EPA Method 14A

APPENDIXATOPART60--TESTMETHODS*****METHOD14A-DETERMINATIONOFTOTALFLUORIDEEMISSIONSFROMSELECTEDSOURCESATPRIMARYALUMINUMPRODUCTIONFACILITIESNOTE:Thismethoddoesnotincludeallthespecifications(e.g.,equipmentandsupplies)andprocedures(e.g.,sampling)essentialtoitsperformance.Somematerialisincorporatedbyreferencefromothermethodsinthispart.Therefore,toobtainreliableresults,personsusingthismethodshouldhaveathoroughknowledgeofatleastthefollowingadditionaltestmethods:Method5,Methods13Aand13B,andMethod14ofthisappendix.1.0ScopeandApplication.1.1Analytes.AnalyteCASNo.SensitivityTotalfluoridesNoneassignedNotdeterminedIncludeshydrogenfluoride007664-39-3Notdetermined1.2Applicability.Thismethodisapplicableforthedeterminationoftotalfluorides(TF)emissionsfromsourcesspecifiedintheapplicableregulation.ThismethodwasdevelopedbyconsensuswiththeAluminumAssociationandtheU.S.EnvironmentalProtectionAgency(EPA).2.0SummaryofMethod.2.1Totalfluorides,intheformofsolidandgaseousfluorides,arewithdrawnfromtheascendingairstreaminsideofanaluminumreductionpotroomand,priortoexitingthepotroomroofmonitor,intoaspecificcassettearrangement.Thecassettesareconnectedbytubingtoflowmetersandamanifoldsystemthatallowsfortheequaldistributionofvolumepulledthrougheachcassette,andfinallytoadrygasmeter.Thecassetteshaveaspecificinternalarrangementofoneunalteredcellulosefilterandsupportpadinthefirstsectionofthecassetteforsolidfluorideretentionandtwocellulosefilterswithsupportpadsthatareimpregnatedwithsodiumformateforthechemicalabsorptionofgaseousfluoridesinthefollowingtwosectionsofthecassette.Aminimumofeightcassettesshallbeusedforapotlineandshallbestrategicallylocatedatequalintervalsacrossthepotroomroofsoastoencompassaminimumof8percentofthetotallengthofthepotroom.Agreaternumberofcassettesmaybeusedshouldtheregulatedfacilitychoosetodoso.Themassflowrateofpollutantsisdeterminedwithanemometersandtemperaturesensingdeviceslocatedimmediatelybelowtheopeningoftheroofmonitorandspacedevenlywithinthecassettegroup.3.0Definitions.3.1Cassette.Asegmented,styreneacrylonitrilecassetteconfigurationwiththreeseparatesegmentsandabase,forthepurposeofthismethod,tocaptureandretainfluoridefrompotroomgases.3.2Cassettearrangement.Thecassettes,tubing,manifoldsystem,flowmeters,drygasmeter,andanyotherrelatedequipmentassociatedwiththeactualextractionofthesamplegasstream.3.3Cassettegroup.Thatsectionofthepotroomroofmonitorwhereadistinctgroupofcassettesislocated.3.4Potline.Asingle,discretegroupofelectrolyticreductioncellselectricallyconnectedinseries,inwhichaluminaisreducedtoformaluminum.3.5Potroom.Abuildingunitthathousesagroupofelectrolyticreductioncellsinwhichaluminumisproduced.3.6Potroomgroup.Anuncontrolledpotroom,apotroomthatiscontrolledindividually,oragroupofpotroomsorpotroomsegmentsductedtoacommonprimarycontrolsystem.3.7Primarycontrolsystem.Theequipmentusedtocapturethegasesandparticulatemattergeneratedduringthereductionprocessandtheemissioncontroldevice(s)usedtoremovepollutantspriortodischargeofthecleanedgastotheatmosphere.3.8Roofmonitor.Thatportionoftheroofofapotroombuildingwheregases,notcapturedatthecell,exitfromthepotroom.3.9Totalfluorides(TF).ElementalfluorineandallfluoridecompoundsasmeasuredbyMethods13Aor13Bofthisappendixorbyanapprovedalternativemethod.4.0InterferencesandKnownLimitations.4.1Therearetwoprincipalcategoriesoflimitationsthatmustbeaddressedwhenusingthismethod.Thefirstcategoryissamplingbiasandthesecondisanalyticalbias.Biasesinsamplingcanoccurwhenthereisaninsufficientnumberofcassetteslocatedalongtheroofmonitorofapotroomorifthedistributionofthosecassettesisspatiallyunequal.Knownsamplingbiasesalsocanoccurwhenthereareleakswithinthecassettearrangementandifanemometersandtemperaturedevicesarenotprovidingaccuratedata.Applicableinstrumentsmustbeproperlycalibratedtoavoidsamplingbias.Analyticalbiasescanoccurwheninstrumentationisnotcalibratedorfailscalibrationandtheinstrumentisusedoutofpropercalibration.Additionally,biasescanoccurinthelaboratoryiffusioncruciblesretainresidualfluoridesoverlengthyperiodsofuse.Thisconditioncouldresultinfalselyelevatedfluoridevalues.Maintainingacleanworkenvironmentinthelaboratoryiscrucialtoproducingaccuratevalues.4.2Biasesduringsamplingcanbeavoidedbyproperlyspacingtheappropriatenumberofcassettesalongtheroofmonitor,conductingleakchecksofthecassettearrangement,calibratingthedrygasmeterevery30days,verifyingtheaccuracyofindividualflowmeters(sothatthereisnomorethan5percentdifferenceinthevolumepulledbetweenanytwoflowmeters),andcalibratingorreplacinganemometersandtemperaturesensingdevicesasnecessarytomaintaintruedatageneration.4.3Analyticalbiasescanbeavoidedbycalibratinginstrumentsaccordingtothemanufacturer'sspecificationspriortoconductinganyanalyses,byperforminginternalandexternalauditsofupto10percentofallsamplesanalyzed,andbyrotatingindividualcruciblesastheblankcrucibletodetectanypotentialresidualfluoridecarry-overtosamples.Shouldanycontaminationbediscoveredintheblankcrucible,thecrucibleshallbethoroughlycleanedtoremoveanydetectedresidualfluoridesandablankanalysisconduc