Hydrogenation of 2-ethylanthraquinone with bimetallic monolithic catalysts: An experimental and DFT study

Article abstract of DOI:10.1016/S1872-2067(18)63035-9

We studied the hydrogenation of 2-ethylanthraquinone (eAQ) over Pd/SiO₂/COR (COR = cordierite) monometallic and Pd-M/SiO₂/COR (M = Ni, Fe, Mn, and Cu) bimetallic monolithic catalysts, which were prepared by the co-impregnation method. Detailed investigations showed that the particle sizes and structures of the Pd-M (M = Ni, Fe, Mn, and Cu) bimetallic monolithic catalysts were greatly affected by the second metal M and the mass ratio of Pd to the second metal M. By virtue of the small particle size and the strong interaction between Pd and Ni of Pd-Ni alloy, Pd-Ni bimetallic monolithic catalysts with the mass ratio of Pd/Ni = 2 achieved the highest H₂O₂ yield (7.5 g/L) and selectivity (95.3%). Moreover, density functional theory calculations were performed for eAQ adsorption to gain a better mechanistic understanding of the molecule-surface interactions between eAQ and the Pd(1 1 1) or PdM(1 1 1) (M = Ni, Fe, Mn, and Cu) surfaces. It was found that the high activity of the bimetallic Pd-Ni catalyst was a result of strong chemisorption between Pd₃Ni₁ (1 1 1) and the carbonyl group of eAQ.

Full text of DOI:10.1016/S1872-2067(18)63035-9

ChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080
ꢀ
ꢀ ꢀ ꢀ
ꢀ ꢀ ꢀ ꢀ
a v a i l a b l e a t w w w . s c i e n c e d 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
Hydrogenationꢀofꢀ2‐ethylanthraquinoneꢀwithꢀbimetallicꢀmonolithicꢀ
catalysts:ꢀAnꢀexperimentalꢀandꢀDFTꢀstudyꢀ
YanyanꢀGuo,ꢀChengnaꢀDai,ꢀZhigangꢀLeiꢀ*
ꢀ
StateꢀKeyꢀLaboratoryꢀofꢀChemicalꢀResourceꢀEngineering,ꢀBeijingꢀUniversityꢀofꢀChemicalꢀTechnology,ꢀBeijingꢀ100029,ꢀChinaꢀ
A R T I C L E ꢀ I N F O ꢀ
A B S T R A C T ꢀ
Articleꢀhistory:ꢀ
Weꢀstudiedꢀtheꢀhydrogenationꢀofꢀ2‐ethylanthraquinoneꢀ(eAQ)ꢀoverꢀPd/SiO /CORꢀ(CORꢀ=ꢀcordierite)ꢀ
2
Receivedꢀ27ꢀDecemberꢀ2017ꢀ
Acceptedꢀ24ꢀJanuaryꢀ2018ꢀ
Publishedꢀ5ꢀJuneꢀ2018ꢀ
monometallicꢀandꢀPd‐M/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀbimetallicꢀmonolithicꢀcatalysts,ꢀwhichꢀ
2
wereꢀ preparedꢀ byꢀ theꢀ co‐impregnationꢀ method.ꢀ Detailedꢀ investigationsꢀ showedꢀ thatꢀtheꢀ particleꢀ
sizesꢀandꢀstructuresꢀofꢀtheꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀbimetallicꢀmonolithicꢀcatalystsꢀwereꢀgreat‐
lyꢀaffectedꢀbyꢀtheꢀsecondꢀmetalꢀMꢀandꢀtheꢀmassꢀratioꢀofꢀPdꢀtoꢀtheꢀsecondꢀmetalꢀM.ꢀByꢀvirtueꢀofꢀtheꢀ
smallꢀparticleꢀsizeꢀandꢀtheꢀstrongꢀinteractionꢀbetweenꢀPdꢀandꢀNiꢀofꢀPd‐Niꢀalloy,ꢀPd‐Niꢀbimetallicꢀ
Keywords:ꢀ
Bimetallicꢀmonolithicꢀcatalystꢀ
Alloyꢀ
monolithicꢀcatalystsꢀwithꢀtheꢀmassꢀratioꢀofꢀPd/Niꢀ=ꢀ2ꢀachievedꢀtheꢀhighestꢀH
2
O ꢀyieldꢀ(7.5ꢀg/L)ꢀandꢀ
2
selectivityꢀ(95.3%).ꢀMoreover,ꢀdensityꢀfunctionalꢀtheoryꢀcalculationsꢀwereꢀperformedꢀforꢀeAQꢀad‐
sorptionꢀtoꢀgainꢀaꢀbetterꢀmechanisticꢀunderstandingꢀofꢀtheꢀmolecule‐surfaceꢀinteractionsꢀbetweenꢀ
eAQꢀandꢀtheꢀPd(1ꢀ1ꢀ1)ꢀorꢀPdM(1ꢀ1ꢀ1)ꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀsurfaces.ꢀItꢀwasꢀfoundꢀthatꢀtheꢀhighꢀ
2‐Ethylanthraquinoneꢀhydrogenationꢀ
DFTꢀcalculationꢀ
Synergisticꢀeffect
activityꢀofꢀtheꢀbimetallicꢀPd‐NiꢀcatalystꢀwasꢀaꢀresultꢀofꢀstrongꢀchemisorptionꢀbetweenꢀPd
3
Ni ꢀ(1ꢀ1ꢀ1)ꢀ
1
andꢀtheꢀcarbonylꢀgroupꢀofꢀeAQ.ꢀ
©ꢀ2018,ꢀDalianꢀInstituteꢀofꢀChemicalꢀPhysics,ꢀChineseꢀAcademyꢀofꢀSciences.
PublishedꢀbyꢀElsevierꢀB.V.ꢀAllꢀrightsꢀreserved.
1.ꢀ ꢀ Introductionꢀ
noneꢀprocessꢀ[9],ꢀareꢀproduced.ꢀ ꢀ
Currently,ꢀtheꢀmostꢀwidelyꢀusedꢀcatalystsꢀinꢀtheꢀanthroqui‐
noneꢀprocessꢀareꢀPd‐basedꢀpelletꢀcatalysts.ꢀTheseꢀshowꢀbetterꢀ
activitiesꢀwhenꢀcomparedꢀwithꢀotherꢀmetal‐basedꢀorꢀmetal‐freeꢀ
pelletꢀcatalystsꢀinꢀtheꢀhydrogenationꢀofꢀeAQ.ꢀInterestingly,ꢀitꢀhasꢀ
beenꢀshownꢀthatꢀmassꢀtransferꢀisꢀaꢀkeyꢀcontrollingꢀfactorꢀinꢀtheꢀ
catalyticꢀhydrogenationꢀofꢀeAQꢀ[10–15].ꢀInꢀcontrastꢀtoꢀtheꢀtra‐
ditionalꢀPd‐basedꢀpelletꢀcatalysts,ꢀmonolithicꢀcatalystsꢀcanꢀde‐
creaseꢀtheꢀinternalꢀandꢀexternalꢀdiffusionꢀresistances,ꢀandꢀthusꢀ
improveꢀtheꢀoverallꢀreactionꢀrate.ꢀFurthermore,ꢀsupportedꢀPdꢀ
monolithicꢀ catalystsꢀ haveꢀ beenꢀ investigatedꢀ byꢀ severalꢀ re‐
searchersꢀforꢀtheꢀcatalyticꢀhydrogenationꢀofꢀeAQ.ꢀForꢀexample,ꢀ
Hydrogenꢀ peroxideꢀ (H
rawꢀ materialꢀ andꢀ anꢀ environment‐friendlyꢀ oxidantꢀ usedꢀ forꢀ
manyꢀindustrialꢀprocessesꢀ[1–8].ꢀAtꢀpresent,ꢀmoreꢀthanꢀ98%ꢀofꢀ
2
O )ꢀisꢀanꢀimportantꢀgreenꢀchemicalꢀ
2
H
2
2
2
O
2
ꢀ isꢀ producedꢀ byꢀ theꢀ anthraquinoneꢀ process,ꢀ inꢀ whichꢀ
‐ethyl‐9,10‐anthraquinoneꢀ (eAQ)ꢀ isꢀ firstꢀ hydrogenatedꢀ toꢀ
‐ethyl‐9,10‐anthrahydroquinoneꢀ (eAQH )ꢀ andꢀ subsequentꢀ
ꢀproducesꢀeAQꢀandꢀH .ꢀTheꢀcatalyticꢀhy‐
drogenationꢀofꢀeAQꢀisꢀtheꢀkeyꢀstep,ꢀinꢀwhichꢀtheꢀdesiredꢀprod‐
uctsꢀ (eAQH ꢀ andꢀ H eAQH )ꢀ areꢀ producedꢀ (Schemeꢀ 1).ꢀ Mean‐
2
oxidationꢀofꢀeAQH
2
O
2 2
2
4
2
while,ꢀ variousꢀ otherꢀ sideꢀ reactionsꢀ occurꢀ inꢀ thisꢀ processꢀ andꢀ
severalꢀ degradationꢀ productsꢀ (e.g.,ꢀ 2‐ethylanthroneꢀ (eAN),ꢀ
theꢀ Pd/Al
Pd/Al ꢀpelletꢀcatalystꢀwereꢀusedꢀforꢀtheꢀcatalyticꢀhydrogena‐
tionꢀ ofꢀ eAQ.ꢀ Itꢀ wasꢀ foundꢀ thatꢀ theꢀ monolithicꢀ catalystꢀ gaveꢀ aꢀ
higherꢀspaceꢀtimeꢀyieldꢀ(STY)ꢀofꢀH ꢀcomparedꢀtoꢀtheꢀpelletꢀ
2 3
O /CORꢀ (cordierite)ꢀ monolithicꢀ catalystꢀ andꢀ
2
‐ethyloxoanthroneꢀ (OXO),ꢀ 2‐ethylꢀ tetrahydroanthroneꢀ
eAN),ꢀ andꢀ octahydroanthrahydroquinoneꢀ (H eAQH )),ꢀ
whichꢀ cannotꢀ beꢀ oxidizedꢀ toꢀ produceꢀ H ꢀ inꢀ theꢀ anthraqui‐
O
2 3
(H
4
8
2
O
2 2
O
2 2
*
ꢀCorrespondingꢀauthor.ꢀTel:ꢀ+86‐10‐64433695;ꢀE‐mail:ꢀleizhg@mail.buct.edu.cnꢀ
ThisꢀworkꢀwasꢀsupportedꢀbyꢀtheꢀNationalꢀNaturalꢀScienceꢀFoundationꢀofꢀChinaꢀ(21476009,ꢀ21406007,ꢀU1462104).ꢀ ꢀ
DOI:ꢀ10.1016/S1872‐2067(18)63035‐9ꢀ|ꢀhttp://www.sciencedirect.com/science/journal/18722067ꢀ|ꢀChin.ꢀJ.ꢀCatal.,ꢀVol.ꢀ39,ꢀNo.ꢀ6,ꢀJuneꢀ2018ꢀ ꢀ ꢀ

ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
1071ꢀ
Pd‐Coꢀ bimetallicꢀ catalystꢀ withꢀ theꢀ massꢀ ratioꢀ ofꢀ Pd/Coꢀ =ꢀ 4ꢀ
demonstratedꢀtheꢀhighestꢀactivityꢀandꢀselectivityꢀ[22].ꢀHowev‐
er,ꢀPd‐basedꢀbimetallicꢀsupportedꢀmonolithicꢀcatalystsꢀhaveꢀnotꢀ
beenꢀstudiedꢀasꢀyetꢀforꢀtheꢀcatalyticꢀhydrogenationꢀofꢀeAQ.ꢀ ꢀ
Therefore,ꢀ inꢀ theꢀ presentꢀ work,ꢀ weꢀ attemptedꢀ toꢀ improveꢀ
theꢀcatalyticꢀperformanceꢀinꢀtheꢀhydrogenationꢀofꢀeAQꢀbyꢀadd‐
ingꢀNi,ꢀFe,ꢀMn,ꢀorꢀCuꢀintoꢀtheꢀPd/SiO /CORꢀmonolithicꢀcatalyst.ꢀ
2
AꢀseriesꢀofꢀPd,ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCuꢀmonometallic,ꢀandꢀPd‐Mꢀ(Mꢀ=ꢀ
Ni,ꢀ Fe,ꢀ Mn,ꢀ Cu)ꢀ bimetallicꢀ catalystsꢀ wereꢀ preparedꢀ viaꢀ theꢀ
co‐impregnationꢀ methodꢀ forꢀ theꢀ catalyticꢀ hydrogenationꢀ ofꢀ
eAQ.ꢀ Theꢀ monometallicꢀ andꢀ bimetallicꢀ monolithicꢀ catalystsꢀ
wereꢀ characterizedꢀ byꢀ scanningꢀ electronꢀ microscopyꢀ (SEM)ꢀ
withꢀenergyꢀdispersiveꢀspectroscopyꢀ(EDS),ꢀtransmissionꢀelec‐
tronꢀ microscopyꢀ (TEM),ꢀ H
tionꢀ (H ‐TPR),ꢀ X‐rayꢀ photoelectronꢀ spectroscopyꢀ (XPS),ꢀ andꢀ
‐O ꢀ titrations.ꢀ Finally,ꢀ DFTꢀ calculationsꢀ wereꢀ performedꢀ toꢀ
2
‐temperatureꢀ programmedꢀ reduc‐
2
H
2
2
gainꢀaꢀdeepꢀmechanisticꢀunderstandingꢀofꢀtheꢀhydrogenationꢀofꢀ
eAQꢀ overꢀ differentꢀ monometallicꢀ andꢀ bimetallicꢀ monolithicꢀ
catalysts.ꢀ ꢀ
2.ꢀ ꢀ Experimentalꢀ ꢀ
Schemeꢀ1.ꢀAnthraquinoneꢀhydrogenation/oxidationꢀprocessꢀforꢀmanu‐
2.1.ꢀ ꢀ Materialsꢀ
facturingꢀH
2
O ꢀ[10].ꢀ
2
Cordieriteꢀ wasꢀ purchasedꢀ fromꢀ Nanningꢀ Yilaiteꢀ Environ‐
catalyst,ꢀwhichꢀwasꢀmainlyꢀattributedꢀtoꢀtheꢀshorterꢀdiffusionꢀ
distanceꢀinꢀtheꢀcaseꢀofꢀtheꢀformerꢀleadingꢀtoꢀanꢀimprovedꢀdiffu‐
mentalꢀ Protectionꢀ Technologyꢀ Co.ꢀ FeCl
fromꢀ Tianjinꢀ Damaoꢀ Reagentꢀ Factory.ꢀ PdCl
Mn(CH COO) ·4H Oꢀ wereꢀ sourcedꢀ fromꢀ Tianjinꢀ Guangfuꢀ Fineꢀ
Chemicalꢀ Researchꢀ Institute.ꢀ C CuSO ·H Oꢀ wasꢀ obtainedꢀ
3
·6H
2
Oꢀ wasꢀ procuredꢀ
2
,ꢀ NiCl ·6H O,ꢀ andꢀ
2
2
2 2 3
sionꢀprocessꢀ[16].ꢀTheꢀPd/SiO /CORꢀ(PSC)ꢀandꢀPd/Al O /CORꢀ
3
2
2
(PAC)ꢀmonolithicꢀcatalystsꢀwereꢀalsoꢀsubjectedꢀtoꢀtheꢀ1000ꢀhꢀ
H
4 6
4
2
longꢀstabilityꢀtestsꢀinꢀtheꢀcatalyticꢀhydrogenationꢀofꢀeAQ.ꢀTheꢀ
resultsꢀshowedꢀthatꢀtheꢀdurabilityꢀofꢀPSCꢀwasꢀsuperiorꢀtoꢀthatꢀ
ofꢀPAC,ꢀwhichꢀwasꢀattributedꢀtoꢀtheꢀregularꢀstructureꢀandꢀweakꢀ
acidityꢀofꢀtheꢀformerꢀmonolithicꢀcatalystꢀ[17].ꢀ ꢀ
fromꢀ Shanghaiꢀ Aladdinꢀ Chemistryꢀ Co.ꢀ andꢀ SiO2ꢀ solꢀ wasꢀ pur‐
chasedꢀfromꢀSigmaꢀAldrichꢀ(Shanghai)ꢀTradingꢀCo.,ꢀLtd.ꢀ ꢀ
2.2.ꢀ ꢀ Catalystꢀpreparationꢀ ꢀ
Furthermore,ꢀ itꢀ wasꢀ evidentꢀ thatꢀ Pd‐basedꢀ bimetallicꢀ sup‐
portedꢀpelletꢀcatalystsꢀexhibitꢀsuperiorꢀperformanceꢀinꢀvariousꢀ
Theꢀ monolithicꢀ catalystsꢀ wereꢀ preparedꢀ accordingꢀ toꢀ theꢀ
followingꢀthreeꢀprocedures.ꢀ ꢀ
hydrogenationꢀ reactions.ꢀ Forꢀ example,ꢀ Pd‐Ni/SiO
excellentꢀ activityꢀ inꢀ theꢀ catalyticꢀ hydrogenationꢀ ofꢀ 5‐
droxymethyl‐2‐furaldehydeꢀ (HMF)ꢀ andꢀ gaveꢀ aꢀ higherꢀ yieldꢀ ofꢀ
,5‐bis(hydroxymethyl)tetrahydrofuranꢀ asꢀ comparedꢀ toꢀ theꢀ
commercialꢀPd/CꢀcatalystꢀandꢀRaneyꢀNiꢀunderꢀtheꢀsameꢀoper‐
atingꢀconditionsꢀ[18].ꢀSimilarly,ꢀPd‐Fe/SiO ꢀandꢀPd/SiO ꢀcata‐
2
ꢀ showedꢀ
ꢀ
hy‐
(i)ꢀ Pretreatmentꢀ ofꢀ monolithicꢀ honeycombꢀ cordierite:ꢀ Cy‐
lindricalꢀcordieritesꢀ(ꢀ101.6ꢀ×ꢀ127ꢀmm,ꢀandꢀ400ꢀchannelsꢀperꢀ
squareꢀinchꢀ(cpsi))ꢀwereꢀcutꢀintoꢀpiecesꢀofꢀmonolithicꢀsupportsꢀ
(ꢀ20ꢀ×ꢀ10ꢀmm)ꢀandꢀthenꢀtreatedꢀwithꢀ15ꢀwt%ꢀnitricꢀacidꢀatꢀ80ꢀ
°Cꢀforꢀ4ꢀh.ꢀSubsequently,ꢀtheꢀpretreatedꢀcordieriteꢀmonolithicꢀ
samplesꢀ wereꢀ washedꢀ usingꢀ deionizedꢀ waterꢀ untilꢀ theyꢀ wereꢀ
neutral,ꢀdriedꢀatꢀ100ꢀ°Cꢀforꢀ4ꢀh,ꢀandꢀcalcinedꢀinꢀaꢀmuffleꢀfurnaceꢀ
atꢀ550ꢀ°Cꢀforꢀ4ꢀh.ꢀ ꢀ
2
2
2
lystsꢀ wereꢀ testedꢀ inꢀ theꢀ liquid‐phaseꢀ hydrogenationꢀ ofꢀ phe‐
nylacetyleneꢀandꢀitꢀwasꢀfoundꢀthatꢀtheꢀformerꢀbimetallicꢀcata‐
lystꢀ exhibitedꢀ betterꢀ catalyticꢀ activityꢀ andꢀ selectivityꢀ thanꢀ theꢀ
latterꢀ monometallicꢀ catalystꢀ [19].ꢀ Inꢀ theꢀ selectiveꢀ hydrogena‐
tionꢀofꢀmixedꢀC4ꢀcompoundsꢀwithꢀaꢀhighꢀcontentꢀofꢀvinylꢀacety‐
2
(ii)ꢀ Preparationꢀ ofꢀ SiO /cordierite.ꢀ Theꢀ pretreatedꢀ mono‐
2
lithicꢀsupportsꢀwereꢀimmersedꢀintoꢀtheꢀSiO ꢀsolꢀsolutionꢀforꢀ5ꢀ
lene,ꢀPd‐Mn/Al
showedꢀ increasedꢀ activity,ꢀ selectivity,ꢀ andꢀ yieldꢀ ofꢀ
,3‐butadieneꢀasꢀcomparedꢀtoꢀtheꢀPd‐Ni/Al ꢀandꢀPd‐Ag/Al
catalystsꢀ[20].ꢀAꢀseriesꢀofꢀPd,ꢀCu,ꢀandꢀPd‐Cuꢀbimetallicꢀcatalystsꢀ
withꢀ differentꢀ metalꢀ loadingsꢀ andꢀ supportsꢀ wereꢀ investigatedꢀ
forꢀ theꢀ aqueousꢀ phaseꢀ hydrogenationꢀ ofꢀ furfuralꢀ toꢀ furfurylꢀ
2
O
3
ꢀcatalystꢀwithꢀtheꢀmassꢀratioꢀofꢀPd:Mnꢀ=ꢀ1:2,ꢀ
minꢀandꢀtheꢀcoatedꢀsupportꢀwasꢀthenꢀdriedꢀatꢀ100ꢀ°C.ꢀThisꢀpro‐
cedureꢀ wasꢀ repeatedꢀ severalꢀ timesꢀ untilꢀ theꢀ desiredꢀ loadingꢀ
wasꢀobtained.ꢀTheꢀcoatedꢀsupportsꢀwereꢀsubsequentlyꢀcalcinedꢀ
inꢀaꢀmuffleꢀfurnaceꢀatꢀ550ꢀ°Cꢀforꢀ4ꢀh.ꢀ ꢀ
1
O
2 3
O
2 3ꢀ
(iii)ꢀ Formationꢀ ofꢀ Pd‐M/SiO
)/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu).ꢀForꢀtheꢀpreparationꢀ
ofꢀbimetallicꢀmonolithicꢀcatalysts,ꢀPdCl ꢀpowderꢀwasꢀdissolvedꢀ
inꢀHClꢀsolution.ꢀTheꢀsecondꢀmetalꢀprecursorsꢀwereꢀthenꢀaddedꢀ
toꢀtheꢀPdCl ꢀaqueousꢀsolution.ꢀThisꢀmixedꢀsolutionꢀwasꢀstirredꢀ
2
/CORꢀ precursorsꢀ andꢀ PdOꢀ
(M O
x y
2
alcohol.ꢀ Itꢀ wasꢀ foundꢀ thatꢀ 5%ꢀ Pd‐5%ꢀ Cu/MgO(Mg(OH)
2
)ꢀ asꢀ aꢀ
2
catalystꢀledꢀtoꢀcompleteꢀconversionꢀofꢀfurfuralꢀwithꢀhighꢀselec‐
tivityꢀ(>98%)ꢀtowardꢀfurfurylꢀalcoholꢀafterꢀ80ꢀminꢀ[21].ꢀCata‐
lyticꢀ hydrogenationꢀ ofꢀ eAQꢀ wasꢀ performedꢀ withꢀ (Co,ꢀ Ag,ꢀ
2
usingꢀ anꢀ agitatorꢀ forꢀ 1ꢀ h.ꢀ Subsequently,ꢀ theꢀ coatedꢀ supportsꢀ
wereꢀplacedꢀintoꢀtheꢀaboveꢀsolutionꢀandꢀheatedꢀatꢀ100ꢀ°Cꢀforꢀ18ꢀ
Cu)‐dopedꢀPd/Al
2
O ꢀpelletꢀcatalystsꢀandꢀitꢀwasꢀfoundꢀthatꢀtheꢀ
3

1072ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
‐
3
h.ꢀAfterꢀimpregnation,ꢀtheꢀsupportsꢀwereꢀdriedꢀatꢀ100ꢀ°Cꢀforꢀ4ꢀ
h,ꢀfollowedꢀbyꢀcalcinationꢀinꢀaꢀmuffleꢀfurnaceꢀatꢀ550ꢀ°Cꢀforꢀ4ꢀh.ꢀ
2V M10
ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
ꢀ
ꢀ
ꢀ
H
DPd ꢀ(%)
100
ꢀ (1)
322.4WP
‐
3
2

3
VH

N
 
22.4WP
ꢀ usedꢀ forꢀ theꢀ titrationꢀ withꢀ O
10
Forꢀtheꢀpreparationꢀofꢀmonometallicꢀmonolithicꢀcatalysts,ꢀPdCl
2
ꢀ
2
ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
(2)
S ꢀ(m /g) 
100
Pd
powderꢀ wasꢀ firstꢀ dissolvedꢀ inꢀ theꢀ HClꢀ solution.ꢀ Theꢀ coatedꢀ
supportsꢀwereꢀthenꢀplacedꢀintoꢀtheꢀaboveꢀsolutionꢀandꢀheatedꢀ
atꢀ100ꢀ°Cꢀforꢀ18ꢀh.ꢀAfterꢀimpregnation,ꢀtheꢀsupportsꢀwereꢀdriedꢀ
atꢀ100ꢀ°Cꢀforꢀ4ꢀhꢀandꢀlaterꢀcalcinedꢀinꢀaꢀmuffleꢀfurnaceꢀatꢀ550ꢀ°Cꢀ
forꢀ4ꢀh.ꢀ ꢀ
whereꢀ V
(
H
ꢀ isꢀ theꢀ volumeꢀ ofꢀ H
2
2
mL),ꢀMꢀisꢀtheꢀrelativeꢀmolecularꢀmassꢀofꢀPdꢀ(g/mol),ꢀWꢀisꢀtheꢀ
23
catalystꢀmassꢀ(g),ꢀPꢀisꢀtheꢀmassꢀfractionꢀofꢀPd,ꢀNꢀ(6.023ꢀ×ꢀ10 )ꢀ
isꢀAvogadro’sꢀconstant,ꢀandꢀσꢀisꢀtheꢀcross‐sectionalꢀareaꢀofꢀaꢀPdꢀ
atomꢀ(8.97ꢀ×ꢀ10−20ꢀm ).ꢀ ꢀ
2
2
Herein,ꢀtheꢀSiO ꢀloadingꢀamountꢀwasꢀcalculatedꢀbasedꢀonꢀtheꢀ
totalꢀcatalystꢀmassꢀconsistingꢀofꢀCORꢀandꢀSiO
theꢀsecondꢀmetal)ꢀloading,ꢀwhichꢀwasꢀanalyzedꢀbyꢀinductivelyꢀ
coupledꢀplasma‐atomicꢀemissionꢀspectrometryꢀ(ICP‐AES),ꢀwasꢀ
2
.ꢀTheꢀmetalꢀ(Pdꢀorꢀ
2.4.ꢀ ꢀ Catalyticꢀactivityꢀtestꢀ ꢀ
Theꢀ catalyticꢀ hydrogenationꢀ ofꢀ eAQꢀ wasꢀ performedꢀ inꢀ aꢀ
stainlessꢀsteelꢀfixed‐bedꢀreactorꢀ(20ꢀmmꢀinꢀinnerꢀdiameterꢀandꢀ
50ꢀmmꢀinꢀheight)ꢀwithꢀaꢀcentralꢀthermocoupleꢀtoꢀmeasureꢀtheꢀ
temperatureꢀofꢀtheꢀreactionꢀzoneꢀatꢀatmosphericꢀpressure.ꢀBe‐
foreꢀtheꢀreaction,ꢀtheꢀmonolithicꢀcatalystsꢀwereꢀreducedꢀunderꢀ
H ꢀgasꢀatꢀaꢀcertainꢀtemperatureꢀforꢀ3ꢀhꢀinꢀtheꢀfixed‐bedꢀreactor.ꢀ
Afterꢀreduction,ꢀtheꢀreactorꢀwasꢀcooledꢀtoꢀ80ꢀ°Cꢀandꢀ45ꢀmLꢀofꢀ
theꢀworkingꢀsolution,ꢀconsistingꢀofꢀaꢀmixtureꢀofꢀindustrialꢀgradeꢀ
C9ꢀaromaticsꢀandꢀtrioctylꢀphosphateꢀinꢀtheꢀvolumeꢀratioꢀofꢀ3:1,ꢀ
alongꢀwithꢀ2.7ꢀgꢀeAQꢀ(60ꢀg/L)ꢀwereꢀpumpedꢀintoꢀtheꢀfixed‐bedꢀ
calculatedꢀbasedꢀonꢀtheꢀmassꢀofꢀSiO
.3.ꢀ ꢀ Catalystꢀcharacterizationꢀ ꢀ
TheꢀSEMꢀmicrographsꢀandꢀtheꢀEDSꢀspectraꢀofꢀtheꢀmonolithicꢀ
catalystsꢀ wereꢀ examinedꢀ onꢀ aꢀ JEOLꢀ JSM‐6701Fꢀ microscopeꢀ
operatingꢀatꢀanꢀacceleratingꢀvoltageꢀofꢀ5.0ꢀkV.ꢀBeforeꢀobserva‐
tion,ꢀtheꢀsamplesꢀwereꢀsputter‐coatedꢀwithꢀaꢀthinꢀgoldꢀfilmꢀbyꢀ
anꢀionꢀsputteringꢀinstrumentꢀtoꢀmakeꢀtheꢀsamplesꢀconductive.ꢀ ꢀ
Theꢀ catalystsꢀ wereꢀ alsoꢀ characterizedꢀ byꢀ TEMꢀ (JEM‐2000ꢀ
EX,ꢀJEOL,ꢀJapan).ꢀTheꢀelectronꢀmicroscopeꢀwasꢀoperatedꢀatꢀanꢀ
acceleratingꢀ voltageꢀ ofꢀ 120ꢀ kV.ꢀ Beforeꢀ analysis,ꢀ theꢀ catalystꢀ
samplesꢀwereꢀgroundꢀintoꢀpowderꢀandꢀdispersedꢀinꢀethanolꢀbyꢀ
ultrasonication.ꢀDropsꢀofꢀtheseꢀsuspensionsꢀwereꢀappliedꢀontoꢀ
aꢀcopperꢀgrid‐supportedꢀtransparentꢀcarbonꢀfoilꢀandꢀdriedꢀinꢀ
air.ꢀ ꢀ
2
.ꢀ ꢀ
4
2
2
reactorꢀatꢀaꢀflowꢀrateꢀofꢀ0.7ꢀmL/min.ꢀMeanwhile,ꢀH
intoꢀ theꢀ fixed‐bedꢀ reactorꢀ byꢀ aꢀ massꢀ flowmeterꢀ (MFC/MFMꢀ
07‐7,ꢀSevenstar,ꢀChina)ꢀatꢀaꢀrateꢀofꢀ10ꢀmL/min.ꢀTheꢀoperatingꢀ
parametersꢀwereꢀchosenꢀaccordingꢀtoꢀtheꢀresultsꢀofꢀtheꢀcatalyt‐
icꢀ activityꢀ testꢀ ofꢀ theꢀ Pd/SiO /CORꢀ monolithicꢀ catalystꢀ asꢀ de‐
2
ꢀgasꢀwasꢀfedꢀ
0
2
scribedꢀinꢀourꢀpreviousꢀworkꢀ[24].ꢀ ꢀ
Afterꢀ hydrogenation,ꢀ theꢀ hydrogenatedꢀ workingꢀ solutionꢀ
wasꢀ oxidizedꢀ withꢀ airꢀ producedꢀ byꢀ anꢀ oil‐freeꢀ airꢀ generatorꢀ
Theꢀreducibilitiesꢀofꢀtheꢀmonolithicꢀcatalystsꢀwereꢀmeasuredꢀ
byꢀH
‐TPRꢀexperimentsꢀwithꢀaꢀThermoꢀElectronꢀTPD/R/Oꢀ1100ꢀ
2
(GC‐readyꢀSPB‐5000ꢀAutomaticꢀAirꢀSource,ꢀChina)ꢀatꢀ25ꢀ°Cꢀandꢀ
seriesꢀinstrumentꢀequippedꢀwithꢀaꢀthermalꢀconductivityꢀdetec‐
torꢀ(TCD).ꢀ ꢀ
1
01ꢀkPaꢀforꢀ30ꢀmin.ꢀTheꢀH ꢀproducedꢀinꢀtheꢀoxidizedꢀworkingꢀ
2 2
O
solutionꢀwasꢀextractedꢀwithꢀdeionizedꢀ waterꢀ severalꢀ timesꢀ toꢀ
obtainꢀitsꢀaqueousꢀsolution.ꢀFinally,ꢀ10ꢀmLꢀofꢀ25.2ꢀwt%ꢀsulfuricꢀ
TheꢀXPSꢀfilesꢀwereꢀobtainedꢀwithꢀanꢀESCALABꢀ250ꢀphotoe‐
lectronꢀ spectrometerꢀ (ThermoFisherꢀ Scientific,ꢀ USA)ꢀ usingꢀ
acidꢀsolutionꢀwasꢀaddedꢀtoꢀtheꢀH
ꢀwasꢀdeterminedꢀbyꢀtitrationꢀwithꢀtheꢀstandardꢀpotassiumꢀ
permanganateꢀsolution.ꢀ ꢀ
Theꢀ catalystꢀ selectivityꢀ (S),ꢀ H O ꢀ yieldꢀ (B),ꢀ andꢀ turnoverꢀ
2
O ꢀsolution,ꢀandꢀtheꢀamountꢀofꢀ
2
monochromatedꢀAlꢀK ꢀexcitationꢀatꢀpassꢀenergiesꢀofꢀ200ꢀeVꢀforꢀ
α
H O
2 2
surveyꢀandꢀ30ꢀeVꢀforꢀhigh‐resolutionꢀscans.ꢀTheꢀbindingꢀenergyꢀ
calibrationꢀ ofꢀ allꢀ spectraꢀ wasꢀ referencedꢀ toꢀ theꢀ adventitiousꢀ
carbonꢀ(Cꢀ1s)ꢀsignalꢀatꢀ284.6ꢀeVꢀtoꢀreduceꢀtheꢀchargingꢀeffectꢀofꢀ
theꢀsamples.ꢀ ꢀ
Theꢀ actualꢀ Pdꢀ contentsꢀ ofꢀ theꢀ monolithicꢀ catalystsꢀ wereꢀ
measuredꢀbyꢀICP‐AESꢀwithꢀaꢀShimadzuꢀCorporationꢀICP‐7500ꢀ
spectrometer.ꢀTheꢀdegreeꢀofꢀPdꢀdispersionꢀandꢀtheꢀPdꢀsurfaceꢀ
2
2
frequencyꢀ(TOF)ꢀwereꢀcalculatedꢀasꢀfollows.ꢀ ꢀ
neAQ  nH4
eAQ
ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
ꢀ
(3)
S(%) 
100
n0(eAQꢀ)
5
CKMnO4 VKMnO4 MH2O2
ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
ꢀ
ꢀ
B
(4)
2
V
2 2
areaꢀofꢀPd/SiO /CORꢀandꢀPd‐M/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀ
n0C ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
ꢀ (5)
Cu)ꢀmonolithicꢀcatalystsꢀwereꢀmeasuredꢀbyꢀH
2
–O ꢀtitrationꢀonꢀaꢀ
2
TOF 
n t
cat
Thermoꢀ Electronꢀ TPD/R/Oꢀ 1100ꢀ seriesꢀ instrumentꢀ equippedꢀ
withꢀaꢀTCD.ꢀTheꢀcatalystꢀ(100ꢀmg)ꢀwasꢀplacedꢀinꢀaꢀquartzꢀtubeꢀ
reactor,ꢀpretreatedꢀwithꢀ50ꢀmL/minꢀofꢀhigh‐purityꢀHeꢀatꢀ450ꢀ°Cꢀ
forꢀ0.5ꢀh,ꢀandꢀthenꢀcooledꢀtoꢀ50ꢀ°C.ꢀNext,ꢀtheꢀcatalystꢀwasꢀre‐
WP
MPd
ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ ꢀ
D
Pd
ꢀ
(6)
ncat

whereꢀ Sꢀ isꢀ theꢀ selectivityꢀ towardꢀ activeꢀ quinonesꢀ (eAQꢀ andꢀ
eAQ),ꢀn ꢀandꢀnꢀareꢀtheꢀmolarꢀconcentrationsꢀofꢀtheꢀcompo‐
nentsꢀ inꢀ theꢀ initialꢀ workingꢀ solutionꢀ andꢀ oxidizedꢀ solutionꢀ
mol/L),ꢀrespectively,ꢀCKMnO4ꢀisꢀtheꢀconcentrationꢀofꢀtheꢀKMnO
solutionꢀ (mol/L),ꢀ andꢀ VKMnO4ꢀ andꢀ Vꢀ representꢀ theꢀ volumesꢀ ofꢀ
theꢀ KMnO ꢀ solutionꢀ andꢀ H ꢀ solutionꢀ (mL),ꢀ respectively.ꢀ Inꢀ
H
4
0
ducedꢀwithꢀ50ꢀmL/minꢀofꢀ5%ꢀH /Arꢀatꢀ120ꢀ°Cꢀforꢀ2ꢀh,ꢀfollowedꢀ
2
byꢀtreatmentꢀwithꢀ50ꢀmL/minꢀofꢀhigh‐purityꢀHeꢀforꢀ2ꢀh.ꢀSubse‐
quently,ꢀ pulsesꢀ ofꢀ oxygenꢀ wereꢀ introducedꢀ untilꢀ theꢀ catalystꢀ
wasꢀfullyꢀsaturated.ꢀTheꢀchemisorbedꢀoxygenꢀwasꢀtitratedꢀwithꢀ
hydrogen,ꢀandꢀthisꢀtitrationꢀwasꢀrepeatedꢀtwiceꢀtoꢀensureꢀtheꢀ
(
4
ꢀ
4
O
2 2
Equationꢀ(5),ꢀCꢀrefersꢀtoꢀtheꢀconversionꢀofꢀeAQꢀatꢀreactionꢀtimeꢀ
tꢀandꢀncatꢀisꢀtheꢀmoleꢀnumberꢀofꢀtheꢀexposedꢀPdꢀatoms,ꢀgivenꢀbyꢀ
Equationꢀ(6),ꢀwhereꢀMPdꢀisꢀtheꢀmolarꢀmassꢀofꢀPd,ꢀWꢀisꢀtheꢀcata‐
lystꢀmassꢀ(g),ꢀPꢀisꢀtheꢀmassꢀfractionꢀofꢀPd,ꢀandꢀDPdꢀisꢀtheꢀdegreeꢀ
accuracyꢀofꢀtheꢀvolumeꢀofꢀH .ꢀTheꢀPdꢀdispersionꢀ(DPd)ꢀandꢀPdꢀ
2
metalꢀ surfaceꢀ areaꢀ (SPd)ꢀ wereꢀ calculatedꢀ usingꢀ theꢀ followingꢀ
equationsꢀ[23]:ꢀ ꢀ

ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
1073ꢀ
ofꢀPdꢀdispersion.ꢀ
whereꢀ EPd(111)/Pd3M1(111)–eAQ,ꢀ EPd(111)/Pd3M1(111),ꢀ andꢀ EeAQꢀ areꢀ theꢀ
totalꢀ energyꢀ forꢀ eAQꢀ adsorbedꢀ onꢀ eitherꢀ theꢀ Pd(1ꢀ 1ꢀ 1)ꢀ orꢀ
2.5.ꢀ ꢀ Computationalꢀmethodꢀ ꢀ
Pd
3
M
1
(1ꢀ 1ꢀ 1)ꢀ surface,ꢀ theꢀ energyꢀ ofꢀ eitherꢀ freeꢀ Pd(1ꢀ 1ꢀ 1)ꢀ orꢀ
(1ꢀ1ꢀ1),ꢀandꢀtheꢀenergyꢀofꢀfreeꢀeAQ,ꢀrespectively.ꢀ
Pd
3
M
1
ꢀ
Theꢀfirst‐principlesꢀcalculationsꢀbasedꢀonꢀdensityꢀfunctionalꢀ
theoryꢀ(DFT)ꢀwithꢀaꢀplaneꢀwaveꢀbasisꢀsetꢀwereꢀcarriedꢀoutꢀtoꢀ
investigateꢀ theꢀ mechanismꢀ ofꢀ theꢀ hydrogenationꢀ ofꢀ eAQꢀ overꢀ
differentꢀ monometallicꢀ andꢀ bimetallicꢀ monolithicꢀ catalystsꢀ
3.ꢀ ꢀ Resultsꢀandꢀdiscussionꢀ
3.1.ꢀ ꢀ SEM,ꢀTEM,ꢀHRTEM,ꢀandꢀEDSꢀresultsꢀ ꢀ
[25,26].ꢀInꢀthisꢀwork,ꢀallꢀDFTꢀcalculationsꢀwereꢀcarriedꢀoutꢀwithꢀ
theꢀ Viennaꢀ abꢀinitioꢀ simulationꢀ packageꢀ (VASP)ꢀ toꢀ obtainꢀ theꢀ
optimizedꢀPd/Pd‐Mꢀsurfacesꢀandꢀtheꢀcorrespondingꢀadsorptionꢀ
energiesꢀ ofꢀ eAQ.ꢀ Theꢀ exchange‐correlationꢀ functionalꢀ wasꢀ de‐
scribedꢀ byꢀ theꢀ GGA‐PBEꢀ functionalꢀ [27]ꢀ andꢀ all‐electronꢀ
plane‐waveꢀbasisꢀsetꢀwasꢀusedꢀwithꢀanꢀenergyꢀcutoffꢀofꢀ400ꢀeV.ꢀ
Inꢀaddition,ꢀ aꢀ projectorꢀ augmentedꢀ waveꢀ (PAW)ꢀ methodꢀ wasꢀ
adoptedꢀtoꢀimproveꢀ theꢀcomputationalꢀ efficiencyꢀ[28].ꢀ AꢀBril‐
louin‐zoneꢀofꢀp(3ꢀ×ꢀ2)ꢀlateralꢀsupercellꢀwasꢀsampledꢀbyꢀ1×1×1ꢀ
k‐pointsꢀ withꢀ theꢀ Monkhorstꢀ Packꢀ scheme.ꢀ Forcesꢀ blowꢀ 0.05ꢀ
eV·Å ꢀandꢀtheꢀenergyꢀdifferencesꢀlessꢀthanꢀ4ꢀ×ꢀ10 ꢀeVꢀwereꢀ
usedꢀasꢀtheꢀcriterionꢀforꢀconvergence.ꢀ ꢀ
Pdꢀandꢀtheꢀsecondꢀmetalꢀ(Ni,ꢀFe,ꢀMn,ꢀorꢀCu)ꢀwereꢀreducedꢀtoꢀ
theirꢀmetallicꢀstateꢀ(zeroꢀvalence)ꢀatꢀ700ꢀ°C.ꢀInꢀtheꢀcalculation,ꢀ
SEMꢀ wasꢀ usedꢀ toꢀ observeꢀ theꢀ microscopicꢀ changesꢀ ofꢀ theꢀ
surfaceꢀofꢀtheꢀmonolithicꢀcatalyst.ꢀTEM,ꢀHRTEM,ꢀandꢀEDSꢀwereꢀ
usedꢀtoꢀinvestigateꢀtheꢀalloyingꢀeffectꢀofꢀtheꢀsecondꢀmetalꢀ(Ni,ꢀ
Fe,ꢀMn,ꢀCu)ꢀonꢀPd.ꢀTheꢀresultsꢀareꢀshownꢀinꢀFigs.ꢀ1ꢀandꢀ2.ꢀTheꢀ
SEMꢀimagesꢀshowꢀthatꢀtheꢀwallsꢀofꢀtheꢀPd‐Ni/SiO /CORꢀmono‐
2
lithicꢀcatalystꢀwereꢀcoveredꢀwithꢀaꢀthinꢀlayerꢀofꢀactiveꢀwashcoatꢀ
[33,34].ꢀTheꢀblackꢀsphericalꢀparticlesꢀobservedꢀinꢀtheꢀTEMꢀim‐
agesꢀwereꢀidentifiedꢀasꢀtheꢀPdꢀandꢀPd‐Mꢀspecies,ꢀindicatingꢀthatꢀ
theꢀ PdOꢀ andꢀ Pd‐Mꢀ oxideꢀ speciesꢀ disappearedꢀ toꢀ someꢀ extentꢀ
−1
−6
afterꢀH
TheꢀHRTEMꢀimagesꢀofꢀtheꢀPd‐M/SiO
lystꢀ showꢀ thatꢀ theꢀ metalꢀ particlesꢀ wereꢀ dispersedꢀ uniformlyꢀ
withoutꢀaggregationꢀafterꢀH ꢀreduction.ꢀInꢀFig.ꢀ2(a),ꢀtheꢀlatticeꢀ
2
ꢀreduction.ꢀ ꢀ
2
/CORꢀmonolithicꢀcata‐
2
Pd
3
M
1
ꢀandꢀpureꢀPdꢀwereꢀusedꢀtoꢀmodelꢀtheꢀPd‐MꢀalloysꢀandꢀPdꢀ
fringesꢀofꢀPdꢀwithꢀaꢀdistanceꢀofꢀ0.227ꢀnmꢀcorrespondedꢀtoꢀtheꢀ
Pd(1ꢀ1ꢀ1)ꢀplane.ꢀMeanwhile,ꢀtheꢀlatticeꢀfringesꢀinꢀFig.ꢀ2(b)–(e)ꢀ
withꢀdistancesꢀofꢀ0.224,ꢀ0.214,ꢀ0.216,ꢀandꢀ0.225ꢀnm,ꢀindicatedꢀ
theꢀ formationꢀ ofꢀ PdNi,ꢀ PdFe,ꢀ PdMn,ꢀ andꢀ PdCuꢀ nanoalloys,ꢀ re‐
spectively.ꢀ Theꢀ particleꢀ sizeꢀ distributionꢀ measurementsꢀ showꢀ
metal.ꢀTheꢀlatticeꢀconstantsꢀofꢀtheꢀoptimizedꢀstructuresꢀwereꢀaꢀ
=
3
3
ꢀbꢀ=ꢀcꢀ=ꢀ3.94ꢀÅꢀforꢀPd,ꢀaꢀ=ꢀbꢀ=ꢀcꢀ=ꢀ3.87ꢀÅꢀforꢀPd
.85ꢀÅꢀforꢀPd Fe ,ꢀaꢀ=ꢀbꢀ=ꢀcꢀ=ꢀ3.86ꢀÅꢀforꢀPd Mn ,ꢀandꢀaꢀ=ꢀbꢀ=ꢀcꢀ=ꢀ
.88ꢀÅꢀforꢀPd Cu ,ꢀwhichꢀareꢀcloseꢀtoꢀtheꢀvaluesꢀreportedꢀinꢀtheꢀ
(1ꢀ1ꢀ1)ꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀ
3 1
Ni ,ꢀaꢀ=ꢀbꢀ=ꢀcꢀ=ꢀ
3
1
3
1
3
1
literatureꢀ[29–32].ꢀTheꢀPd(1ꢀ1ꢀ1)ꢀandꢀPd
M
3 1
2 2
thatꢀtheꢀmeanꢀparticleꢀsizesꢀofꢀPd/SiO /COR,ꢀPd‐Ni/SiO /COR,ꢀ
Mn,ꢀCu)ꢀsurfacesꢀwereꢀmodeledꢀbyꢀaꢀsupercellꢀ(3 2)ꢀconsistingꢀ
Pd‐Fe/SiO /COR,ꢀPd‐Mn/SiO /COR,ꢀandꢀPd‐Cu/SiO
2
2
2
/CORꢀwereꢀ
ofꢀaꢀthree‐layerꢀPdꢀslab.ꢀForꢀstructuralꢀoptimizations,ꢀtheꢀatomsꢀ
inꢀtheꢀbottomꢀtwoꢀlayersꢀwereꢀfixedꢀandꢀtheꢀatomsꢀinꢀtheꢀup‐
permostꢀ layerꢀ alongꢀ withꢀ theꢀ adsorbedꢀ eAQꢀ moleculeꢀ wereꢀ
relaxed.ꢀ Theꢀ twoꢀ successiveꢀ slabsꢀ wereꢀ separatedꢀ byꢀ aꢀ 10ꢀ Åꢀ
vacuumꢀ regionꢀ toꢀ ensureꢀ thatꢀ theꢀ adsorbateꢀ (eAQ)ꢀ andꢀ theꢀ
subsequentꢀslabꢀwouldꢀnotꢀinteract.ꢀTheꢀadsorptionꢀenergyꢀofꢀ
eAQꢀwasꢀdefinedꢀasꢀfollows:ꢀ ꢀ
7.5,ꢀ4.0,ꢀ4.6,ꢀ5.1,ꢀandꢀ6.0ꢀnm,ꢀrespectively,ꢀasꢀcalculatedꢀfromꢀ50ꢀ
metalꢀparticlesꢀselectedꢀfromꢀdifferentꢀregionsꢀ[35,36].ꢀItꢀwasꢀ
2
evidentꢀthatꢀtheꢀmeanꢀparticleꢀsizesꢀofꢀPd‐M/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀ
Fe,ꢀ Mn,ꢀ andꢀ Cu)ꢀ wereꢀ smallerꢀ thanꢀ thatꢀ ofꢀ Pd/SiO
2
/COR.ꢀ Theꢀ
additionꢀ ofꢀ theꢀ secondꢀ metalꢀ reducedꢀ theꢀ aggregationꢀ ofꢀ Pdꢀ
particles,ꢀwhichꢀledꢀtoꢀtheꢀbetterꢀdispersionꢀandꢀsmallerꢀparti‐
cleꢀsizesꢀofꢀPd.ꢀTheꢀinteractionꢀofꢀPdꢀandꢀtheꢀsecondꢀmetalꢀMꢀinꢀ
theꢀ Pd‐Mꢀ alloysꢀ alsoꢀ inhibitedꢀ metalꢀ sinteringꢀ becauseꢀ ofꢀ theꢀ
ꢀ
ꢀ ꢀ ꢀ ꢀ ꢀ
ꢀ
Eads 

EPd(111)/Pd3M1(111)eAQ EPd(111)/Pd3M1(111) EeAQ

ꢀ (7)
(
a)
(b)
(d)
(
c)
(
e)
(f)
(g)
Fig.ꢀ 1.ꢀ SEMꢀ imagesꢀ ofꢀ Pd‐Ni/SiO
Pd‐Mn/SiO /CORꢀ(f),ꢀandꢀPd‐Cu/SiO
2
/CORꢀ catalystꢀ (a)ꢀ andꢀ (b);ꢀ TEMꢀ imagesꢀ ofꢀ Pd/SiO
/CORꢀ(g).ꢀ
2
/CORꢀ (c),ꢀ Pd‐Ni/SiO
2
/CORꢀ (d),ꢀ Pd‐Fe/SiO /CORꢀ (e),ꢀ
2
2
2

1074ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
d = 0.227 nm
d = 0.224 nm
d = 0.214 nm
d = 0.216 nm
d = 0.225 nm
(a)
(b)
(c)
(d)
(e)
3
3
2
2
1
1
5
0
5
0
5
0
5
0
70
60
50
40
30
20
10
70
60
50
40
30
20
10
45
35
30
25
20
15
10
5
(
f)
(g)
(h)
(i)
(j)
dm = 6.0 nm
40
dm = 7.5 nm
dm = 4.0 nm
dm = 4.6 nm
35
dm = 5.1 nm
30
25
20
1
5
0
5
0
1
0
0
0
2
4
6
8
10
12
14
2
4
6
8
10
2
4
6
8
10
2
4
6
8
10
1
2
3
4
5
6
7
8
9
10 11 12 13 14
Diameter (nm)
Diameter (nm)
Diameter (nm)
Diameter (nm)
Diameter (nm)
Fig.ꢀ2.ꢀHRTEMꢀimagesꢀofꢀPd/SiO
2
/CORꢀ(a),ꢀPd‐Ni/SiO
2
/CORꢀ(b),ꢀPd‐Fe/SiO
2
/CORꢀ(c),ꢀPd‐Mn/SiO
/CORꢀ(i),ꢀandꢀPd‐Cu/SiO
2
/CORꢀ(d),ꢀandꢀPd‐Cu/SiO
2
/CORꢀ(e);ꢀandꢀPdꢀparticleꢀ
sizeꢀdistributionsꢀofꢀPd/SiO /CORꢀ(f),ꢀPd‐Ni/SiO
2
2
/CORꢀ(g),ꢀPd‐Fe/SiO /CORꢀ(h),ꢀPd‐Mn/SiO
2
2
2
/CORꢀ(j).ꢀ
(a)
(b)
(4)
(4)
387
440
280
423
210
(
3)
2)
454
(3)
295
298
445
613
(
(2)
412
563-633
416 457
(1)
(1)
100
200
300
400
500
600
700
800
100
200
300
400
500
600
700
800
T (C )
T (C)
(
d)
(c)
(
3)
2)
(
4)
193
2
53
8
5
2
62
504
(
3)
331
(
548
275
(2)
554
3
00 356
423
(1)
(1)
100
200
300
400
500
600
700
800
100
200
300
400
500
600
700
800
T (C)
‐TPRꢀspectraꢀofꢀdifferentꢀmonolithicꢀcatalysts.ꢀ(a)ꢀPd‐Ni/SiO
/COR;ꢀ(4)ꢀ0.8%ꢀPd/SiO /COR.ꢀ(b)ꢀPd‐Fe/SiO /COR.ꢀ(1)ꢀ0.4%ꢀFe/SiO
.8%ꢀ Pd/SiO /COR.ꢀ (c)ꢀ Pd‐Mn/SiO /COR.ꢀ (1)ꢀ 0.4%ꢀ Mn/SiO /COR;ꢀ (2)ꢀ 0.8%ꢀ Pd‐2%ꢀ Mn/SiO
Pd/SiO /COR.ꢀ(d)ꢀPd‐Cu/SiO /COR.ꢀ(1)ꢀ0.8%ꢀPd‐2%ꢀCu/SiO /COR;ꢀ(2)ꢀ0.8%ꢀPd‐0.4%ꢀCu/SiO /COR;ꢀ(3)ꢀ0.8%ꢀPd/SiO
T (C)
Fig.ꢀ3.ꢀH
Ni/SiO
0
2
2
/COR.ꢀ(1)ꢀ0.4%ꢀNi/SiO
/COR;ꢀ(2)ꢀ0.8%ꢀPd‐2%ꢀFe/SiO
/COR;ꢀ (3)ꢀ 0.8%ꢀ Pd‐0.4%ꢀ Mn/SiO
/COR.ꢀ
2
/COR;ꢀ(2)ꢀ0.8%Pd‐2%ꢀNi/SiO
/COR;ꢀ(3)ꢀ0.8%ꢀPd‐0.4%ꢀFe/SiO
/COR;ꢀ (4)ꢀ 0.8%ꢀ
2
/COR;ꢀ(3)ꢀ0.8%ꢀPd‐0.4%ꢀ
2
2
2
2
2
2
/COR;ꢀ(4)ꢀ
2
2
2
2
2
2
2
2
2
2
stabilizingꢀeffectꢀofꢀtheꢀtwoꢀmetalsꢀ[30].ꢀ
Fig.ꢀS1ꢀinꢀSupportingꢀInformation).ꢀ
3.2.ꢀ ꢀ H ‐TPRꢀresultsꢀ ꢀ
Theꢀ EDSꢀ resultsꢀ ofꢀ theꢀ monolithicꢀ catalystsꢀ confirmedꢀ theꢀ
presenceꢀofꢀdifferentꢀmetalsꢀ(Pd,ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu).ꢀNeitherꢀPdꢀ
norꢀ theꢀ secondꢀ metalsꢀ existedꢀ inꢀ regionsꢀ farꢀ awayꢀ fromꢀ theꢀ
investigatedꢀmetalꢀparticles,ꢀindicatingꢀthatꢀtheꢀsecondꢀmetalꢀisꢀ
locatedꢀnearbyꢀtheꢀPdꢀparticles,ꢀformingꢀPd‐Mꢀalloysꢀ[30]ꢀ(seeꢀ
2
2
H ‐TPRꢀmeasurementsꢀwereꢀusedꢀtoꢀanalyzeꢀtheꢀreducibili‐
tiesꢀ ofꢀ theꢀ differentꢀ monometallicꢀ andꢀ bimetallicꢀ monolithicꢀ

ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
1075ꢀ
catalysts,ꢀ andꢀ theꢀ resultsꢀ areꢀ shownꢀ inꢀ Fig.ꢀ 3.ꢀ Inꢀ theꢀ caseꢀ ofꢀ
Pd/SiO /COR,ꢀnoꢀreductionꢀpeakꢀwasꢀ foundꢀbecauseꢀtheꢀPdOꢀ
speciesꢀonꢀtheꢀsurfaceꢀofꢀtheꢀmonolithicꢀcatalystꢀwasꢀreducedꢀ
readilyꢀ andꢀ evenꢀ atꢀ aꢀ lowꢀ temperatureꢀ withꢀ relativeꢀ easeꢀ
3
d
5/2
3d3/2
2
3
35.8 336.4
340.4 341.3
(5)
3
35.4
2
[37,38].ꢀ Forꢀ theꢀ Ni/SiO /CORꢀ monolithicꢀ catalyst,ꢀ thereꢀ wasꢀ
340.6
336.2
3
41.5
oneꢀreductionꢀpeakꢀatꢀ412ꢀ°C,ꢀwhichꢀwasꢀassignedꢀtoꢀtheꢀNiOꢀ
speciesꢀ onꢀ theꢀ catalystꢀ surface.ꢀ Inꢀ contrast,ꢀ thereꢀ wereꢀ threeꢀ
(4)
(3)
335.5
3
40.7
3
36.3
341.4
reductionꢀpeaksꢀatꢀ566–633,ꢀ457,ꢀandꢀ416ꢀ°CꢀforꢀFe/SiO
TheꢀfirstꢀpeakꢀwasꢀassignedꢀtoꢀtheꢀreductionꢀofꢀFeOꢀtoꢀFe,ꢀtheꢀ
secondꢀcorrespondedꢀtoꢀtheꢀreductionꢀofꢀFe ꢀtoꢀFeO,ꢀandꢀtheꢀ
thirdꢀpeakꢀwasꢀattributedꢀtoꢀtheꢀreductionꢀofꢀFe ꢀtoꢀFe .ꢀInꢀ
theꢀ caseꢀ ofꢀ theꢀ Mn/SiO /CORꢀ monolithicꢀ catalyst,ꢀ thereꢀ wereꢀ
fourꢀreductionꢀpeaksꢀatꢀ554,ꢀ423,ꢀ356,ꢀandꢀ300ꢀ°C,ꢀwhichꢀwereꢀ
assignedꢀtoꢀtheꢀreductionꢀofꢀMnOꢀtoꢀMn,ꢀMn ꢀtoꢀMnO,ꢀMn
toꢀMn ,ꢀandꢀMnO ꢀtoꢀMn ,ꢀrespectively.ꢀ
Forꢀ0.8%ꢀPd‐2%ꢀM/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀCu)ꢀbimetallicꢀ
monolithicꢀcatalysts,ꢀtheꢀnumberꢀofꢀreductionꢀpeaksꢀwereꢀtheꢀ
sameꢀasꢀthatꢀofꢀtheꢀM/SiO /CORꢀmonometallicꢀmonolithicꢀcat‐
alysts,ꢀalthoughꢀtheꢀreductionꢀ temperaturesꢀ wereꢀlowerꢀ[39].ꢀ
However,ꢀoneꢀpeakꢀinꢀtheꢀH ‐TPRꢀplotsꢀofꢀtheꢀPd‐M/SiO /CORꢀ
2
/COR.ꢀ
3
35.4
3
40.6
336.2
O
3 4
341.4
(2)
O
2 3
O
3 4
335.5
2
340.7
(1)
O
3 4
2 3
O ꢀ
O
3 4
2
O
2 3
2
332
334
336
338
340
342
344
Binding energy (eV)
2
Fig.ꢀ 4.ꢀ XPSꢀ spectrumꢀ ofꢀ differentꢀ monolithicꢀ catalysts.ꢀ (1)ꢀ 0.8%ꢀ
Pd/SiO /COR;ꢀ (2)ꢀ 0.8%ꢀ Pd‐0.4%ꢀ Ni/SiO /COR;ꢀ (3)ꢀ 0.8%ꢀ Pd‐0.4%ꢀ
/COR;ꢀ (4)ꢀ 0.8%ꢀ Pd‐0.4%ꢀ Mn/SiO /COR;ꢀ (5)ꢀ 0.8%ꢀ Pd‐0.4%ꢀ
/COR.ꢀ
2
2
Fe/SiO
Cu/SiO
2
2
2
2
2
bimetallicꢀmonolithicꢀcatalystsꢀdecreasedꢀremarkably,ꢀindicat‐
ingꢀtheꢀformationꢀofꢀPd‐Mꢀalloysꢀ[40].ꢀComparedꢀtoꢀtheꢀreduc‐
tionꢀpeaksꢀofꢀtheꢀ0.8%ꢀPd‐2%ꢀM/SiO
2
/CORꢀmonolithicꢀcatalyst,ꢀ
bridizationꢀ betweenꢀ Pdꢀ andꢀ theꢀ secondꢀ metalꢀ uponꢀ alloyingꢀ
[49,50].ꢀItꢀcanꢀbeꢀseenꢀthatꢀtheꢀoverallꢀintensityꢀofꢀtheꢀPdꢀpeaksꢀ
decreasedꢀuponꢀadditionꢀofꢀtheꢀsecondꢀmetal,ꢀwhichꢀsuggestedꢀ
aꢀlowerꢀdegreeꢀofꢀexposureꢀofꢀtheꢀPdꢀatomsꢀinꢀtheꢀPd‐Mꢀbime‐
tallicꢀmonolithicꢀcatalysts.ꢀ ꢀ
thoseꢀofꢀtheꢀ0.8%ꢀPd‐0.4%ꢀM/SiO /CORꢀcatalystsꢀwereꢀshiftedꢀ
2
toꢀlowerꢀtemperaturesꢀbecauseꢀofꢀtheꢀlowerꢀcontentꢀofꢀtheꢀsec‐
ondꢀmetalꢀinꢀtheꢀlatterꢀcatalysts.ꢀItꢀisꢀworthꢀnotingꢀthatꢀtheꢀre‐
ductionꢀtemperaturesꢀdecreasedꢀwithꢀtheꢀadditionꢀofꢀtheꢀsec‐
ondꢀmetalꢀandꢀtheꢀincreasingꢀPd/Mꢀmassꢀratio,ꢀimplyingꢀthatꢀPdꢀ
seeminglyꢀpromotedꢀtheꢀreductionꢀofꢀtheꢀsecondꢀmetalꢀoxidesꢀ
inꢀ aꢀ hydrogenationꢀ spilloverꢀ fromꢀ Pdꢀ toꢀ theꢀ secondꢀ metalꢀ Mꢀ
3.4.ꢀ ꢀ H
2 2
‐O ꢀtitrationꢀ ꢀ
[41].ꢀ
2 2
H ‐O ꢀtitrationsꢀwereꢀperformedꢀtoꢀdetermineꢀtheꢀdegreesꢀofꢀ
Itꢀisꢀnotedꢀthatꢀaꢀnegativeꢀpeakꢀwasꢀobservedꢀatꢀ100ꢀ°Cꢀforꢀ
Pdꢀ dispersionꢀ andꢀ specificꢀ surfaceꢀ areasꢀ ofꢀ theꢀ monometallicꢀ
andꢀ bimetallicꢀ monolithicꢀ catalysts,ꢀ whichꢀ wereꢀ affectedꢀ re‐
markablyꢀbyꢀtheꢀadditionꢀofꢀtheꢀsecondꢀmetalꢀ[23].ꢀTheꢀdetailedꢀ
dataꢀareꢀlistedꢀinꢀTableꢀ1.ꢀItꢀcanꢀbeꢀseenꢀthatꢀtheꢀPdꢀdispersionꢀ
severalꢀ monolithicꢀ catalysts,ꢀ whichꢀ correspondedꢀ toꢀ theꢀ de‐
compositionꢀ ofꢀ theꢀ PdHꢀ speciesꢀ thatꢀ wasꢀ formedꢀ duringꢀ theꢀ
initialꢀhydrogenationꢀstageꢀofꢀPdꢀ[42,43].ꢀItꢀhasꢀbeenꢀreportedꢀ
thatꢀ theꢀ formationꢀ ofꢀ alloysꢀ uponꢀ theꢀ additionꢀ ofꢀ theꢀ secondꢀ
metalꢀ (Bi,ꢀ Cu,ꢀ Ag)ꢀ inꢀ bimetallicꢀ catalystsꢀ suppressesꢀ theꢀ for‐
mationꢀ ofꢀ PdHꢀ [41,44–46].ꢀ Forꢀ allꢀ theꢀ bimetallicꢀ monolithicꢀ
catalystsꢀ investigatedꢀ inꢀ thisꢀ work,ꢀ theꢀ PdHꢀ peaksꢀ eitherꢀ de‐
creasedꢀinꢀintensityꢀorꢀevenꢀcompletelyꢀdisappearedꢀwhenꢀtheꢀ
secondꢀ metalꢀ (Mꢀ =ꢀ Ni,ꢀ Fe,ꢀ Mn,ꢀ Cu)ꢀ wasꢀ added,ꢀ whichꢀ furtherꢀ
confirmedꢀtheꢀformationꢀofꢀalloyꢀcatalystsꢀ
andꢀ specificꢀ surfaceꢀ areaꢀ inꢀ theꢀ Pd‐Ni/SiO /CORꢀ wereꢀ higherꢀ
2
thanꢀthoseꢀofꢀotherꢀbimetallicꢀmonolithicꢀcatalysts,ꢀevenꢀthoughꢀ
theꢀ Pdꢀ loadingsꢀ wereꢀ identical.ꢀ Theseꢀ resultsꢀ confirmedꢀ thatꢀ
thereꢀwereꢀstrongꢀinteractionsꢀbetweenꢀPdꢀandꢀNi,ꢀasꢀtheꢀaddi‐
tionꢀofꢀtheꢀsecondꢀmetal,ꢀi.e.,ꢀnickel,ꢀimprovedꢀtheꢀPdꢀdispersionꢀ
andꢀ preventedꢀ theꢀ migrationꢀ andꢀ aggregationꢀ ofꢀ Pdꢀ particles.ꢀ
Thus,ꢀ theꢀ resultsꢀ ofꢀ H
2
‐O ꢀtitrationsꢀwereꢀconsistentꢀwithꢀ theꢀ
2
HRTEM,ꢀH ‐TPR,ꢀandꢀXPSꢀcharacterizations.ꢀ ꢀ
2
3.3.ꢀ ꢀ XPSꢀresultsꢀ ꢀ
3.5.ꢀ ꢀ CatalyticꢀhydrogenationꢀofꢀeAQꢀ ꢀ
Toꢀinvestigateꢀtheꢀsurfaceꢀpropertiesꢀandꢀchemicalꢀstatesꢀofꢀ
Pdꢀ inꢀ Pd‐Mꢀ bimetallicꢀ monolithicꢀ catalysts,ꢀ XPSꢀ analysisꢀ wasꢀ
performed.ꢀ Theꢀ bindingꢀ energiesꢀ ofꢀ Pdꢀ 3d5/2ꢀ andꢀ Pdꢀ 3d3/2ꢀ inꢀ
pureꢀPdꢀ(Fig.ꢀ4)ꢀwereꢀ335.5ꢀandꢀ340.7ꢀeV,ꢀrespectively,ꢀwhichꢀisꢀ
consistentꢀwithꢀpreviousꢀreportsꢀ[47,48].ꢀInꢀtheꢀcaseꢀofꢀPd‐Mꢀ
alloys,ꢀi.e.,ꢀafterꢀdopingꢀPdꢀwithꢀtheꢀsecondꢀmetal,ꢀnewꢀpeaksꢀ
appearedꢀforꢀtheꢀbindingꢀenergiesꢀofꢀPdꢀ3d5/2ꢀandꢀPdꢀ3d3/2ꢀatꢀ
TheꢀhydrogenationꢀofꢀeAQꢀoverꢀmonometallicꢀandꢀbimetallicꢀ
monolithicꢀ catalystsꢀ wasꢀ performedꢀ inꢀ aꢀ stainlessꢀ steelꢀ
fixed‐bedꢀreactor.ꢀInꢀthisꢀexperiment,ꢀPdꢀloadingꢀwasꢀfixedꢀatꢀ
0.8ꢀwt%.ꢀTheꢀeffectꢀofꢀloadingꢀofꢀtheꢀsecondꢀmetalꢀonꢀtheꢀselec‐
tivityꢀandꢀH
gated,ꢀasꢀshownꢀinꢀFig.ꢀ5ꢀandꢀTableꢀ1.ꢀItꢀcanꢀbeꢀseenꢀthatꢀtheꢀ
ꢀyieldꢀachievedꢀbyꢀusingꢀNi/SiO /CORꢀwasꢀlowerꢀthanꢀthatꢀ
obtainedꢀwithꢀPd/SiO /COR.ꢀFurthermore,ꢀtheꢀPd‐Ni/SiO /CORꢀ
bimetallicꢀ monolithicꢀ catalystꢀ demonstratedꢀ higherꢀ selectivityꢀ
andꢀ gaveꢀ betterꢀ H ꢀ yieldsꢀ thanꢀ theꢀ Pd/SiO /CORꢀ andꢀ
Ni/SiO /CORꢀ monometallicꢀ monolithicꢀ andꢀ otherꢀ bimetallicꢀ
2 2
O ꢀyieldꢀinꢀtheꢀhydrogenationꢀofꢀeAQꢀwasꢀinvesti‐
336.3±0.1ꢀandꢀ341.4ꢀ±0.1ꢀeV,ꢀrespectively.ꢀTheꢀincreaseꢀinꢀtheꢀ
H
2
O
2
2
bindingꢀ energyꢀ forꢀ Pdꢀ didꢀ notꢀ exceedꢀ 1.0ꢀ eVꢀ asꢀ comparedꢀ toꢀ
pureꢀPd,ꢀwhichꢀisꢀconsistentꢀwithꢀtheꢀpreviousꢀworkꢀ[22].ꢀTheꢀ
shiftꢀinꢀtheꢀbindingꢀenergiesꢀoriginatesꢀfromꢀtheꢀdecreaseꢀinꢀtheꢀ
densityꢀofꢀstatesꢀatꢀtheꢀFermiꢀlevelꢀbecauseꢀofꢀtheꢀd‐bandꢀhy‐
2
2
O
2 2
2
2

1076ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
Tableꢀ1ꢀ
CatalyticꢀperformanceꢀandꢀtexturalꢀparametersꢀofꢀdifferentꢀmonolithicꢀcatalystsꢀforꢀtheꢀhydrogenationꢀofꢀeAQꢀtoꢀeAQH
2
.ꢀ
Pdꢀefficiencyꢀ DPdꢀaꢀ
S
ꢀ
TOFꢀ
Washcoatꢀ
loadingꢀ(wt%)ꢀ
Pdꢀloading
(wt%)ꢀ
Secondꢀmetal Selectivityꢀ YieldꢀofꢀH
O
2 2
Pd
Sampleꢀ
Washcoatꢀ
‒
2
1
loadingꢀ(wt%)
(%)ꢀ
42.2ꢀ
35.0ꢀ
95.3ꢀ
80.6ꢀ
63.7ꢀ
40.4ꢀ
(g/L)ꢀ
3.82ꢀ
2.84ꢀ
ꢀ 7.5ꢀ
ꢀ 5.5ꢀ
4.35ꢀ
ꢀ 3.1ꢀ
(g/(g·h))ꢀ
793ꢀ
(%)ꢀ (m /g) (s )ꢀ
Pd/SiO
Ni/SiO
Pd‐Ni/SiO
Pd‐Fe/SiO
Pd‐Mn/SiO
2
/CORꢀ
/CORꢀ
/CORꢀ ꢀ ꢀ ꢀ ꢀ
/CORꢀ
/CORꢀ
/CORꢀ
aꢀDPdꢀwasꢀdeterminedꢀbyꢀH
SiO
SiO
SiO
SiO
SiO
SiO
2
2
2
2
2
2
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
30ꢀ
30ꢀ
30ꢀ
30ꢀ
30ꢀ
30ꢀ
0.8ꢀ
0.8ꢀ
0.8ꢀ
0.8ꢀ
0.8ꢀ
0.8ꢀ
—ꢀ
—ꢀ
16ꢀ
—ꢀ
45ꢀ
36ꢀ
32ꢀ
26ꢀ
81ꢀ
—ꢀ
228ꢀ
183ꢀ
162ꢀ
132ꢀ
0.69ꢀ
0.49ꢀ
1.21ꢀ
0.99ꢀ
0.78ꢀ
0.56ꢀ
2
—ꢀ
2
0.4ꢀ
0.4ꢀ
0.4ꢀ
0.4ꢀ
1349ꢀ
1141ꢀ
ꢀ 902ꢀ
ꢀ 643ꢀ
2
2
Pd‐Cu/SiO
2
2
‐O ꢀtitrationꢀanalysis.ꢀ
2
ꢀ
1
00
8
Fe
Mn
Ni
Cu
Pd
(a)
Fe
Mn
Ni
Cu
Pd
(b)
7
6
5
4
3
2
1
8
6
4
2
0
0
0
0
0
0
(A)
(C)
Catalyst
Fig.ꢀ5.ꢀEffectꢀofꢀtheꢀsecondꢀmetalꢀloadingꢀonꢀselectivityꢀ(a)ꢀandꢀH
trationꢀofꢀeAQꢀsolutionꢀ=ꢀ60ꢀg/L,ꢀflowꢀrateꢀofꢀeAQꢀsolutionꢀ=ꢀ0.7ꢀmL/min,ꢀandꢀflowꢀrateꢀofꢀH
M/SiO /COR;ꢀ(C)ꢀ0.8%ꢀPd‐0.4%ꢀM/SiO /COR;ꢀ(D)ꢀ0.8%ꢀPd‐0.8%ꢀM/SiO /COR;ꢀ(E)ꢀ0.8%ꢀPd‐1%ꢀM/SiO
(B)
(D)
(E)
(F)
(A)
(B)
(C)
(D)
(E)
(F)
Catalyst
ꢀyieldꢀ(b).ꢀReactionꢀconditions:ꢀpressureꢀ=ꢀ101ꢀkPa,ꢀtemperatureꢀ=ꢀ80ꢀ°C,ꢀconcen‐
ꢀgasꢀ=ꢀ10ꢀmL/min.ꢀ(A)ꢀPd/SiO /COR;ꢀ(B)ꢀ0.8%ꢀPd‐0.2%ꢀ
/COR;ꢀ(F)ꢀ0.8%ꢀPd‐2%ꢀM/SiO /COR.ꢀ
2 2
O
2
2
2
2
2
2
2
monolithicꢀ catalysts.ꢀ Oneꢀ reasonꢀ forꢀ theseꢀ resultsꢀ isꢀ thatꢀ theꢀ
formationꢀ ofꢀ Pd‐Niꢀ alloysꢀ ledꢀ toꢀ theꢀ betterꢀ catalyticꢀ perfor‐
mance,ꢀasꢀprovenꢀbyꢀXPS,ꢀTEM,ꢀandꢀTPRꢀresults.ꢀTheꢀotherꢀrea‐
sonꢀisꢀtheꢀstrongꢀinteractionꢀbetweenꢀtheꢀcarbonylꢀbondꢀofꢀeAQꢀ
andꢀtheꢀPd‐Niꢀalloys,ꢀasꢀconfirmedꢀbyꢀDFTꢀcalculations.ꢀ
wasꢀattributedꢀtoꢀtheꢀinhibitionꢀadsorptionꢀofꢀeAQꢀonꢀtheꢀPd‐Cuꢀ
nanoalloyꢀsurfaceꢀ[22].ꢀ
3.6.ꢀ ꢀ DFTꢀcalculationsꢀonꢀPd(1ꢀ1ꢀ1)ꢀandꢀPd
3
M (1ꢀ1ꢀ1)ꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀ
1
Mn,ꢀandꢀCu)ꢀsurfacesꢀ ꢀ
SinceꢀH
2
O ꢀyieldꢀisꢀstronglyꢀinfluencedꢀbyꢀtheꢀmetalꢀloading,ꢀ
2
aꢀseriesꢀofꢀcatalystsꢀwithꢀdifferentꢀPd/Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀCu)ꢀ
ratiosꢀandꢀatꢀaꢀfixedꢀPdꢀloadingꢀofꢀ0.8ꢀwt%ꢀwereꢀpreparedꢀandꢀ
theirꢀcatalyticꢀactivitiesꢀinꢀtheꢀhydrogenationꢀofꢀeAQꢀwereꢀin‐
vestigatedꢀ (Fig.ꢀ 5).ꢀ Itꢀ wasꢀ foundꢀ thatꢀ theꢀ bimetallicꢀ
Itꢀ isꢀ generallyꢀ believedꢀ thatꢀ theꢀ hydrogenationꢀ ofꢀ eAQꢀ onꢀ
monometallicꢀ Pdꢀ andꢀ bimetallicꢀ Pd‐Mꢀ pelletꢀ catalystsꢀ followsꢀ
theꢀ Langmuir‐Hinshelwoodꢀ (L‐H)ꢀ mechanism,ꢀ inꢀ whichꢀ H ꢀ isꢀ
2
preferentiallyꢀadsorbedꢀonꢀtheꢀtopꢀofꢀaꢀPdꢀatomꢀandꢀthenꢀdisso‐
ciatedꢀ withꢀ aꢀ lowꢀ overallꢀ activationꢀ barrierꢀ (3.9ꢀ kcal/mol)ꢀ
[22,51].ꢀThisꢀimpliesꢀthatꢀtheꢀadsorptionꢀofꢀeAQꢀonꢀtheꢀsurfacesꢀ
ofꢀPdꢀorꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀmonolithicꢀcatalystsꢀwasꢀ
crucialꢀforꢀtheꢀhydrogenationꢀofꢀeAQ.ꢀ ꢀ
Pd‐Ni/SiO
Pd‐Cu/SiO
achievedꢀtheꢀhighestꢀselectivityꢀandꢀH
ofꢀPd/Mꢀ=ꢀ2.ꢀTheꢀselectivityꢀandꢀH
2
/COR,ꢀ Pd‐Fe/SiO
/CORꢀ monolithicꢀ catalystsꢀ reducedꢀ atꢀ 700ꢀ °Cꢀ
ꢀyieldꢀatꢀtheꢀmassꢀratioꢀ
ꢀyieldꢀincreasedꢀwithꢀtheꢀ
2
/COR,ꢀ Pd‐Mn/SiO
2
/COR,ꢀ andꢀ
2
O
2 2
O
2 2
increasingꢀloadingꢀamountsꢀofꢀtheꢀsecondꢀmetalꢀ(i.e.,ꢀNi,ꢀFe,ꢀMn,ꢀ
andꢀCu)ꢀbelowꢀaꢀfinalꢀamountꢀofꢀ0.4ꢀwt%.ꢀHowever,ꢀtheꢀtrendꢀ
wasꢀreversedꢀatꢀloadingꢀamountsꢀaboveꢀ0.4ꢀwt%.ꢀ ꢀ
Inꢀthisꢀwork,ꢀDFTꢀcalculationsꢀwereꢀconductedꢀtoꢀcompareꢀ
theꢀadsorptionꢀenergeticsꢀforꢀeAQꢀonꢀbothꢀPdꢀandꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀ
Fe,ꢀMn,ꢀandꢀCu)ꢀsurfacesꢀtoꢀexplainꢀtheꢀdifferencesꢀinꢀtheꢀcata‐
lyticꢀactivityꢀofꢀmonometallicꢀPdꢀandꢀbimetallicꢀPd‐Mꢀmonolith‐
Inꢀaddition,ꢀtheꢀTOFsꢀofꢀ0.8%ꢀPd‐0.4%ꢀM/SiO
Fe,ꢀMn)ꢀbimetallicꢀmonolithicꢀcatalystsꢀwereꢀhigherꢀthanꢀthatꢀofꢀ
theꢀ 0.8%ꢀ Pd/SiO /CORꢀ monometallicꢀ monolithicꢀ catalystꢀ (Ta‐
2
/CORꢀ(Mꢀ=ꢀNi,ꢀ
icꢀcatalysts.ꢀThus,ꢀtheꢀfourꢀcrystalꢀstructuresꢀofꢀPd
3
‐Ni
1
,ꢀPd
3
‐Fe ,ꢀ
1
2
Pd ‐Mn ,ꢀandꢀPd ‐Cu ,ꢀalongꢀwithꢀthatꢀofꢀpureꢀPd,ꢀwereꢀusedꢀtoꢀ
3
1
3
1
bleꢀ 1),ꢀ whichꢀ wasꢀ attributedꢀ toꢀ theꢀ improvedꢀ Pdꢀ dispersionꢀ
afterꢀtheꢀadditionꢀofꢀtheꢀsecondꢀmetal.ꢀHowever,ꢀwhileꢀtheꢀPdꢀ
dispersionꢀ andꢀ specificꢀ surfaceꢀ areaꢀ inꢀ theꢀ caseꢀ ofꢀ theꢀ 0.8%ꢀ
modelꢀtheꢀPd‐Mꢀalloyꢀ(Fig.ꢀ6).ꢀ
TheꢀcalculationsꢀforꢀtheꢀadsorptionꢀofꢀeAQꢀonꢀtheꢀPd(1ꢀ1ꢀ1)ꢀ
andꢀ Pd
3
M (1ꢀ 1ꢀ 1)ꢀ surfacesꢀ wereꢀ somewhatꢀ complicatedꢀ be‐
1
Pd‐0.4%ꢀ Cu/SiO
2
/CORꢀcatalystꢀwereꢀhigherꢀthanꢀthoseꢀofꢀtheꢀ
causeꢀofꢀtheꢀnumerousꢀpossibleꢀconfigurations.ꢀInꢀtermsꢀofꢀpre‐
viousꢀworkꢀ[51], theꢀmostꢀrelevantꢀstructuresꢀofꢀeAQꢀonꢀtheseꢀ
ꢀ
0
.8%ꢀPd/SiO /CORꢀmonometallicꢀmonolithicꢀcatalyst,ꢀtheꢀTOFꢀ
2
ofꢀtheꢀformerꢀwasꢀlowerꢀthanꢀthatꢀofꢀtheꢀlatterꢀmonometallicꢀ
monolithicꢀ catalyst.ꢀ Theꢀ lowerꢀ TOFꢀ ofꢀ theꢀ bimetallicꢀ catalystꢀ
surfacesꢀwereꢀchosenꢀforꢀcalculations.ꢀTheꢀpossibleꢀstructuresꢀ
areꢀ shownꢀ inꢀ Fig.ꢀ S2ꢀ (seeꢀ Supportingꢀ Information),ꢀ andꢀ theꢀ

ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
1077ꢀ
Pd
Pd Ni
electronꢀback‐donation),ꢀasꢀreportedꢀinꢀliteratureꢀ[38,51].ꢀTheꢀ
adsorptionꢀenergyꢀofꢀeAQꢀonꢀPd(1ꢀ1ꢀ1)ꢀwasꢀcalculatedꢀasꢀ61.4ꢀ
kcal/mol.ꢀTheꢀaromaticꢀringꢀofꢀeAQꢀplayedꢀanꢀimportantꢀroleꢀinꢀ
promotingꢀaꢀflatꢀadsorptionꢀonꢀtheꢀPd(1ꢀ1ꢀ1)ꢀsurface.ꢀThisꢀwasꢀ
similarꢀtoꢀtheꢀadsorptionꢀofꢀ furanꢀonꢀPd(1ꢀ1ꢀ 1)ꢀ describedꢀbyꢀ
Bradleyꢀetꢀal.ꢀ[52]ꢀonꢀtheꢀbasisꢀofꢀDFTꢀcalculations.ꢀThisꢀresultꢀ
showedꢀthatꢀtheꢀpreferredꢀadsorptionꢀstructureꢀwasꢀoneꢀwithꢀ
theꢀaromaticꢀringꢀessentiallyꢀparallelꢀtoꢀtheꢀmetalꢀsurface.ꢀ ꢀ
Theꢀ additionꢀ ofꢀ theꢀ secondꢀ metalꢀ (i.e.,ꢀ Ni,ꢀ Fe,ꢀ Mn,ꢀ andꢀ Cu)ꢀ
weakenedꢀtheꢀadsorptionꢀofꢀtheꢀbenzeneꢀrings,ꢀwhichꢀwasꢀex‐
plainedꢀbyꢀtheꢀstrongꢀaffinityꢀofꢀtheꢀaromaticꢀringsꢀtowardꢀPdꢀ
andꢀnotꢀforꢀtheꢀsecondꢀmetalsꢀ[40].ꢀAsꢀaꢀresult,ꢀtheꢀadsorptionꢀ
3
1
3 1
Pd Fe
Pd
3
Mn
1
3 1
Pd Cu
energiesꢀofꢀeAQꢀonꢀPd
andꢀ Pd Cu (1ꢀ 1ꢀ 1)ꢀ decreasedꢀ toꢀ 59.8,ꢀ 50.7,ꢀ 44.5,ꢀ andꢀ 35.5ꢀ
kcal/mol,ꢀ respectively.ꢀ However,ꢀ theꢀ adsorptionꢀ energiesꢀ ofꢀ
eAQꢀonꢀPd ꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀandꢀMn)ꢀdidꢀnotꢀchangeꢀasꢀmuchꢀasꢀ
3 1 3 1 3 1
Ni (1ꢀ1ꢀ1),ꢀPd Fe (1ꢀ1ꢀ1),ꢀPd Mn (1ꢀ1ꢀ1),ꢀ
3
1
M
3 1
thatꢀonꢀPd(1ꢀ1ꢀ1),ꢀbecauseꢀofꢀtheꢀstrongꢀbondingꢀofꢀtheꢀcarbonylꢀ
groupꢀwithꢀtheꢀsecondꢀmetalꢀatoms,ꢀasꢀshownꢀinꢀFig.ꢀ7.ꢀAtꢀtheꢀ
sameꢀ time,ꢀ theꢀ strongꢀ bindingꢀ ofꢀ theꢀ carbonylꢀ groupꢀ onꢀ theꢀ
Pd‐Mꢀ(Mꢀ=ꢀNi,ꢀFeꢀandꢀMn)ꢀalloyꢀsurfaceꢀfacilitatedꢀtheꢀhydro‐
Pd
Ni
Fe
Mn
Cu
Fig.ꢀ6.ꢀBulkꢀmonometallicꢀPdꢀandꢀbimetallicꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀand
genationꢀ ofꢀ eAQꢀ toꢀ eAQH ꢀ [30].ꢀ Thisꢀ wasꢀ inꢀ goodꢀ agreementꢀ
2
Cu)ꢀalloyꢀstructures.ꢀ
withꢀtheꢀexperimentalꢀresultsꢀofꢀenhancedꢀcatalyticꢀactivityꢀofꢀ
theꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀandꢀMn)ꢀalloys.ꢀHowever,ꢀtheꢀadsorptionꢀ
mostꢀstableꢀstructuresꢀareꢀshownꢀinꢀFig.ꢀ7.ꢀTheꢀadsorptionꢀen‐
ergiesꢀ(Eads)ꢀandꢀbondꢀlengthsꢀareꢀlistedꢀinꢀTableꢀ2ꢀ(forꢀmoreꢀ
details,ꢀseeꢀTableꢀS1ꢀinꢀtheꢀSupportingꢀInformation).ꢀ ꢀ
energyꢀofꢀeAQꢀ onꢀPd
posedꢀtoꢀthatꢀonꢀPd(1ꢀ1ꢀ1),ꢀwhichꢀwasꢀattributedꢀtoꢀtheꢀinhibi‐
tionꢀofꢀadsorptionꢀofꢀeAQꢀonꢀPd Cu (1ꢀ1ꢀ1)ꢀ[22].ꢀThisꢀprovedꢀtoꢀ
3
Cu
1
(1ꢀ1ꢀ 1)ꢀchangedꢀ aꢀlittleꢀ moreꢀ asꢀop‐
3
1
TheꢀadsorptionꢀofꢀeAQꢀonꢀtheꢀoptimizedꢀPd(1ꢀ1ꢀ1)ꢀsurfaceꢀ
wasꢀsuchꢀthatꢀtheꢀtwoꢀbenzeneꢀringsꢀandꢀcarbonylꢀoxygenꢀat‐
omsꢀofꢀeAQꢀwereꢀlocatedꢀatꢀtheꢀbridgeꢀsites,ꢀwhileꢀtheꢀcarbonylꢀ
bondꢀwasꢀelongatedꢀ(0.048ꢀÅ),ꢀwhichꢀwasꢀlikelyꢀdueꢀtoꢀtheꢀin‐
teractionꢀofꢀtheꢀcarbonylꢀπꢀbondꢀwithꢀtheꢀPd(1ꢀ1ꢀ1)ꢀsurfaceꢀ(i.e.,ꢀ
beꢀunfavorableꢀforꢀtheꢀhydrogenationꢀofꢀC=Oꢀgroupsꢀandꢀledꢀtoꢀ
lowerꢀcatalyticꢀactivity.ꢀ ꢀ
Theꢀbenzeneꢀringsꢀgreatlyꢀenhancedꢀthisꢀinteractionꢀonꢀpureꢀ
Pd.ꢀHowever,ꢀwhenꢀtheꢀsecondꢀmetalsꢀwereꢀadded,ꢀtheꢀstabilityꢀ
3 1
ofꢀeAQꢀonꢀPd M (1ꢀ1ꢀ1)ꢀdecreasedꢀdramatically.ꢀComparedꢀtoꢀ
1.237
(d)
(a)
2.459
1.291
2.271
2
.243
.331
.007
1
2
(b)
E
ads = 53.3 kcal/mol
2.382
(e)
1
.285
2.536
.275
.362
1
2
.276
.268
.372
2
2
2.267
1
.286
2
1.341
1.948
(c)
E
ads = 61.4 kcal/mol
(f)
E
ads = 44.5 kcal/mol
2.035
2.516
1
.284
1.313
2
.379
2
2
.235
.270
2.325
.288
1
1.286
2.492
2.579
E
ads = 35.5 kcal/mol
E
ads = 59.8 kcal/mol
C
O
H
Pd
Ni
Fe
Mn
Cu
Fig.ꢀ7.ꢀDFTꢀoptimizedꢀstructuresꢀofꢀeAQꢀ(a)ꢀandꢀeAQꢀadsorbedꢀonꢀPd(1ꢀ1ꢀ1)ꢀ(b),ꢀPd
)ꢀ(f).ꢀ
3
Ni
1
(1ꢀ1ꢀ1)ꢀ(c),ꢀPd
3
Fe
1
(1ꢀ1ꢀ1)ꢀ(d),ꢀPd
3
Mn
1
(1ꢀ1ꢀ1)ꢀ(e),ꢀandꢀPd
3
Cu (1ꢀ1ꢀ
1
1

1078ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
Tableꢀ2ꢀ
AdsorptionꢀenergiesꢀandꢀoptimizedꢀbondꢀlengthsꢀofꢀeAQꢀonꢀPd(1ꢀ1ꢀ1)ꢀandꢀonꢀPd
3 1
M (1ꢀ1ꢀ1)ꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀslabs.ꢀ
a
b
c
d
Modelꢀ
eAQꢀ
eAQ/Pd(1ꢀ1ꢀ1)ꢀ
eAQ/Pd
eAQ/Pd
eAQ/Pd
E
adsꢀ(kcal/mol)ꢀ
C−Oꢀ(Å)ꢀ
1.237ꢀ
Pd−Cꢀ ꢀ(Å)ꢀ
—ꢀ
Pd−Oꢀ ꢀ(Å)ꢀ
—ꢀ
M−Cꢀ ꢀ(Å)ꢀ
M−Oꢀ ꢀ(Å)ꢀ
—ꢀ
61.4ꢀ
59.8ꢀ
53.3ꢀ
44.5ꢀ
35.5ꢀ
—ꢀ
—ꢀ
—ꢀ
—ꢀ
—ꢀ
—ꢀ
—ꢀ
2.035ꢀ
2.007ꢀ
1.948ꢀ
—ꢀ
1.285/1.286ꢀ
1.313/1.286ꢀ
1.291/1.331ꢀ
1.275/1.341ꢀ
1.284/1.288ꢀ
2.276/2.268ꢀ
2.235/2.270ꢀ
2.271/2.243ꢀ
2.362/2.267ꢀ
2.325ꢀ
2.382/2.372ꢀ
2.579ꢀ
2.459ꢀ
2.536ꢀ
2.516/2.492ꢀ
3
3
3
3
Ni
Fe
Mn
Cu
1
(1ꢀ1ꢀ1)ꢀ
1
(1ꢀ1ꢀ1)ꢀ
1
(1ꢀ1ꢀ1)ꢀ
eAQ/Pd
1
(1ꢀ1ꢀ1)ꢀ
2.379ꢀ
aꢀ
Pd−C:ꢀtheꢀdistanceꢀbetweenꢀCꢀandꢀtheꢀnearestꢀPdꢀatom,ꢀasꢀshownꢀinꢀFig.ꢀ7.ꢀ ꢀ
Pd−O:ꢀtheꢀdistanceꢀbetweenꢀOꢀandꢀtheꢀnearestꢀPdꢀatom,ꢀasꢀshownꢀinꢀFig.ꢀ7.ꢀ ꢀ
bꢀ
cꢀ
M−C:ꢀtheꢀdistanceꢀbetweenꢀCꢀandꢀtheꢀnearestꢀsecondꢀmetalꢀatom,ꢀasꢀshownꢀinꢀFig.ꢀ7.ꢀ ꢀ
dꢀ
M−O:ꢀtheꢀdistanceꢀbetweenꢀOꢀandꢀtheꢀnearestꢀsecondꢀmetalꢀatom,ꢀasꢀshownꢀinꢀFig.ꢀ7.ꢀ
ꢀ
eAQꢀonꢀPdꢀ(1ꢀ1ꢀ1),ꢀtheꢀbondsꢀ(C=O,ꢀPd−C/M−C,ꢀandꢀPd−O)ꢀ(Mꢀ=ꢀ
Ni,ꢀ Fe,ꢀ andꢀ Mn)ꢀ wereꢀ elongated,ꢀ whichꢀ indicatedꢀ aꢀ stronglyꢀ
bindingꢀcarbonylꢀgroupꢀandꢀweaklyꢀbindingꢀbenzeneꢀringsꢀ(Ta‐
bleꢀ2ꢀandꢀTableꢀS1).ꢀAtꢀtheꢀsameꢀtime,ꢀtheꢀM–Oꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀandꢀ
Mn)ꢀ bondꢀ lengthsꢀ wereꢀ shortenedꢀ asꢀ comparedꢀ toꢀ Pd–O.ꢀ
Therefore,ꢀtheꢀstrongerꢀadsorptionꢀbetweenꢀtheꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀ
Fe,ꢀandꢀMn)ꢀalloyꢀsurfaceꢀandꢀtheꢀcarbonylꢀgroupꢀledꢀtoꢀhigherꢀ
catalyticꢀ activityꢀ inꢀ comparisonꢀ withꢀ theꢀ monometallicꢀ Pdꢀ
monolithicꢀcatalysts.ꢀ ꢀ
activity.ꢀForꢀexample,ꢀPd‐CuꢀinhibitedꢀtheꢀadsorptionꢀofꢀeAQꢀonꢀ
theꢀPdCuꢀsurfaceꢀandꢀthisꢀprovedꢀtoꢀbeꢀunfavorableꢀforꢀtheꢀhy‐
drogenationꢀofꢀtheꢀC=Oꢀmoiety.ꢀForꢀtheꢀdifferentꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀ
Fe,ꢀ andꢀ Mn)ꢀ alloys,ꢀ aꢀ strongerꢀ adsorptionꢀ betweenꢀ theꢀ Pd‐Mꢀ
alloyꢀsurfaceꢀandꢀtheꢀcarbonylꢀgroupꢀimpliedꢀhigherꢀadsorptionꢀ
energyꢀ forꢀ theꢀ eAQꢀ chemisorptionꢀ onꢀ Pd‐M,ꢀ whichꢀ explainedꢀ
theꢀ higherꢀ catalyticꢀ activity.ꢀ Furthermore,ꢀ theoreticalꢀ insightsꢀ
wereꢀalsoꢀgainedꢀintoꢀtheꢀreactionꢀperformanceꢀofꢀtheꢀmono‐
metallicꢀ Pdꢀ andꢀ Pd‐Mꢀ bimetallicꢀ monolithicꢀ catalystsꢀ forꢀ theꢀ
hydrogenationꢀofꢀeAQ.ꢀ ꢀ
4.ꢀ ꢀ Conclusionsꢀ ꢀ
SupportingꢀInformationꢀ
Inꢀthisꢀwork,ꢀaꢀseriesꢀofꢀPd‐M/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀ
2
Cu)ꢀ bimetallicꢀ andꢀ Pd/SiO
prepared,ꢀ characterized,ꢀ andꢀ evaluatedꢀ forꢀ theꢀ hydrogenationꢀ
ofꢀeAQ.ꢀTheꢀbimetallicꢀ0.8%ꢀPd‐0.4%ꢀNi/SiO /CORꢀmonolithicꢀ
catalystꢀ gaveꢀ theꢀ highestꢀ H ꢀ yieldꢀ (7.5ꢀ g/L)ꢀ andꢀ selectivityꢀ
95.3%)ꢀ amongꢀ allꢀ theꢀ monolithicꢀ catalystsꢀ investigated.ꢀ Theꢀ
2
/CORꢀ monolithicꢀ catalystsꢀ wereꢀ
Supportingꢀ Informationꢀ associatedꢀ withꢀ thisꢀ articleꢀ canꢀ beꢀ
foundꢀinꢀtheꢀonlineꢀversion.ꢀ
2
2
O
2
Referencesꢀ
(
higherꢀactivityꢀofꢀthisꢀcatalystꢀwasꢀattributedꢀtoꢀtheꢀformationꢀ
ofꢀPdNiꢀalloys.ꢀTheꢀresultsꢀofꢀtheꢀdifferentꢀcatalystꢀcharacteriza‐
[1] C.ꢀSamanta,ꢀAppl.ꢀCatal.ꢀA,ꢀ2008,ꢀ350,ꢀ133–149.ꢀ ꢀ
[
[
[
[
[
2] E.ꢀ Santacesaria,ꢀ M.ꢀ D.ꢀSerio,ꢀR.ꢀVelotti,ꢀ U.ꢀ Leone,ꢀInd.ꢀ Eng.ꢀ Chem.ꢀ
Res.,ꢀ1994,ꢀ33,ꢀ277–284.ꢀ ꢀ
3] H.ꢀShang,ꢀH.ꢀJ.ꢀZhou,ꢀZ.ꢀH.ꢀZhu,ꢀW.ꢀH.ꢀZhang,ꢀJ.ꢀInd.ꢀEng.ꢀChem.,ꢀ2012,ꢀ
tionꢀmethodsꢀ(e.g.,ꢀSEM,ꢀEDS,ꢀTEM,ꢀH
2
‐TPR,ꢀXPS,ꢀandꢀH
2
‐O ꢀti‐
2
tration)ꢀ wereꢀ consistentꢀ withꢀ theꢀ activityꢀ test.ꢀ Toꢀ theꢀ bestꢀ ofꢀ
ourꢀknowledge,ꢀthisꢀisꢀtheꢀfirstꢀworkꢀtoꢀstudyꢀtheꢀcatalyticꢀper‐
18,ꢀ1851–1857.ꢀ
4] J.ꢀM.ꢀCampos‐Martin,ꢀG.ꢀBlanco‐Brieva,ꢀJ.ꢀL.ꢀG.ꢀFierro,ꢀAngew.ꢀChem.ꢀ
Int.ꢀEd.,ꢀ2006,ꢀ45,ꢀ6962–6984.ꢀ ꢀ
5] V.ꢀPaunovic,ꢀV.ꢀOrdomsky,ꢀM.ꢀF.ꢀN.ꢀD’Angelo,ꢀJ.ꢀC.ꢀSchouten,ꢀT.ꢀA.ꢀ
Nijhuis,ꢀJ.ꢀCatal.,ꢀ2014,ꢀ309,ꢀ325–332.ꢀ ꢀ
6] L.ꢀK.ꢀOuyang,ꢀG.ꢀJ.ꢀDa,ꢀP.ꢀF.ꢀTian,ꢀT.ꢀY.ꢀChen,ꢀG.ꢀD.ꢀLiang,ꢀJ.ꢀXu,ꢀY.ꢀF.ꢀ
Han,ꢀJ.ꢀCatal.,ꢀ2014,ꢀ311,ꢀ129–136.ꢀ ꢀ
formanceꢀonꢀPd‐M/SiO /CORꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀMn,ꢀandꢀCu)ꢀbimetal‐
2
licꢀmonolithicꢀcatalystsꢀforꢀtheꢀhydrogenationꢀofꢀeAQ.ꢀ ꢀ
DFTꢀ calculationsꢀ revealedꢀ thatꢀ theꢀ strongꢀ adsorptionꢀ be‐
tweenꢀtheꢀPd‐Mꢀ(Mꢀ=ꢀNi,ꢀFe,ꢀandꢀMn)ꢀalloyꢀsurfaceꢀandꢀtheꢀcar‐
bonylꢀ groupꢀ ofꢀ eAQꢀ wasꢀ responsibleꢀ forꢀ theꢀ higherꢀ catalyticꢀ
GraphicalꢀAbstractꢀ
Chin.ꢀJ.ꢀCatal.,ꢀ2018,ꢀ39:ꢀ1070–1080ꢀ ꢀ ꢀ doi:ꢀ10.1016/S1872‐2067(18)63035‐9
Hydrogenationꢀofꢀ2‐ethylanthraquinoneꢀwithꢀbimetallicꢀ ꢀ
monolithicꢀcatalysts:ꢀAnꢀexperimentalꢀandꢀDFTꢀstudyꢀ
M = Ni, Fe, Mn, and Cu
Pd/PdM
Hydrogenation
YanyanꢀGuo,ꢀChengnaꢀDai,ꢀZhigangꢀLeiꢀ*ꢀ
BeijingꢀUniversityꢀofꢀChemicalꢀTechnologyꢀ
Oxidization
Oxidization
2
‐Ethylanthraquinone(EAQ)
2‐Ethylanthrahydroquinone(EAQH2)
ꢀ
Pd/PdM
M = Ni, Fe, Mn, and Cu
Hydrogenation
Theꢀ highꢀ catalyticꢀ activityꢀ ofꢀ Pd‐Ni/SiO /cordieriteꢀ monolithicꢀ
catalystꢀisꢀattributedꢀtoꢀtheꢀsmallꢀparticleꢀsize,ꢀtheꢀstrongꢀinterac‐
tionꢀ betweenꢀ Pdꢀ andꢀ Niꢀ ofꢀ Pd‐Niꢀ alloyꢀ andꢀ strongerꢀ adsorptionꢀ
2
3 1
betweenꢀPd Ni ꢀ(1ꢀ1ꢀ1)ꢀandꢀtheꢀcarbonylꢀgroupꢀofꢀeAQ.ꢀ
4
Tetrahydro 2‐Ethylanthraquinone(H EAQ)
4 2
Tetrahydro 2‐Ethylanthrahydroquinone(H EAQH )

ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
1079ꢀ
[
7] R.ꢀ Ciriminna,ꢀ L.ꢀ Albanese,ꢀ F.ꢀ Meneguzzo,ꢀ M.ꢀ Pagliaro,ꢀ
Ruban,ꢀ P.ꢀ J.ꢀ Uggowitzer,ꢀ E.ꢀ Kozeschnik,ꢀ Calphad,ꢀ 2015,ꢀ 51,ꢀ
314–333.ꢀ ꢀ
ChemSusChem,ꢀ2016,ꢀ9,ꢀ3374–3374.ꢀ ꢀ
[
8] R.ꢀB.ꢀRankin,ꢀJ.ꢀGreeley,ꢀACSꢀCatal.,ꢀ2012,ꢀ2,ꢀ2664–2672.ꢀ ꢀ
9] H.ꢀM.ꢀChen,ꢀD.ꢀP.ꢀHuang,ꢀX.ꢀY.ꢀSu,ꢀJ.ꢀL.ꢀHuang,ꢀX.ꢀL.ꢀJing,ꢀM.ꢀM.ꢀDu,ꢀD.ꢀ
H.ꢀSun,ꢀL.ꢀS.ꢀJia,ꢀQ.ꢀB.ꢀLi,ꢀChem.ꢀEng.ꢀJ.,ꢀ2015,ꢀ262,ꢀ356–363.ꢀ ꢀ
[32] S.ꢀSitthisa,ꢀT.ꢀPham,ꢀT.ꢀPrasomsri,ꢀT.ꢀSooknoi,ꢀR.ꢀG.ꢀMallinson,ꢀD.ꢀE.ꢀ
Resasco,ꢀJ.ꢀCatal.,ꢀ2011,ꢀ280,ꢀ17–27.ꢀ ꢀ
[33] O.ꢀSanz,ꢀF.ꢀJ.ꢀEchave,ꢀM.ꢀSánchez,ꢀA.ꢀMonzon,ꢀM.ꢀMontes,ꢀAppl.ꢀCatal.ꢀ
A,ꢀ2008,ꢀ340,ꢀ125–132.ꢀ ꢀ
[34] B.ꢀYuan,ꢀB.ꢀZhang,ꢀZ.ꢀL.ꢀWang,ꢀS.ꢀM.ꢀLiu,ꢀJ.ꢀLi,ꢀY.ꢀLiu,ꢀC.ꢀLi,ꢀChin.ꢀJ.ꢀ
Catal.,ꢀ2017,ꢀ38,ꢀ440–446.ꢀ ꢀ
[35] M.ꢀQ.ꢀChai,ꢀX.ꢀY.ꢀLiu,ꢀL.ꢀLi,ꢀG.ꢀX.ꢀPei,ꢀY.ꢀJ.ꢀRen,ꢀY.ꢀSu,ꢀH.ꢀK.ꢀCheng,ꢀA.ꢀQ.ꢀ
Wang,ꢀT.ꢀZhang,ꢀChin.ꢀJ.ꢀCatal.,ꢀ2017,ꢀ38,ꢀ1338–1346.ꢀ
[36] B.ꢀ Y.ꢀ Bai,ꢀ Q.ꢀ Qiao,ꢀ J.ꢀ H.ꢀ Li,ꢀ J.ꢀ M.ꢀ Hao,ꢀ Chin.ꢀ J.ꢀ Catal.,ꢀ 2016,ꢀ 37,ꢀ
102–122.ꢀ
[37] G.ꢀ Agostini,ꢀ C.ꢀ Lamberti,ꢀ R.ꢀ Pellegrini,ꢀ G.ꢀ Leofanti,ꢀ F.ꢀ Giannici,ꢀ A.ꢀ
Longo,ꢀE.ꢀGroppo,ꢀACSꢀCatal.,ꢀ2014,ꢀ4,ꢀ187–194.ꢀ ꢀ
[38] E.ꢀGroppo,ꢀG.ꢀAgostini,ꢀA.ꢀPiovano,ꢀN.ꢀB.ꢀMuddada,ꢀG.ꢀLeofanti,ꢀR.ꢀ
Pellegrini,ꢀ G.ꢀ Portale,ꢀ A.ꢀ Longo,ꢀ C.ꢀ Lamberti,ꢀ J.ꢀCatal.,ꢀ2012,ꢀ 287,ꢀ
44–54.ꢀ ꢀ
[
[
10] Y.ꢀY.ꢀGuo,ꢀC.ꢀN.ꢀDai,ꢀZ.ꢀG.ꢀLei,ꢀChem.ꢀEng.ꢀSci.,ꢀ2017,ꢀ172,ꢀ370–384.ꢀ ꢀ
11] A.ꢀ Drelinkiewicz,ꢀ A.ꢀ Waksmundzka‐Góra,ꢀ J.ꢀ Mol.ꢀ Catal.ꢀ A,ꢀ 2006,ꢀ
[
2
58,ꢀ1–9.ꢀ ꢀ
12] R.ꢀKosydar,ꢀA.ꢀLinkiewicz,ꢀE.ꢀLalik,ꢀJ.ꢀGurgul,ꢀAppl.ꢀCatal.ꢀA,ꢀ2011,ꢀ
02,ꢀ121–131.ꢀ ꢀ
13] C.ꢀShen,ꢀY.ꢀJ.ꢀWang,ꢀJ.ꢀH.ꢀXu,ꢀY.ꢀC.ꢀLu,ꢀG.ꢀS.ꢀLuo,ꢀChem.ꢀEng.ꢀJ.,ꢀ2011,ꢀ
73,ꢀ226–232.ꢀ ꢀ
14] Y.ꢀC.ꢀZhu,ꢀL.ꢀY.ꢀYu,ꢀX.ꢀF.ꢀWang,ꢀY.ꢀH.ꢀZhou,ꢀH.ꢀQ.ꢀYe,ꢀCatal.ꢀCommun.,ꢀ
013,ꢀ40,ꢀ98–102.ꢀ ꢀ
15] A.ꢀDrelinkiewicz,ꢀM.ꢀHasik,ꢀJ.ꢀMol.ꢀCatal.ꢀA,ꢀ2001,ꢀ177,ꢀ149–164.ꢀ ꢀ
16] J.ꢀ G.ꢀ Zhang,ꢀ D.ꢀ F.ꢀ Li,ꢀ Y.ꢀ J.ꢀ Zhao,ꢀ Q.ꢀ D.ꢀ Kong,ꢀ S.ꢀ D.ꢀ Wang,ꢀ Catal.ꢀ
Commun.,ꢀ2008,ꢀ9,ꢀ2565–2569.ꢀ ꢀ
[
[
[
4
1
2
[
[
[
17] X.ꢀ T.ꢀ Li,ꢀ H.ꢀ J.ꢀ Su,ꢀ G.ꢀ Y.ꢀ Ren,ꢀ S.ꢀ D.ꢀ Wang,ꢀ RSCꢀ Adv.,ꢀ 2015,ꢀ 5,ꢀ
[39] G.ꢀX.ꢀPei,ꢀX.ꢀY.ꢀLiu,ꢀM.ꢀQ.ꢀChai,ꢀA.ꢀQ.ꢀWang,ꢀT.ꢀZhang,ꢀChin.ꢀJ.ꢀCatal.,ꢀ
2017,ꢀ38,ꢀ1540–1548.ꢀ ꢀ
100968–100977.ꢀ ꢀ
[
18] Y.ꢀNakagawa,ꢀK.ꢀTomishige,ꢀCatal.ꢀCommun.,ꢀ2010,ꢀ12,ꢀ154–156.ꢀ ꢀ
19] A.ꢀA.ꢀShesterkina,ꢀO.ꢀA.ꢀKirichenko,ꢀL.ꢀM.ꢀKozlova,ꢀG.ꢀI.ꢀKapustin,ꢀI.ꢀ
V.ꢀMishin,ꢀA.ꢀA.ꢀStrelkova,ꢀL.ꢀM.ꢀKustov,ꢀMendeleevꢀCommun.,ꢀ2016,ꢀ
[40] S.ꢀSitthisa,ꢀT.ꢀPham,ꢀT.ꢀPrasomsri,ꢀT.ꢀSooknoi,ꢀR.ꢀG.ꢀMallinson,ꢀD.ꢀE.ꢀ
Resasco,ꢀJ.ꢀCatal.,ꢀ2011,ꢀ280,ꢀ17–27.ꢀ ꢀ
[41] R.ꢀ R.ꢀ Hong,ꢀ Y.ꢀ F.ꢀ He,ꢀ J.ꢀ T.ꢀ Feng,ꢀ D.ꢀ Q.ꢀ Li.ꢀ AIChEꢀ J.,ꢀ 2017,ꢀ 63,ꢀ
3955–3965.ꢀ
[
26,ꢀ228–230.ꢀ ꢀ
[
20] P.ꢀInsorn,ꢀB.ꢀKitiyanan,ꢀCatal.ꢀToday,ꢀ2015,ꢀ256,ꢀ223–230.ꢀ ꢀ
21] K.ꢀ Fulajtárova,ꢀ T.ꢀ Soták,ꢀ M.ꢀ Hronec,ꢀ I.ꢀ Vohra,ꢀ E.ꢀ Dobrocka,ꢀ M.ꢀ
Omastová,ꢀAppl.ꢀCatal.ꢀA,ꢀ2015,ꢀ502,ꢀ78–85.ꢀ ꢀ
[42] Y.ꢀN.ꢀLiu,ꢀY.ꢀF.ꢀHe,ꢀD.ꢀR.ꢀZhou,ꢀJ.ꢀT.ꢀFeng,ꢀD.ꢀQ.ꢀLi,ꢀCatal.ꢀSci.ꢀTechnol.,ꢀ
2016,ꢀ6,ꢀ3027–3037.ꢀ ꢀ
[43] J.ꢀSá,ꢀG.ꢀD.ꢀArteaga,ꢀR.ꢀA.ꢀDaley,ꢀJ.ꢀBernardi,ꢀJ.A.ꢀAnderson,ꢀJ.ꢀPhys.ꢀ
Chem.ꢀB,ꢀ2006,ꢀ110,ꢀ17090–17095.ꢀ ꢀ
[44] S.ꢀKarski,ꢀI.ꢀWitońska,ꢀJ.ꢀRogowski,ꢀJ.ꢀGołuchowska.ꢀJ.ꢀMol.ꢀCatal.ꢀA,ꢀ
2005,ꢀ240,ꢀ155–163.ꢀ
[45] S.ꢀKarski,ꢀI.ꢀWitońska.ꢀJ.ꢀMol.ꢀCatal.ꢀA,ꢀ2003,ꢀ191,ꢀ87–92.ꢀ
[46] Q.ꢀ Zhang,ꢀ J.ꢀ Li,ꢀ X.ꢀ X.ꢀ Liu,ꢀ Q.ꢀ M.ꢀ Zhu.ꢀ Appl.ꢀ Catal.ꢀ A,ꢀ 2000,ꢀ 197,ꢀ
221–228.ꢀ
[
[
[
[
22] E.ꢀX.ꢀYuan,ꢀC.ꢀWu,ꢀX.ꢀHou,ꢀM.ꢀB.ꢀDou,ꢀG.ꢀZ.ꢀLiu,ꢀG.ꢀZ.ꢀLi,ꢀL.ꢀWang,ꢀJ.ꢀ
Catal.,ꢀ2017,ꢀ347,ꢀ79–88.ꢀ ꢀ
23] J.ꢀT.ꢀFeng,ꢀH.ꢀY.ꢀWang,ꢀD.ꢀG.ꢀEvans,ꢀX.ꢀDuan,ꢀD.ꢀQ.ꢀLi,ꢀAppl.ꢀCatal.ꢀA,ꢀ
2
010,ꢀ382,ꢀ240–245.ꢀ ꢀ
24] Y.ꢀY.ꢀGuo,ꢀC.ꢀN.ꢀDai,ꢀZ.ꢀG.ꢀLei,ꢀB.ꢀH.ꢀChen,ꢀX.ꢀC.ꢀFang,ꢀCatal.ꢀToday,ꢀ
016,ꢀ276,ꢀ36–45.ꢀ ꢀ ꢀ
2
[
25] R.ꢀCar,ꢀM.ꢀParrinello,ꢀPhys.ꢀRev.ꢀLett.,ꢀ1985,ꢀ55,ꢀ2471–2474.ꢀ ꢀ
26] M.ꢀC.ꢀPayne,ꢀJ.ꢀD.ꢀJoannopoulos,ꢀD.ꢀC.ꢀAllan,ꢀM.ꢀP.ꢀTeter,ꢀD.ꢀH.ꢀVan‐
derbilt,ꢀPhys.ꢀRev.ꢀLett.,ꢀ1986,ꢀ56,ꢀ2656.ꢀ ꢀ
[47] A.ꢀS.ꢀIvanova,ꢀE.ꢀM.ꢀSlavinskaya,ꢀR.ꢀV.ꢀGulyaev,ꢀV.ꢀI.ꢀZaikovskii,ꢀO.ꢀA.ꢀ
Stonkus,ꢀI.ꢀG.ꢀDanilova,ꢀL.ꢀM.ꢀPlyasova,ꢀI.ꢀA.ꢀPolukhina,ꢀA.ꢀI.ꢀBoronin,ꢀ
Appl.ꢀCatal.ꢀB,ꢀ2010,ꢀ97,ꢀ57–71.ꢀ ꢀ
[
[27] J.ꢀ P.ꢀ Perdew,ꢀ K.ꢀ Burke,ꢀ M.ꢀ Ernzerhof,ꢀ Phys.ꢀ Rev.ꢀ Lett.,ꢀ 1996,ꢀ 77,ꢀ
[48] V.ꢀJohánek,ꢀI.ꢀStará,ꢀV.ꢀMatolı ń ,ꢀSurf.ꢀSci.,ꢀ2002,ꢀ507–510,ꢀ92–98.ꢀ ꢀ
3
865–3868.ꢀ ꢀ
[49] S.ꢀ Maheswari,ꢀ S.ꢀ Karthikeyan,ꢀ P.ꢀ Murugan,ꢀ P.ꢀ Sridhar,ꢀ S.ꢀ
Pitchumani,ꢀPhys.ꢀChem.ꢀChem.ꢀPhys.,ꢀ2012,ꢀ14,ꢀ9683–9695.ꢀ ꢀ
[50] N.ꢀLopez,ꢀJ.ꢀK.ꢀNørskov,ꢀSurf.ꢀSci.,ꢀ2001,ꢀ477,ꢀ59–75.ꢀ ꢀ
[51] T.ꢀKamachi,ꢀT.ꢀOgata,ꢀE.ꢀMori,ꢀK.ꢀIura,ꢀN.ꢀOkuda,ꢀM.ꢀNagata,ꢀK.ꢀYo‐
shizawa,ꢀJ.ꢀPhys.ꢀChem.ꢀC,ꢀ2015,ꢀ119,ꢀ8748–8754.ꢀ ꢀ
[
28] G.ꢀKresse,ꢀD.ꢀJoubert,ꢀPhys.ꢀRev.ꢀB,ꢀ1999,ꢀ59,ꢀ1758–1775.ꢀ ꢀ
29] F.ꢀCinquini,ꢀF.ꢀDelbecq,ꢀP.ꢀSautet,ꢀPhys.ꢀChem.ꢀChem.ꢀPhys.,ꢀ2009,ꢀ
[
11,ꢀ11546–11556.ꢀ ꢀ
[
30] N.ꢀPino,ꢀS.ꢀSitthisa,ꢀQ.ꢀH.ꢀTan,ꢀT.ꢀSouza,ꢀD.ꢀLópez,ꢀD.ꢀE.ꢀResasco,ꢀJ.ꢀ
Catal.,ꢀ2017,ꢀ350,ꢀ30–40.ꢀ ꢀ
31] E.ꢀ Povoden‐Karadeniz,ꢀ P.ꢀ Lang,ꢀ F.ꢀ Moszner,ꢀ S.ꢀ Pogatscher,ꢀ A.ꢀ V.ꢀ
[52] M.ꢀK.ꢀBradley,ꢀJ.ꢀbinꢀRoson,ꢀD.ꢀP.ꢀWoodruff,ꢀSurf.ꢀSci.,ꢀ2010,ꢀ604,ꢀ
920–925.
[
2
-乙基蒽醌在双金属整体式催化剂上的氢化反应: 实验和 DFT 研究
*
郭燕燕, 代成娜, 雷志刚
北京化工大学化工资源有效利用国家重点实验室, 北京100029
摘要: 过氧化氢(H
2
O
2
2 2
)是一种绿色化工原料和环境友好氧化剂. 目前, 超过 98% 的H O 是通过蒽醌法生产. 蒽醌法主要包
括 2-乙基蒽醌氢化生成 2-乙基氢蒽醌和 2-乙基氢蒽醌氧化生成 2-乙基蒽醌和H
2
O
2
的过程. 其中, 2-乙基蒽醌氢化是关键
步骤. 在氢化过程中, 生成的 2-乙基氢蒽醌和四氢-2-乙基氢蒽醌是目标产物, 同时生成许多副产物. 目前, Pd 颗粒催化剂
是广泛使用的催化剂, 但是蒽醌氢化过程中, 质量传递是主要的控制因素. 与颗粒催化剂对比, 整体式催化剂可以减弱整
个反应的内外扩散, 提高反应速率. 很多研究结果显示, 整体式催化剂的传质优于颗粒催化剂, 可以提高催化效率. 近期许
多研究显示, 双金属颗粒催化剂在很多氢化反应中体现出优异的催化性能. 本工作制备了双金属整体式催化剂, 考察了其
在蒽醌氢化过程中的催化性能.
首先, 通过浸渍法制备了4 种双金属整体式催化剂 Pd-M/SiO
2
/COR (M = Ni, Fe, Mn和 Cu)以及Pd/SiO
产量低于 Pd/SiO
2
/COR和
o
Ni/SiO /COR两种单金属整体式催化剂. 催化活性结果显示, Ni/SiO /COR的H
2
2
2
O
2
2
/COR, 而且在 700 C还

1080ꢀ
YanyanꢀGuoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1070–1080ꢀ
2 2 2
原的 Pd-Ni/SiO /COR 整体式催化剂在 Pd/M = 2 时取得了最高选择性 (95.3%) 和H O 产量 (7.5 g/L). 然后, 考察了金属负
载量的影响. 结果显示, 在金属负载量低于 0.4% 时, 随着金属负载量增加, 选择性和H O 产量增加, 在金属负载量高于
2
2
0
.4% 时, 随着金属负载量增加, 选择性和H
TEM结果表明, 添加第二种金属后, 双金属整体式催化剂颗粒尺寸变小, 分布更均匀. EDS结果显示, 双金属形成了合
金. H -TPR结果显示, 随着Pd/M比率增加, 还原温度降低, 说明Pd有助于第二种金属氧化物的还原. 这可能是由于 Pd 表面
2 2
O 产量降低.
2
的氢溢流到第二种金属 (Ni, Fe, Mn和 Cu) 表面. 此外, 文献结果表明, 合金的形成能够抑制 PdH 的形成. 本工作表明添加
第二种金属 (Ni, Fe, Mn和Cu) 后, PdH 的峰强度减弱或者峰消失, 也说明形成了合金. XPS 结果显示, 添加第二种金属后,
在 336.3 ± 0.1 和 341.4 ± 0.1 eV 出现了新的 Pd 3d5/2 和 Pd 3d3/2 峰, 说明形成了合金. H -O 滴定结果表明, Pd-Ni/SiO /COR
2 2 2
的Pd分散度和Pd比表面积都高于其他双金属催化剂, 说明第二种金属 Ni 更有利于促进 Pd 的分散, 减弱颗粒集聚, 揭示了
Pd 和 Ni 之间强烈的相互作用.
DFT 计算结果显示, Pd
3 1 2
M (M = Ni, Fe, Mn和Cu) 双金属整体式催化剂和 2-乙基蒽醌之间的结合能低于 Pd/SiO /COR
和 2-乙基蒽醌之间的结合能, 但是 Pd M (M = Ni, Fe和Mn) 双金属催化剂和 2-乙基氢蒽醌之间的结合能减小得很少, 这可
3
1
能是由于 2-乙基蒽醌的 C=O 和第二种金属之间具有强烈相互作用的缘故. Pd
结合能减小很多, 主要是由于 Pd Cu 表面不利于 2-乙基氢蒽醌的吸附.
因此, Pd-Ni/SiO /COR 比 Pd/SiO /COR, Ni/SiO /COR 和其他的双金属整体式催化剂具有更高的选择性和H O 产量, 主
3 1
Cu 双金属催化剂和 2-乙基氢蒽醌之间的
3
1
2
2
2
2
2
要是由于合金的形成以及 2-乙基氢蒽醌的 C=O 双键和 2-乙基氢蒽醌强烈的相互作用.
关键词: 双金属整体式催化剂; 合金; 2-乙基氢蒽醌; DFT 计算; 协同作用
收稿日期: 2017-12-27. 接受日期: 2018-01-24. 出版日期: 2018-06-05.
*
通讯联系人. 电话: (010)64433695; 电子信箱: leizhg@mail.buct.edu.cn
基金来源: 国家自然科学基金 (21476009, 21406007, U1462104).
本文的电子版全文由Elsevier出版社在ScienceDirect上出版(http://www.sciencedirect.com/science/journal/18722067).