WPC  .\TZ;*nlwh|Hg59|M*G !?\xc/_[DxQCDHVM Yo( D$jő2;^H}<@=;n .i48K淖S``h]bq|\@vȇ9`2r6MBZ@`1bEB\@m*w>j"s,I2m>Uy5at0GH0D'aM7"RpJj;FqlWd=J.M!GnBXy:q7 |B[Y.^@+ ?#@UYzOMvcbeņ:,R׈ S6ed)]CɉÔ+4GJ/: ʕwYi]Hk4".2$|o^ Kv_]*q.`$X &81qĤa_c 0~* 0x 0k  0k 0 0~ 0 0 0 0i 0 ` 0j 01 0C 0V 0R 0! 0 0i 0, 0 0 0|! 0"U>" %## 0d%U :H'U:' 0' 0z( 08) 0) 0* 0{+ 0o9, 0, 0f- 0$. 0. 0/ 0G0 0Z1 0z'3 03 4 0=4 0F4 0P5 1i5 05 16 078 0[3<E< 0<UB> 0T> 0?UBy@ 0@ 0j_A 0eA 0W.B 0WB 0B 0j`C 0C 0D 0OE 0E 0F 0G 0H 0-I 0@K 0BL 0L 0M 0N 0N 0lO 0P 0P 0bQ 0}Q 0xmR 0R 0|rS 0rT 0l`V 0gV 0Y3W 0YW 0W 0lkX 0X 0Y 0`Z 0Z 0[ 0\ 0] 0-^ 0@` 0Ua 0b 0b 0,c 0c 0d 0$e 0e 0uf 0tg 0wh 0t'i 0tj 0xl 0tm 0n 0p 0xr 0gt 0av 0[x 0Uz 0O| 0I~ 0 C 0N 0K 0|H 0|ć 0@ 0|= 0 0 0 0@ 0 0 0Γ 0Y 0 0v 0xU> D+ 0 [ 0E A YO%dz͘"Gdi?d@ C#f=a? BL*S B}.dț+Ud8dR-qd*d5tdzf#a%%%%%%%%%% h9 h?? 0\EE 1An 0@# U> uHw4UFիUFUFad ,UF7V}UFӰVUFoooUFUFAUFQQQUFݴUFqUFǷUF UFSSSUF߸UFUF5{UFUFѻUF'''UFmUFľV& 8Document[8]Document Style0..8` ..` V8Document[4]Document Style.. . V 8Document[6]Document Style8..V 8Document[5]Document Style0..V/8Document[2]Document Style 2A.3  Ԁ   V& 8Document[7]Document Style0..0` ..` zU :Right Par[1]Right-Aligned Paragraph Numbers..2I.3  Ԁ..0..zh :Right Par[2]Right-Aligned Paragraph Numbers..` ..2A.3  Ԁ..0` ..` V?8Document[3]Document Style.. 21.3  Ԁ   z{ :Right Par[3]Right-Aligned Paragraph Numbers..` ..`  ..P 21.3  Ԁ` ..` 0 .. z :Right Par[4]Right-Aligned Paragraph Numbers..` ..`  .. .. 2a.3  Ԁ .. 0..z :Right Par[5]Right-Aligned Paragraph Numbers..` ..`  .. ..h..2(1)3  Ԁ..0h..hz :Right Par[6]Right-Aligned Paragraph Numbers..` ..`  .. ..h..h..2(a)3  Ԁh..h0..z :Right Par[7]Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....2i)3  Ԁ..0..z :Right Par[8]Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....p..2a)3  Ԁ..0p..pVX8Document[1]Document Style  @..^  2I.3  Ԁ     Ԉ l2:Technical[5]Technical Document Style.. 2(1)3  Ԁ. l2:Technical[6]Technical Document Style.. 2(a)3  Ԁ. l/%:Technical[2]Technical Document Style 2A.3  Ԁ   .. l,!:Technical[3]Technical Document Style 21.3  Ԁ   .. l(!:Technical[4]Technical Document Style 2a.3  Ԁ   .. l:0:Technical[1]Technical Document Style  2I.3  Ԁ     .. l1:Technical[7]Technical Document Style.. 2i)3  Ԁ. l1:Technical[8]Technical Document Style.. 2a)3  Ԁ. ,xA`&UniversScalable'     T. J. Geyer 0N.FTIR CEM Performance Specification FTIR, CEM h~~:Default ParaDefault Paragraph FontXXXO<6X9`("Courier NewTTO        O<6X9`("Courier 12cpiO<6X9`("Courier NewTT<6X9`("Courier 12cpi8;K,toc 1toc 1  PP!"        5+ ` hp x (#58;K,toc 2toc 2  P !"        5+ ` hp x (#58>K,toc 3toc 3  P X !"        5+ ` hp x (#58>K,toc 4toc 4  PX !"        5+ ` hp x (#58>K,toc 5toc 5  P !"        5+ ` hp x (#58;K,toc 6toc 6  PP!"        5+ ` hp x (#58!,toc 7toc 7          8;K,toc 8toc 8  PP!"        5+ ` hp x (#58;K,toc 9toc 9  PP!"        5+ ` hp x (#5@>K0index 1index 1  PP !"        5+ ` hp x (#5@;K0index 2index 2  P !"        5+ ` hp x (#5P8K8toa headingtoa heading   PP!"        5+ ` hp x (#5@~0captioncaption  XXXO<6X9`("Courier NewTTO        O<6X9`("Courier 12cpiO^~~:_Equation Ca_Equation CaptionXXXO<6X9`("Courier NewTTO        O<6X9`("Courier 12cpiOR& 8BibliogrphyBibliography0....fp2Doc InitInitialize Document Style/S 3    I. 1. A. a.(1)(a) i) a)S012($0 ($0 0 (($0 0 0   A_ekqwDocumentDocument StyleI.1.A.a.(1)(a)i)a)jo4Tech InitInitialize Technical StyleS 5  1 .1 .1 .1 .1 .1 .1 .1 S CuyTechnicalTechnical Document Style11.11.1.11.1.1.11.1.1.1.11.1.1.1.1.11.1.1.1.1.1.11.1.1.1.1.1.1.1x?+u2PleadingHeader for numbered pleading paper 7  8G(   >RXXX?9'*dE*??9'*dE*?PP 1PP 2PP 3PP 4PP 5PP 6PP 7PP 8PP 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28  .+(32G$ !     0.HEADER  2( ` hp!(#2W;* `*Times New RomanTTW5+ ` hp x (#5W;* `*Times New RomanTTW* `*Times New RomanTT20 Normal  5+ 4 <DL!(#5O<6X9`("Courier 12cpiO5+ ` hp x (#5W;* `*Times New RomanTTW6p4Body TextXXXW>*\  `*Times New RomanTTW*\  `*Times New RomanTT4p2DivisionW>*\  `*Times New RomanTTWB& $1Document Style0..8` ..` B $2Document Style.. . B $3Document Style8..B $4Document Style0..B-$5Document Style 23  Ԁ   B& $6Document Style0..0` ..` dS $7Right-Aligned Paragraph Numbers..23  Ԁ..0..df $8Right-Aligned Paragraph Numbers..` ..` 23  Ԁ..0` ..` B=$9Document Style.. 23  Ԁ   fy &10Right-Aligned Paragraph Numbers..` ..`  .. 23  Ԁ` ..` 0 .. f &11Right-Aligned Paragraph Numbers..` ..`  .. ..23  Ԁ .. 0..f &12Right-Aligned Paragraph Numbers..` ..`  .. ..h..h23  Ԁ..0h..hf &13Right-Aligned Paragraph Numbers..` ..`  .. ..h..h..23  Ԁh..h0..f &14Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....23  Ԁ..0..f &15Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....p..p23  Ԁ..0p..pDV!&16Document Style  @..  23  Ԁ     Ԉ X/&17Technical Document Style.. 23  Ԁ. X/&18Technical Document Style.. 23  Ԁ. X-(&19Technical Document Style 23  Ԁ   .. X*$&20Technical Document Style 23  Ԁ   .. X)$&21Technical Document Style 23  Ԁ   .. X83&22Technical Document Style  23  Ԁ     .. X/&23Technical Document Style.. 23  Ԁ. X/&24Technical Document Style.. 23  Ԁ. p4Heading 2Underlined Heading Flush LeftV 4Heading 1Centered Heading@..  b* 8Bullet ListIndented Bullet List0..0` ..` T&25Default Paragraph FontXXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOJ&26_Equation CaptionXXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOD& &27Document Style0..8` ..` D &28Document Style.. . D &29Document Style8..D &30Document Style0..D-&31Document Style 23  Ԁ   D& &32Document Style0..0` ..` fS &33Right-Aligned Paragraph Numbers..23  Ԁ..0..ff &34Right-Aligned Paragraph Numbers..` ..` 23  Ԁ..0` ..` D=&35Document Style.. 23  Ԁ   fy &36Right-Aligned Paragraph Numbers..` ..`  .. 23  Ԁ` ..` 0 .. f &37Right-Aligned Paragraph Numbers..` ..`  .. ..23  Ԁ .. 0..f &38Right-Aligned Paragraph Numbers..` ..`  .. ..h..h23  Ԁ..0h..hf &39Right-Aligned Paragraph Numbers..` ..`  .. ..h..h..23  Ԁh..h0..f &40Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....23  Ԁ..0..f &41Right-Aligned Paragraph Numbers..` ..`  .. ..h..h....p..p23  Ԁ..0p..pDV!&42Document Style  @..  23  Ԁ     Ԉ X/&43Technical Document Style.. 23  Ԁ. X/&44Technical Document Style.. 23  Ԁ. X-(&45Technical Document Style 23  Ԁ   .. X*$&46Technical Document Style 23  Ԁ   .. X)$&47Technical Document Style 23  Ԁ   .. X83&48Technical Document Style  23  Ԁ     .. X/&49Technical Document Style.. 23  Ԁ. X/&50Technical Document Style.. 23  Ԁ. L6Document 8Document 8XXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOL226Document 4Document 4                  L6Document 6Document 6XXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOL6Document 5Document 5XXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOL6Document 2Document 2S,xA`&UniversScalableS                O<6X9`("Courier 12cpiOL6Document 7Document 7XXXO<6X9`("Courier NewTTO                O<6X9`("Courier 12cpiOP8Right Par 1Right Par 1  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 2Right Par 2  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5L6Document 3Document 3S,xA`&UniversScalableS                O<6X9`("Courier 12cpiOP8Right Par 3Right Par 3  >4X` P  hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 4Right Par 4  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 5Right Par 5  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 6Right Par 6  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 7Right Par 7  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Right Par 8Right Par 8  >4X` hp x (#>S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5L6Document 1Document 1&    ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5'  P8Technical 5Technical 5  ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Technical 6Technical 6  ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Technical 2Technical 2S,xA`&UniversScalableS                O<6X9`("Courier 12cpiOP8Technical 3Technical 3S,xA`&UniversScalableS                O<6X9`("Courier 12cpiOP8Technical 4Technical 4  ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Technical 1Technical 1S,xA`&UniversScalableS                O<6X9`("Courier 12cpiOP8Technical 7Technical 7  ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5P8Technical 8Technical 8  ;1X` hp x (#;S,xA`&UniversScalableS                O<6X9`("Courier 12cpiO5+ ` hp x (#5<Q.footerfooter   (#        L&51_Equation Caption1XXXO<6X9`("Courier NewTTO        P8page numberpage numberXXXO<6X9`("Courier NewTTO        L~&52_Equation Caption1XXXO<6X9`("Courier NewTTO    T!:endnote textendnote text          ^:endnote refeendnote reference        V!:footnote texfootnote text          `:footnote reffootnote reference        L&53_Equation Caption2        < 9Z &Courier Regular'dxd Level 1 Level 2 Level 3 Level 4 Level 5(32G$ !     ($ (    ) M << deUU func{TC~=~V_celloverR_s} XX TCXX\  XJ3XXiVfcellXX}R^8s func{B~=~S_m~~M_m~~CS} XX}BXX}XX}S18mXX}XX}M&8mXX}XX}CS 5func{DF~=~{[SF_6]_direct}over{[SF_6]_spiked}} XX DFXX\ XJ3XXi[XXSFf6XX!]fdirectXXi}[XX}SF86XX!}]8spiked dTable_B&0 d d d $func{CS~=~{[anal]_dir}overDF} XXCSXX\XJXXix[XXxanalXXx]93dirXXJDF !func{CF~=~1over{1~+~Bover{CS}}} XXCFXX\xXJXX@1XXi1XX9X'XXEBXXFJCS .func{SD_s~=~sqrt{{sum{d_i}^2}over{2n}}} XXSD sXX X3XXTXXSXXESXXSXXSXXR XQXXpEIXXxd3iP2XXJ2XXJn #func{RSD~=~left(SDoverS_mright)} XX RSDXX XXEhXXjXXjXXiXX_EoXX_qXX_qXX_pX13XXPxSDXXa}S8m func{SD~=~sqrt{SD_s^2~+~SD_u^2}} XX}SDXX\}VXV,PXX}SD2OsXXS}XX#}SD32Ou +func{t~=~{left|~B~RIGHT|}overSDM}̀ XXtXXFXXA XA1 XXaEBXXJSDMTable_C(43$ !       '   XXX ` 218.InPart60,AppendixBisamendedbyaddingPerformanceSpecification15asfollows:#XXXT##q5# 8  APPENDIXBPERFORMANCESPECIFICATIONS   L *****  CRight ParRight-Aligned Paragraph NumbersI.A.1.a.(1)(a)i)a)3|T,xA`&UniversScalableixP7P<6X9`("Courier NewTTeXXx6X@DQX@<6X9`("Courier 12cpied6X@8;@* `*Times New RomanTTd PP*\  `*Times New RomanTTXx\  PCXP*\  `*Times New RomanTTd\  PCP<6X9`("Courier 10cpieTXx6X@8;X@($    HP LaserJet 5/5M - EnhancedHPBXLA0nLH< 9Z &Courier Regular k G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ Xk XXXX XXk Eq.1 #Xk XXXX##q XXk#qj% &36<9Z .Courier New Regular<9Z .Courier New Regular<09Z .Courier New Regular {FUNC{RMSD~=~SQRT{{left(1overnright)}~{SUMFROM{i~=~1}ton}{left(Asubi~~AsubMright)^2}}}   XXRMSDXXLHXXXTXXSXXESXXSXXSXXRXX hXXjXXiXX oXXqXXp5X[XXzF1XXzKnXX+Iqn8iq881X.pfX.b pmXXAiXXXX A M %2<09Z .Courier New Regular L G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.2 #XXX##q#<9Z .Courier New Regular L G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.3 #XXX##q#<9Z .Courier New Regular<09Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.4 <9Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_  XXXEq.5#XXX# #q#qXXX<9Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.6 <9Z .Courier New Regular<9Z .Courier New Regular<09Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.7 <9Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.8 <9Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.9 <9Z .Courier New Regular<09Z .Courier New Regular  G#   p$X` t  Mh<p x (#%'0*,.8135@8:<H?ACX_ XXX Eq.10 <9Z .Courier New Regular !     JXXXTRY3'3'3' Letter3'T7XXXdd7'XX XX'  @ 1 PERFORMANCESPECIFICATION15 "  @$PERFORMANCESPECIFICATIONFOREXTRACTIVEFTIRCONTINUOUS̜@  )EMISSIONSMONITORSYSTEMSINSTATIONARYSOURCES ' DXXX X'i  1.0ScopeandApplication .  Л 1.1Analytes.Thisperformancespecificationisapplicableformeasuringallhazardousairpollutants(HAPs)whichabsorbintheinfraredregionandcanbequantifiedusingFourierTransformInfraredSpectroscopy(FTIR),aslongastheperformancecriteriaofthisperformancespecificationaremet.ThisspecificationistobeusedforevaluatingFTIRcontinuousemissionmonitoringsystemsformeasuringHAPsregulatedunderTitleIIIofthe1990CleanAirActAmendments.ThisspecificationalsoappliestotheuseofFTIRCEMsformeasuringothervolatileorganicorinorganicspecies. 1.2Applicability.AsourcewhichcandemonstratethattheextractiveFTIRsystemmeetsthecriteriaofthisperformancespecificationforeachregulatedpollutantmayusetheFTIRsystemtocontinuouslymonitorfortheregulatedpollutants. ,$ 2.0SummaryofPerformanceSpecification. \  Z  ForcompoundspecificsamplingrequirementsrefertoFTIRsamplingmethods(e.g.,reference1).FordatareductionproceduresandrequirementsrefertotheEPAFTIRProtocol(reference2),hereafterreferredtoasthe"FTIRProtocol."Thisspecificationdescribessamplingandanalyticalproceduresforqualityassurance.Theinfraredspectrumofanyabsorbingcompoundprovidesadistinctsignature.Theinfraredspectrumofamixturecontainsthesuperimposedspectraofeachmixturecomponent.Thus,anFTIRCEMprovidesthecapabilitytocontinuouslymeasuremultiplecomponentsinasampleusingasingleanalyzer.Thenumberofcompoundsthatcanbespeciatedinasinglespectrumdepends,inpractice,onthespecificcompoundspresentandthetestconditions.  3.0Definitions.  | Л ForalistofdefinitionsrelatedtoFTIRspectroscopyrefertoAppendixAoftheFTIRProtocol.Unlessotherwisespecified,spectroscopicterms,symbolsandequationsinthisperformancespecificationaretakenfromtheFTIRProtocolorfromdocumentscitedintheProtocol. Additionaldefinitionsaregivenbelow. b+%*  3.1FTIRContinuousEmissionMonitoringSystem(FTIR CEM). 3.1.1FTIRSystem.Instrumenttomeasurespectrainthemidinfraredspectralregion(500to4000cm1).It  L containsaninfraredsource,interferometer,samplegascontainmentcell,infrareddetector,andcomputer.Theinterferometerconsistsofabeamsplitterthatdividesthebeamintotwopaths,onepathafixeddistanceandtheotheravariabledistance.Thecomputerisequippedwithsoftwaretoruntheinterferometerandstoretherawdigitizedsignalfromthedetector(interferogram).Thesoftwareperformsthemathematicalconversion(theFouriertransform)oftheinterferogramintoaspectrumshowingthefrequencydependentsampleabsorbance.Allspectraldatacanbestoredoncomputermedia. 3.1.2GasCell.Agascontainmentcellthatcanbeevacuated.Itcontainsthesampleastheinfraredbeampassesfromtheinterferometer,throughthesample,andtothedetector.Thegascellmayhavemultipassmirrorsdependingontherequireddetectionlimit(s)fortheapplication. 3.1.3SamplingSystem.Equipmentusedtoextractsamplefromthetestlocationandtransportthegastothe 2-&, FTIRanalyzer.Samplingsystemcomponentsincludeprobe,heatedline,heatednonreactivepump,gasdistributionmanifoldandvalves,flowmeasurementdevicesandanysampleconditioningsystems. 3.2ReferenceCEM.AnFTIRCEM,withsamplingsystem,thatcanbeusedforcomparisonmeasurements. 3.3InfraredBand(alsoAbsorbanceBandorBand).Collectionoflinesarisingfromrotationaltransitionssuperimposedonavibrationaltransition.Aninfraredabsorbancebandisanalyzedtodeterminetheanalyteconcentration. 3.4SampleAnalysis.Interpretinginfraredbandshapes,frequencies,andintensitiestoobtainsamplecomponentconcentrations.Thisisusuallyperformedbyasoftwareroutineusingaclassicalleastsquares(cls),partialleastsquares(pls),orKorPmatrixmethod. 3.5(Target)Analyte.Acompoundwhosemeasurementisrequired,usuallytosomeestablishedlimitofdetectionandanalyticaluncertainty. 3.6Interferant.Acompoundinthesamplematrixwhoseinfraredspectrumoverlapsatleastpartofananalytespectrumcomplicatingtheanalytemeasurement.Theinterferantmaynotpreventtheanalytemeasurement,but 2-&, couldincreasetheanalyticaluncertaintyinthemeasuredconcentration.Referencespectraofinterferantsareusedtodistinguishtheinterferantbandsfromtheanalytebands.Aninterferantforoneanalytemaynotbeaninterferantforotheranalytes. 3.7ReferenceSpectrum.Infraredspectraofananalyte,orinterferant,preparedundercontrolled,documented,andreproduciblelaboratoryconditions(seeSection4.6oftheFTIRProtocol).Asuitablelibraryofreferencespectracanbeusedtomeasuretargetanalytesingassamples. 3.8CalibrationSpectrum.InfraredspectrumofacompoundsuitableforcharacterizingtheFTIRinstrumentconfiguration(Section4.5intheFTIRProtocol). 3.9Onehundredpercentline.Adoublebeamtransmittancespectrumobtainedbecombiningtwosuccessivebackgroundsinglebeamspectra.Ideally,thislineisequalto100percenttransmittance(orzeroabsorbance)ateverypointinthespectrum.ThezeroabsorbancelineisusedtomeasuretheRMSnoiseofthesystem. 3.10BackgroundDeviation.Anydeviation(from100percent)intheonehundredpercentline(orfromzeroabsorbance).Deviationsgreaterthan5percentinany 2-&, analyticalregionareunacceptable.Suchdeviationsindicateachangeintheinstrumentthroughputrelativetothesinglebeambackground.̜ 3.11BatchSampling.Agascellisalternatelyfilledandevacuated.ASpectrumofeachfilledcell(onediscreetsample)iscollectedandsaved. 3.12ContinuousSampling.Sampleiscontinuouslyflowingthroughagascell.Spectraoftheflowingsamplearecollectedatregularintervals. 3.13ContinuousOperation.IncontinuousoperationanFTIRCEMsystem,withoutuserintervention,samplesfluegas,recordsspectraofsamples,savesthespectratoadisk,analyzesthespectraforthetargetanalytes,andprintsconcentrationsoftargetanalytestoacomputerfile.Userinterventionispermittedforinitialsetupofsamplingsystem,initialcalibrations,andperiodicmaintenance. 3.14SamplingTime.Inbatchsamplingthetimerequiredtofillthecellwithfluegas.Incontinuoussamplingthetimerequiredtocollecttheinfraredspectrumofthesamplegas. 3.15PPMMeters.Sampleconcentrationexpressedastheconcentrationpathlengthproduct,ppm(molar) 2-&, concentrationmultipliedbythepathlengthoftheFTIRgascell.Expressingconcentrationintheseunitsprovidesawaytodirectlycomparemeasurementsmadeusingsystemswithdifferentopticalconfigurations.Anotherusefulexpressionis(ppmmeters)/K,whereKistheabsolutetemperatureofthesampleinthegascell. 3.16CEMMeasurementTimeConstant.TheTimeConstant(TC,minutesforonecellvolumetoflowthroughthecell)determinestheminimumintervalforcompleteremovalofananalytefromtheFTIRcell.Itdependsonthesamplingrate(RsinLpm),theFTIRcellvolume(VcellinL)  andthechemicalandphysicalpropertiesofananalyte.̜RB>zx pX @Xdddddddd@E 'dYMD_ y   (#(#      (#(#  Forexample,ifthesampleflowrate(throughtheFTIRcell)is5Lpmandthecellvolumeis7liters,thenTCisequalto1.4minutes(0.71cellvolumesperminute).  This "@  performancespecificationdefines5*TCastheminimumintervalbetweenindependentsamples.̜ 3.17IndependentMeasurement.Twoindependentmeasurementsarespectraoftwoindependentsamples.Twoindependentsamplesareseparatedby,atleast5cellvolumes.Theintervalbetweenindependentmeasurements 2-&, dependsonthecellvolumeandthesampleflowrate(throughthecell).Thereisnomixingofgasbetweentwoindependentsamples.Alternatively,estimatetheanalyteresidencetimeempirically:(1)fillcelltoambientpressurewitha(knownanalyteconcentration)gasstandard,(2)measurethespectrumofthegasstandard,(3)purgethecellwithzerogasatthesamplingrateandcollectaspectrumeveryminuteuntiltheanalytestandardisnolongerdetectedspectroscopically.Ifthemeasuredtimecorrespondstolessthan5cellvolumes,use5*TCastheminimumintervalbetweenindependentmeasurements.Ifthemeasuredtimeisgreaterthan5*TC,thenusethistimeastheminimumintervalbetweenindependentmeasurements. 3.18TestCondition.Aperiodofsamplingwhereallprocess,andsamplingconditions,andemissionsremainconstantandduringwhichasinglesamplingtechniqueandasingleanalyticalprogramareused.OneRunmayincluderesultsformorethanonetestcondition.Constantemissionsmeansthatthecompositionoftheemissionsremainsapproximatelystablesothatasingleanalyticalprogramissuitableforanalyzingallofthesamplespectra.Agreaterthantwofoldchangeinanalyteorinterferantconcentrationsortheappearanceofadditionalcompoundsin 2-&, theemissions,mayconstituteanewtestconditionandmayrequiremodificationoftheanalyticalprogram. 3.19Run.AsingleRunconsistsofspectra(onespectrumeach)ofatleast10independentsamplesoveraminimumofonehour.Theconcentrationresultsfromthespectracanbeaveragedtogethertogivearunaverageforeachanalytemeasuredinthetestrun.     4.0Interferences. \  Л Severalcompounds,includingwater,carbonmonoxide,andcarbondioxide,areknowninterferencesintheinfraredregioninwhichtheFTIRinstrumentoperates.FollowtheproceduresintheFTIRprotocolforsubtractingorotherwisedealingwiththeseandotherinterferences. , ~$ 5.0Safety. B Л Theproceduresrequiredunderthisperformancespecificationmayinvolvehazardousmaterials,operations,andequipment.ThisperformancespecificationmaynotϜaddressallofthesafetyproblemsassociatedwiththeseprocedures.Itistheresponsibilityoftheusertoestablishappropriatesafetyandhealthpracticesanddeterminetheapplicableregulatorylimitationspriorto performingtheseprocedures.TheCEMSusersmanualand V+$* materialsrecommendedbythisperformancespecificationshouldbeconsultedforspecificprecautionstobetaken.   6.0EquipmentandSupplies.   Л 6.1InstallationofsamplingequipmentshouldfollowrequirementsofFTIRtestMethodssuchasreferences1and3andtheEPAFTIRProtocol(reference2).Selecttestpointswherethegasstreamcompositionisrepresentativeoftheprocessemissions.Ifcomparingtoareferencemethod,theprobetipsfortheFTIRCEMandtheRMshouldbepositionedclosetogetherusingthesamesampleportifpossible. 6.2FTIRSpecifications.TheFTIRCEMmustbeequippedwithreferencespectrabracketingtherangeofpathlengthconcentrations(absorbanceintensities)tobemeasuredforeachanalyte.Theeffectiveconcentrationrangeoftheanalyzercanbeadjustedbychangingthepathlengthofthegascellorbydilutingthesample.TheopticalconfigurationoftheFTIRsystemmustbesuchthatmaximumabsorbanceofanytargetanalyteisnogreaterthan1.0andtheminimumabsorbanceofanytargetanalyteisatleast10timestheRMSDnoiseintheanalyticalregion.Forexample,ifthemeasuredRMSDinananalyticalregionisequalto103,thenthepeakanalyteabsorbanceisrequired b+%* tobeatleast0.01.Adequatemeasurementofallofthe &-&, targetanalytesmayrequirechangingpathlengthsduringarun,conductingseparaterunsfordifferentanalytes,dilutingthesample,orusingmorethanonegascell. 6.3DataStorageRequirements.Thesystemmusthavesufficientcapacitytostorealldatacollectedinoneweekofroutinesampling.Datamustbestoredtoawriteprotectedmedium,suchaswriteoncereadmany(WORM)opticalstoragemediumortoapasswordprotectedremotestoragelocation.Abackupcopyofalldatacanbetemporarilysavedtothecomputerharddrive.Thefollowingitemsmustbestoredduringtesting."0 AtleastonesampleinterferogrampersamplingRunor l oneinterferogramperhour,whicheverisgreater.Thisassumesthatnosamplingoranalyticalconditionshavechangedduringtherun.  (# (# "0 Allsampleabsorbancespectra(about12perhr,288  | perday).  (# (# "0 Allbackgroundspectraandinterferograms(variable, ^$" butabout5perday).  (# (# "0 AllCTSspectraandinterferograms(atleast2each24 '!& hourperiod).  (# (# "0 Documentationshowingarecordofresolution,path n+%* length,apodization,samplingtime,sampling 2-&, conditions,andtestconditionsforallsample,CTS,calibration,andbackgroundspectra.  (# (# Л Usingaresolutionof0.5cm1,withanalyticalrange   of3500cm1,assumingabout65Kbytesperspectrumand130  L Kbperinterferogram,thestoragerequirementisabout164Mbforoneweekofcontinuoussampling.Lowerspectralresolutionrequireslessstoragecapacity.Alloftheabovedatamustbestoredforatleasttwoweeks.Aftertwoweeks,storagerequirementsinclude:(1)allanalyticalresults(calculatedconcentrations),(2)atleast1samplespectrumwithcorrespondingbackgroundandsampleinterferogramsforeachtestcondition,(3)CTSandcalibrationspectrawithatleastoneinterferogramforCTSandallinterferogramsforcalibrations,(4)arecordofanalyticalinputusedtoproduceresults,and(5)allotherdocumentation.Thesedatamustbestoredaccordingtotherequirementsoftheapplicableregulation. "@  7.0ReagentsandStandards.[Reserved] ^$" 8.0SampleCollection,Preservation,Storage,andTransport.[Reserved] '!& 9.0QualityControl.   )D#( Л TheseproceduresshallbeusedforperiodicquarterlyorsemiannualQA/QCchecksontheoperationoftheFTIRCEM. -&, Someprocedurestestonlytheanalyticalprogramandarenotintendedasatestofthesamplingsystem.  9.1AuditSample lA.Thiscanserveasacheckonboththesamplingsystemandtheanalyticalprogram. 9.1.1SampleRequirements.Theauditsamplecanbeamixtureorasinglecomponent.Itmustcontaintargetanalyte(s)atapproximatelytheexpectedfluegasconcentration(s).Ifpossible,eachmixturecomponentconcentrationshouldbeNISTtraceable(2percentaccuracy).Ifacylindermixturestandard(s)cannotbeobtained,then,alternatively,agasphasestandardcanbegeneratedfromacondensedphaseanalytesample.AuditsamplecontentsandconcentrationsarenotrevealedtotheFTIRCEMoperatoruntilaftersuccessfulcompletionofproceduresin5.3.2. 9.1.2TestProcedure.AnauditsampleisobtainedfromtheAdministrator.SpiketheauditsampleusingtheanalytespikeprocedureinSection11.TheauditsampleismeasureddirectlybytheFTIRsystem(undiluted)andthenspikedintotheeffluentataknowndilutionratio.Measureaseriesofspikedandunspikedsamplesusingthesameproceduresasthoseusedtoanalyzethestackgas.AnalyzetheresultsusingSections12.1and12.2.Themeasured 2-&, concentrationofeachanalytemustbewithin5percentoftheexpectedconcentration(plustheuncertainty),i.e.,thecalculatedcorrectionfactormustbewithin0.93and1.07foranauditwithananalyteuncertaintyof2percent.  9.2AuditSpectra %G.AuditspectracanbeusedtotesttheanalyticalprogramoftheFTIRCEM,butprovidenotestofthesamplingsystem. 9.2.1DefinitionandRequirements.Auditspectraareabsorbancespectrathat;(1)havebeenwellcharacterized,and(2)containabsorbancebandsoftargetanalyte(s)andpotentialinterferantsatintensitiesequivalenttowhatisexpectedinthesourceeffluent.Auditspectraareprovidedbytheadministratorwithoutidentifyinginformation.MethodsofpreparingAuditspectrainclude;(1)mathematicallyaddingsamplespectraoraddingreferenceandinterferantspectra,(2)obtainingsamplespectraofmixturespreparedinthelaboratory,or(3)theymaybesamplespectracollectedpreviouslyatasimilarsource.Inthelastcaseitmustbedemonstratedthattheanalyticalresultsarecorrectandreproducible.ArecordassociatedwitheachAuditspectrumdocumentsitsmethodofpreparation.ThedocumentationmustbesufficienttoenableanindependentanalysttoreproducetheAuditspectra. 2-&,  9.2.2TestProcedure.AuditspectraconcentrationsaremeasuredusingtheFTIRCEManalyticalprogram.Analyticalresultsmustbewithin5percentofthecertifiedauditconcentrationforeachanalyte(plustheuncertaintyintheauditconcentration).Iftheconditionisnotmet,demonstratehowtheauditspectraareunrepresentativeofthesamplespectra.Iftheauditspectraarerepresentative,modifytheFTIRCEManalyticalprogramuntilthetestrequirementismet.UsethenewanalyticalprograminsubsequentFTIRCEManalysesofeffluentsamples.  9.3SubmitSpectraForIndependentAnalysis M.ThisproceduretestsonlytheanalyticalprogramandnottheFTIRCEMsamplingsystem.TheanalystcansubmitFTIRCEMspectraforindependentanalysisbyEPA.Requirementsforsubmissioninclude;(1)threerepresentativeabsorbancespectra(andstoredinterferograms)foreachtestperiodtobereviewed,(2)correspondingCTSspectra,(3)correspondingbackgroundspectraandinterferograms,(4)spectraofassociatedspikedsamplesifapplicable,and(5)analyticalresultsforthesesamplespectra.Theanalystwillalsosubmitdocumentationofprocesstimesandconditions,samplingconditionsassociatedwitheach 2-&, spectrum,filenamesandsamplingtimes,methodofanalysisandreferencespectraused,opticalconfigurationofFTIRCEMincludingcellpathlengthandtemperature,spectralresolutionandapodizationusedforeveryspectrum.IndependentanalysiscanalsobeperformedonsiteinconjunctionwiththeFTIRCEMsamplingandanalysis.SamplespectraarestoredontheindependentanalyticalsystemastheyarecollectedbytheFTIRCEMsystem.TheFTIRCEMandtheindependentanalysesarethenperformedseparately.Thetwoanalyseswillagreetowithin20percentforeachanalyteusingtheprocedureinSection12.3.Thisassumesbothanalyticalroutineshaveproperlyaccountedfordifferencesinopticalpathlength,resolution,andtemperaturebetweenthesamplespectraandthereferenceϜspectra.  10.0CalibrationandStandardization.  | Л 10.1CalibrationTransferStandards.ForCTSrequirementsseeSection4.5oftheFTIRProtocol.AwellcharacterizedabsorbancebandintheCTSgasisusedtomeasurethepathlengthandlineresolutionoftheinstrument.TheCTSmeasurementsmadeatthebeginningofevery24hourperiodmustagreetowithin5percentaftercorrectionfordifferencesinpressure. &-&, VerifythatthefrequencyresponseoftheinstrumentandCTSabsorbanceintensityarecorrectbycomparingtootherCTSspectraorbyreferringtotheliterature. 10.2AnalyteCalibration.IfEPAlibraryreferencespectraarenotavailable,usecalibrationstandardstopreparereferencespectraaccordingtoSection6oftheFTIRProtocol.Asuitablesetofanalytereferencedataincludesspectraofatleast2independentsamplesateachofatleast2differentconcentrations.Theconcentrationsbracketarangethatincludestheexpectedanalyteabsorbanceintensities.Thelinearfitofthereferenceanalytebandareasmusthaveafractionalcalibrationuncertainty(FCUinAppendixFoftheFTIRProtocol)ofnogreaterthan10percent.ForrequirementsofanalytestandardsrefertoSection4.6oftheFTIRProtocol. 10.3SystemCalibration.Thecalibrationstandardisintroducedatapointonthesamplingprobe.Thesamplingsystemispurgedwiththecalibrationstandardtoverifythattheabsorbancemeasuredinthiswayisequaltotheabsorbanceintheanalytecalibration.Notethatthesystemcalibrationgivesnoindicationoftheabilityofthesamplingsystemtotransportthetargetanalyte(s)underthetestconditions. 2-&,  10.4AnalyteSpike.Thetargetanalyte(s)isspikedattheoutletofthesamplingprobe,upstreamoftheparticulatefilter,andcombinedwitheffluentataratioofabout1partspiketo9partseffluent.Themeasuredabsorbanceofthespikeiscomparedtotheexpectedabsorbanceofthespikeplustheanalyteconcentrationalreadyintheeffluent.Thismeasuressamplingsystembias,ifany,asdistinguishedfromanalyzerbias.Itisimportantthatspikedsamplepassthroughallofthesamplingsystemcomponentsbeforeanalysis. 10.5SignaltoNoiseRatio(S/N).ThemeasureofS/Ninthisperformancespecificationistherootmeansquare(RMS)noiselevelasgiveninAppendixCoftheFTIRProtocol.TheRMSnoiselevelofacontiguoussegmentofaspectrumisdefinedastheRMSdifference(RMSD)betweenthencontiguousabsorbancevalues(Ai)whichformthesegment  | andthemeanvalue(AM)ofthatsegment. "@  М\LHzx  x pX @Xdddddddd@E%k$ 8d pJ*_ ^$   (#(#           (#(#  AdecreaseintheS/Nmayindicatealossinoptical throughput,ordetectororinterferometermalfunction. n+%*  10.6BackgroundDeviation.The100percentbaseline mustbebetween95and105percenttransmittance(absorbanceof0.02to0.02)ineveryanalyticalregion.Whenbackgrounddeviationexceedsthisrange,anewbackgroundspectrummustbecollectedusingnitrogenorotherzerogas. 10.7DetectorLinearity.MeasurethebackgroundandCTSatthreeinstrumentaperturesettings;oneattheaperturesettingtobeusedinthetesting,andoneeachatsettingsonehalfandtwicethetestaperturesetting.ComparethethreeCTSspectra.CTSbandareasshouldagreetowithintheuncertaintyofthecylinderstandard.Iftestapertureisthemaximumaperture,collectCTSspectrumatmaximumaperture,thenclosetheaperturetoreducetheIRthroughputbyhalf.CollectasecondbackgroundandCTSatthesmalleraperturesettingandcomparethespectraasabove.Insteadofchangingtheapertureneutraldensityfilterscanbeusedtoattenuatetheinfraredbeam.SetuptheFTIRsystemasitwillbeusedinthetestmeasurements.CollectaCTSspectrum.Useaneutraldensityfiltertoattenuatetheinfraredbeam(eitherimmediatelyafterthesourceortheinterferometer)toapproximatelyitsoriginalintensity.CollectasecondCTSspectrum.Useanotherfiltertoattenuatetheinfraredbeamto 2-&, approximately1/4itsoriginalintensity.CollectathirdbackgroundandCTSspectrum.ComparetheCTSspectraasabove.Anothercheckonlinearityistoobservethesinglebeambackgroundinfrequencyregionswheretheopticalconfigurationisknowntohaveazeroresponse.Verifythatthedetectorresponseis"flat"andequaltozerointheseregions.Ifdetectorresponseisnotlinear,decreaseaperture,orattenuatetheinfraredbeam.Repeatthelinearitycheckuntilsystempassestherequirement. z  11.0AnalyticalProcedure. > Л 11.1InitialCertification.First,performtheevaluationproceduresinSection6.0oftheFTIRProtocol.TheperformanceofanFTIRCEMcanbecertifieduponinstallationusingEPAMethod301typevalidation(40CFR,Part63,AppendixA),orbycomparisontoareferenceMethodifoneexistsforthetargetanalyte(s).Detailsofeachprocedurearegivenbelow.Validationtestingisusedforinitialcertificationuponinstallationofanewsystem.Subsequentperformancecheckscanbeperformedwithmorelimitedanalytespiking.Performanceoftheanalyticalprogramischeckedinitially,andperiodicallyasrequired byEPA,byanalyzingauditspectraorauditgases. b+%*    11.1.1Validation j.UseEPAMethod301typesampling (reference4,Section5.3ofMethod301)tovalidatetheFTIRCEMformeasuringthetargetanalytes.Theanalytespikeprocedureisasfollows;(1)aknownconcentrationofanalyteismixedwithaknownconcentrationofanonreactivetracergas,(2)theundilutedspikegasissentdirectlytotheFTIRcellandaspectrumofthissampleiscollected,(3)preheatthespikedgastoatleastthesamplelinetemperature,(4)introducespikegasatthebackofthesampleprobeupstreamoftheparticulatefilter,(5)spikedeffluentiscarriedthroughallsamplingcomponentsdownstreamoftheprobe,(6)spikeataratioofroughly1partspiketo9partsfluegas(ormoredilute),(7)thespiketofluegasratioisestimatedbycomparingthespikeflowtothetotalsampleflow,and(8)thespikeratioisverifiedbycomparingthetracerconcentrationinspikedfluegastothetracerconcentrationinundilutedspikegas.Theanalytefluegasconcentrationisunimportantaslongasthespikedcomponentcanbemeasuredandthesamplematrix(includinginterferences)issimilartoitscompositionundertestconditions.ValidationcanbeperformedusingasingleFTIRCEManalyzingsamplespectracollectedsequentially.Sincefluegasanalyte(unspiked) 2-&, concentrationscanvary,itisrecommendedthattwoseparatesamplinglines(andpumps)areused;onelinetocarryunspikedfluegasandtheotherlinetocarryspikedfluegas.Evenwithtwosamplinglinesthevariationinunspikedconcentrationmaybefastcomparedtotheintervalbetweenconsecutivemeasurements.Alternatively,twoFTIRCEMscanbeoperatedsidebyside,onemeasuringspikedsample,theotherunspikedsample.Inthisarrangementspikedandunspikedmeasurementscanbesynchronizedtominimizetheaffectoftemporalvariationintheunspikedanalyteconcentration.Ineithersamplingarrangement,theintervalbetweenmeasuredconcentrationsusedinthestatisticalanalysisshouldbe,atleast,5cellvolumes(5*TCinequation1).Avalidationrunconsistsof,atleast,24independentanalyticalresults,12spikedand12unspikedsamples.SeeSection3.17fordefinitionofan"independent"analyticalresult.TheresultsareanalyzedusingSections12.1and12.2todetermineifthemeasurementspassedthevalidationrequirements.Severalanalytescanbespikedandmeasuredinthesamesamplingrun,butaseparatestatisticalanalysisisperformedforeachanalyte.Inlieuof24independentmeasurements,averagedresultscanbeusedinthestatisticalanalysis. 2-&, Inthisprocedure,aseriesofconsecutivespikedmeasurementsarecombinedoverasamplingperiodtogiveasingleaverageresult.Therelatedunspikedmeasurementsareaveragedinthesameway.Theminimum12spikedand12unspikedresultaveragesareobtainedbyaveragingmeasurementsoversubsequentsamplingperiodsofequalduration.TheaveragedresultsaregroupedtogetherandstatisticallyanalyzedusingSection12.2. 11.1.1.1ValidationwithaSingleAnalyzerandSamplingLine.Ifonesamplinglineisused,connectthesamplingsystemcomponentsandpurgetheentiresamplingsystemandcellwithatleast10cellvolumesofsamplegas.Beginsamplingbycollectingspectraof2independentunspikedsamples.Introducethespikegasintothebackoftheprobe,upstreamoftheparticulatefilter.Allow10cellvolumesofspikedfluegastopurgethecellandsamplingsystem.Collectspectraof2independentspikedsamples.Turnoffthespikeflowandallow10cellvolumesofunspikedfluegastopurgetheFTIRcellandsamplingsystem.Repeatthisprocedure6timesuntilthe24samplesarecollected.Spikedandunspikedsamplescanalsobemeasuredingroupsof4insteadofinpairs.AnalyzetheresultsusingSections12.1and12.2.Ifthestatistical 2-&, analysispassesthevalidationcriteria,thenthevalidationiscompleted.Iftheresultsdonotpassthevalidation,thecausemaybethattemporalvariationsintheanalytesamplegasconcentrationarefastrelativetotheintervalbetweenmeasurements.Thedifficultymaybeavoidedby:(1)averagingthemeasurementsoverlongsamplingperiodsandusingtheaveragedresultsinthestatisticalanalysis,(2)modifyingthesamplingsystemtoreduceTCby,forexample,usingasmallervolumecellorincreasingthesampleflowrate,(3)usingtwosamplelines(4)usetwoanalyzerstoperformsynchronizedmeasurements.ThisperformancespecificationpermitsmodificationsinthesamplingsystemtominimizeTCiftheotherrequirementsofthevalidationsamplingprocedurearemet. 11.1.1.2ValidationWithaSingleAnalyzerandTwoSamplingLines.Analternativesamplingprocedureusestwoseparatesamplelines,onecarryingspikedfluegas,theothercarryingunspikedgas.Avalveinthegasdistributionmanifoldallowstheoperatortochooseeithersample.AshortheatedlineconnectstheFTIRcelltothe3wayvalveinthemanifold.Bothsamplinglinesarecontinuouslypurged.Eachsamplelinehasarotameterandabypassventlineaftertherotameter,immediatelyupstream 2-&, ofthevalve,sothatthespikeandunspikedsampleflowscaneachbecontinuouslymonitored.Beginsamplingbycollectingspectraof2independentunspikedsamples.TurnthesamplingvalvetocloseofftheunspikedgasflowandallowthespikedfluegastoentertheFTIRcell.Isolateandevacuatethecellandfillwiththespikedsampletoambientpressure.(Whiletheevacuatedcellisfilling,preventairleaksintothecellbymakingsurethatthespikesamplerotameteralwaysindicatesthataportionoftheflowisdirectedoutthebypassvent.)Openthecelloutletvalvetoallowspikedsampletocontinuouslyflowthroughthecell.Measurespectraof2independentspikedsamples.Repeatthisprocedureuntilatleast24samplesarecollected. 11.1.1.3SynchronizedMeasurementsWithTwoAnalyzers.UsetwoFTIRanalyzers,eachwithitsowncell,toperformsynchronizedspikedandunspikedmeasurements.Ifpossible,useasimilaropticalconfigurationforbothsystems.TheopticalconfigurationsarecomparedbymeasuringthesameCTSgaswithbothanalyzers.EachFTIRsystemusesitsownsamplingsystemincludingaseparatesamplingprobeandsamplingline.Acommongasdistributionmanifoldcanbeusedifthesamplesarenevermixed.One 2-&, samplingsystemandanalyzermeasuresspikedeffluent.Theothersamplingsystemandanalyzermeasuresunspikedfluegas.Thetwosystemsaresynchronizedsothatsothateachmeasuresspectraatapproximatelythesametimes.Thesampleflowratesarealsosynchronizedsothatbothsamplingratesareapproximatelythesame(TC1~TC2in .  equation1).Startbothsystemsatthesametime.Collectspectraofatleast12independentsampleswitheach(spikedandunspiked)systemtoobtaintheminimum24measurements.AnalyzetheanalyticalresultsusingSections12.1and12.2.Runaveragescanbeusedinthestatisticalanalysisinsteadofindividualmeasurements.   11.1.1.4ComparetoaReferenceMethod(RM). ԀObtainEPAapprovalthatthemethodqualifiesasanRMfortheanalyte(s)andthesourcetobetested.FollowthepublishedproceduresfortheRMinpreparingandsettingupequipmentandsamplingsystem,performingmeasurements,andreportingresults.SinceFTIRCEMShavemulticomponentcapability,itispossibletoperformmorethanoneRMsimultaneously,oneforeachtargetanalyte.Conductatleast9runswheretheFTIRCEMandtheRMaresamplingsimultaneously.EachRunisatleast30minuteslongandconsistsofspectraofatleast5independentFTIRCEM 2-&, samplesandthecorrespondingRMmeasurements.Ifmorethan9runsareconducted,theanalystmayeliminateupto3runsfromtheanalysisifatleast9runsareused. 11.1.1.4.1RMsUsingIntegratedSampling.PerformtheRMandFTIRCEMsamplingsimultaneously.TheFTIRCEMcanmeasurespectraasfrequentlyastheanalystchooses(andshouldobtainmeasurementsasfrequentlyaspossible)providedthatthemeasurementsincludespectraofatleast5independentmeasurementsevery30minutes.ConcentrationresultsfromalloftheFTIRCEMspectrawithinarunmaybeaveragedforuseinthestatisticalcomparisonevenifallofthemeasurementsarenotindependent.WhenaveragingtheFTIRCEMconcentrationswithinarun,itispermittedtoexcludesomemeasurementsfromtheaverageprovidedtheminimumof5independentmeasurementsevery30minutesareincluded:TheRunaverageoftheFTIRCEMmeasurementsdependsonboththesampleflowrateandthemeasurementfrequency(MF).TherunaverageoftheRMusingtheintegratedsamplingmethoddependsprimarilyonitssamplingrate.Ifthetargetanalyteconcentrationfluctuatessignificantly,thecontributiontotherunaverageofalargefluctuationdependsonthesamplingrateandmeasurementfrequency,andonthedurationandmagnitudeof 2-&, thefluctuation.Itis,therefore,importanttocarefullyselectthesamplingrateforboththeFTIRCEMandtheRMandthemeasurementfrequencyfortheFTIRCEM.Theminimumof9runaveragescanbecomparedaccordingtotherelativeaccuracytestprocedureinPerformanceSpecification2forSO2andNOxCEMs(40CFR,Part60,App.B). .   11.1.1.4.2RMsUsingaGrabSamplingTechnique.SynchronizetheRMandFTIRCEMmeasurementsascloselyaspossible.ForagrabsamplingRMrecordthevolumecollectedandtheexactsamplingperiodforeachsample.SynchronizetheFTIRCEMsothattheFTIRmeasuresaspectrumofasimilarcellvolumeatthesametimeastheRMgrabsamplewascollected.Measureatleast5independentsampleswithboththeFTIRCEMandtheRMforeachoftheminimum9Runs.ComparetheRunconcentrationaveragesbyusingtherelativeaccuracyanalysisprocedurein40CFR,Part60,App.B. 11.1.1.4.3ContinuousEmissionMonitors(CEMs)asRMs.IftheRMisaCEM,synchronizethesamplingflowratesoftheRMandtheFTIRCEM.Eachrunisatleast1hourlongandconsistsofatleast10FTIRCEMmeasurementsandthecorresponding10RMmeasurements(oraverages).Forthestatisticalcomparisonusetherelativeaccuracy 2-&, analysisprocedurein40CFR,Part60,App.B.IftheRMtimeconstantis<12theFTIRCEMtimeconstant,brief  fluctuationsinanalyteconcentrationswhicharenotadequatelymeasuredwiththeslowerFTIRCEMtimeconstantcanbeexcludedfromtherunaveragealongwiththecorrespondingRMmeasurements.However,theFTIRCEMrunaveragemuststillincludeatleast10measurementsovera1hrperiod. \  12.0CalculationsandDataAnalysis. z  Л 12  .1SpikeDilutionRatio ,ExpectedConcentration.TheMethod301biasiscalculatedasfollows.̜RB>zx pX @Xdddddddd@E dJ_ Kߛ   (#(#      (#(#  where̜'D4 LXDX'B 4 = Biasatthespikelevel  МSm= Meanoftheobservedspikedsampleconcentrations  p МMm=0 Meanoftheobservedunspikedsampleconcentrations"4  (# (# МCS= Expectedvalueofthespikedconcentration. R$" Л 4 TheCSisdeterminedbycomparingtheSF6tracer &$ concentrationinundilutedspikegastotheSF6tracer '!& concentrationsinthespikedsamples;̜RB>zx pX @Xdddddddd@Eb+ dM@_ b+yߛ   (#(#      &-&,   Theexpectedconcentration(CS)isthemeasuredconcentrationoftheanalyteinundilutedspikegasdividedbythedilutionfactor̜RB>zx pX @Xdddddddd@E $d\@_  F   (#(#      (#(#  where:̜3D4 N XD4 LX3[anal]dir󛀜=0 N Theanalyteconcentrationinundilutedspikegas   measureddirectlybyfillingtheFTIRcellwiththespikegas. N (#N (# МIfthebiasisstatisticallysignificant(Section12.2),Method301requiresthatacorrectionfactor,CF,bemultipliedbytheanalyticalresults,andthat0.7CF l 1.3.̜RB>zx pX @Xdddddddd@EN d@c_ N   (#(#           (#(#    * DXD4 N X* 12.2StatisticalAnalysis ԀofValidationMeasurements.Arrangetheindependentmeasurements(ormeasurementaverages)asinTable1.Morethan12pairsofmeasurementscanbeanalyzed.ThestatisticalanalysisfollowsEPAMethod301,Section6.3.Section12.1ofthisperformancespecificationshowsthecalculationsforthebias,expectedspikeconcentration,andcorrectionfactor.ThisSections 2-&, showsthedeterminationofthestatisticalsignificanceofthebias.Determinethestatisticalsignificanceofthebiasatthe95percentconfidencelevelbycalculatingthetvalueforthesetofmeasurements.First,calculatethedifferences,di,foreachpairofspikedandeachpairof j  unspikedmeasurements.Thencalculatethestandarddeviationofthespikedpairsofmeasurements.̜RB>zx pX @Xdddddddd@E 8hdp@*_ ߛ   (#(#           (#(#  where̜*D XDX*di  =0 Thedifferencesbetweenpairsofspikedmeasurements.l (# (# МSds  = Thestandarddeviationinthedivalues. 0 Мn  =0 Thenumberofspikedpairs,2n=12fortheminimum N of12spikedand12unspikedmeasurements.  (# (# Calculatetherelativestandarddeviation,RSD,usingSDs  | andthemeanofthespikedconcentrations,Sm.TheRSDmust "@  be50%. ^$" МRB>zx pX @Xdddddddd@E"& d$@_ "&P   (#(#      (#(#  Repeatthecalculationsinequations7and8todetermine SDuandRSD,respectively,fortheunspikedsamples. n+%* МCalculatethestandarddeviationofthemeanusingSDsand Z  SDufromequation7.  МRB>zx pX @Xdddddddd@E Nd@_     (#(#      (#(#  Thetstatisticiscalculatedasfollowstotestthebiasforstatisticalsignificance;̜R B>zx pX @Xdddddddd@E d@_     (#(#      (#(#  wherethebias,B,andthecorrectionfactor,CF,aregiveninSection12.1.For11degreesoffreedom,andaonetaileddistribution,Method301requiresthatt2.201.If  thetstatisticindicatesthebiasisstatisticallysignificant,thenanalyticalmeasurementsmustbemultipliedbythecorrectionfactor.Thereisnolimitationonthenumberofmeasurements,buttheremustbeatleast12independentspikedand12independentunspikedmeasurements.Refertothetdistribution(Table2)atthe95percentconfidencelevelandappropriatedegreesoffreedomforthecriticaltvalue. "&$ 16.0References. '!& М* DXD X* 1.Method318,40CFR,Part63,AppendixA(Draft), "MeasurementofGaseousFormaldehyde,PhenolandMethanol b+%* EmissionsbyFTIRSpectroscopy,"EPAContractNo.68D20163,WorkAssignment218,February,1995.  2."EPAProtocolfortheUseofExtractiveFourierTransformInfrared(FTIR)SpectrometryinAnalysesofGaseousEmissionsfromStationaryIndustrialSources,"February,1995.̜ 3."MeasurementofGaseousOrganicandInorganicEmissionsbyExtractiveFTIRSpectroscopy,"EPAContractNo.68D20165,WorkAssignment308.̜ 4."Method301FieldValidationofPollutantMeasurementMethodsfromVariousWasteMedia," 40CFR63,  AppA . p 17.0Tables,Diagrams,Flowcharts,andValidationData. 8 М  J    TABLE1.ARRANGEMENTOFVALIDATIONMEASUREMENTSFOR Z STATISTICALANALYSIS.  @ SӀ*q ddd Xdd Xdd X(#(#q,dd , d ,td ,)td ,td ,td +  /   ,l/Measurement(oraverage) ;*Z " ;Time ;*v " ;Spiked(ppm) ;*Z " ;dispiked ;*v  " ;Unspiked(ppm) ;*Z  " ;diunspiked OB*v  " l0 ,  LO1 9* ^ " 9 9* ^" 9S1 9* ^" 9 9* ^" 9U1 9* ^" 9 OB* ^" L   LO2 9*\ " 9 9*\ " 9S2 9*\ " 9S2󀄀S1 9*\ " 9U2 9*\ " 9U2󀄀U1 OB*\ " L   LO3 9*" 9 9*" 9S3 9*" 9 9*" 9U3 9*" 9 OB*" L   LO4 9*J " 9 9*J " 9S4 9*J !" 9S4󀄀S3 9*J "" 9U4 9*J #" 9U4󀄀U3 OB*J $" L   LO5 9*H %" 9 9*H &" 9S5 9*H '" 9 9*H (" 9U5 9*H )" 9 OB*H *" L   LO6 9* +" 9 9* ," 9S6 9* -" 9S6󀄀S5 9* ." 9МU6 9* /" 9U6󀄀U5 OB* 0" L   LO7 9*61" 9 9*62" 9S7 9*63" 9 9*64" 9U7 9*65" 9 OB*66" L   LO8 9*47" 9 9*48" 9S8 9*49" 9S8󀄀S7 9*4:" 9U8 9*4;" 9U8󀄀U7 OB*4<" L   LO9 9*~=" 9 9*~>" 9S9 9*~?" 9 9*~@" 9U9 9*~A" 9 OB*~B" L   LO10 9*|"C" 9 9*|"D" 9S10 9*|"E" 9S10󀄀S9 9*|"F" 9U10 9*|"G" 9U10U9 OB*|"H" L   LO11 9* I" 9 9* J" 9S11 9* K" 9 9* L" 9U11 9* M" 9 O>* N" L   LO12 ;*jO" ; ;*jP" ;S12 ;*jQ" ;S12󀄀S11 ;*jR" ;U12 ;*jS" ;U12U11 O>*jT" L0   LOAverage> ;*hU" ; ;*hV" ;Sm ;*hW" ; ;*hX" ;Mm ;*hY" ;:0.hZ" L0   :  !J[   TABLE2.tVALUES.5  Z j* dddd d td )td td td (#(#,!<< ,L< ,l< ,L< ,(l< ,L< ,l< ,L< +  /" $ p/n1a 9*r" 9tvalue 9*r" 9n1a 9*r" 9tvalue 9*r" 9n1a 9*r" 9tvalue 9*r" 9n1a 9*r " 9tvalue OB*r " p  PO11 9*6  " 92.201 9*6  " 917 9*6  " 92.110 9*6 " 923 9*6 " 92.069 9*6 " 929 9*6 " 92.045 LB*6 " P  @xL12 , 6" ,2.179 9* 6" 918 9* 6" 92.101 9* 6" 924 9* 6" 92.064 9* 6" 930 9* 6" 92.042 LB* 6" x @  @xL13 , " ,2.160 9* " 919 9* " 92.093 9* " 925 9* " 92.060 9*  " 940 9* !" 92.021 LB* "" x @  @xL14 ,D#" ,2.145 9*D$" 920 9*D%" 92.086 9*D&" 926 9*D'" 92.056 9*D(" 960 9*D)" 92.000 LB*D*" x @  @xL15 ,D +" ,2.131 9*D ," 921 9*D -" 92.080 9*D ." 927 9*D /" 92.052 9*D 0" 9120 9*D 1" 91.980 O>*D 2" x @  xO16 ;* 3" ;2.120 ;* 4" ;22 ;* 5" ;2.074 ;* 6" ;28 ;* 7" ;2.048 ;* 8" ; ;* 9" ;1.960:0. :" x0    :(a)nisthenumberofindependentpairsofmeasurements(apairconsistsofonespikedanditscorrespondingunspikedmeasurement).Eitherdiscreet(independent)measurementsinasinglerun,orrunaveragescanbeused.*****