011 CH&S_July_2004

8/12/2019 011 CH&S_July_2004

1/7

Learning from reactivechemistry incidents

The general lessons from an incident are often more important for preventing futureoccurrences than the details of what actually happened in that particular event. An incidentis always a specific example of a more general category of potential incidents. If the investigationand follow up consider the more general lessons, you can prevent other similar events. Severalspecific reactive chemistry incidents will be discussed, with an emphasis on general lessonswhich can be applied to a broad range of processes.

By J. Kent Carpenter,Dennis C. Hendershot,and Steven J. Watts

INTRODUCTIONIn the past several years there has been a renewed focus on reactivechemistry incidents by the chemicalindustry, and from government andregulatory agencies. The US Environ-mental Protection Agency issued cas estudies related to runaway reactio ns1

and reactive chemical explosions 2 in1999 and 2000. The United StatesChemical Safety and Hazard Investi -gation Board (CSB) issued a report 3

describing the results of a study of reactive chemistry incidents. TheCenter for Chemical Process Safety(CCPS) has published an alert 4 whichintroduces basic tools for understand-ing and managing reactive chemistryhazards, and a book with more exten-sive guidance f or managing reactivechemical risks. 5

Understandingthe causes of pastreactive chemistry

incidents is animportant tool for preventing future

incidents.Understanding the causes of past

reactive chemistry incidents is animportant tool for preventing futureincidents. The details of a particularincident are specic to the process

and chemistry in which the incidentoccurred. However, all incidents pro-vide general lessons which are usefulfor preventing future events, regardlessof the technology and chemistry. Thispaper will review three reactive chem-istry incidents, focusing on general les-sons which are applicable to a widerange of processes and technologies:

1. An unexpected runaway reactionduring decomposition of an organicperoxide dissolved in an organic

solvent2. A decomposition reaction while blending two organic compounds,a process in which no chemicalreaction was intended

3. An uncontrolled release from aneutralization vessel caused by in-sufcient solvent during the reac-tion

We will also discuss tools which can be used to understand the potential forreactive chemistry incidents, and toavoid them in the future.

We will also discusstools which can beused to understand

the potential for reactive chemistryincidents, and toavoid them in the

future.

INCIDENT 1ORGANICPEROXIDE DECOMPOSITION

Process DescriptionThe plant equipment involved in thisincident consisted of a batch solutionpolymerization reactor, which wasprotected by a rupture disk which dis-charged to a catch tank, and then toatmosphere, as shown in Figure 1 .

The rupture disk was properlydesigned to protect the reactor fromoverpressure in case of a runaway poly-merization reaction for the worst caseof many polymer products which weremanufactured in the reactor. The catchtank was a at top vessel not rated forany pressure, which had been in placefor a number of years. This catch tankhadbeenidentied as potentially inade-quate,andanupgradeprojectwas inthedesign stage at the time of the incident.

Incident DescriptionBecause of an instrument failure, a batch of polymerization initiator, adilute solution of 2-amyl-peroxy-2-ethylhexanoate (CAS 686-31-7) dis-solvedin a mixture of common solvents

FEATURE

10 Division of Chemical Health and Safety of the American Chemical Society 1074-9098/$30.00Elsevier Inc. All rights reserved. doi:10.1016/j.chs.2004.01.014

J. Kent Carpenter, Dennis C. Hendershot (e-mail:[email protected] ), andSteven J. Watts are afliated withRohm and Haas Company,Engineering Division, 3100 StateRoad, Croydon, PA 19021.

Prepared for Presentation at American Institute of ChemicalEngineers, 2003 Spring NationalMeeting, Process Plant SafetySymposium, Paper No. 27a, NewOrleans, LA, March 31April 2, 2003.
http://[email protected]/http://[email protected]/

8/12/2019 011 CH&S_July_2004

2/7

including ketones and aromatic hydro-carbons, was improperly prepared. Itwas notpossible to re-work theinitiatorsolution so that it could be used for

production. Because of the dif cultyof disposing of the peroxide solution,the plant decided to use a laboratoryprocedure to decompose the peroxidein the solution, allowing the remainingsolvent to be burned as normal wastesolvent. The procedure was as follows:

charge the off grade peroxide solu-tion to the reactor

add sufficient additional solvent sothat the batch level reaches the agi-tator blades

turn on the agitator and heat the batch to a specified temperature( 100 8 C)

hold for a specified time at that tem-perature

cool and discharge the material fordisposal as solvent waste

The temperature and time for theperoxide decomposition were deter-mined by the laboratory based on theperoxide half-life to ensure completedecomposition of the peroxide. Sup-plier literature reported the half-lifefor 2-amyl-peroxy-2-ethylhexanoate asabout 6 min at about 100 8 C, althoughthe half-life can vary depending on fac-tors such as the solvent and concentra-tion. This peroxide decompositionprocedure had been used many timesin the laboratory without incident.Unfortunately, the half-life of a perox-ideonly provides information about therate of the decomposition reaction. Toensurea safe scale up of thedecomposi-tion process, additional informationis needed, including:

heat of reaction (decomposition) decomposition products (particu-larly non-condensable gases)

maximum adiabatic temperature

and pressure from the decomposi-tion reaction

In the plant, the operators had dif -culty in heating the batch not surpris-ing because it just barely reached theagitator blades and only a small por-tion of the reactor heating jacket wasactually in contact with the small batch. The operators increased theheating, and the temperature beganto rise more rapidly. As the batch tem-perature approached 90 8 C, the opera-

tors reduced the heating, but thetemperature continued to rise rapidly.Pressure was observed on the reactorpressure gage, and cooling wasapplied, but it had little effect notmuch of the jacket area was exposedto the batch and available for coolingeither. The batch temperature contin-ued to rise, and pressure began to buildup in the reactor. Eventually the pres-sure reached the rupture disk burstpressure of 135 psig, and the diskopened, discharging to the catch tank.

The rupture disk size was adequateto prevent overpressurization of thereactor, and there was no damageinside the building. However, thiswas by good luck, not by design. Therupture disk was sized based on a run-away polymerization reaction for anormal production batch, not for aperoxide decomposition reaction. Out-side the building, the catch tank wasoverpressurized, and its top was blownoff. The catch tank top was propelledinto the pipe bridge above the tank,and it damaged or ruptured several

pipes in the pipe bridge. Fortunately,the spilled material was contained in apaved area. Some ammable materialswere released, but did not ignite. Thecatchtank top did nothit anybody, andthere were no injuries.

Causes

Inadequateunderstandingof reactivechemistryhazards .Theperoxidehalf-life provides some information aboutthe rate of decomposition, but noinformation about the heat of reac-tion, the decomposition products, orthe potential maximum temperatureand pressure from the reaction.

Insufficient understanding of theeffects of scale up on the behavior

of the reaction . This decompositionhad been carried out numerous timesin thelaboratory on a small scale withno problem. But this is not an indica-tion that the reaction can be donesafely on a large scale. The relativelylarge heat removal capability, andeven heat leakage to the surroundingenvironment, allows a reaction to becontrolled in the laboratory when it isnot possible to control the same reac-tion in plant scale equipment.

This batch process contained all of

the possible chemical reactionenergy at the start of the reaction .The process depended on controlof the batch temperature to limitthe rate of reaction. Once that tem-perature control was lost, the run-away could not be prevented. A better process would have been togradually add the peroxide solutionto a reactor containing hot solvent,at a temperature high enough thatthe peroxide would have decom-posed immediately. If there wereany problems with temperature orpressure control, the feed of perox-ide solution could be stopped, andthere is no potential chemical reac-tion energy present in the batch.

Inadequate catch tank design . Whilethe catch tank had been recognizedas inadequate, it was not recognizedthat it could fail as badly as it didwhen the incident occurred. A newcatch tank was designed using cur-rent methodologies as developed byDIERS (Design Institute for Emer-gency Relief Systems).

Reactor

Catch Tank

Building

Pipe Bridge

RuptureDisk

Figure 1. Polymerization reactor equipment layout.

Chemical Health & Safety, July/August 2004 11

8/12/2019 011 CH&S_July_2004

3/7

Other causes not directly relatedto reactive chemistry hazards . Thisunusual operation was not captured by the plant Management of Change(MOC) process, because the MOCsystem was directed toward normal

production changes. Also, therewere no detailed operating instruc-tions for this operation.

General Lessons

Treat all chemical reaction processesthe same, whether they are partof themain production process or not.Make sure that Management of Change procedures, process hazardanalysis programs, and operatinginstructions cover all reaction pro-

cesses. Understand the maximum adiabatictemperature and pressure of all reac-tion processes, and make sure vesselpressure relief systems (including allequipment downstream of the emer-gency relief device) are adequate.

Good experience in the laboratorydoes not mean that a reaction can be safely run on a large scale. Youmust understand the reactionkinetics and thermodynamics, andthe heat removal capabilities of the

plant reactor. Gradual addition, or semi-batch processes are usually preferable to batch reactions for exothermic reac-tions, or reactions which producegaseous products. Gradual additionto a reactor, at a temperature highenough to ensure immediate or rapidreaction, limits the available chemi-cal energy in the reactorif something

goes wrong. If the process deviatesfrom intended conditions, the feedof the limiting reactant can bestopped, and the reaction will stop because the reactor contains little orno unreacted material.

INCIDENT 2 DECOMPOSITIONDURING A BLENDINGOPERATION

Process DescriptionThe process involved in this incidentrequired mixing two compounds andheating them in a mixing vessel tomake a solution. The exact chemicalsinvolved in the incident are not rele-vant to the major lesson that reactive

chemistry incidents are possible evenwhen the process is a simple physicalmixing operation with no intendedchemical reaction. The mixing vesselwas equipped for heating only, no cool-ing was available on the vessel jacket.The solution was normally fed from themixing vessel to a heat exchanger forcooling and nal product pack out, asshown in Figure 2a . No chemical reac-tions were intended in this process theproduct was a physical mixture of thetwo components. This was an existing

product and process that had been runtwenty years without issue.

Incident DescriptionDuring the incident, the product wasrun in a new mixing vessel. In the newvessel, the mixture was heated to atemperature 35 8 C higher than the mix-ing temperature in the old equipment.This was done so that the feed tempera-

ture where the solution entered thedownstream heat exchanger, whichwas further away, would be the same(Figure 2b ). At the higher mixing tem-perature an unexpected reactionoccurred, generated an exotherm, and

blew the rupture disk on the vessel.The rupture disk was designed for are case (material vapor pressure and boiling), not an exothermic reaction.Fortunately, the rupture disk was largeenough to prevent overpressure fromthe decomposition reaction, and thevessel did not rupture. The dischargedmaterial was carried up to 0.7 km awayfrom the plant, and deposited on plant buildings, neighboring buildings, and asoccer eld (which, fortunately, wasnot being used at the time).

Incident CausesThereaction was caused by thedecom-position of the two organic raw mate-rials in the vessel. The manufacturer sMaterial Safety Data Sheet (MSDS) forone of the materials indicated thatdecomposition of the compound couldoccur above 180 8 C. Although a Man-agement of Change process was fol-lowed before running the product onthe new equipment, decomposition orside reactions of the materials were not

considered an issue because the pro-cess temperature was signi cantly below the decomposition temperature.

After the incident, the reaction wasreproduced in a laboratory usingretained samples of the same rawmate-rials. An Accelerated Rate Calorimetry(ARC) test was run to determine thedecomposition temperature of the mix-ture. The more reactive of the two raw

Heat exchanger

Jacket - heating only

Relief valve dischargeto atmosphere

Mixingvessel

Water cooling

Product

Short distance, steamtraced pipe

Batch Temp. ~ 95 C

Heat exchanger

Jacket - heating only

Relief valve dischargeto atmosphere

Mixingvessel

Water cooling

Product

Longer distance,steam traced pipe

Batch Temp. ~ 130 C

(a) (b)

Figure 2. (a) Original mixing vessel conguration, (b) modied mixing vessel conguration.

12 Chemical Health & Safety, July/August 2004

8/12/2019 011 CH&S_July_2004

4/7

materials was also run separately in the

ARC test. The pure component decom-position temperature was about thesame as the temperature indicated inthe vendor s MSDS the vendor s infor-mation wasaccuratewith respect to thedecomposition temperature ( Figure 3 ).However, the mixture of componentsunderwent a reaction at a temperatureof about 130 8 C, or 50 8 C less than thedecompositiontemperatureof the morereactive component ( Figure 4 ). Thesolution was less stable than either of the pure components.

General LessonsWhen considering decomposition re-actions, a mixture may react or decom-pose at a lower temperature than any of the pure components. Stability mea-

surements on the actual solution pro-

vide the best information for deter-mining a safe operating temperature.Reactive chemistry incidents can

occur even when there is no inten-tional chemistry in the process. Thisprocess was a simple physical blend-ing operation, no chemical reactionwas intended.

INCIDENT 3 VESSELOVERPRESSURIZED DURINGNEUTRALIZATION

Process DescriptionThe process involved dissolving solidsodium hydroxide (NaOH) in deio-nized water, followed by neutraliza-tion with a strong aqueous acid to

form the desired product. In this case,theacid was methane sulfonic acid, buta similar incident could have occurredduring neutralization of any strongacid with a base. The neutralizationreaction occurred in a plastic vessel

with a loose tting top and no relief system. It was not rated for pressureand had no temperature control.

Incident DescriptionA power failure delayed production forseveral days and shut down the deio-nized (DI) water unit. When produc-tion resumed, there was urgentdemand for product due to theunscheduled downtime, and DI waterwas not available to dilute the NaOH.The production team decided to add

pure, undiluted NaOH directly to thevessel containing the aqueous acid.They thought that DI water would beavailable later, and the product could be diluted after the neutralization wascomplete. A sudden, severe exother-mic reaction occurred, boiling the mix-ture and the vapors overpressurizedand cracked the mixing vessel. Theoperators were sprayed with thereleased solution. Minor injuries wereincurred, but everyone involved wasable to return to work the same day.

Incident CausesThe more severe exothermic reactionwas a direct result of not diluting theNaOH with DI water. The productionteam failed to recognize the potential

Figure 3. ARC test results for the more reactive component of the mixture.

Figure 4. ARC test results for the two component mixture.


8/12/2019 011 CH&S_July_2004

5/7

hazard of the exothermic reaction withsmaller quantities of water present.Not knowing the reaction thermody-namics, their decision to proceed with-out DI water eliminated part of theheat sink for the heat of reaction and

mixing. The heat of neutralization andmixing of the acid and base was suf -cient to raise the mixture temperatureabove its boiling point without theadditional dilution water present.

Additionally, there was no Manage-ment of Change (MOC) process car-ried out. The production team did notfollow standard operating procedure(SOP) when they charged undilutedNaOH to the vessel. Also, the opera-tors were not wearing the appropriatePersonal Protective Equipment (PPE).

General Lessons

Make sure that personnel at ALLlevels have an awareness of thepotential reactive chemical hazards.This includes the chemists, engi-neers, operators, shift foremen andproduction supervisors, and anybody else involved with operations. Know the heat of reaction and themaximum adiabatic temperature forany reaction or exothermic dilution

process. Recognize the impact of changes inconcentration on maximum adiabatic temperature. If the presence of a diluent or solventis critical to maintaining tempera-ture control in an exothermic reac-tion process, consider using one ormore of the reactants already dis-solved in the diluent or solvent. Inthis case, aqueous NaOH is aninherently safer reactant, becausedilution with DI water is not neces-sary. Using aqueous reactant elimi-nates the dilution step and wouldhave avoided this incident.

Do not compromise process safetyor quality to meet short term produc-tion demands, or demands forincreased productivity. This culturemust exist at all levels in the organi-zation. In this incident, the opera-tors were trying to do a good job thecustomer wanted the productquickly and they thought they knewhow they could satisfy the need.They did not understand the impact

that their deviation from the SOPwould have on the batch tempera-ture, and they did not seek advice orapproval from technical manage-ment.

Strictly follow the SOP at all times.If

changes are needed, conduct MOCreviews, develop modified proce-dures, and train personnel on thenew procedure. Consult with theappropriate expertise about theseprocess changes make sure thatthe participants in the MOC reviewhave the necessary knowledge tounderstand the potential hazards.

Protect vessel integrity with a prop-erly designed overpressure relief sys-tem.

Wear proper PPE at all times during

the process.

HOW CAN REACTIVECHEMISTRY INCIDENTS SUCHAS THESE BE AVOIDED?There are a number of available toolsand resources which can help to avoidreactive chemistry incidents such asthose described.

Review of Material Safety DataSheets

A key followup action in Incident 2 wasto reviewall rawmaterial combinationsfor other undesirable reactive situa-tions. In order to quickly narrow downthe hundreds of raw material mixturesused in the plant, we reviewed the Sta- bility and Reactivity section of eachvendor s MSDS. In particular, we werelooking for chemical incompatibilitieswith other categories of materials (suchas acids, bases, oxidizing agents, redu-cing agents, and other material types)which might be present in formulationsusing the material being reviewed, orwhich might come into contact with thematerial through use of shared equip-ment or nearby storage.Decompositiontemperatures, if reported in the MSDS,were compared to maximum tempera-tures attainable in storage or proces-sing. In addition, the review of chemical mixtures by knowledgeablechemists who could identify potentialreactions was critical remember thatthe individual material MSD Sheetsfor the materials involved in the inci-dent did not suggest a reactivity pro-

blem. Using this information, theoriginal group of raw materials could be reduced to a smaller set of materialsand mixtures with potential reactivityhazards, which received a moredetailed review and testing.

Other information was used todetermine whether undesired reac-tions were possible with the reducedset of raw materials. Raw material ven-dors were contacted to determinewhether stability testing of their mate-rial had been performed. Processes andequipment which used the reactivematerials were reviewed, to determinelikelihood of a reaction. In situationswhere no information existed, stabilitytesting of theraw material or raw mate-rial mixture was performed to deter-

mine decomposition temperature of the compound or mixture. By analyz-ing the raw materials using existinginformation we avoided the time andexpense of testing hundreds of mixturecombinations in the laboratory.

Suspect Chemical StructuresMany chemical structures are knownto be particularly energetic or reactive.Certain functional groups indicateinstability or a tendency to polymerize.For example, many chemicals with

strained rings, double bonds, or het-erogeneous ring structures are knownto be unstable or reactive. Figure 5 listssome examples of highly reactive func-tional groups. This table is not a com-plete list of unstable or reactivefunctional groups and should only beused as an example of the types of materials which should be subject toadditional review. Processes usingmaterials with energetic structuresshould be thoroughly evaluated forpotential reactivity hazards, even if

there is no chemical reaction intendedin the process.

Literature Resources, HazardAnalysis Tools, and ComputerToolsThere are many published resourcesand tools to help in understandingand managing chemical reactivityhazards. Some particularly usefulsources are listed below.

Center for Chemical Process Safetypublications


8/12/2019 011 CH&S_July_2004

6/7

Safety Alert: Reactive MaterialHazards 4 (2001)

Guidelines for Safe Storage an dHandling of Reactive Materials 6

(1995) Guidelines for Chemical Reactiv-ity Evaluation and Application to Process Design 7 (1995) Essential Practices for Managin gChemical Reactivity Hazards 5

(2003) Bretherick s Hand book of ReactiveChemical Hazards 8 (1999)

Chemical Reactivity Worksheet ,9

U.S. National Oceanic and Atmo-spheric Administration, http://res-ponse.restoration.noaa.gov/che-maids/react.html

United Kingdom Health and SafetyExecutive, Designing and Operating

Safe Chemical Reaction Processes 10

(2000) Barton and Rogers, Chemical Rea c-tion Hazards: A Guide to Safety 11

(1997) Descriptions and examples on theapplication of process hazard analy-sis (PHA) tools such as HAZOP,interaction matrices, and checkliststo reacti ve chemistry hazard man-agement. 12,13

Hazard AwarenessThe three incidents detailed in thispaper all could have been preventedwith knowledge of the chemistry, pro-cess,and thepotential hazards. So whois responsible for knowing theseaspects of production? Everybody,from of ce management to engineers,

chemists, and plant operators. Differ-ent levels of understanding, however,are required at each level.

Managers must implement safetyprograms to maintain awareness or

chemical reactivity hazards, includ-ing everybody involved in manufac-turing engineers, chemists, andoperators. They must maintain asafety culture which includes rigor-ously following written operatingprocedures and management of change of any deviations.

Engineers and chemists are respon-sible for the chemistry and processunderstanding. They must look forsuspect chemical structures, andtrain operators to report any unex-

pected reaction behavior, such astemperature and pressure excur-sions. They must review the reac-tivity and incompatibility sectionsof all MSD Sheets to confirm thatthey are handling the material cor-rectly. If still unsure, they shouldsolicit the material s thermal stabi-lity data from the manufacturerand confirm the process conditionswill not cause reactivity issues, orconduct appropriate calorimetrytesting to obtain the needed infor-

mation. Operators and other production per-sonnel should know the steps tomaintain control of the reactionand the consequences of losing con-trol of the reaction. Operators mustfollow written procedures withoutdeviation, and be aware of thehazards of the materials they areusing. If the process involves reac-tive materials or operations, opera-tors should feel free to questionhandling procedures and seek safety

information from the engineers andchemists. They must report any unu-sual behavior of the process, parti-cularly regarding temperature andpressure control.

To summarize, safeguard proce-dures must be coordinated for allpotential hazards and open commu-nication between all levels of the orga-nization is crucial to the safety andsuccess of production. Reactive chem-istry hazard management is every-body s business .

Figure 5. Some potentially unstable or reactive chemical structures (compiledbased on information in References 6,8,10,11 ).

http://res-/http://ponse.restoration.noaa.gov/che-http://maids/react.htmlhttp://maids/react.htmlhttp://ponse.restoration.noaa.gov/che-http://res-/

8/12/2019 011 CH&S_July_2004

7/7

SUMMARYIncidents which have occurred in thepast are an important source of theknowledge required to prevent futureincidents. In this paper, we haveshared three reactive chemistry inci-

dents, and, more importantly, the gen-eral lessons from them which can beapplied to many chemical processingoperations. We have also shared sometools and techniques which we haveused to avoid this kind of incident inthe future. We hope that this informa-tion will be helpful to others in pre-venting reactive chemistry events, andencourage others to also share inci-dents in an effort to improve the over-all safety of the chemical industry.

AcknowledgementsThe authors acknowledge Dr.

Joseph Magee, Mr. Paul Morton, andMr. Alan Keiter for their work in con-ducting the calorimetry studies andemergency relief system modeling insupport of investigation of the inci-dents reported in this paper. We alsoacknowledge the work of many Rohmand Haas research, production, andplant support personnel who wereinvolved in the investigations. Their

contribution was essential to a com-plete understandingof what happened,

without which we would not have been able to share the lessons learned.

References1. Chemical Safety Case Study: How to

Prevent Runaway Reactions . UnitedStates Environmental ProtectionAgency, Chemical Emergency Prepa-redness and Prevention Of ce: Wa-shington, DC, August 1999 (EPA 550-F99-004).

2. Chemical Safety Case Study: Preven-tion of Reactive Chemical Explosions .United States Environmental Protec-tion Agency, Chemical EmergencyPreparedness and Prevention Of ce:Washington, DC, April 2000 (EPA550-F00-001).

3. Improving Reactive Hazard Manage-ment . United States Chemical Safetyand Hazard Investigation Board: Wa-shington, DC, October 2002 (ReportNo. 2001-01-H).

4. Center for Chemical Process Safety(CCPS). Safety Alert: Reactive Chem-istry Hazards . American Institute of Chemical Engineers: New York, Oc-tober 2001 ( www.aiche.org/ccps/safe-tyalerts ).

5. Johnson, R. W.; Rudy, S. W.; Unwin,S. D. Essential Practices for ManagingChemical Reactivity Hazards . Amer-ican Institute of Chemical Engineers:New York, 2003.

6. Guidelines for Safe Storage and Hand-ling of Reactive Materials . American

Institute of Chemical Engineers: NewYork, 1995.

7. Guidelines for Chemical ReactivityEvaluation and Application to Pro-cess Design . American Institute of Chemical Engineers: New York, 1995.

8. Urben, P. G. (Ed.). Bretherick sHandbook of Reactive Chemical Ha-zards . Butterworth-Heinemann: Bos-ton, MA, 1999.

9. Farr, J. K.; Freeman, W.; Odojewski, S.S. New program for chemical compat-ibility. Chem. Health Safety 1998 ,(6, November/December), 33 6.

10. United Kingdom Health and SafetyExecutive. Designing and OperatingSafe Chemical Reaction Processes .HSE Books: Sudbury, Suffolk, UK,2000.

11. Barton, J.; Rogers, R. Chemical Reac-tion Hazards: A Guide to Safety , 2nd

ed. Gulf Publishing Company: Hous-ton, TX, 1997.12. Mosley, D. W.; Ness, A. I.; Hender-

shot, D. C. Screen reactive chemicalhazards early in process development.Chem. Eng. Prog. 2000 , 96(11 (No-vember)), 51 63.

13. Hendershot, D. C. A checklist forinherently safer chemical reactionprocess design and operation; in 17th Annual International Conference andWorkshop: Risk, Reliability, and Se-curity , October 8 11, 2002, Jackson-ville, FL. American Institute of Chemical Engineers: New York, 2002,pp. 211 218.

http://www.aiche.org/ccps/safe-http://tyalerts/http://tyalerts/http://www.aiche.org/ccps/safe-

011 CH&S_July_2004

Documents

Transcript of 011 CH&S_July_2004