Multiscale Modeling for Process Safety Applications PDF
Preview Multiscale Modeling for Process Safety Applications
Multiscale Modeling for Process Safety Applications Arnab Chakrabarty Sam Mannan Tahir Cagin AMSTERDAM(cid:129)BOSTON(cid:129)HEIDELBERG(cid:129)LONDON NEWYORK(cid:129)OXFORD(cid:129)PARIS(cid:129)SANDIEGO SANFRANCISCO(cid:129)SINGAPORE(cid:129)SYDNEY(cid:129)TOKYO Butterworth-HeinemannisanimprintofElsevier Butterworth-HeinemannisanimprintofElsevier TheBoulevard,LangfordLane,Kidlington,OxfordOX51GB,UK 225WymanStreet,Waltham,MA02451,USA Copyright©2016ElsevierInc.Allrightsreserved. ThisbookandtheindividualcontributionscontainedinitareprotectedundercopyrightbythePublisher (otherthanasmaybenotedherein). Nopartofthispublicationmaybereproducedortransmittedinanyformorbyanymeans,electronicor mechanical,includingphotocopying,recording,oranyinformationstorageandretrievalsystem,without permissioninwritingfromthepublisher.Detailsonhowtoseekpermission,furtherinformationaboutthe Publisher’spermissionspoliciesandourarrangementswithorganizationssuchastheCopyrightClearance CenterandtheCopyrightLicensingAgency,canbefoundatourwebsite:www.elsevier.com/permissions. Notices Knowledgeandbestpracticeinthisfieldareconstantlychanging.Asnewresearchandexperiencebroadenour understanding,changesinresearchmethods,professionalpractices,ormedicaltreatmentmaybecome necessary. Practitionersandresearchersmustalwaysrelyontheirownexperienceandknowledgeinevaluatingandusing anyinformation,methods,compounds,orexperimentsdescribedherein.Inusingsuchinformationormethods theyshouldbemindfuloftheirownsafetyandthesafetyofothers,includingpartiesforwhomtheyhavea professionalresponsibility. Tothefullestextentofthelaw,neitherthePublishernortheauthors,contributors,oreditors,assumeany liabilityforanyinjuryand/ordamagetopersonsorpropertyasamatterofproductsliability,negligenceor otherwise,orfromanyuseoroperationofanymethods,products,instructions,orideascontainedinthe materialherein. ISBN:978-0-12-396975-0 BritishLibraryCataloguing-in-PublicationData AcataloguerecordforthisbookisavailablefromtheBritishLibrary LibraryofCongressCataloging-in-PublicationData AcatalogrecordforthisbookisavailablefromtheLibraryofCongress ForinformationonallButterworth-Heinemannpublications visitourwebsiteathttp://store.elsevier.com/ Publisher:JoeHayton AcquisitionsEditor:FionaGeraghty EditorialProjectManager:LindsayLawrence ProductionProjectManager:LisaJones Designer:MatthewLimbert TypesetbyTNQBooksandJournals www.tnq.co.in PrintedandboundintheUnitedStates Preface Computational modeling has emerged as a powerful partner to experimental and theoretical studies. Itisafamiliarsubjectinseveralfieldsbuthassofarbeenconsideredofsecondaryimportanceinpro- cesssafetyapplications,whichmostlyreliesonplantpersonnelexperienceandaccumulatedlearning ofthecommunityfromhistoricalincidences.Thecurrentstandardpracticesinassessmentofhazard- ousscenariosorrelevantparameterscanbesignificantlyaidedbythewiderimplementationofdetailed computationalmethods,atvarioustimeandlengthscales.Executionoftheseapproaches,asexplored inthisbook,hasbecomemorepracticalowingtothesignificantincreaseincomputationalpowerand ease of access to appropriate computational tools. Researchers across the globe have worked in exploitingthesemethodsforprocesssafetyapplications,butuntilnow,theensembleoftheseefforts has remained mostly as isolated studies. The main objective of this book is to group these existing efforts under a common platform by providing the current status of this novel area and discussing the potential implementation of multiscale modeling approaches for process safety applications. Professionals,inbothindustryandacademia,canusethisbook,ascangraduateresearcherswork- inginanydomainwhereprocesssafetychallengesareimplicitlyorexplicitlyembedded.Toaddress such challenges, the book has attempted to cover applications of quantum mechanics, molecular dynamics, quantitative structureeproperty relationship, quantitative structureeactivity relationship, computationalfluiddynamics,finiteelementanalysis,chaostheory,statisticalanalysis,anddynamic simulation as applicable. The problems dealt with are consequence modeling, risk assessment, and related parameter estimation for potential hazards resulting from fire, explosion, and dispersion of toxic gasses. Inthisbook,thebridgebetweenvariousmodelingmethodologiesisnotdiscussed.Inthatsense,the individual chapters are essentially independent. Nor is the mathematical background of these multi- scale modeling approaches discussed in detail. Fortunately, for each of these topics, several good booksareavailableonthemarket.Thesesourcescanbeusedtosupplementone’straininginmathe- matical fundamentals. Typically, problems at the level of quantum mechanics and molecular dynamics have dealt with safety-related concerns for a specific molecular system. At the molecular modeling level, it is more about material characterization in a given environment. Limited by time and length scale, these methods are not suitable for answering safety concerns for the whole process domain. However, knowledge gained at these scales, such as estimation of parameters at the molecular scale, can be passed on to a modeling environment with a lesser resolution for an atomistically informed holistic model. Another example of using two different scales in modeling approaches for addressing the same concern could be the use of a conservative low-resolution model for screening analysis and then the performanceofdetailed modeling of selected scenarios, similar to the Pareto principle. Emerging tools and increasing computational power modeling in process safety have improved much from the case of a spherical cow, a metaphor used for highly simplified scientific models of complex physical phenomena. Accordingly, the use of detailed computational modeling approaches incombinationwithaccumulatedknowledgeofsafetyexpertscanonlycontributeinapositivemanner towardimprovingthesafetyperformanceofchemicalfacilities.Thisbookiswrittenwiththehopethat it will contribute toward promoting the use of multiscale modeling methodologies in addressing processsafety problems. xi Acknowledgments It is our pleasure to acknowledge here some debts of gratitude. Without the tremendous help of the following, given below in no particular order, completion of this book would have faced much hindrance: Monir Ahammad, Xiaodan Gao, Ning Gan, Zhe Han, Brian Harding, Logan Hatanaka, Sunder Janardanan, Xinrui Li, Yan-Ru Lin, Ruochen Liu, Yi Liu, Ha Nguyen, William Pittman, Samina Rahmani, Olga Reyes, Josh Richardson, Camilo Rosas, Nitin Roy, Sonny Sachdeva, BinZhang,andJiaqiZhang,allatMaryKayO’ConnorProcessSafetyCenter,TexasA&MUniversity, for their help with preparation of the chapters. Se´bastien Canneaux, Fre´de´ric Bohr, LISM, and Eric He´non, ICMR, at the University of Reims Champagne-Ardenne,for preparationof the KiSThelP tutorial problem. Fiona Geraghty, Cari Owen, Natasha Welford, Lindsay Lawrence, Lisa Jones, Matthew Limbert and the Elsevier production team, for their patience and enormous help throughout the life of the project. Our debts toward our families are too personal to acknowledge here entirely. As we struggled to complete our manuscript, they havesufferedsomeneglect.It is tothem, we dedicate thisbook with love andaffection. xiii CHAPTER 1 INTRODUCTION Concernformanhimselfandhissafetymustalwaysformthechiefinterestofalltechnical endeavors.Neverforgetthisinthemidstofyourdiagramsandequations. AlbertEinstein Numerous incidents such as the Flixborough explosion resulting from the release of flammable hydrocarbons in June 1974, the methyl isocyanate release in Bhopal in December 1984, and the Macondodisasterin2010continuetoremindusoftheimportantroleofprocesssafetyinthedesign andoperationsofprocessfacilities.Overtheyears,processsafetyhasemergedasadisciplineinitself andhascontinuedtoplayadominantroleinanyprocesstechnologicaldevelopment.Whileafocuson process safety model development, experiments at various scales have gained momentum over the yearsandhadapositiveimpactontheindustry,thesafetyincidentdatabasesareanythingbutstagnant (MARSH,2014).Evenrecently,inAugust2013,inresponsetorecentcatastrophicchemicalfacility incidents in the United States, President Obama issued an executiveorder to enhance the safety and securityofchemicalfacilities,andconsequentlyreducetherisksassociatedwithhazardouschemicals, for owners, operators, workers,and communities. Given the complexity of existing and emerging technologies, interactions of process parameters, andintangibleandhiddencorrelations,quantitativelydefiningtheunderlyingrootofaprocesssafety phenomenon, or predicting the same scenario, requires a great deal of experience, extensive knowl- edge, and a structured approach. Additionally, while experimentally quantifying process safety parameters is often insightful, it is resource intensive and often not comprehensive owing to the number of experiments that would be required to achievethe same. Experimentally validated multi- scale modeling methodologies are powerful tools that have the potential to offer answers which are otherwiseoftenexpensiveorunreachablethroughexperiments.Forexample,thinkaboutlarge-scale jet fires or deflagration-to-detonation events. Typical current practice in modeling process safety scenariosistosubstitutethelackofunderstandingbyintroducingconservativenesstoasafetymodel. Unfortunately, conservative safety parameters not only increase plant costs, but may not ensure the safetyofafacility.Theinsufficientdetailsorcoarserresolutionofthesolutionhasthepotentialtooffer more risk than is mitigated by adding conservativeness to it. As we have demonstrated in a later chapter,mixturesexhibitingminimumflashpointbehaviorsposesucharisk,andthereforeexercising conservativenessalone may notbe sufficient. Withsignificantriseofcomputationalpowerinrecenttimesalongwithbetterunderstandingofthe underlying physics, implementation of higher-resolution models in order to gain insights and refine existingmodelstoassesshazardousscenarioshasbecomemorepracticalandachievable.Owingtothe multiscale nature of processes, multiscale modeling has emerged as a new set of tools that provides insight into important features at multiple times and lengths of a physical phenomenon. Thus, in 1 MultiscaleModelingforProcessSafetyApplications.http://dx.doi.org/10.1016/B978-0-12-396975-0.00001-2 Copyright©2016ElsevierInc.Allrightsreserved. 2 CHAPTER 1 INTRODUCTION severalapplications,ithasbecomecrucialtoincorporateinformationfromarangeoflengthandtime scalesintoamodel(CameronandGani,2011).Asanoutcomeofmodelingproductandprocessissues, agrowingnumberofmodelingeffortsareresultinginmultiscalemodelingapproaches.Thestrategies discussed throughout the book, albeit in the context of process safety, attempt to capture inherently important properties at various scales of a system and correlate them accurately to the system’s macroscaleproperties.Areasonableamountofworkhas been doneinthatrespect inotherareas,as demonstrated in several publications (Le´pinoux, 2000; Kwon et al., 2007; Derosa and Cagin, 2010; Maekawaetal.,2008).Multiscalemodelingtechniqueshavebeensuccessfullyemployedinmaterial design to rationally develop and accurately predict the performance of systems with the building blocksoftheirmacroscopic-levelperformanceresidingatmuchsmallerscales.Similartootherfields, itisonlyappropriatetoextendtheadvancementinmaterialstheoryatdifferenttimeandlengthscales toaddresslessunderstoodsafetyconcernsbyincorporatinganadequatelevelofdetail.Inthecontext of process safety, while some sporadic work exists to assess problems at different time and length scales,todatethosehavebeenmostlyisolatedefforts.Theprimaryobjectiveinthisbookistogroup theseexistingeffortsonacommonplatform.Thisbookaimstoprovideareviewofthecurrentstatusin this area, discuss potential implementation of multiscale modeling, and help refine existing compu- tational approaches used for safety analysis. It attempts to provide an overall picture of how safety issues are addressed at all scales ofmodeling,and discussesthe latest methods inthe field. Chapter 2 serves as the introduction to process safety. It touches upon the status of current industriallyacceptedstate-of-the-artmodelingapproachesinprocesssafetyanalysis.Asappropriate, thechapterintroducesorreintroducesthereadertoprocesssafetyfundamentalssuchasthephysics, consequences, and risks of fire, explosion, and toxic hazards in process industries. Concepts of flammability, ignition phenomena, fire, dispersion of flammable and toxic gases, deflagration and detonation, risk assessment of fire, toxic, and explosion hazards, are also covered in this chapter. Chapters 3–6 deal with the use of modeling methodologies as applicable to process safety applicationswithinvarioustimeandlengthscales.Chapter3demonstratestheapplicability,relevance, and benefits of molecular modeling methods such as quantum mechanics, molecular dynamics, quantitative structure–property relationship (QSPR), and quantitative structure–activity relationship (QSAR) in process safety applications at various capacities. Chapter 4 moves into greater time and length scales and examines the effectiveness and benefits of implementing computational fluid dynamics(CFD)inthedevelopmentofconsequencemodelsofprocessfacilities.Itlooksintouseof CFDinassessingtheconsequenceofvarioustypesoffiressuchasjet,pool,andflashfires.Italsolooks atthemodelingofexplosionandblastwavesusingCFD.Inasimilarfashion,Chapter5illustratesthe practicality of using finite element methods in process safety applications. Finite element methods have been implemented in understanding flare systems, storage and transportation of flammable materials, and other concerns in process hazard analysis. Accessing larger time and length scales phenomena are demonstrated through implementation of dynamic process simulations in Chapter 6. The transient nature of a process is typically crucial in modeling plant start-up and shutdown phenomena.Inthesamechapter,chaostheoryandstatisticalanalysisareintroducedwithinthecontext ofaddressingprocesssafety concerns. Chaos theory can beapplied toinvestigaterunawayreactions andhasthepotentialtoprovideearlywarningdetectionofsame.Statisticalanalysishasbeenutilized tomonitorreal-timeplantdata.Multivariatestatisticalanalysiscanbeappliedtoplantdatathatcanin turn help inincidentinvestigation. REFERENCES 3 The chapters referred to above insights into implementing modeling approaches for the accurate estimationofconsequencesfromfire,explosion,ortoxicemissionatvariousscales.Riskassessment, thesucceedingpartofaconsequencemodelingstudythatestimatesthelikelihoodofaconsequence,is covered in Chapter 7. The chapter illustrates approaches to gain insight into risk-related parameters through multiscale modeling. Approaches such as Bayesian logic, Bayesian-LOPA methodology for risk assessment are analyzed in this chapter. Process and material characteristics that directly affect equipmentfailurerate,suchasfracture,corrosionandsimilarphenomenaareexaminedinthischapter inthecontextofquantitativeriskassessment.Forexample,beyondpurelymonetaryvalue,corrosion profoundlyaffectsthesafetyofadvancedproductsandprocesses.Chapter8dealswithinherentlysafer design,anemergingandincreasinglyimportantprocesssafetytopic.Addressingsafetyconcernsatthe design level has gained momentum in recent years and is a promising path forward for the process safety community. The subject of industrial hygiene is covered in Chapter 9. While the chapter typically addresses molecular modeling approaches in assessing the toxicity of plant chemicals, the perspectiveisdifferentthanthatofChapter3,partofwhichalsoaddressesassessingtoxicitythrough implementationof molecular modeling. ThebookconcludeswithChapter10,followedbyexerciseproblemsatvariousscalesrelatedtothe topics discussed in the book. The aim is to represent multiscale modeling methodology as a set of crucialtools for answeringquestions andgaininginsightsin process safety applications. REFERENCES Cameron,IanT.,Gani,Rafiqul,2011.ProductandProcessModelling.Elsevier. Derosa,Pedro,Cagin,Tahir,2010.MultiscaleModeling.CRCPress. Kwon, Young, Allen, David H., Talreja, Ramesh R., 2007. Multiscale Modeling and Simulation of Composite MaterialsandStructures.SpringerScience&BusinessMedia. Le´pinoux,Joe¨l,2000.MultiscalePhenomenainPlasticity:FromExperimentstoPhenomenology,Modellingand MaterialsEngineering.SpringerScience&BusinessMedia. Maekawa, Koichi, Ishida, Tetsuya, Kishi, Toshiharu, 2008. Multi-scale Modeling of Structural Concrete. CRCPress. MARSH, 2014. The 100 Largest Losses (In the Hydrocarbon Industry) 1974-2013. Uk.Marsh.com. http://uk. marsh.com/NewsInsights/Articles/ID/37406/The-100-Largest-Losses-in-the-Hydrocarbon-Industry-1974- 2013.aspx(AccessedMarch26.). CHAPTER 2 PROCESS SAFETY 2.1 FIRE Theconceptoffireisonewhichthecommonpersonisusuallyfamiliar.Onemightdescribeafireas “hot”or“capableofdestroyingbuildings,”butwhatreallyisafire?Whatneedstobeknownaboutfire hazardstopreventincidents?Thesecomplexquestionshighlightakeyconceptaboutfiresafety:fireis a complicated phenomenon, but to prevent harm to workers and facilities, a fundamental under- standing offire hazards at allscalesis required. Fire, explosions, and toxic releases are considered to be the three most commonly encountered types of hazards that cause severe incidents and loss in process industries. Compared with toxic releasesandexplosions,firesgenerallycauselessdamagebutoccurmorefrequently(CocoandMarsh Risk,2001).However,becauseafireoftenprecedesanexplosionortoxicrelease,thedamagesandloss fromfirescanhavedireconsequences.Itis,therefore,importanttounderstandthephenomenaoffire to properly addressthe hazardspresented byit. Fire is a specific case of oxidation reaction that involves the rapid reaction of a fuel with an oxidizer.Thischemicalreactionresultsinanetreleaseofheat,alsoknownasanexothermicreaction. Inafire,anignitionsourcesuppliesenergytotheseaforementionedcombustionreactionsinitially,and iftheexcessheatfromtheoxidationreactionsissufficienttodrivethecombustionofmorematerial, the flame becomes self-sustaining(Crowl and Louvar, 2011). Generallyspeaking,theoxidizingreactionsthatdrivethefireoccurinthevaporphase.Therefore, flammablesolidsandliquidmustfirstheatupandvaporizebeforefullycombusting.Foraliquid,thisis assimpleasheatingandevaporatingor,inmoreextremecases,boiling.Forsolidflammablematerials, it is more complicated and usually first involves thermal degradation, known as pyrolysis, then vaporizationandcombustion(Drysdale,2011).Althoughgeneralinformationaboutfiresisinsightful, furtherdiscussion is requiredtofullyunderstand the behaviorandhazards offire at allscales. 2.1.1 THE FIRE TRIANGLE Forafiretooccur,threecomponentsmustbepresent:fuel,anoxidizer(oftenoxygen),andasourceof ignition,asdepictedinFigure2.1.Ifanysideofthefiretriangleisremoved,afirewillnotform(Crowl andLouvar,2011).Ifaflammablematerialisstoredinthepresenceofoxygenbutnoignitionsourceis present,afirecannotoccur.Similarly,ifafuelisheatedunderaninertgas,afirewillnothaveoxygen andwillnotburn.Last,itissimpletoseethatafirewillnotoccurifthereisnofuel.Regardlessofthe sizeof asystem, without the three components ofthe fire triangle, afirewill nottake place. 5 MultiscaleModelingforProcessSafetyApplications.http://dx.doi.org/10.1016/B978-0-12-396975-0.00002-4 Copyright©2016ElsevierInc.Allrightsreserved. 6 CHAPTER 2 PROCESS SAFETY FIGURE2.1 Thefiretriangle(AmericanInstituteofChemicalEngineers,CenterforChemicalProcessSafety,2003). While the absence of any side of the fire triangle prevents ignition, the presence of all three componentsofthefiretriangledoesnotnecessarilyensurethatafirewilloccur.Manyfactorsrelating to the three fire triangle component dictate whether a fire will occur, including the amount of each component present in the fire triangle (Crowl and Louvar, 2011). For example, a reduction in the oxygenconcentrationpastacertainpointwillextinguishliquidflammablefires.Theconcentrationof oxygen in airis an important factor inthe existence ofa fire. Similarly,iftheheatsourceusedtoinitiallybegincombustionisnotstrongenough,afiremayfail to start. The minimum ignition energy (MIE) for a material is heavily dependent on material type, physicaldistribution,andthephysicalconditions,butmosthydrocarbonshaveaMIEintherangeof 0.01–2mJ(GlassmanandYetter,2008).However,increasesinpressurecandecreasetheMIE,soitis importanttoaccountforareducedignitionenergyinpressurizedsystems(GlassmanandYetter,2008). Thesefactorsplayaroleinestimationofignitionprobability(CCPS,2014;Moosemiller,2011),which inturnpredicts thechancesofafire asoppose toexplosionorviceversainthe case ofaflammable release. The physical phase of a fuel also affects whether it can be ignited. As previously discussed, a flammablevapormixturedoesnotneedtovaporizebeforecatchingfire,butsolidandliquidmustfirstbe vaporized to burn, requiring a higher ignition energy (Crowl and Louvar, 2011). In addition, solids generally go through a thermal decomposition known as pyrolysis to produce a volatile flammable vapor(Drysdale,2011).Consequently,theenergyrequiredtosustainasolid,liquid,orvaporfirevaries. Last,thephysicalgeometryoftheflammablefuelaffectsthechancesthatafirewillmanifest.This iseasilyvisualizedwhencomparingtheflamefromamatchburningtop-to-bottomtoonethatburns from the bottom-to-top. It is clear that the fire triangle is greatly influenced by a large number of complicated factors. 2.1.2 IGNITION PHENOMENA Fires are initiated by an external heat source but the source can sometimes be silent and discrete. Although some fires can be started by a simple ignition source such as a spark or hot surface, protectingagainstignitionsourcesisnottheonlyprecautionthatmustbetakentopreventignition. 2.1 FIRE 7 Ignitionsourceawarenessensuresthattheproperprecautionscanbetakentocontrolignitionsources and prevent an incident in the future. The autoignition of a flammable mixture occurs when the temperature of the environment is highenoughtoprovidetheheatrequiredforcombustion.Instudyingthesephenomena,itisuseful to define the autoignition temperature for a vapor mixture, which is the minimal temperature at which a flammable mixture can undergo a rapid combustion process without any other ignition source (Crowl and Louvar, 2011). By maintaining a flammable mixture’s temperature below its autoignition temperature, the hazard of spontaneous ignition is reduced greatly. Once above this temperature,theflammablemixturecanhaveenoughthermalenergytoignite,evenwithoutasource of heat. Another sourceofignition thatisoftenoverlookedisthe auto-oxidizationofa flammableliquid. This occurs when a flammable liquid has a high boiling point and is stored without temperature control.Theslowoxidationofthematerialovertimecausesthetemperatureoftheliquidtoincrease slowly.Involatileliquids,thisheatisreleasedbysmallamountofvaporization,butinlow-volatility liquids,thisheatcanbuildupuntilitissufficienttoignitevaporabovetheflammableliquid(Crowland Louvar, 2011). Temperature controls or other precautions must be taken with similar low-volatility flammable organicliquids toavoidauto-oxidation. 2.1.3 FLAMMABILITY LIMITS OF GASES AND VAPORS Flammablevaporsandgasesareamajorfirehazardand,unlikemostsolidsandliquids,canignitewith very little ignition energy. The ease with which flammable gas and vapor mixtures ignite warrants hazard awareness when handling these materials. However, real-world mixtures of gases and vapors arehandledatarangeoftemperatures,pressures,andcompositions,andtherefore,understandinghow thesefactorsaffect flammablegases and vapors is important. The concentration of flammable vapor plays a key role in whether a fire will occur. If the con- centrationofaflammablevaporinairistoohigh,itissaidtobetoorichtoburn.Similarly,ifthevapor concentrationistoolow,itissaidtobetooleantoburn.Theseupperandlowerconcentrationbounds offlammabilityinairaredefinedastheupperflammabilitylimit(UFL)andlowerflammabilitylimit (LFL),respectively (Crowl andLouvar, 2011). Whenamixtureofflammablegasesishandled,theLFLandUFLofthemixturearedifferentthan the LFL and UFL of the separate components. One simple way to calculate LFL and UFL of flam- mablegasmixturesistouseLeChatelier’sequation,showninEqns(2.1)and(2.2)(Chatelier,1891). 1 LFL ¼P ð2:1Þ mix n yi i¼1LFL i where LFL is the LFL of chemical species i if it was pure, and LFL is the LFL of the mixture i mix consistingof chemical species ithroughn. 1 UFL ¼P ð2:2Þ mix n yi i¼1UFLi where UFL is the UFL of chemical species i if it was pure, and UFL is the UFL of the mixture i mix consistingof chemical species ithroughn.
The list of books you might like
