1

2 / Director’s Corner 3 / Recent Publications 4 / Precis of Excellence

5 / Dust Explosions 8 / ATEX and REACH 10 / LNG Safety 11 / Data Mining

12 / Isolation Technology 14 / Safety Alert 17 / Cont. Education 18 / Symposium

INSIDE

Vol. 8, No. 2SUMMER 2004

Mary KayO’ConnorProcessSafetyCenter

Chemical EngineeringDivision of the

Texas EngineeringExperiment Station

The Texas A&MUniversity System

Major General Hessto present Frank Lees LectureMajor General Kenneth W. Hess, member of the Columbia Accident

Investigation Board, will present the Frank P. Lees Memorial Lectureentitled "The Columbia Disaster, NASA Culture, and Lessons Learned," at the2004 Mary Kay O’Connor Process Safety Center Symposium.

In its foreword to the Columbia Accident Investigation Report, theBoard opined, “The organizational causes of this accident are rooted in theSpace Shuttle Program’s history and culture, including the original compromisesthat were required to gain approval for the Shuttle,subsequent years of resource constraints, fluctuatingpriorities, schedule pressures, mischaracterization ofthe Shuttle as operational rather than developmental,and lack of an agreed national vision for humanspace flight. Cultural traits and organizationalpractices detrimental to safety were allowed todevelop, including: reliance on past success as asubstitute for sound engineering practices (such astesting to understand why systems were notperforming in accordance with requirements);organizational barriers that prevented effective communication of critical safetyinformation and stifled professional differences of opinion; lack of integratedmanagement across program elements; and the evolution of an informal chain ofcommand and decision-making processes that operated outside theorganization’s rules.” Major General Hess’ keynote address will provide detailsof the Board’s findings and recommendations and how these lessons learnedcan be applied not only to the space shuttle program but to other high hazardprocesses as well.

Maj. Gen. Hess is also the Air Force Chief of Safety, HeadquartersU.S. Air Force, Washington, D.C., and Commander, Air Force Safety Center,Kirtland Air Force Base, N.M. The Lees Memorial Lecture will be held onWednesday, October 27, day two of the Symposium.

Ms. Kathleen Shaver, president of The Chlorine Institute, will presentthe keynote speech on the first day of the Symposium. Her lecture is entitled"Process Safety Challenges for the Chlorine Industry." Also presenting ingeneral session is Mr. Scott Berger, CCPS director. Papers will be presentedin many areas over the two day conference, including case histories, safetyculture, reactive chemicals, LNG, security, and management systems.

Maj. Gen. Kenneth W. Hess

2

Director’sCorner

The Alliance on Chemical Reactivity Hazards (CRH) is anagreement between the Occupational Safety and HealthAdministration, the U.S. Environmental ProtectionAgency, the American Chemistry Council, the Center forChemical Process Safety, the Chlorine Institute, theNational Association of Chemical Distributors, theSynthetic Organic Chemical Manufacturers Association,and the Mary Kay O’Connor Process Safety Center towork towards safer and more healthful Americanworkplaces and communities through better identificationand management of chemical reactivity hazards. Throughthis Alliance, the Signatories aim to (1) increase awarenessof the need to identify and manage CRH among thosewho manufacture, distribute, use and store chemicals; (2)provide chemical reactivity hazards managementinformation, methods and tools to a variety of audiences inmeaningful and useful forms to those audiences; and (3)gain experience in the use of methods and tools tocontinuously improve identification and management ofCRH.

The Alliance has started work in developingawareness and training programs, development of CRHtools, lessons learned and case history modules, anddevelopment of metrics to measure progress inidentification and management of CRH. A number ofactivities are going on and planned with regard to all theseareas, however; I feel it is pertinent to summarize theactivities of the Mary Kay O’Connor Process SafetyCenter in this regard. The Center is a strong supporter ofthe goals of the Alliance and is committed to conductingactivities to support all aspects of the Alliance goals. Inbrief, the Center is currently involved in the following CRHactivities:

The Center offers continuing education courseson CRH. These courses vary in content startingfrom management of CRH up to the fundamentalsof CRH and experimental and theoretical methodsfor CRH analysis. The courses are also offeredon-site and can be tailored to specificorganizational needs.

The Center conducts experimentalmeasurements related to CRH analysis. Wehave the capability to conduct both screeningmeasurements (e.g., reactive systems screeningtool) and advanced measurements (e.g.,automatic pressure tracking adiabaticcalorimeter). Results of these experimentalmeasurements are published in journals,proceedings, and conference presentations assoon as reasonably practicable. The Center is involved in development oftheoretical methods for CRH analysis. Thesemethods focus on the application of thermalanalysis techniques, computational chemistrymodels, and thermodynamic-energy relationshipsto estimate reactivity evaluation parameters. Thecombination of these techniques helps to minimizethe amount of needed experimental work andprovides the required parameters for evaluatingreactivity hazards and a more comprehensiveunderstanding of process chemistry. The Center also conducts incidentanalysis and develops lessons learned andcase histories from incidents that have alreadyoccurred. In this regard, the Center is currentlydeveloping brief incident analysis and lessonslearned synopsis for the 167 incidents reported inthe special report of the US Chemical Safety andHazard Investigation Board. The objective ofthese analyses is to provide an objective analysisof each incident based on known data andscientific information. However, moreimportantly, these analyses point out whatinformation should be known or collected tounderstand the CRH of the system underquestion. These analyses and other CRHinformation are available on the CRH web pageof the Center (http://process-safety.tamu.edu).

Continued on Page 3

3

RECENT PUBLICATIONS

n Cisneros, L.O., W.J. Rogers, M.S. Mannan, “Comparison of the thermal decomposition behaviorfor members of the hydroxylamine family,” Thermochimica Acta, vol. 414, no. 2, May 2004, pp.177-183.

n Triplett, T.L, Y. Zhou, and M.S. Mannan, “Application of Chain of Events Analysis to Process SafetyManagement,” Process Safety Progress, vol. 23, no. 2, June 2004, pp. 132-135.

n Krishna, K., W.J. Rogers, and M.S. Mannan, “Prediction of Aerosol Formation for Safe Utilizationof Industrial Fluids, Chemical Engineering Progress, vol. 100, no. 7, July 2004, pp. 25-28.

n Mannan, M.S., W.J. Rogers, Y.-S. Liu, S.R. Saraf, and A.A. Aldeeb, “Regulatory Initiatives in theUnited States With Regard to Reactive Chemicals,” Proceedings of the 11th InternationalSymposium on Loss Prevention and Safety Promotion in the Process Industries, Prague, CzechRepublic, May 31-June 3, 2004, pp. 1287-1294.

n Qiao, Y., M. Gentile, and M.S. Mannan, “Fuzzy Logic Methodology for Accident FrequencyAssessment in Hazardous Materials Transportation,” Proceedings of the CCPS 19th AnnualInternational Conference, Orlando, Florida, June 29-July 1, 2004, pp. 215-224.

The Annual Symposium of the Center dedicates several sessions to CRH issues.The topics covered range from CRH management, experimental and theoretical methods,structured approaches, lessons learned, and case histories. The Center also continues to conduct graduate research in the area of CRH. Inaddition to several researchers who have completed their PhD in CRH topics, five moreresearchers are currently involved in PhD work related to CRH topics. The results of theseresearch activities are disseminated through dissertations, publications, conferenceproceedings, presentations at conferences, newsletters, and personal interactions withCenter personnel.

In summary, the Center is conducting a number of activities that support the CRH Alliance goals.However, a lot more needs to be done and stakeholders need to work together to solve the CRH problemsand challenges.

M. Sam Mannan

Continued from Page 1

4

I graduated with a Bachelor degree in ChemicalEngineering at Zhejiang University, China in 1997. I joined theChemical Engineering Department at Texas A&M University asa PhD student in 1999. Since then, I have been working in thearea of quantitative risk analysis under the guidance of Dr. SamMannan.

As I conclude my graduate work, I still remember thenight that I arrived at Houston International Airport without knowingwhere to sleep. Looking back at my five years here, I have beenblessed to be part of the Mary Kay O’Connor Process SafetyCenter. The Center provides me opportunities to work withtalented people from all over the world. We bounce creative ideasback and forth in research work and help each other. Warm-hearted staff are always standing with you and ready to helpwhenever needed. Studying in such a unique and remarkable placeas the Center, I have been inspired to devote myself to processsafety.

Research:Computer-Aided Fault Tree Synthesis for Quantitative Risk Analysis in theChemical Process Industry

Along with the rapid progress of industrialization, the risk of incidents (such as fire, explosion,and chemical release) also is increasing. There has been growing public concern regarding the threat topeople and to environment from industrial activities, thus more rigorous regulations. The investigation ofalmost all the major incidents shows that we could have avoided those tragedies with an effective riskanalysis and safety management program. High-quality risk analysis is absolutely necessary forsustainable development. As a powerful and systematic tool, Fault Tree Analysis has been adapted to theparticular need of chemical process quantitative risk analysis and found great applications. However, theapplication of FTA in the chemical process industry has been limited. One major bottleneck is thetraditional manual synthesis of fault trees. Manual construction of fault trees is a difficult and time-consuming task, which requires specially-trained analysts who understand the methodology and thesystem under study as well. The quality of FTA is highly subjective and variable.

The availability of a computer-based analysis methodology will greatly increase the attractivenessof FTA and significantly benefit the CPI. The primary theme of my research is to develop a practical andefficient computer-aided fault tree synthesis methodology for quantitative risk analysis in the chemical processindustry. Ideally, the procedure of quantitative risk analyses can be standardized through a computer packageand guide decision makers toward more formal and more cost-effective solutions for the allocation ofresources for incident prevention and risk reduction.

Précis of Excellence: Yanjun Wang

5

Dust explosions present a substantial risk for anymanufacturing industry that produces powder fromcombustible solids consisting of chemicals that cancombine with oxygen. Even metals such as aluminum,which in bulk form appears noncombustible, can result indeadly explosions if finely divided in air and dispersed toan ignition source.

An Italian flourmill in 1785 may have been the firstfacility to document a dust explosion [2], but thedevelopment of the process industries has been consistentwith increasingly common dust explosions. The mostvulnerable industries to dust explosions include chemical,pharmaceutical, agricultural, metallurgical, and woodprocessing, because they include combustible feeds,intermediates, and products. Due to reportinginconsistencies, dust explosion incidents statistics areinaccurate, but several recent incidents have attractedattention to the devastating consequences of dustexplosions.

On February 20, 2003, a dust explosion and firecaused seven deaths at the CTA Acoustics plant inCorbin, Kentucky. This plant produced fiberglassinsulation using a resin dust, which had accumulated in aproduction area and could have been ignited by flamesfrom a malfunctioning oven. Also, on January 29, 2003, adust explosion and fire destroyed the WestPharmaceutical Services plant in Kinston, North Carolina,and resulted in six deaths, dozens of injuries, andhundreds of lost jobs. The dust hazard in this incident wasless noticeable, because combustible polyethylene used toproduce rubber products had accumulated above asuspended ceiling over a manufacturing area at the plant.An event puffed the dust into a cloud, which was ignitedby an energy source.

In a statement released on January 28, 2004, theU.S. Chemical Safety and Hazard Investigation Board(CSB) Chairman Carolyn Merritt stated, “The explosionat West Pharmaceuticals and a similar incident a fewweeks later in Corbin, Kentucky, raise safety questions ofnational significance. Our investigators have found thatboth disasters resulted from accumulations of combustibledust. Workers and workplaces need to be protected fromthis insidious hazard.” [3] Both events are still underinvestigation by the CSB, but it is clear that dust hazards

were not identified or understood sufficiently to avoid theincidents. Although the dust cloud formation mechanismand the ignition source of the Kinston, North Carolina,incident may not yet be known, an inherently saferapproach would have included methods to identify andavoid the dust accumulations that led to both incidents.

The CSB is also investigating another major U.S.dust incident that occurred on October 29, 2003, at theHayes Lemmerz plant in Huntington, Indiana, where aseries of explosions killed one worker, severely burnedanother worker, injured a third, and damaged property.The principal hazard was aluminum dust that hadaccumulated from the manufacture of cast aluminumautomotive wheels. A fourth major dust incidentoccurred at ICL Plastics and Stockline Plastics Ltd. inGlasgow, Scotland, on May 11, 2004, and caused 9deaths and 37 injuries. The exploding dust was fromchemical powders and fine spray from spray-paint guns,solvent vapors, and gas ovens with cracks in doors oropen doors through which dusts probably entered. Inthese cases also, following an inherently safer approachconsistent with good housekeeping would have avoidedthe hazardous dust accumulations that led to the incidents.

Dust CharacteristicsAs defined by the industrial dust standards of

NFPA 654, dust refers to particles with diameters below420 m. Compared to vapor cloud explosions (VCE),analysis of dust explosion incidents is more involved dueto the complexity of dust behavior leading to combustion.Dust explosions have traditionally received less attentionthan VCE, because they are less frequent and occur onlyin certain types of process plants. Compared to a VCE,however, a dust mixture with air can contain more fueland explosive energy than the same volume of gasmixture.

Explosions require three components: a fuel, anoxidizer, and ignition energy, which combination isreferred to as the fire triangle. This triangular structuresymbolizes the source of prevention and protectionstrategies, because removing or diminishing any of thethree components reduces the risk of explosions. ForVCE, the fuel is the vapor, and for combustible dusts thefuel also is vapor formed from the dust when heated by

Dust Deposits Lead toDust Cloud Explosions

6

an ignition source. Following formation, the vapor mixeswith air to produce a combustible mixture.

The equilibrium state of a gas is determined by itschemical composition, temperature, and pressure. For adust cloud, however, the state is dynamic with particles inmotion and suspended in air. Also, the dust must havesufficiently low moisture content to combust. Therefore astructure to symbolize prevention strategies for dusts canbe a pentagon to represent the original three elements offuel, air, and ignition energy together with these twoadditional elements with reductions of risk consistent withdust not airborne and with increased moisture content.

The suspended dust particles must also be in acombustible concentration range with air for flamepropagation beyond the ignition source, so sufficientoxygen levels in air provide the oxidizer for ordinarycombustible dusts. Although ignition sensitivity andexplosibility is each dependant on particle size and area,dispersibility of the particles determines effective particlesize due to agglomeration. So agents that decreasedispersibility and enhance particle agglomeration, such asincreased moisture content, can greatly decrease the riskof explosion.

The severity level of an explosion is determined bythe flame propagation development. Dust explosionsconsist of flame fronts that are limited by mass andthermal transport, so they are usually deflagrations withpropagation rates that do not exceed the speed of sound.The potential damage caused by a dust explosion can besimilar to damage from a VCE, except that multiple dustexplosions can occur with the first explosion makingairborne and distributing additional dust that explodes andproduces a pressure wave to generate additional dustclouds by entrainment of accumulated dust and dispersionin air. Sequential explosions can continue until the dust inthe propagation path is consumed.

An ABS (acrylontrile-butadiene-styrene) plantincident in Taiwan exhibited the multiple explosionphenomena in 1997. [4] Six of eight connected silosholding the plastic powder were likely ignited to explosionby bulk electrostatic discharges between the compactedpowders and the silo walls. The primary explosionoccurred in the first silo, and the flame propagated toproduce explosions in each of the following five silos.

Ignition sourcesIgnition sources consist of various forms of energy

and are sometimes not easily identified, but theunderstanding of potential ignition sources is necessary toreduce explosion risk. Open flames are the most obvious

ignition source and exist where there is welding or cutting.Open flame ignition examples include the Corbin,Kentucky, and the Glasgow, Scotland, incidentsdiscussed above. Hot surfaces, such as heaters, lightbulbs, and overheated bearings, often can supplysufficient energy to ignite dust. Drying processes involvea substantial number of heated dryer wall surfaces forwhich inherently safer methods include lower dryingtemperatures and reduced time at higher temperatures tominimize fire and explosion risk.

Heat from mechanical impact also can provide theenergy from electric sparks to initiate explosion.Electrical and electrostatic discharges are perhaps themost difficult sources to eliminate and are of varioustypes. Smoldering and burning dust deposits demonstratemultiple hazards, because the dust can become airborneand be self-ignited or it can contact another cloud andprovide the ignition for an explosion.

MKOPSC Dust ResearchA proliferation of finer solid powders in industry

has caused an increase in dust explosion risks, but thecombustion behavior of most powders or dustscommonly occurring in industry have been insufficientlytested. As indicated by the U.S. Chemical Safety Boardin regard to the June 18, 2003, dust explosion inKingston, NC, there is a critical need for research tocharacterize the explosion behavior of dusts to makepossible strategies for handling dusts safely and tominimize the possibility and severity and therefore the riskof industrial dust explosions. The MKOPSC is dedicatedto current issues and to research the fundamental factorsthat characterize dust explosion behavior together withrisk assessment for plants that handle or producecombustible solids.

Although dust explosion tests can involve a widerange of scales, a 1.2 liter Hartmann vessel and a 20 literexplosion test chamber with a detonator, mixer,temperature sensors and pressure transducers, are usefulfor many types of measurements to include maximumexplosion pressures, maximum rates of pressure rise,minimum explosion dust concentrations, minimum ignitionenergies, effects of inert components, minimum oxygenconcentrations, and suppression effects.

Reproducible control of the dispersion will ensurephysical and chemical homogeneity of the dust particlesand uniform dispersion of the dust throughout the testapparatus. The use of highly characterized and narrowparticle size and specific area distributions in air will makepossible development of reliable correlations between

7

explosion characteristics, including maximum pressuresand maximum rates of pressure rise, and dust properties.

Models of explosion blast waves are needed forsafer designs of facilities and protection of personnel.Characterization of blast waves and their dependence ondust cloud properties can help predict the consequencesof an incident. The initial state of a cloud has a significanteffect on the ignition sensitivity and burning rate of a dustcloud. Therefore, understanding the transformation of dustlayers and deposits into dust clouds from blast waves willallow development of more effective safeguards fromsecondary explosions. For experimental studies tovalidate model predictions, it is important to study and testsystems and configurations that relate to industrialpractices.

A major goal of tests is explosibility. For dustscreening tests to conclude whether a particular mixture isexplosive, the limiting dust composition for flamepropagation is determined using characterized dust withsize below 75 m consistent with the ASTM standard.This determination yields also the minimum explosive dustconcentration for the material and the limiting maximumpermissible oxygen concentration for inerting procedures.

A second major goal is the dynamic combustionbehavior. A more turbulent dust cloud produces morehomogeneous dust concentrations in air in which the flamefront moves more quickly and a more robust explosion isgenerated. For a model to predict the dynamic state of acloud and its combustion rate, an understanding of thebasic microscopic turbulence mechanisms that promotecombustion is needed. [5] An important application ofthis information is design of vents for pressure relief andautomatic suppression systems.

ConclusionsThe CSB is currently investigating the number and

severity of U.S. dust explosions during the past fewdecades and is examining whether the hazards ofcombustible dust have been adequately identified andcontrolled through codes, standards, and good operatingpractices. Stephen Selk, the CSB lead investigator in theWest Pharmaceuticals incident, asked “How much dustaccumulation can be tolerated before it becomeshazardous? And if it is hazardous, what are theappropriate controls?”[3] With the release of CSB reportson these incidents, the pressure for more stringent hazardassessment programs and discussion of additionalregulations is likely to increase, so a more comprehensiveunderstanding of dust behavior is urgently needed.Proposed research will relate explosion behavior tofundamental parameters that can be scaled from laboratorytests to industrial scale. The necessary research,information, and modeling to meet industrial needs will beaccomplished through collaboration with industry.

References1. BBC News, http://news.bbc.co.uk/2/hi/

in_pictures/3704367.stm2. George Williamson 2002, http://

www.chemeng.ed.ac.uk/~emju49/SP2001/webpage/

3. CSB News Release, January 28, 2004, http://www.csb.gov/news_releases/

4. Kao, Chen-Shan and Yih-Shing Duh, “Accidentinvestigation of an ABS plant,” Journal of LossPrevention in the Process Industries, 15 (1), 223-232, 2002

5. Eckoff, R.K., “Dust Explosions in the ProcessIndustries,” 3rd ed., Elsevier, 2003

Root Cause:Poor Management of Change

A poor Management of Change (MOC) program is frequently found as the root cause of incidents in theprocess industries. What is wrong with these programs? What are the features that these programs lack that leadto incidents?

The Center will review a large number of incident investigation reports in order to characterize the mostcommon elements in MOC programs that lead to incidents. Information from the reports will be compared to thefundamental structure of MOC programs. The results of this study will create a baseline and an opportunity foran open dialog on MOC programs.

Dust, Continued

8

ATEX and REACH: European Regulatory InitiativesTwo recent regulatory initiatives by the European

Commission may provide a clue of things to come. Inthe global economy combined with the expandedmembership of the European Union, these newinitiatives will likely be felt worldwide and the waveswill slowly but surely reach across the Atlantic and thePacific. The two recent European initiatives discussedhere are ATEX and REACH.

“Atmosphères Explosibles” (ATEX) guidelinesprovide details and directions on implementation ofEuropean Parliament and Council Directive 94/9/ECconcerning equipment and protective systems intendedfor use in potentially explosive atmospheres. Anexplosive atmosphere for the purposes of this standardis defined as a mixture:

i) of flammable substances in the form ofgases,vapors, mists, or dusts;

ii) with air;

iii) under atmospheric conditions;

iv) in which, after ignition, the combustionspreads to the entire unburned mixture

An atmosphere, which could become explosivedue to local and/or operational conditions, is called apotentially explosive atmosphere. It is onlyequipment and products intended for use in this kind ofpotentially explosive atmosphere, which are covered bydirective 94/9/EC and ATEX. It is important to notethat products are not covered by this directive whenthey are intended for use in relation to atmospheres thatmight potentially be explosive, but one or more of thedefining elements i) to iv) above are not present. To bewithin the scope of the directive, a product has to be:

a) equipment; or

b) a protective system; or

c) a component; or

d) a safety, controlling, or regulating device.

In order to determine the appropriate conformityassessment procedure, the next step is to decide basedon the intended use, as to which Group and Categorythe product belongs. Group I comprises equipmentintended for use in underground parts of mines, and to

those parts of surface installations of such mines, likelyto become endangered by firedamp and/or combustibledust. Group II comprises equipment intended for usein other places likely to become endangered byexplosive atmospheres. The groups are againsubdivided into categories. Group I has twocategories, M1 and M2. Products in category M1 arerequired to remain functional for safety reasons whenan explosive atmosphere is present and is characterizedby integrated explosion protection measures functioningin such a way that:

• in the event of failure of one integratedmeasure, at least a second means ofprotection provides for a sufficient level ofsafety; or

• in the event of two faults occurringindependently of each other, a sufficient levelof safety is ensured

Products in category M2 are intended to be de-energized in the event of an explosive atmosphere.

As mentioned earlier, Group II comprisesequipment intended for use in other places likely tobecome endangered by explosive atmospheres. Thisgroup would thus cover most of the process industry.Group II has three categories, 1, 2, and 3. Category 1comprises products designed to be capable ofremaining within its operational parameters, stated bythe manufacturer, and ensuring a very high level ofprotection, for its intended use in areas in whichexplosive atmospheres caused by mixtures of air andgases, vapors, mists, or air/dust mixtures are highlylikely to occur and are present continuously, for longperiods of time or frequently. Equipment in thiscategory are characterized by:

• in the event of failure of one integratedmeasure, at least a second independent meansof protection provides for a sufficient level ofsafety; or

• in the event of two faults occurringindependently of each other, a sufficient level ofsafety is ensured

Category 2 comprises products designed to becapable of remaining within its operational parameters,stated by the manufacturer, and based on a very high

9

level of protection, for its intended use in areas inwhich explosive atmospheres caused by mixtures of airand gases, vapors, mists, or air/dust mixtures arelikely to occur. The explosion protection relating tothis category must function in such a way as to providea sufficient level of safety even in the event ofequipment with operating faults or in dangerousoperating conditions that normally have to be takeninto account.

Category 3 comprises products designed to becapable of remaining within its operational parameters,stated by the manufacturer, and based on a normallevel of protection, for its intended use, consideringareas in which explosive atmospheres caused bymixtures of air and gases, vapors, mists, or air/dustmixtures are less likely to occur and if they do occur,do so infrequently and for a short period of time only.The design of products of this category must provide asufficient level of safety during normal operation.

To meet the ATEX requirements, it is essential toconduct a risk assessment. In principle, the riskassessment consists of four steps: hazardidentification, risk estimation, risk evaluation, and riskreduction options analysis. Article 8 of the directivedescribes the procedures whereby the manufacturerensures and declares that the product is in compliancewith the directive.

Registration, Evaluation, Authorization, andRestrictions of Chemicals (REACH) is a proposedregulation now making its way through the EuropeanParliament and Council. The proposed REACHregulation will replace some 40 existing EuropeanUnion directives and will be applied homogeneouslythrough the newly expanded European Unioncomprising 25 Member states. The proposalestablishes the REACH system and creates aEuropean Chemicals Agency to administer theregulation. The REACH regulation will apply tomanufacturers and importers within the community whomanufacture or import a substance in quantities startingat one tonne (2,240 pounds) per year. In a nutshell,REACH consists of the following elements:

Registration requires industry to obtainrelevant information on their substances and touse that data to manage them safely. Failure toregister means that the substance cannot bemanufactured or imported at quantities aboveone tonne (2,240 pounds) per year. The

registration provisions oblige manufacturers andimporters of substances to obtain (wherenecessary by performing appropriate tests)knowledge on the substances they manufactureor import and to use this knowledge to ensureresponsible and well-informed management ofthe risks that the substances may present. Theinformation required is modulated by tonnage,since this gives an indication of the potential forexposure, with a requirement for a chemicalsafety assessment that documents the choice ofrisk management measures for tonnages above10 tonnes (22,400 pounds) per year. Forpurposes of enforcement as well as for reasonsof transparency, the technical dossier containingthe registration information (including thechemical safety assessment when appropriate)is to be submitted to the authorities. There areprovisions on generation of information, whichaim to ensure that it is of acceptable quality.Currently, polymers are exempted from therequirement to register. If a downstream useris using a substance in a way not covered by amanufacturer’s or importer’s chemical safetyassessment or if he intends to use different riskmanagement measures, then he must send ashort report to the Agency.

Evaluation provides confidence thatindustry is meeting its obligations and consistsof a dossier evaluation and a substanceevaluation. The dossier evaluation is intendedto prevent unnecessary animal testing and giveauthorities the task of checking compliance ofregistration dossiers with the registrationrequirements. Substance evaluation providesa mechanism for an authority to require industryto obtain and submit more information in caseof suspicion of a risk to human health or theenvironment. Evaluation may lead authoritiesto the conclusion that action should be takenunder the restrictions or authorizationsprocedures in REACH.

The substances selected for theAuthorization system have hazardousproperties of such high concern that it is essentialto regulate them through a mechanism thatensures the risks related to their use areassessed, weighed, and then decided upon by

10

LNG SafetyLNG safety is back in the headlines as over 30 LNG

importation terminals have recently been proposed to meetthe projected USA natural gas demand. Controversy oversiting LNG facilities has focused attention on LNG safetyissues, particularly the potential impact of large fires onadjacent areas.

Participation in LNG projects over the past twentyyears sparked the interest in process safety for severalMKOPSC staff. Mike O’Connor, Sam Mannan andHarry West have a long history in LNG research,education and specific project experience, resulting in acontinuing involvement in LNG safety issues.

Foremost in the MKOPSC LNG safety activity is thecontinued expansion of the LNG safety library. The libraryof Professor Cedomir M. Sliepcevich, renowned as thefather of modern LNG technology with his work in the1950s and 1960s as the research director of ConchMethane, was transferred from the University ofOklahoma to MKOPSC. Several other LNG pioneerscontinue to donate resources to the LNG safety library. Inaddition to its value as a student resource, many industrialgroups have sought information from the library. TheLNG library (both books and reports as well as individualpapers and literature) has been cataloged and can beviewed or searched via the web at: http://process-safety.tamu.edu/publications/library.htm.

Education of LNG safety is a hallmark of MKOPSC.In addition to the LNG safety continuing education course(offered in the Far East in recent years), MKOPSC isdeveloping a two-week in-depth course on LNG safetyand technology for the new TAMU Doha Qatar campus.

Research projects underway or under considerationinclude:

• Review of LNG-Water explosion (also calledrapid phase transitions) experiments andpredictive models

• Implication of scaling laws in consequence modelson extrapolation of experimental data to predictvery large LNG release scenarios.

• Pool spreading models

• Aerosol formation

• Pool fire size correlations

• Alarm management issues

• Wireless ship to shore ESD technologies

• Update of LNG fire suppression data

• In co-operation with the TAMU FireSchool

the community prior to their actual use. Risks associated with uses of substances with properties of veryhigh concern will be reviewed, and if they are adequately controlled, or if socio-economic benefits outweighthe risks and there are no suitable alternative substitute substances or technologies, the uses will begranted an authorization. The authorization provisions require those using or making available substanceswith properties of high concern to apply for an authorization for each use within deadlines set by theCommission. The burden of proof is placed on the applicant to demonstrate that the risk from the use isadequately controlled or that the socio-economic benefits outweigh the risks. Downstream users mayuse a substance for an authorized use provided they obtain the substance from a company for whom anauthorization has been granted and that they keep within the conditions of that authorization. Thedownstream users will have to, however, inform the Agency of the authorized use so that the authoritiesare fully aware of how and where substances of high concern are being used.

The Restrictions procedure provides a safety net to manage risks that have not been adequatelyaddressed by another part of the REACH system. Proposals for restrictions may consist of conditionsfor the manufacture, use(s), and/or placing on the market of a substance or of the prohibition of theseactivities, if necessary. The restrictions provisions: address the need to ensure that action is taken whererequired as quickly as possible, provide a scientific basis for any restrictions, and enable all interestedparties to participate in the procedure.

European, Continued

11

Harnessing Data Mining to ExploreIncident Databases

Data mining is a technology used intensively for creating higher-value fromotherwise ill-prepared and ill-organized data sets. Data mining consists of apowerful set of tools that work with very large databases. Recently, data mininghas been used to support research in medicine as well.

Since many entities collect data on incidents, the cumulative number ofrecords may be in the magnitude of 107. Therefore, these sources are candidatesfor analyses by data mining. The Center used a subset of the National ResponseCenter incident database in order to identify the usefulness of data mining tools inexploring these types of datasets. The analysis of this subset aimed at identifyingrelationships between types of equipment and the chemical involved in theincidents. This analysis reveals the level of elevation (the Lift) in the probability ofincidents with a combination of chemical-equipment, compared to the generalprobability of incident of the specific equipment only. Since the lift valuerepresents an elevation in probability of an equipment-related incident withrespect to a specific chemical, multiplying the average failure rates by the liftvalue will produce an annual failure probability that represents the chemical in theprocess as well.

The efforts following this study will be devoted to establishing a decision-support platform that uses results from data mining of incident databases.

New Web Resourcesfor Members

The Center has implemented several new features in the “Members Area” of theprocess safety web site, located at http://process-safety.tamu.edu.

One of the new elements is a pop-up window that appears upon entering the area.This window changes with each entry and contains current news and events in processsafety. This change provides Mary Kay O’Connor Process Safety Center consortiummembers with a view of the latest happenings at the Center and in industry.

The Members Area also features an “Ask the Experts” field. This section contains anon-line form that when submitted, is received at a centralized location and then directed tothe relevant expert for quick response via email.

The final update to the Members Area is the addition of the Symposium Proceedings,available on-line in a read only format. The Center believes providing the Proceedings,along with other Center publications, is a valuable information resource for its members.

The new features are meant to provide members with even more access to the Centerthrough education, research, and expertise.

12

Though it has been the topic of discussion formany years, process isolations continue to be aleading cause of incidents in industry today.Notwithstanding the many workshops, seminars,safety meetings and job hazard analyses dedicated tothe task, we must still strive to improve ourprocedures when it comes to isolating piping systemsand equipment for hot work or for confined spaceentry. As a safety community, we must be resolute inour mission to control the risks associated with thisactivity. Recent developments in isolation technologywill better equip us to manage risk and eliminatematerial and human loss.

In late December of 2002, a welder in Alaskawas decapitated by a projected plumber’s plug. Theblown-out plug struck the welder and killed himinstantly. Two other workers suffered injuries(broken wrist, cuts and bruises) because of theincident. In this case, water condensed and frozeinside the nitrogen vent line, blocking the ventingcapabilities of the plug during the application.Similarly, a welder at a gulf coast refinery wasrecently impaled when a 20-inch plumber’s plug blewout of a line and the vent pipe struck him in thestomach and penetrated his body until it protrudedthrough his backside. Luckily, this craftsman lived.However, his injuries seriously inhibited his quality oflife thereafter.

Another current incident involved a failed seal ina large-diameter line isolation within a different gulfcoast refinery. During welding, the rubber seal of theplug was disturbed by either the heating of the pipefrom welding or from the heat of the sun’s rays. Theplug leaked and leaching vapors and gases wickedthrough this breach of seal. The welder was engulfedin flames immediately.

Even people within the field of safety have beenon the receiving end of a plugging occurrence. Alatest incident involves a welder noticing movementof a plug while performing his work on a platformthat was 80 feet above ground level. The welderclimbed down the scaffold to locate a safetyprofessional in order to receive counsel on the

Recent Advances inIsolation Technology and Procedures

circumstances. He and the safety representativereturned to the scaffold platform and the safetyexpert positioned himself in front of the plug andstooped down in order to look into the pipe at theplug in question. As he did so, the plug displaceditself, striking him with incredible force in theshoulder and chest. The force from the blow threwhim into the railings of the platform. The only thingthat kept him from falling 80 feet to his death was theretention of the scaffold railings. He suffered injuriesthat remained with him for quite some time.

An official safety memo dated August 23, 1990and issued by the Minerals Management Service, adivision of the U.S. Department of Interior, alertedindustry of an incident where a welder suffered third-degree burns to his face, hands and ears when theseal of a plug failed. His cutting torch ignited thevapors that were leaking from within the line. Thememo calls upon industry to look at better isolationprocedures.

In each of these incidents, there was no warningto the workers. However, the element of surprisewas, at the very least, alarming. Moreover, in theextreme case, it was fatal. There are countless otherstories that remain within the fence line of facilitiesacross the world, mostly unspoken of and archivedin memories hopefully to be forgotten.

Safety personnel within many facilities are stillencumbered with the challenge of trying to changethe traditions of tradespeople and their respectiveorganizations, as well as other departments chargedwith the responsibility of field operations: it is “hardto teach old dogs new tricks.” Sadly, somedecisions are based primarily upon budgetaryconcerns.

Many devices have been developed over time inorder to provide what any one individual deems as asafe isolation tool. From rubber tires framing acircular steel plate to various creations of rags, mud,sand, clay and just plain dirt, innovations haveflowed freely when dealing with this quandary. Morecommonplace is the current trend to freeze water ina line with nitrogen or to use plumber’s plugs, sewer

13

component to the tool. A vent line, coupled with agauge and 50-100 foot vent hose, ensure that anyupstream vapors, gases, product, etc… are ventedfar away from any ignition source, while enablingupstream pressure to be monitored. An increase ofpressure upstream of the tool will alarm the attendingcertified technician and the call to action is again tostop hot work and evacuate the area until the matteris addressed – an incident intercepted or a lifesaved.

The ability to monitor both the seal of the tooland the upstream pressure on the tool during hotwork, coupled with the ability to vent potentiallyvolatile vapors or product far away from theirignition source is what makes this technology mostcompelling. Historically, industry has engaged manypractices in order to react to or prepare one for anincident (i.e., fire watch, hole watch, safety glasses,hard hats, 100% tie-off & harnesses, FRC’s). Ineach of these measures, the action is reactive,recognizing that an incident has already occurred.What sets this technology apart is that it prevents theincident from occurring in the first place.

This same technology is versatile enough to beapplied to challenging applications, for instance,where a nitrogen purge must be initiated andmaintained on a line or on equipment or where anisolation must be preserved during heat treatment,pre-heat or bake-out procedures (wheretemperatures may exceed 1400°F). The tools can beused to isolate within and through elbows and tees,an application that is impossible with the variousalternative apparatus. In addition, this isolationtechnology can be used for confined space entry invarious situations.

The battle to conquer unsafe isolationprocedures is not over. Circulating knowledge ofgenerally accepted practices and best practicesamongst respected peers is tedious. The wearisomenature of promoting any worthwhile cause oftenimpedes momentum. However, information sharingand an open mind on the receiving end is a suremethodology.

More information about this technology isavailable at http://www.carbertesting.com or bycontacting CAR-BER TESTING SERVICES at(800) 592-8378.

plugs, high-pressure test plugs, flange testers andinflatable bladders in desperate search of the idealisolation mechanism.

Using these technologies for isolations inpetrochemical process systems is like using a bungeecord as a safety lanyard - one may not bounce back!Welders and others that routinely work around suchisolation measures certainly relate to what is commonlyreferred to in industry as the “pucker factor.” The risksinherent in these solutions are two-fold: (1) there is nomechanism built into these devices whereby the sealsof any of them can be monitored and continuallyverified during hot work, and (2) proper ventingprocedures are seldom adhered to when using them. Itis an oversight of one or both of these principles thatcan be considered the sole cause of each of theheretofore-mentioned incidents.

A more recent development delivers a more highlydeveloped way to perform isolations; it provides a100-percent guaranteed vapor-free work environment.This isolation tool providesdouble o-ring seals and addsa third positive vapor-barrierseal in between the o-rings.This third seal is establishedby pressurizing medium(water, peanut oil, glycol,nitrogen, etc…) betweenand against the two o-ringseals and against the walls oftheir tool and the pipe orequipment. By holding aconstant pressure in thisliquid or gas-filled vapor-barrier, the isolation’s sealcan be continually monitoredby viewing a pressure gauge,thus conquering the proverbial problem of beingunable to “sniff and weld at the same time.” In theevent of the loss of one of the two o-ring seals, thepressure gauge will immediately alarm a certifiedtechnician of the circumstances because of its loss ofpressure. Upon such, the hot work will be shut downimmediately and the area will be evacuated until thecertified technician reestablishes isolation.

In addition to a guaranteed positive seal, thisleading-edge technology also includes a venting

14

Shell Safety Newsletter, Dated 19th January 2004, Issue No. 04 #01. We would like to share withyou the experience on a Hydrotreater unit that suffered failure of an advanced back-flow protection system inthe feed pump discharge. This resulted in a serious threat to personal safety and the loss of the unit for anextended period of time due to severe equipment damage.

The Incident

A middle distillate hydrotreater charge pump was stopped forstrainer cleaning. After the pump came to a standstill, the operatorheard a loud humming sound due to pump back spinning. Hot, high-pressure liquid and hydrogen from the reactor circuit was flowingbackward through the advanced backflow protecting system andthrough the pump, pressurising the upstream feed vessel. Thepressure in the feed vessel rose quickly from 7 barg to the designpressure of 10 barg (RV lifted). The operator stopped the backflow,by closing the remote operated valve present in the pump suction.This quick and good operator intervention prevented the feed vesselfrom being over pressurised to far above the vessel design pressure and possible catastrophic vessel failure.The consequence of the backflow was limited to damages to the pump and motor. The coupling between thepump and motor broke and was hurled away some 40 meters - the photo shows the bent shaft. Hydrocarbonsleaking from the severely damaged pump seal ignited and caused a fire. The unit had to be shut down anddepressurised. Luckily nobody was injured.

Line-up

SAFETY ALERT:Failure of the Backflow Protection System in a Hydrotreater

10% OPEN

100% STUCK OPEN PASSING.

Activated +2s Closed +4s

OK

50% OPEN

50% STUCK OPEN

OK CLOSED after 3 min

TO E301/ R301

H2 GAS 55 barg

TCV

FCV

P301B

TO E302/ R301

ROV Strainer

Feed Vessel

ESD

FRESH H2

TO HLPS

MID DIST Feed

A-pump

Legend : Pump suction side not in isometric

TO E301/ R301

H2 GAS 55 barg

TCV

FCV

P301B

TO E302/ R301

ROV Strainer

FRESH H2

TO TO E301/ R301

H2 GAS 55 barg

TCV

FCV

B-pump

TO E302/ R301

ESD

ROV Strainer

FRESH H2

T

Legend : Pump suction side not in isometric

15

The schematic figure above shows the line-up and the valve positions at the time of the backflow incident.Investigations revealed that the back-flow protection system was not effective because:w The non return valve (NRV) at the pump discharge was found stuck and 50% openw The pump discharge block valve was in the process of being closed and was still about 50% open.w The butterfly type emergency shut down (ESD) valve, tightness class IV, was in the closed position. When

tested, the valve was found to be non-tight and installed in the wrong flow direction.w The second NRV (downstream of the ESD valve) was found stuck and 100% openw The flow control valve (FCV) was 10 % open.The pump feed strainer, the NRVs and the ESD valve had all been severely affected by corrosion productsoriginating from the feed vessel. The FCV was kept open by an aberration in the DCS system.

Causes:

1. A Plant ChangeThe first step leading towards this incident was taken some 7 years before, when a plant change was initiatedand implemented to install an equalising line between the Hot Low Pressure Separator (HLPS) and the feedvessel. The carbon steel feed vessel had been designed for a normal operating temperature of 237°C and a lowH2S concentration in the feed. With the implementation of the plant change, gas with a high H2S content wasintroduced into the feed vessel, which was being operated at a higher temperature (250-260°C). The newoperating conditions caused heavy FeS scaling in the feed vessel.

Some 5 years later another plant change was raised to prevent FeS formation. Fresh gas (H2) was allowed intothe feed vessel, to prevent the high H2S containing gas from the HLPS going to the feed vessel. However, duringplant upsets the HLPS pressure was higher than the feed vessel and high H2S containing gas still entered thefeed vessel resulting in further FeS scaling. The scaling was dislodged from the vessel during shutdown andstart-up, causing fouling of all elements of the reactor feed system such as pump strainer, pump sealing systemand elements of the back flow protection system.

2. Testing of the Backflow Protection System

In a backflow protection systemthe ESD valve and FCV aretripped close, activated by eitherthe PdZA or the FZA initiator(see the figure). The design oftwo independent initiators,activating two final elements asbackflow protection, should besufficient to prevent back-flow.The back-flow protectionprotects the feed vessel againstthe potential of severeoverpressure, and the feedvessel relief system against toohigh temperatures and anexcessive amount of hydrogenfrom the reactor. Therefore theprotection has an IPFsafeguarding integrity level of

16

SIL-3. The NRV has been added to reduce the probability of backflow in general and especially during start-upof the pump. When the ESD valve is open and the FCV is opened in the start sequence, the FZA is overridden toallow the pump to build up flow.

In order to maintain the high, established SIL of the back flow protection, a full system test is typically requiredevery 3 months. The system test shall be thoroughly carried out to achieve a test coverage factor as close as one tocover all the dangerous failures possibly present in the IPF. The non-return valve downstream of the ESD valve ispart of the safeguarding system and should also be checked regularly (at least every shutdown).The aberration in the DCS system that kept the FCV 10 % open should have been detected as part of the IPFtesting.

Learning points:

The learning points from this incident have been summarised below:

1. Despite the correct basic design, backflow may occur and threaten unit safety if the system has not beenproperly engineered, installed, or is not tested according to the design frequency.

2. All the elements installed to avoid backflow in this backflow protection system failed due to a mechanismthat nobody identified in a plant change approval round.

When the content of a proposed plant change is judged, it should be ensured that the selected basic designprinciples taken into account in the original design are reviewed again. There is an inherent danger thatthe originally built-in design features could be ignored or not recognised, resulting in unexpected failurescenarios like this common mode failure due to corrosion products.

In this particular case the original plant change should not have been implemented.

3. Although specified as a Tight Shut Off type, the ESD butterfly valve installed could never be tight the wayit was installed. The ESD valve had been installed in the normal flow direction, while for this application itshould have been reversed. Correct installation of valves needs to be confirmed every time they areremoved for maintenance reasons.

4. Regular testing of the IPF is required. Only by doing so, the required SIL can be maintained. Testing ofthis IPF may be cumbersome, since the ESD valve and NRV may have to be taken out for tightnesstesting. These operational and financial consequences of the mandatory testing should be taken intoconsideration during the design phase and may result in additional hardware provisions enabling therequired tests at minimum cost impact.

The contents of this newsletter represent Shell Global Solutions International and Shell International Chemicals bestprofessional judgment of the matters dealt with. However, it is offered for information only and should not be relied uponas authoritative guidance in any particular situation. Recipients of this newsletter should seek advice from their owntechnical advisers and the vendors of their specific equipment. Shell Global Solutions International and Shell InternationalChemicals accept no liability whatsoever for any loss or damage arising out of or in connection with the contents of thisnewsletter, no matter how it arises and even if it is wholly or partly caused by any negligence of Shell Global SolutionsInternational and Shell International Chemicals.

DISCLAIMER

This safety alert has been provided by the Mary Kay O’Connor Process Safety Center with permission from Shell.

This alert is being provided by the Mary Kay O’Connor Process Safety Center as a service. Users of this informationshould make appropriate analysis and check the information to their own satisfaction. The Center does not warrant orrepresent, expressly or implied, the correctness or accuracy of the content or the information presented in this alert,nor can they accept liability or responsibility whatsoever for the consequences of its use or misuse by anyone.

17

Mary Kay O'Connor Process Safety CenterFALL 2004 Continuing Education

Please send registration form and check or fax credit card(American Express, Diners Club, MasterCard, or Visa) to:

Mary Kay O’Connor Process Safety CenterAttention: Mary CassTexas A&M University

3574 TAMUCollege Station, TX 77843-3574

Phone: (979) 458-1863 Fax: (979) 458-0422

Circle one: w t y e

Total $ __________________

CC# ______________________________ Exp. ______

Card Holder __________________________________

To register online, go to: http://www.texasonline.state.tx.us/NASApp/tamu/ODEManagerand select courses offered by the Texas Engineering Experiment Station and then you will be linked to the site listing all our ourcourses. Follow the instructions and be sure to wait for a confirmation that your registration was received before exiting the site.

COURSE TITLE COURSE DATE FEE

Last Name First Name MI

Company Name

Mailing Address

City State Zip

Telephone Fax E-Mail Address

Continuing Education Registration Form

Locations: See above listings Class Time: 9:00AM - 4:00PM (*except where noted)

Registration Fees: Early Registration (4 weeks prior) - $495.00 per person (†except where noted)Regular Registration - $550.00 per person

For more information or to register,contact Mary Cass at 979-458-1863 or mary-cass@tamu.edu

November

18 · †*Management of Change - Harry West - GSWEC, Katy (1-day Course, $295 Early, $350 Regular) Class time: 8AM-4PM

October

5 - 6 · Inherently Safer Design - Dennis Hendershot - SIS-Tech facility

12 - 13 · Process Safety Management - Adrian L. Sepeda - GSWEC, Katy

September

14 - 15 · *Root Cause Incident Investigation - Jack Philley - Baker Engineering & Risk Consultants facility Class time: 8AM-4PM

14 - 15 · Process Hazard Analysis Leadership Training - William (Skip) Early - SIS-Tech facility28 - 29 · Systematic Assessment of Reactive Chemical Hazards - Dr. Sam Mannan & Dr. William Rogers -

Great Southwest Equestrian Center (GSWEC)

18

Mary Kay O’Connor Process Safety CenterBEYOND REGULATORY COMPLIANCE,MAKING SAFETY SECOND NATURE

October 26 - 27, 2004

Reed Arena • Texas A&M UniversityCollege Station, Texas

Registration

Early registration is currently underway. Symposium registrations received by September27, 2004 are $495.00 per person and include refreshments, lunch, handouts and proceeding. Thereis a $50 per person additional discount available when registering five or more people from the samecompany. Registration can be done via phone, fax, or on-line. On-line registration is available at:http://www.texasonline.state.tx.us/NASApp/tamu/ODEManager by clicking on Texas Engineer-ing Experiment Station for the Center's events list. Select: 2004 Beyond Regulatory Compliance,Making Safety Second Nature International Symposium 081161A.

ExhibitsThe symposium will feature an exhibit area where companies can show products, technology,

and software related to process safety. The booth fee of $1250 includes the 10'X10' booth, electri-cal hookup, ethernet line, table, curtain backdrop, and one complimentary symposium registration.

Sponsorship

Your organization can become an Official Sponsor of the Symposium. Sponsorship levelsfor a break, lunch or reception are $2500 and $5000. As a sponsor you receive 1) recognition at thesymposium as an official sponsor. A sign will be prominently placed at your particular event recognizingyour organization as the sponsor; 2) your organization’s logo will be included in the list of sponsors,in Symposium publicity material (printed after receipt of sponsorship agreement) and on theSymposium web page; 3) your logo and a short description of your organization, with contactinformation, will be included in the official Symposium Proceedings; and 4) an exhibit booth (regularrental $1250) will be provided to your organization, in the exhibit area, at no additional cost.

More information on any of the above items is available by contacting DonnaStartz at donnas@tamu.edu or, phone 979/845-5981 and also, by visiting our website at: http://process-safety.tamu.edu/symposium/2004/info.htm

19

MORNING

8:00 -10:00AM

10:30 -12 Noon

AFTER-NOON

1:00 -2:30PM

3:00 -5:00PM

Keynote: “Process Safety Challenges for the Chlorine Industry,”Ms. Kathleen Shaver, President, The Chlorine Institute

“CCPS,” Scott Berger, Director, CCPS

State of the Center: Research Program, Current Activities, and Future DirectionDr. Sam Mannan, Director, Mary Kay O’Connor Process Safety Center

Time Tuesday, October 26, 2004

Track I Track II Track III

5:00 -7:00PM - Reception

Co-Sponsors

MARY KAY O’CONNOR PROCESS SAFETY CENTER - 2004 SYMPOSIUM PROGRAM

8:00AM -

9:00AM -

9:30AM -

Risk Assessment/Risk Mgmt I• ”Assembling an Equipment Failure

Rate Database for RiskAssessments,” M. Moosemiller,BakerRisk

• “Analysis of Rotating EquipmentReliability in Chemical ProcessIndustry,” K-S. Park, J-G. Son,and Y-D. Jo, Korea Gas SafetyCorporation, Seoul, South Korea

• ”Managing Industrial Risk –Having a Tested and ProvenSystem to Prevent and AssessRisk,” S. Heller, Stephen Heller& Associates, Inc.

Risk Assessment/Risk Mgmt II• ”Updated Hazard Rate Equation.”

M. Rothschild, Rohm and HaasCompany

• ”One stone for two birds: early-stageprocess operability with processsafety in tandem,” F. L. Wu,C. Krakowski, and D. Woods,OmniCal Technologies Inc.

• ”Spontaneous Combustion ofOrganic Materials by Water,” Z-M.Fu, Chinese People’s Armed PoliceForce Academy, and H. Koseki, Y.Iwata, National Research Institute ofFire and Disaster, Tokyo, Japan

General Session - Case Histories and Learning from Incident Investigations• “Accident Investigation: Keep Asking “Why?”,” T.A. Kletz, Loughborough University, Leicestershire, UK• “DMAI2C: A Novel Approach to Incident Reporting and Investigation,” E. Miles, Johnson and Johnson• “Case Study: Flame Arresters and Exploding Gasoline Containers,” L.C. Hasselbring, Stress Engineering Services

Lessons Learnedfrom Incident Databases

• ”Surveillance of Hazardous SubstancesReleases Due to System Interruptions,2002,” M. F. Orr, Agency for ToxicSubstances and Disease Registry

• ”Harnessing Data Mining to ExploreIncident Databases,” S. Anand, N.Keren, T.M. O’Connor, and M.S.Mannan, Mary Kay O’Connor ProcessSafety Center

• ”Lessons Learned from ProcessIncident Databases: The Process SafetyIncident Database (PSID) ApproachSponsored by the Center for ChemicalProcess Safety,” A.L. Sepeda, CCPS

Management Systems - I• “Applying Inherent Safety to Mitigate

Offsite Impact of a Toxic LiquidRelease,” D. Ferguson, DuPontEngineering Technology

• ”Managing and Integrating ProcessSafety: Manage PSM, Quality,Environmental Under One System,”J. Chosnek, KnowledgeOne

• ”Database Management Systems forProcess Safety,” W.F. Early, EarlyConsulting, L.C., and K. Thaxton,Advanced Aromatics, L.P.

Flammability and Combustion• ”Combustible Dust: Hazards and

Precautions,”G. Joseph and F.AltamiranoInvestigator, U. S.Chemical Safety and HazardInvestigation Board

• ”Evaluation of Lower FlammableLimits of Fuel-Air-DiluentMixtures,” M. Vidal, W. Wong,M. S. Mannan, and W. J. Rogers,MKOPSC

• ”Challenges in Fire, Dispersion andExplosion Consequence ModelingDuring Preliminary Design Stage ofan Offshore Platform,” G. Sharmaand P. Fletcher, Granherne Inc.

Safety Culture• ”Creating and Maintaining the Right

Safety Culture,” C. Olive, T.M.O’Connor, and M.S. Mannan, MaryKay O’Connor Process Safety Center

• “Benchmarking Best Practices inSafety Management,” M. Southard,The SRI Group, Inc.

• ”Quantification of the Intangible –Dealing with Safety Culture,”I. Rosenthal, University ofPennsylvania

Sima Chervin, David Chung, Marc Levin Skip Early, Mike Marshall, Kathy Shell George King, Scott Ostrowski, Rob Smith Chairs: Chairs: Chairs:

20

Security• ”Application of the API/NPRA SVA

Methodology to TransportationSecurity Issues,” D.A. Moore,AcuTech Consulting Group

• ”Emergency Response to Chemical/Biological Terrorist Incidents,” K.Scyoc, DNV

• ”An Alternative Approach toConducting SVA’s at ChemicalFacilities,” W.S. Effron and T. Low,RRS Engineering

•“Development of a Cable Designed toFunction Electrically During a FireEmergency”, S. Stephan, DekoronWire & Cable

Control Systems• ”Potential Use of Advanced Process

Control for Safety Purposes DuringAttack of a Process Plant,” J.R.Whiteley, Oklahoma State University

• ”IEC 61511 – Integration with theCapital Project Process,” A.E.Summers, SIS-TECH Solutions

• ”A Methodology for Fault Detection,Isolation, and Identification forNonlinear Processes with ParametricUncertainties,” S. Rajaraman, J.Hahn, and M.S. Mannan, MKOPSC

Reactive Chemicals - I• ”Integrating Reactive Chemical

Management Practices: Philosophy andApproach,” A.A. Aldeeb, Berwanger

• ”Study on Decomposition Mechanismof Solid and Application in the SADTPrediction using Highly SensitiveCalorimeter,” X. Li and H. Koseki,National Research Institute of Fire andDisaster, Tokyo, Japan

• ”Kinetics of Acid-Catalyzed CumeneHydroperoxide Decomposition,” M.E.Levin, Shell Global Solutions (US)

• ”Decomposition of Energetic Chemi-cals Contaminated withIron or Stain-less Steel,” S. Chervin, G.T. Bodman,and R.W. Barnhart, Eastman Kodak Co.

Reactive Chemicals - II• ”Global Kinetic Model: n-Oxidation

of Alkylpyridines, J. Gao and M.Papadaki, University of Leeds, UK

• ”Thermal Decomposition HazardEvaluation of Hydroxylamine Nitrate,”C. Wei, W. J. Rogers, and M. SamMannan, MKOPSC

• ”Process Hazard Screening Analysisfor a Reactive Chemical Process,”J. Philley, BakerRisk

Management Systems - II• ”Utilizing the Human, Machine, and

Environment Matrix in Investigations,”D. Curry, J.M. McKinney, and S.Paugels, Packer Engineering

• ”Serial Killers in the Process Plant,”G. S. Price, Plant Safety Analysts Inc.

• ”Digital Dashboard for Process SafetyManagement (PSM) Key PerformanceMetrics,” D. Drerup, Data Systems &Solutions

Frank P. Lees Memorial Lecture

“The Columbia Disaster, NASA Culture, and Lessons Learned,”Major General Kenneth W. Hess, Columbia Accident Investigation Board

Time Wednesday, October 27, 2004

MORNING

8:00 -10:00AM

9:30 -11:30AM

AFTER-NOON

12:30 -2:00PM

2:30 -4:00PM

MARY KAY O’CONNOR PROCESS SAFETY CENTER - 2004 SYMPOSIUM PROGRAM

8:00 - 9:00AM

LNG• ”LNG Project,” H. Pateo, KBR

• ”The Role of Consequence Modelingin LNG Siting,”D.W. Taylor, Bechtel Corp.

• ”LNG Decision Making ApproachesCompared,” R. Pitblado, DNV

• ”Effect of Tank Configuration &Rupture Variables on the Consequenceof LNG Spillage onto Water,” M.S.Mannan, Y. Qiao, H.H. West,MKOPSC, and J.B. Cornwell, D.W.Johnson, Quest Consultants Inc.

Consequence Analysis• ”Modeling the Initial Velocity of Two-

Phase Jets: Initiating a NewExperimental Program for ModelVerification,” T.O. Spicer and J.Havens, University of Arkansas

• ”Near and Far Field Plume Dispersionin a Large Scale Fire Scenario: CFDcomparisons with experimentalobservations,” N.L. Ryder and C.F.Schemel, Packer Engineering, Inc.

• “Effects of Heating Rate, Temperatureand Iron Catalysis on the ThermalBehavior and Decomposition of 2-Nitrobenzoyl Chloride,” S. D. Leverand M. Papadaki, University of Leeds

Learning from Experience• “Identifying the Direct and Underlying Human Causes of Accidents:

Developing an Evidence Base to Prioritize Accident Reduction Efforts,”D. Embrey and J. Henderson, Human Reliability Associates, Ltd

• “Bulk Transportation of Hazardous Materials by Rail: Lessons Learned fromNon-Collision Accidents,” M.J. Viz, R.A. Ogle, A.R. Carpenter, and D.T.Morrison, Exponent Failure Analysis Associates

• “When Process Safety Management Breaks Down: A Case Study of the LomacTNM Explosion,” C.K. Kaijala, Process Technology Consulting

Track I Track II Track IIISima Chervin, David Chung, Marc Levin Skip Early, Mike Marshall, Kathy Shell George King, Scott Ostrowski, Rob Smith

Chairs: Chairs: Chairs:

21

2004 SYMPOSIUM REGISTRATIONMary Kay O’Connor Process Safety Center

BEYOND REGULATORY COMPLIANCE, MAKING SAFETY SECOND NATUREOctober 26-27, 2004

Reed Arena • Texas A&M University • College Station, TexasPlease Print Clearly

q Payment by Check(payable to Mary Kay O’Connor Process Safety Center)

Total Enclosed $___________________q Payment by Credit Card q MasterCard q Visa q American Express q Diners Club

CC# _______________________________________________________

Card Holder _________________________________ Exp. ___________

Total Charge $____________________

Send payment to:

Texas A&M UniversityMary Kay O’Connor Process Safety Center

3574 TAMU916 Richardson PE Bldg.

College Station, TX 77843-3574-or-

Phone: (979) 845-5981Fax: (979) 458-1493

E-mail: donnas@tamu.edu

Travel:You can travel to College Station by flying into theEasterwood Airport in College Station from the HoustonIntercontinental or the Dallas/FortWorth Airport. Also, youcan drive from Houston Intercontinental, which is aboutan hour and a half drive.

Parking: Complimentary parking is available atReed Arena during the Symposium.

• • • • • •

For more information: Contact Donna StartzE-mail: donnas@tamu.edu • Phone: (979) 845-5981

http://process-safety.tamu.edu

Last Name First Name MI

Company Name

Mailing Address

City State Zip

Telephone Fax E-Mail Address

Additional Persons Registering: ($50.00 discount per person when registering five or more from the same organization)

2) _______________________________________________ 4) ________________________________________________

3) _______________________________________________ 5) ________________________________________________

Please indicate preferred track for session attendance:

Day 1 - First Session: Track I Track II Track III Day 2 - First Session: Track I Track II Track III Second Session: Track I Track II Track III Second Session: Track I Track II Track IIIGeneral Session Third Session: Track I Track II & III

REGISTRATION FEES {Fee includes refreshments, lunch, handouts and proceedings}

• Received by September 27, 2004- $495.00 per person • After September 27, 2004 - $550.00 per person

q Proceedings only (Book/CD-Rom set) - $65.00

Cancellation Policy: Cancellations must be received ten working days prior to the symposium to receive a full refund. After that time, there will be a 30% penalty.All refunds will incur a $25 service charge.

Accommodations:Please indicate you are attending the Mary Kay O’ConnorProcess Safety Center Symposium when making reservations.  Rooms have been blocked at the following hotels.

HOTEL PHONE RATE DeadlineHampton Inn (979) 846-0184 $68.00 10/9/2004Hilton (979) 693-7500 $80.00 10/3/2004Holiday Inn (979) 693-1736 $62.00 10/11/2004Holiday Inn Express (979) 846-8700 $70.00 9/24/2004LaQuinta Inn (979) 696-7777 $52.00 10/15/2004Manor House Inn (979) 764-9540 $58.00 10/15/2004Quality Suites (979) 695-9500 $80.00 9/27/2004Ramada Inn (979) 693-9891 $59.00 10/10/2004

22

Contact:Mary Kay O’Connor Process Safety CenterTexas A&M University3574 TAMUCollege Station, TX 77843-3574Phone: 979/845-3489Fax: 979/458-1493http://process-safety.tamu.edu

FALL 2004 CALENDAR

Monday, October 25, 2004Executive Forum Meeting

10 AM - 5 PMTexas A&M University

Jack E. Brown Building -(New ChemE Bldg.)

October 26-27, 20042004 MKOPSC Symposium

Texas A&M UniversityReed Arena

Thursday, October 28, 2004Technical Advisory Committee Meeting

9 AM - 4 PMTexas A&M University

Jack E. Brown Building -(New ChemE Bldg.)