion protection systems employed over and over again in the same fa-
Cutting edge centrifuge technology for
in chemical and petrochemical indus- cility. People are alert for a few years filtration, washing and drying of solid/liquid
tries. These systems are developed after an accident. Then safety reports suspensions
after years of plant experience. This get filed away and forgotten. A new t Widest range of applications - hardest to
easiest filtering products can be handled
is an area that is often ignored when company taking over old facilities is t No residual heel for exact repeatable
chemical plants or petroleum refiner- busy with big ticket items. Plant safety batches and no loss of product
t PAC technology allows drying of the
ies are sold to new owners. records and reports are reviewed at a product inside of the centrifuge
I have seen plants with near-perfect much later stage. t Thin cake filtration operation allows for
safety records lose that record within I recommend a thorough review of a improved quality and production rates
t Full containment eliminates operator
months of acquisition by new own- plants past safety history, followed by exposure
ers. Looking deeply into the reasons, I a plant safety audit that looks at the t Effective automated CIP
found that plant engineers and opera- particular areas where past accidents
tors did not understand the function of have occurred.
Kilo-Lab Conical
Vacuum Dryer-Mixer
various corrosion protection systems. 7. Implementation of EPA (U.S.
In a vinyl chloride plant, for example, Environmental Protection Agency)
the overhead condenser for an ethyl- and OSHA (U.S. Occupational
ene dichloride distillation column was Safety and Health Admin.) Stan- Advanced technology
protected by the circulation of dilute dards. This should be a priority in for real Kilo size
caustic solution. After a new owner any plant, and needs to be given top drying research
took over the facility, the circulation priority during the plant shutdown
of caustic was stopped because the in- period. I have seen vent scrubber sys- and development
experienced engineers did not under- tems shut down because a plant is
stand its role in preventing corrosion not routinely sending any hazardous
in the condenser and overhead piping. vapors to the vent scrubber. However, t Utilizes interchangeable agitator systems
5. Training of plant engineers and there may be some storage tanks or either orbiting arm screw or central shaft
operators. This area also gets side- vessels containing hazardous materi- t Flexible small scale volume of 150ml to
1500ml
lined, particularly in old facilities. Ex- als. These materials are vented to the t Plastic view through vessel available
perienced engineers take it for granted scrubbers in overpressure conditions. t Designed for real laboratory requirements
that young engineers will learn with If the vent scrubbers are shut down, of size, with full instrument & data
recording
hands-on experience. They can cer- then severe environmental and health t Direct scale up to production or pilot size
tainly learn the plant operation, but related incidents can occur. units
may not understand why the system Edited by Dorothy Lozowski Horizontal & Vertical
was designed in a certain way. Centrifuges
There should be an ongoing training Author
Nasim Hassan is a con-
program for new engineers and even sultant (HSN Associates
experienced engineers if new tech- of Delaware; 7 Signal Hill
Drive, Hockessin, DE, 19707;
nology is introduced in a plant. The Phone: 302-239-9272; Email:
review of various reactions and side nhassa@yahoo.com; Website:
www.hsnrcs.com) in chemical
reactions, optimum operating condi- technology licensing, plant
tions, limits and safety precautions safety, process analysis, pro-
cess simulation and process t Size ranges from 200mm to 1800mm
should be essential for every person. design. He has over 35 years t Wide range of standard & custom designs
experience in the design and t Laboratory size equipment
In older facilities, there is a wealth operation of chemical and petrochemical plants.
of information available but only ex- Hassan has worked for BE&K Engineering (a
Lab Testing Available
KBR subsidiary), ICI Americas and Stauffer
perienced people know how to access Chemical Co. as senior staff engineer and proj- Rental & Lease Machines Available
it. Training materials should be easily ect manager. He has designed and started up a
number of chemical and petrochemical plants
accessible to all new operators. for licensees in Africa, Asia, Europe and the U.S. www.heinkelusa.com
6. Review of past incidents and in- Hassan has an M.S.Ch.E. from the Institute of
Chemical Engineering, Punjab University, Paki- Tel: 856-467-3399
vestigative reports. My experience stan and is a member of AIChE.
Circle 11 on p. 58 or go to adlinks.che.com/35061-11
Suzanne Shelley
Figure 1. Rising temperature is often Figure 2. Problems such as over-
Precision Prose, Inc. an indicator of operational problems in heated bearings can be diagnosed with
many types of machinery. The ability to a thermal imaging camera, which pro-
R
ising temperatures and rapid gather and analyze temperature data in vides an alternative to direct-contact
or excessive heat buildup are realtime using non-contact options can temperature sensors, especially for
help operators to pinpoint issues and components that may be hard to reach,
useful markers for determin- act accordingly inaccessible or potentially hazardous
ing the operational health
of many types of machinery and
components that are used in a wide ance, misalignments, insufficient ous. In this way, these temperature-
array of industrial and manufactur- cleaning or lubrication, friction and monitoring devices help to enable
ing settings. The types of mechani- electrical problems. realtime temperature measure-
cal and electrical systems for which The traditional approach to tem- ment while ensuring worker safety
temperature increases often signal perature monitoring in industrial (Figures 1 and 2).
problems include (but are not lim- settings relies on direct-contact
ited to): rotating machinery, such as temperature sensors, such as ther- Technology options
motors, turbines, compressors, and mocouples and resistance tempera- The two types of remote thermal-
their bearings, couplings and gear- ture detectors (RTDs). While these sensing options infrared ther-
boxes; other types of process equip- devices are certainly proven and ac- mal-imaging cameras and IRTs
ment, such as pumps, valves, heat curate, they are not appropriate for are widely used for tempera-
exchangers, steam traps, heaters, use with certain types of equipment ture assessment in industrial and
conveyors belts, rollers, furnaces components or in some types of in- manufacturing facilities. In addi-
and more; steam and electrical dustrial settings. tion, two different types of IRTs are
heat-tracing systems; insulation on By contrast, remote or non-con- available conventional IRTs, and
pipes and vessels; refractory lining tact temperature-measurement so-called scanning IRTs. Used sepa-
systems for high-temperature sys- devices, such as infrared thermal- rately or together, these devices can
tems and much more. imaging cameras and infrared help users across numerous indus-
Because equipment malfunctions thermometers (IRTs) allow use- try sectors to quickly and easily
and abnormal or fault conditions in ful temperature data to be easily assess the thermal condition of ma-
mechanical and electrical systems gathered from remote locations chinery, process systems, pipelines
are often forewarned by a rise in and thus offer a useful alternative and more.
temperature, the ability to gather to direct-contact temperature sen- Conventional IRTs are best
and analyze temperature data in sors. The ability to safely carry out suited for applications that re-
realtime can often help operators to temperature sensing from a dis- quire accurate spot-temperature
both pinpoint existing performance- tance, using either a thermal cam- readings, while scanning IRTs and
related issues and identify the era or an IRT, is particularly use- thermal cameras are useful for ap-
onset of incipient problems. These ful for machinery components and plications for which knowing the
problems can be caused by an array systems that may be hard to reach, absolute temperature of a surface
of issues, including wear, imbal- inaccessible, or potentially hazard- is less important than knowing the
48 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
Figure 3. Thermal-imaging cameras through color variations in the aging cameras, IRTs offer an ideal
measure the surface temperature of the rendered image. way for operators to determine the
objects or areas being analyzed in terms Such visual displays of relative temperature of hot or cold surfaces
of the amount of infrared (IR) energy
that is emitted, transmitted or reflected temperature variations across the remotely, which is especially useful
by the object. The data can then be ren- surface of the objects gives opera- for inaccessible or hard-to-reach ob-
dered as still or video images to help tors and technicians unprecedented jects or areas.
operators interpret the result insight into the health of equip- As noted earlier, two types of
ment and systems, and help to ad- IRTs are available conventional
temperature of a particular surface dress emerging problems efficiently. spot IRTs and so-called scanning
relative to other surfaces around it. Thermograms are especially useful IRTs. Scanning IRTs allow users
The primary advantage of a ther- when thermal imaging is used as to scan an entire area or system
mal camera is its ability to display part of regular inspections, because and quickly identify those sections
the thermal spectrum of an entire they allow engineers to quickly rec- where there is a significant tempera-
area, as seen in Figure 3. ognize changes that may signal an ture differential between the actual
Infrared thermal-imaging cam- emerging problem. Thermal images temperature of that section and a
eras. Portable thermal-imaging that are captured and analyzed over pre-set temperature setpoint value
cameras are easy to use, and typically time for the same component (for that the user has programmed into
come with a pistol-grip design, as instance, a given motor or pump) the device.
seen in Figure 4. They use infrared- can help users to identify locations Using a conventional IRT with
imaging techniques to measure the of incipient malfunction or progres- an appropriate distance-to-spot
surface temperatures of the objects sive wear or deterioration. (D:S) ratio one that allows the
or areas being analyzed and can ren- Creating such a record of heat measurement area to be entirely
der the data in the form of two-di- buildup due to deteriorating con- focused within the object being
mensional images or videos images ditions allows operators or tech- measured plant personnel can
to illustrate the data. Specifically, nicians to dispatch the most ap- determine the temperature of an
these specialized cameras measure propriate intervention in a timely object at a single spot. A built-in
surface temperature in terms of the manner. These interventions in- laser-beam sighting source helps
amount of infrared (IR) energy that clude detailed inspection, trouble- the user to focus the device on
is emitted, transmitted and reflected shooting and diagnostic efforts, and the target precisely, to ensure
by the object or area being analyzed. strategic maintenance and repair measurement accuracy.
The temperature data are displayed activities. Because thermal imag- While conventional IRTs are use-
as an IR heat spectrum, using a ing is carried out at a distance, it ful for remotely gathering point-
range of colors that are correlated to enables the capture of thermal data source data about the absolute tem-
specific temperature ranges. Todays from components that are in remote perature of a given spot, scanning
thermal cameras have great sensi- or hazardous areas, thereby ensur- IRTs are useful for applications
tivity, and provide measurement ac- ing worker safety. where it is not necessarily impor-
curacies of up to 2F. Infrared thermometers. IRTs are tant to determine the absolute tem-
Thermal cameras produce a portable, non-contact devices perature of a surface, but it is useful
thermal signature or thermogram, again, typically with a user-friendly, to determine the relative tempera-
which is a two-dimensional vi- pistol-grip design. IRTs use a spe- ture of a surface or area compared
sual display that depicts the rela- cial lens to focus the thermal radia- to its surroundings.
tive temperature variations across tion that is being emitted by the ob- Todays scanning IRTs are not
the objects surface. These images ject (in the form of IR energy) onto only very affordable (typically
allow operators to quickly pinpoint an IR sensor. The embedded soft- under a hundred dollars, compared
problem areas, since temperature ware correlates those IR readings to more than a thousand for thermal
excursions, such as areas of heat to the temperature of the object cameras), they are also extremely
loss or heat gain relative to the using information about the mate- easy to use. With a point-and-shoot
surroundings, are easily displayed rials emissivity. Like thermal im- design, the user first establishes a
Chemical Engineering www.che.com January 2014 49
Engineering Practice
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 is the probable length of exposure? per million. While this is indeed a high
know much more than just the name The longer workers are exposed to a concentration for toluene in ground-
of the chemical to be handled. If an certain chemical, the greater the level water, in this case, gloves that protect
employee, for example, is working of protection required. against water would be more appropri-
with nitric acid, then PVC (polyvi- The chemical concentration must ate and cost-effective than gloves that
nyl chloride) and neoprene are the also be considered. Is the compound protect against pure toluene.
preferred glove materials. Several present only as a trace impurity? Is
questions, however, must still be an- the employee working with a pure Glove types
swered. Is the worker cleaning up chemical or a dilute solution? If the Gloves are generally categorized ac-
spills? If so, the person will require chemical is diluted, what solvent was cording to the materials used and
highly chemical-resistant gloves with used for dilution? whether they are supported or un-
good storage stability because the Companies sometimes over-specify supported (see definitions below). The
gloves may be kept in a spill-control gloves, which can lead to unecessary following glove types are available in
cabinet for several years until they expense. Once for example, a request a range of lengths, thicknesses and
are urgently needed. from a chemical company indicated finishes, while some styles are offered
Is the person handling sealed bot- the need for gloves that would resist with special modifications.
tles, which will require minimal pro- high concentrations of toluene. When
tection unless the bottle breaks? Could questioned about the application, it Unsupported vs. supported
a spill be caused by breaking a bottle? was revealed that workers would be Unsupported gloves are made of
If so, the worker will need protection exposed to groundwater with a toluene materials such as natural or syn-
from cuts as well as nitric acid. What concentration of less than 100 parts thetic latex, nitrile and neoprene and
52 Chemical Engineering www.che.com July 2008
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 Fabric gloves rather than woven fabric, because knits
hand forms directly into the glove This category includes general pur- have a natural tendency to stretch and
compound without a supporting liner pose gloves made with polyester, fit closely. Natural rubber latex is used
or fabric. This type of glove generally nylon and cotton; cut-resistant gloves to make the least expensive additive
provides the dexterity and tactile constructed with Kevlar, Dyneema elastic yarn for making tighter-fitting
sensitivity required for many chemi- and steel; stretch gloves made with gloves and cuffs. Workers, who need a
cal applications. small percentages of natural latex and tighter fit and know or suspect they
Unsupported gloves offer a broad Lycra yarns added to other fibers; and have a latex allergy, should consider
spectrum of chemical resistance, special purpose gloves with materi- switching to gloves made with a syn-
based on the material used. Nitrile als, such as thermal foam or vibration thetic alternative, such as spandex.
gloves, for instance, are excellent for foam. Special purpose gloves include
many chemical processing, oil refin- extra clean gloves and sterile gloves. Special purpose gloves
ing, food processing and petrochemi- General purpose gloves. Nylon, Specialty gloves include those made
cal applications. Gloves made with a polyester and cotton gloves are com- with thermal foam for protection from
neoprene and natural-rubber-latex fortable and protect against snags, heat or cold, vibration-absorbing foam
blend provide the protection workers punctures, cuts and abrasion. They do for workers handling power tools and
need in food processing facilities and not, however, protect against chemi- other vibrating machinery, or products
certain chemical- and pesticide-manu- cals and liquids, which is why chemi- with radiation-absorbing additives
facturing plants. cally resistant gloves may need to be such as lead or bismuth. This cat-
Supported gloves are made by dip- worn as liners in some applications. egory also includes extra-clean gloves
ping a hand form wearing a knit- Nylon is an alternative fabric that for electronics cleanrooms and sterile
ted- or woven-cloth liner into a glove may be used in cleanrooms, automo- gloves for pharmaceutical facilities.
compound such as nitrile. The liner tive paint rooms and inspection sta-
supports the compound and adds tions where there is concern about Chemical Barriers
strength to the glove. Some supported lint contamination. Descriptions of glove materials that
styles have continuous coatings to en- Cut resistant gloves. Gloves made may be used with various chemicals
sure protection from chemicals. Cotton with Kevlar typically provide the best follow, with advantages and disadvan-
and polyester may be used in various cut protection based on price, and tages listed. In addition, the Table above
combinations as a yarn for woven or gloves made with Spectra or Dyneema summarizes some chemical-resistance
knitted products, coated with vari- ultra-high-molecular-weight polyeth- information. Gloves are listed below in
ous compounds including natural ylene typically offer the best cut pro- order, by degree of specialized use.
latex, neoprene and PVC to protect tection based on weight. Steel assures
against petrochemicals, oils, acids, al- the best cut protection based on bulk. Natural rubber (latex)
cohols and solvents. None of these materials, however, Natural-rubber latex gloves (Figure 2)
Supported gloves with non-con- will protect the hands from liquids. are generally unsupported and avail-
tinuous coatings are better suited Liquid-protective cut-resistant gloves able in many styles, including clean-
for applications that require comfort need to be coated with a compound, room and sterile styles. These gloves
and grip. Many supported gloves also such as nitrile. provide excellent protection from bases,
offer cut, snag, puncture and abra- Stretch gloves. Most of todays fabric alcohols, and dilute water solutions of
sion resistance. gloves are made from knitted fabric many chemicals, with fair protection
Chemical Engineering www.che.com July 2008 53
Environmental Manager
D
nance of a respiratory protection
ust in the chemical pro- ing workers from the dangers of program. The program shall cover
cessing industries (CPI) is potentially dusty environments, each employee required by this sec-
present in numerous op- including contact irritations and in- tion to use a respirator.
erations, including powder halation exposure, is an important Additionally, the National Fire
processing; the transport of mate- factor in determining the meth- Protection Association (NFPA;
rials on belt or rotary-screw con- ods and extent to which the par- Quincy, Mass.; www.nfpa.org) has
veyors; grinding materials in giant ticulate matter or dust needs to be stepped in and set consensus stan-
shredders and pulverizers; machin- suppressed or eliminated. dards, based upon good engineering
ing, sawing, grinding or sanding op- practices, to control and prevent
erations; dumping bags of materials Regulations combustible-dust related hazards.
into reactors; handling coating ma- Dust hazards are so problematic NFPA 654 (Standard for the Pre-
terials; as well as the processing of to employees that the U.S. Occupa- vention of Fire and Dust Explosions
pharmaceuticals and foods. An in- tional Safety and Health Adminis- from the Manufacturing, Process-
tegral industry-safety issue arises tration (OSHA; Washington D.C.; ing and Handling of Combustible
when we consider the handling www.osha.gov) has implemented Particulate Solids) applies to all
and suppression of the dust and rules for personal protective equip- phases of the manufacture, process-
particulate matter associated with ment (PPE) specifically dust ing, blending, pneumatic conveying,
these processes, namely prevent- masks and respirators, under stan- repackaging and handling of com-
ing the formation of combustible dard 29CFR 1910.134 to protect bustible particulate solids or hybrid
clouds that can create a fire hazard workers [1]. This standard states: mixtures, regardless of concentra-
or even trigger an explosion. Such In the control of those occupational tion or particle size, where the ma-
a catastrophic dust explosion in a diseases caused by breathing air terials present a fire or explosion
manufacturing facility is depicted contaminated with harmful dusts, hazard [2]. In addition to this stan-
in Figure 1. Additionally, protect- fogs, fumes, mists, gases, smokes, dard, the NFPA has also set stan-
Chemical Engineering www.che.com July 2014 59
Environmental Manager
able individuals to provide testing, throughout an organization with the workforce, facility and the envi-
hazard assessments, risk analysis a mechanism for longevity. Or, put ronment, while maintaining stake-
studies and engineered solutions more simply, organizations must holder confidence and encouraging
to minimize or control danger- have the right people, with the compliance with all legislation and
ous dusts and particulate matter right skills, implementing appropri- avoiding regulatory intervention.
in manufacturing. ately designed process-safety pro- Eliminating or minimizing dust is
To achieve process-safety excel- grams, motivated by the right orga- a small investment compared to the
lence, employees must be genuinely nizational culture, in the right way. costs of the harm that can occur if
proficient and competent in their Results are far-reaching and broad, not appropriately addressed.
requisite technical disciplines, and affecting finances, the environment Edited by Mary Page Bailey
appropriate levels of knowledge and most of all, workforce safety.
must be embedded in key positions Properly managing dust protects Author
Walter S. Kessler is a se-
nior process engineer and
safety specialist at Chilworth
References 5. U.S. Chemical Safety and Hazard Investiga-
tion Board, Investigation Report: Combusti-
Technology, Inc., a DEKRA
Company (113 Campus Drive,
1. OSHA Standard 1910: Personal Protective ble Dust Hazard Study, Report No. 2006-H-1, Princeton, N.J. 08540; Phone:
Equipment, 19982011. November 2006. 609-799-4449; Email: walter.
kessler@dekra.com). He has
2. NFPA 654: Standard for the Prevention of 6. OSHA General Duty Act of 1970, Sec. 5 20 years of experience in pro-
Fire and Dust Explosions from the Manufac- Duties, 1970. cess design, development and
turing, Processing and Handling of Combus- 7. OSHA Status Report on Combustible Dust improvement, process control,
tible Particulate Solids, 2013. NEP, October 2009. logic-controller programming,
3. OSHA, Combustible Dust: an Explosion Haz- 8. National Fire Incident Reporting System. process-hazard analysis, layers-of-protection
ard, OSHA website, https://www.osha.gov/ (NFIRS) Combustible Dust Policy Institute, analysis (LOPA), safety-integrity levels (SIL)
dsg/combustibledust. 2011 Combustible Dust Related Incidents and process safety management in petroleum
Fact Sheet, April 2013. refineries, gas and chemical plants, as well as
4. Amyotte, P., An Introduction to Dust Ex- pharmaceutical, manufacturing and wastewater
plosions: Understanding the Myths and Re- 9. OSHA, Combustible Dust in Industry: Pre- facilities. Kessler holds a B.S.Ch.E. from Penn-
alities of Dust Explosions for a Safer Work- venting and Mitigating the Effects of Fire sylvania State University and is currently ob-
place, Elsevier, 2013. and Explosions, SHIB 07-31-2005. taining a Ph.D. at the University of Houston.
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
Circle 3 on p. 68 or go to adlinks.che.com/50978-03
Chemical Engineering www.che.com July 2014 63
Feature Report
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-
0
BATCH 1 BATCH 2 BATCH 3
Circle 13 on p. 62 or go to adlinks.chemengonline.com/56199-13
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
sources of these incidents. This leads to the commissioned, because no part of the plant
larger question, Why do bad things happen had been shut down.
to good steam equipment? When the in-service quantity of CDLs is
The technical answer lies in understanding lower than the design calls for, then it may
that the cause for a large percentage of failures be that the reduction was due to a misun-
might be due to the steam system not being derstanding of the importance in maintain-
maintained to the as-built or design speci- ing the design total, or possibly from general
fication. The original designers analyzed the neglect of the trap population. If the discrep-
plant requirements and determined the most ancy is not explainable, then the systems
suitable design, according to their expertise drainage is restricted from the originally re-
and standards. That original design included quired capability. Here, the target for total in-
the correct number of steam traps and the as- service CDLs should be increased to equal
semblage of piping and components that help the original design total of 11,000 CDLs.
to drain condensate from the system. Com- When routinely administered, steam-
monly referred to as a condensate discharge trap surveys conducted by plant personnel Figure 3. Accurate and
regularly sustained diagnosis
location (CDL), the total assembly is required (Figure 3) provide invaluable information for of steam traps operating
to remove condensate and effectively maintain evaluating the health of the steam system. conditions is essential to
the design performance of the steam system. As it turns out, in this hypothetical example, determining the populations
Once built, the plant is handed over to the end several years had passed since the last trap current health (state of the
population). Once failures
user for operation and maintenance with survey was completed. So, when the current are identified, the informa-
the expectation of sustaining the initial design. survey was finished, it was found that there tion is valuable for allocating
Unfortunately, that is when operational bud- were 3,000 trap failures (both hot leaking resources to restore all con-
densate discharge locations
gets, personnel turnover and, in some cases, failures and cold-blocked low-temperature to an as designed operating
inexperience with steam systems may play a failures occurred); and 7,000 traps were condition for safety and reli-
causal role in negative events. considered to be in good condition. This sit- able performance
Example: 10,000 traps / 1,000 traps failed in a year = Example: 3,000 failures / 10,000 traps = 30% state
10 year trap life expectancy of failure
7,000 good traps / 10,000 traps = 70% good state
uation creates a significant operational and to grow until catastrophic events occur.
maintenance dilemma for the site. One takeaway from this example is that for
If the goal is to have a zero reset sce- every 1,000 traps that must be repaired in
nario, in which all failed traps must be re- a year, there is a requirement for the plan-
paired, and the cost to repair the average ning and repair resources to correct 84 traps
failure is $600, then $1.8 million is needed monthly without fail. That replacement re-
from the years budget to accomplish the quirement equates to repairing 45 CDLs
target. To achieve this goal, 250 steam traps per day, with any lesser amount creating a
must be replaced each month, which repre- gap at year-end; the difference here equaling
sents two or three maintenance crews work- the next years carryover failures.
ing full time for a year, assuming that every Steam traps must be replenished or re-
trap is accessible and can be isolated for paired in order to maintain a sustainable
repair or replacement. It is certainly a monu- operation. Not taking the highest-priority
mental task that may have been caused by corrective action with regard to steam trap
an improper course of action a number of failures is somewhat comparable to reusing
years before, when trap replacements were a teabag even though it no longer dispenses
not completed in sufficient quantity to keep flavor although there is a teabag in ser-
pace with the annual failure rate (amount of vice, it no longer provides a useful purpose.
failures per year) of the population. Similarly, if a steam trap has failed par-
Consider that the average annual failure ticularly via a cold failure then the CDL is
rate of a steam trap population in a mature no longer serving its intended purpose, and
plant can be estimated from historical re- must be repaired.
cords, provided that there are at least two Is there any question about what outcome
survey events within a period of 45 years. should be expected if all of the steam traps
Simply subtract the carryover failures (failed and related CDLs in a system were simply iso-
traps recorded from a prior survey for which lated by valving, thereby completely remov-
no action was taken to repair) from the re- ing their drainage capability? There would be
ported failures that were recorded in the no way to automatically remove condensate
survey. The remainder is the quantity of new from the system, creating a highly dangerous
failures. The number of new failures can be situation. What about if just 30% were shut
divided by the number of years since the prior off, or 50%? It is akin to gambling with the
survey to provide an estimate of the average safety of the plant, as the potential forma-
annual new failures of a steam trap popula- tion of condensate slugs in the pipeline can
tion. Several related useful calculations are lead to unstable, hazardous conditions. It is
shown in the box above. extremely distressing to consider a site that
In this hypothetical plant history, if the is not replacing trap failures due to a bud-
management of the trap population had get constraint, because the timely repair of
been continuous and sustainable at an an- traps allows the system to operate at the
nual failure rate of 1,000 (10%), then the intended conditions. This must be seen as
plant should only have to support the repair an inflexible demand. Timely action should
of 84 traps per month, not 250 traps. How- be mandatory, not optional, to optimize the
ever, in this case, the trap population was al- operation of a steam system.
lowed to deteriorate to 3,000 failures (30%
failure state), and this situation places a tre- Management prioritization
mendous burden on resources. Worse still, if In a properly drained and maintained steam
not corrected, the failures can be expected system, it is critical that the steam flowing
10,000
400
5,000
200
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12
Month
1,000
Carryover
500 failures
Target monthly replacements
Actual replacements
0
0 1 2 3 4 5 6 7 8 9 10 11 12
Month
the hot failures, because these repairs can a plant with an average steam-trap lifecycle
be readily justified by simple energy-cost of ten years, the actual failures could be es-
analysis. Figure 4 illustrates a typical close timated as 2,000, consisting of the 1,000
relationship between hot-failure traps and re- carryover failures just identified and 1,000
duced steam loss. Especially in times of very new failures. While the diagnosis of each
high energy prices, incredible emphasis is trap in the population is performed at regular
placed on reducing the cost of operations by intervals (usually annually or semi-annually),
fixing leaking steam traps. A progress chart, new failures are constantly occurring, as il-
similar to the one shown in Figure 4, can be lustrated in Figure 6.
generated and tracked. Thus, if the project If carryover failures are included as part of
plan is to eliminate 3,000 failures from the the repair strategy, a significant number of
hypothetical plant, then the corresponding CDLs will be operating improperly, thereby
requirement is to repair 250 CDLs per month, increasing the chances of a debilitating in-
otherwise there will be a carryover failures cident occurring within the plant. For this
gap (see Figure 5 for additional analysis). reason, it is not recommended to adopt a
The carryover failures represent the real work process that allows carryover failures
world; rarely do plants correct all or even and focuses only on those traps that have
nearly all of the failures. The result is that already been fixed. Instead, a paradigm shift
a significant number of CDLs are not re- is required.
stored to proper drainage operation, and it
is not uncommon to carry a sizable number System goals
of failures over to the following year. Once Applying maintenance action to individual
carryover failures are accepted in a plants steam traps is a path action, but not an
operations, the steam system becomes overarching system goal. Risk lies in the
destined for sub-optimal and potentially false sense of security that is given when
unreliable operation. the goal is simply to repair a given number
What follows when 1,000 carryover failures of steam traps, and the real goal of achiev-
are extended into the following year? Instead ing an optimized steam system is neglected.
of correcting all 3,000 failed steam traps (a So, if the annual target is established to fix
zero reset mentality), suppose that the plant 2,000 steam traps and only 1,000 traps are
management allocated a budget for repair repaired, there might be some explanation of
of only 2,000 failures instead. Perhaps this mitigating events that explains the lapse. In
thought process stems from the expecta- such a situation, the carryover failures might
tion that with 2,000 failures corrected, the be considered acceptable under the circum-
next annual period will only require budget stances, and a new goal is assigned for the
for 1,000 failures. However, that is not an next year. However, this is a dangerous situ-
accurate scenario, because those 1,000 ation, because while the potential for dam-
CDLs represent carryover failures only, and age is not visible, it is prevalent in the unad-
the plant also must consider new failures. In dressed sections of the steam system, and
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
4,000
Accumulated repairs
2,000
New accumulated failures
0
0 6 12 18 24 30 36
Month
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
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 See how the details on the inside make all the difference on the outside.
for testing activities, and must be Call a piping systems consultant at 1.216.447.7397
suitably educated and certified for or visit corzancpvc.com to learn more.
these tasks. Most critically, there
should exist a regular audit process
to check and confirm the accuracy
of the surveyors qualification cer-
2014 The Lubrizol Corporation, all rights reserved. All marks are the property of The Lubrizol Corporation.
tificates, the actual judgments per- The Lubrizol Corporation is a Berkshire Hathaway company.
formed by those surveyors, and in GC 140681
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).
Circle 42 on p. 82 or go to adlinks.che.com/56195-42
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
Keep an eye on key variables tance thermal detectors (RTDs; also give operators early indication of de-
Vibration monitoring is an important known as resistance thermometers), terioration in the machines operating
indicator that can help operators to thermistors, filled thermal systems, condition.
identify machinery problems at an pyrometers, infrared thermography Control valves are vital compo-
early stage. The location of vibration techniques and glass thermometers. nents in every process and in many
sensors is vital, to avoid introducing All are widely used to provide types of equipment packages where
errors during the measurement. In process and machinery temperature fluid is handled and regulated. How-
general, the best location for vibra- measurements. ever, the most common malfunction
tion sensors in rotating machinery The measurement of flowrate pro- with control valves is internal leak-
is at the bearing or bearing housing vides another useful indicator for the age, which cannot be detected eas-
(Figure 2). Vibration measurements health of many processes and ma- ily. Internal leakage can result from
at the machinery casing, in addition chinery systems. Traditionally, tech- one of several factors, including an
to bearing vibration, can provide niques based on differential pres- eroded valve plugs or seats, or insuf-
useful data for identifying and solv- sure, turbine and vortex shedding ficient seat load.
ing machinery problems. have been used to measure fluid
However, monitoring vibration flowrate. However, these methods Five case studies
alone is not sufficient to gauge the might not be suitable if an accurate Case study 1: Insects in an air-
health of rotating machinery espe- measurement is required for gases, compressor inlet filter. In a CPI
cially when large, critical machines since errors can be introduced in the plant in a remote area, insects
are concerned. For instance, in a re- measurement due the compressible caused problems for the inlet-air fil-
cent case, it was observed that one nature of gases. To use these meth- tration unit of a large process-type
electric motors temperature and ods with gas flows, it would be nec- air compressor. The accumulation of
current increased when the machine essary to compensate for the gas insects on the filters led to high dif-
was overloaded well before any density fluctuation in order to reduce ferential pressure in the unit, leading
vibration was detected. This under- the errors. to periodic compressor shutdowns.
scores the importance of monitoring Modern Coriolis mass flowme- Multiple cases of surge and surge-
other parameters (such as tempera- ters are a more appropriate choice related trips occurred because of
ture and current), in addition to vibra- for measuring gas flowrate, which is this issue. Regarding a problem of
tion, to allow for early fault diagnosis usually independent of temperature, persistent insects impacting the air-
of electric machines. pressure, density and composition. compressor, two solutions should be
When vibration analysis is used This meter uses the Coriolis effect to considered:
alone, it may be difficult to estab- measure the amount of mass mov- 1. A multi-sided insect screen. For
lish the real cause of the vibration. ing through the element. In a simple example, three-, four- or five-
In general, it is better to monitor form, the fluid to be measured runs sided insect screens can be con-
both vibration and overall machin- through a U-shaped tube that is structed (for instance, in a cube or
ery performance (such as delivered caused to vibrate in a perpendicu- similar configuration) in the front
pressure and flow in pumps or com- lar direction to the flow. Fluid forces of the air filter, with a surface area
pressors) to obtain a more reliable running through the tube interact roughly three to four times that of
fault diagnosis. And, it is important with the vibration, causing it to twist. the front face of the air filter inlet.
to realize that the vibration produced The greater the angle of the twist, The rationale is that insects will
by the machinery is not a fault on its the higher the flow. These flowme- gather on the larger-surface area
own but rather a symptom (and ters have no rotating parts. Modern of the screen (far from actual in-
early warning sign) of a developing Coriolis flowmeters are highly accu- take) and thus experience less
failure condition. rate (providing accuracy to around intake air velocity, allowing them
Temperature is another critical 0.5% of the mass flowrate). to fly away, reducing their oppor-
variable, and the ability to moni- For relatively low-speed rotating tunity to trap and plug the air filter.
tor it accurately is of paramount machines, vibration analysis may not When a screen is built with three
importance in CPI plants. Today, a be suitable to detect degradation times the area of the air intake,
wide range of temperature sensors or operational problems. However, the air velocity across the insect
and measurement systems exists, acoustic-emission sensors, strate- screen would be around 33% of
including thermocouples, resis- gically located on the machine, can the original velocity. When the
Febru
a
2015 ry
02
www
He .che
Fundam
Two-P at Exchange
men
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
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
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 guage in place that describes either the from existing test-method standards.
are available in a wide vari- overall performance of the protective Many companies are multinational
ety of fabrics and designs that ensemble or the performance of the and do not have access to the same
provide varying levels of pro- individual components against types chemical protective apparel in each
tection for chemical process industries of different chemical hazards. While location even though they may be
(CPI) workers, who may be exposed to some standards do exist for hazardous engaged in the same work tasks and
chemicals in the form of vapors, liq- materials (Hazmat) emergencies, little therefore have the same needs for
uids and particles. While other forms regulation exists regarding routine in- protective clothing. The ISO 16602
of personal protective equipment dustrial workplace applications. This description of chemical-protective-
such as safety glasses and hard hats void in recognized standards for every- clothing performance is vital to these
are highly regulated in the U.S. and day, industrial-use chemical protective businesses as it provides a common
Canada, the performance of chemical clothing can lead to confusion and un- description that can be used to se-
protective clothing is not. An interna- certainty for the safety professionals lect the appropriate, locally available
tional standard for classifying chem- who must select the appropriate ap- products. In addition, ISO 16602s ob-
ical-protective-clothing performance, parel for a given hazard or application. jective performance levels allow safety
however, is available in ISO 16602. The Established by the International Or- professionals a means of identifying
ISO 16602 Standard (Protective cloth- ganization for Standardization (ISO), and selecting higher performing gar-
ing for protection against chemicals the requirements outlined in ISO ments if a specific selection of chemi-
Classification, labelling and per- 16602 provide a common language for cal protective clothing fails to perform
formance requirements) provides an the performance of chemical protec- as intended in the work environment
objective system to appropriately test, tive clothing. While ISO 16602 was in- for which it was chosen.
classify and label chemical-protective- troduced fairly recently (in 2007), it is
apparel products. This article shows already recognized and accepted inter- Clothing selection
how ISO 16602 can be used as a tool to nationally, and is poised to become the The first, and most critical, step to
help with the selection of appropriate global language for expressing chemi- selecting appropriate chemical pro-
chemical protective clothing by provid- cal-protective-clothing performance. tective clothing is the completion of
ing an objective means of defining the By offering a common, global clas- an accurate hazard assessment. By
performance of chemical protective ap- sification system and a standard pre- identifying the task, the surround-
parel in situations involving potential sentation of the performance data, ings, the chemical(s) and any hazards
exposure to chemical hazards. ISO 16602 can simplify the selection of beyond chemical exposure, a hazard
Since, in North America, chemical chemical protective apparel by safety assessment provides a basis for un-
protective clothing is not assessed by and occupational health professionals. derstanding what chemical protective
a comprehensive standard that quan- Standardizing the protective cloth- clothing and other personal protective
tifies the performance of garments, ing performance data allows objective equipment (PPE) are necessary to pro-
safety and health professionals select- evaluation and comparison of the vari- tect the workers operating in specific
ing chemical protective clothing can ous garments available in the market- situations. The environment, chemi-
only rely upon manufacturers claims place, rather than relying solely upon cal hazard and work activity must be
to choose the appropriate garments. vendors claims. Although ISO 16602 considered in order to be able to se-
Unfortunately, the performance of permits a range of performance levels lect the most appropriate protective
chemical protective clothing can vary for a series of key properties, it also clothing including the fabric, seam
by manufacturer and there is no es- establishes a minimum level of perfor- type and garment design most suit-
tablished means to compare the perfor- mance for each major type of hazard. able. Without adequate protection
mance of the various garments on the ISO 16602 defines performance re- from chemical protective apparel and
market. There is also no standard lan- quirements based on results obtained proper work practices, the likelihood
Chemical Engineering www.che.com March 2011 51
Environmental Manager
of a worker being exposed to poten- testing is used to validate the barrier specific type requirements. A higher
tially harmful chemicals increases. performance of the entire ensemble class rating denotes a higher level of
Conversely, overprotecting a worker against a specific type of chemical performance for that property.
with unnecessary clothing can be bur- threat (gas, liquid or particle). Whole- The flammability requirements
densome and result in heat stress, re- garment testing is conducted for each outlined in ISO 16602 establish a
duced field of vision, restricted mobil- ISO 16602 type, and the exposure con- minimum performance level of flame
ity and increased physical exertion for ditions for the tests vary according to spread once the material is ignited; it
the wearer, as well as possibly higher the hazard defined by that type. Note does not qualify an ensemble as suit-
purchasing costs for the employer. again that the test chemicals are non- able for protection against heat and
Every work situation is unique, hazardous and are used to essentially flame hazards. Specific evaluation of
and the ISO 16602 Standard does not determine how much of a similar phase chemical protective clothing for heat
consider all specific hazards that may chemical will leak into the suit. The and flame protection is not in the pur-
be present in the work environment. whole-garment tests do not evaluate view of ISO 16602. In North America,
Safety and occupational health profes- the chemical permeation properties of relevant standards specifically related
sionals need to consider hazards that the garments. This is assessed in the to protective clothing for use near fire
are specific to their work conditions class testing portion of ISO 16602. and electric-arc hazards are NFPA
and then look to ISO 16602 to find the For example, the Type 3 whole-gar- 2112 (Standard on Flame-Resistant
minimum requirements for chemical ment test is conducted by having the Garments for Protection of Industrial
protective garments for that type of test subject march in place with ex- Personnel Against Flash Fire) and
situation. ISO 16602 focuses exclu- aggerated arm-pumping movements, NFPA 70E (Standard for Electrical
sively on typical chemical hazards while slowly turning, in front of a high- Safety in the Workplace).
and protective clothing requirements pressure, high-volume liquid spray The class tests are as important as
for such hazards so the work situation nozzle. This liquid is tinted with a dark the whole-garment tests described ear-
may also require additional forms of color to be visible if it penetrates the lier. Both are integral to evaluating the
PPE, such as footwear, gloves, face pro- test garment. Once the high-pressure overall integrity and expected perfor-
tection, fall protection and respirators. liquid exposure portion of the test is mance of chemical protective clothing.
When evaluating protective clothing completed, the outer test garment is This integrated rating system aids the
performance under ISO 16602, gloves, carefully removed and an inner in- selection of the most appropriate chem-
footwear and respirators must be in- dicator garment is checked for evi- ical protective apparel for specific work
cluded in the whole-garment ensem- dence of liquid penetration, which is tasks. Testing the whole garment and
ble testing. It is important to validate evident by staining from the dark liq- the individual components separately
that any additional PPE components uid. Beyond the whole-garment tests, ensures the sum of the parts is able to
will work properly with the chemical additional tests are conducted on the provide the appropriate protection.
protective clothing selected for use. garments fabric(s) and components to
qualify the class performance level for Applying ISO 16602
Types of clothing the garment against specific chemical While understanding ISO 16602 is
The ISO 16602 standard is based upon challenges and the physical demands critical for safety and occupational
a series of types and classes. ISO 16602 of the task and work surroundings. health professionals, it must also be
designates minimum performance lev- ISO 16602 defines different class lev- applied consistently and correctly in
els for six types of chemical hazards. els of performance for each type of pro- the workplace. Consider the task of
The type is the overarching system, tection as described below. opening a flange in a process pipe con-
with chemical protective clothing fit- taining a hazardous liquid under pres-
ting into one of the six different types. Classes of clothing sure. Even though the pipe should be
The garment type designation is based Within each of the six garment types, depressurized and drained as much as
upon the physical state of the hazard, there are also requirements directed possible before the flange is removed,
for example vapors, liquids, aerosols at the mechanical, barrier and basic this activity would likely require an
or particles (Figure 1). flammability properties of the fabrics ISO 16602 Type 3 garment for protec-
When setting the requirements for and components used to make chemical tion against the possibility that pres-
each type in ISO 16602, the entire gar- protective clothing. Laboratory tests surized liquid exposure might occur.
ment is tested as well as the individual are used to determine the mechanical The Type 3 whole-garment jet test
components. During the whole-garment durability, the barrier against specific would have demonstrated the liquid-
tests, a human subject wears the test chemical hazards, and ease of ignition penetration resistance of the ensem-
garment and accompanying PPE, such of the garment materials. The results ble and the performance of the connec-
as gloves, boots and respirator. The of these tests will fall into a unique tions between the garment and other
human subject is exposed to non-haz- performance class. Each type within items of PPE, such as gloves, boots
ardous test chemicals while in an en- ISO 16602 specifies a combination and respirator. The chemical barrier
closed chamber performing a series of of barrier and durability tests levels, performance of the suit material and
movements meant to simulate actual establishing a minimum performance components should have been deter-
work activities. This whole-garment class for each of the tests to meet the mined with permeation tests using
52 Chemical Engineering www.che.com March 2011
Six types of CPC outlined in ISO 16602
Type 1
Type 2
Type 3 Gas-tight
Type 5
Type 4
Protection
Non
gas-tight
protection
against
chemicals
The Chemical
Engineering
Type 6 against protection
Protection and vapors
against pressurized and toxic
Protection liquid
liquid particles
bookstore offers
Limited against chemicals
airborne, aerosols
protection
against solid
a variety of
particulate Type indication doesnt tell you everything
liquid mist
chemicals about the performance of the CPC. It only
tells that it meets the minimum require-
industry topics
ments of a particular type certification
Fresh ethylene
Fresh butene
1
2
3 3'
For
disposal
A
mong pilot plants, those in petro- those for full-scale plants
chemical or petroleum-refining
service pose special risks of fire
and/or explosion. The key strate- less tested and/or less robust than pilot plant design and operations,
gies to minimize those risks are these: conventional process equipment. are fully aware of how a pilot plant
properly conducted preconstruction In order to ensure a safe operating en- differs from a full-scale plant, and
safety reviews; leaktight design and vironment, a comprehensive approach are able to address the research-spe-
construction; close attention to area is required. No single activity can en- cific needs knowledgeably
electrical classification; well-designed sure safety; there is no magic bullet. The predetermined basis reflects the
ventilation; well-planned storage; Instead, many complementary and fact that the review requires struc-
properly conducted maintenance; and overlapping approaches are needed. ture and organization, both of which
strictly followed procedures for the And in addition to the seven strategies need to be determined and set before
control of change. listed at the beginning of this article the start.
Intelligent application of those seven and elaborated in the rest of it, behav- The review should scrutinize all facets
strategies can be aided by a closer look ioral based safety (BBS) efforts and of the proposed operation, including
at just why petrochemical and refin- overall company safety programs and startup, shutdown, material charg-
ing pilot facilities are so risky. procedures are also critical; but they lie ing, sampling, waste disposal, routine
They routinely handle significant outside the scope of this discussion. operation and all conceivable process
volumes of flammable materials, upsets. It should examine each com-
many at elevated temperatures Preconstruction safety review ponent of the proposed design, among
and pressures. While most pilot Sometimes called a pre-design or pre- them the layout, detailed piping, wir-
plants deal with only small volumes approval safety review, a preconstruc- ing, and safety-shutdown logic and
of these materials at a given time, tion safety review aims to ensure that implementation.
the overall quantity of flammable any proposed design or modification Every component, from valves to fit-
and combustible materials used and is safe before a final commitment to tings, controllers to indicators, pumps
stored in most pilot plant areas is carry it out is made. This review may to analyzers and mixers to flanges,
far from trivial often be preceded by other safety re- should be reviewed and examined in
In comparison with full-scale pro- views (such as conceptual ones) for detail for potential hazards. For each
cess plants, pilot plants are usually relatively hazardous operations, but potential hazard uncovered, the team
characterized by a high density of the preconstruction safety review re- then needs to assess the risk associ-
operating equipment, so the risk to mains the most comprehensive, de- ated, and decide whether the proposed
personnel is significant in the event tailed and critical review. design and/or operation mitigates the
of even a small fire or explosion Properly conducted, it is a long and risk to an acceptable level or, instead,
Pilot plants often involve novel op- detailed exercise, which typically takes requires additional appropriate cor-
erations, whose nature and hazards 4 to 12 hours to complete. To be effec- rective measures.
are imperfectly known tive, a preconstruction safety review The preconstruction-safety-review
Pilot plants frequently switch from should be conducted by an experienced committee should also make sure to
one process operation to another, and group of trained individuals familiar address compliance with all standards
sometimes the scale from one opera- with pilot plant design, equipment and legal, external and companywide
tion to the next differs by orders of operations, on a predetermined basis. and recommend or approve variances
magnitude, to a greater extent than Each of those three attributes is impor- or deviations where they seem war-
is true for a full-scale plant tant: ranted or necessary. The committee
The processes studied in pilot plants Trained implies that the reviewers then approves the project advancing
are typically still evolving and being must understand and be able to ef- to the construction phase.
modified. Similarly, the equipment, fectively apply risk assessment and Keep in mind that the purpose of
while admittedly small in size, is hazard analysis methodology the preconstruction safety review is
often home-made, one of a kind, and Familiar means they understand to identify areas where the risk is un-
52 CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006
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
acceptable and needs to be mitigated. like, often with unfortunate results. inadequate personnel is as likely
While the committee conducting the Instead, a well scheduled review is to generate a major overkill in the
review may chose to recommend cor- held as late as possible in the design safety requirements as it is to miss
rective actions when these are obvious effort, generally just before construc- a key hazard, because the review-
and straightforward, it should not at- tion and materials procurement ers recognize their lack of skill and
tempt to design the mitigative mea- Holding the review after construction seek to compensate for it by play-
sures in the review. Rarely are such has started (or in extreme cases ing it safe. This state of affairs can
spur-of-the-moment corrective designs after it has ended!). This delay result in requirements that greatly
as effective and economical as those always leads to a tendency to accept increase the cost and complexity of
created by the project team later. (Re- work already completed, or to tolerate the design and decrease its operabil-
call the old advice about a camel being equipment already ordered. The engi- ity, in many cases with little to no
a horse designed by a committee.) neer should always keep in mind that commensurate increase in safety
A portion of a typical review sum- a given design is either acceptably Inadequate or incomplete analysis,
mary is shown in Table 1. And a typi- safe or unacceptably risky based on usually due either to a lack of time
cal matrix for the associated hazard an analysis of the hazard (if any), the or to a desire to rush the always-
analysis appears in Table 2. potential consequence and the per- tedious review process. Failure to
Poor practices with preconstruction ceived probability, and not on whether allow enough time leads to a rushed
safety reviews include the following: that design is already being installed process. Conversely, scheduling one
Holding the review before the de- and would be costly to modify marathon session almost always en-
sign of the pilot plant is complete. Lacking adequate reviewer exper- sures less rigor as the day drags to a
This timing forces the review team tise in a particular area of concern, close. The hours needed to properly
to guess at the details of the final such as vessel design, safety shut- conduct a preconstruction safety
design, and is always guaranteed to down, or process chemistry. With review of a large, complex or par-
produce potential problems. At best, this lack, accurate evaluation of the ticularly hazardous pilot unit may
the review ends up wasting time by risks can be difficult, if not impos- best be addressed over several days
working some of the design, or hav- sible. The reviewers either have to or, even more effectively, in several
ing to reconvene later to review the make their best guess as to the haz- smaller and more bearable sessions.
missing parts; at worst, the review as- ards, or rely solely on the laboratory Anyone who has ever been trapped
sumes what the final design will look personnels assessment. Review with in the eleventh hour of a 12-to-14-
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006 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 Review Level Comment
Change in kind None if failure was non-hazardous Only if component is identical in all respects.
and routine (such as a clogged filter) A switch to a different manufacturer or a dif-
ferent model should require review
First-line supervisor if failure was non- Cause of failure should be examined; deter-
hazardous but unexpected mine if it could have led to a hazard
Safety department if failure was haz-
ardous
Change of specific model but pur- First-line supervisor or engineering Review all specifications in detail and ad-
pose and operating conditions re- department dress any differences
main unchanged
Change of purpose but operating Engineering department or safety de- Confirm new purpose does not produce new
conditions remain unchanged partment concerns or modify original assumptions
Change of operating conditions Safety department
within pilot plant design parameters
Change of operating conditions out- Safety department and engineering
side pilot plant design parameters department
Process change Safety department, and usually an-
other preconstruction review
Safety system change Safety department, and usually an- Checkout of entire safety system (includ-
other preconstruction review ing all unmodified parts) is mandatory after
change
Piping or wiring change Safety department, and usually an-
other preconstruction review
Whenever a pilot plant is built or re- are often able to spot bad practices or API-540 Electrical Installations In
vamped, ensure that the construction inadequate methods well before these Petroleum Refineries
is carried out soundly. All too many become major problems. Sadly, none of these are specifically
pilot plants are built either by re- Of course, problems identified dur- intended to address pilot plant con-
search personnel who are not as famil- ing construction, start up or operations cerns, and all of them fail to address
iar with good construction techniques, should not be solved and then forgot- issues that are major for pilot-plant-
or by contractors with little experi- ten. Rather, they should lead to re- scale work. Thus, additional hazard
ence in pilot-scale equipment. Either vised internal company specifications/ analysis, consultation with experi-
alternative is likely to lead to unsound standards, modified future designs, enced pilot plant safety personnel and
construction decisions; for example, a improved training and/or appropriate alternate methods will almost always
failure to weld a long line that, during communications. The amount of hard- be needed with respect to electricity-
service, will be frequently heated and won and costly experience that is lost related safety.
cooled, and is not only covered with in- each day through not communicating All petrochemical and petroleum-
sulation but also situated in a location the lessons learned is substantial in refining sites should generate and
that is difficult for leak-testing. Even most pilot plant organizations. maintain detailed site standards that
explicit safety issues can arise; for in- show the area electrical classification
stance, when a section of piping that Electrical classification of all pilot plant areas, site-wide. In
must be removed frequently for clean- Proper area electrical classification, general, at least Class I Division 2
ing or charging is fabricated from com- made initially and then maintained construction is prudent for all pilot
mon pipe fittings and in such a way diligently, is a key factor in prevent- plant areas where units are handling
that it is difficult to remove and even ing fires and explosions, even though it any appreciable amounts of flam-
more difficult to routinely reseal. somewhat raises the cost and difficulty mable materials. For some research
Proper construction quality also of pilot-plant construction. Many codes situations, alternative or special de-
implies continual and ongoing quality and standards nominally address this signs may be required in order to meet
control, including uniform construc- area. Among them are these: the intent of the area electrical clas-
tion procedures, appropriate site-wide NFPA-497A Classification of Class sification. Examples include purging
training, proper component selection, I Hazardous Locations for Electrical in accordance with NFPA-496, small
proper installation, and ongoing con- Installations in Chemical Plants positive-air-flow enclosures similar to
struction supervision. There should NFPA-30 Flammable and Combus- control rooms per NFPA-496, ventila-
be spot tests of completed assemblies tible Liquids Code tion barriers, equipment separation,
or components during construction, NFPA-70 The National Electric barricades, and special conditions and
the gauging of compression fittings, Code equipment to prevent ignition. All
and the X-raying of welds and/or the NFPA-496 Purged and Pressurized such measures require additional haz-
hydrostatic testing of welded compo- Enclosures for Electrical Equip- ard analysis and assessment of their
nents. Equally important is, simply, ment own, and they usually result in addi-
insisting that knowledgeable and ex- API-500A Classification of Locations tional design and construction effort.
perienced pilot plant personnel visit for Electrical Installations In Petro- As emphasized in next section,
the construction site frequently. They leum Refineries proper ventilation is essential for
CHEMICAL ENGINEERING WWW.CHE.COM MAY 2006 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 safety. However, reliance equipment that takes up a lot of space. should be used to capture any emis-
on a high ventilation rate alone as an For pilot plant areas, both high- and sions or leakage at the source before it
alternative to proper area electrical low-point ventilation are required, to becomes a hazard.
classification is risky it can lead capture and exhaust light and heavy Equipment that handles very toxic
to unanticipated fires and explosions, gases alike. materials may need to be placed
because the concentration of flam- Smoke tests are often needed, to wholly or partially in hoods or simi-
mable material always has to descend confirm that the design is ventilating lar ventilated enclosures that meet
through the explosive range before the properly. This is especially true after ASHRAE 110 and AIHA ventilation
ventilation-induced dilution brings the bulky new equipment is installed, or requirements.* This requirement in-
concentration below the lower explo- in difficult-to-assess situations. volves significant design work, both
sive limit. Even when the boundaries Most pilot-plant and laboratory areas for the hood itself and to ensure that
of this explosive range are relatively should also be continuously monitored the pilot plant layout is not keeping
narrow, the sparks or other sources of via fixed sensors for combustible and/ the hood from functioning properly.
ignition from nearby electrical devices or toxic gases. This monitoring should Figure 2 shows potential problems
(in the absence of proper area electri- alert personnel to the onset of a poten- related to hoods. Operator training
cal classification) can cause a disaster. tial hazard and shut off the hazardous should address the safe use of large
An electrically heat-traced line, for feed at the source until the problem is hoods. And an ongoing maintenance
example, virtually always includes located and corrected. program is required to ensure that
heater elements that are too close to Typically, monitoring takes place in- hoods continue to function properly.
allow dilution before contact. side the pilot plant area but outside of
In summary, with careful design, any hoods or ventilated enclosures; in Storage, maintenance
proper area electrical classification other words, it takes place where the To safely store hazardous materials
minimizes the risk of explosions and operators are likely to be. Monitoring in pilot plants, full compliance with
reduces the risk of fire (usually to a inside the ventilation duct should be NFPA-30 as well as local building/fire
minor hazard at worst) while accom- avoided. Sensors placed there are dif- codes is required; and in some cases
modating almost all research needs ficult to maintain and are subject to involving highly flammable or explo-
with minimal adverse operational and problems with dirt and dust. More sig- sive materials, it is prudent to go even
economic impact. nificantly, they can be prone either to beyond those code requirements.
numerous false trips or to a failure to Gas cylinders should not be placed
Ventilation detect anything short of a catastrophic inside pilot plant areas; instead, they
All pilot plant and laboratory areas release, due to dilution (see Figure 1 for should be stored outside or in sepa-
should be provided with exhaust ven- an example related to this point). rate cylinder storage areas. Solvents
tilation. The optimal rate depends on In addition to general ventilation, should be stored in flammable storage
the operations involved and the tox- local exhaust should be provided cabinets whenever possible; and not
icity of the chemicals handled but, where needed, such as locations that just if required by code. Try to keep
in general, rates of at least 3 cfm per involve sampling, vessel filling, ves- only small amounts of hazardous ma-
square foot of floor area are prudent. sel draining, product collection, areas terials on hand; the quantities that
The ventilation rate alone is not suf- of relatively high leak potential (such operators or engineers declare to be
ficient; an appropriate ventilation pat- as seals), equipment that needs to be necessary for operations should be pe-
tern (directional sweep) is also neces- removed or opened frequently (such riodically challenged and reassessed.
sary. For instance, take care to ensure as reactors, filters, or feed cases) and Routine maintenance of all safety
that the flow is not disrupted by large equipment requiring frequent mainte-
*ASHRAE is the American Soc. of Heating,
openings, such as overhead doors or re- nance. Ventilation employing fixed en- Refrigerating and Air-Conditioning Engineers;
movable panels, or by newly installed closures and flexible arms/hoses alike AIHA is the American Industrial Hygiene Assn.
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 detection
Notification
r 4NPLF BQQMJBODFT
LOGIC
r )FBU CFBDPOT
SOLVER
r 'MBNF TPVOEFST
r .BOVBMDBMM
Fire and gas
point
DPOUSPMMFS
Fire
TVQQSFTTJPO
or
Gas detection TZTUFN
r 5PYJD 'JSFBMBSN
Integrating fire-and-gas r 'MBNNBCMF
r "DPVTUJD
QBOFM
Safety
TZTUFN
detectors and mitigation MFBLBHF
r .BOVBM
TIVUEPXO
TZTUFN
systems into overall process DBMMQPJOU
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 of fuel. Water and land quality were of this issue is the state of the global
abnormal situations, such as a adversely affected across southern economy; manufacturers simply can-
fire, and combustible or toxic England, resulting in a clean-up ef- not afford the monetary losses com-
gas releases, fire and gas (F&G) fort that cost hundreds of millions of monly associated with plant safety
detection and mitigation systems are pounds. Such incidents illustrate the incidents.
the linchpins to maintaining overall potential threats to the worlds oil- Amid this backdrop, integrated
safety and operation of industrial fa- and-gas supply chain and other vital process control and safety systems
cilities. The premise of these systems process industry resources. They also have emerged over the last few years
is to protect processes and the envi- serve as a critical reminder that, due as effective solutions for providing
ronment by providing early warning to the very nature of the business and safer working conditions in plants. In
of incidents and enabling actions that the products associated with it, the providing operators better visibility
prevent them from escalating. Also, CPI is inherently fraught with risks. across process and safety, CPI compa-
implementing an integrated F&G These risks range from raw mate- nies are realizing that integrated sys-
strategy based on the latest automa- rial and intermediate toxicity and tems can reduce costs by minimizing
tion technology, the chemical process reactivity, to energy release from equipment damage and incidents that
industries (CPI) not only meet pro- chemical reactions, high tempera- impact people and the environment.
tection requirements, but also have tures and high pressures. With all At the same time, the reflection on
potential to enhance business success. too frequent news reports of cata- corporate image is positive.
strophic incidents occurring at man-
F&G safety awareness ufacturing facilities across the globe, Elements and integration
In December 2005, an explosion at the safety has never been a more scru- International standards organize
Buncefield fuel terminal in Hertford- tinized issue than it is today. Com- safety implementation under a series
shire, U.K. ignited millions of gallons pounding the already high visibility of six protection layers:
40 Chemical Engineering www.che.com May 2009
Plant evaluation system/
Plant security center,
(normally not in F&G scope) F&G operator DCS/SIS operator
station station
F&G mimic
Plant wide panel
historian
Fault-tolerant
Ethernet ethernet
Serial interface
Digital Battery
video server backup DCS
Addressable F&G logic solver ESD safety
fire alarm panel safety manager manager
Digital
video cameras
Addressable fire detectors, F&G
modules, manual call points F&G detectors, manual call points Sounders, beacons mimic panel
Inherently safe plant design The logic solver acts as the central warnings of explosive and health haz-
Process control systems that are control unit (Figure 1), which receives ards, including combustible and toxic
safe and secure alarm and status or analog signals gas releases, thermal radiation from
Automatic safety, security and from field monitoring devices such as fires and minute traces of smoke
mitigation systems detectors or manual pull stations, ini- in sensitive equipment enclosures.
Work procedures tiates alarms and mitigates the haz- They also provide audible and visual
Alarm systems ard. Correct and proven connection alarm indications to ensure that op-
Mechanical protection systems of detectors to plant safety systems erators and personnel are informed
Acting as the prevention safety layer, is an important factor in reliable per- of hazardous situations. With these
the safety shutdown system takes formance of the F&G system and for improved detection capabilities, F&G
automatic and independent action to establishing the desired SIL. systems automatically initiate execu-
prevent a hazardous incident from oc- In the past, a proprietary F&G sys- tive actions, such as deluge systems
curring and to protect personnel and tem was standalone or a hardwired and evacuation procedures. This
plant equipment against serious harm. mimic overview panel that was linked minimizes the escalation of safety in-
Still, in reality, incidents unfortunately to a control system. With these older cidents and protects personnel, prop-
escalate far beyond this layer. systems, fire-control measures had to erty and the environment.
This problem is the main reason be manually activated, which was far Further value can be realized an
that F&G systems should remain at from an ideal practice. Today, however, F&G solution is plugged into an inte-
the core of any integrated solution. F&G detection systems are gener- grated system that provides common
Used for automating emergency ac- ally programmable electronic systems tools, operating interface and network-
tions with high-integrity safety and (PES) with high safety availability and ing, resulting in independent systems
control solutions, F&G systems act as mitigation effectiveness. As modern that are tied together on a common
the mitigation safety layer that takes F&G systems are tightly integrated platform. A recommended approach
action to lessen the consequences of with the overall process safety strategy, is integration at the controller level,
hazardous events. They are also criti- mitigation is either enacted through which provides: plant-wide safety in-
cal to resuming full production when an emergency shutdown (ESD) system strumented system (SIS) point data;
recovering from these incidents. or directly by the F&G system itself. diagnostics and system information;
A good F&G system combines state- Using these new techniques and add- alarms and events, operator displays;
of-the-art detectors, conventional and ing intelligence to instruments to re- and sequence of event information to
analog addressable fire panels, clean- duce the number of spurious alarms any station. This minimizes manual
agent and inert-gas fire suppression has greatly improved detection rates. intervention and drastic plant shut-
systems, and a safety integrity level Thanks to these advances in F&G downs, reduces hardware costs, and
(SIL) certified logic solver (Figure 2). detectors, the systems can detect early allows plants to recover more quickly
Chemical Engineering www.che.com May 2009 41
Feature Report Secure process control
and easily from process upsets and ab- Abnormal situation management
normal situations. Effective operating environment
Modern safety systems, when seam-
Boundary management
lessly integrated with the plant au-
tomation system through a secure Emergency shutdown system
communication network, will transfer
Physical protection
alarm signals, fault and trip signals
and system diagnostics. Information Emergency response
from all related systems can be man-
aged from the same location, enabling
that additional layer to monitor the
Figure 3. Industrial operations benefit from a holistic approach to safety that
status and operability of the total supports a secure process control network to the perimeter of the plant to protect
F&G detection and control system. people, assets and profitability. A layered safety strategy encompasses process and
Overall SIS technology integrates system technology and the people who interact with that technology to help
safety measures dispersed throughout plants achieve their safety objectives
a plant to reduce risk to employees
and assets, increase process availabil- lated layers of protection (Figure 3) many cases, reduce overall wall-to-
ity, and improve regulatory compli- plays an important role in improving wall project costs by 25%. Seamless
ance. SIS solutions can be integrated safety and efficiency. integration with the ESD and DCS
with F&G detectors for increased pro- The F&G system should have com- through a common network proto-
tection, and unified with third-party munications integration with the col provides a safe landing in case of
systems to reduce validation and ac- plant distributed control system emergencies and eliminates the need
ceptance testing costs. (DCS) in order to have F&G graphics for additional equipment or engineer-
Combined with the new generation and alarms displayed to the operator. ing. Integration of fire detection and
of F&G solutions, an integrated sys- However, there also should be inde- security systems for offsites and utili-
tem provides alerts of abnormal situ- pendent displays, such as independent ties with the plant automation infra-
ations in a fast, accurate and struc- human-machine interfaces (HMIs), structure further improves operator
tured way, giving personnel time to for plant operators to respond to F&G efficiency, through single-window ac-
decide upon the correct course of ac- excursions when the DCS HMIs are cess for alarm visualization, diagnos-
tion. These solutions, which include not available. The plant F&G system, tics, and events and historians.
new integration capabilities with with a fire system for occupied build- These truly integrated safety sys-
process-simulation tools, F&G detec- ings, should also be integrated with tems deliver several benefits, includ-
tors and control communication pro- the plant evacuation and site security ing the following.
tocols, enable safety engineers to de- center for efficient plant-evacuation Integrated operational interface
sign large integrated and distributed procedures. This enables plant man- Integrated peer control
plant-wide safety strategies. With in- agers to keep better tabs on personnel Integrated diagnostics
novative simulation solutions, safety and efficiently coordinate with first Integrated postmortem analysis
engineers can easily test the impact responders during emergencies. Integrated F&G system
of safety strategies on the overall As part of an overall plant-safety Integrated power supplies
plant design and operations before strategy, end users need a unified Integrated modifications
implementation. This reduces overall platform for emergency shutdown and Integrated simulation and
risk and the impact of system modifi- F&G detection. A single window for optimization
cations and ultimately increases prof- operators and a common tool for engi- Operational integration provides
itability by bringing new plants into neering and maintenance drives down a seamless interface to the process
full production much faster. operational risk and costs. under control, and at the same time,
In addition, new field-device con- Integrated control and safety sys- maintains safe separation. From an
figuration tools allow plant personnel tems (ICSS) provide multiple benefits operational perspective, it makes no
to automatically configure intelligent to process plants. They help opera- difference where the application is
safety devices and integrate them tors minimize intervention and shut- running. All required information is
into the control system database. downs, and recover more easily from available to the operator. This allows
Facilities subsequently save money process upsets. They also allow facili- applications ranging from rotating
by using a single tool to manage all ties to reduce hardware and installa- equipment and compressor-protective
equipment assets. tion costs, and ensure easier system systems, to emergency shutdown sys-
configuration with preconfigured tems and large plant-wide F&G ap-
Toward integrated systems function block selections. plications to be monitored from any
As more plant owners move toward Plants implementing an ICSS plat- operator console.
highly integrated plant-production form for F&G, ESD and DCS systems Additionally, industrial operations
systems, appropriate integration with can significantly lower their opera- benefit from a holistic approach to
multiple interdependent yet interre- tion and maintenance costs, and in safety that supports everything from a
42 Chemical Engineering www.che.com May 2009
Review
potential F&G incidents as
Integrated safety in oil and gas
L
part of risk assessment
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-
Define loading operations all require F&G and safety systems to protect personnel, assets and
role required of F&G the environment. The consequences of incidents at oil and gas terminals, as illustrated at the
system in risk reduction Buncefield fuel terminal, can be enormous (Figure 5).
In oil and gas terminal applications, operators not only need overfill protection but also an
Select 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
appropriate system type
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
Set less expensive implementation and maintenance, since all components work together.
performance standard 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
Develop detectors, integrated with the overfill shutdown system, to detect incidents before they esca-
specification test, mainten-
late. The system will mitigate safety incidents by ensuring that all personnel are informed of
ance, inspection procedure 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
Implement protection, but will also integrate security access control, personnel mustering systems and
performance standard 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.
Figure 4. An effective project
strategy starts with an assessment
of future or existing fire and gas per-
formance according to functional safety tools and procedures for managing safety lifecycle as outlined in safety
standards. Based on this assessment, abnormal situations and reducing standards such as IEC 61511. An in-
end users have a detailed roadmap for incidents. When an abnormal situa- tegrated main automation contractor
installing new equipment or updating
obsolete infrastructure to an optimal
tion occurs, alarm management, early (I-MAC) can help identify F&G haz-
level of safety. The main automation con- event detection, and abnormal situa- ard points and possible risks, and then
tractor can help to identify F&G hazard tion management (ASM)-designed develop basic design packages and re-
points and possible risks, and develop displays ensure operators have the lated acceptance test criteria to meet
basic design packages and related ac- information available in the context safety requirements. This results in
ceptance test criteria to meet safety
requirements they need it. This enables faster reac- optimized risk reduction and opera-
tion to hazardous situations. tional performance, better compliance
secure process control network (PCN) Next, properly designed emergency- with safety standards, and increased
to the perimeter of the plant. This lay- shutdown systems and automated lifecycle sustainability.
ered safety strategy encompasses pro- procedures can move a plant to a safe Supplier assistance can extend
cess and system technology and the state in the event that an incident es- to implementing SIS solutions; live
people who interact with that tech- calates beyond the inner sphere of pro- hot cutover, implementation and ex-
nology to help plants achieve their tection. Should an incident occur, F&G ecution of revamps; and installation,
safety objectives. detection solutions, coupled with rapid commissioning and safety validation.
A layered safety strategy unifies location of individuals and a carefully To sustain the end users F&G sys-
all plant protection layers (including designed emergency response proce- tem performance, leading automa-
basic control, prevention and mitiga- dure, will help contain the impact. tion contractors also provide lifecycle
tion, as outlined in IEC 61511 stan- Finally, a layered approach to safety support services that include periodic
dard) required for achieving optimum protects the perimeter of the plant proof testing; system maintenance;
functional safety. It also provides the using physical security that safe- training programs on safety, code and
required functional safety with a high guards access to structures and moni- standard compliance; and spare parts
SIL. This includes superior visualiza- tors traffic approaching the facility. management.
tion and logging facilities enabling op- When designing a truly integrated
timal operator response and accurate system, an effective project strategy Challenges and the road ahead
evaluations. By integrating basic con- starts with an assessment of future Industrial standards such as IEC
trol, prevention and mitigation com- or existing F&G performance accord- 61511 (ANSI/ISA S84.01 in the U.S.)
ponents, a CPI company can vastly ing to functional safety standards play a large role in developing, imple-
reduce its overall project costs and (Figure 4). Based on this assessment, menting and installing F&G systems.
ongoing maintenance expenses. end users have a detailed roadmap for The overall safety lifecycle model de-
At the core of a layered safety strat- installing new equipment or updating scribed in the IEC standard lists all of
egy is process design the embodi- obsolete infrastructure to an optimal the necessary project activities, from
ment of the business, safety and pro- level of safety. This process begins the concept (definition) phase to the
duction considerations necessary for with a hazard and risk assessment decommissioning phase, necessary to
effective operations. At the next layer, (such as HAZOP) and then contin- ensure the functional safety of equip-
the layered approach implements ues through the various steps of the ment under control (EUC). These ac-
Chemical Engineering www.che.com May 2009 43
Feature Report
Amin Almasi
WorleyParsons Services Pty. Ltd.
O
ne of the most significant limi- collapses, and a sudden backward flow both hardware and software, should
tations on the operation of dy- of gas occurs from the discharge to the be based on a design algorithm that
namic compressors (axial and suction side of the compressor. operates very quickly when opening
centrifugal compressors) is This surge phenomenon can cause a the anti-surge (recycle) valve to avoid
the low-flow limit known as surge. reversal of the thrust loads (large dy- the surge. However, this anti-surge
A sudden valve closing, such as for namic/transient forces on the rotor), system should be designed and pro-
emergency shutdowns or for a suc- which can result in damages to bear- grammed so as to not make the circuit
tion/discharge valve, presents sig- ings (particularly thrust bearings), (or the compression unit) unstable.
nificant challenges to anti-surge seals and other parts of the compres- Centrifugal and axial compressors
valve sizing and selection. Anti-surge sor. For these reasons a surge and experience a surge whenever they
valve requirements depend to a large particularly high-energy surges are suddenly tripped. Controlling the
degree on the compressor type, the should be avoided. Recent innovations head and total energy of the surge
compression system arrangement, in control technology, dynamic simula- event, particularly a surge at the trip,
and particularly the volume of the tions, rotating machinery knowledge is the main objective of dynamic simu-
discharge piping downstream of the and control-valve design (the anti- lations and anti-surge valve sizing
compressor. Important aspects of surge valve) have made it possible to and selection. Meaningful gains can
anti-surge valve sizing and selection supply anti-surge systems that are be made by a better understanding of
for plants in the chemical process in- capable of coping with rapid flow fluc- the interactions between the compres-
dustries (CPI), as well as examples, tuations and process gas variations. sor, the anti-surge valve(s) and the
are discussed in this article. A dynamic compressor, also known facility piping and layout (including
as a turbo-compressor, is generally coolers, scrubbers, check valves, vent
The basics provided with an anti-surge valve (Fig- valves and more). Key parameters for
At low flows, the performance of a ure 1) in order to keep the compres- anti-surge valve selection and sizing
dynamic compressor particularly sor in a stable operating range. This include the following: fast response,
centrifugal and axial compressors is achieved by assuring a suction flow optimum size, high integrity, accuracy
is limited by the occurrence of aero- that is higher than the corresponding and noise abatement.
dynamic flow instabilities that could flow at the surge. An anti-surge valve
lead to a catastrophic failure of the is placed in the anti-surge circuit (a The anti-surge valve
compressor due to mechanical or ther- recycle circuit). The size of an anti-surge valve influ-
mal loads. The surge is an unstable The main purpose of this valve is to ences the amount of fluid that can be
flow situation that occurs when the avoid a surge and surge-related dam- moved from the discharge side of the
flow is too low. When a dynamic com- ages, particularly irreparable dam- compressor to the suction side. Larger
pressor reaches its surge limit, the ages (Figure 2). The anti-surge system valves allow more flow. The speed
flow pattern through the compressor and the controller, which comprise with which an anti-surge valve opens
Chemical Engineering www.che.com November 2012 43
Engineering Practice
is also important in rapidly reducing valves with the same opening time tem, the valve dissipates a large por-
the discharge pressure (the head and (around one second) were evaluated tion of operating power of the compres-
stored energy at the compressor dis- and compared. The small anti-surge sor. A portion of the power dissipated
charge). One important trade-off in valve resulted in a drop in flowrate by the valve is converted into acoustic
relation to anti-surge valves is that without much change in the pressure energy, which becomes valve noise.
larger valves exhibit slower opening head. The shutdown surge occurred at Apart from the noise, considerations
rates. There are certainly cases when a moderately high discharge pressure. should also be given to the fluid veloc-
a faster, smaller valve is better than a The large anti-surge valve with a ity in the valve outlet. This should be
larger, slower valve, but in all cases, valve flow coefficient (Cv) about two kept within some limits to avoid pipe
the anti-surge valve should be large times that of the small valve re- vibrations. Typically, CPI plants use
enough to handle the flow supplied by sulted in a major reduction in the an anti-surge-valve noise limit range
the compressor at a sufficiently low head. The shutdown surge occurred (before the external attenuation) from
differential pressure (head). at a discharge pressure just above the 85 to 100 dB(A). The anticipated noise
Another important factor is control- suction pressure. The surge event in level (before the external attenuation)
lability. Larger valves are more dif- this case was not even noticeable. The should not exceed 105 dB(A) with fluid
ficult to control. The faster the valve flow reversed through the compressor velocities below Mach 0.3.
can be opened, the more flow can pass at a low head point. This case study Anti-surge valves should be de-
through it and the more effective it is shows the surge-preventing effects of signed to increase the total system
in surge prevention. There are limits an anti-surge valve with a higher Cv efficiency over a wide range of the op-
to the valve opening speed, dictated compared to a smaller valve with a eration. The inline and symmetrical
by the need to control intermediate lower Cv. flow path eliminates indirect flows
positions of the valve. There are some Additional requirements for anti- and unnecessary changes in the flow
practical limits to the power of the ac- surge valves include the following: directions through the valve. Axial-
tuator. The situation may be improved 1. The reduction of stroke time (in flow anti-surge valves are most often
by using a valve that is boosted to response to control signal step specified. This design is rugged and
open, combining a high opening speed changes) offers high performance. It is highly
for the surge avoidance with the capa- 2. A stable response reliable and requires little mainte-
bility to avoid oscillations. 3. Minimized overshooting during nance. The breaker vanes are often
During a surge event, the flow in- the valve adjustment steps used in the downstream section of the
creases rapidly at first and then de- The requirements for an anti-surge valve body (downstream of the valves
creases more slowly as the pressure valve should always be optimized. Too flow-control internals), which cut and
difference across the compressor fast a response can result in excessive streamline any flow turbulence. This
and consequently the anti-surge overshoot and poor accuracy. How- can result in significant reductions in
valve, too decreases. With a rela- ever, too slow a response could result the noise and the turbulence, elimi-
tively large anti-surge valve, the en- in sluggish opening of the anti-surge nating excessive vibrations in the
ergy level at which the surge occurs valve and possible inadequate pro- valve and the associated piping, sup-
is much lower compared to a smaller tection of a compressor from a surge. ports and structures.
anti-surge valve. An anti-surge valve Most information required for the siz- The anti-surge valve should be ca-
with a relatively large capacity (com- ing of anti-surge valves is available pable of changing its position in a
pared to the compressors volumetric on the compressor map (the compres- short period of time. For reasons of
flowrate), but with the same opening sor curves). Well-known compressor reliability, a spring-loaded actuator is
time (compared to relatively small manufacturers and anti-surge system desirable. The typical control instru-
anti-surge valves), will allow the suppliers have their own standards mentation and accessories provided
shutdown surge to be avoided (or the for the safety margins on anti-surge on an anti-surge valve can include a
shutdown surge will occur at a lower valve capacity. This figure is usually valve positioner, a volume booster and
head condition). On the other hand, a based on the compression system dy- a solenoid. In other words, usually a
relatively small anti-surge valve could namics, the valve response and the solenoid valve with a well-designed
cause a more energetic and more po- system characteristics. booster device (that accelerates the
tentially damaging shutdown. The anti-surge valve should be ca- positioner operation) is used.
The ability to quantify these ef- pable of passing 100% of the surge The correct size and configuration of
fects and select a proper valve size flowrate at around 50% of the valve the required actuators, instrumenta-
by dynamic modeling can be very opening. In other words, the anti- tion and accessories could guarantee
useful. Because of this, accurate surge valves Cv is selected to be ap- an anti-surge-valve response time of
simulations are encouraged in order proximately two times that of the re- less than 2 s (to fully open). An over-
to select an anti-surge valve with an quired Cv, based on the compressor shoot on intermediate changes should
optimum size. surge flow on the highest compressor be kept to an absolute minimum.
The effect of valve flow coefficient. speed curve (on the compressor map). Control range. The anti-surge valve
In a case study for a centrifugal com- Valve noise. When the anti-surge should have a wide control range. The
pressor in a CPI plant, two anti-surge valve is opened by the anti-surge sys- whole valve system should operate
44 Chemical Engineering www.che.com November 2012
Figure 2. position and not by the torque. Rapid
Axial com- changes in the differential pressure
pressors are
very vulnera-
across the anti-surge valve should
ble to damage have no effect on the stability of the
from surges. valve position. A fast response is usu-
This photo de- ally achieved with properly sized ac-
picts the type tuators in contrast to those fitted on
of blade dam-
age that can conventional control valves. These
be expected advanced actuators are usually very
as a result of efficient and compact. They cannot
a surge event be bulky or massive. Modern designs
in an axial
compressor
contain a minimum number of moving
parts and a short valve stroke. These
advanced designs permit fast actions
very quickly in case of an emergency roethylene (PTFE) with a special and reliable operation.
(particularly an emergency-trip surge design is specified. For low molecu- Various studies, different simula-
event). A very high operating range- lar-weight gases (such as hydrogen) tions and extensive operational ex-
ability is usually required. Typically, a high-quality graphite packing is periences pertain to a typical one
rangeability of 150:1 is specified. The commonly specified anti-surge valve, one compressor
rangeability parameter is defined as The trim materials should be cho- casing arrangement. More com-
the ratio between the rated Cv for a sen for good corrosion-proofing and plex systems of cascaded valves (or
completely open valve and the mini- erosion-proofing. For usual services, valves around multiple compressors)
mum Cv that the valve can control. a proper stainless-steel alloy is typi- require a more detailed analysis and
A high rangeability value means suc- cally specified. For a special appli- sophisticated provisions.
cessful control even with a high differ- cation, a special, sintered tungsten Valve and size selection. An anti-
ential pressure and a low flow. carbide could be a good selection. surge valve should be sized to meet
To increase the rangeability, special The trim design should allow for ex- two diverse objectives: steady-state
trims are used where the area distri- cellent sealing operation and transient cases.
bution is made according to the valve For high-pressure anti-surge valves, During steady-state recycling, the
characteristic curve. For example, in additional requirements should be required capacity of the anti-surge
some valve designs, until 40% of the considered. For example, the welded valve can be directly derived from
stroke, the Cv is typically limited to seat is replaced by a high-quality the compressor map (the compressor
1525% of its maximum value to ob- cage seat for superior sealing, better curve). Typically, smaller valves ex-
tain superior controllability. Too often, centering and easier maintenance hibit smoother control. During tran-
a special trim with multiple jumps is and operation sient conditions, however, the required
specified for an anti-surge valve. Leakage. Leaks across an anti-surge valve capacity increases. To avoid a
Service reliability. Good service re- valve can influence the efficiency of a surge during an emergency shutdown
liability even under very severe condi- compressor system. Even under the (or similar transient events, such as
tions (for example, compression ratios most severe working conditions, high- a sudden suction or discharge valve
lower than 0.3, a discharge pressure quality anti-surge valves should main- closing, or others), a bigger valve will
ranging up to 600 bars or more, and tain a tight shut-off (within practical give better performance. To fulfill both
other difficult situations) is required. limits) over the full pressure range. above-mentioned requirements, an
In order to obtain these requirements, With this feature, the compressor ef- anti-surge valve with an equal per-
the following considerations should ficiency is maintained at the highest centage characteristic (defined below)
be respected: levels during the normal operating is recommended. Two types of anti-
Anti-surge valves in various trim mode. In a case study for a medium- surge valves are generally used: globe
styles are usually fitted with pres- size high-pressure centrifugal com- valves and ball valves (noise-attenu-
sure balanced pistons. The required pressor (8-in. ASME Class 900 recycle ating ball valves).
thrust should be virtually indepen- control valve), with an ASME Class The equal percentage characteris-
dent of the differential pressure IV leakage rate and a differential tic spreads the first half (50%) of the
across the valve pressure of around 70 bars, the valve valves fully open capacity over the
Bushings should be anti-seize and leaked approximately 150 Nm/h of a first two thirds (66%) of the valves
self-lubricated. The internal slid- compressed gas. travel for a globe valve, and about one
ing area requires high-quality and The leakage feature should be inde- third (33%) of the valves fully open
special design and materials (usu- pendent of the actuation method. New capacity over the first two thirds of
ally chromium-plated internals are designs do not need a higher torque as the valves travel for a ball valve. An
specified) was required in older designs. In mod- anti-surge valve should be described
Proper packings should be used. ern designs, the ASME Class IV seal- by its maximum capacity (Cv), and by
Typically, a charged polytetrafluo- ing class is usually achieved by the its capacity as a function of the valve
Chemical Engineering www.che.com November 2012 45
Engineering Practice
travel, and the opening behavior. The should be evaluated. Parametric stud- surge on a compressor shutdown, and
globe valves Cv approximately var- ies of the anti-surge valve size and also holds the anti-surge valve in a
ies with the square of the percentage opening time can yield optimum sizes fully open position for a startup.
travel. The ball valves capacity varies for an anti-surge valve (and a hot-gas- Extreme care has to be taken when
roughly with the cube of the percent- bypass valve, if required). testing an anti-surge system during
age travel. A ball valve will have more Overheating. A large portion of the operation. The compressor should
capacity to depressurize the discharge mechanical energy produced by a com- not accidentally be pushed into a
volume compared to a globe valve of pressor is converted into heat in the surge. The controller should take
the same size. discharged gas. In an uncooled recy- over if something causes the oper-
In a case study for a 6-in. (150 mm) cle-loop system, this heat is recycled ating point to head toward a surge.
anti-surge valve, the Cv of a ball valve into the compressor suction end, and Such features should be available,
was more than 2.5 times of the Cv then more heat energy is added to the enabled and working correctly. A sud-
for a same-sized (6-in.) globe valve. recycling gas. In a typical natural-gas den movement of the anti-surge valve
At two thirds of valve travel, the ball centrifugal compressor at a 100% re- could have an adverse impact on the
valve flowrate was more than 50% cycle without recycle cooling, the tem- remaining gas system (the upstream
higher than that for the same-sized perature of the gas increases about and the downstream of the com-
globe valve. This additional flow ca- 1C per second. Eventually, this could pressor). All of these control actions
pacity would sometimes, theoretically, lead to overheating at the compressor should be evaluated properly.
make the ball valve a better choice in discharge side. It could be useful to assess how the
an anti-surge application. This is par- In theory, the recycling can lead to anti-surge valve behaves in response
ticularly true where there is only a heating whenever the compressed gas to a small 12% change in the com-
single anti-surge valve per stage (for cannot be cooled to the initial operat- mand signal. Bad signs are hunting,
example, there is no hot-gas bypass) ing temperature at the compressor stiction (the static friction that needs
and the discharge volumes are large suction side. This is the case both for a to be overcome when the anti-surge
(especially when the anti-surge loop hot-gas bypass and a partially cooled valve-position should be changed), hys-
is taken from downstream of the dis- bypass (if the cooled bypass is not teresis and delayed lifting, all of which
charge cooler). However, this is just a designed properly, for example, when could make the control more difficult.
theoretical and textbook idea. sufficient cooling capacity is not pro- The anti-surge step test should be
From a practical standpoint, the vided). The overheating could be de- performed with the controller in the
globe valve behavior is more predict- fined as the gas temperature within manual mode and with the valve off
able. The globe valve is more comfort- the compressor exceeding 150190C, its seat, preferably somewhere be-
able to control and manage. It is rela- depending on the process. The same tween 1585% open. The compressor
tively less nonlinear. The ball valve problem may also occur if there is a will be on partial-recycle through-
usually offers a highly nonlinear be- long period of time between the start out the test (the remaining process
havior (Cv varies with the cube of the of rotation (the compressor startup) should be able to cope). The first
percentage of travel). and overcoming the pressure down- valve step could probably be in the
The anti-surge valve should be stream of the discharge check valve, open direction so as to move the com-
matched to the compressor system. when some kind of hot-gas bypass is pressor further from the surge. If the
Both globe and ball valves are used in used for the startup. response is unknown, a small valve
modern anti-surge systems (depend- movement (even typically 12%)
ing on the application). Practically, a Control and testing should be selected.
globe valve is more common in anti- An anti-surge-system configuration One important consideration when
surge systems. usually allows an operator to open operating in the recycle mode is that
In some compressor installations, the anti-surge valve further than the the gas could thin out due to loss of
it is necessary to have both a nor- valve position that is defined by the the heavier components. The surge
mal anti-surge valve (the recycle line anti-surge controller (based on the al- line of the lower molecular-weight
taken from downstream of the after- gorithm) in manual-mode operation. gas may be rather different from the
cooler or the anti-surge line that in- However, the operator cannot close normal gas. Some amounts of new gas
cludes a cooler) and a hot-gas bypass the anti-surge valve any further than should always be fed through the com-
valve, which can be opened to rapidly the surge controller permits. The con- pressor. Another useful test is to in-
transfer the compressed gas back to troller can override the manual mode, troduce a disturbance to the compres-
the suction side of the compressor if necessary, to open the anti-surge sor, such as cutting back the suction
without cooling (usually for a short valve. The fail safe philosophy that throttle valve slightly or changing the
period of time). A hot-gas-bypass al- is commonly employed is that the driver speed (if it is a variable-speed
lows the pressure head across a com- anti-surge valve should be an air-to- compressor train). It should be veri-
pressor to be rapidly reduced during close valve (fails open). The control fied that the anti-surge valve changes
a trip. Various options for anti-surge is used to force the anti-surge valve position without any delay and also
valve size, opening rate, recycle loop fully open whenever the compressor without a significant undershoot of
arrangement and operation sequence is shutdown. This helps to prevent a the setpoint.
46 Chemical Engineering www.che.com November 2012
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
Figure 1. This photo
Preventing
shows the effects of
pneumatic transport of
powder to a silo through
a hose not designed
Dust Explosions
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
Risk management programs are critical potential ignition hazards,
including the control
for safe handling and processing of combustible dust of static electricity and
process equipment hot
as well as for OSHA regulatory compliance spots, similar to the pic-
tured example
Geof Brazier and Mitch Rooker PAs current standards define best en- of housekeeping and dust collection-
BS&B Safety Systems, LLC gineering practices and are designed to and-filtration measures. These mea-
H
protect both personnel and plant from sures are combined with inspection to
ousehold items, such as combustible dust explosions. These ensure that they remain effective. A
breakfast cereal, flour, corn standards are typically adopted by state layer of dust that is only 1/32-in. thick
starch and sugar are so com- fire marshals, insurance companies, is a concern when it covers 5% or more
mon that many were left consultants and their legal representa- of a workplace floor area that has
surprised after the 2008 combustible- tives and are referenced by the OSHA dust-laden structural members, such
dust explosion at a sugar refinery in NEP. OSHA can and does issue safety as joists and I-beams. Successful dust
Georgia. Heard in much of the early citations in relation to combustible dust control reduces fugitive emissions of
media coverage was the question How risks. Its citations are based on the Gen- combustible dusts, which in turn cuts
could sugar explode? eral Duty Clause whereby the owner/ the dust explosion risk.
Plant explosions are, thankfully, not operator of a facility shall provide a safe Quantification of a combustible
routine occurrences. However, when workplace, as well as by reference to dust risk requires that the charac-
they do occur, it is the unusual event specific NFPA standards including 61, teristics of the material be identified
often a combination of abnormal 68, 69, 86, 484, 499, 654, and 664. by test. Key parameters include the
events that is typically the trigger. maximum explosion pressure (Pmax),
This article outlines some of the guide- OSHAs NEP the deflagration index (Kst) and the
lines that are available to help prevent OSHA introduced the Combustible minimum ignition energy (MIE) of the
dust explosions. It also offers a list of Dust National Emphasis Program in dust. Prevention and protection strat-
factors to consider when choosing ex- October 2007 issued by the Directorate egies must also be based on knowledge
plosion vent technology to minimize of Enforcement Programs and revised of process conditions, such as dust con-
combustible-dust-explosion hazards. it on March 11, 2008 with directive centration, airflow velocity, operating
The facts are that in the U.S. and number CPL 03-00-008*. This docu- pressure, temperature and humidity.
elsewhere, the chemical process in- ment describes OSHAs policies and The higher the Kst value, the faster
dustries (CPI) are working to mini- procedures regarding inspection of fa- the rate of pressure rise due to com-
mize the risk of combustible dust ex- cilities that handle combustible dust bustion. Kst values above 600 are
plosions by following a revised suite of as well as their expectations regarding considered extremely explosive. Most
National Fire Protection Association owner/operator compliance with U.S. common plant dusts have Kst values
(NFPA) standards, that are recog- safety standards. The scope is national between 100 to 150. Even if a plant
nized by the U.S. Occupational Safety and administered by national, regional owner/operator knows the characteris-
and Health Administration (OSHA). and area offices of OSHA or the offi- tics of a particular dust, a sample must
A documented combustible-dust, risk- cials of participating state programs. be tested under NFPA 68. Particle size
management program that is regu- Its focus includes the following three can profoundly affect explosive proper-
larly updated, forms the cornerstone of technical areas of risk management: ties, as the finer the dust, the higher
a plants owner/operator responsibility 1) Dust control the Kst value. For example, sugar has
to provide a safe workplace. Identifica- 2) Ignition control a recorded Kst value of about 138. Of
tion of areas of risk, along with imple- 3) Injury and damage control more concern with regard to the po-
mentation of prevention and protec- Dust control. To summarize, dust tential damage to process equipment
tion strategies, achieves compliance control is achieved by the implemen- is the fact that a fine-sugar-dust explo-
with these safety standards. tation of appropriate combinations sion will generate a pressure in excess
OSHAs Combustible Dust National of 100 psi within an enclosed volume in
Emphasis Program (NEP) and the NF- * Available for review on www.bsbipd.com less than 100 milliseconds. Other key
Chemical Engineering www.che.com October 2009 49
Environmental Manager
F
consequence
rom plastics and synthetics to
fertilizers and fuel, the gen-
eral public takes for granted
so much of what the chemical Figure 1. The Swiss Cheese model involves the implementation of various safe-
process industries (CPI) produces, guards and barriers to prevent the propagation of a hazardous situation however,
and the risks associated with CPI gaps exist in this setup and must be addressed
production facilities are not typi-
cally considered. That is, until the facility operators by notifying them cedures; poor communication causes
next headline-creating industrial when risk increases, and enabling insufficient worker-to-worker han-
catastrophe occurs. The human toll them to respond before abnormal doffs; stress and excessive fatigue
of such disasters plays out across operation escalates uncontrollably impair decision-making and con-
ruined lives, devastated communi- and causes injuries, monetary loss tribute to procedural missteps; and
ties and obliterated opportunities. or environmental damage. This ar- poor human-machine interfacing
The effects can stretch out for years ticle examines how key elements, can lead to operator confusion and
in the form of chronic health condi- such as automation and informa- delayed responses.
tions, diminished earning capacities tion-management systems, as well Many companies minimize the
and contaminated environments. as operating discipline, can begin to contribution of human error in in-
However, those connected to the converge to make this proactive ap- cident initiation through the imple-
CPI do not forget the risks, nor can proach a reality. mentation of safeguards or barri-
they afford to. Catastrophic incidents, ers based on the so-called Swiss
large and small, have increased the Cascading causes of incidents Cheese model of accident causation
focus on process safety management Numerous studies and analyses have used in risk management (Figure
in production facilities around the concluded that the majority of safety 1). The purpose of this model is to
world. Although enormous progress incidents result from human error. reduce the likelihood of an incident
has been made in preventing them, Operating discipline is typically at occurring, or to reduce the impact
incidents continue to occur. In fact, fault, since these errors spring pri- if an incident does occur. However,
while safety incidents have been de- marily from defects and deficiencies once these safeguards and barriers
clining in number since 2008, those in following operating or mainte- are introduced into operations, rig-
that do happen have been increasing nance procedures, and in applying orous procedural controls are then
in severity, according to the Ameri- necessary administrative controls necessary to ensure their integrity.
can Institute of Chemical Engineers to ensure competency, effective com- Otherwise, inevitable human errors
(AIChE; www.aiche.org). munications, performance measure- and equipment degradation reduce
Today, the industry is on a cusp, ment and change management. For the effectiveness of the model.
seeking the next breakthrough im- example, work instructions that are Shifting tasks from manual to
provement in process safety man- incomplete, inaccessible or illegible automated operation reduces the
agement one that can empower lead to inconsistent execution of pro- potential for direct human errors in
Chemical Engineering www.chemengonline.com october 2014 53
Feature Report
Da
Whether manual or automated, a Procedures IInformation
In
ta
Inspections
consistent and accurate execution Maintenace Dat
a
of safety-critical tasks requires op- DCS
Information
I
Info
erating discipline and the ability to PLC Data
monitor for changes in risk. Incident SIS
SCADA ta
investigations reveal that deteriora- Da
HAZOP In
nt
nteg
Integrated P
Procedures
tion of the barriers and safeguards LOPA Operational efficiency
often start long before the accident SIL/SIF
SRS Actions
occurs, and that no systems existed
to detect and report their loss. There Figure 2. Various sources of information can come together to provide a realtime
are three main contributors to this visualization of operational risks
deterioration of barriers and safe-
guards: the passage of time, covert Holistic systems integration stamped signatures, prescribed re-
risks and complacency. Non-integrated process-safety pro- cords and quality tolerances. Any
The passing of time. Just because grams may address some of the low procedure can be further supported
a safety incident has not occurred hanging fruit, but they do not pre- with safety rules and automated in-
for some time does not mean that vent everything, as is evident from terlocking functions, to ensure that
all is well. If assets are poorly main- the prevalence of catastrophic in- manual tasks are sequenced prop-
tained and operating processes not cidents. The integration of manual erly, that the proper individuals are
regularly checked for safety effec- tasks with automated procedures notified for certain tasks, and that
tiveness, they will eventually stop and smart interface design aids in detailed findings are consistently
providing the level of risk reduc- avoiding deficiencies in procedure recorded for monitoring purposes.
tion they were originally designed implementation and communica- Further, workflow applications can
to provide. tion, as well as minimizing the po- be deployed to aggregate the results
Covert risk. Risk has a propensity tential for human error. The key to the desired level of granularity
to emerge from the least-expected pieces already exist in the opera- to facilitate adequate information
places. The inability to visualize tional-technology (OT) and informa- sharing and reduce the chances of
where the risks are and the source tion-technology (IT) realms and now human error.
of the next incident is an open door must be pulled into conjunction to Most promising of all is the oppor-
to disaster. comprehensively address the grow- tunity to integrate realtime analysis
Complacency. Statements like ing complexity of the human factors into the overall picture to identify
This is how we always do it and that contribute to safety incidents. operational risk before it translates
Dont fix it if it isnt broken are For example, on the operational into incidents, and to drive the most
heard frequently. As months and technology side, companies can effective risk-mitigation schemes.
years pass without incidents, it implement state-based unit-control Figure 2 illustrates how data can be
is all too easy to become compla- schemas to address a range of nor- leveraged through proper integrated
cent, especially when it is not read- mal and abnormal operating condi- procedures to help visualize risk.
ily apparent where hazards can tions. They can also employ auto-
stealthily develop. That is when mation systems to detect abnormal Static versus dynamic risk
poor habits can infiltrate processes, events and take preemptive actions Risk is not static; things are al-
and running overtime becomes the to stop incident propagation. Addi- ways changing. Safeguards can de-
new normal. tionally, they can implement abnor- velop faults, or they can be down for
Overall, the Swiss Cheese model mal situation management (ASM) maintenance. New and different ac-
alone is too porous and static to graphical standards to optimize op- tivities may be taking place during
achieve the next safety perfor- erator navigation during abnormal installations. Organizations, people,
mance breakthrough. Instead, the events and alarm management to resources and logistics can easily
confluence of human error, operat- promptly focus attention on safety- and quickly shift. In short, nothing
ing data and information-manage- critical issues. should be taken for granted. Rather,
ment systems must be addressed. On the IT side, companies can organizations must bolster safe-
Process-safety challenges must be integrate standard operating pro- guards and barriers in a dynamic
viewed as a holistic system-inte- cedures (SOPs) with automated fashion. To do this, they can moni-
gration problem in order to make delivery of the latest revision of tor leading indicators of increased
meaningful progress. the required tasks, electronic time- vulnerability to incidents on a day-
54 Chemical Engineering www.chemengonline.com october 2014
Feature Report
dent, and perhaps one of higher se- Daily review of cumulative Bi-weekly critical review
risk to understand the with asset managers and
verity, than was originally planned
impact of new information technical authorities
into the design of the unit. In effect, Break-ins to the plan, which will
the operator is viewing a window of defer planned work
a units behavior as operating ver- Defective devices or
sus as designed. equipment are found
and cannot be
Examining further, the operator
immediately repaired
consults a safety rulebook to iden- Unplanned changes Monthly
tify the cause of the increasing risk to organization governance
and to see what tasks are necessary capability, such as review with asset
to resolve the problem, ensure con- vacancies, sickness leadership team
or absence
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 Daily discussion between Quarterly reporting
operational integrity to group executive
unit or some of its equipment offline committee
manager and operations
if required by procedure. superintendent
Safety-critical and operational risk
Multilevel utilization of data assessment raised as required with task-
The same collected operational analysis support
data, along with its attendant data- Figure 5. Day-to-day information aggregation can help in scheduling periodic re-
integration infrastructure, that views regarding process safety and asset management
drives realtime analytics and im-
proved situational awareness can in which the data are collected and gate abnormal situations effectively.
also be leveraged in other ways to presented. This context is a necessity Figure 5 illustrates some ways that
increase operational risk aware- for comparing best-practice histori- aggregated data can be used to
ness, power predictive insights and cal data with realtime operational schedule reviews and discussions
enable proactivity. performance. Within this process- about safety-critical tasks.
By definition, realtime data have state context, events (planned or
a limited shelf life and quickly be- abnormal) can be collated, analyzed Employing cloud and mobile
come historical data. As todays re- and acted upon quickly. The previously discussed emerging
altime data become tomorrows his- Procedural, state-based process- paradigm for process safety man-
torical data, they can be compiled control standards, such as ISA 106, agement is proactive and driven by
to inform periodic critical reviews have been established to provide data from a myriad of sources. Now,
with asset managers and technical the appropriate data context and it is time to discuss how to best de-
authorities, monthly governance re- process-control vehicle for realtime liver the data to the appropriate lo-
views with asset leadership teams event mitigation of an abnormal cation, using three technologies that
and quarterly reporting to executive situation. Without providing such have proven to be game-changing in
committees. Engineers are able to context keys, it is very difficult to many industries: mobile (display),
analyze the accumulated data to en- provide the multilevel datasets re- cloud (database) and industrial Eth-
hance designs and processes, while quired to enable the analytics engine ernet (IE) communications.
managers and executives can utilize to work. However, with state-based Mobile technology has been prob-
the data to improve process safety process automation standards in lematic from a control standpoint
and enterprise asset-management place, basic process control is fully for the CPI. There is a natural pro-
strategies. Meanwhile, analyzing re- integrated with safety shutdown pensity for individuals to use mobile
altime data in conjunction with his- systems, providing a key piece of technologies to move things along
torical data can help organizations enablement technology for rapid, outside of normal workflows. Some-
more effectively identify and track closed-loop response to an abnor- times this works to great effect;
trends, not just within one facility mal situation. State-based control sometimes it does not. Nonetheless,
but across many, and help drive suc- strategies also provide the event there is no arguing the value that
cessful preemptive actions. context to continuous, streaming in- the correct application of mobile
However, time-based information formation, which is vital to analysis technology can have in maximizing
alone is not enough to present a and reporting. asset health and process safety.
complete picture of operational risk. Overall, interactions with inte- Areas where mobile is particularly
The attendant data-integration in- grated data, aggregated at the ap- valuable include field-data capture,
frastructure mentioned above is propriate level, improves the ability event logging, operator rounds,
required to provide the context (for to benchmark performance, plan al- safety inspections and audits and
example, the unit processing state) ternative courses of action and miti- communicating operations and
56 Chemical Engineering www.chemengonline.com october 2014
Feature Report
maintenance instructions, among protocols, IE enables a flexible, re- dation, whether it is mobile data, or
others. The larger the facility, and sponsive, end-to-end networking ar- data created automatically by in-
the more dispersed the resources chitecture that provides connectiv- strumentation and systems
and assets are, the more valuable ity, collaboration and integration. As Receive instantaneous alerts on de-
mobile capabilities become. Mobile a result, data can be delivered from veloping issues through visualization
technology is also economical, elimi- the shop floor to the cloud, and then and alarm or alert technologies on top
nates paperwork, enhances compli- from the cloud to a mobile device of realtime analytic applications
ance and because it is so perva- efficiently and securely. Use mobile and cloud technolo-
sive in everyday life the learning gies to deploy and access data
curve is far from steep. Operational integrity windows from the point of work or the
The advantages of cloud deploy- The new imperative for CPI com- process edge
ments are clear as well. Cloud-based panies is knowing with certainty Implement realtime detection
process-safety management enables the source of the next incident, and and automated response to ab-
organizations to quickly roll out being able to avoid it or at least min- normal conditions, be they partial
new process-safety capabilities, and imize its effects. Regulatory agen- process impairment or emergency
can enhance collaboration between cies, shareholders and employees shutdown events
individuals and departments. The demand this. Doing this will require Have a single window into op-
cloud can also reduce the costs of ac- the ability to: erational integrity powered by dy-
cessing and integrating data, help Reduce operational risk by un- namic risk analysis and leverag-
eliminate information silos and en- locking data for realtime insights ing existing risk data, automation
able cost-effective enterprise-wide and enterprise-level visibility systems, maintenance and opera-
visibility. Importantly, the cloud is Examine the big picture tions procedures and business sys-
also pivotal for transferring data in through enterprise-level bench- tems for planning, scheduling and
a controlled and secure fashion. marking and trend analyses to cost control
Another key enabler is IE. Sup- guide safety strategies and keep When all these abilities fall into
ported by wireless-mesh network- incidents at bay place, a company can capitalize in
ing and secure communications Trust the data by leveraging vali- new ways on existing infrastructure
PREMIUM
EXPLOSION
Ensure your PROTECTION
www.rembe.com
www.rembe.us
and Dust Retainer
Q-Box
Q-Rohr-3
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
Authors
Andrew Soignier is serving Angela Summers is presi- Mike K. Williams is a consul-
as vice president of Oil, Gas & dent of SIS-TECH (12621 tant with Modern Automation
Petrochemical Solutions for Featherwood Drive, Suite Consulting Services, LLC
Ventyx, an ABB company (800 120, Houston, TX 77034; (6204 Woodview Pass, Mid-
Town & Country Blvd. Suite Phone: 281-922-8324; Web- land, MI 48642; Email:
500, Houston, TX 77024; site: www.sis-tech.com). She mike.k.williams1952@gmail.
Phone: 225-751-3348; Email: has more than 25 years of ex- com). With over 39 years of ex-
andrew.soignier@ventyx.abb. perience in safety controls, perience in the CPI, Williams
com). In this role, he directly alarms and interlocks, pro- provides work-process guid-
oversees Ventyx Level 3 solu- cess engineering and environ- ance in the automation and
tions in the areas of opera- mental engineering. She is an operation of batch and contin-
tions, safety and asset health. active participant in stan- uous facilities. Prior to start-
Soignier joined Ventyx in 2007 and has held sev- dards committees and has written over 60 pa- ing external consulting, Williams worked for The
eral positions in executive account management, pers, technical reports and book chapters. She is Dow Chemical Company in the Specialty and Ad-
and as the sales director of Oil, Gas & Petrochem- currently editing the Center for Chemical Pro- vanced Materials divisions, where he consulted
ical Solutions. Prior to joining Ventyx, he spent cess Safety (CCPS) book Guidelines for Safe on and provided analysis and investment plan-
11 years in the process automation industry, with Automation of Chemical Processes. Summers ning for a number of new and rehabilitation capi-
a focus on electrical, process safety, critical con- received a B.S.Ch.E. from Mississippi State Uni- tal projects. He has a B.S.Ch.E. from the Univer-
trols and rotating equipment. He holds a B.S. in versity and a Ph.D. in chemical engineering from sity of Michigan, Ann Arbor and is certified in
electrical engineering and an executive MBA the University of Alabama. Lean-Six Sigma methodologies.
from Louisiana State University.
and benefit from a breakthrough quick enough. This is where pre- in state-based control can respond
in process safety performance. A programmed safety systems pro- in advance of the last line of defense
process-safety paradigm shift is vide the final layer of protection to and address the abnormal event,
achieved through integration of return a process to a safe state. when it is less than critical, before
realtime information, best operat- Unfortunately, safety systems shutting down the entire process.
ing practices and closed-loop con- are conservatively designed to fail This example of man-machine-
trol. Mobile procedure assistants safe. In some cases, this mode of ac- method integration provides the op-
can help avoid human error. How- tion may be premature, resulting in timum response to ensure high lev-
ever, in todays complex processes, false process trips, causing undesir- els of both safety and productivity.
human response time may not be able loss of productivity. Advances Edited by Mary Page Bailey
PROTECT PUMPS
DRY RUNNING CAVITATION BEARING FAILURE OVERLOAD
PUMPING
POWER
AMPS
Circle 18 on p. 72 or go to adlinks.che.com/50981-18
58 Chemical Engineering www.chemengonline.com october 2014
Solids
Environmental
Processing
Manager
Selecting Laboratory
Exhaust Systems
With upfront thought, exhaust systems can ensure
personnel safety and improve building efficiency
Figure 1. Roofline aesthetics are
Paul A. Tetley among a host of considerations when
selecting a fume-hood exhaust system
Met-Pro Corp. for a laboratory facility
A
s components of chemical re- stream) is also a key consideration. at the sash (damper) can occur, which
search facilities, fume-hood Other important considerations in- may push fumes into the room. Airflow
exhaust systems for labora- clude: system efficiency, initial and oper- velocities that optimize safety can vary
tory workstations are often ating costs and roofline aesthetics (Fig- by manufacturer, but typically are in
taken for granted. Yet the choice of ure 1). For example, quiet, vibration-free the range of 100 ft/min.
exhaust system, when designing or performance can be important to protect Exhaust flow from a fume hood should
specifying a laboratory workstation, delicate laboratory instruments. be greater than the supply air to create
can have a tremendous impact on the Before selecting a laboratory work- outside-in air movement and thereby
safety of the laboratory buildings oc- station fume-hood exhaust system, contain airborne contaminants.
cupants as well as those in adjacent the location of the workstation should Workstation fume-hood exhaust sys-
buildings. The choice of a fume-hood be carefully thought out. Its location tems can be either constant-air-volume
exhaust system can also affect a build- can significantly affect efficiency and (CAV) or variable-air-volume (VAV) de-
ings longterm operating costs. safety. For example, occupants in the signs. A CAV fume hood exhausts air at
A better upfront understanding of laboratory can create air turbulence a constant rate, with an airflow velocity
the various types of exhaust systems simply by walking past the worksta- that increases as the sash is lowered. A
and their operation can eliminate po- tion, causing contaminants to be pulled VAV laboratory fume hood attempts to
tentially significant safety problems outside of its air space. For a work- eliminate the variable velocity by vary-
in the future while optimizing the effi- space designed with an air diffuser di- ing the air volume exhausted through
ciency of the building and the reliabil- rectly above the fume hood, significant the hood. In doing so, it can reduce the
ity of the system. air turbulence can cause contaminants total supply of exhaust air, and con-
to escape into the laboratory. tribute to reduced operating costs. VAV
Exhaust system basics In addition, airflow into the labora- fume hoods use a closed-loop exhaust
The main purpose of a fume-hood ex- tory will affect the performance of a system to monitor the amount of air
haust system in a laboratory is to re- fume-hood exhaust system. Since the being exhausted in order to maintain
move contaminated air resulting from laboratory should maintain negative the required average face velocity. The
chemical fumes, reactions and other air pressure with respect to the out- control system also helps monitor the
laboratory activities. The system must side, the doors, windows, air vents and presence of contaminants. In addition,
provide maximum protection while other openings should remain closed many VAV hoods use visual and audi-
using the least possible amount of air. during normal operation to ensure that ble alarms and gages to warn of inad-
It must control the flow of two types any contaminants will be exhausted equate air velocity or exhaust system
of air: workstation exhaust, and the through the fume hood, rather than malfunctions.
clean air that replaces it. by other laboratory openings.
When designing new laboratory facil- The effectiveness of a laboratory Exhaust system fans
ities, or upgrading existing ones, safety fume-hood exhaust system can be Laboratories can choose from a number
with regard to re-entrainment of gauged in terms of safety provided. of fume-hood exhaust-system designs
workstation exhaust into the building The principle design parameter is to based on the requirements of the facil-
or adjacent buildings and indoor air allow safe removal of fumes away from ity. Exhaust systems rely on hidden,
quality to protect building occupants, the operator and from the occupants of but essential subsystems, including ex-
are the most important considerations the laboratory. The velocity of the air haust fans and ventilation systems.
for a fume-hood exhaust system. In entering the hood from the room is the Fume-hood exhaust systems typi-
addition, the systems ability to com- key factor in achieving this safety. If cally employ a network of air ducts
ply with appropriate pollution-control the velocity is too low, the fumes can (plenums) leading to a roof-mounted
standards (including elimination of escape the hood and enter the room. If exhaust system that is driven by one
unpleasant odors in the discharge the velocity is too high, air turbulence of three fan technologies: axial, cen-
Chemical Engineering www.che.com September 2010 43
Environmental Manager
FIGURE 2. A difference
trifugal, and mixed-flow impeller. The between exhaust system
fans in each of these categories can types is the height of
be direct-drive designs, with the fan rooftop stacks, a charac-
wheel mounted directly on the motor teristic with value in ac-
shaft, or belt-driven, with the fan wheel commodating space and
mounted on a separate drive shaft and weight restrictions
driven by power from the motor trans-
ferred through one or more belts. fans, and greater air pressure than into a network of ducts that transport
Axial fan systems Axial fan blades axial fans. These systems eliminate re- it to the outside of the facility, usually
draw air into fan housings and dis- entrainment of exhaust air and odors to the roof, where it is mixed with am-
charge air in the same axial direction into a facility and surrounding build- bient air. In some cases, particularly
ideal for moving large volumes of air, ings with improved indoor air quality where local ordinances dictate rooftop
rated at cubic feet per minute (ft3/min) (IAQ). This technology typically incor- appearances, mixed-flow impeller sys-
at low pressures, and also rated in static porates low-profile, roof-mounted ex- tems may be mounted in an enclosed
pressure (SP). Fan blades used in axial haust systems with lower-speed fans facility (this is usually done for aes-
fans are typically plastic, aluminum, as compared to the other technolo- thetic reasons).
stainless steel or fiberglass. Different gies. To minimize noise levels, these Using base-mounted bypass damp-
types of axial fans used for industrial systems employ low-vibration, direct- ers, a mixed-flow impeller can draw in
applications include panel fans, which drive motors to draw in the exhaust twice as much fresh air as exhaust air,
are wall-mounted to draw air into or air from a laboratory fume hood. resulting in a much lower concentra-
exhaust air out of a room, and tube A noticeable difference between ex- tion of contaminants and odor-causing
axial fans, which are connected to a se- haust systems with centrifugal fans elements to be treated from the labo-
ries of ductwork conduits for exhausting and those with mixed-flow impellers ratory fume hood.
contaminated or hot air from a location. is the height of the exhaust stacks:
Centrifugal fan systems Centrifugal mixed-flow impeller systems present a Release to atmosphere
fans have blade wheels mounted inside lower profile and smaller footprint on Because a mixed-flow-impeller ex-
a housing. The blades can be forward- the roof (Figure 2). This characteristic haust system can generate relatively
curved, backward-curved or straight. has obvious value in accommodating high escape velocities, even at fairly
Exhaust air is pulled into the inlet of space and weight restrictions, as well low fan speeds, additional fresh air is
the blower housing and deflected by the as in its aesthetic qualities (more and entrained into the exhaust mixture
fan blades into a spiral pattern. This more communities are passing ordi- when it is expelled from the rooftop
pattern of airflow is then focused by nances limiting overall building height, stack by means of a nozzle/windband
the scroll housing and directed into a including rooftop appurtenances). section. This mechanism creates a pro-
single outlet airstream. The airstream As the name implies, a mixed-flow tective envelope around the discharged
leaves the fans housing at a right impeller system combines the best at- exhaust-air mixture and leaves the
angle to the axis of the housing assem- tributes of both axial and centrifugal surrounding air stationary to improve
bly. A variation on the traditional, cen- air movement technologies. Impellers entrainment. The exit velocities of
trifugal scroll-type fan housing is the in this approach have high efficiency, these escape gases can exceed 6,300
tubular centrifugal fan, which uses a consume less energy and are typically ft/min and can send an exhaust plume
cylindrical housing rather than a scroll quieter. In addition, true mixed-flow more than 120 ft into the air, even into
housing and cutoff assembly. technology is stable throughout the steady crosswinds. Of course, a variety
A centrifugal fan is designed to work impeller performance curve. Therefore, of factors can determine the effective-
at fairly high revolutions per minute these impellers do not have a stall or un- ness of the system, including building
and with a tall, roof-mounted exhaust stable performance region, as is found architecture, surrounding terrain and
stack in order to discharge the exhaust on axial and centrifugal fan curves in even weather conditions.
air at sufficient height above a facility regions of higher pressure. The concern As an example, a mixed-flow impel-
to prevent re-entrainment of exhaust is that systems operating at or near ler moving an exhaust and clean air
air (and its possible noxious fumes or this unstable performance region can mixture at 80,000 ft3/min will typi-
unpleasant odors) into the building give rise to significant performance, cally generate 6,300 ft/min exit veloc-
and surrounding facilities. Because of sound and vibration problems. ity. According to wind tunnel research,
the size and height of the stack, it re- Mixed-flow impellers blend the ex- additional air is entrained into an air
quires expensive and complex mount- haust air from a laboratory fume hood plume through aspiration, even at exit
ing hardware for stability and to mini- with outside, ambient air to reduce the velocities of 3,000 ft/min (the mini-
mize vibration. This includes elbows, concentration of contaminated particu- mum velocity specified by the ANSI
flex connectors, spring vibration isola- lates, as measured in parts per million Z9.5 standard for clearing chemical
tors and supporting guy wires. (ppm) or milligrams per cubic meter contaminants in laboratory ventila-
Mixed-flow impeller systems Mixed- (mg/m3) to a level that can be safely tion systems).
flow impeller systems generally pro- and legally exhausted to the atmo- A belt-driven, centrifugal-fan ex-
vide greater airflow than centrifugal sphere. These fans draw exhaust air haust system and duct network can
44 Chemical Engineering www.che.com September 2010
also mix contaminated and clean air systems, compared to axial and cen- fuel costs for the building (for every 1F
similar to a mixed-flow impeller, with trifugal fan systems, usually translate in added heat, there is a corresponding
useful dilution effects when released into lower noise and vibration levels. In 3% reduction in heating fuel costs).
into the atmosphere. But the centrif- addition, for applications where noise
ugal fan approach requires a rooftop must be tightly controlled, mixed-flow Ductless laboratory fume hoods
stack with sufficient height, some- impeller systems can be designed with The three major laboratory fume-hood
times as high as 40 ft or more. For the integrated noise attenuators that do exhaust systems discussed so far as-
same dilution and operating efficiency, not increase the overall height of the sume the use of ductwork networks to
a mixed-flow impeller system will typ- roof exhaust fan. vent contaminated air from the labo-
ically rise only about 15 ft above the Axial and centrifugal fan exhaust ratory. In recent years, ductless labo-
roofline. Because of the lower height, systems can also be equipped with ratory fume hoods have emerged as
the mixed-flow impeller exhaust sys- noise-reduction attachments, such as lower-profile alternatives. In ductless
tem would not require the extra stabi- a discharge silencer, added onto the designs, the hood incorporates a carbon-
lizing hardware needed with centrifu- exhaust stack. But this adds to the al- bed or other type of filter to remove and
gal fan exhaust systems. ready tall stack height, and requires control contaminants. The filter media
Tall exhaust stacks are not the only additional mounting hardware to mini- must be matched to the types of chemi-
structures associated with centrifugal mize vibration induced in the building. cals used in a laboratory. General-pur-
fan systems. Often, centrifugal fan Even at their lower motor speeds, there pose filters will adsorb a percentage of
systems also need a rooftop penthouse, is little sacrifice in performance with most chemicals and gases, while par-
mainly to protect service personnel. a mixed-flow impeller. It can provide ticulate matter filters may be needed
Because centrifugal fan systems rely greater airflow than a centrifugal fan, for some applications. In some cases, a
on belt-drive motors in contrast to and greater air pressure than an axial chemical impregnate may be needed in
the more reliable direct-drive motors fan, at lower rotational speeds. the filter to alter the target chemical to
of mixed-flow impellers periodic Compared to belt-driven centrifugal a less hazardous material.
maintenance visits are required to ad- and axial fans, direct-drive mixed-flow A ductless, laboratory fume hood has
just belt tension or to replace pulleys impeller fans are energy-efficient and no external connections other than its
or failed belts. The penthouse enclo- cost-effective choices for fume hoods, power cord, making it extremely por-
sure maintains safe, workable condi- with about 25% greater energy effi- table and useful for facilities that re-
tions for maintenance personnel. ciency than belt-driven centrifugal fan quire more than occasional redesign.
In terms of reliability, laboratory systems. Improperly aligned shafts, The power cord connects to electronic
workstation fume-hood exhaust sys- pulleys, pillow blocks and motors in a systems that monitor airflow and filter
tems based on mixed-flow impellers belt-driven system can rob it of power, conditions for operator safety, as well
are relatively simple and maintenance- and can introduce reliability issues. as provide visual and audible alarms
free compared to the belt-driven fans Also, the outlet nozzle of a mixed-flow when needed. A ductless approach is
and blowers of typical centrifugal and impeller system can be designed for not a substitute for all ducted systems,
axial fan systems. The belts alone in high exhaust-gas exit velocities even and should never be used in any chemi-
a belt-drive system, which can stretch, at lower fan speeds and corresponding cal laboratories involving unknown
loosen and break over time, require reduced power requirements. results (for which the filter may not
regular maintenance and repair, re- For further energy efficiency, it may work). Admittedly, ductless laboratory
sulting in shorter expected operating be necessary to specify a variable-speed fume hoods sacrifice efficiency for por-
lifetimes. For example, the minimum drive system. The use of variable-speed tability, especially compared to a cen-
expected lifetimes (L10) of bearings drives can contribute to lower energy trifugal or mixed-flow impeller system,
in belt-driven systems are typically costs, running at slower fan speeds at and they require careful monitoring of
40,000 h, while the expected operat- night when laboratories are unoccu- the carbon filters condition so that re-
ing lifetime of direct-drive motor bear- pied. In addition, if one fan fails in a placement can occur once the filter has
ings can be at least 200,000 h for some multiple-fan exhaust system, the re- become saturated.
mixed-flow impeller systems. maining fans increase speed to com- Edited by Scott Jenkins
pensate. This is critical in many bio-
A quieter solution safety level (BSL) laboratories, where Author
Paul A. Tetley is execu-
Noise management is also an important redundancy is mandatory. tive vice president of Met-
consideration for most laboratories and In addition, energy savings are also Pro Corp. (160 Cassell Rd.,
Harleysville, PA 19438;
areas in some cases requiring firms possible by recovering some of the heat Phone: 215-723-4700; Email:
to perform noise studies at their prop- normally lost through exhausted air. 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,
erty lines to determine the noise levels Mixed-flow impellers can be equipped Tetley was vice president and
general manager at Strobic
of a facilitys various systems, includ- with unique heat-recovery modules Air Corp., a Met-Pro subsid-
ing the laboratory exhaust system. The that extract warmed air from the ex- iary. Since joining Strobic Air
in 1989 as engineering pro-
lower speeds and inherently quieter haust flow and transfer the heat energy duction manager, he has designed or invented
design configuration of the direct-drive to the intake of a buildings ventilation many innovative Tri-Stack fan systems, an
acoustical silencer nozzle, and a unique multi-
motors in mixed-flow-impeller exhaust system, significantly reducing heating fan plenum system.
Vacuum Systems:
PI
X
Cooling water
Recommendations
return
T
YZ intercondenser
Y
Z
Aftercondenser
For Safe Operation
Condensate
systems, mechanical vacuum pumps and integrated Gate valve PI Pressure indicator
Four-way valve TI Temperature indicator
vacuum systems are designed, operated and T Steam trap Piccolo connection
T
his article is the first in a two- Steam ejector systems contains a water-reactive chemical
part series, in which the types Steam ejector systems are gener- (such as acetic anhydride), a steam
of process vacuum equipment ally categorized into one of four basic bleed is not acceptable. If the ejector
(steam ejector systems and types: single-stage, multi-stage non- should backfire, this will push steam
mechanical vacuum pumps) are dis- condensing, multi-stage condensing, back into the process and could result
cussed.1 Emphasis is given on how and multi-stage with both condensing in a possibly violent reaction, which
they are constructed and operate, and and non-condensing stages. could overpressure the system.
how to design associated parts of the For many years, steam ejector sys- To avoid this problem, a nitrogen
systems (such as foundations, suction tems dominated process applications bleed can be used to control the suc-
piping, discharge piping, and so on), to thanks to their simplicity, low pur- tion pressure. If the jet should back
ensure safe operation. chase cost, good reliability and their fire, nitrogen will prevent the steam
Many unit operations including ability to be designed for very large ca- from reaching the process. However,
distillation, evaporation, drying, crys- pacities (in excess of 1,000,000 acfm) using nitrogen to control an existing
tallization, filtration, and others are and be operated at very low pressures multi-stage ejector may not be techni-
often conducted under vacuum condi- in the micron range (for instance, six- cally feasible. If nitrogen is to be used
tions. To achieve the desired vacuum, stage units are routinely designed for to control a multi-stage jet, the user
the following three process- vacuum suction pressures in the range of 310 must indicate this when writing the
systems can be used: microns). They also are available in a performance and equipment specifica-
Steam ejector systems variety of materials of construction. tions for the steam ejector.
Mechanical vacuum pumps Steam ejector systems are ideal for Using nitrogen to control the suc-
Integrated vacuum systems (which wet-vacuum and highly corrosive ap- tion pressure of the first stage that
combine steam ejectors and me- plications. As a result, they are still which is closest to the process will
chanical vacuum pumps) being used instead of mechanical affect the design of subsequent stages,
Available capacities and operating vacuum pumps in applications that and using nitrogen to control suction
ranges for vacuum pumps and vacuum require a reliable vacuum system that pressure can significantly affect the
pumping systems that are most often can tolerate corrosive chemicals, liq- design of the inter-condensers. If the
used in process operations throughout uid slugs and solids carryover. back flow of steam or water into the
the chemical process industries (CPI) Safety considerations. In gen- process equipment from the ejector
are listed in the Table on p. 60 [1]. eral, there are very few safety prob- system can cause a serious safety or
This article reviews the hazards as- lems associated with the operation quality problem, then a liquid separa-
sociated with steam ejector systems of steam ejector vacuum systems. tor (a so-called knockout pot) may be
and mechanical vacuum pumps, and However, one safety issue associated installed between the ejector system
recommends design and operating with their design is the potential for and the process equipment.
practices that can be taken to prevent back-streaming of steam into the pro- Another possible safety problem
or mitigate these dangers. cess equipment. may result from the loss of the inter-
For instance, steam ejectors nor- condenser cooling medium (usually
1. Editors note: This is the first half of a two-
part article. Part 2 is scheduled to appear in the mally use a steam bleed to control water), which could lead to overpres-
October 2012 issue. the suction pressure. If the process sure of the system.
Chemical Engineering www.che.com september 2012 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 Discharge
baffle with
Seal screen
0.0001-in.
clearance Inlet
Outlet
liquids and particulates in the suc-
Oil level tion stream. The degree of protection
view port
Discharge necessary will depend on the process
valve
application. If the upstream knock-
out pot is not continuously drained, a
Inlet
high-level sensor should be installed
Piston to automatically stop the pump or
close an isolation valve to prevent any
overflow from the knockout pot from
Automatic
decanting reaching the pump.
Slide Cam
Rotor Vanes pin
arrangement Oil-sealed vacuum pumps must be
Stator
protected from flooding if the oil flow
does not automatically stop when the
FIGURE 2. Shown here is a schematic of the internals of a rotary-vane vacuum pump is shut down. To prevent flood-
pump (left) and a rotary-piston vacuum pump (right) (Reprinted with permission ing, a valve can be installed in the
from Ref. 1) suction line to admit air or nitrogen
to break the vacuum when the pump
selection of an appropriate oil for the When either an oil-sealed rotary- stops. This prevents the siphoning of
application is critical for good pump vane or rotary-piston vacuum pump oil from the reservoir when the oil is
maintenance. is improperly installed a hazardous not being discharged from the pump
This dependency on the oil for inter- condition can result. The following by the normal pumping action. A zero-
nal lubrication is a limiting factor in installation practices will minimize speed switch or solenoid valve is used
the use of these types of pumps in the potential problems with these types of to actuate the air or nitrogen bleed
CPI. The integrity of the oil must be oil-sealed vacuum pumps [1]: and close the isolation valve. Closing
maintained to avoid internal damage Foundations. The foundation design the isolation valve allows the process
that could cause contaminant buildup. will depend on the type and size of the to remain under vacuum and allows
Particulates in the oil must be filtered oil-sealed pump. If the pump is inher- the vacuum in the pump to be broken
out ahead of the vacuum pump, and ently balanced, it can be mounted on more quickly.
water or solvent vapors must either be any floor that is able to support the Flooding may be avoided with some
removed ahead of the pump by using weight of the pump. However, pumps oil-sealed pumps by sealing the dis-
pre-condensers, or prevented from that are partially balanced, should charge line using a barometric leg
condensing within the pump by the be installed with vibration mounts or a discharge check valve. This al-
use of gas ballasting (air stripping) or or matting between the base and the lows the system to pull a vacuum on
oil distillation. floor. If the pump is not balanced, the the oil reservoir so that oil siphoning
For oil-sealed, rotary-vane vacuum base should be securely fastened to a stops. A check valve or knockout pot
pumps, a variety of oils are available, concrete foundation with anchor bolts must be installed in the suction line to
each with very different properties. and grouting. The pump manufacturer prevent reverse flow from carrying oil
The sealing oil used in such pumps should be consulted when designing into the process vessel. A knockout pot
that are destined for chemical service the pump foundation. It is always good is recommended for critical applica-
has several additional requirements. engineering practice to install flexible, tions where carryover of oil into the
For instance, it must: bellows-type connectors in the suction process vessel might result in a fire
Be inert to chemical attack by and discharge piping; these can help or explosion.
acids, bases, halogens and halogen- to isolate the rest of the system from Discharge piping. Condensable vapors
induced aging) pump vibration. that pass through an oil-sealed pump
Be resistant to strong oxidants such Suction piping. Certain precautions tend to condense in the discharge line.
as oxygen, chlorine and fluorine must be taken when installing the suc- The discharge piping should slope
Possess good solvent power for prob- tion piping for an oil-sealed vacuum away from the pump, and a drip-leg
lematic, mostly organic constituents pump. Care must be taken to prevent or knockout pot should be installed
of pumped media, such as oligomers, any backstreaming of pump oil into in the discharge line to prevent the
polymers, resins, crystalline decom- the process vessel and to prevent pro- backflow of condensate into the pump.
position or reaction products cess liquids and solids from reaching Backpressure on the pump should
Both oil-sealed, rotary-vane and pis- the pump. Backstreaming of oil is es- be minimized.
ton vacuum pumps need periodic oil pecially significant with larger pumps. Serious mechanical damage can re-
changes. The chief purpose of chang- Baffles and traps are frequently in- sult if the pump is operated against
ing of the oil is to remove particles that stalled in the suction line to prevent a closed discharge valve. Discharge
either originate from aging of the oil this from happening. valves are therefore not recommended
or particles that are generated by the Cold traps, knockout pots, wet scrub- unless the pump must discharge into
process (such as dust, decomposition bers, and bag filters are some of the a common exhaust header. If a dis-
products, and other contaminants). devices used to protect the pump from charge valve is required, an interlock
Chemical Engineering www.che.com september 2012 61
Environmental Manager
(that ensures that the exhaust valve is out of the flammable range. It is criti- improved safety, especially when com-
open before the pump can be started) cal that the nitrogen supply source be pressing explosive gases and fluids.
is recommended; in fact, this is a con- free of oxygen. There have been a few Generally, internal temperatures
dition of the warranty for some vac- instances where the nitrogen supply in liquid-ring vacuum pumps are the
uum-pump manufacturers. was compromised, allowing in air that lowest of most mechanical vacuum
A fine mist of oil drops is emitted can lead to an explosion [3]. For pyro- pumps. Hence, gases exit at almost
with the exhaust gases that are dis- phoric gases, such as silane, special the same temperature, or in some
charged from an oil-sealed pump. Ex- care must be taken to prevent air or cases, at a lower temperature than
haust filters are normally installed oxygen from entering the system. they enter the vacuum pump. Further,
to eliminate this oil mist. The most since the gases handled are in direct
popular design has removable filter Liquid-ring vacuum pumps contact with large amounts of seal-
elements and can be mounted any- Liquid-ring vacuum pumps are widely ant liquid, even if touching of inter-
where in the discharge line. Units are used in the CPI. They are available nal parts (the rotor with the casing)
available that can remove up to 99.9% as single-stage (one or two impellers occurs, the resulting sparks are not
of the exhausted oil mist. Oil mist fil- in parallel) or two-stage (two impel- hot enough or of sufficient duration
ters should not be allowed to become lers in series) units. This type of pump to cause explosions or ignition of even
blocked, as hazardous over-pressures does not require internal lubrication the most hazardous gases. This pro-
could occur. A differential pressure of the impellers, since they do not con- vides a significant safety advantage,
sensor and alarm can be installed on tact the housing. The sealant liquid, especially when handling easily ignit-
oil mist filters to monitor their opera- used for both sealing and cooling, can able gas streams. The occurrence of an
tion and detect potential blockages. be virtually any liquid that is compat- explosion within liquid-ring vacuum
If the oil mist is contaminated with ible with the process gas. For instance, pumps handling flammable vapors is
organic compounds from the process water, low-viscosity oils, glycols and very unlikely as the sealant liquid in
gas, especially if they are toxic, then many process solvents such as tolu- the pump casing acts as a hydraulic
an activated carbon filter should be in- ene, xylene, methanol, ethanol, pro- flame arrester, quenching any flame
stalled after the oil mist filter. The ac- panol, butanol and ethylbenzene have that may be generated and preventing
tivated carbon filter will remove these been used as sealants in liquid-ring it from propagating.
contaminants by adsorption. pumps. Other liquids have also been However, one of the major drawbacks
Pump cooling. Oil-sealed pumps must used, such as ethylene glycol and pro- of liquid-ring vacuum pumps is that
be cooled to remove the absorbed heat pylene glycol, and even sulfuric acid. they must cope with cavitation when
of compression. Both water- and air- The motor horsepower must be cor- running at low inlet pressure. If cavi-
cooled models are available. Air-cooled rected for the sealant viscosity and tation is allowed to continue over long
models should be installed in an area density (see Reference 1, p. 273). These periods of time, serious damage can be
that will allow an ample flow of fresh sealants can be recirculated in a full done to the pump. The amount of cavi-
air to reach the pump. The jackets on recovery system that includes a gas- tation can be affected by the sealant
water-cooled pumps usually contain liquid separator tank and a heat ex- liquid, sealant temperature, impeller
small clearances that can plug with changer to cool the sealant liquid. Even speed, blade angle and inlet pressure.
solids. A strainer or filter should be higher vapor-pressure sealants can For a given pump and sealant liquid,
installed in the cooling water supply be used if a low-temperature coolant cavitation can normally be suppressed
line to remove any solids. A regulating (such as glycol solution) is used in the by bleeding air (or nitrogen) into the
valve should be installed in the cool- heat exchanger to reduce the sealant pump inlet to raise its total pressure
ing-water supply line to control the liquid temperature. This recovery sys- above the vapor pressure of the seal-
pump operating temperature. A high- tem allows process materials to be col- ant at operating temperatures.
temperature cutout switch is some- lected in the pump and either returned All liquid-ring pump units have aux-
times installed to protect the vacuum to the process or collected for disposal, iliary equipment such as a vapor-liq-
pump from overheating. thereby minimizing contamination of uid separator, a sealant recirculation
Movement of oil and flammable other liquids or the environment. pump, a sealant cooler, and various
vapor droplets from oil-sealed vacuum The most advantageous property of instrumentation items, relief devices
pumps has been known to generate a liquid-ring vacuum pumps is the fact and valving.
static charge, which could lead to a fire that they are nearly isothermal. The Some advantages and disadvan-
or explosion [3]. To avoid or minimize liquid in the pump provides an ample tages of liquid-ring pumps are sum-
this potential hazard, it is critical to heat sink for the removal of the heat marized below:
inert the exhaust stream with nitro- of compression. This means that the Advantages:
gen to eliminate oxygen or reduce it to temperature rise in the casing is very Has a simple design that uses only
below the limiting oxygen concentra- small. It is a function of the charac- one rotating assembly
tion (LOC). teristics of the sealant fluid used, its Can be fabricated from any castable
Also, in many cases, it is prudent to temperature, flowrate, and of course, material
use nitrogen to dilute the gas stream the process gas conditions. The result- Creates minimal noise and
upstream of the pump inlet to move it ing low internal temperatures provide vibration
62 Chemical Engineering www.che.com september 2012
Inlet
1 In this sector liquid moves outward,
draws gas from inlet ports
2 In this sector liquid moves inward,
compresses gas in rotor chambers
connections
into rotor chambers
Body
Liquid
Inlet Inlet
port port
Rotating
liquid
compressant
Rotor
Discharge
3 In this sector
compressed gas
ports
Discharge
escapes at discharge ports connections
September 2012
Literature Review
CHE.COM
W
orkers in the chemical process in- Six types of CPC outlined in ISO 16602
dustries (CPI) can wear chemical Type 1
protective clothing to protect them
Type 2
from exposure to potentially hazardous
chemicals in the form of vapor, liquid Type 3
and solid particles. While chemical pro- Gas-tight
tective clothing has not been regulated Type 4
Non protection
to the extant that safety glasses, hard Type 5 gas-tight against
hats and some other forms of personal Protection chemicals
against protection
protective equipment (PPE) have been, Type 6 Protection and vapors
against pressurized and toxic
an international standard is available Protection liquid
liquid particles
for classifying the performance of chemi- against chemicals
Limited aerosols
cal protective equipment. airborne,
protection
against solid
ISO 16602 liquid mist particulate Type indication doesnt tell you everything
ISO 16602 (Protective clothing for protec- chemicals about the performance of the CPC. It only
tion against chemicalsclassification, tells that it meets the minimum require-
labeling and performance requirements) ments of a particular type certification
provides an objective system to appro-
priately test, classify and label chemical- Figure 1. Do not assume that the terms gas-tight and liquid-tight refer to absolute barrier
protective apparel. ISO 16602 designates performance in chemical protective clothing (CPC). They can oversimplify the level of barrier
minimum performance levels of chemical provided by such garments. Instead, these terms should be considered as vapor-protective
protective clothing for six types of chemical and liquid-protective, respectively
hazards. The garment type designation
is based upon the physical state of the haz- Hazard assessment there are also requirements directed
ard, for example vapors, liquids, aerosols A hazard assessment provides a basis for at the mechanical, barrier and basic
or particles (Figure 1). The garment types understanding what chemical protective
are as follows: flammability properties of the fabrics
clothing and other PPE are necessary to and components used to make chemi-
Type 6: Limited protection against protect the workers operating in specific
liquid mist cal protective clothing. Laboratory tests
situations. The environment, chemical haz- are used to determine the mechanical
Type 5: Protection against airborne ard and work activity must be considered
solid particulate chemicals durability, the barrier against specific
in order to be able to select the most ap-
Type 4: Protection against liquid aero- chemical hazards, and ease of ignition
propriate protective clothing including
sols of the garment materials. The results
the fabric, seam type and garment design
Type 3: Protection against pressurized most suitable. The goal is to select cloth- of these tests will fall into a unique
liquid chemicals ing with adequate protection, but without performance class. Each type within
Type 2: Non-gas-tight protection overprotecting a worker with unnecessary ISO 16602 specifies a combination of
Type 1: Gas-tight (vapor-protective) pro- clothing, which can result in heat stress, barrier and durability tests levels, estab-
tection against chemicals and vapors reduced field of vision, restricted mobility lishing a minimum performance class
and toxic particles and increased physical exertion for the for each of the tests to meet the specific
The requirements outlined in ISO 16602 wearer, as well as higher costs for the type requirements. A higher class rating
provide a common language for the per- employer. denotes a higher level of performance
formance of chemical protective cloth- for that property.
ing. ISO 16602 defines performance Performance testing The flammability requirements outlined
When setting the requirements for each in ISO 16602 establish a minimum per-
requirements based on results obtained
type in ISO 16602, the entire garment formance level of flame spread once the
from existing test-method standards. is tested as well as the individual compo-
Although ISO 16602 permits a range material is ignited; it does not qualify
nents. an ensemble as suitable for protection
of performance levels for a series of key Whole-garment tests. During the whole-
properties, it also establishes a minimum against heat and flame hazards. Spe-
garment tests, a human subject wears cific evaluation of chemical protective
level of performance for each major the test garment and accompanying PPE,
type of hazard. clothing for heat and flame protection is
such as gloves, boots and respirator. The
Each workplace environment is not in the purview of ISO 16602, but is
subject is exposed to non-hazardous test
unique, and the ISO 16602 Standard covered within NFPA standards.
chemicals while in an enclosed chamber
does not consider all specific hazards performing a series of movements meant to
simulate actual work activities. This whole- Conclusion
that may be present. Safety and oc-
garment testing is used to validate the Both testing approaches are integral
cupational health professionals should to evaluating the overall integrity and
consider hazards that are specific to barrier performance of the entire ensemble
against a specific type of chemical threat expected performance of chemical protec-
their work conditions and then consult tive clothing. This integrated rating system
the ISO 16602 standard to determine (gas, liquid or particle). Test chemicals are
used to determine how much of a similar aids the selection of the most appropriate
the minimum requirements of chemical chemical protective apparel for specific
protective garments used in the particu- phase chemical will leak into the suit.
Class tests. Whole-garment tests do not work tasks. Testing the whole garment
lar situation. and the individual components separately
evaluate the chemical permeation proper-
ISO 16602 focuses exclusively on ensures the sum of the parts is able to
ties of the garments. This is assessed in
typical chemical hazards and protective the class testing portion of ISO 16602. provide the appropriate protection.
clothing requirements for those hazards, Beyond the whole-garment tests, additional
so each work situation will likely require tests are conducted on the garments Editor's note: This Facts at your Fingertips
additional forms of PPE, possibly includ- fabric(s) and components to qualify the column was adapted from Lovasic, S., Chemi-
ing footwear, gloves, face protection, class performance level. cal Protective Clothing, Chem. Eng., March
fall protection and respirators. 2011, pp. 51-53.
Within each of the six garment types,
Static Electricity
Discharge and Fire
Department Editor: Scott Jenkins Prevention
W
hen flammable or combus- insulating, so as to to possibly prevent Table 1. Typical MIE values
tible atmospheres are present, the proper dissipation of static charge. Material Minimum
uncontrolled discharges of static (gas/vapor or pow- Ignition Energy
electricity can be potentially dangerous Static discharge sources der/dust) (MIE), mJ
or even catastrophic. A significant por- Static discharges come in several forms, Carbon disulfide 0.009
tion of industrial explosions and fires are the most important for CPI interests being Methanol 0.14
attributable to static electricity each year. spark and brush discharges. A spark Xylene 0.20
In theory, controlling static electricity by is a discharge from a charged isolated Toluene 0.24
grounding potential sources is simple, but conductor to another conductor at lower Propane 0.25
in practice, doing so effectively requires potential. A brush discharge occurs from
Ethyl acetate 0.46
thorough knowledge of processes and an electrostatically charged insulator to a
Zirconium 5.00
operations, sound engineering controls, grounded conductor.
Epoxy resin 9.00
properly specified safety equipment and Typical possible sources of static elec-
Aluminum 10.00
properly trained operational staff. This trical discharges include the following:
Sugar 30.00
column provides information on potential Spark discharges from any conductive
Wheat flour 50.00
sources of static electrical discharge. but not earthed (grounded) bag, bin,
drum, container and so on, from which Note 1: Minimum ignition energy (MIE) is defined
as the minimum energy that can ignite a mixture
Static risk the powder is transferred into the reac- of a specified flammable material with air or oxy-
Static electricity discharges are possible tor gen, measured by a standard procedure.
almost continuously in the CPI, because Brush discharges from any non-conduc- Note 2: MIE values are provided for guidance
static electricity is generated whenever tive bag, bin, drum, container, and so only specific MIE data for any material
should be verified. Source: NFPA, IchemE
surfaces come into contact and then on, from which the powder is trans-
separate. In most cases, the charge ferred into the reactor been formulated with some added semi-
generation in industrial processes is Spark discharges from any conductive conductive additives. Bonding means
small typically no greater than 1 but not earthed auxiliary device used linking these objects together by means
104 Amp. However, even small charges in the transfer procedure, including, but of a suitably strong conductor (wire), and
can be a problem in hazardous areas not limited to, shovels, funnels, chutes grounding refers to a true ground/
when allowed to accumulate on objects and pipes earth connection that is applied to one
that are not at ground (earth) potential. Spark discharges from the operator, if or more of the bonded objects.
If no ground is present, voltages in he is not adequately earthed When one or both of these techniques
excess of 30 kV can develop. Depend- Brush discharges from any non-conduc- is applied, and while a low resistance
ing on the capacitance of the object, tive auxiliary devices, such as shovels, connection between the objects and
this may result in significant levels of funnels, chutes and pipes ground is maintained, operators are
energy being available for discharge. Brush discharge from the dust cloud able to prevent dangerous levels of static
If the energy equals or exceeds the formed within the reactor during pow- charge from accumulating.
minimum ignition energy (MIE) of the der transfer In the case of fixed installations such
surrounding flammable atmosphere, the Spark discharges from any conductive as pipe work, storage tanks and so forth,
potential for an explosion and fire exist. but not earthed fixtures and fittings this is relatively simple to implement.
Many commonly used solvents and other within the reactor However, these preventive measures
flammable chemicals have MIEs that are Brush discharges from the charged sol- are more difficult to implement with
relatively low on the order of 1 mJ or vent, suspension or emulsion preloaded mobile/portable objects, such as drums,
less (Table 1). in the reactor IBCs and tankers. In these instances,
Brush discharges from the powder purpose-designed temporary ground-
Isolated conductors heap formed on top of the liquid phase ing and bonding devices must be used,
Isolated conductors are conductive within the reactor with strict procedures to ensure that they
objects that are either inherently or ac- Cone discharges from the powder heap are always in place prior to starting the
cidentally insulated from earth. During formed on top of the liquid phase process. For instance, specific types of
day-to-day operations at industrial facili- Liquids flowing through pipelines or clamps and devices for grounding and
ties, isolated conductors are probably filling into drums and tanks bonding portable or mobile plant equip-
the most likely source of static ignition Persons walking across an insulating ment, drums and containers are recom-
incidents. floor mended in NFPA 77, and such ground-
The insulation effectively keeps any ing clamps and devices generally should
static electricity buildup from safely Static discharge prevention employ sharp contact points. These
discharging, thereby resulting in ac- Where recommendations tend to contact points should be made of a wear-
cumulation of charge on the object. If converge is in the recommendation to al- resistant material, have positive spring
the isolated conductor then comes into ways use conductive or static dissipative pressure, and be universally adaptable to
proximity with another object at a lower materials, and to ensure effective bond- a wide range of plant objects.
potential, energy could be released in ing and grounding. For information on
the form of an incendive spark. grounding best practices and examples References
Isolated conductors may arise from of preventing static electrical discharge,
1. Tyers, G., Avoiding Static Sparks in Hazardous
metal flanges, fittings or valves in pipe- consult National Fire Protection Associa-
Atmospheres, Chem. Eng., June 2009, pp.
work systems; portable drums, containers tion (NFPA; Quincy, Mass.; www.nfpa. 4449.
or vessels; tanker trucks, railcars and org) standards 77 and 30.
2. Glor, M., Preventing Explosions During the
intermediate bulk containers (IBCs); and In this context, the term conductive Transfer of Solids into Flammable Solvents,
even people. would apply to metal materials, such Chem. Eng., October 2007, pp. 8895.
Many modern industrial paints, as stainless or carbon steel, aluminum
coatings, gaskets, seals and other and others, and static-dissipative may Editors note. This edition of "Facts at your Finger-
non-conductive materials are sufficiently indicate rubber or plastics that have 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
Zone 2
Vent
Outdoors Zone 0
Gasoline
storage
10 ft (3m) tank 10 ft (3m) Gasoline
without storage
floating tank
roof without
Circular dike floating
around tank roof
Class 1, division 2
FIGURE 1. Shown here is a typical example of a Class I hazardous area uti- FIGURE 2. The example hazardous area shown in Figure 1 is here classified ac-
lizing division methods of area classification cording to the zones
Robert Schosker stances, where there exists the principles for working or gathering
Pepperl+Fuchs risk of explosion or fire that can be information out of a hazardous area
E
caused by an electrical spark or hot remain the same. Its just that today
xplosion protection is essen- surface, requires specifically defined we have multiple options. In order to
tial for many companies, and instrumentation located in a hazard- exercise those options, we must first
those companies have deci- ous location. It also requires that determine if the danger of an explo-
sion makers. But before any the interfacing signals coming from sion exists and how severe it may be.
decisions can be made, there are a hazardous location are unable to
some important factors one must create the necessary conditions to What is a hazardous area?
consider. These factors include what ignite and propagate an explosion. Hazardous areas are most frequently
is most efficient and economical, as This risk of explosion or fire has found in places where there is a pos-
well as knowing the basics of explo- been the limiting factor when using sibility of an emission of flammable
sion protection; so the decision mak- electrical instrumentation because gas or dust. A hazardous area can
ers are headed in the right direction. energy levels were such that the occur in normal operation, in the
We will highlight many of the different energy limitation to the hazardous event of a fault, or due to wear and
things to know, but first, lets step location was difficult, if not impossi- tear of seals or other components.
back in time and take a look at the ble, to obtain. For this reason, those Now the risk of an ignition of an
background of explosion protection. parts of the process that were con- air/gas mixture in this hazardous
sidered risky were controlled with area depends on the probability of
Backdrop pneumatic instrumentation. the simultaneous presence of the
After World War II, the increased following two conditions:
use of petroleum and its derivatives Moving forward Formation of flammable or explo-
brought the construction of a great Now lets move forward 70 years, sive vapors, liquids or gases, or
number of plants for extraction, refin- where almost everything you can combustible dusts or fibers with
ing and transformation of the chemi- think of can be found at the touch of a atmosphere or accumulation of
cal substances needed for techno- finger. From pneumatics to quad core explosive or flammable material
logical and industrial development. processors, information gathering Presence of an energy source (elec-
The treatment of dangerous sub- has definitely changed, but the same trical spark, arc or surface temper-
50 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM JULY 2015
TABLE 1. DEFINING AREAS FOR DIVISIONS TABLE 2. THE BREAKDOWN OF CLASSES INTO SUBGROUPS
Class Type of Material Class Subgroup Atmospheres
Class I Locations containing flammable Class I Group A Atmospheres containing acetylene
gases, flammable liquid-produced va- Group B Atmospheres containing hydrogen and flammable process gases with
pors, or combustible liquid-produced more than 30 vol.% H2, or gases or vapors posing a similar risk level,
vapors such as butadiene and ethylene oxide
Class II Locations containing combustible Group C Atmospheres such as ether, ethylene or gases or vapors posing a similar
dusts risk
Class III Locations containing fibers and Group D Atmospheres such as acetone, ammonia, benzene, butane, cyclopropane,
flyings ethanol, gasoline, hexane, methanol, methane, natural gas, naphtha, pro-
pane or gases or vapors posing a similar threat
ature) that is capable of igniting the Class II Group E Atmospheres containing combustible metal dusts, including aluminum,
explosive atmosphere present magnesium, and their commercial alloys, or other combustible dusts
Determining hazardous areas in whose particle size, abrasiveness and conductivity present similar haz-
ards in the use of electronic equipment
a plant is normally performed by
experts from various disciplines. It Group F Atmospheres containing combustible carbonaceous dusts, including
carbon black, charcoal, coal or coke dusts that have more than 8% total
may be necessary for chemists, pro- entrapped volatiles, or dusts that have been sensitized by other materials
cess technologists, and mechani- so that they present an explosion hazard
cal engineers to cooperate with an Group G Atmospheres containing combustible dusts not included in Group E or
explosion-protection expert in order Group F, including flour, grain, wood, plastic and chemicals
to evaluate all hazards. The possible
presence of a potentially explosive TABLE 3. THE DIVISION METHOD
atmosphere as well as its properties Division Class I Class II Class III
and the duration of its occurrence
(gases and vapors) (flammable dust or powder) (flammable fibers or sus-
must be established. pended particles)
Also understanding terms such as In accordance with NEC In accordance with NEC
minimum ignition energy (MIE), upper 500.5 and CEC J18-004 500.6 and CEC 18-008 In accordance with NEC
and lower explosive limit (UEL/LEL), 500.5 and CEC 18-010
flash point, and ignition temperature Division 1 Areas containing dangerous Areas containing dangerous Areas containing dangerous
in the evaluation of your hazardous concentrations of flam- concentrations of flammable concentrations of flammable
mable gases, vapors or mist dusts continuously or oc- fibers or suspended particles
area will also provide a clearer direc- continuously or occasionally casionally under normal continuously or occasion-
tion on how severe a hazardous area under normal operating operating conditions ally under normal operating
might be. conditions conditions
In any situation involving an ex- Division 2 Areas probably not contain- Areas probably not contain- Areas probably not contain-
plosive material, the risk of ignition ing dangerous concentra- ing dangerous concentra- ing dangerous concentra-
must be taken into account. In addi- tions of flammable gases, tions of flammable dusts tions of flammable fibers or
tion to the nominal rating of materi- vapors or mist under normal under normal operating suspended particles under
operating conditions conditions normal operating conditions
als under consideration, parameters
related to the process involved are and utilized worldwide. may exist due to flammable gases,
especially important in the evalua- In rather simple terms, we can dif- vapors, or liquids within a plant or
tion. For example, the risk of explo- ferentiate between the International working environment. These are ex-
sion may be caused by the evapora- Electrotechnical Commission (IEC; plained by defining the type of haz-
tion of a liquid or by the presence of Geneva, Switzerland) (zones) and ardous material present, severity of
liquid sprayed under high pressure. the North American (division) proce- the hazard, and probability of the
It is also important to know which dures. The differences lie in the cat- hazard. It may also depend on the
atmospheric conditions are present egorization of hazardous areas, the likelihood of the hazard, risk of an
normally and abnormally. The range design of apparatus, and the installa- explosion, and the boundaries of the
of concentration between the ex- tion technology of electrical systems. hazardous location.
plosion limits generally increases as The categorization of these areas is This is usually determined by a
the pressure and temperature of the carried out in North America in ac- HAZOP (hazard and operability)
mixture increases. cordance with the National Electrical study and documented on a set
Code (NEC) NFPA 70, article 500. of electrical plot plans on record in
Divisions and zones The European Zone practice is de- every plant.
Once it has been determined that a scribed in IEC/EN 60079-10. For divisions, the type of material
hazardous area exists, it now needs So how does each work? First lets is given by a class designation, as
to be classified. While the physical start at the basics, and then well shown in Table 1. These can be bro-
principles of explosion protection cover each individually. ken down further into sub-groups,
are the same worldwide and are as shown in Table 2.
not differentiated, there are two dif- Defining the area Once we have determined the haz-
ferent and distinct models to define Hazardous location or area clas- ardous material we are working with,
your hazardous area divisions and sification methods specify the dan- the probability of an explosion and
zones both of which are accepted ger of fire or explosion hazards that boundaries must also be taken in to
Io L R
Interstice
U
C
Length of
S Uo
junction
P
FIGURE 3. Explosion-proof protection is based on FIGURE 4. In purging or pressurization protec- FIGURE 5. Intrinsic safety is based on the prin-
the explosion-containment concept, whereby the tion, a dangerous air/gas mixtures is not allowd ciple of preventing an effective source of ignition
enclosure is built to resist the excess pressure to penetrate the enclosure containing the electri-
created by an internal explosion cal parts that can generate sparks or dangerous
temperatures
ated, T1 to T6. The classes T2, T3
and T4 are divided into further sub-
consideration. The division method better rationalized by looking at the classes, as indicated in Table 7.
is divided into two areas: Division 1 example shown in Figure 2. In Europe, the apparatus are
and Division 2 (Table 3). These were With a slightly different approach, certified on the basis of design
created in 1947 when the NEC first IEC 600079-0 requires apparatus to and construction characteristics.
recognized that different levels of risk be subdivided into two groups, as From a practical point of view, the
exist in hazardous locations. Figure 1 shown in Table 5. two systems are equivalent, even
shows a typical example of a Class The groups indicate the types if there are minor differences, but
I hazardous area utilizing Division of danger for which the apparatus before you run out and choose the
methods of area classification. has been designed. Group I is in- most convenient method for you, it
In comparison to the division- tended for mines. Group II concerns is important that you consult your
based area classification, which is above-ground industries (electrical local authority having jurisdiction to
prevalent throughout North America, apparatus for hazardous areas with learn what method is allowed or, in
the zone-based architecture prevails potentially explosive gas (dust) at- fact, preferred.
in the rest of the world. mosphere except firedamp hazard- The initial steps to determine
Zones are similar in nature to divi- ous mining areas) and is subdivided whether a hazardous area exists and
sions where type of hazardous ma- into II G (gases) and II D (dusts). classify that area may seem rudimen-
terial present, severity of the hazard, Similar to divisions, the zones tary to some, but they are important
and probability of the hazard and offer a sub material classification as they now open up the multiple
boundaries must be determined. as well. Table 6 shows how this methods of protection, which may
Zones are in accordance with IEC/ approach compares to the North or may not be allowed, depending
EN 60079-10, which states that any American equivalent. on whether you classified your area
area in which there is a probability of Finally, when classifying your haz- by divisions or zones.
a flammable gas or dispersed dust ardous area, whether it be division
must be classified into one of the or zones, you must also classify the Protection methods
areas shown in Table 4. maximum surface temperature that There are three basic methods of
Similar to the division method of can go in to the hazardous area. The protection explosion containment,
area classification, zones can be maximum surface temperature must segregation and prevention.
be below the minimum ignition tem- Explosion containment. This is the
TABLE 4. DEFINING AREAS BY ZONES
perature of the gas/dust present. only method that allows the explosion
Zone Type of material In North America as in Europe, six to occur, but confines it to a well-de-
Zone 0 An area in which an explosive air/gas temperature classes are differenti- fined area, thus avoiding the propaga-
mixture is continuously present or tion to the surrounding atmosphere.
present for long periods of time
TABLE 5. APPARATUS GROUPS PER IEC Flameproof and explosion-proof en-
Zone 1 An area in which an explosive air/gas 600079-0 closures are based on this method.
mixture is likely to occur in normal
operation Group Apparatus Segregation. This method attempts
Group I Apparatus to be used in mines where to physically separate or isolate the
Zone 2 An area in which an explosive air/gas
mixture is unlikely to occur; but if it the danger is represented by methane electrical parts or hot surfaces from
does, only for short periods of time gas and coal dust the explosive mixture. This method
Zone 20 An area in which a combustible dust Group II Apparatus to be used in surface includes various techniques, such
cloud is part of the air permanently, industries where the danger is rep- as pressurization, encapsulation,
over long periods of time or frequently resented by gas and vapor that has and so on.
been subdivided into three groups:
Zone 21 An area in which a combustible dust A, B and C. These subdivisions are
Prevention. Prevention limits the
cloud in air is likely to occur in normal based on the maximum experimental energy, both electrical and thermal,
operation safe gap (MESG) for an explosion- to safe levels under both normal op-
Zone 22 An area in which a combustible dust proof enclosure or the minimum igni- eration and fault conditions. Intrinsic
cloud in air may occur briefly or dur- tion current (MIC) for intrinsically safe safety is the most representative
ing abnormal operation electrical apparatus
technique of this method.
52 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM JULY 2015
TABLE 6. SUB MATERIAL CLASSIFICATION FOR ZONES TABLE 7. TEMPERATURE CLASSES
Material Apparatus clasification Apparatus classification Ignition energy Tmax, C Tmax, F T Class in N.A.*
450 842 T1
Europe (*IEC) North America
300 572 T2
Methane Group I (mining) Class I, Group D
280 536 T2A
Acetylene Group IIC Class I, Group A > 20 J
260 500 T2B
Hydrogen Group IIC Class I, Group B > 20 J
230 446 T2C
Ethylene Group IIB Class I, Group C > 60 J
215 419 T2D
Propane Group IIA Class I, Group D > 180 J
200 392 T3
Conductive dust (metal) Group IIIC* Class II, Group E
180 356 T3A
Non-conductive dust Group IIIB* Class II, Group F
(carbon) 165 329 T3B
Cereal/flour Group IIIB* Class II, Group G 160 320 T3C
Fibers/suspended Group IIIA* Class III 135 275 T4
particles 120 248 T4A
100 212 T5
My application requirements fication, application, and cost in
85 185 T6
Now the questions really start racing helping users find the ideal solution
in: Which should I use? Which one to match their needs. *N.A. = North America
offers the best protection? What if In addition to considering the nor-
all of my equipment is not low pow- mal functioning of the apparatus, area classification, there are many
ered? My plant is already using a eventual malfunctioning of the appa- other options when an area is classi-
technique; can I use another pro- ratus due to faulty components must fied using zones, but for now we will
tection method? Can they co-exist? be a consideration. And finally, all concentrate on the above as they
Who makes that decision? Why those conditions that can acciden- are most commonly used.
should I use one method over the tally occur, such as a short circuit, XP. The explosion-proof protection
other? Can I use two methods at open circuit, grounding and errone- method is the only one based on the
the same time? So many questions, ous wiring of the connecting cables, explosion-containment concept. In
all of which are very important, and must be evaluated. The choice of a this case, the energy source is permit-
with a little understanding of your specific protection method depends ted to come in contact with the dan-
process, they will guide you to best on the degree of safety needed for gerous air/gas mixture. Consequently,
method(s) to use. the type of hazardous location con- the explosion is allowed to take place,
Hazardous-area protection method sidered in such a way as to have the but it must remain confined in an en-
selection depends on three impor- lowest probable degree of an even- closure built to resist the excess pres-
tant factors: (1) area classification, (2) tual simultaneous presence of an sure created by an internal explosion,
the application and (3) the cost of the adequate energy source and a dan- thus impeding the propagation to the
protection method solution. gerous concentration level of an air/ surrounding atmosphere.
Area. Area classification depends on gas mixture. The theory supporting this method
the type of hazardous substances None of the protection methods is that the resultant gas jet coming
used, operating temperature, and can provide absolute certainty of from the enclosure is cooled rap-
explosion risk due to how often the preventing an explosion. Statistically, idly through the enclosures heat
dangerous substance is present in the probabilities are so low that not conduction and the expansion and
the atmosphere and the boundary even one incident of an explosion dilution of the hot gas in the colder
of the substance from various parts has been verified when a standard- external atmosphere. This is only
of the process. Area classification ized protection method has been possible if the enclosure openings
is determined by either the division properly installed and maintained. or interstices have sufficiently small
method or zone method. The first precaution is to avoid dimensions (Figure 3).
Application. Application character- placing electrical apparatus in haz- In North America, a flameproof en-
istics also affect which protection ardous locations. When designing closure (in accordance with IEC) is, as
method is used. For example, some a plant or factory, this factor needs a rule, equated with the "flameproof"
methods are more appropriate for to be considered. Only when there is designation. In both considerations,
large equipment protection, while no alternative should this application the housing must be designed for
others are more appropriate for high- be allowed. a x1.5 explosion overpressure. The
power applications. North American version Explosion
Cost. Cost is also an important fac- Choosing the best method proof (XP) must withstand a maxi-
tor for many engineers. For example, After carefully considering the above, mum explosion overpressure of x4.
if their application requires Division we can look at three more popular Furthermore, in North America,
2 protection, they may not want to methods of protection, XP (explo- the installation regulations (NEC 500)
purchase more expensive equipment sion proof/flameproof), purging and specify the use of metal conduit for
rated for Division 1. For that reason, pressurization, and intrinsic safety. the field wiring installation. It is also
it is important to understand the in- Although these are the most com- assumed here that the air-gas mixture
terplay of all three factors classi- monly used methods in the division can also be present within the con-
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
FIGURE 1. The concept and components of closed-loop
Vienna University of Technology control are shown here
F
uel combustion in engines, process itself, as well as fuel trans- sends a signal to an actuator for re-
power plants, boilers, furnaces portation and storage, and the use of action (Figure 1). Closed-loop con-
and other equipment provides its outcome (for instance, in cogene- trol can steer a process better than
energy for transportation, heat- ration, waste heat is being utilized). one using an open-loop system [3
ing, electricity and goods manufac- Concerning pollutants, measures to 4]. This article focuses on advanced
turing. Optimizing efficiency while limit or reduce their impact can be sensors and decision-maker algo-
lowering pollutant emissions is the taken pre-combustion, during com- rithms, because these control-sys-
main goal of industrial combustion, bustion and post-combustion. Prom- tem elements have seen explosive
and reaching those goals, coupled inent examples of these strategies development in the past two de-
with climate-change issues, have include sulfur-free diesel fuel, air or cades, whereas valves and actua-
triggered a great deal of research in fuel staging and selective catalytic tors have changed comparatively
this field. One approach to improv- reduction (SCR) of NOx. little over that time. For more in-
ing the efficiency and emissions Energy-efficiency and emissions formation on the fundamentals of
performance of a combustor is to strategies depend on the ability to process control, see Refs. 56.
apply advanced control techniques control combustion processes. Clas- Advanced combustion control can
for both new and existing installa- sic combustion controls are based on be applied to virtually all types of
tions. This article presents several temperature (for example, adjust- combustors, including the following:
methods of advanced control for ment of the air-to-fuel ratio by peak gas turbines, furnaces, boilers and
combustion, ranging from new diag- exhaust gas temperature), flame reciprocating internal-combustion
nostic techniques to enhanced con- emissions (such as the detection of engines, to name a few. Often, it is
trol schemes. Key benefits are cost misfiring in stationary engines by possible to retrofit existing instal-
savings through higher efficiency, ultraviolet light detection), mea- lations so that significant economic
environmental benefits through surement of in-cylinder pressure in and environmental benefits can be
lower levels of pollutants, and in- engines and measurement of CO, achieved [7].
creased safety. CO2 or O2 in the exhaust gases.
Advanced combustion-control Modern combustion options
Classic versus advanced control strategies generally represent Based on the huge variety of com-
Combustion proccesses account for extensions and improvements to bustion applications, modern com-
85% of global primary energy pro- classical control methods. An ad- bustion processes differ signifi-
duction [12], from electricity and vanced control system has a closed cantly from each other. Distinctions
heat generation to propulsion in feedback loop (open-loop control can be made among combustion
terrestrial, marine and aerial trans- systems do not have feedback). In processes of gaseous, liquid and
portation. Measures to increase ef- combustion control, a sensor deliv- solid fuels. The two major modes of
ficiency encompass the combustion ers data to a decision maker, which combustion are those with the fuel
34 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-
FIGURE 2. Oxygen concentration can be measured by tunable diode-laser absorp- rosion. TDLS has become a mature
tion spectroscopy [10]
technology in advanced combustion
and oxidizer premixed and those mensional, spatially resolved or in- control. A related approach, infrared
where they are not premixed. There tegrated signal) and cost. absorption tomography for active
are also differences between lami- Optical, and particularly laser- combustion control, is described in
nar and turbulent combustion. For based methods, lend themselves Ref. 12.
technical applications, turbulent to in-situ measurements. A major Advanced sensors based on light
combustion dominates, because it advantage of optical techniques is emission. A technique based on
speeds up the process. However, that they do not perturb the probed light emission that combustion re-
non-premixed combustion is some- system, hence avoiding measure- searchers frequently use is laser-
times preferred for safety reasons. ment errors from sampling. Light induced fluorescence spectroscopy
The following combustion processes can interact with matter in three (LIF). It can be deployed to obtain
can be termed advanced: basic modes: absorption, emission two-dimensional images (planar-
Lean combustion (to achieve low- and scattering. LIF, P-LIF) of species concentra-
NOx emissions) Advanced light-absorption sen- tions, including radical species such
Carbon capture and storage sors. The technique of tunable as OH. However, the setup is com-
(sequestration of CO2 from diode-laser absorption spectros- plex, so this technique has not yet
combustion) copy (TDLS) can measure species found industrial use for combus-
Chemical looping combustion concentrations and temperatures tion-control purposes. The test of a
Oxyfuel combustion of simple gases, such as CO, CO2, closed-loop equivalence ratio con-
Co-combustion of fossil fuels with O2, NH3 and CH4, quantitatively trol of premixed combustors using
biomass by selective light absorption in the a chemiluminescence signal is de-
Combustion of alternative fossil infrared spectral region by the tar- scribed in Ref. 13.
fuels, such as methane hydrate get molecules. By tuning the laser Advanced sensors based on
HCCI (homogeneous charge com- wavelength around the respective light scattering. Laboratory-
pression ignition) [8] absorption feature, users can com- based methods that rely on scat-
Advanced combustion is also dis- pensate for non-specific effects, tering are Raman scattering and
cussed in Refs. 2 and 9. such as beam steering, background Cars (coherent anti-Stokes Raman
radiation and partial beam blockage scattering) spectroscopy. Like LIF,
Advanced sensors by soot particles. A typical setup in due to their complex experimental
Combustion processes are difficult the stack of a combustion process is setup and difficult data evaluation
to probe, since high temperatures depicted in Figure 2. It consists of a and interpretation, they are not
are involved. Often, conditions are transmitter and a receiver unit. yet used for industrial combustion
transient, high pressures are pres- In an exhaust stack, this type of control applications.
ent and multiple interferences, such sensor can measure NH3, for ex- Advanced sensors based on other
as soot particles and background ample, so it can minimize the use of signals. In Ref. 14, an advanced
radiation, disturb measurements. urea supplied to the NOx removal closed loop combustion sensor based
Parameters to be measured include unit further upstream. Other exam- on ionization is described. Ref. 15
temperature, pressure and species ples are given for a municipal waste uses pressure signals, and Ref. 16
concentrations. Generally, one can incineration plant (Figure 3). In this image-based controls for a closed-
distinguish in-situ measurements example, oxygen and temperature loop setup. A review of sensors for
(which probe the combustion pro- are measured in the combustion combustion control is provided else-
cesses directly at the spot) and chamber. The measurement range where [17].
ex-situ measurements, which draw for O2 is 015% (accuracy 0.2%),
a sample that is analyzed in a con- and that for the temperature 750 Advanced decision makers
trolled manner outside of the pro- 1,100 C (accuracy 30C). The re- The term decision maker refers
cess (for example, in a gas chroma- sponse time is 13 s. The signal can to the computer hardware and soft-
tography instrument). The criteria be used to control the feed of fresh ware necessary to run the control
for selecting suitable measurement air. This is important because the algorithms that govern the adjust-
methods are time resolution, sensi- composition of the fuel varies a great ment of the combustion system pa-
tivity, type of measurement (such deal and the measurement can help rameters. The hardware basis of the
as point-measurement or two-di- achieve optimal combustion. central, decision-making part of ad-
CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014 35
Feature Report
A
dvanced control techniques offer opportunities for effectively ous limitation of traditional control may not necessarily be present,
handling combustion systems that have strong nonlinearities but where advanced control methods would assure significant in-
or internal couplings, and that therefore cannot be satisfac- creases in efficiency and flexibility, or decreases in lifetime con-
torily handled by the commonly applied traditional PID (propor- sumptions of critical parts.
tional, integral derivative) controller, which is characterized by For collecting the answers to this question, and for formulating
linear and scalar behaviors. Published data are available on the some advice on overcoming most blocking factors, a workshop
shares of advanced control methods for several years up to 2008, took place in May 2012 at the Budapest University of Technology
which clearly show that the application of advanced control meth- and Economics. A group of experts in three categories plant
ods for power plants and industry boilers are far below those in operators, equipment suppliers and researchers came together
the chemical, oil-and-gas and petroleum refining industries, for from four countries to discuss the issue in an unstructured, inter-
example. In 2008, the power industry spent the most on DCS (dis- active format. From the workshop, a number of basic statements
tributed control system) upgrades and replacements, while it spent emerged. They are summarized here.
the least on advanced process control (see Table 1 in Ref. 18). An important reason for the low level of advanced control meth-
What reasons could explain these data? Instead of exact an- ods in combustion processes in the energy industry is the presence
swers, some differences will be identified and discussed here. It of gaps between several actors. A lack of communication exists
may be that combustion processes do not dictate so definitely the among people and institutions working on the same topic. Key
applications of advanced controllers, because some other chemi- people involved working on industrial combustion projects in the
cal processes are characterized by much stronger nonlinearities. areas of control, energy, management and financials are often
But in reality, combustion processes are also nonlinear, especially unable to balance benefits and drawbacks from the perspective of
when considering their entire load ranges. It is true that these non- each of their separate fields. Several platforms for bringing people
linearities are less significant, and they predominantly can still be together could help the situation, as could the use of a common
kept in operation by means of traditional linear tools. However, language. This common language could be money. However, ex-
industrial practice shows clearly that this approach leads to ef- pressing certain benefits (such as safety, usability and lifetime re-
ficiency reductions already in the simplest case of a power plant duction) in the context of money is not straightforward. Education
boiler running below its base load. could also help in connecting control theory with system theory,
university with industry, and practice with theory.
TABLE 1. RATIOS OF WORLDWIDE SALES OF APC Safety is another issue that slows down the utilization of advanced
RELATIVE TO WORLDWIDE SALES OF DCS* control in combustion processes. Safety is the highest priority,
Industry Refining Oil and Chemical Power however, it can be warranted not only by traditional controllers
gas but also by means of a proper structure in the control system. Tra-
Ratio of APC 0.092 0.061 0.050 0.010 ditional controllers should be used on a higher, supervising level,
sales over and very clear interfaces should be built with the advanced control
DCS sales elements [21]. Consider that this approach will also assure further
benefits. Additional certification costs for safety-related control
* Ratios of worldwide sales of APC (advanced process control) relative
to worldwide sales of DCS (distributed control systems) in several in- elements can be avoided, as can some of the risk arising from
dustries in 2008. Adapted from Ref. 18. frequent version changes of the hardware and software platforms
running the advanced control computation methods.
In some cases of combustion technologies, the suboptimal efficiency Software elements of advanced decision makers must follow the
associated with traditional control methods is no longer acceptable, same safety-related requirements as their hardware components,
or the traditional control methods simply do not work. In these cases, as discussed above. Thats why, also the above discussed consid-
the application of advanced controllers is dictated already. Many erations in applying them should be followed in this case es-
such examples were reported together with success stories of solving pecially regarding the structure and clear interfaces joining tradi-
them [2124]. They include a wide range of combustion processes, tional and advanced elements. The same idea also suggests that
from compensating for unacceptable fluctuations of biomass boilers the range of control tasks to be realized on the basis of advanced
with soft sensors, to lambda control of internal combustion engines methods must be selected carefully. Positioners of actuators are, for
by means of a robust predictive controller. example, advised to be kept unchanged for simplicity and reliabil-
A further question is, of course, what factors block the dissemina- ity. Also, simple tasks of low significance should not be involved in
tion of advanced control methods in those cases in which an obvi- the area of tasks to be realized by advanced control methods.
more, MPC is also able to consider trol and lambda control in a boiler. Intelligent control methods.
other constraints without direct re- The basic MPC algorithm itself With respect to control methods,
lationships to the process in ques- is a discrete time on-line optimi- the term intelligent suggests
tion. If the control system is well de- zation, which requires a dynamic those with origins in artificial intel-
signed, this ability can be perfectly process model. The more accurate ligence research. Two basic control
utilized for limiting temperature the model, the better the control methods fall into this category: ar-
stresses in key structural elements performance that can be expected. tificial neural networks and fuzzy
of a combustion system during load However, model inaccuracies will be control. Both are excellent tools for
shifts, for example. effectively compensated by the con- controller design in cases where the
Another advantage of MPC troller. Prior to the controller design, formulation of control-system rules
against the traditional PID is that the constraints must be given, along on the basis of traditional mathe-
is it multidimensional. As a result, with two further parameters, the so- matical formalism is very difficult.
internal cross-couplings of the com- called weight matrices. They form a In many situations, the control
bustion process can be handled very clear representation of the de- rule is available by verbal state-
inherently. There is no need for sign criteria for balancing between ments only, which arise from human
figuring out the significance of the two contradictory requirements of thinking. This set of control rules can
cross-couplings. An example is the all control tasks accuracy and be called expert knowledge, and a
cross-couplings between load con- quiet actuator movements. very effective way of representing
CHEMICAL ENGINEERING WWW.CHE.COM AUGUST 2014 37
Feature Report
FIGURE 4. Conventional
it is a new direction of mathemati- control (left) and advanced
cal set theory called fuzzy logic. control (right) of a power
Artificial neural networks are very plant. The latter allows a
higher average boiler operat-
effective when the learning capa- ing temperature, which trans-
bilities and methods of the human lates into efficiency gains [21]
brain need to be applied on the basis
of corresponding input and output crete groups (sets) that correspond possibility of using it in low-cost ap-
data sets of a system. Although sev- to intermediate, high, and very plications. Limitations of fuzzy con-
eral similarities exist between these high. An important point is that trollers arise from the basic princi-
two intelligent control methods, and the borders of these groups (or sets) ple, which may result in unsmooth
their combined application is a pop- are not crisp, like they would be in outputs. Clear stability criteria for
ular solution, they will be discussed classical set theory. This grouping those are still lacking.
further separately. is much closer to human thinking, The basic concept behind artifi-
The central element of a fuzzy which generally does not draw a cial neural networks (ANN) was
controller is a rule base, containing crisp borderline between the sets inspired by the human nervous
rules formulated on an if ... then ... distinguishing the age of a person system, the basic element of which
logic scheme. (Example: if the mix- to sets like youth or middle-aged, is the neuron. Like its biological
ture temperature is very high, then for example. Defuzzification is a counterpart, an artificial neuron
set coolant valve to totally open.) procedure in the opposite direction. is a multi-input, single-output ele-
The previous fuzzy control element The characteristics of a fuzzy con- ment, which is organized into net-
is called fuzzification, while the troller make it effective for control- works that give rise to multi-input,
following one is defuzzification. ler design. The simple realizability multi-output (MIMO) systems. The
In the prior one, analog quantities of each element of the fuzzy con- strengths between the neuron in-
(such as temperature) will be fuzzi- troller is another advantage. This terconnections can be varied, which
fied; that is, ordered into the dis- simplicity of operation allows the establishes the basis for the ability
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- refineries, either partial power failure planned shutdown of the affected
ties has typically resulted in an array or cooling water failure is typically the units; whereas emergency flare
of independent flare systems built controlling scenario. maintenance would require either an
to meet the requirements of specific Such events are generally local- unplanned shutdown or turndown of
expansion projects. ized to a particular area of the plant. a portion of the plant. In both cases,
Flare consolidation is one attrac- For instance, if a chemical facility production would likely be curtailed
tive solution that is based on a ho- has three independent cooling-water and would thereby result in higher
listic approach to updating the exist- systems, the loss of any one system overall costs.
ing infrastructure to meet growing should not result in releases through- The ability to perform emergency
demands. Integrating separate flare out the entire facility. or scheduled maintenance on the
systems into a single flare network Thus, the availability of a plant-wide flare system with minimal impact on
can be an economically and envi- flare system can provide increased operations could lead to a signifi-
ronmentally viable solution to meet relief capacity for such scenarios to cant increase in overall up-time for
challenging operational and regula- support the expansion of processing many facilities.
tory requirements. capacity. Since safe operating limits
can be set by relief- and flare-system EPA requirements
The benefits capacities, a well-implemented flare The U.S. Environmental Protection
Generally speaking, independent consolidation may lead to an in- Agency (EPA; Washington, D.C.) has
flare systems could be integrated crease in the plant capacity through released a National Petroleum Re-
through the installation of cross- modest upgrades to debottleneck finery Initiative to encourage refiners
over piping jumpers at appropriate the flare systems. to make a commitment to emission
location(s) between the main lines reductions. Since 2000, 109 refiner-
or headers of two existing systems. Ease of maintenance ies in 32 states and comprising over
The rationale and engineering re- With a single large network of mul- 90% of the total U.S. refining capac-
quirements for these modifications tiple flare systems servicing an entire ity have agreed to comply [3]. As
are described below. facility, some of the system capacity part of this initiative, a covered facil-
Though many factors can affect gained may allow for any one of the ity may not emit more than 500 lb of
the sizing of flare systems and relief multiple flares to be removed from sulfur dioxide in a 24-h period during
headers, it is usually a type of utility service for maintenance without sig- non-emergency situations, based
failure, such as total or partial power nificantly impacting the operation of on Emergency Planning and Com-
failure or a loss of cooling water flow, the facility. munity Right-to-Know Act (EPCRA)
that dictates the design basis of the With independent flares, flare Section 304.
disposal system. In most petroleum maintenance would require a One option to minimize emissions
The consolidation of multiple flare FIGURE 1. In the U.S., the total crude-oil processing capacity has increased steadily as the number of
systems into a single integrated operable refienries declined over the years
network requires a thorough under-
standing of the design basis for the loads (due to area fires, power fail- that there are no localized areas of
flare systems. There is no reasonable ures, cooling water failures and high velocity in the flare network that
means of combining multiple inde- major equipment failures) must be could cause mechanical problems
pendent systems without a concrete considered, documented, and eval- due to excessive momentum forces
and well-documented design basis uated. To properly size the FGRU or acoustically induced vibrations
for the flare systems, which in turn and isolation seal drums for com- 3. Flare equipment sizing To en-
is part of the process safety infor- pliance with the EPA requirements, sure satisfactory performance of all
mation requirement per the Process releases during routine operations relevant flare equipment, such as
Safety Management (PSM) Standard and maintenance must also be ac- liquid knockout pots (for vapor-liquid
29 CFR 1910.119(d)(3)(i)(D) [4] of the counted for and documented. separation), flare seals (including
U.S. Occupational Safety and Health seal height evaluation), flare tip per-
Admin. (OSHA; Washington, D.C.). Develop a model formance (in terms of flare radiation,
Evaluation of the location of the ex- noise, flame stability and so on), va-
Methodologies isting flare headers on a plot plan or porizers and heaters and more
All credible flare loads should be other facility map will allow for a quick 4. Mechanical limits of the header
considered to ensure that the modi- overview of the system connections To ensure that the flare header itself
fied system can safely, economically and aid in the design of proposed can adequately handle the expected
and efficiently handle the expected crossovers. Once the consolidated operating requirements
loads. It is also necessary to under- flare system has been developed, a 5. Liquid removal facilities To en-
stand the current layout of the flare hydraulic model of this flare network sure that the integrated flare loads
system to optimize flare header con- can be developed to verify that it will not result in excessive mechani-
nections. A thorough understanding can handle the expected loads ad- cal stress due to undesirable flow
of the current system and the pro- equately. Three types of modeling patterns (such as high-velocity liq-
posed changes allows the designer will be necessary: emergency load, uid slugs)
to confirm that the modified system non-emergency load and mainte- Non-Emergency load model-
can operate safely and to validate nance load. ing. All releases to the flare system
that flare maintenance requirements Emergency load modeling. A during normal operation or routine
can be met. standard method, such as the one maintenance of the facility should be
outlined in American Petroleum In- accounted for. Critical items to verify
Identify and document stitute (API) Standard 521 [5], is ap- include the following:
All potential loads to the flare sys- plied to ensure that the system can 1. FGRU To ensure that the normal
tem must be analyzed and docu- handle the expected emergency flare loads from all sources such
mented. The loads can originate loads. Critical items to verify include as leaking relief valves or normally
from a variety of sources (such as the following: open vent lines plus intermittent re-
relief valves, control valves, bleeder 1. Backpressure limits To ensure leases during maintenance (equip-
valves, emergency shutdown [ESD] that the relief capacities for indi- ment blowdown), and operations
systems and so on) and can enter vidual sources will remain adequate (backpressure control valves, startup/
the flare system due to a variety of despite the pressures built-up in the shutdown requirements, purging)
causes. To ensure that the system disposal system can be handled by the FGRU. This
is adequately sized, all emergency 2. Header velocities To ensure information is used to generate the
Ap
components can be quite Tip
maintenance and
Mixer
inspection best
Orifice
practices encourages
FIGURE 2. Pilot burners provide both
longterm operational Steam orifices
ignition and stabilization for the
main flame
reliability outer, secondary air register. Pre-
mix burners require less furnace
Mohammed H. Al-Hajji Fuel oil orifices draft than raw-gas burners. Pre-
Saudi Arabian Oil Co. mix burners can produce a wide
(Saudi Aramco) range of flame shapes.
B
urners are mechanical devices Oil burners
that are utilized for mixing Steam is always added to assist in
proper quantities of fuel and Oil
the atomization of liquid fuels. In
air, and also for maintaining Steam Steam
oil burners, the steam and oil are
a stable flame inside fired equip- mixed in specially designed fuel at-
ment. Burners are critical compo- FIGURE 1. Fuel atomizers are designed omizers, where the kinetic energy
nents that must be periodically to provide proper mixing between the oil of the steam jets breaks up the fuel
maintained to ensure the reliability and steam in oil burners into small droplets. The resulting
of fired equipment in the chemical mixture of steam and finely dis-
process industries (CPI), as well as lizer cone is located just below the persed oil is then released into the
many other industries. This article gas to improve combustion stabil- airstream through a number of ori-
discusses the internal components ity. These burners are suitable for fices in the burner tip. A typical at-
and applications for different types mounting in plenum chambers (the omizer is shown in Figure 1.
of burners (gas, oil and combina- area where air enters the burner), The steam used for atomization
tion), as well as the maintenance and can be used with preheated must be perfectly dry. If there is
procedures that are required to en- combustion air. moisture in the steam, this mois-
sure the integrity and reliability of Pre-mix burners are sometimes ture will flash when it mixes with
burners in the CPI. used in specialized applications. In the oil, causing erratic oil flow.
these burners, the kinetic energy Furthermore, the atomizing steam
Gas burners made available by the expansion should be superheated.
There are two principal types of of the fuel gas through the fuel- Mechanical atomization can be
gas-fired burners used in the CPI: gas orifice introduces about half of used when steam is not available.
raw gas and pre-mix. the combustion air (called primary In this method, the kinetic energy
Raw-gas burners are used for air) into the Venturi mixer. This in the oil itself is used for atomiza-
most applications. In these burn- mixture then exits through a large tion by releasing the oil through
ers, the fuel gas passes through burner tip, where it is mixed with the tip under very high pressure.
orifices in the gas tip and is injected the balance of the combustion air Mechanical atomization is usually
directly into the combustion zone, (secondary air). This secondary air- used only in very large burners or
where it mixes with air. A stabi- flow enters the burner through the with extremely clean fuels, since
40 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
the very small orifices required in mation is reduced. Secondary fuel
smaller burners can become plugged is added to complete the combus-
by small dirt or coke particles tion process. Staged fuel burners
in the fuel. provide greater NOx reduction, as
the fuel supply has a larger effect
Combination burners on NOx formation. This method is
The principles of gas and oil burn- more commonly used when a con-
ers can be combined into a third sistent fuel supply is available.
category of burner, the combination Internal fluegas recirculation
burner. A typical combination gas- burners. This method combines ei-
and-oil burner can be used to fire ther staged air or staged fuel with
FIGURE 3. Pilot burners use electrical
liquid, gas or a combination of liq- internal fluegas recirculation to current to detect the flames presence
uid and gas fuels, depending upon help reduce NOx formation. The
the fuel systems supplied with the best results are obtained where in-
burner. Liquid fuel is fired through ternal fluegas recirculation is used Burner components
a centrally located oil gun. The to dilute the fuel gas in a staged Burners are very complex devices
oil and atomizing steam are fed fuel burner, creating a gas with a not only are there many varieties
through separate pipes in the feed low calorific value. of burners utilized in the CPI, each
tube to the atomizer and burner tip. burner is equipped with a number
The fuel-steam mixture is injected Pilot burner of integral components that are
through orifices in the burner tip Pilot burners (or, simply pilots) are critical to operations.
into the primary airstream, where small burners used for ignition of Burner air registers. Air enters
combustion begins. the main burner flame. In some the burner through the air register.
systems, a pilot can also act as a Airflow can be controlled by adjust-
Reducing NOx in burners stabilizer for the main flame. Pilots ing the size of the openings in the
Combustion operations frequently in process-heater burners usually register. For natural-draft burners,
create nitrogen oxides (NOx), spe- operate continuously. Individual pi- the most common type of air regis-
cifically NO and NO2, which must lots should be removable for mainte- ter consists of fixed and moveable
be highly controlled, as they can nance, even while the process heater concentric cylinders, each with
be harmful pollutants. Currently, remains in operation. A typical pilot slots. For forced-draft burners, air
three burner configuration meth- burner is shown in Figure 2. registers are controlled either man-
ods exist for reducing nitrogen ox- More and more process heaters ually or automatically. Burner reg-
ides in burners: staged air, staged require the presence of a permanent isters may also be a single-bladed or
fuel and internal fluegas recircu- pilot with electrical ignition. These twin-bladed opposed butterfly type.
lation combined with staged air types of pilots use a low-tension ig- This variety is used for greater con-
or staged fuel. nition rod and usually combine the trol of the combustion air pressure.
Staged air burners. These types function of the low-tension rod with Plenum chamber. Sometimes
of burners work by introducing ionization detection. called a windbox, the plenum
100% of the fuel into the burner When the ignition transformer is chamber is where all the air for the
and only part of the combustion air connected directly to the end of the burners enters the device. The ple-
(primary air), thus creating a sub- pilot, standard low-tension cable num chamber may contain control
stoichiometric flame. This flame can be used. This eliminates the vanes, as well as the air registers
has a reduced temperature and need for high-tension cable and, as for the burners. The plenum cham-
therefore inhibits NOx formation. a result, longer cable runs can be ber can serve several purposes,
The flame is completed with the ad- used from the pilot to the distrib- including noise and emissions re-
dition of the secondary air to com- uted control system (DCS). duction, single-source combustion
plete the combustion process. This An electrical current is con- air supply and total airflow to the
process allows for greater control at ducted between the flame and the furnace controls.
lower burner loads and also accom- flame-retention head. The flame is Burner fuel-gas piping. Burner
modates a wider range of fuels. ionized, and the ionization part of fuel-gas piping consists of the man-
Staged fuel burners. This burner the flame rod detects this change ifolds and piping that deliver fuel
method introduces 100% of the in the current. This electrical cur- to the burner tips. The fuel passes
combustion air into the burner and rent change is then relayed by a through one or more openings in
splits the fuel supply into primary lamp to indicate the presence of the the tips, which act as restriction or-
and secondary volumes. The primary flame (Figure 3). It is crucial to en- ifices. Here, the fuel is injected into,
fuel mixes with the combustion air sure that the pilot flame has a very and mixes with, the airstream. Var-
to create a flame. Again, as with strong core to enable consistent de- ious types of burner tips are used,
staged air burners, the peak flame tection of the flames presence in depending upon the type of fuel
temperature is lower, and NOx for- the pilot burner. and the flame pattern desired.
CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014 41
Cover Story
rial that could decrease the effec- in heavy fuel-oil service will usu- must be removed and cleaned.
tive port size. If the ports become ally have stubborn hard deposits Flame stabilizer. Different
partially or completely plugged, on the surface and in the ports flame-stabilizer designs are used
the quantity and distribution of Use steam cleaning or a twist to maintain a stable flame in the
fuel entering the burning zone may drill, in combination or individu- combustion zone. For natural-draft
vary from the design objective, and ally, to remove harder deposits. burners, these designs take into ac-
combustion problems may occur. It is not recommended to use a count the diffuse cones, ledges and
The type of material plugging the power drill or tool to clean ports tapered tiles of the burner. Poor
ports determines how it should be or oil tip surfaces, as any nick or conditions or the absence of a sta-
cleaned. Foreign material must be notch on the oil tips surface acts bilizer can result in unsatisfactory
carefully removed from the fuel as a site to collect oil and acceler- flame shape or an unsafe flame that
orifices so as not to affect the ori- ate tip-coking problems can lift off the burner and leave the
fice dimensions. Fuel tips should Soak the tip in a solvent, fol- zone where combustion is initiated.
be replaced if any orifices exceed lowed by blowing steam through Operations suffer when part of the
the specified diameter by more the orifice to remove the remain- flame lifts off the burner due to a
than one to two twist-drill sizes. ing oil or solid deposits damaged stabilizer because irregu-
If the material is not easily re- Handle the atomizer carefully lar, erratic flame patterns occur,
moved, some cleaning guidelines when removed from the oil gun and new risks are introduced into
should be followed: to protect the labyrinth seal. the process. A damaged stabilizer
Soak the burner part in a solvent This seal separates the oil from must be replaced.
to loosen the deposit the steam by a series of rings on Air registers and dampers. All
Use a twist drill to remove the the atomizer with a tight toler- registers and dampers should be
deposit. Never use a power tool ance between the atomizer and exercised periodically to ensure
with the twist drill because it is the gun body. If the atomizer is that they are functional. If damp-
likely to enlarge the ports continuously getting stuck and ers are inoperable, check the actua-
Use a welders file to gently re- cannot be removed from the gun tor and repair if necessary. Damp-
move the scale from the orifices. without breaking the atomizer, ers are often inaccessible during
If the scale cannot be removed, the metallurgy may be at fault. operation and repairs must wait
the fuel tip or orifice must In this case, it is recommended until the equipment has been shut
be replaced to exchange the atomizer metal- down. Clean the register of foreign
Use wet steam or hot water if the lurgy with a different metallurgy materials, spilled oil, sand, rust and
source of the foreign material is than that of the oil tip insulation that blocks airflow. If the
an amine compound, because Burner tiles. The unique profile of register is deformed, possibly due
amines are water-soluble a burner tiles surface is challeng- to incorrect operating procedures,
Shut down the fuel gas valve ing to reproduce, making tiles diffi- or the operating handle is broken,
and inject steam into the burner cult to fix and return to service suc- repair or replace the part.
if the amine plugging occurs fre- cessfully. Also, the refractory of the Pilot burners. The most common
quently without removing the tile undergoes phase transforma- pilot burners used in the CPI are
burner. Note that injection clean- tion while in service, and refractory small, heat-release pre-mix burn-
ing may not clear all tip ports repairs usually do not adhere to the ers. The basic parts of these pilot
equally surface for very long periods of time. burners include the pilot tip, gas
Soak the tips in a hydrocarbon Burner tiles should be replaced if mixer, mixing tube and orifice. The
solvent, followed by cleaning they experience any cracks or de- gas orifice is commonly 1/16 in. in
with a twist drill if the tips are formations. Especially of concern diameter and can be easily plugged
plugged with polymers are major cracks, mainly in wall- or by pipe scale. This element is
Remove the coking, polymer or roof-mounted burner tiles, where cleaned by hand with a twist drill,
solid deposit by oxidation in a the tile pieces are expected to fall. as previously described. If inspec-
small, high-temperature furnace. Also, one should look carefully for tion reveals a damaged pilot tip,
Many tips can be cleaned at the multiple cracks in a section, which the tip should be replaced.
same time with this technique present evidence of crushing due to Flare pilot burners. When the
Oil tip and atomizer. Fuel-oil restrained expansion, and are rea- flare system is in service, monitor
tips are harder to maintain than son for replacement. The primary the supply pressure and clean the
gas tips and require more frequent or oil tile shall be checked as well. fuel strainers, knockout pots and
cleaning. Tips in light fuel-oil ser- If it is badly pitted or cracked, the drains that are accessible from
vice may have only a slight carbon- recirculation of gases within the grade. When the flare system is
like deposit on the surface. To clean tile is uneven, and coking can occur not in service, it is recommended to
fuel-oil tips, do the following: on the tile. This coking can lead to clean the strainer upstream of the
Use a wire brush to remove oil dripping and spilling from the orifice and to inspect the orifice.
slight carbon-like deposits. Tips burner. If this is observed, the tile Flame-front generators require
44 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM NOVEMBER 2014
Cover Story
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.
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 erations personnel hold the key to delivering David
as an owner implies certain re- optimal business objectives, through their ef- Rosenthal
sponsibilities. Consider the range forts related to the ongoing operation of the Reliability Strategy
of behaviors demonstrated by indi- assets. and Implementation
viduals when it comes to the automobiles To further explain this important con- Consultancy, LLC
they own. For some, car ownership leads cept, we must first understand the bathtub
to a relentless pursuit of caring for every curves, developed by Nolan and Heap in
aspect of the car, from operation to main- the 1960s and 1970s [1], which have driven
tenance. For others, it is a daily wish that maintenance practices in the airline industry
their cars will simply start when the ignition for decades. The authors developed six fail- IN BRIEF
is engaged. The outcome of the diverse be- ure curves that demonstrate how the prob-
OPERATIONS
haviors along this continuum of ownership ability of failure is a function of run time for OWNERSHIP
will have a direct impact on the reliability, machine components. The major finding
longevity and cost of ownership of these was that 89% of these failure modes occur THE PATH FORWARD
complex machines. randomly often with little to no warning. DEVELOPING THE VALUE
Throughout the chemical process indus- At the time of these findings, industrys ap- PROPOSITION
tries (CPI), owners of complex, costly ma- proach to maintaining industrial and other
ESTABLISHING METRICS
chines and systems (assets) tend to act along assets had been to rely heavily on preven-
this same continuum. In general, CPI opera- tive or time-based activities, such as planned CHANGING THE
tors typically want to ensure the delivery of overhauls. However, given that the majority EQUIPMENT-CONDITION
performance levels in terms of three impor- of failures occur randomly, it is not practi- MINDSET
tant measures reduced lifecycle costs, cal to expect that a time-based approach to TRAINING AND
improved reliability, and increased longevity equipment maintenance will detect or iden- TROUBLESHOOTING
before replacement. However, different indi- tify all potential failures. While online moni-
INTEGRATING
viduals will go about achieving these objec- toring options can provide a close proxy for
MAINTENANCE WORK
tives in different ways. Within any CPI facility, realtime surveillance in some instances, we
PROCESSES
the quest to ensure reliability is thought to cannot place a mechanic at each machine
require three partners personnel from the to constantly monitor its condition on a real- PERFORMING
operations, maintenance, and engineering time basis. The operations department is the EQUIPMENT ROUNDS
Depts. All three sets of individuals play vital only group with enough continual exposure AUTONOMOUS
roles in helping the asset to meet its impor- to the assets on the plant floor to be able to MAINTENANCE
tant objectives, via their interactions through- detect the earliest signs of many impending
out the lifecycle of each asset. failures. So why do many CPI facilities still
The efforts of the engineering department experience relatively high levels of reactive or
should be building in reliability since the de- breakdown-related maintenance, and fail to
sign itself has a greater impact on reliability effectively deploy their operational personnel
compared to the efforts by maintenance to provide close ongoing surveillance of the
and operations depts. combined. For many, assets in the field?
maintenance department efforts are thought
to be the primary element responsible for Operations ownership
the reliability of the installed assets. How- In recent decades, there has been a transition
ever, from our knowledge of various paths in operations department culture. Many retir-
for equipment failure (the majority of which ees lament the bygone era when operators
are random in nature), it turns out that op- knew not only how to operate their equipment
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-
COUNTLESS
through greater uptime, reduced
maintenance costs and a safer
working environment. As operators
are closest to the daily operation of
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 FIGURE 1. Nurturing a culture of cohesive teamwork among consultants, project
conditions, yet not be overdesigned teams and end-users is difficult, but can generate benefits for equipment reliability
in a way that leads to higher overall
lifecycle costs. To sucessfully design End-user interactions The mode of interaction among the
and commission plant assets, two The end-users of the plant asset can end-users and project-management
critical objectives come into play: es- provide a great deal of data to the en- group for the information transfer
tablishing a culture of cohesive team- gineering and construction team, and described above is of immense impor-
work among the consultant, project that information should be captured tance, because the two groups have
and end-user teams; and following a in the right perspective, and then for- different core objectives to achieve.
structured procedure with the help of malized and structured in a way that It is often a difficult task to align the
equipment-specific checklists. promotes understanding. If that is mindset of both groups and to bring
The teamwork component of this achieved, the result is a high degree them under a common umbrella of un-
approach is intended to systematically of practicality and flexibility in the derstanding, so that the exchange of
capture on-the-ground realities of way the information is implemented. knowledge becomes collaborative, fo-
the plant and to realize longterm ben- During the data collection process be- cused and effective, resulting in real-
efits by incorporating improvements tween the engineering and end-user ization of an integrated common goal.
into the design, construction and com- groups, there is potential for a great In other words, a mutually agreed-
missioning processes. To do this, the deal of irrelevant and impractical in- upon structured methodology must
active participation of the end-users formation to be exchanged. Hence, the be in place for communication and
is essential throughout the process organization and use of the data must knowledge management that takes
(Figure 1). The second component, be carried out by a team of subject- into account the profile of participat-
following a structured set of detailed matter experts with visionary wis- ing teams, the nature of the informa-
checklists, is intended to optimize the dom, so that the basis of action items tion available, and the areas where
equipment design and setup at the is justified from an economic, as well that information will be applied.
outset, to allow maximum reliability. as a reliability, point of view. When this structured methodology is
52 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013
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-
established, the project workflow will tem, including all mechanical and hydraulic forces acting upon it
remain on a progressive track. 2 Design provides inherent protection to components from corro-
sive or erosive environments to which it is, or can be, exposed as
This process of value addition by a part of the operating process
effectively merging end-users experi- 3 Design provides a correct and realistic input process condition
ence from the design selection stage to for the equipment
commissioning of the asset is unique Design provides a correct and realistic equipment output perfor-
for each project and has to be engi- mance within the limit of specified input condition
neered to allow for modifications to 4 Design provides the most suitable hydraulic balancing for equip-
ment, so as to minimize the forces acting on components
accommodate the systems and proce-
5 Design provides a specified trouble-free operating zone without
dures for different organizations. If affecting the desired performance and component life
the goal of smooth knowledge trans- 6 Design provides correct and robust shaft-sealing system that
fer is achieved, a safe, on-time, cost- can withstand the most adverse operating conditions possible in
effective execution of the design and the system without failure
installation process will translate into 7 Design provides an appropriate bearing-support system to
provide maximum stiffness to the rotating mass at minimum fric-
a sustained reliable and productive tional loss
operation thereafter. 8 Design provides proper lubrication system that is most suitable
to protect the bearings at minimum friction loss
Defining the five Gs 9 Design provides the most suitable lubricant for the bearing sys-
While several methodologies can be tem, considering the load and service conditions
derived for the interaction of end-users 10 Design provides for bearings and seals with wear- or failure-
monitoring capability to permit scheduling of maintenance prior
and project management consultants to actual component failure or component damage
(PMCs), it is extremely important to 11 Design provides suitable casing design pertinent to service con-
convert the information into execut- ditions to provide highest built-in reliability
able tasks that keep the inherent val- 12 Design provides that all running clearances are minimized with
applicable tolerances for thermal expansion
ues intact. The most popular and ef-
13 Design provides monitoring facilities for crucial performance
fective roadmap for structuring these and physical dynamic parameters of the equipment, so as to as-
interactions is to develop checklists sess the health of the machine
and procedures for the design selec- 14 Design provides suitable bearing-housing seals to avoid ingress
tion, pre-commissioning and commis- of contamination and leaking of the lubricating oil
sioning activities and to assign respon- 15 Design provides the base frame that enforces adequate rigidity
to dampen the vibration generated in the system and prevent
sibility to the teams in the loop. Such the transmitted vibration from other sources to enter
checklists play vital roles in ensuring 16 Design provides an appropriate piping support system that pre-
that proactive measures are taken for vents piping stress to the equipment and also to eliminate the
all asset-reliability issues, operability potential of generating unsymmetrical dynamic forces in the
system during operation
issues and maintainability issues. To
17 Design provides suction, discharge and auxiliary piping system
the extent that those measures can that will not cause excitation of flow-related problems and sizing
be taken, a more trouble-free commis- shall not be over-designed or under-designed in relation to the
sioning is possible, and an enhanced maximum rated flow, NPSH (net positive suction head) required
effective service life of the equipment 18 Design provides a coupling assembly between the driver and
driven system that should be rated for a minimum of 1.5 times
will follow. The process flow can be ex- the service factor and above the maximum starting torque re-
pressed as the five Gs: quired, as per the speedtorque curve of the machine
1) Get. Get end-users expertise and 19 For single-stage rotor, the entire assembled rotor, along with cou-
participation pling hub mounted, should be balanced in accordance with API
684 or ISO 1940 grade 2.5
2) Gather. Gather information and
20 For multistage rotors, individual impellers should be balanced be-
data on reliability, operability, fore assembly. The assembled rotor should be balanced thereafter
maintainability and safety issues. 21 Design provides the driver rating for end-of-the-curve operation
Develop procedures and a check- 22 Design provides adequate external protection from weather and
list system to incorporate proac- harmful interference from other equipment and systems
tive mitigating measures in design, 23 Design provides a firm identification tag posted at a suitable
location close to the equipment that mentions rated operating
erection and commissioning of the parameters, allowable physical operating parameters and alarm
new asset or facility, in consulta- limits of critical performance parameters
tion with the PMC, licenser or OEM 24 Design provides applicable interlocks, alarm annunciation and
(original equipment manufacturer) a trip system for crucial parameters to mitigate unplanned cata-
strophic failure of equipment
3) Grow. Grow the new facility by com-
25 Design provides color-coding of equipment and connected pip-
mencing the design, erection and ing to identify its service media
pre-commissioning in line with the 26 Design provides vortex breakers in the suction and discharge
developed system and procedures nozzles to avoid separation
4) Gain. Gain an on-time, safe and 27 Design provides a provision for an oil-mist lubrication system for
smooth commissioning by closely antifriction bearings of pumps
sioning centrifugal pumps, but they 7 Design provides an applicable in-built protection system for elec-
trical equipment and instruments in line with hazardous area clas-
can also be applied to other pieces of sification API RP 500, IS 5572, IEC 79-10 :1995, NFPA 69
rotary equipment beyond pumps. 8 Design provides that service-specific emergency handling for
The purpose of the checklists found safety equipment and systems are readily available and stepwise
procedures are displayed close to equipment
in these tables is to provide a sum-
mary of the design review points for 9 Design provides a data historian system for critical process and
environment safety information
assessing the maintainability of new
10 Design provides for automatic actuation of safety-protection sys-
or existing rotary equipment for petro- tems for equipment that handles flammable, toxic, auto-igniting
leum refineries or other CPI facilities. and other hazardous media
The checklists specifically focus on 11 Design provides equipment-specific safe operating and mainte-
nance procedures
the identification of equipment-design
features, tasks, or procedures that
impact equipment downtime, repair TABLE 3. DESIGN STANDARDIZATION FEATURES FOR CENTRIFUGAL PUMPS
Description of check point Adequate Remark
costs, labor hours and the skill level no.
requirements for maintenance staff. Yes No
The specific objectives for each table 1 Design provides interchangeability of equipment that is in similar
or nearly similar service to all new pieces of equipment that are
are as follows: under procurement
Table 1. General maintenance 2 Design provides for interchangeability of equipment that functions
reduction similarly with equipment that already exist in the operating facility
Table 2. Safety and environmental 3 Design describes the full scope of standardization and inter-
changeability of components, such as shafts, impellers, wear
system design features parts, mechanical seals, seal spares, bearings, couplings, gaskets,
Table 3. Design standardization hardware, valves, connectors and others
features 4 Design provides vendor-qualification criteria for maintaining stan-
Table 4. Design features for routine dardization as a nodal decision-making point
maintenance 5 Design provides for interchangeability of monitoring, control in-
struments, piping support system and so on
Table 5. Design features for trouble-
shooting
Table 6. Design features for repair Implementing the five Gs the basic specifications and execution
and replacement The remaining portion of this article plans, while the licensor would pro-
Table 7. Accessibility for visual in- elaborates on the methodology, and vide the most applicable technology
spections and monitoring provides examples of checklists using and operating profiles.
Table 8. Design for physical centrifugal pumps as the asset. There Under the guidance of a companys
accessibility are three broad phases for the process. management, one or more teams from
Table 9*. General precommissioning Phase 1 Get and gather. Each various disciplines would be formed,
checklist and commissioning proce- participating group should prepare making sure to involve the relevant
dure for centrifugal pumps its own objectives to share, depend- knowledgeable persons and having
The Description of check point column ing upon the project type. The project- that particular project department as
in each table covers the essential fea- management department prepares the principal facilitator.
tures for completion of items required the project details, including the back- In the first phase, the end-users
for the safe and first-time-right com- ground, business objectives, location, feedback and participation comes into
missioning of rotary equipment in the mandate and so on. This would be the play in the form of input to the project
given category. primary input and the basis for select- management consultant on the practi-
ing the consultant, technology licenser cal problems of reliability, operability,
*Editor's note: Table 9 can be viewed in the or plant-equipment system or asset in- maintainability and safety, so that the
online version of this article (www.che.com) volved. The consultant would prepare weak points can be addressed at the
54 CHEMICAL ENGINEERING WWW.CHE.COM SEPTEMBER 2013
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
reviewed. These include construc- Documenting and record keeping are required. This can save a great
tion and testing records for inter- for commissioning activities and in- deal of time and money.
connected piping, systems, electrical cidents with absolute transparency
circuits and control loops Managing progress updates Closing comment
Ensure adequacy of fire protection The process described here has become
systems, for example, fire-water sys- Benefits of giving an inherent part of any project man-
tem, gas-leak detection system, fire The benefits for following through on agement because it nurtures a win-
extinguishers, and so on the giving component of the process win bond between the project group
Ensure safe movement and ap- are felt by end-users, and also extend and end-users group deriving sub-
proach of equipment to the the organization as a whole. The stantial benefit for the organization.
Note equipment-specific hazards benefits include the following: Without such a structured approach,
Determine deployment of trained Improved reliability projects could draw displeasure from
manpower Increased safety the end users as well as from manage-
Ensure use of personal protective On-time delivery of quality product ment for any design, operability and
equipment (PPE) Higher customer confidence maintainability issues in the newly
Ensure that equipment installation Improvement through learning commissioned plant. Futhermore, end-
procedures are consistent with re- Fostering long-lasting mutual re- users could struggle to sustain the op-
spective equipment standards spect among teams eration with a problematic system, re-
Apart from technical activities, the Solidifying ownership concept among sulting in a state of unreliability. With
most important areas that the proj- the post-commissioning team this system in place, projects are likely
ect management team must strive to Fostering collaborative culture to have reduced post-commissioning
implement are the following: Increased plant familiarity and problems and more reliable plants.
Preparing a daily activity list and hands-on training of end-user per- Edited by Scott Jenkins
review of progress sonnel (competency development) Author
Adhering to safety guidelines Optimize inventory management Sourav Kumar Chatterjee is
chief manager of rotary equip-
throughout the process through standardization and inter- ment at Mumbai Refinery of
Establishing an effective good- changeability of asset-equipment Hindustan Petroleum Corp.
ltd. India (Petroleum House 17,
housekeeping procedure system spare parts Jamshedji Tata Road, Mumbai
Maintaining 360-deg communica- In addition, another benefit of end-user 400020; E-mail: skchatterjee@
hpcl.co.in; Phone: +022 228-
tions with all stakeholders participation in precommissioning and 63900). Chatterjee is a char-
tered mechanical engineer
Holding a toolbox meeting to ap- commissioning is that, because end and qualified boiler proficiency
praise the respective execution per- users hold good rapport with internal engineer with a diploma in
electrical engineering. He has more than 23 years
sonnel on the scope of the job, safety and external local resource providers, experience in thermal power plant operation and
rules, job-specific safety precautions they can guide the executing contrac- maintenance, and design maintenance operation
of refinery rotary equipment. He specializes in root
and PPE tor on the availability of quality fab- cause failure analysis and equipment reliability. He
Adhering to the work permit system, ricators in cases where urgent site had presented many papers at national and inter-
national conferences, and has published papers in
without exception modifications or hardware changes various journals and magazines.
E
Fred Ashmore very year the process industries Equally important: Writing down the rea-
International Fire suffer hundreds of fires, some seri- sons for choosing certain fire protection is a
Investigators and ous and most minor. Fire protec- record for those who come afterwards. It is
Consultants tion matched to the risk is what easy to forget the reasons behind important
makes the difference between the fire that safety systems, causing accidents to be re-
torpedoes your business and the one that peated.
doesn't even make the local newspaper.
For most process engineers, fire protec- What have we got?
tion is something that specialists deal with. The list of fire hazards in a process plant is
But wrong judgments or decisions make nearly always topped by flammable liquids.
the difference between a controlled incident Flammable gases are also common, but the
and one that escalates. Because specialists fire potential can often be mitigated by dis-
tend to plan for the worst-case scenario, the posing of the inventory by flaring.
plant may overdo fire-protection systems. Process plants dealing with materials other
Although this sounds like a good idea, it than flammable liquids may have problems,
means investment and maintenance effort too. For example, paper plants feature high
that could be better directed. in the list of major loss by fire, often involv-
So it is important for the non-specialist to ing storage, but also including incidents in-
work with fire-protection designers and ask volving paper machines. Food plants handle
the right questions. The fire protection of a combustible materials, often with potential
process plant should be based on a realistic for dust explosion.
assessment of fire cases. Other areas may have hazards that seem
Fire-risk assessment is a hot topic today less likely to be spectacular but could be
and there is a mass of guidance, systems, equally damaging to the business. For exam-
publications and consultants eager to help ple, a fire in a warehouse full of agrochemi-
(see box, page 78). Performing a fire-risk as- cals can lead to runoff of contaminated-fire-
sessment comes down to asking the ques- fighting water or a toxic-smoke plume drifting
tions: off site. Or the loss of a thermal-fluid heater
What have we got? could stop production for months. Although
What can go wrong? preventing such incidents is the priority, one
What should we do about it? needs to understand what fires could occur
Are we really doing what we should? and be able to deal with the consequences.
A useful approach is to prepare an inven-
The fourth question is especially important tory of fire hazards for the area or site under
for the operating plant. concern, and to list possible fire scenarios.
If an assessment exercise is carried out With this, you can talk seriously with fire-pro-
properly, there should be a better under- tection specialists to determine what protec-
standing of fire hazards and precautions in tion is essential and what is optional.
the plant, and why they are in place. This is A fire-hazards inventory should specify the
desirable because fire-protection systems combustibles, define their fire properties,
are non-productive and often lie idle for and assess how they might be involved in
years. Their maintenance priority can eas- a fire. The hazards of process materials are
ily slip unless all concerned understand the defined in the material safety data sheet of
systems. commercial products.
2 CHEMICAL ENGINEERING WWW.CHEMENGONLINE.COM MARCH 1998
tions may already be available for the maxi-
Pinpointing the fire hazards of solids can mum overpressures that could develop from
be more difficult compared with liquids, be- an explosion. For the purposes of assess-
cause the form affects how they respond to ing fire protection needs, note explosions
ignition sources. For instance, lump sugar is as something that could disable protective
not readily ignited, but a cloud of icing sugar systems. An explosion is more immediately
dust can cause a severe explosion. hazardous to life than is fire, because it can
When a flammable gas or vapor is in- happen too fast to permit either reaction or
volved in a confined area, there is an explo- flight. Fire protection does not control explo-
sion hazard. Explosion and fire hazards are sion hazards.
linked, so be sure to note these in screening For a fire scenario with process fluids, as-
fire problems. sess how long it could last if there is no re-
As a simple example, consider the small sponse. Both jet and pool fires can impose a
batch plant shown in Figure 1 with its inven- heat flux of around 200 kW/m2to structures
tory of materials with fire hazards. The plant or equipment that are engulfed. This could
is an open process area with enclosed con- raise the temperature of a major steel mem-
trol room and offices. For the process area, ber to failure point in a few minutes. It can
there will be releases in various forms, and also weaken process vessels so that they fail
the fire scenarios arise following ignition. For under internal pressure, even with pressure
this set of material's hazards, the following relief.
scenarios are realistic: The duration of a pool fire depends on a
number of factors, including the inventory
A flammable liquid fire, either from a of flammable solvents, burning rates, leak
pressurized release or from flowing liquids. rates, and drainage. It may be helpful to use
The total inventory of flammable and com- one of the many computer packages for
bustible liquids is 19 m.t., including the heat process hazard analysis which are available.
transfer fluid. This fire could be: a pool fire; For this example, suppose that you as-
a jet fire involving sprayed liquids or vapor; sess the maximum area of a pool fire as 50
a flash fire of a vapor cloud; or a running- m2. If the solvent has a pool-regression rate
liquids fire of 5 mm/min this is a typical value for or-
Possible explosion if there is a big re- ganic solvents and hydrocarbons the total
lease of hot solvent burning rate would be 0.25 m superscrip 3/
min. With a 12-m superscrip 3 inventory, the
Fire at the heat transfer fluid heater. It
pool fire could last 48 min.
could involve the whole inventory of the
In contrast to a pool fire, jet fires might
system (5 m.t.)
only last a few minutes if there is an effec-
Fire involving the bag storage of chemi- tive blowdown system. However, even with
cal C. Check whether it's likely to melt and an overpressure of one or two barsg, there
spread to other storage can still be a substantial rate of release, so a
Fire involving ordinary combustibles or jet release could last longer than the 15-min
maintenance items in storage area nominal blowdown time under API 521.
Control room or office fire Other types of plant can be analyzed in
What can go wrong? the same way, one scenario at a time. Pro-
With the inventory of combustibles and cesses with complex chemistry and batch
notes on how they are arranged, the next plants with the potential for exotherm require
step is to determine what can go wrong. For more detail and may have to consider ad-
this type of exercise it is wise to take a team ditional types of fire scenario. In addition to
approach to make sure all the major prob- simple leaks, a runaway reaction could lead
lems are caught. The team must include to major failure of a reactor or pipework. And
people who understand the engineering, changes in process conditions from unfors-
operations and safety issues. eeable upsets could lead to major change in
From the list of fire hazards, the team de- fire potential. A systematic analysis will help
velops a list of fire scenarios. Possible igni- identify all fire cases.
tion sources are not considered initially be-
cause the concern is with consequences, Check the engineering
not probabilities. Before fire protection is considered, the
A possible explosion the second item scenarios need to be checked against en-
above should be assessed as part of the gineering safeguards and shut-down sys-
overall safety analysis of the plant. Calcula- tems. Do these safeguards place a limit on