NaOH modified WO3/SiO2 catalysts for propylene production from 2-butene and ethylene metathesis

Article abstract of DOI:10.1016/S1872-2067(12)60760-8

A WO₃/SiO₂ catalyst is used in industry to produce propylene from 2-butene and ethylene metathesis. Catalysts with various WO ₃ loading (4% to 10%) were prepared by impregnation and tested for the metathesis of ethene and trans-2-butene. Ion exchange of NaOH onto the WO₃/SiO₂ catalyst was used to mitigate the acidity of the catalysts in a controlled way. At low WO₃ loading, the treatment with large amounts of NaOH resulted in a significant decrease in metathesis activity concomitant with significant W leaching and marked structural changes (XRD, Raman). At higher WO₃ loading (6% to 10%), the treatment with NaOH mainly resulted in a decrease in acidity. FT-IR experiments after adsorption of pyridine showed that the Lewis acidic sites were poisoned by sodium. Nevertheless, the metathesis activity remained constant after the NaOH treatment. This suggested that the remaining acidity on the catalyst was enough to ensure the efficient formation of the carbene active sites. Interestingly, Na poisoning resulted in some modification of the selectivity. The mitigation of acidity was shown to favor propene selectivity over the formation of isomerization products (cis-2-butene, 1-butene, etc.). Moreover, treatment with NaOH led to a shorter induction period and reduced coke formation on the WO ₃/SiO₂ catalyst.

Full text of DOI:10.1016/S1872-2067(12)60760-8

ChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241
ꢀ
ꢀ ꢀ ꢀ
ꢀ ꢀ ꢀ ꢀ
ava i l a b l e a t w w w. s c i e n c ed i r e c t . c o m
ꢀ
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e /c h n j c
ꢀ
ꢀ
Article
NaOHꢀmodifiedꢀWO
3
/SiO ꢀcatalystsꢀforꢀpropyleneꢀproductionꢀfromꢀ
2
2‐buteneꢀandꢀethyleneꢀmetathesisꢀ
ꢀ
a
ꢀb
ꢀa,
ꢀa
SurasaꢀMaksasithorn ,ꢀDamienꢀP.ꢀDebecker ,ꢀPiyasanꢀPraserthdam *,ꢀJoongjaiꢀPanpranot ,ꢀ
ꢀc
ꢀc
KongkiatꢀSuriye ,ꢀSirachayaꢀKunjaraꢀNaꢀAyudhya
ꢀ
aꢀCenterꢀofꢀExcellenceꢀonꢀCatalysisꢀandꢀCatalyticꢀReactionꢀEngineering,ꢀDepartmentꢀofꢀChemicalꢀEngineering,ꢀFacultyꢀofꢀEngineering,ꢀChulalongkornꢀ ꢀ
University,ꢀBangkokꢀ10330,ꢀThailandꢀ
bꢀInstituteꢀofꢀCondensedꢀMatterꢀandꢀNanoscienceꢀ–ꢀMolecules,ꢀSolidsꢀandꢀReactivityꢀ(IMCN/MOST),ꢀUniversitéꢀcatholiqueꢀdeꢀLouvain,ꢀCroixꢀduꢀSudꢀ2/17,ꢀ
1348ꢀLouvain‐La‐Neuve,ꢀBelgiumꢀ
cꢀSCGꢀChemicals,ꢀCo.,ꢀLtd.,ꢀ1ꢀSiamꢀCementꢀRoad,ꢀBangsue,ꢀBangkokꢀ10800,ꢀThailandꢀ
A R T I C L E ꢀ I N F O ꢀ
A B S T R A C T ꢀ
/SiO
Articleꢀhistory:ꢀ
AꢀWO
sis.ꢀCatalystsꢀwithꢀvariousꢀWO
theꢀmetathesisꢀofꢀetheneꢀandꢀtrans‐2‐butene.ꢀIonꢀexchangeꢀofꢀNaOHꢀontoꢀtheꢀWO
usedꢀtoꢀmitigateꢀtheꢀacidityꢀofꢀtheꢀcatalystsꢀinꢀaꢀcontrolledꢀway.ꢀAtꢀlowꢀWO ꢀloading,ꢀtheꢀtreatmentꢀ
withꢀlargeꢀamountsꢀofꢀNaOHꢀresultedꢀinꢀaꢀsignificantꢀdecreaseꢀinꢀmetathesisꢀactivityꢀconcomitantꢀ
withꢀsignificantꢀWꢀleachingꢀandꢀmarkedꢀstructuralꢀchangesꢀ(XRD,ꢀRaman).ꢀAtꢀhigherꢀWO ꢀloadingꢀ
6%ꢀtoꢀ10%),ꢀtheꢀtreatmentꢀwithꢀNaOHꢀmainlyꢀresultedꢀinꢀaꢀdecreaseꢀinꢀacidity.ꢀFT‐IRꢀexperimentsꢀ
3
2
ꢀcatalystꢀisꢀusedꢀinꢀindustryꢀtoꢀproduceꢀpropyleneꢀfromꢀ2‐buteneꢀandꢀethyleneꢀmetathe‐
ꢀloadingꢀ(4%ꢀtoꢀ10%)ꢀwereꢀpreparedꢀbyꢀimpregnationꢀandꢀtestedꢀforꢀ
/SiO ꢀcatalystꢀwasꢀ
Receivedꢀ25ꢀJulyꢀ2013ꢀ
Acceptedꢀ31ꢀOctoberꢀ2013ꢀ
Publishedꢀ20ꢀFebruaryꢀ2014ꢀ
3
3
2
3
3
Keywords:ꢀ
Propyleneꢀ
(
afterꢀadsorptionꢀofꢀpyridineꢀshowedꢀthatꢀtheꢀLewisꢀacidicꢀsitesꢀwereꢀpoisonedꢀbyꢀsodium.ꢀNever‐
theless,ꢀtheꢀmetathesisꢀactivityꢀremainedꢀconstantꢀafterꢀtheꢀNaOHꢀtreatment.ꢀThisꢀsuggestedꢀthatꢀtheꢀ
remainingꢀacidityꢀonꢀtheꢀcatalystꢀwasꢀenoughꢀtoꢀensureꢀtheꢀefficientꢀformationꢀofꢀtheꢀcarbeneꢀactiveꢀ
sites.ꢀInterestingly,ꢀNaꢀpoisoningꢀresultedꢀinꢀsomeꢀmodificationꢀofꢀtheꢀselectivity.ꢀTheꢀmitigationꢀofꢀ
acidityꢀ wasꢀ shownꢀ toꢀ favorꢀ propeneꢀ selectivityꢀ overꢀ theꢀ formationꢀ ofꢀ isomerizationꢀ productsꢀ
Olefinꢀmetathesisꢀ
Tungstenꢀtrioxideꢀ
Siliconꢀdioxideꢀ
Sodiumꢀhydroxideꢀtreatmentꢀ
Acidityꢀ
(cis‐2‐butene,ꢀ1‐butene,ꢀetc.).ꢀMoreover,ꢀtreatmentꢀwithꢀNaOHꢀledꢀtoꢀaꢀshorterꢀinductionꢀperiodꢀandꢀ
Pyridine
reducedꢀcokeꢀformationꢀonꢀtheꢀWO /SiO ꢀcatalyst.ꢀ
3
2
©
ꢀ2014,ꢀDalianꢀInstituteꢀofꢀChemicalꢀPhysics,ꢀChineseꢀAcademyꢀofꢀSciences.
PublishedꢀbyꢀElsevierꢀB.V.ꢀAllꢀrightsꢀreserved.
1.ꢀ ꢀ Introductionꢀ
olefinsꢀ conversionꢀ technologyꢀ (OCT)ꢀ byꢀ ABBꢀ Lummusꢀ Global,ꢀ
Houstonꢀ(USA)ꢀforꢀtheꢀproductionꢀofꢀpropeneꢀ[11].ꢀSinceꢀthen,ꢀ
Dueꢀtoꢀtheꢀrisingꢀdemandꢀforꢀpropyleneꢀinꢀtheꢀpetrochemi‐
calꢀ industry,ꢀ theꢀ metathesisꢀ ofꢀ 2‐buteneꢀ andꢀ ethyleneꢀ isꢀ nowꢀ
usedꢀforꢀon‐purposeꢀpropyleneꢀproductionꢀ[1,2].ꢀSeveralꢀcata‐
manyꢀstudiesꢀhaveꢀbeenꢀcarriedꢀoutꢀtoꢀdevelopꢀWO
lystsꢀwithꢀimprovedꢀperformance.ꢀTheꢀmechanismꢀofꢀtheꢀme‐
tathesisꢀreactionꢀofꢀ2‐buteneꢀandꢀethyleneꢀoverꢀtheꢀWO /SiO
catalystꢀ involvesꢀ theꢀ formationꢀ ofꢀ W‐carbeneꢀ speciesꢀ onꢀ theꢀ
WO ꢀ surface,ꢀ followedꢀ byꢀ metathesisꢀ onꢀ theseꢀ activeꢀ sitesꢀ toꢀ
yieldꢀ propyleneꢀ [12].ꢀ Manyꢀ factorsꢀ affectꢀ catalyticꢀ activityꢀ in‐
cludingꢀtheꢀmethodꢀofꢀcatalystꢀpreparation,ꢀreactionꢀconditions,ꢀ
structureꢀ ofꢀ theꢀ support,ꢀ andꢀ natureꢀ ofꢀ theꢀ surfaceꢀ tungstenꢀ
3
/SiO
2
ꢀcata‐
3
2
ꢀ
lystsꢀ includingꢀ supportedꢀ MoO3ꢀ [3,4],ꢀ Re
8–10]ꢀ areꢀ knownꢀ toꢀ catalyzeꢀ theꢀ reaction.ꢀ Industrially,ꢀ theꢀ
WO /SiO ꢀsystemꢀisꢀtheꢀmostꢀcommon.ꢀTheꢀmetathesisꢀofꢀlightꢀ
2 7
O
ꢀ [5–7],ꢀ andꢀ WO ꢀ
3
[
3
3
2
olefinsꢀhasꢀbeenꢀdevelopedꢀsinceꢀ1966ꢀinꢀaꢀprocessꢀknownꢀasꢀ
theꢀPhillipsꢀtriolefinꢀprocess,ꢀwhichꢀwasꢀadaptedꢀasꢀaꢀlicensedꢀ
*ꢀCorrespondingꢀauthor.ꢀTel:ꢀ+662‐218‐6861;ꢀE‐mail:ꢀpiyasan.p@chula.ac.thꢀ
DOI:ꢀ10.1016/S1872‐2067(12)60760‐8ꢀ|ꢀhttp://www.sciencedirect.com/science/journal/18722067 |ꢀChin. J. Catal.,ꢀVol.ꢀ35,ꢀNo.ꢀ2,ꢀFebruary 2014ꢀ ꢀ

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
ꢀ
species.ꢀ Experimentalꢀ studiesꢀ haveꢀ reportedꢀ thatꢀ tetrahedralꢀ
tungstenꢀoxideꢀspeciesꢀareꢀtheꢀactiveꢀsitesꢀ[11,13].ꢀ
withꢀ aꢀ solutionꢀ ofꢀ NaOH,ꢀ whichꢀ couldꢀ titrateꢀ halfꢀ ofꢀ theꢀ acidꢀ
sitesꢀpresentꢀonꢀtheꢀcatalyst.ꢀ.ꢀ
AsꢀcomparedꢀtoꢀMoO
3
‐ꢀandꢀRe
2
O ‐basedꢀcatalysts,ꢀsupportedꢀ
7
tungstenꢀoxideꢀcatalystsꢀareꢀactiveꢀatꢀhigherꢀreactionꢀtempera‐
turesꢀ(300–500ꢀ°C)ꢀ[1,14].ꢀAtꢀtheseꢀtemperaturesꢀhowever,ꢀsideꢀ
reactionsꢀsuchꢀasꢀisomerization,ꢀoligomerization,ꢀandꢀcrackingꢀ
ofꢀalkenesꢀalsoꢀoccur,ꢀsoꢀtheꢀselectivityꢀtoꢀprimaryꢀmetathesisꢀ
productsꢀisꢀusuallyꢀdecreasedꢀ[2].ꢀHuangꢀetꢀal.ꢀ[15–17] foundꢀ
thatꢀ Brönstedꢀ acidityꢀ promotedꢀ theꢀ formationꢀ ofꢀ activeꢀ me‐
tathesisꢀsites.ꢀAndreiniꢀetꢀal.ꢀ[18]ꢀsuggestedꢀthatꢀtheꢀformationꢀ
ofꢀ theꢀ initialꢀ metalꢀ carbeneꢀ neededꢀ acidityꢀ onꢀ theꢀ catalyst.ꢀ
However,ꢀvanꢀRoosmalenꢀandꢀMolꢀ[13]ꢀreportedꢀthatꢀisomeri‐
2.2.ꢀ ꢀ Characterizationꢀ
Theꢀspecificꢀsurfaceꢀarea,ꢀporeꢀvolume,ꢀandꢀporeꢀdiameterꢀofꢀ
theꢀ catalystsꢀ wereꢀ determinedꢀ byꢀ N2ꢀ adsorptionꢀ onꢀ aꢀ Tristarꢀ
(Micromeritics)ꢀ instrument.ꢀ Measurementsꢀ wereꢀ performedꢀ
withꢀnitrogenꢀasꢀtheꢀadsorbateꢀatꢀ–196ꢀ°Cꢀafterꢀpretreatmentꢀofꢀ
theꢀsampleꢀatꢀ200ꢀ°Cꢀunderꢀnitrogenꢀflowꢀforꢀ12ꢀh.ꢀTheꢀactualꢀ
metalꢀloadingꢀ wasꢀdeterminedꢀ byꢀinductively‐coupledꢀ plasmaꢀ
opticalꢀemissionꢀspectroscopyꢀ(ICP‐OES)ꢀusingꢀaꢀPerkinꢀElmerꢀ
Optimaꢀ 2100ꢀ DVꢀ instrument.ꢀ Beforeꢀ theꢀ measurement,ꢀ theꢀ
sampleꢀ wasꢀ dissolvedꢀ withꢀ hydrofluoricꢀ acidꢀ atꢀ 60ꢀ °C.ꢀ X‐rayꢀ
diffractionꢀ(XRD)ꢀwasꢀperformedꢀtoꢀdetermineꢀtheꢀbulkꢀphaseꢀ
ofꢀtheꢀcatalystsꢀusingꢀaꢀSIEMENSꢀDꢀ5000ꢀinstrumentꢀandꢀCuꢀKꢀ
radiationꢀwithꢀaꢀNiꢀfilterꢀinꢀtheꢀ2ꢀrangeꢀofꢀ10°ꢀtoꢀ80°ꢀwithꢀaꢀ
resolutionꢀ ofꢀ 0.04°.ꢀ Theꢀ crystalliteꢀ sizeꢀ wasꢀ calculatedꢀ byꢀ
Scherrer’sꢀequation.ꢀ
zationꢀoccursꢀonꢀtheꢀBrönstedꢀacidꢀsitesꢀofꢀtheꢀWO
3
/SiO ꢀcata‐
2
lysts.ꢀThisꢀledꢀtoꢀseveralꢀrecentꢀstudiesꢀthatꢀfocusedꢀonꢀpoison‐
ingꢀtheꢀacidꢀsitesꢀwithꢀsmallꢀamountsꢀofꢀanꢀalkaliꢀmetalꢀsuchꢀasꢀ
Na,ꢀ K,ꢀ Rb,ꢀ andꢀ Csꢀ [19,20].ꢀ Spamerꢀ etꢀ al.ꢀ [20]ꢀ foundꢀ thatꢀ
branchedꢀmetathesisꢀproductsꢀwereꢀdecreasedꢀbyꢀtheꢀadditionꢀ
ofꢀsmallꢀamountsꢀofꢀanꢀalkaliꢀmetalꢀion,ꢀthatꢀis,ꢀisomerizationꢀ
wasꢀ reduced.ꢀ However,ꢀ excessiveꢀ amountsꢀ ofꢀ alkaliꢀ ledꢀ toꢀ aꢀ
sharpꢀdecreaseꢀinꢀtheꢀmetathesisꢀrate.ꢀBecauseꢀtheꢀWO
3
ꢀcatalystꢀ
Ramanꢀspectraꢀofꢀtheꢀsamplesꢀwereꢀcollectedꢀusingꢀaꢀcon‐
tinuousꢀwaveꢀYAGꢀlaserꢀofꢀNdꢀ(810ꢀnm)ꢀofꢀ100ꢀmWꢀpowerꢀwithꢀ
theꢀ sampleꢀ atꢀ roomꢀtemperature.ꢀAꢀscanningꢀrangeꢀofꢀ200ꢀtoꢀ
isꢀ activeꢀ atꢀ 300–500ꢀ °C,ꢀ theꢀ formationꢀ ofꢀ cokeꢀ isꢀ expectedꢀ toꢀ
causeꢀdeactivation.ꢀSeveralꢀstudiesꢀlookedꢀatꢀcokeꢀformationꢀonꢀ
theꢀ catalystsꢀ duringꢀ olefinꢀ metathesisꢀ [21,22].ꢀ Theyꢀ showedꢀ
thatꢀtheꢀmechanismꢀofꢀcokeꢀformationꢀisꢀpromotedꢀbyꢀacidityꢀ
onꢀtheꢀcatalystꢀ[21–24].ꢀInꢀthisꢀcaseꢀalso,ꢀalkalineꢀdopingꢀcanꢀ
haveꢀaꢀbeneficialꢀimpactꢀ[22,25].ꢀ ꢀ
–1
–1
2000ꢀcm ꢀwithꢀaꢀresolutionꢀofꢀ16ꢀcm ꢀwasꢀused.ꢀTheꢀnumberꢀ
ofꢀ totalꢀ acidꢀ sitesꢀ wasꢀ determinedꢀ byꢀ aꢀ methodꢀ involvingꢀ anꢀ
+
+
aqueousꢀ ionꢀ exchangeꢀ ofꢀ theꢀ catalystꢀ H ꢀ ionsꢀ withꢀ Na ꢀ ions,ꢀ
followedꢀbyꢀtheꢀtitrationꢀofꢀtheꢀresultingꢀsolutionꢀ[26].ꢀCatalystꢀ
(0.2ꢀg)ꢀwasꢀaddedꢀtoꢀ10ꢀmLꢀofꢀaqueousꢀsolutionꢀofꢀNaClꢀ(3.42ꢀ
mol/L)ꢀunderꢀstirring.ꢀAfterꢀ30ꢀhꢀofꢀionꢀexchangeꢀatꢀroomꢀtem‐
perature,ꢀtheꢀsolidꢀparticleꢀwasꢀfilteredꢀoff,ꢀandꢀtheꢀfiltrateꢀwasꢀ
titratedꢀwithꢀaqueousꢀNaOHꢀsolutionꢀ(0.05ꢀmol/L)ꢀtoꢀdetermineꢀ
Inꢀtheꢀpresentꢀstudy,ꢀweꢀtestedꢀWO
3
/SiO ꢀcatalystsꢀthatꢀhaveꢀ
2
differentꢀloadingꢀunderꢀconditionsꢀcloseꢀtoꢀtheꢀindustrialꢀonesꢀ
toꢀevaluateꢀtheꢀimpactꢀofꢀNaꢀpoisoning.ꢀInfraredꢀspectroscopyꢀ
ofꢀ pyridineꢀ adsorptionꢀ wasꢀ usedꢀ toꢀ evaluateꢀ theꢀ impactꢀ ofꢀ
BrönstedꢀandꢀLewisꢀsitesꢀonꢀmetathesisꢀactivity.ꢀWeꢀdiscussꢀtheꢀ
effectꢀofꢀNaꢀpoisoningꢀonꢀtheꢀundesiredꢀsideꢀreactions,ꢀcatalystꢀ
stability,ꢀandꢀcokeꢀformation.ꢀ ꢀ
+
theꢀ numberꢀ ofꢀ exchangeableꢀ acidꢀ sitesꢀ inꢀ molꢀ H /g‐catalyst.ꢀ
Fourierꢀ transformꢀ infraredꢀ (FT‐IR)ꢀ spectroscopyꢀ ofꢀ pyridineꢀ
adsorptionꢀ wasꢀ usedꢀ toꢀ characterizeꢀ theꢀ Lewisꢀ andꢀ Brönstedꢀ
acidꢀsites.ꢀEachꢀsampleꢀwasꢀpressedꢀintoꢀaꢀthinꢀwaferꢀ(~30ꢀmg).ꢀ
Theꢀwaferꢀwasꢀoutgassedꢀatꢀ300ꢀ°Cꢀforꢀ2ꢀhꢀunderꢀvacuumꢀandꢀ
thenꢀ cooledꢀ toꢀ roomꢀ temperature,ꢀ andꢀ theꢀ IRꢀ spectrumꢀ wasꢀ
recordedꢀasꢀtheꢀbackgroundꢀspectrum.ꢀSubsequently,ꢀpyridineꢀ
vaporꢀwasꢀadmittedꢀintoꢀtheꢀsampleꢀchamberꢀatꢀ50ꢀ°Cꢀforꢀ15ꢀ
min.ꢀ Next,ꢀ pyridineꢀ wasꢀ desorbedꢀ underꢀ vacuumꢀ forꢀ 0.5ꢀ hꢀ toꢀ
removeꢀphysisorbedꢀpyridine,ꢀandꢀanꢀIRꢀspectrumꢀwasꢀrecord‐
ed.ꢀ Allꢀ IRꢀ spectraꢀ ofꢀ theꢀ adsorbedꢀ speciesꢀ wereꢀ obtainedꢀ byꢀ
subtractingꢀ theꢀ backgroundꢀ spectrum.ꢀ Theꢀ cokeꢀ contentꢀ wasꢀ
determinedꢀbyꢀthermogravimetricꢀanalysisꢀ(TGA)ꢀusingꢀanꢀSDQꢀ
Tꢀ600ꢀequipment.ꢀTheꢀusedꢀcatalystsꢀwereꢀheatedꢀinꢀaꢀstreamꢀofꢀ
2
.ꢀ ꢀ Experimentalꢀ ꢀ
.1.ꢀ ꢀ Catalystꢀpreparationꢀ ꢀ
WO /SiO ꢀ catalystsꢀ withꢀ differentꢀ tungstenꢀ loadingꢀ wereꢀ
2
3
2
preparedꢀbyꢀimpregnationꢀofꢀsilicaꢀgelꢀ(DavisilꢀGradeꢀ646,ꢀporeꢀ
3
volumeꢀ1.15ꢀcm /g,ꢀsuppliedꢀbyꢀAldrich)ꢀwithꢀanꢀaqueousꢀsolu‐
tionꢀ containingꢀ theꢀ desiredꢀ amountꢀ ofꢀ ammoniumꢀ meta‐
tungstateꢀ hydrateꢀ (Aldrich,ꢀ 99.9%).ꢀ Theꢀ impregnatedꢀ sampleꢀ
wasꢀdriedꢀinꢀambientꢀairꢀforꢀ2ꢀhꢀandꢀsubsequentlyꢀinꢀanꢀovenꢀatꢀ
1
10ꢀ°Cꢀforꢀ24ꢀh,ꢀfollowingꢀbyꢀcalcinationꢀatꢀ550ꢀ°Cꢀinꢀairꢀforꢀ8ꢀh.ꢀ
Theꢀ catalystsꢀ wereꢀ denotedꢀ asꢀ xWO /SiO ,ꢀ whereꢀ xꢀ wasꢀ theꢀ
WO ꢀweightꢀloadingꢀ(%)ꢀandꢀreferredꢀtoꢀasꢀpristineꢀcatalysts.ꢀ
10%O
2
‐90%Heꢀatꢀaꢀ10ꢀ°C/minꢀheatingꢀrateꢀtoꢀ700ꢀ°C.ꢀ
3
2
3
2.3.ꢀ ꢀ Catalyticꢀtestingꢀ
Eachꢀcalcinedꢀpristineꢀcatalystsꢀ(3.5ꢀg)ꢀwasꢀionꢀexchangedꢀwithꢀ
anꢀaqueousꢀNaOHꢀsolutionꢀ(300ꢀmL).ꢀTheꢀconcentrationꢀofꢀtheꢀ
solutionꢀwasꢀsuchꢀthatꢀtheꢀnumberꢀofꢀNaOHꢀmoleculesꢀintro‐
ducedꢀinꢀtheꢀionꢀexchangeꢀwasꢀequalꢀtoꢀ0.5,ꢀ1.0,ꢀorꢀ1.5ꢀtimesꢀtheꢀ
numberꢀofꢀacidꢀsitesꢀpresentꢀonꢀtheꢀcatalystꢀ(asꢀdeterminedꢀbyꢀ
titration,ꢀvideꢀinfra).ꢀTheꢀsuspensionꢀwasꢀstirredꢀatꢀroomꢀtem‐
peratureꢀforꢀ2ꢀh.ꢀAfterꢀthat,ꢀtheꢀsolidꢀwasꢀfilteredꢀoff,ꢀandꢀitꢀwasꢀ
driedꢀinꢀanꢀovenꢀatꢀ110ꢀ°Cꢀforꢀ24ꢀhꢀfollowedꢀbyꢀcalcination.ꢀTheꢀ
sampleꢀ obtainedꢀ byꢀ thisꢀ procedureꢀ wasꢀ designated,ꢀ e.g.,ꢀ asꢀ
Forꢀactivityꢀtesting,ꢀ3.0ꢀgꢀofꢀcatalystꢀwasꢀpackedꢀinꢀtheꢀcenterꢀ
ofꢀaꢀstainlessꢀsteelꢀtubularꢀreactor.ꢀTheꢀcatalystꢀwasꢀpretreatedꢀ
atꢀ550ꢀ°Cꢀunderꢀnitrogenꢀflowꢀforꢀ1ꢀh,ꢀandꢀtheꢀreactionꢀcondi‐
tionsꢀ were:ꢀ temperatureꢀ 400ꢀ °C,ꢀ pressureꢀ 0.1ꢀ MPa,ꢀ
C H
2
4
:trans‐2‐C
4
H
8
ꢀ molarꢀ ratioꢀ 2:1ꢀ (4%ꢀ ofꢀ C
2
H ꢀ andꢀ 2%ꢀ ofꢀ
4
trans‐2‐C
4
H
8
)ꢀ balancedꢀ withꢀ N ,ꢀ andꢀ theꢀ totalꢀ flowꢀ rateꢀ 90ꢀ
2
3
cm /min.ꢀ Theꢀ compositionꢀ ofꢀ theꢀ productꢀ andꢀ feedꢀ streamsꢀ
wereꢀ analyzedꢀ byꢀ aꢀ Shimadzuꢀ GC‐2014ꢀ gasꢀ chromatographꢀ
equippedꢀwithꢀanꢀFIDꢀdetector.ꢀ
4
W+0.5NaOH,ꢀwhichꢀdenotesꢀthatꢀ4%WO
3
/SiO ꢀwasꢀexchangedꢀ
2

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
3
.ꢀ ꢀ Resultsꢀandꢀdiscussionꢀ ꢀ
3
WO crystal
3.1.ꢀ ꢀ Catalystꢀcharacterizationꢀ ꢀ
Theꢀtotalꢀacidityꢀofꢀtheꢀpristineꢀcatalystsꢀwasꢀdeterminedꢀbyꢀ
(
4)
titrationꢀ (Tableꢀ 1).ꢀ Theseꢀ valuesꢀ allowedꢀ usꢀ toꢀ performꢀ theꢀ
NaOHꢀ ionꢀ exchangeꢀ inꢀ aꢀ controlledꢀ wayꢀ byꢀ addingꢀ aꢀ specificꢀ
amountꢀofꢀNaOH.ꢀTheꢀcompositionsꢀofꢀtheꢀpristineꢀandꢀmodi‐
(
3)
fiedꢀcatalystsꢀwereꢀinvestigatedꢀbyꢀICP‐OESꢀ(Tableꢀ1).ꢀTheꢀWO ꢀ
3
(
2)
contentꢀ wasꢀ lowerꢀ forꢀ theꢀ NaOH‐treatedꢀ catalysts,ꢀ indicatingꢀ
thatꢀtheꢀtreatmentꢀinducedꢀWꢀleaching.ꢀ ꢀ However,ꢀtheꢀleachingꢀ
effectꢀ wasꢀ pronouncedꢀ onlyꢀ forꢀ theꢀ catalystꢀ withꢀ theꢀ lowestꢀ
(
1)
WO
loading
3
ꢀloadingꢀ(4%WO
3
/SiO
2
),ꢀforꢀwhichꢀaꢀsharpꢀdecreaseꢀinꢀWO ꢀ
3
2
0
30
40
50
60
70
80
o
ꢀ
(byꢀ75%)ꢀwasꢀobservedꢀafterꢀtheꢀionꢀexchangeꢀwithꢀtheꢀ
2/( )
Fig.ꢀ1.ꢀXRDꢀpatternsꢀofꢀpristineꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀload‐
ing.ꢀ (1)ꢀ 4%WO /SiO ;ꢀ (2)ꢀ 6%WO /SiO ;ꢀ (3)ꢀ 8%WO /SiO ;ꢀ (4)ꢀ
0%WO /SiO .ꢀ
solutionꢀwithꢀtheꢀhighestꢀNaOHꢀconcentration.ꢀForꢀallꢀtheꢀotherꢀ
catalystsꢀwithꢀhigherꢀtungstenꢀloading,ꢀevenꢀwithꢀtheꢀtreatmentꢀ
withꢀ theꢀ solutionꢀ withꢀ theꢀ highestꢀ NaOHꢀ concentration,ꢀ theꢀ
3
2
3
2
3
2
1
3
2
amountꢀofꢀWO ꢀleachedꢀwasꢀlessꢀthanꢀ~15%.ꢀ
3
Theꢀtexturalꢀstructureꢀofꢀtheꢀcatalystsꢀwasꢀcharacterizedꢀbyꢀ
ꢀphysisorptionꢀ(Tableꢀ1).ꢀTheꢀspecificꢀsurfaceꢀareaꢀandꢀporeꢀ
thatꢀtheꢀNaOHꢀionꢀexchangeꢀledꢀtoꢀaꢀsignificantꢀlossꢀofꢀWO3ꢀspe‐
ciesꢀonlyꢀwithꢀtheꢀ4%WO /SiO ꢀcatalyst.ꢀ
N
2
3
2
volumeꢀ decreasedꢀ slightlyꢀ withꢀ increasingꢀ tungstenꢀ loadingꢀ
whileꢀ theꢀ averageꢀ poreꢀ sizeꢀ remainedꢀ unchanged.ꢀ Thusꢀ theꢀ
impregnationꢀ didꢀ notꢀ affectꢀ significantlyꢀ theꢀ textureꢀ ofꢀ theꢀ
Ramanꢀspectroscopyꢀwasꢀusedꢀtoꢀdetermineꢀtheꢀstructureꢀofꢀ
theꢀtungstenꢀspeciesꢀpresentꢀonꢀtheꢀsupportedꢀcatalystsꢀ(Fig.ꢀ3).ꢀ
Theꢀ Ramanꢀ bandsꢀ atꢀ 498,ꢀ 602,ꢀ 808,ꢀ andꢀ 1060ꢀ cm–1ꢀ wereꢀ as‐
support,ꢀandꢀitꢀcanꢀbeꢀconcludedꢀthatꢀWO
3
ꢀwasꢀmainlyꢀdepos‐
signedꢀ toꢀ theꢀ characteristicꢀ vibrationsꢀ ofꢀ four‐ꢀ andꢀ
three‐memberꢀsilicaꢀringsꢀandꢀSi–O–Siꢀstretching,ꢀrespectivelyꢀ
[30].ꢀ Asꢀ expectedꢀ fromꢀ theꢀ XRDꢀ results,ꢀ theꢀ Ramanꢀ bandsꢀ ofꢀ
itedꢀonꢀtheꢀouterꢀsurfaceꢀofꢀtheꢀsupportꢀ(andꢀnotꢀinꢀtheꢀpores).ꢀ
Theꢀtexturalꢀpropertiesꢀwereꢀnotꢀalteredꢀbyꢀtheꢀtreatmentꢀwithꢀ
NaOH.ꢀ ꢀ
crystallineꢀWO ꢀwereꢀobservedꢀforꢀallꢀtheꢀpristineꢀcatalysts.ꢀTheꢀ
3
TheꢀXRDꢀofꢀtheꢀpristineꢀcatalystsꢀshowedꢀthatꢀthereꢀwasꢀon‐
lyꢀpartialꢀdispersionꢀofꢀtungstenꢀoxideꢀbecauseꢀtheꢀcharacteris‐
bandsꢀatꢀ263–275,ꢀ707–720,ꢀandꢀ807–808ꢀcm–1ꢀwereꢀassignedꢀ
toꢀ theꢀ deformationꢀ modeꢀ ofꢀ W–O–W,ꢀ bendingꢀ modeꢀ ofꢀ W–O,ꢀ
ticꢀpeaksꢀofꢀWO
4.38°ꢀ(Fig.ꢀ1).ꢀAsꢀexpectedꢀ[27–29],ꢀtheꢀintensityꢀofꢀtheꢀWO
peaksꢀ increasedꢀ withꢀ WO ꢀ loading.ꢀ Figureꢀ 2ꢀ showsꢀ theꢀ XRDꢀ
patternsꢀofꢀtheꢀNaOH‐treatedꢀcatalystsꢀcomparedꢀwithꢀtheꢀpris‐
tineꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀloading.ꢀTheꢀintensityꢀofꢀ
theꢀ XRDꢀ peaksꢀ decreasedꢀ afterꢀ theꢀ NaOHꢀ treatment.ꢀ Aꢀ sharpꢀ
3
ꢀcrystalsꢀwereꢀdetectedꢀatꢀ23.12°,ꢀ23.60°,ꢀandꢀ
andꢀsymmetricꢀstretchingꢀmodeꢀofꢀW–O,ꢀrespectivelyꢀ[28].ꢀTheꢀ
broadꢀbandꢀatꢀ970ꢀcm–1ꢀwasꢀassignedꢀtoꢀtheꢀO=W=Oꢀbandꢀofꢀ
isolatedꢀsurfaceꢀtetrahedralꢀtungstenꢀoxideꢀspeciesꢀ[31],ꢀwhichꢀ
wasꢀreportedꢀasꢀtheꢀactiveꢀspeciesꢀforꢀmetathesisꢀ[17].ꢀAsꢀre‐
portedꢀbyꢀAndreiniꢀetꢀal.ꢀ[18]ꢀandꢀDebeckerꢀetꢀal.ꢀ[10],ꢀtheꢀac‐
tiveꢀ tungstenꢀ compoundꢀ wasꢀ difficultꢀ toꢀ reduceꢀ whereasꢀ theꢀ
2
3
ꢀ
3
decreaseꢀinꢀtheꢀintensityꢀofꢀtheꢀWO
3
ꢀcrystalꢀpeaksꢀwasꢀobservedꢀ
crystallineꢀWO
hedralꢀtungstenꢀoxideꢀspeciesꢀhasꢀaꢀstrongerꢀinteractionꢀwithꢀ
theꢀsupportꢀthanꢀdoesꢀtheꢀcrystallineꢀWO .ꢀInꢀtheꢀpresentꢀwork,ꢀ
theꢀcatalystsꢀexhibitedꢀdifferentꢀintensityꢀforꢀtheꢀbandsꢀatꢀ710,ꢀ
3
ꢀcanꢀbeꢀreducedꢀeasily,ꢀimplyingꢀthatꢀtheꢀtetra‐
forꢀmodifiedꢀ4%WO /SiO .ꢀInꢀcontrast,ꢀonlyꢀsmallꢀchangesꢀwereꢀ
3
2
observedꢀ forꢀ theꢀ catalystsꢀ withꢀ higherꢀ tungstenꢀ loading.ꢀ Theꢀ
XRDꢀresultsꢀwereꢀinꢀgoodꢀagreementꢀwithꢀtheꢀICP‐OESꢀresultsꢀ
3
Tableꢀ1ꢀ
MeasuredꢀcompositionꢀofꢀWO
3
,ꢀtexturalꢀcharacterization,ꢀandꢀtotalꢀacidityꢀ(ionꢀexchangeꢀtitration)ꢀofꢀtheꢀpristineꢀcatalystsꢀandꢀmodifiedꢀcatalysts.ꢀ
ActualꢀWO
3
ꢀcompositionꢀ ꢀ
S
BETꢀ ꢀ
Poreꢀvolumeꢀ
(cm /g)ꢀ
Averageꢀporeꢀsizeꢀ
(nm)ꢀ
Totalꢀacidityꢀ
(µmolꢀH /g‐catalyst)ꢀ
Catalystꢀ
2
3
+
(
wt%ꢀ)ꢀ
3.68ꢀ
1.17ꢀ
1.10ꢀ
0.94ꢀ
5.05ꢀ
4.60ꢀ
4.03ꢀ
7.02ꢀ
6.58ꢀ
6.08ꢀ
9.14ꢀ
8.82ꢀ
8.68ꢀ
8.25ꢀ
(m /g)ꢀ
281ꢀ
283ꢀ
285ꢀ
288ꢀ
278ꢀ
279ꢀ
280ꢀ
277ꢀ
280ꢀ
279ꢀ
269ꢀ
269ꢀ
271ꢀ
273ꢀ
4
4
4
4
6
6
6
8
8
8
1
1
1
1
%WO
3
/SiO
2
ꢀ
0.98ꢀ
0.99ꢀ
0.99ꢀ
1.01ꢀ
0.95ꢀ
0.95ꢀ
0.97ꢀ
0.92ꢀ
0.94ꢀ
0.95ꢀ
0.87ꢀ
0.88ꢀ
0.88ꢀ
0.90ꢀ
9.32ꢀ
9.38ꢀ
9.37ꢀ
9.38ꢀ
9.29ꢀ
9.29ꢀ
9.32ꢀ
9.28ꢀ
9.29ꢀ
9.31ꢀ
9.27ꢀ
9.25ꢀ
9.27ꢀ
9.26ꢀ
ꢀ 74ꢀ
ꢀ 61ꢀ
ꢀ 51ꢀ
ꢀ 40ꢀ
ꢀ 83ꢀ
ꢀ 58ꢀ
ꢀ 35ꢀ
ꢀ 95ꢀ
ꢀ 67ꢀ
ꢀ 42ꢀ
101ꢀ
ꢀ 82ꢀ
ꢀ 65ꢀ
ꢀ 46ꢀ
W+0.5NaOHꢀ
W+1.0NaOHꢀ
W+1.5NaOHꢀ
%WO
3
/SiO ꢀ
2
W+1.0NaOHꢀ
W+1.5NaOHꢀ
%WO
W+1.0NaOHꢀ
W+1.5NaOHꢀ
0%WO
3
/SiO ꢀ
2
3
/SiO ꢀ
2
0W+0.5NaOHꢀ
0W+1.0NaOHꢀ
0W+1.5NaOHꢀ

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
ꢀ
4
% WO
3
6% WO
3
(4)
(4)
(3)
(3)
(2)
(1)
(1)
8% WO
3
1
0% WO
3
(4)
(4)
(3)
(2)
(3)
(1)
(1)
20
30
40
50
60
70
80
20
30
40
50
60
70
80
o
o
2
/( )
2/( )
Fig.ꢀ2.ꢀXRDꢀpatternsꢀofꢀpristineꢀcatalystsꢀcomparedꢀwithꢀmodifiedꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀloading.ꢀ(1)ꢀPristineꢀcatalyst;ꢀ(2)ꢀModifiedꢀwithꢀ
0
.5NaOH;ꢀ(3)ꢀModifiedꢀwithꢀ1.0NaOH;ꢀ(4)ꢀModifiedꢀwithꢀ1.5NaOH.ꢀ
–1
–1
8
07,ꢀandꢀ970ꢀcm .ꢀWhenꢀtheꢀtungstenꢀloadingꢀwasꢀincreased,ꢀ
isolatedꢀ speciesꢀ (970ꢀ cm ).ꢀ Thisꢀ effectꢀ wasꢀ largerꢀ whenꢀ theꢀ
sharpꢀincreasesꢀinꢀtheꢀintensityꢀwereꢀobservedꢀatꢀtheꢀ710ꢀandꢀ
concentrationꢀ ofꢀ NaOHꢀ wasꢀ increased.ꢀ Forꢀ theꢀ catalystꢀ withꢀ
higherꢀloading,ꢀtheꢀeffectꢀofꢀNaOHꢀtreatmentꢀwasꢀnotꢀveryꢀpro‐
–1
8
07ꢀ cm ꢀ bandsꢀ (crystallineꢀ species),ꢀ whileꢀ thereꢀ wasꢀ onlyꢀ aꢀ
–1
slightꢀincreaseꢀinꢀtheꢀbandꢀatꢀ970ꢀcm ꢀ(isolatedꢀspecies)ꢀwithꢀ
theꢀmaximumꢀintensityꢀforꢀ8%ꢀWO ꢀloading.ꢀ ꢀ
TheꢀsameꢀWO ꢀspeciesꢀwasꢀobservedꢀonꢀallꢀtheꢀNaOH‐
edꢀcatalystsꢀ(Fig.ꢀ4).ꢀTheꢀspectrumꢀofꢀtheꢀ4%WO /SiO ꢀcatalystꢀ
wasꢀ significantlyꢀ modified.ꢀ Theꢀ intensityꢀ ofꢀ theꢀ bandsꢀ corre‐
nounced.ꢀOnlyꢀtheꢀbandsꢀcorrespondingꢀtoꢀtheꢀcrystallineꢀWO ꢀ
3
3
speciesꢀ wereꢀ progressivelyꢀ smaller,ꢀ whileꢀ theꢀ bandꢀ corre‐
spondingꢀtoꢀtheꢀisolatedꢀspeciesꢀwasꢀhardlyꢀaffected.ꢀ ꢀ
x
ꢀ
treat‐
3
2
ThusꢀtheꢀICP‐OESꢀdata,ꢀXRDꢀdata,ꢀandꢀRamanꢀspectroscopyꢀ
consistentlyꢀ indicatedꢀ that:ꢀ (i)ꢀ 4%WO
3
/SiO ꢀ wasꢀ stronglyꢀ af‐
2
–1
spondingꢀ toꢀ theꢀ crystallineꢀ speciesꢀ (710ꢀ andꢀ 807ꢀ cm )ꢀ wasꢀ
decreasedꢀ drastically,ꢀ asꢀ wasꢀ theꢀ bandꢀ correspondingꢀ toꢀ theꢀ
fectedꢀbyꢀtheꢀNaOHꢀtreatment,ꢀ(ii)ꢀcatalystsꢀwithꢀhigherꢀloadingꢀ
wereꢀ onlyꢀ slightlyꢀ affected,ꢀ andꢀ (iii)ꢀ theꢀ mainꢀ effectꢀ wasꢀ thatꢀ
partꢀ ofꢀ theꢀ crystallineꢀ speciesꢀ presentꢀ onꢀ theꢀ catalystsꢀ wasꢀ
leachedꢀoff.ꢀ
807
970
Theꢀ NH ‐TPDꢀ experimentsꢀ demonstratedꢀ theꢀ effectꢀ ofꢀ theꢀ
3
NaOHꢀtreatmentꢀonꢀtheꢀacidityꢀofꢀtheꢀcatalystsꢀ(Fig.ꢀ5).ꢀBecauseꢀ
theꢀsilicaꢀsupportꢀisꢀnotꢀacidicꢀ[18],ꢀtheꢀacidꢀsitesꢀwereꢀdueꢀtoꢀ
theꢀtungstenꢀoxide.ꢀTheꢀlowꢀtemperatureꢀpeakꢀofꢀtheꢀpristineꢀ
catalystsꢀatꢀ110–130ꢀ°Cꢀcorrespondedꢀtoꢀweakꢀacidꢀsitesꢀ[28].ꢀ
Theꢀacidityꢀofꢀtheꢀpristineꢀcatalystsꢀincreasedꢀwithꢀincreasingꢀ
tungstenꢀloading,ꢀwhichꢀisꢀinꢀgoodꢀagreementꢀwithꢀtheꢀlitera‐
(
4)
3)
2)
1)
900 1000 1100
(
tureꢀ[29].ꢀBothꢀtitrationꢀ(Tableꢀ1)ꢀandꢀNH ‐TPDꢀdataꢀagreedꢀinꢀ
3
showingꢀ thatꢀ theꢀ acidityꢀ wasꢀ decreasedꢀ afterꢀ theꢀ treatmentꢀ
withꢀNaOH.ꢀItꢀisꢀinterestingꢀtoꢀnoteꢀthatꢀtheꢀlowꢀtemperatureꢀ
peakꢀofꢀtheꢀmodifiedꢀcatalystsꢀwasꢀshiftedꢀtoꢀaꢀhigherꢀtempera‐
ture.ꢀ Thus,ꢀ whileꢀ poisoningꢀ appearsꢀ toꢀ affectꢀ bothꢀ typesꢀ ofꢀ
acidicꢀ sites,ꢀ itꢀ mustꢀ beꢀ noticedꢀthatꢀ partꢀofꢀtheꢀ strongerꢀ sitesꢀ
remainedꢀonꢀtheꢀcatalyst.ꢀ ꢀ
(
(
4
00
500
600
700
Raman shift (cm )
Fig.ꢀ3.ꢀRamanꢀspectraꢀofꢀtheꢀpristineꢀcatalystsꢀwithꢀdifferentꢀtungsten
loading.ꢀ (1)ꢀ 4%WO /SiO ;ꢀ (2)ꢀ 6%WO /SiO ;ꢀ (3)ꢀ 8%WO /SiO ;ꢀ (4)ꢀ
0%WO /SiO .ꢀ
800

1
3
2
3
2
3
2
InꢀorderꢀtoꢀgainꢀfurtherꢀinsightꢀintoꢀtheꢀrolesꢀofꢀtheꢀLewisꢀ
1
3
2

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
4
% WO
3
6% WO
3
(4)
(3)
(4)
(2)
(3)
(1)
(1)
10% WO
3
8% WO
3
(4)
(4)
(3)
(3)
(2)
(1)
(1)
400
500
600
700
800
900 1000 1100
400
500
600
700
800
900 1000 1100
1
1
Raman shift (cm )
Raman shift (cm )
Fig.ꢀ4.ꢀRamanꢀspectraꢀofꢀtheꢀpristineꢀcatalystsꢀcomparedꢀtoꢀtheꢀmodifiedꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀloading.ꢀ(1)ꢀPristineꢀcatalyst;ꢀ(2)ꢀModifiedꢀ
withꢀ0.5NaOH;ꢀ(3)ꢀModifiedꢀwithꢀ1.0NaOH;ꢀ(4)ꢀModifiedꢀwithꢀ1.5NaOH.ꢀ
andꢀBrönstedꢀacidꢀsitesꢀonꢀtheseꢀmetathesisꢀcatalysts,ꢀaꢀstudyꢀ
wasꢀ performedꢀ byꢀ infraredꢀ spectroscopyꢀ ofꢀ pyridineꢀ adsorp‐
sitesꢀ increasedꢀ markedlyꢀ withꢀ increasingꢀ WO3ꢀloading,ꢀ whichꢀ
wasꢀ inꢀ agreementꢀ withꢀ theꢀ literatureꢀ [29,34].ꢀ Theꢀ Brönstedꢀ
acidityꢀpeaks,ꢀhowever,ꢀwereꢀveryꢀweak.ꢀTinyꢀpeaksꢀwereꢀseenꢀ
−1
tionꢀ (Fig.ꢀ 6).ꢀ Theꢀ peakꢀ atꢀ 1596ꢀ cm ꢀ wasꢀ dueꢀ toꢀ hydrogenꢀ
−1
−1
bondedꢀpyridine.ꢀTheꢀpeaksꢀatꢀ1623,ꢀ1575,ꢀandꢀ1445ꢀcm ꢀcor‐
respondedꢀtoꢀLewisꢀacidꢀsites.ꢀBrönstedꢀacidꢀsitesꢀwereꢀshownꢀ
atꢀ 1545ꢀ andꢀ 1639ꢀ cm ꢀ withꢀ theꢀ catalystsꢀ withꢀ 8%ꢀ andꢀ 10%ꢀ
WO ꢀloading.ꢀThusꢀBrönstedꢀsitesꢀwereꢀeitherꢀnotꢀabundantꢀorꢀ
3
−1
−1
byꢀtheꢀ peaksꢀ atꢀ1545ꢀ andꢀ 1639ꢀcm .ꢀTheꢀ peakꢀatꢀ1492ꢀcm ꢀ
wasꢀ attributedꢀ toꢀ pyridineꢀ adsorbedꢀ onꢀ bothꢀ Brönstedꢀ andꢀ
Lewisꢀ acidꢀ sitesꢀ [32,33].ꢀ Figureꢀ 6ꢀ showsꢀ thatꢀ theꢀ Lewisꢀ acidꢀ
relativelyꢀweakꢀonꢀtheseꢀcatalysts.ꢀThisꢀisꢀconsistentꢀwithꢀpre‐
viousꢀ reportsꢀ showingꢀ thatꢀ significantꢀ Brönstedꢀ acidityꢀ wasꢀ
onlyꢀgeneratedꢀonꢀcatalystsꢀwithꢀhighꢀWO
3
ꢀcontentsꢀ[34].ꢀ ꢀ
Figureꢀ 7ꢀ showsꢀ theꢀ FT‐IRꢀ pyridineꢀ spectraꢀ ofꢀ theꢀ NaOH‐
ꢀ
treatedꢀ catalystsꢀ asꢀ comparedꢀ toꢀ theꢀ pristineꢀ catalysts.ꢀ Theꢀ
signalsꢀ correspondingꢀ toꢀ theꢀ Brönstedꢀ sitesꢀ wereꢀ nowꢀ totallyꢀ
(8)
(7)
(6)
1445
1596
1639
(5)
1492
1623
1545
(4)
(4)
(3)
(3)
(
2)
(2)
(1)
(
1)
1
00
200
Temperature ( C)
‐TPDꢀ profilesꢀ ofꢀ theꢀ pristineꢀ catalystsꢀ comparedꢀ withꢀ the
modifiedꢀ catalystsꢀ withꢀ differentꢀ tungstenꢀ loading.ꢀ (1)ꢀ 4%WO /SiO ;ꢀ
2)ꢀ 6%WO /SiO ;ꢀ (3)ꢀ 8%WO /SiO ;ꢀ (4)ꢀ 10%WO /SiO ;ꢀ (5)ꢀ 4W+
.5NaOH;ꢀ(6)ꢀ6W+1.5NaOH;ꢀ(7)ꢀ8W+1.5NaOH;ꢀ(8)ꢀ10W+1.5NaOH.ꢀ
300
400
o
1
400
1450
1500
1550
1600
1650
1
Wavenumber (cm )
Fig.ꢀ6.ꢀFT‐IRꢀspectraꢀofꢀpyridineꢀadsorbedꢀatꢀ50ꢀ°Cꢀonꢀcatalystsꢀwithꢀ
differentꢀ tungstenꢀ loading.ꢀ (1)ꢀ 4%WO /SiO ;ꢀ (2)ꢀ 6%WO /SiO ;ꢀ (3)ꢀ
8%WO /SiO ;ꢀ(4)ꢀ10%WO /SiO .ꢀ
Fig.ꢀ 5.ꢀ NH
3
3
2
(
1
3
2
3
2
3
2
3
2
3
2
3
2
3
2

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
ꢀ
4
% WO
3
6% WO3
(4)
(4)
(3)
(3)
(2)
(1)
(1)
8
% WO
3
10% WO
3
(4)
(4)
(3)
(3)
(2)
(1)
(1)
1400
1450
1500
1550
1600
1650
1400
1450
1500
1550
1600
1650
1
1
Wavenumber (cm )
Wavenumber (cm )
Fig.ꢀ7.ꢀFT‐IRꢀspectraꢀofꢀpyridineꢀadsorbedꢀatꢀ50ꢀ°Cꢀonꢀpristineꢀcatalystsꢀcomparedꢀwithꢀmodifiedꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀloading.ꢀ(1)ꢀPris‐
tineꢀcatalyst;ꢀ(2)ꢀModifiedꢀwithꢀ0.5NaOH;ꢀ(3)ꢀModifiedꢀwithꢀ1.0NaOH;ꢀ(4)ꢀModifiedꢀwithꢀ1.5NaOH.ꢀ
absent,ꢀevenꢀforꢀtheꢀcatalystsꢀwithꢀtheꢀhighestꢀloading.ꢀForꢀtheꢀ
Lewisꢀsites,ꢀitꢀwasꢀclearꢀthatꢀtheꢀionꢀexchangeꢀwithꢀNaOHꢀhadꢀaꢀ
markedꢀeffect.ꢀInꢀparticular,ꢀtheꢀintensityꢀofꢀtheꢀpeakꢀatꢀ1455ꢀ
3.2.ꢀ ꢀ Metathesisꢀactivityꢀandꢀselectivityꢀ ꢀ
TheꢀmetathesisꢀactivityꢀofꢀdifferentꢀWO
3
/SiO
2
ꢀcatalystsꢀwasꢀ
−1
ꢀ
cm ꢀwasꢀprogressivelyꢀreducedꢀasꢀtheꢀNaOHꢀsolutionꢀusedꢀforꢀ
ionꢀexchangeꢀwasꢀmoreꢀconcentrated.ꢀTheꢀintensityꢀofꢀtheꢀpeakꢀ
evaluatedꢀatꢀ400 °Cꢀforꢀ10ꢀh.ꢀTheꢀresultsꢀareꢀshownꢀinꢀFig.ꢀ8.ꢀ
Metathesisꢀ activity,ꢀ expressedꢀ byꢀ theꢀ conversionꢀ ofꢀ trans‐2‐
butene,ꢀ increasedꢀ withꢀ increasingꢀ tungstenꢀ loadingꢀ andꢀ
reachedꢀ aꢀ maximumꢀ ofꢀ 80%ꢀ conversionꢀ withꢀ 8%WO /SiO .ꢀ
Furtherꢀincreasingꢀofꢀtheꢀtungstenꢀloadingꢀtoꢀ10%W0 /SiO ꢀdidꢀ
ꢀ
−1
atꢀ 1492ꢀ cm ꢀ (whichꢀ representedꢀ bothꢀ Brönstedꢀ andꢀ Lewisꢀ
sites)ꢀalsoꢀdecreasedꢀafterꢀNaOHꢀtreatment.ꢀ ꢀ
3
2
Byꢀ correlatingꢀ theꢀ resultsꢀ ofꢀ FT‐IR‐pyridineꢀ withꢀ theꢀ
3
2
NH
3
‐TPDꢀresults,ꢀitꢀcanꢀbeꢀconcludedꢀthatꢀtheꢀtreatmentꢀwithꢀ
notꢀ resultꢀ inꢀ aꢀ furtherꢀ performanceꢀ increase.ꢀ Itꢀ isꢀ widelyꢀ ac‐
ceptedꢀ thatꢀ dispersedꢀ tetrahedralꢀ tungstenꢀ oxideꢀ speciesꢀ areꢀ
theꢀ activeꢀ sitesꢀ forꢀ theꢀ metathesisꢀ reactionꢀ [11,17,27].ꢀ Here,ꢀ
Ramanꢀresultsꢀindicatedꢀaꢀ relativelyꢀhighꢀ proportionꢀofꢀtetra‐
NaOHꢀ resultedꢀ inꢀ aꢀ progressiveꢀ decreaseꢀ inꢀ totalꢀ acidity.ꢀ Theꢀ
remainingꢀ acidityꢀ wasꢀ attributedꢀ toꢀ Lewisꢀ acidꢀ sitesꢀ withꢀ aꢀ
strongꢀinteractionꢀwithꢀtheꢀsilicaꢀsupport.ꢀ ꢀ
1
00
100
(b)
80
(
a)
8
6
4
2
0
0
0
0
0
60
40
4
%WO
3
/SiO
2
8%WO
3 2
/SiO
4W+1.5NaOH
8W+1.5NaOH
4
%WO
3
/SiO
2
3 2
8%WO /SiO
4
W+1.5NaOH
8W+1.5NaOH
20
0
6%WO /SiO2
10%WO /SiO2
3
3
6%WO
3
/SiO
2
3 2
10%WO /SiO
10W+1.5NaOH
6W+1.5NaOH
10W+1.5NaOH
6W+1.5NaOH
0
2
4
6
8
10
0
2
4
6
8
10
Time on stream (h)
Time on stream (h)
Fig.ꢀ8.ꢀ2‐Buteneꢀconversionꢀ(a)ꢀandꢀpropyleneꢀselectivityꢀ(b)ꢀoverꢀdifferentꢀcatalysts.ꢀ

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
–
1
hedralꢀtungstenꢀoxideꢀspeciesꢀ(bandꢀatꢀ970ꢀcm )ꢀonꢀtheꢀmostꢀ
activeꢀcatalystꢀ(8%WO ).ꢀ
AfterꢀtheꢀtreatmentꢀwithꢀNaOH,ꢀtheꢀmetathesisꢀactivityꢀwasꢀ
significantlyꢀ decreasedꢀ onlyꢀ forꢀ theꢀ 4%WO /SiO ꢀ catalyst.ꢀ
pointedꢀoutꢀthatꢀLewisꢀacidꢀsitesꢀwereꢀshownꢀtoꢀbeꢀnecessaryꢀ
forꢀ theꢀ formationꢀ ofꢀ theꢀ initialꢀ metalꢀ carbeneꢀ speciesꢀ forꢀ theꢀ
metathesisꢀ reactionꢀ [13,18].ꢀ Here,ꢀ althoughꢀ theꢀ Lewisꢀ acidityꢀ
wasꢀdecreasedꢀbyꢀtheꢀNaOHꢀtreatment,ꢀtheꢀmetathesisꢀactivityꢀ
remainedꢀunchanged.ꢀThisꢀsuggestedꢀthatꢀevenꢀwithꢀtheꢀcata‐
lystsꢀwithꢀtheꢀlowestꢀacidity,ꢀtheꢀLewisꢀacidityꢀwasꢀstillꢀenoughꢀ
toꢀconvertꢀtheꢀreactantꢀolefins.ꢀ ꢀ
Tableꢀ 2ꢀ summarizesꢀ theꢀ selectivityꢀ toꢀ propyleneꢀ andꢀ theꢀ
otherꢀproductsꢀofꢀtheꢀreaction.ꢀTheꢀpropyleneꢀselectivityꢀoverꢀ
theꢀpristineꢀcatalystsꢀdecreasedꢀfromꢀ81.1%ꢀtoꢀ72.4%ꢀwhenꢀtheꢀ
tungstenꢀloadingꢀwasꢀincreasedꢀfromꢀ4%ꢀtoꢀ10%.ꢀTheꢀhighestꢀ
3
3
2
However,ꢀoneꢀcannotꢀdrawꢀdefinitiveꢀconclusionsꢀbecauseꢀoth‐
erꢀparametersꢀwereꢀchangedꢀasꢀwell.ꢀInꢀparticular,ꢀtheꢀtungstenꢀ
loadingꢀwasꢀdramaticallyꢀdecreasedꢀ(byꢀleaching).ꢀInteresting‐
ly,ꢀtheꢀmetathesisꢀactivityꢀofꢀtheꢀmodifiedꢀcatalystsꢀwithꢀhigherꢀ
tungstenꢀloadingꢀ(6%–10%)ꢀremainedꢀalmostꢀunchangedꢀafterꢀ
theꢀ NaOHꢀ treatment.ꢀ Thusꢀ theꢀ smallꢀ proportionꢀ ofꢀ theꢀ WO
speciesꢀthatꢀwereꢀleachedꢀfromꢀtheseꢀcatalystsꢀ(atꢀmostꢀ15%ꢀofꢀ
theꢀoriginalꢀWO ,ꢀandꢀmostꢀprobablyꢀtheseꢀwereꢀmainlyꢀcrystal‐
lineꢀWO )ꢀwereꢀspeciesꢀthatꢀwereꢀnotꢀactiveꢀinꢀtheꢀmetathesisꢀ
x
ꢀ
3
propyleneꢀ selectivityꢀ wasꢀ obtainedꢀ overꢀ 4%WO
wasꢀdueꢀprobablyꢀtoꢀtheꢀhigherꢀdispersionꢀofꢀtungstenꢀandꢀtheꢀ
lowerꢀproportionꢀofꢀWO ꢀcrystals.ꢀInterestingly,ꢀafterꢀtreatmentꢀ
withꢀ NaOH,ꢀ propeneꢀ selectivityꢀ increasedꢀ forꢀ allꢀ theꢀ catalystsꢀ
exceptꢀ4%WO /SiO .ꢀAtꢀtheꢀsameꢀtime,ꢀselectivityꢀtoꢀisomeriza‐
3
/SiO ,ꢀ whichꢀ
2
3
reaction.ꢀInꢀotherꢀwords,ꢀtheꢀactiveꢀspecies,ꢀisolatedꢀtetrahedralꢀ
speciesꢀwithꢀaꢀstrongꢀinteractionꢀwithꢀtheꢀsupport,ꢀremainedꢀonꢀ
theꢀcatalystꢀafterꢀNaOHꢀtreatment.ꢀThisꢀwasꢀalsoꢀsuggestedꢀbyꢀ
3
3
2
theꢀH
peakꢀwasꢀslightlyꢀdecreasedꢀandꢀshiftedꢀtoꢀ aꢀhigherꢀ tempera‐
tureꢀwithꢀtheꢀtreatedꢀcatalysts.ꢀHowever,ꢀonꢀtheseꢀNaOH‐treat‐
edꢀcatalysts,ꢀtheꢀacidityꢀwasꢀsignificantlyꢀmodifiedꢀasꢀcomparedꢀ
toꢀtheꢀpristineꢀcatalysts.ꢀTheꢀfactꢀthatꢀtheꢀactivityꢀremainedꢀatꢀ
theꢀsameꢀlevelꢀshowedꢀthatꢀtheꢀacidityꢀofꢀtheseꢀtreatedꢀcatalystꢀ
wasꢀ stillꢀ sufficientꢀtoꢀgenerateꢀactiveꢀmetathesisꢀsitesꢀ andꢀ forꢀ
theꢀmetathesisꢀreactionꢀtoꢀoccur.ꢀ ꢀ ꢀ
AsꢀcanꢀbeꢀseenꢀinꢀFig.ꢀ9,ꢀafterꢀpretreatmentꢀwithꢀNaOH,ꢀtheꢀ
totalꢀ acidityꢀ (measuredꢀ byꢀ liquidꢀ phaseꢀ titration)ꢀ andꢀ Lewisꢀ
acidityꢀ(evaluatedꢀbyꢀFT‐IRꢀofꢀpyridineꢀadsorption)ꢀdecreasedꢀ
progressivelyꢀwhenꢀhigherꢀconcentrationsꢀofꢀNaOHꢀwereꢀusedꢀ
forꢀ theꢀ ionꢀ exchange.ꢀ Atꢀ theꢀ highestꢀ concentrationꢀ ofꢀ NaOHꢀ
used,ꢀtheꢀLewisꢀacidityꢀwasꢀdecreasedꢀbyꢀaboutꢀ50%ꢀ(evaluatedꢀ
byꢀ theꢀ decreaseꢀ ofꢀ theꢀ FT‐IRꢀ peakꢀ intensity).ꢀ Itꢀ shouldꢀ beꢀ
2
‐TPRꢀdataꢀ(notꢀshownꢀinꢀthisꢀpaper)ꢀwhereꢀtheꢀreductionꢀ
tionꢀproductsꢀ(1‐butene,ꢀcis‐2‐butene)ꢀslightlyꢀdecreased.ꢀThusꢀ
theꢀ desiredꢀ metathesisꢀreactionꢀwasꢀ slightlyꢀ favoredꢀoverꢀtheꢀ
undesiredꢀsideꢀreactions.ꢀThisꢀwasꢀattributedꢀtoꢀtheꢀremovalꢀofꢀ
residualꢀBrönstedꢀacidityꢀonꢀtheꢀNaOH‐treatedꢀcatalysts.ꢀHow‐
ever,ꢀthisꢀwasꢀdifficultꢀtoꢀdemonstrateꢀbecauseꢀtheꢀFT‐IRꢀspec‐
troscopyꢀ ofꢀ adsorbedꢀ pyridineꢀ wasꢀ notꢀ sensitiveꢀ enoughꢀ toꢀ
probeꢀ quantitativelyꢀ theꢀ weakꢀ orꢀ scarceꢀ Brönstedꢀ acidꢀ sites.ꢀ
Comparedꢀ withꢀ Spamerꢀ etꢀ al.ꢀ [20]ꢀ whoꢀ aimedꢀ toꢀ reduceꢀ theꢀ
occurrenceꢀofꢀisomerizationꢀinꢀtheꢀliquidꢀ phaseꢀmetathesisꢀofꢀ
1‐octene,ꢀtheꢀobservedꢀeffectꢀofꢀalkalineꢀdopingꢀhereꢀwasꢀmod‐
est.ꢀNevertheless,ꢀtheꢀpropyleneꢀyieldsꢀincreasedꢀfromꢀ52.7%ꢀtoꢀ
55.3%,ꢀ fromꢀ 59%ꢀ toꢀ 63.0%,ꢀ andꢀ fromꢀ 57.2%ꢀ toꢀ 61.2%ꢀ whenꢀ
NaOHꢀionꢀexchangeꢀwasꢀappliedꢀtoꢀtheꢀ6%,ꢀ8%,ꢀandꢀ10%ꢀload‐
edꢀ catalysts,ꢀ respectively.ꢀ Thus,ꢀ whenꢀ Wꢀ leachingꢀ isꢀ notꢀ tooꢀ
pronounced,ꢀNaOHꢀionꢀexchangeꢀappearsꢀtoꢀbeꢀaꢀgoodꢀwayꢀtoꢀ
ꢀ
7
6
5
4
3
2
100
8
7
6
5
4
3
2
100
80
60
40
20
4
% WO
3
6
% WO
3
80
60
40
20
80
40
50
60
70
30
40
50
60
70
80
90
Total acidity (mol/g)
Total acidity (mol/g)
8
7
6
5
4
100
80
60
40
20
12
100
8
% WO
3
10% WO
3
1
1
0
9
8
7
6
5
80
60
40
20
1
40
50
60
Total acidity (mol/g)
Fig.ꢀ9.ꢀCorrelationꢀbetweenꢀtotalꢀacidity,ꢀLewisꢀacidity,ꢀandꢀtrans‐2‐buteneꢀconversion.ꢀ
70
80
90
100
40
50
60
70
80
90
100 110
Total acidity (mol/g)

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
ꢀ
Tableꢀ2ꢀ
Activityꢀofꢀpristineꢀcatalystsꢀandꢀmodifiedꢀcatalystsꢀwithꢀdifferentꢀtungstenꢀloading.ꢀ
Conversionꢀofꢀ ꢀ
trans‐2‐buteneꢀ(%)ꢀ
Selectivityꢀ(%)ꢀ
1,3‐Butadieneꢀ Cis‐2‐buteneꢀ
Yieldꢀofꢀpropylene
(%)ꢀ
Catalystꢀ
Propyleneꢀ
81.1ꢀ
64.0ꢀ
67.2ꢀ
78.5ꢀ
76.5ꢀ
76.7ꢀ
77.2ꢀ
74.6ꢀ
77.6ꢀ
76.6ꢀ
72.4ꢀ
73.6ꢀ
75.0ꢀ
75.3ꢀ
1‐Buteneꢀ
ꢀ 8.2ꢀ
17.8ꢀ
16.5ꢀ
10.2ꢀ
10.5ꢀ
11.0ꢀ
10.6ꢀ
10.8ꢀ
10.1ꢀ
ꢀ 9.9ꢀ
12.1ꢀ
11.8ꢀ
10.7ꢀ
10.9ꢀ
Othersꢀ
1.3ꢀ
2.1ꢀ
1.9ꢀ
2.5ꢀ
2.2ꢀ
2.3ꢀ
2.1ꢀ
2.7ꢀ
2.6ꢀ
2.8ꢀ
3.0ꢀ
2.9ꢀ
2.9ꢀ
2.8ꢀ
4
4
4
4
6
6
6
8
8
8
1
1
1
1
%WO
3
/SiO
2
ꢀ
58.2ꢀ
48.6ꢀ
31.7ꢀ
26.2ꢀ
68.9ꢀ
72.1ꢀ
70.4ꢀ
79.3ꢀ
81.2ꢀ
80.4ꢀ
79.0ꢀ
80.4ꢀ
81.6ꢀ
80.8ꢀ
0ꢀ
ꢀ 9.4ꢀ
16.1ꢀ
14.4ꢀ
ꢀ 8.8ꢀ
10.6ꢀ
ꢀ 9.8ꢀ
10.0ꢀ
11.5ꢀ
ꢀ 9.5ꢀ
10.4ꢀ
12.0ꢀ
11.4ꢀ
11.1ꢀ
10.7ꢀ
47.2ꢀ
31.1ꢀ
21.3ꢀ
20.6ꢀ
52.7ꢀ
55.3ꢀ
54.3ꢀ
59.2ꢀ
63.0ꢀ
61.6ꢀ
57.2ꢀ
59.2ꢀ
61.2ꢀ
60.8ꢀ
W+0.5NaOHꢀ
W+1.0NaOHꢀ
W+1.5NaOHꢀ
%WO
W+1.0NaOHꢀ
W+1.5NaOHꢀ
0ꢀ
0ꢀ
0ꢀ
3
/SiO
2
ꢀ
0.2ꢀ
0.2ꢀ
0.1ꢀ
0.4ꢀ
0.3ꢀ
0.3ꢀ
0.5ꢀ
0.4ꢀ
0.3ꢀ
0.3ꢀ
%WO
3
/SiO ꢀ
2
W+1.0NaOHꢀ
W+1.5NaOHꢀ
0%WO
0W+0.5NaOHꢀ
0W+1.0NaOHꢀ
0W+1.5NaOHꢀ
3
/SiO ꢀ
2
Allꢀreactionsꢀwereꢀonlineꢀforꢀ10ꢀh.ꢀReactionꢀconditionsꢀwereꢀ400ꢀ°Cꢀandꢀ0.1ꢀMPa.
increaseꢀtheꢀpropeneꢀyieldꢀaꢀlittleꢀbitꢀmore.ꢀ ꢀ
theꢀpristineꢀcatalystꢀandꢀmodifiedꢀcatalystꢀreachedꢀstableꢀcon‐
Theꢀmetathesisꢀreactionꢀmechanismꢀincludesꢀtheꢀformationꢀ
ofꢀ aꢀ surfaceꢀ carbeneꢀ byꢀ theꢀ reactionꢀ ofꢀ tungstenꢀ oxoꢀ speciesꢀ
versionꢀ andꢀ selectivityꢀ andꢀ didꢀ notꢀ showꢀ deactivationꢀ inꢀ theꢀ
timeꢀ frameꢀ ofꢀ ourꢀ experiment.ꢀ However,ꢀ thisꢀ stableꢀ behaviorꢀ
wasꢀonlyꢀreachedꢀafterꢀanꢀinductionꢀperiodꢀofꢀaboutꢀ10ꢀhꢀwithꢀ
theꢀ pristineꢀ catalystꢀ whileꢀ itꢀ isꢀ almostꢀ immediatelyꢀ reachedꢀ
withꢀtheꢀmodifiedꢀcatalystꢀasꢀdescribedꢀabove.ꢀ ꢀ
withꢀtheꢀolefinsꢀinꢀtheꢀfeed.ꢀTheꢀstabilityꢀofꢀWO ꢀsurfaceꢀspeciesꢀ
x
andꢀtheirꢀtendencyꢀtoꢀreactꢀwithꢀtheꢀolefinsꢀisꢀalsoꢀdictatedꢀbyꢀ
theꢀ structureꢀofꢀtheꢀ surfaceꢀtungstenꢀoxideꢀ speciesꢀ[15].ꢀWellꢀ
dispersedꢀ tetrahedralꢀ tungstenꢀ speciesꢀ reactꢀ betterꢀ withꢀ theꢀ
olefinsꢀ toꢀ yieldꢀ theꢀ desiredꢀ carbeneꢀ speciesꢀ [11,13,35].ꢀ Onꢀ
comparingꢀ theꢀ catalyticꢀ performanceꢀ ofꢀ theꢀ pristineꢀ catalystsꢀ
TGAꢀanalysisꢀwasꢀusedꢀforꢀdeterminingꢀcokeꢀformationꢀonꢀ
theꢀspentꢀcatalystsꢀcomparedꢀtoꢀtheꢀpristineꢀcatalystsꢀandꢀmod‐
ifiedꢀcatalysts.ꢀCokeꢀonꢀoxideꢀcatalystsꢀisꢀthoughtꢀtoꢀariseꢀfromꢀ
alkeneꢀ polymerization,ꢀ alkeneꢀ cyclization,ꢀ andꢀ polyaromaticꢀ
formationꢀ [23].ꢀ Theseꢀ reactionsꢀ areꢀ catalyzedꢀ byꢀ acidicꢀ sites,ꢀ
especiallyꢀ Brönstedꢀ acidꢀ sitesꢀ [23,24].ꢀ Cokeꢀ formationꢀ canꢀ beꢀ
reducedꢀ byꢀ addingꢀ promoters,ꢀ whichꢀ canꢀ partiallyꢀ eliminateꢀ
acidicꢀsitesꢀ[25],ꢀorꢀbyꢀanꢀacidꢀleachingꢀpost‐treatmentꢀ[22].ꢀInꢀ
theꢀpresentꢀwork,ꢀtheꢀeffectꢀofꢀcokeꢀformationꢀafterꢀtheꢀtreat‐
mentꢀwithꢀNaOHꢀwasꢀalsoꢀstudied.ꢀAfterꢀ10ꢀhꢀmetathesisꢀreac‐
tionꢀatꢀ 400ꢀ°C,ꢀ theꢀtemperatureꢀwasꢀdecreasedꢀtoꢀroomꢀtem‐
peratureꢀandꢀthenꢀaꢀnitrogenꢀflowꢀwasꢀadmitted.ꢀTableꢀ3ꢀshowsꢀ
thatꢀtheꢀcokeꢀcontentsꢀonꢀallꢀsamplesꢀwereꢀquiteꢀlowꢀasꢀcom‐
paredꢀ toꢀ theꢀ longꢀ runꢀ experimentsꢀ reportedꢀ inꢀ theꢀ literatureꢀ
[21].ꢀTheꢀresultꢀalsoꢀshowedꢀclearlyꢀthatꢀtheꢀcokeꢀcontentꢀonꢀ
theꢀmodifiedꢀcatalystꢀwasꢀlowerꢀthanꢀthatꢀonꢀtheꢀpristineꢀcata‐
lyst.ꢀ Thisꢀ wasꢀ alsoꢀ seenꢀ withꢀ theꢀ TGAꢀ andꢀ DTGꢀ profilesꢀ (notꢀ
shownꢀhere).ꢀTheꢀDTGꢀofꢀbothꢀcokedꢀcatalystsꢀshowedꢀaꢀsimilarꢀ
patternꢀ ofꢀ massꢀ loss.ꢀ Twoꢀ massꢀ lossꢀ peaksꢀ wereꢀ observedꢀ atꢀ
(
Fig.ꢀ8),ꢀ4%WO
shortꢀ inductionꢀ period.ꢀ Catalystsꢀ withꢀ higherꢀ loadingꢀ
6%–10%)ꢀhadꢀaꢀlongerꢀinductionꢀperiod,ꢀsuggestingꢀthatꢀtheꢀ
3
/SiO
2
ꢀreachedꢀmodestꢀactivityꢀafterꢀaꢀrelativelyꢀ
(
formationꢀofꢀtheꢀcarbeneꢀspeciesꢀisꢀfacilitatedꢀwhenꢀtheꢀdisper‐
sionꢀisꢀhigher.ꢀMoreꢀimportantꢀwasꢀthatꢀtheꢀNaOH‐treatedꢀcat‐
alystsꢀreachedꢀtheꢀsteadyꢀstateꢀmuchꢀfasterꢀthanꢀdidꢀtheꢀpris‐
tineꢀcatalystsꢀ(Fig.ꢀ8).ꢀThisꢀwasꢀcorrelatedꢀwithꢀthatꢀtheꢀdisper‐
sionꢀofꢀtungstenꢀinꢀtheꢀformꢀofꢀtetrahedralꢀspeciesꢀwasꢀfavoredꢀ
byꢀtheꢀNaOHꢀtreatment.ꢀ
Figureꢀ10ꢀdepictsꢀtheꢀeffectꢀofꢀNaOHꢀtreatmentꢀonꢀtheꢀstabil‐
ityꢀ ofꢀ theꢀ catalystꢀ inꢀ theꢀ firstꢀ stageꢀ ofꢀ theꢀ reactionꢀ andꢀ alsoꢀ
showsꢀcatalystꢀstabilityꢀoverꢀaꢀlongerꢀreactionꢀrunꢀ(30ꢀh).ꢀBothꢀ
9
8
7
6
5
4
0
0
0
0
0
0
90
80
70
60
50
40
280ꢀandꢀ460ꢀ°C,ꢀwhichꢀcorrespondedꢀtoꢀaliphaticꢀhydrocarbonsꢀ
(softꢀcoke)ꢀandꢀpolyaromaticsꢀ(hardꢀcoke),ꢀrespectivelyꢀ[36,37].ꢀ
Tableꢀ3ꢀ
CokedꢀWO
3
/SiO ꢀcatalystsꢀforꢀcomparingꢀtheꢀcokeꢀcontentꢀonꢀtheꢀpris‐
2
tineꢀandꢀmodifiedꢀcatalysts.ꢀ
Catalystꢀ
Cokeꢀ(%)ꢀ
0.14ꢀ
0.06ꢀ
Pristine catalyst
Modified catalyst
6%WO
6W+1.5NaOHꢀ
%WO /SiO
8W+1.5NaOHꢀ
10%WO /SiO
0W+1.5NaOHꢀ
3
/SiO
2
ꢀ
8
3
2
ꢀ
0.16ꢀ
0.08ꢀ
0.22ꢀ
0.11ꢀ
0
5
10
Time on stream (h)
Fig.ꢀ10.ꢀStabilityꢀofꢀpristineꢀandꢀmodifiedꢀcatalystꢀwithꢀ8%ꢀWO
15
20
25
30
3
2
ꢀ
1
3
.ꢀ

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
Theꢀ intensityꢀ ofꢀ bothꢀ massꢀ lossꢀ peaksꢀ decreasedꢀ afterꢀ theꢀ
treatmentꢀwithꢀNaOH,ꢀindicatingꢀthatꢀtheꢀacidityꢀthatꢀisꢀrespon‐
sibleꢀforꢀcokeꢀformationꢀwasꢀreduced.ꢀThusꢀtheꢀbenefitsꢀofꢀacidꢀ
siteꢀpoisoningꢀalsoꢀincludedꢀaꢀprolongedꢀlifetimeꢀofꢀtheꢀcatalyst.ꢀ
Noteꢀthatꢀcokeꢀformationꢀdidꢀnotꢀresultꢀinꢀanyꢀsignificantꢀmodi‐
ficationꢀofꢀtheꢀcatalystꢀtextureꢀ(dataꢀnotꢀshown).ꢀMoodleyꢀetꢀal.ꢀ
Acknowledgementsꢀ
AuthorsꢀthankꢀtheꢀSCGꢀChemicalꢀCo.,ꢀLtd.ꢀforꢀtheirꢀfinancialꢀ
support.ꢀ
Referencesꢀ
[21]ꢀusedꢀEFTEMꢀtoꢀdetermineꢀtheꢀlocationꢀofꢀtheꢀcokeꢀonꢀtheꢀ
WO /SiO ꢀcatalystꢀandꢀsuggestedꢀthatꢀtheꢀcokeꢀlaydownꢀratherꢀ
3
2
[1] vanꢀ Schalkwykꢀ C,ꢀ Spamerꢀ A,ꢀ Moodleyꢀ Dꢀ J,ꢀ Dubeꢀ T,ꢀ Reynhardtꢀ J,ꢀ
BothaꢀJꢀM.ꢀApplꢀCatalꢀA,ꢀ2003,ꢀ255:ꢀ121ꢀ
2] LokhatꢀD,ꢀStarzakꢀM,ꢀStelmachowskiꢀM.ꢀApplꢀCatalꢀA,ꢀ2008,ꢀ351:ꢀ
thanꢀtheꢀamountꢀofꢀcokeꢀwasꢀdecisiveꢀinꢀdeterminingꢀtheꢀactiv‐
ity.ꢀ Inꢀ theꢀ presentꢀ study,ꢀ theꢀ activityꢀ ofꢀ theꢀ modifiedꢀ catalystꢀ
wasꢀmaintainedꢀatꢀtheꢀsameꢀlevelꢀforꢀ30ꢀh,ꢀindicatingꢀgoodꢀsta‐
bility,ꢀasꢀcomparedꢀtoꢀaꢀpreviousꢀreportꢀ[8].ꢀNevertheless,ꢀcokeꢀ
formationꢀ duringꢀ prolongedꢀ catalyticꢀ reactionꢀ shouldꢀ beꢀ fur‐
therꢀinvestigated.ꢀ ꢀ
[
[
[
[
137ꢀ
3] DebeckerꢀDꢀP,ꢀHauwaertꢀD,ꢀStoyanovaꢀM,ꢀBarkschatꢀA,ꢀRodemerckꢀ
U,ꢀGaigneauxꢀEꢀM.ꢀApplꢀCatalꢀA,ꢀ2011,ꢀ391:ꢀ78ꢀ
4] DebeckerꢀDꢀP,ꢀStoyanovaꢀM,ꢀRodemerckꢀU,ꢀGaigneauxꢀEꢀM.ꢀJꢀMolꢀ
CatalꢀA,ꢀ2011,ꢀ340:ꢀ65ꢀ
5] Phongsawatꢀ W,ꢀ Netivorruksaꢀ B,ꢀ Suriyeꢀ K,ꢀ Dokjampaꢀ S,ꢀ
PraserthdamꢀP,ꢀPanpranotꢀJ.ꢀJꢀNatꢀGasꢀChem,ꢀ2012,ꢀ21:ꢀ83ꢀ
Inꢀsummary,ꢀNaꢀdopingꢀdidꢀnotꢀaffectꢀsignificantlyꢀtheꢀme‐
tathesisꢀ activityꢀ ofꢀ WO
3
/SiO ꢀ catalystsꢀ butꢀ reducedꢀ slightlyꢀ
2
theirꢀisomerizationꢀactivity.ꢀAsꢀaꢀresult,ꢀNaOHꢀionꢀexchangeꢀisꢀaꢀ
goodꢀwayꢀtoꢀincreaseꢀpropyleneꢀyield.ꢀTheꢀeffectꢀofꢀthisꢀtreat‐
mentꢀ wasꢀ beneficialꢀ bothꢀ inꢀ helpingꢀ theꢀ catalystꢀ reachꢀ theꢀ
steadyꢀstateꢀinꢀaꢀshorterꢀtimeꢀandꢀminimizingꢀcokeꢀdeposition,ꢀ
[6] Phongsawatꢀ W,ꢀ Netiworaruksaꢀ B,ꢀ Suriyeꢀ K,ꢀ Dokjampaꢀ S,ꢀ
PraserthdamꢀP,ꢀPanpranotꢀJ.ꢀJꢀNatꢀGasꢀChem,ꢀ2012,ꢀ21:ꢀ158ꢀ
[7] Bouchmellaꢀ K,ꢀ Mutinꢀ Pꢀ H,ꢀ Stoyanovaꢀ M,ꢀ Poleunisꢀ C,ꢀ Eloyꢀ P,ꢀ
RodemerckꢀU,ꢀGaigneauxꢀEꢀM,ꢀDebeckerꢀDꢀP.ꢀJꢀCatal,ꢀ2013,ꢀ301:ꢀ
233ꢀ
whichꢀincreasedꢀtheꢀlifetimeꢀofꢀtheꢀWO
3
/SiO ꢀcatalyst.ꢀ
2
[8] ChaemchuenꢀS,ꢀPhatanasriꢀS,ꢀVerpoortꢀF,ꢀSae‐MaꢀN,ꢀSuriyeꢀK.ꢀKinetꢀ
Catal,ꢀ2012,ꢀ53:ꢀ247ꢀ
4
.ꢀ ꢀ Conclusionsꢀ
[9] Chaemchuenꢀ S,ꢀ Limsangkassꢀ W,ꢀ Netiworaraksaꢀ B,ꢀ Phatanasriꢀ S,ꢀ
Sae‐MaꢀN,ꢀSuriyeꢀK.ꢀBulgꢀChemꢀCommun,ꢀ2012,ꢀ44:ꢀ87ꢀ
IonꢀexchangeꢀwithꢀNaOHꢀwasꢀusedꢀtoꢀmitigateꢀtheꢀacidityꢀofꢀ
WO /SiO ꢀmetathesisꢀcatalysts.ꢀThisꢀtreatmentꢀhadꢀaꢀlargeꢀef‐
fectꢀonꢀtheꢀcatalystꢀwithꢀaꢀlowꢀWO ꢀloading,ꢀesp.ꢀinꢀcausingꢀaꢀ
markedꢀ leachingꢀ ofꢀ theꢀ Wꢀ species.ꢀ Onꢀ theꢀ catalystsꢀ withꢀ 6%ꢀ
WO ꢀtoꢀ10%ꢀWO ,ꢀtheꢀionꢀexchangeꢀmainlyꢀhadꢀanꢀeffectꢀonꢀtheꢀ
acidity.ꢀAlthoughꢀbothꢀLewisꢀandꢀBrönstedꢀacidꢀsitesꢀwereꢀde‐
creased,ꢀtheꢀmetathesisꢀacidityꢀremainedꢀtheꢀsame.ꢀThisꢀindi‐
catedꢀ thatꢀ theꢀ acidꢀ sitesꢀ onꢀ theꢀ NaOH‐treatedꢀ catalystsꢀ wereꢀ
stillꢀenoughꢀforꢀtheꢀactiveꢀspeciesꢀtoꢀform.ꢀHowever,ꢀtheꢀisom‐
erizationꢀ activityꢀ wasꢀ slightlyꢀ decreasedꢀ byꢀ theꢀ NaOHꢀ treat‐
ment.ꢀThusꢀNaOHꢀtreatmentꢀisꢀaꢀgoodꢀwayꢀtoꢀtuneꢀtheꢀselectiv‐
[
10] Debeckerꢀ Dꢀ P,ꢀ Stoyanovaꢀ M,ꢀ Rodemerckꢀ U,ꢀ Colbeau‐Justinꢀ F,ꢀ
BoissèreꢀC,ꢀChaumonnotꢀA,ꢀBonduelleꢀA,ꢀSanchezꢀC.ꢀApplꢀCatalꢀA,ꢀ
http://dx.doi.org/10.1016/j.apcata.2013.06.041ꢀ
3
2
3
[11] MolꢀJꢀC.ꢀJꢀMolꢀCatalꢀA,ꢀ2004,ꢀ213:ꢀ39ꢀ
[12] HerissonꢀJꢀL,ꢀChauvinꢀY.ꢀMakromolꢀChem,ꢀ1971,ꢀ141:ꢀ161ꢀ
[13] VanꢀRoosmalenꢀAꢀJ,ꢀMolꢀJꢀC.ꢀJꢀCatal,ꢀ1982,ꢀ78:ꢀ17ꢀ
3
3
[
[
[
[
[
14] GrünertꢀW,ꢀ MörkeꢀW,ꢀ FeldhausꢀR,ꢀ AndersꢀK.ꢀ JꢀCatal,ꢀ 1989,ꢀ 117:ꢀ
485ꢀ
15] HuangꢀSꢀJ,ꢀLiuꢀSꢀL,ꢀXinꢀWꢀJ,ꢀBaiꢀJ,ꢀXieꢀSꢀJ,ꢀWangꢀQꢀX,ꢀXuꢀLꢀY.ꢀJꢀMolꢀ
CatalꢀA,ꢀ2005,ꢀ226:ꢀ61ꢀ
16] HuangꢀSꢀJ,ꢀLiuꢀSꢀL,ꢀZhuꢀQꢀJ,ꢀZhuꢀXꢀX,ꢀXinꢀWꢀJ,ꢀLiuꢀHꢀJ,ꢀFengꢀZꢀC,ꢀLiꢀC,ꢀ
XieꢀSꢀJ,ꢀWangꢀQꢀX,ꢀXuꢀLꢀY.ꢀApplꢀCatalꢀA,ꢀ2007,ꢀ323:ꢀ94ꢀ
17] HuangꢀSꢀJ,ꢀChenꢀFꢀC,ꢀLiuꢀSꢀL,ꢀZhuꢀQꢀJ,ꢀZhuꢀXꢀX,ꢀXinꢀWꢀJ,ꢀFengꢀZꢀC,ꢀLiꢀC,ꢀ
WangꢀQꢀX,ꢀXuꢀLꢀY.ꢀJꢀMolꢀCatalꢀA,ꢀ2007,ꢀ267:ꢀ224ꢀ
18] AndreiniꢀA,ꢀMolꢀJꢀC.ꢀJꢀChemꢀSoc,ꢀFaradayꢀTransꢀ1,ꢀ1985,ꢀ81:ꢀ1705ꢀ
ityꢀ ofꢀ WO
3
/SiO ꢀ catalystsꢀ toꢀ giveꢀ aꢀ littleꢀ moreꢀ propeneꢀ yieldꢀ
2
withoutꢀ affectingꢀ theirꢀ metathesisꢀ activity.ꢀ Otherꢀ beneficialꢀ
effectsꢀ ofꢀ theꢀ NaOHꢀ treatmentꢀ wereꢀ toꢀ shortenꢀ theꢀ inductionꢀ
periodꢀandꢀtoꢀdecreaseꢀdeactivationꢀbyꢀcokeꢀformation.ꢀ
[19] KawaiꢀT,ꢀGotoꢀH,ꢀYamazakiꢀY,ꢀIshikawaꢀT.ꢀ JꢀMolꢀCatal,ꢀ1988,ꢀ46:ꢀ
GraphicalꢀAbstractꢀ
Chin.ꢀJ.ꢀCatal.,ꢀ2014,ꢀ35:ꢀ232–241ꢀ ꢀ ꢀ doi:ꢀ10.1016/S1872‐2067(12)60760‐8
NaOHꢀmodifiedꢀWO /SiO ꢀcatalystsꢀforꢀpropyleneꢀproductionꢀfromꢀ
3
2
80
2
‐buteneꢀandꢀethyleneꢀmetathesisꢀ
7
0
NaOH‐treated catalyst
Pristine catalyst
ꢀ
SurasaꢀMaksasithorn,ꢀDamienꢀP.ꢀDebecker,ꢀPiyasanꢀPraserthdam*,ꢀ ꢀ
JoongjaiꢀPanpranot,ꢀKongkiatꢀSuriye,ꢀSirachayaꢀKunjaraꢀNaꢀAyudhyaꢀ
ChulalongkornꢀUniversity,ꢀThailand;ꢀ
UniversitéꢀcatholiqueꢀdeꢀLouvain,ꢀBelgium;ꢀ
SCGꢀChemicals,ꢀCo.,ꢀLtd.,ꢀThailandꢀ
60
5
4
0
0
30
W0 /SiO2
3
ꢀ
2
1
0
0
0
ꢀ
AfterꢀtreatmentꢀofꢀaꢀWO
conversionꢀwasꢀnotꢀchangedꢀbutꢀisomerizationꢀwasꢀslightlyꢀreduced,ꢀandꢀ
theꢀfinalꢀpropeneꢀyieldꢀandꢀoverallꢀcatalystꢀstabilityꢀwereꢀimproved.ꢀ
3
/SiO ꢀcatalystꢀwithꢀNaOHꢀtoꢀreduceꢀitsꢀacidity,ꢀtheꢀ
2
0
5
10
15
20
25
30
Time on stream(h)

ꢀ
SurasaꢀMaksasithornꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ35ꢀ(2014)ꢀ232–241ꢀ
ꢀ
157ꢀ
[29] VerpoortꢀF,ꢀBossuytꢀAꢀR,ꢀVerdonckꢀL.ꢀJꢀMolꢀCatalꢀA,ꢀ1995,ꢀ95:ꢀ75ꢀ
[
[
[
20] SpamerꢀA,ꢀDubeꢀTꢀI,ꢀMoodleyꢀDꢀJ,ꢀvanꢀSchalkwykꢀC,ꢀBothaꢀJꢀM.ꢀApplꢀ
[30] ChenꢀJꢀQ,ꢀBozzanoꢀA,ꢀGloverꢀB,ꢀFuglerudꢀT,ꢀKvisleꢀS.ꢀCatalꢀToday,ꢀ
2005,ꢀ106:ꢀ103ꢀ
[31] LeeꢀEꢀL,ꢀWachsꢀIꢀE.ꢀJꢀPhysꢀChemꢀC,ꢀ2008,ꢀ112:ꢀ6487ꢀ
[32] BarzettiꢀT,ꢀSelliꢀE,ꢀMoscottiꢀD,ꢀForniꢀL.ꢀJꢀChemꢀSoc,ꢀFaradayꢀTrans,ꢀ
1996,ꢀ92:ꢀ1401ꢀ
[33] Martínꢀ C,ꢀ Maletꢀ P,ꢀ Solanaꢀ G,ꢀ Rivesꢀ V.ꢀ JꢀPhysꢀChemꢀ B,ꢀ 1998,ꢀ 102:ꢀ
2759ꢀ
[34] Rodríguez‐RamosꢀI,ꢀGuerrero‐RuizꢀA,ꢀHomsꢀN,ꢀdeꢀlaꢀPiscinaꢀPꢀR,ꢀ
FierroꢀJꢀLꢀG.ꢀJꢀMolꢀCatalꢀA,ꢀ1995,ꢀ95:ꢀ147ꢀ
[35] HeckelsbergꢀLꢀF,ꢀBanksꢀRꢀL,ꢀBaileyꢀGꢀC.ꢀIndꢀEngꢀChemꢀProdꢀResꢀDev,ꢀ
1968,ꢀ7:ꢀ29ꢀ
[36] SiddiquiꢀMꢀRꢀH,ꢀHolmesꢀS,ꢀHeꢀH,ꢀSmithꢀW,ꢀCokerꢀEꢀN,ꢀAtkinsꢀMꢀP,ꢀ
KozhevnikovꢀIꢀV.ꢀCatalꢀLett,ꢀ2000,ꢀ66:ꢀ53ꢀ
CatalꢀA,ꢀ2003,ꢀ255:ꢀ153ꢀ
21] MoodleyꢀDꢀJ,ꢀvanꢀSchalkwykꢀC,ꢀSpamerꢀA,ꢀBothaꢀJꢀM,ꢀDatyeꢀAꢀK.ꢀ
ApplꢀCatalꢀA,ꢀ2007,ꢀ318:ꢀ155ꢀ
22] HuangꢀSꢀJ,ꢀLiuꢀHꢀJ,ꢀZhangꢀL,ꢀLiuꢀSꢀL,ꢀXinꢀWꢀJ,ꢀLiꢀXꢀJ,ꢀXieꢀSꢀJ,ꢀXuꢀLꢀY.ꢀ
ApplꢀCatalꢀA,ꢀ2011,ꢀ404:ꢀ113ꢀ
[23] ForzattiꢀP,ꢀLiettiꢀL.ꢀCatalꢀToday,ꢀ1999,ꢀ52:ꢀ165ꢀ
[24] BartholomewꢀCꢀH.ꢀApplꢀCatalꢀA,ꢀ2001,ꢀ212:ꢀ17ꢀ
[25] HoriuchiꢀT,ꢀSakumaꢀK,ꢀFukuiꢀT,ꢀKuboꢀY,ꢀOsakiꢀT,ꢀMoriꢀT.ꢀApplꢀCatalꢀ
A,ꢀ1996,ꢀ144:ꢀ111ꢀ
[
26] LópezꢀDꢀE,ꢀGoodwinꢀJꢀGꢀJr,ꢀBruceꢀDꢀA.ꢀJꢀCatal,ꢀ2007,ꢀ245:ꢀ381ꢀ
27] ZhaoꢀQꢀF,ꢀChenꢀSꢀL,ꢀGaoꢀJꢀS,ꢀXuꢀCꢀM.ꢀTransitionꢀMetꢀChem,ꢀ2009,ꢀ34:ꢀ
[
621ꢀ
[
28] WangꢀYꢀD,ꢀChenꢀQꢀL,ꢀYangꢀWꢀM,ꢀXieꢀZꢀK,ꢀXuꢀW,ꢀHuangꢀDꢀY.ꢀApplꢀ
[37] LiuꢀHꢀJ,ꢀZhangꢀL,ꢀLiꢀXꢀJ,ꢀHuangꢀSꢀJ,ꢀLiuꢀSꢀL,ꢀXinꢀWꢀJ,ꢀXieꢀSꢀJ,ꢀXuꢀLꢀY.ꢀJꢀ
NatꢀGasꢀChem,ꢀ2009,ꢀ18:ꢀ331ꢀ
CatalꢀA,ꢀ2003,ꢀ250:ꢀ25ꢀ