Selective hydrogenolysis of furfuryl alcohol to 1,5- and 1,2-pentanediol over Cu-LaCoO3 catalysts with balanced Cu0-CoO sites

Article abstract of DOI:10.1016/S1872-2067(18)63110-9

Selective hydrogenolysis of biomass-derived furfuryl alcohol (FFA) to 1,5- and 1,2-pentanediol (PeD) was conducted over Cu-LaCoO₃ catalysts with different Cu loadings; the catalysts were derived from perovskite structures prepared by a one-step citrate complexing method. The catalytic performances of the Cu-LaCoO₃ catalysts were found to depend on the Cu loading and pretreatment conditions. The catalyst with 10 wt% Cu loading exhibited the best catalytic performance after prereduction in 5%H₂-95%N₂, achieving a high FFA conversion of 100% and selectivity of 55.5% for 1,5-pentanediol (40.3%) and 1,2-pentanediol (15.2%) at 413 K and 6 MPa H₂. This catalyst could be reused four times without a loss of FFA conversion but it resulted in a slight decrease in pentanediol selectivity. Correlation between the structural changes in the catalysts at different states and the simultaneous variation in the catalytic performance revealed that cooperative catalysis between Cu⁰ and CoO promoted the hydrogenolysis of FFA to PeDs, especially to 1,5-PeD, while Co⁰ promoted the hydrogenation of FFA to tetrahydrofurfuryl alcohol (THFA). Therefore, it is suggested that a synergetic effect between balanced Cu⁰ and CoO sites plays a critical role in achieving a high yield of PeDs with a high 1,5-/1,2-pentanediol selectivity ratio during FFA hydrogenolysis.

Full text of DOI:10.1016/S1872-2067(18)63110-9

ChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723
ꢀ
ꢀ ꢀ ꢀ
ꢀ ꢀ ꢀ ꢀ
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
Selectiveꢀhydrogenolysisꢀofꢀfurfurylꢀalcoholꢀtoꢀ1,5‐ꢀandꢀ
,2‐pentanediolꢀoverꢀCu‐LaCoO ꢀcatalystsꢀwithꢀbalancedꢀCu ‐CoOꢀ
sitesꢀ
0
1
3
a,b
a
a
a,#
a,
a
FangfangꢀGaoꢀ ,ꢀHailongꢀLiuꢀ ,ꢀXunꢀHuꢀ ,ꢀJingꢀChenꢀ ,ꢀZhiweiꢀHuangꢀ *,ꢀChunguꢀXiaꢀ
ꢀ
aꢀStateꢀKeyꢀLaboratoryꢀforꢀOxoꢀSynthesisꢀandꢀSelectiveꢀOxidation,ꢀSuzhouꢀResearchꢀInstituteꢀofꢀLICP,ꢀLanzhouꢀInstituteꢀofꢀChemicalꢀPhysicsꢀ(LICP),ꢀChineseꢀ
AcademyꢀofꢀSciences,ꢀLanzhouꢀ730000,ꢀGansu,ꢀChinaꢀ
bꢀUniversityꢀofꢀChineseꢀAcademyꢀofꢀSciences,ꢀBeijingꢀ100049,ꢀChinaꢀ
A R T I C L E ꢀ I N F O ꢀ
A B S T R A C T ꢀ
Articleꢀhistory:ꢀ
Selectiveꢀ hydrogenolysisꢀ ofꢀ biomass‐derivedꢀ furfurylꢀ alcoholꢀ (FFA)ꢀ toꢀ 1,5‐ꢀ andꢀ 1,2‐pentanediolꢀ
Receivedꢀ7ꢀAprilꢀ2018ꢀ
Acceptedꢀ25ꢀMayꢀ2018ꢀ
Publishedꢀ5ꢀOctoberꢀ2018ꢀ
(PeD)ꢀwasꢀconductedꢀoverꢀCu‐LaCoO
rivedꢀfromꢀperovskiteꢀstructuresꢀpreparedꢀbyꢀaꢀone‐stepꢀcitrateꢀcomplexingꢀmethod.ꢀTheꢀcatalyticꢀ
performancesꢀofꢀtheꢀCu‐LaCoO ꢀcatalystsꢀwereꢀfoundꢀtoꢀdependꢀonꢀtheꢀCuꢀloadingꢀandꢀpretreatmentꢀ
conditions.ꢀ Theꢀ catalystꢀ withꢀ 10ꢀ wt%ꢀ Cuꢀ loadingꢀ exhibitedꢀ theꢀ bestꢀ catalyticꢀ performanceꢀ afterꢀ
prereductionꢀinꢀ5%H ‐95%N ,ꢀachievingꢀaꢀhighꢀFFAꢀconversionꢀofꢀ100%ꢀandꢀselectivityꢀofꢀ55.5%ꢀforꢀ
,5‐pentanediolꢀ(40.3%)ꢀandꢀ1,2‐pentanediolꢀ(15.2%)ꢀatꢀ413ꢀKꢀandꢀ6ꢀMPaꢀH .ꢀThisꢀcatalystꢀcouldꢀbeꢀ
3
ꢀcatalystsꢀwithꢀdifferentꢀCuꢀloadings;ꢀtheꢀcatalystsꢀwereꢀde‐
3
2
2
Keywords:ꢀ
1
2
Furfurylꢀalcoholꢀ
Selectiveꢀhydrogenolysisꢀ
Pentanediolꢀ
reusedꢀfourꢀtimesꢀwithoutꢀaꢀlossꢀofꢀFFAꢀconversionꢀbutꢀitꢀresultedꢀinꢀaꢀslightꢀdecreaseꢀinꢀpentanediolꢀ
selectivity.ꢀCorrelationꢀbetweenꢀtheꢀstructuralꢀchangesꢀinꢀtheꢀcatalystsꢀatꢀdifferentꢀstatesꢀandꢀtheꢀ
simultaneousꢀvariationꢀinꢀtheꢀcatalyticꢀperformanceꢀrevealedꢀthatꢀcooperativeꢀcatalysisꢀbetweenꢀCu0
3
Cu‐LaCoO ꢀcatalystꢀ
0
andꢀCoOꢀpromotedꢀtheꢀhydrogenolysisꢀofꢀFFAꢀtoꢀPeDs,ꢀespeciallyꢀtoꢀ1,5‐PeD,ꢀwhileꢀCo ꢀpromotedꢀtheꢀ
Perovskiteꢀstructureꢀ
hydrogenationꢀofꢀFFAꢀtoꢀtetrahydrofurfurylꢀalcoholꢀ(THFA).ꢀTherefore,ꢀitꢀisꢀsuggestedꢀthatꢀaꢀsyner‐
0
geticꢀeffectꢀbetweenꢀbalancedꢀCu ꢀandꢀCoOꢀsitesꢀplaysꢀaꢀcriticalꢀroleꢀinꢀachievingꢀaꢀhighꢀyieldꢀofꢀPeDsꢀ
withꢀaꢀhighꢀ1,5‐/1,2‐pentanediolꢀselectivityꢀratioꢀduringꢀFFAꢀhydrogenolysis.ꢀ
©ꢀ2018,ꢀDalianꢀInstituteꢀofꢀChemicalꢀPhysics,ꢀChineseꢀAcademyꢀofꢀSciences.
PublishedꢀbyꢀElsevierꢀB.V.ꢀAllꢀrightsꢀreserved.
1.ꢀ ꢀ Introductionꢀ
vertedꢀintoꢀaꢀvarietyꢀofꢀimportantꢀchemicalsꢀandꢀfuels,ꢀsuchꢀasꢀ
1,2‐pentanediolꢀ (1,2‐PeD),ꢀ 1,5‐pentanediolꢀ (1,5‐PeD),ꢀ furan,ꢀ
Theꢀ conversionꢀ ofꢀ renewableꢀ andꢀ abundantꢀ biomassꢀ feed‐
2‐methylꢀ furan,ꢀ cyclopentanone,ꢀ andꢀ γ‐valerolactoneꢀ [4–7].ꢀ
Particularly,ꢀtheꢀselectiveꢀhydrogenolysisꢀofꢀFAꢀandꢀitsꢀderiva‐
tives,ꢀ furfurylꢀ alcoholꢀ (FFA)ꢀ andꢀ tetrahydrofurfurylꢀ alcoholꢀ
(THFA),ꢀintoꢀusefulꢀdiols,ꢀsuchꢀasꢀ1,2‐PeDꢀandꢀ1,5‐PeD,ꢀisꢀno‐
ticeablyꢀattractiveꢀ[8–12].ꢀTheseꢀPeDsꢀareꢀwidelyꢀusedꢀforꢀtheꢀ
productionꢀofꢀmicrobicides,ꢀcosmetics,ꢀpolyesters,ꢀplastics,ꢀetc.ꢀ
stockꢀavailableꢀinꢀnatureꢀintoꢀchemicalsꢀandꢀfuelsꢀisꢀconsideredꢀ
aꢀ feasibleꢀ techniqueꢀ toꢀ alleviateꢀ theꢀ currentꢀ increasingꢀ envi‐
ronmentalꢀ andꢀ resourceꢀ problemsꢀ [1–3].ꢀ Furfuralꢀ (FA)ꢀ isꢀ anꢀ
importantꢀ platformꢀ chemical,ꢀ industriallyꢀ producedꢀ fromꢀ lig‐
nocellulosicꢀbiomassꢀthroughꢀacidicꢀhydrolysis.ꢀItꢀcanꢀ beꢀ con‐
*
ꢀ
ꢀCorrespondingꢀauthor.ꢀTel:ꢀ+86‐931‐4968070;ꢀFax:ꢀ+86‐931‐4968129;ꢀE‐mail:ꢀzwhuang@licp.cas.cnꢀ
Correspondingꢀauthor.ꢀTel:ꢀ+86‐931‐4968068;ꢀFax:ꢀ+86‐931‐4968129;ꢀE‐mail:ꢀchenj@licp.cas.cn
#
ThisꢀworkꢀwasꢀsupportedꢀbyꢀtheꢀNationalꢀNaturalꢀScienceꢀFoundationꢀofꢀChinaꢀ(21473224),ꢀKeyꢀResearchꢀProjectꢀofꢀFrontierꢀScienceꢀofꢀChineseꢀ
AcademyꢀofꢀSciencesꢀ(QYZDJ‐SSW‐SLH051),ꢀtheꢀYouthꢀInnovationꢀPromotionꢀAssociation,ꢀCASꢀ(2016371),ꢀandꢀtheꢀSuzhouꢀScienceꢀandꢀTechnologyꢀ
DevelopmentꢀPlanꢀ(SYG201626).ꢀ
DOI:ꢀ10.1016/S1872‐2067(18)63110‐9ꢀ|ꢀhttp://www.sciencedirect.com/science/journal/18722067ꢀ|ꢀChin.ꢀJ.ꢀCatal.,ꢀVol.ꢀ39,ꢀNo.ꢀ10,ꢀOctoberꢀ2018ꢀ ꢀ ꢀ

1712ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
[
6,13].ꢀ Nowadays,ꢀ PeDsꢀ areꢀ generallyꢀ producedꢀ viaꢀ aꢀ
overꢀ1,2‐PeD,ꢀwithꢀaꢀ1,5‐PeD/1,2‐PeDꢀselectivityꢀratioꢀofꢀupꢀtoꢀ
3/1;ꢀthisꢀresultꢀisꢀmuchꢀhigherꢀthanꢀthatꢀobtainedꢀinꢀpreviousꢀ
studiesꢀ onꢀ Cu‐basedꢀ catalysts,ꢀ suchꢀ asꢀ copperꢀ chromiteꢀ (3/4)ꢀ
cost‐intensiveꢀ multistepꢀ routeꢀ fromꢀ petroleum‐derivedꢀ feed‐
stocksꢀ involvingꢀ selectiveꢀ oxidationꢀ andꢀ reductionꢀ reactionsꢀ
[9,13].ꢀTheꢀavailabilityꢀofꢀFAꢀfromꢀabundantꢀrenewableꢀligno‐
3 2 3
[29],ꢀ Cu‐Mg AlO4.5ꢀ (3/5)ꢀ [8],ꢀ andꢀ Cu‐Al O ꢀ (1/2.2)ꢀ [27].ꢀ Theꢀ
celluloseꢀ[3],ꢀinꢀcontrastꢀtoꢀnon‐renewableꢀC ꢀpetroleumꢀfeed‐
5
possibleꢀreasonsꢀforꢀtheꢀhigherꢀselectivityꢀofꢀCu‐LaCoO
3
ꢀ cata‐
stocks,ꢀmakesꢀtheꢀselectiveꢀhydrogenolysisꢀofꢀFAꢀandꢀitsꢀderiva‐
tivesꢀ ofꢀ FFA/THFAꢀ aꢀ practicalꢀ pathwayꢀ forꢀ theꢀ sustainableꢀ
productionꢀofꢀPeDsꢀwithꢀhighꢀenergyꢀefficiency.ꢀ ꢀ
lystsꢀ towardsꢀ 1,5‐PeDꢀ areꢀ discussedꢀ inꢀ termsꢀ ofꢀ theꢀ effectꢀ ofꢀ
differentꢀ reductionꢀ atmospheresꢀ andꢀ theꢀ catalystsꢀ character‐
izedꢀatꢀdifferentꢀstatesꢀare.ꢀ
Thusꢀ far,ꢀ theꢀ selectiveꢀ hydrogenolysisꢀ ofꢀ FAꢀ orꢀ itsꢀ deriva‐
tivesꢀatꢀtheꢀC–OꢀbondsꢀinꢀtheꢀfuranꢀringꢀtoꢀproduceꢀPeDsꢀmainlyꢀ
focusedꢀ onꢀ usingꢀ supportedꢀ catalystsꢀ ofꢀ groupꢀ VIIIꢀ preciousꢀ
metalsꢀ (Ruꢀ [13],ꢀ Ptꢀ [12,14,15],ꢀ Rhꢀ [16–21],ꢀ andꢀ Irꢀ [22–26]).ꢀ
2.ꢀ ꢀ Experimentalꢀ
2.1.ꢀ ꢀ Materialsꢀ
Low‐valencyꢀ metallicꢀ oxidesꢀ (suchꢀ asꢀ ReO
x
,ꢀ MoO
x
,ꢀ VO ꢀ orꢀ
x
WO )‐incorporatedꢀ Rhꢀ [16,18,19,21]ꢀ andꢀ Irꢀ [22–24,26]ꢀ cata‐
x
Allꢀtheꢀreagentsꢀwereꢀofꢀanalyticalꢀgradeꢀandꢀdirectlyꢀusedꢀ
lystsꢀattractedꢀmuchꢀattentionꢀforꢀtheꢀhydrogenolysisꢀofꢀTHFAꢀ
toꢀsynthesizeꢀ1,5‐PeD.ꢀTomishige’sꢀgroupꢀ[19–25]ꢀledꢀpioneer‐
ingꢀworkꢀonꢀsuchꢀreactionsꢀandꢀachievedꢀaꢀhighꢀ1,5‐PeDꢀyieldꢀofꢀ
withoutꢀ furtherꢀ pretreatment.ꢀ H
wereꢀ purchasedꢀ fromꢀ Shaanxiꢀ Rockꢀ Newꢀ Materialsꢀ Co.,ꢀ Ltd.,ꢀ
China.ꢀ PdCl ꢀ andꢀ RhCl ·3H Oꢀ wereꢀ purchasedꢀ fromꢀ Beijingꢀ
HWRKꢀChemꢀCo.,ꢀLtd.,ꢀChina.ꢀCuCr ꢀcatalystꢀwasꢀpurchasedꢀ
fromꢀ Yingkouꢀ Tianyuanꢀ Chemicalꢀ Industryꢀ Researchꢀ Instituteꢀ
Co.,ꢀLtd.,ꢀChina.ꢀFFAꢀandꢀTHFAꢀwereꢀpurchasedꢀfromꢀAlfaꢀAesar.ꢀ
2
PtCl
6
·6H
2
Oꢀ andꢀ RuCl
3
·3H
2
Oꢀ
2
3
2
9
4.0%ꢀatꢀ373ꢀtoꢀ393ꢀKꢀandꢀ8ꢀMPaꢀH
suchꢀ asꢀ MnO ,ꢀ CeO ,ꢀ andꢀ hydrotalciteꢀ (HT)‐supportedꢀ nobleꢀ
metalsꢀ ofꢀ Ruꢀ [13]ꢀ andꢀ Ptꢀ [12,14]ꢀ wereꢀ appliedꢀ toꢀ convertꢀ
FA/FFAꢀtoꢀ1,2‐PeDꢀatꢀ423–443ꢀKꢀandꢀ1–2ꢀMPaꢀH ;ꢀ1,2‐PeDꢀwithꢀ
2
.ꢀFurther,ꢀbasicꢀsupports,ꢀ
O
2 4
x
2
2
5%H
wereꢀ obtainedꢀ fromꢀ Lanzhouꢀ Lanmeiꢀ Cryogenicꢀ Productsꢀ Co.,ꢀ
Ltd.,ꢀChina.ꢀ CO ꢀ(99.99%)ꢀwasꢀ obtainedꢀ fromꢀ Lanzhouꢀ Hongliꢀ
2
‐95%Ar,ꢀ5%H
2
‐95%N
2
,ꢀH ꢀ(99.999%),ꢀandꢀHeꢀ(99.999%)ꢀ
2
aꢀyieldꢀofꢀupꢀtoꢀ80%ꢀwasꢀachievedꢀoverꢀaꢀPt/HTꢀcatalystꢀatꢀaꢀ
highꢀPtꢀtoꢀFFAꢀmolarꢀratioꢀofꢀ1/1.ꢀInꢀaddition,ꢀthereꢀareꢀaꢀfewꢀ
reportsꢀ onꢀ theꢀ useꢀ ofꢀ non‐preciousꢀ metals,ꢀ suchꢀ asꢀ Cuꢀ
2
gasꢀCo.,ꢀLtd.,ꢀChina.ꢀ
[8,27–29],ꢀ Coꢀ [10,30],ꢀ andꢀ Ni‐basedꢀ [11,31]ꢀ catalystsꢀ inꢀ theꢀ
hydrogenolysisꢀofꢀFFAꢀandꢀTHFAꢀtoꢀPeDs.ꢀGenerally,ꢀCu‐basedꢀ
catalystsꢀ exhibitꢀ higherꢀ selectivityꢀ towardꢀ 1,2‐PeD,ꢀ whileꢀ Coꢀ
andꢀ Ni‐basedꢀ catalystsꢀ preferꢀ theꢀ generationꢀ ofꢀ 1,5‐PeD.ꢀ Forꢀ
2.2.ꢀ ꢀ Catalystꢀpreparationꢀ
Inꢀthisꢀstudy,ꢀaꢀseriesꢀofꢀxCuO‐LaCoO
3
ꢀ(x/%ꢀ=ꢀ0,ꢀ2,ꢀ5,ꢀ10,ꢀ15,ꢀ
instance,ꢀCu‐Mg
FFAꢀtoꢀ1,2‐PeDꢀandꢀ1,5‐PeDꢀwithꢀyieldsꢀofꢀ51.2%ꢀandꢀ28.8%,ꢀ
respectively,ꢀatꢀ413ꢀKꢀandꢀ6ꢀMPaꢀH ,ꢀwhileꢀaꢀNi‐Y ꢀ[11]ꢀcom‐
positeꢀ catalystꢀ selectivelyꢀ hydrogenolyzedꢀ FFAꢀ toꢀ 1,5‐PeDꢀ
41.9%)ꢀ ratherꢀ thanꢀ 1,2‐PeDꢀ (1.2%)ꢀ atꢀ 423ꢀ Kꢀ andꢀ 2ꢀ MPaꢀ H .ꢀ
Clearly,ꢀstudiesꢀonꢀFAꢀandꢀitsꢀderivativesꢀsubjectedꢀtoꢀhydro‐
genolysisꢀ stillꢀ encounterꢀ lowꢀ activitiesꢀ orꢀ selectivitiesꢀ towardꢀ
targetꢀPeDs,ꢀorꢀtheꢀuseꢀofꢀnobleꢀmetalꢀcatalysts.ꢀItꢀisꢀthusꢀurgentꢀ
toꢀdevelopꢀeffectiveꢀandꢀenvironmentallyꢀbenignꢀnon‐preciousꢀ
metalꢀcatalystsꢀforꢀtheꢀefficientꢀconversionꢀofꢀFAꢀandꢀitsꢀderiva‐
tivesꢀintoꢀvalue‐addedꢀPeDs.ꢀHowever,ꢀtheꢀdevelopmentꢀofꢀef‐
fectiveꢀmethodsꢀ forꢀtuningꢀtheꢀ chemoselectivityꢀofꢀFAꢀandꢀitsꢀ
derivativesꢀforꢀhydrogenolysisꢀisꢀaꢀgreatꢀchallengeꢀatꢀpresent.ꢀ
Recently,ꢀ owingꢀ toꢀ theꢀ flexibilityꢀ ofꢀ theirꢀ electronicꢀ andꢀ
crystalꢀ structureꢀ asꢀ wellꢀ asꢀ theirꢀ chemicalꢀ versatility,ꢀ perov‐
3
AlO4.5ꢀ[8]ꢀwithꢀaꢀbasicꢀsupportꢀhydrogenolyzedꢀ
andꢀ20)ꢀperovskite‐typeꢀmixedꢀoxidesꢀwereꢀsynthesizedꢀbyꢀtheꢀ
citric‐complexingꢀmethodꢀ[32].ꢀTheꢀdesiredꢀamountsꢀofꢀcopper,ꢀ
cobalt,ꢀandꢀlanthanumꢀnitratesꢀwithꢀaꢀLa/Coꢀmolarꢀratioꢀofꢀ1/1ꢀ
wereꢀ dissolvedꢀ inꢀ deionizedꢀ (DI)ꢀ waterꢀ atꢀ aꢀ concentrationꢀ ofꢀ
Cu2+ꢀ=ꢀ0.1ꢀmol/L.ꢀLater,ꢀcitricꢀacidꢀ(20%ꢀexcessꢀoverꢀtheꢀtotalꢀ
molarꢀ contentꢀ ofꢀ metalꢀ cations)ꢀ andꢀ polyethyleneꢀ glycolꢀ 400ꢀ
(24%ꢀmolarꢀamountꢀofꢀcitricꢀacid)ꢀwereꢀaddedꢀtoꢀtheꢀsolution.ꢀ
Theꢀsolutionꢀwasꢀstirredꢀatꢀroomꢀtemperatureꢀforꢀ6ꢀhꢀandꢀthenꢀ
stirredꢀ atꢀ 353ꢀ Kꢀ untilꢀ aꢀ foamyꢀ solidꢀ wasꢀ formed.ꢀ Theꢀ foamyꢀ
solidꢀwasꢀdriedꢀatꢀ383ꢀKꢀforꢀ12ꢀhꢀandꢀthenꢀcalcinedꢀatꢀ973ꢀKꢀforꢀ
2ꢀhꢀinꢀstaticꢀairꢀwithꢀaꢀtemperatureꢀrampingꢀrateꢀofꢀ3ꢀK/min.ꢀ
2
O
2 3
(
2
Theꢀ calcinedꢀ samplesꢀ wereꢀ markedꢀ asꢀ xCuO‐LaCoO
representsꢀtheꢀnominalꢀCuꢀloading.ꢀTheꢀcorrespondingꢀreducedꢀ
catalystsꢀwereꢀlabeledꢀasꢀxCu‐LaCoO .ꢀLaCoO ‐supportedꢀPt,ꢀRu,ꢀ
3
,ꢀ whereꢀ xꢀ
3
3
Pd,ꢀandꢀRhꢀcatalystsꢀwithꢀindividualꢀnominalꢀloadingꢀofꢀ5ꢀwt%ꢀ
wereꢀalsoꢀpreparedꢀbyꢀaꢀsimilarꢀmethod.ꢀ ꢀ
skite‐typeꢀ oxidesꢀ withꢀ aꢀ generalꢀ ABO
ꢀ structureꢀ haveꢀ beenꢀ
3
studiedꢀ forꢀ catalysisꢀ applicationsꢀ [32–37].ꢀ Usingꢀ perov‐
skite‐typeꢀoxidesꢀasꢀprecursorsꢀtoꢀstabilizeꢀmetalꢀparticlesꢀonꢀ
mixedꢀoxidesꢀisꢀalsoꢀanꢀattractiveꢀoptionꢀtoꢀproduceꢀactiveꢀandꢀ
stableꢀcatalystsꢀ[32,33].ꢀInꢀaddition,ꢀtheꢀspecificꢀsurfaceꢀacidi‐
ty/basicityꢀ ofꢀ perovskite‐typeꢀ oxidesꢀ alsoꢀ contributesꢀ toꢀ theꢀ
novelꢀ performanceꢀ ofꢀ theꢀ catalystsꢀ [36,37].ꢀ Takingꢀ theseꢀ ad‐
vantagesꢀ ofꢀ perovskite‐basedꢀ catalystsꢀ andꢀ theꢀ highꢀ C–Oꢀ hy‐
drogenation/hydrogenolysisꢀactivityꢀofꢀtheꢀnon‐nobleꢀmetalꢀCuꢀ
2.3.ꢀ ꢀ Catalystꢀcharacterizationꢀ ꢀ
X‐rayꢀ diffractionꢀ (XRD)ꢀ experimentsꢀ wereꢀ performedꢀ onꢀ aꢀ
PANalyticalꢀX’pertꢀProꢀDiffractometerꢀwithꢀnickel‐filteredꢀCuꢀK ꢀ
radiationꢀ(λꢀ=ꢀ1.05406ꢀnm)ꢀatꢀ40ꢀkVꢀandꢀ40ꢀmA.ꢀTheꢀpatternsꢀ
wereꢀcollectedꢀinꢀtheꢀ2θꢀrangeꢀofꢀ10°–80°ꢀatꢀaꢀscanningꢀspeedꢀofꢀ
10°/min.ꢀ
α
[
8,38,39]ꢀ intoꢀ consideration,ꢀ aꢀ seriesꢀ ofꢀ Cu‐LaCoO
3
ꢀ perov‐
TheꢀBrunauer‐Emmett‐Tellerꢀ(BET)ꢀsurfaceꢀareaꢀofꢀtheꢀcat‐
skite‐typeꢀ mixedꢀ oxidesꢀ withꢀ differentꢀ Cuꢀ loadingsꢀ wereꢀ syn‐
thesizedꢀinꢀtheꢀcurrentꢀstudyꢀandꢀinvestigatedꢀforꢀtheꢀselectiveꢀ
alystsꢀ wasꢀ determinedꢀ usingꢀ theꢀ N ꢀ adsorption‐desorptionꢀ
2
methodꢀandꢀtheꢀexperimentsꢀwereꢀconductedꢀonꢀaꢀMicromerit‐
icsꢀTristarꢀIIꢀ3020ꢀinstrumentꢀatꢀliquidꢀnitrogenꢀtemperaturesꢀ
(77ꢀ K).ꢀ Priorꢀ toꢀ theseꢀ measurements,ꢀ theꢀ samplesꢀ wereꢀ pre‐
hydrogenolysisꢀ ofꢀ FFAꢀ toꢀ value‐addedꢀ PeDs.ꢀ Theꢀ Cu‐LaCoO
3
ꢀ
catalystsꢀwereꢀfoundꢀtoꢀexhibitꢀaꢀhigherꢀselectivityꢀforꢀ1,5‐PeDꢀ

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1713ꢀ
treatedꢀwithꢀN
2
ꢀatꢀ363ꢀKꢀforꢀ2ꢀhꢀandꢀthenꢀmaintainedꢀforꢀ4ꢀhꢀatꢀ
Agilentꢀ 7890A/5975Cꢀ gasꢀ chromatograph‐massꢀ spectrometerꢀ
(GC‐MS)ꢀ withꢀ anꢀ HP‐5MSꢀ column.ꢀ Theꢀ reactantꢀ andꢀ liquidꢀ
productsꢀ wereꢀ analyzedꢀ byꢀ gasꢀ chromatographyꢀ (Agilentꢀ
7890AꢀGC)ꢀwithꢀaꢀPONAꢀcapillaryꢀcolumnꢀ(50ꢀmꢀ×ꢀ0.20ꢀmmꢀ×ꢀ
0.50ꢀμm)ꢀandꢀaꢀflameꢀionizationꢀdetectorꢀ(FID).ꢀConversionꢀandꢀ
productꢀ selectivityꢀ wereꢀ determinedꢀ byꢀ anꢀ internalꢀ standardꢀ
5
73ꢀK.ꢀ
Theꢀreducibilityꢀandꢀsurfaceꢀacidity/basicityꢀofꢀtheꢀcalcinedꢀ
samplesꢀ wereꢀ determinedꢀ byꢀ H ‐temperature‐programmedꢀ
reductionꢀ (TPR)ꢀ andꢀ NH /CO ‐Temperatureꢀ programmedꢀ de‐
sorptionꢀ(TPD)ꢀmeasurements,ꢀrespectively,ꢀwithꢀaꢀDAX‐7000ꢀ
instrumentꢀ (Huasiꢀ Technologyꢀ Co.,ꢀ Ltd,ꢀ China).ꢀ Forꢀ H ‐TPRꢀ
2
3
2
2
methodꢀandꢀcalculatedꢀasꢀfollows:ꢀ
initial moles of FFA  moles of FFA left
experiments,ꢀ~50ꢀmgꢀofꢀtheꢀsamplesꢀwasꢀplacedꢀinꢀaꢀquartzꢀcellꢀ
andꢀpretreatedꢀatꢀ473ꢀKꢀunderꢀHeꢀflowꢀforꢀ1ꢀh.ꢀAfterꢀcoolingꢀtoꢀ
Conversion(%) =
 100%
 100%
initial moles of FFA
moles of a product generated
ꢀ
ꢀ
3
03ꢀ K,ꢀ theꢀ samplesꢀ wereꢀ reducedꢀ inꢀ 5%H
mL/min)ꢀandꢀtheꢀtemperatureꢀwasꢀincreasedꢀlinearlyꢀtoꢀ1073ꢀKꢀ
atꢀaꢀrampingꢀrateꢀofꢀ10ꢀK/min.ꢀH ꢀconsumptionꢀwasꢀmonitoredꢀ
byꢀ thermalꢀ conductivityꢀ detectorꢀ (TCD).ꢀ Forꢀ CO ‐TPDꢀ experi‐
2
‐95%Arꢀ flowꢀ (40ꢀ
Selectivity(%) =
initial moles of FFA  moles of FFA left
2
3
.ꢀ ꢀ Resultsꢀandꢀdiscussionꢀ
2
ments,ꢀ~0.2ꢀgꢀofꢀtheꢀsamplesꢀwasꢀplacedꢀinꢀaꢀquartzꢀcellꢀandꢀ
pretreatedꢀatꢀ473ꢀKꢀunderꢀHeꢀflowꢀforꢀ1ꢀhꢀandꢀthenꢀreducedꢀatꢀ
3
.1.ꢀ ꢀ StructuralꢀcharacterizationꢀofꢀCu‐LaCoO ꢀcatalystsꢀ
3
5
73ꢀKꢀinꢀ5%H
2
‐95%Arꢀflowꢀ(40ꢀmL/min)ꢀforꢀ2ꢀh.ꢀAfterꢀcoolingꢀ
Fig.ꢀ 1ꢀ showsꢀ theꢀ XRDꢀ patternsꢀ ofꢀ calcinedꢀ xCuO‐LaCoO
samplesꢀ withꢀ differentꢀ Cuꢀ loadings.ꢀ Strongꢀ peaksꢀ dueꢀ toꢀ theꢀ
formationꢀ ofꢀ LaCoO ꢀ perovskite‐typeꢀ oxidesꢀ wereꢀ seenꢀ inꢀ theꢀ
patternꢀofꢀtheꢀLaCoO ꢀsupport.ꢀAdditionally,ꢀsegregatedꢀphasesꢀ
dueꢀ toꢀ theꢀ formationꢀ ofꢀ La CO ꢀ (diffractionꢀ peaksꢀ atꢀ 22.3°,ꢀ
5.8°,ꢀ30.4°,ꢀ44.4°,ꢀandꢀ47.4°)ꢀandꢀCo ꢀ(atꢀ36.8ꢀ°)ꢀwereꢀalsoꢀ
observed.ꢀLa CO ꢀwasꢀprobablyꢀformedꢀbyꢀCO ‐inducedꢀde‐
compositionꢀofꢀLaCoO ꢀtoꢀLa ꢀandꢀCo ꢀandꢀtheꢀfurtherꢀre‐
actionꢀ ofꢀ La ꢀ andꢀ CO ꢀ inꢀ airꢀ duringꢀ theꢀ calcinationꢀ processꢀ
33].ꢀTheꢀincorporationꢀofꢀCu,ꢀevenꢀatꢀanꢀamountꢀasꢀsmallꢀasꢀ2ꢀ
wt%,ꢀresultedꢀinꢀtheꢀdecompositionꢀofꢀtheꢀLaCoO ꢀperovskiteꢀ
3
ꢀ
toꢀ313ꢀK,ꢀtheꢀsamplesꢀwereꢀexposedꢀtoꢀCO
andꢀ maintainedꢀ forꢀ 1ꢀ h.ꢀ Subsequently,ꢀ theꢀ temperatureꢀ wasꢀ
increasedꢀlinearlyꢀtoꢀ1173ꢀKꢀatꢀaꢀrampingꢀrateꢀofꢀ5ꢀK/minꢀun‐
2
ꢀflowꢀ(40ꢀmL/min)ꢀ
3
3
derꢀ Heꢀ flow;ꢀ theꢀ desorbedꢀ CO
NH ‐TPDꢀofꢀtheꢀ10ꢀwt%ꢀCu‐LaCoO
CO ‐TPDꢀprocedureꢀaboveꢀbutꢀwithꢀtheꢀsaturationꢀadsorptionꢀ
ofꢀNH ꢀatꢀ373ꢀKꢀforꢀ1ꢀh.ꢀ
Transmissionꢀ electronꢀ microscopyꢀ (TEM)ꢀ testsꢀ wereꢀ per‐
formedꢀonꢀaꢀTECNAIꢀG2ꢀTF20ꢀinstrumentꢀatꢀ200ꢀkV.ꢀTheꢀTEMꢀ
samplesꢀ wereꢀ preparedꢀ byꢀ ultrasonicꢀ dispersionꢀ inꢀ ethanol.ꢀ
Afterꢀultrasonicꢀdispersionꢀofꢀtheꢀcatalysts,ꢀtheꢀsamplesꢀwereꢀ
depositedꢀonꢀcopperꢀgridsꢀwithꢀaꢀporousꢀcarbonꢀfilmꢀsupport.ꢀ
X‐rayꢀphotoelectronꢀspectraꢀ(XPS)ꢀmeasurementsꢀwereꢀcar‐
riedꢀoutꢀonꢀanꢀESCALAB250xiꢀspectrometerꢀequippedꢀwithꢀanꢀ
2
ꢀ wasꢀ monitoredꢀ byꢀ TCD.ꢀ
2
O
2
3
3
3
ꢀcatalystꢀwasꢀsimilarꢀtoꢀtheꢀ
2
3 4
O
2
2
O
2
3
2
3
3
2
O
3
3 4
O
2
O
3
2
[
3

Co O JCPDS 74-2120
3 4
 CuO JCPDS 80-1268
Alꢀ K
α
ꢀ X‐rayꢀ radiationꢀ sourceꢀ (hυꢀ =ꢀ 1486.6ꢀ eV).ꢀ Theꢀ sampleꢀ
20CuO-LaCoO


3
bindingꢀenergyꢀcanꢀbeꢀcalibratedꢀwithꢀCꢀ1sꢀ(E
b
ꢀ=ꢀ284.8ꢀeV)ꢀpeakꢀ
asꢀtheꢀinternalꢀstandard,ꢀtheꢀerrorꢀisꢀaboutꢀ±ꢀ0.2.ꢀ
Theꢀchemicalꢀcompositionꢀofꢀtheꢀcatalystsꢀwasꢀdeterminedꢀ
byꢀ X‐rayꢀ fluorescenceꢀ (XRF)ꢀ onꢀ aꢀ PANalyticalꢀ MagixPW2403ꢀ
instrument.ꢀ
1
5CuO-LaCoO
3
Carbonꢀ depositionꢀandꢀ stabilityꢀofꢀtheꢀusedꢀ catalystsꢀwereꢀ
determinedꢀ onꢀ aꢀ NETZSCHꢀ STA449F3ꢀ thermogravime‐
try‐differentialꢀ scanningꢀ calorimetryꢀ (TG‐DSC)ꢀ instrument.ꢀ
Initially,ꢀ10ꢀmgꢀofꢀtheꢀsamplesꢀwereꢀplacedꢀinꢀanꢀaluminaꢀcruci‐
bleꢀandꢀheatedꢀtoꢀ1073ꢀKꢀatꢀaꢀheatingꢀrateꢀofꢀ10ꢀK/minꢀinꢀaꢀni‐
trogenꢀatmosphere.ꢀ
1
0CuO-LaCoO


3
3
3


5
CuO-LaCoO
2
CuO-LaCoO
2.4.ꢀ ꢀ Catalyticꢀhydrogenolysisꢀofꢀfurfurylꢀalcoholꢀ
TheꢀselectiveꢀhydrogenolysisꢀofꢀFFAꢀwasꢀcarriedꢀoutꢀinꢀaꢀ100ꢀ
LaCoO
3
mLꢀstainlessꢀsteelꢀautoclaveꢀatꢀaꢀstirringꢀspeedꢀofꢀ800ꢀr/min.ꢀAllꢀ
theꢀcalcinedꢀ samplesꢀ wereꢀusedꢀinꢀtheꢀ powderꢀform.ꢀPriorꢀtoꢀ
eachꢀ test,ꢀ theꢀ calcinedꢀ samplesꢀ withꢀ aꢀ granuleꢀ sizeꢀ ofꢀ 60–80ꢀ
La O CO JCPDS 84-1963
2
2
3
meshꢀwereꢀpre‐reducedꢀinꢀ5%H
73ꢀKꢀforꢀ3ꢀh.ꢀInꢀaꢀtypicalꢀrun,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanolꢀ
solutionꢀ togetherꢀ withꢀ theꢀ pre‐reducedꢀ catalystꢀ wereꢀ intro‐
ducedꢀintoꢀtheꢀautoclave.ꢀAfterꢀpurgingꢀthriceꢀwithꢀH ,ꢀtheꢀre‐
actorꢀwasꢀpressurizedꢀtoꢀ6ꢀMPaꢀandꢀheatedꢀtoꢀ413ꢀKꢀtoꢀstartꢀtheꢀ
reaction.ꢀForꢀcomparison,ꢀtheꢀhydrogenolysisꢀofꢀTHFAꢀwasꢀalsoꢀ
2
‐95%N ꢀflowꢀ(40ꢀmL/min)ꢀatꢀ
2
5
LaCoO3 JCPDS 84-0848
2
10
20
30
40
50

60
70
80
2
studiedꢀoverꢀtheꢀ10Cu‐LaCoO
3
ꢀcatalystꢀusingꢀsimilarꢀconditions.ꢀ
Fig.ꢀ1.ꢀXRDꢀpatternsꢀofꢀcalcinedꢀxCuO‐LaCoO ꢀcatalystsꢀwithꢀdifferentꢀCuꢀ
loadings.ꢀ
3
Afterꢀcentrifugation,ꢀtheꢀproductsꢀwereꢀidentifiedꢀusingꢀanꢀ

1714ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
structureꢀ toꢀ La
peaksꢀassociatedꢀwithꢀtheꢀLaCoO
atꢀ5ꢀwt%ꢀCu.ꢀFurthermore,ꢀtheꢀdiffractionꢀpeaksꢀofꢀCo
therꢀintensifiedꢀinꢀtheꢀpatternꢀofꢀ5CuO‐LaCoO .ꢀNoꢀclearꢀdiffrac‐
2
O
2
CO
3
ꢀ andꢀ Co
3
O
4
ꢀ andꢀ almostꢀ noꢀ diffractionꢀ
ꢀperovskiteꢀcouldꢀbeꢀobservedꢀ
ꢀfur‐
(PDF#80‐1268)ꢀ andꢀ Co
showingꢀthatꢀtheꢀcatalystꢀwasꢀhardlyꢀreducedꢀunderꢀtheseꢀcon‐
ditions.ꢀTheseꢀfindingsꢀalsoꢀsuggestꢀthatꢀtheꢀcatalystꢀexhibitedꢀaꢀ
3
O ꢀ (PDF#74‐2120)ꢀ stillꢀ remained,ꢀ
4
3
O
3 4
3
highꢀ reducibilityꢀ inꢀ pureꢀ H
2
.ꢀ Asꢀ forꢀ theꢀ catalystꢀ reducedꢀ inꢀ
tionꢀpeaksꢀcorrespondingꢀtoꢀCuOꢀcouldꢀbeꢀseenꢀinꢀtheꢀsamplesꢀ
withꢀCuꢀloadingꢀbelowꢀ5ꢀwt%,ꢀwhichꢀisꢀprobablyꢀdueꢀtoꢀtheꢀhighꢀ
dispersionꢀofꢀCuꢀinꢀtheseꢀsamples.ꢀIncreasingꢀtheꢀCuꢀloadingꢀtoꢀ
5%H
2
‐95%N
2
ꢀ andꢀ usedꢀ forꢀ oneꢀ time,ꢀ theꢀ diffractionꢀ peaksꢀ ofꢀ
0
Cu ꢀappearedꢀinꢀitsꢀXRDꢀpattern;ꢀnoꢀCuOꢀcouldꢀbeꢀobservedꢀandꢀ
theꢀintensityꢀofꢀtheꢀdiffractionꢀpeaksꢀofꢀCo
2(4)),ꢀindicatingꢀthatꢀCuOꢀwasꢀreducedꢀtoꢀCu ꢀandꢀCo
largelyꢀreducedꢀtoꢀCo ꢀand/orꢀCoOꢀwithꢀhighꢀdispersion.ꢀThisꢀ
findingꢀalsoꢀsuggestsꢀthatꢀtheꢀreactionꢀmediaꢀofꢀFFAꢀinꢀalcoholꢀ
hasꢀ aꢀ strongerꢀ reducibilityꢀ thanꢀ 5%H
diffractionꢀpeaksꢀofꢀCo ꢀalmostꢀdisappearedꢀafterꢀtheꢀcatalystꢀ
wasꢀusedꢀforꢀ4ꢀcyclesꢀ(Fig.ꢀ2(5)),ꢀindicatingꢀthatꢀCo
tinuouslyꢀ reducedꢀ toꢀ Co ꢀ and/orꢀ CoOꢀ duringꢀ theꢀ repeatꢀ reac‐
tions.ꢀTheꢀ reductionꢀofꢀ CoO,ꢀwhichꢀisꢀmoreꢀdifficultꢀtoꢀbeꢀre‐
ducedꢀthanꢀCo
drogenolysisꢀ[41].ꢀTheꢀintensityꢀofꢀtheꢀdiffractionꢀpeaksꢀofꢀCu0ꢀ
asꢀwellꢀasꢀthoseꢀofꢀLa CO ꢀ(Fig.ꢀ2(5))ꢀdecreasedꢀafterꢀtheꢀcat‐
alystꢀwasꢀusedꢀfourꢀtimes.ꢀSuchꢀobviousꢀstructuralꢀchangesꢀinꢀ
theꢀcatalystꢀduringꢀrepeatedꢀrunsꢀmayꢀbeꢀrelatedꢀtoꢀtheꢀchangesꢀ
inꢀitsꢀcatalyticꢀperformance,ꢀasꢀwillꢀbeꢀdiscussedꢀlater.ꢀ ꢀ
3
O ꢀdecreasedꢀ(Fig.ꢀ
4
0
1
0ꢀwt%ꢀledꢀtoꢀtheꢀappearanceꢀofꢀdiffractionꢀpeaksꢀcorrespond‐
ingꢀnotꢀonlyꢀtoꢀtheꢀLaCoO ꢀperovskite,ꢀbutꢀalsoꢀCuO,ꢀwithꢀaꢀsim‐
ultaneousꢀdecreaseꢀinꢀtheꢀCo ꢀdiffractionꢀpeaks.ꢀTheꢀpeaksꢀofꢀ
theꢀLaCoO ꢀperovskiteꢀandꢀCuOꢀfurtherꢀintensifiedꢀwithꢀanꢀin‐
creaseꢀinꢀtheꢀCuꢀloadingꢀtoꢀ20ꢀwt%,ꢀwhileꢀtheꢀpeaksꢀassignableꢀ
toꢀCo ꢀalmostꢀdisappearedꢀinꢀtheꢀpatternꢀofꢀ20CuO‐LaCoO .ꢀItꢀ
seemsꢀthatꢀtheꢀincorporationꢀofꢀsmallꢀamountsꢀ((2–10)ꢀwt%)ꢀofꢀ
Cuꢀ causedꢀ theꢀ segregationꢀ ofꢀ LaCoO ,ꢀ whichꢀ isꢀ similarꢀ toꢀ theꢀ
effectꢀobservedꢀwhenꢀZnꢀwasꢀincorporatedꢀinꢀLaCoO ꢀ[40].ꢀTheꢀ
3
O ꢀwasꢀ
4
0
3
O
3 4
3
2 2
‐95%N ꢀ atꢀ 573ꢀ K.ꢀ Theꢀ
O
3 4
O
3 4
3
3
O ꢀwasꢀcon‐
4
0
3
3
3 4
O ꢀ[10,41],ꢀisꢀalsoꢀobservedꢀduringꢀglycerolꢀhy‐
gradualꢀincreaseꢀinꢀtheꢀintensityꢀofꢀCuOꢀpeaksꢀwithꢀanꢀincreaseꢀ
inꢀtheꢀCuꢀloadingꢀcouldꢀbeꢀascribedꢀtoꢀtheꢀaggregationꢀofꢀCuOꢀ
particles.ꢀ
O
2 2
3
3
Fig.ꢀ2ꢀpresentsꢀtheꢀXRDꢀpatternsꢀofꢀ10Cu‐LaCoO ꢀsamplesꢀatꢀ
differentꢀstates.ꢀAfterꢀreductionꢀinꢀpureꢀH ꢀatꢀ573ꢀK,ꢀtheꢀdiffrac‐
2
tionꢀ peaksꢀ ofꢀ CuOꢀ disappeared;ꢀ meanwhile,ꢀ theꢀ diffractionꢀ
peaksꢀ correspondingꢀ toꢀ cubicꢀ Cu ꢀ (PDF#04‐0836)ꢀ appearedꢀ
Fig.ꢀ3ꢀshowsꢀtheꢀXPSꢀsurveyꢀofꢀtheꢀCuꢀandꢀCoꢀspecies.ꢀAsꢀcanꢀ
beꢀseenꢀinꢀFig.ꢀ3(a),ꢀtheꢀbindingꢀenergyꢀofꢀCuꢀ2p3/2ꢀatꢀ933.5ꢀeVꢀ
alongꢀwithꢀaꢀsatelliteꢀpeakꢀinꢀtheꢀrangeꢀofꢀ940–945ꢀeVꢀindicateꢀ
0
(
Fig.ꢀ2(2)),ꢀsuggestingꢀthatꢀCuOꢀinꢀtheꢀsampleꢀwasꢀreducedꢀtoꢀ
0
2+
Cu .ꢀTheꢀdisappearanceꢀofꢀCo
3 4
O ꢀdiffractionꢀpeaksꢀsuggestsꢀthatꢀ
theꢀexistenceꢀofꢀCu ꢀinꢀtheꢀcalcinedꢀsampleꢀ[42].ꢀTheꢀdecreaseꢀ
Co
3
O
4
ꢀ wasꢀ reducedꢀ toꢀ Coꢀ speciesꢀ withꢀ lowꢀ valencies,ꢀ i.e.,ꢀ Co0ꢀ
inꢀtheꢀbindingꢀenergyꢀofꢀCuꢀ2p3/2ꢀtoꢀ932.5ꢀeVꢀandꢀtheꢀweaken‐
and/orꢀCoO.ꢀNonetheless,ꢀnoꢀdiffractionꢀpeaksꢀassociatedꢀwithꢀ
Co ꢀ orꢀ CoOꢀ couldꢀ beꢀ observed,ꢀ whichꢀ isꢀ probablyꢀ dueꢀ toꢀ theꢀ
ingꢀorꢀevenꢀdisappearanceꢀofꢀtheꢀsatelliteꢀpeakꢀsuggestsꢀthatꢀCuꢀ
0
0
speciesꢀwereꢀreducedꢀtoꢀCuꢀwithꢀlowꢀvalenciesꢀ(Cu ꢀorꢀCu
2
O)ꢀinꢀ
highꢀ dispersionꢀ ofꢀ theseꢀ species.ꢀ Afterꢀ reductionꢀ inꢀ
bothꢀreducedꢀsamplesꢀasꢀwellꢀasꢀtheꢀsamplesꢀusedꢀforꢀoneꢀtime.ꢀ
TheꢀsimulationꢀofꢀCuꢀ2pꢀ(Fig.ꢀ3(a)–(2))ꢀforꢀtheꢀreducedꢀcatalystꢀ
confirmedꢀtheꢀcoexistenceꢀofꢀlargeꢀamountsꢀofꢀCuꢀspeciesꢀwithꢀ
lowꢀvalencies.ꢀTheꢀdecreaseꢀinꢀtheꢀintensityꢀofꢀtheꢀCuꢀ2pꢀspectraꢀ
withꢀaꢀhighꢀbindingꢀenergyꢀcorrespondingꢀtoꢀtheꢀusedꢀcatalystꢀ
5
%H
2
‐95%N
2
ꢀ atꢀ theꢀ sameꢀ temperature,ꢀ however,ꢀ almostꢀ noꢀ
0
diffractionꢀ peaksꢀ associatedꢀ withꢀ Cu ꢀ and/orꢀ Cu
2
Oꢀ wereꢀ ob‐
servedꢀ (Fig.ꢀ 2(3)),ꢀ whileꢀ theꢀ diffractionꢀ peaksꢀ ofꢀ CuOꢀ
2+
indicatedꢀaꢀdecreaseꢀinꢀtheꢀCu ꢀamountꢀ[42].ꢀAsꢀforꢀtheꢀCoꢀ2pꢀ
spectrumꢀofꢀtheꢀcalcinedꢀsample,ꢀtheꢀasymmetricꢀpeakꢀatꢀ780.5ꢀ
eVꢀcorrespondingꢀtoꢀCoꢀ2p3/2ꢀandꢀtheꢀpresenceꢀofꢀaꢀweakꢀpeakꢀ



Co O JCPDS 74-2120
3 4
(5)
CuO JCPDS 80-1268
Cu
JCPDS 04-0836
3+
ofꢀCoꢀ2p1/2ꢀatꢀaboutꢀ796.6ꢀeVꢀareꢀcharacteristicꢀofꢀCo ꢀ[43].ꢀTheꢀ
peaksꢀofꢀCoꢀ2p3/2ꢀatꢀ782.3ꢀeVꢀandꢀtheꢀweakꢀpeakꢀatꢀ798.5ꢀeVꢀ
(4)


2+
wereꢀassociatedꢀwithꢀCo .ꢀTheꢀsimulationꢀofꢀCoꢀ2pꢀ((b)–(2))ꢀforꢀ
3+
2+
theꢀreducedꢀcatalystꢀconfirmedꢀtheꢀcoexistenceꢀofꢀCo ꢀandꢀCo ꢀ
speciesꢀ[43].ꢀFromꢀFig.ꢀ3(b),ꢀweꢀcanꢀseeꢀthatꢀafterꢀbeingꢀusedꢀ
once,ꢀtheꢀCoꢀ2pꢀsignalꢀweakenedꢀ(Fig.ꢀ3(b)–(3))ꢀandꢀexhibitedꢀaꢀ
tendencyꢀtoꢀshiftꢀtowardsꢀlowerꢀbindingꢀenergies,ꢀindicatingꢀanꢀ
increaseꢀinꢀlow‐stateꢀCoꢀspecies.ꢀTheseꢀfindingsꢀareꢀinꢀlineꢀwithꢀ
theꢀresultsꢀofꢀXRDꢀcharacterizationꢀshownꢀinꢀFig.ꢀ2.ꢀ
(3)


(2)
Fig.ꢀ4ꢀshowsꢀtheꢀTEMꢀimagesꢀofꢀcalcinedꢀCuO‐LaCoO
plesꢀwithꢀCuꢀloadingsꢀofꢀ(2,ꢀ10,ꢀandꢀ20)ꢀwt%ꢀasꢀwellꢀasꢀthoseꢀofꢀ
reducedꢀandꢀusedꢀ10Cu‐LaCoO ꢀcatalysts.ꢀTheꢀcalcinedꢀsamplesꢀ
3
ꢀsam‐
(1)




3
La O CO JCPDS 84-1963
generallyꢀ exhibitedꢀ anꢀ amorphousꢀ structure,ꢀ whichꢀ isꢀ inꢀ lineꢀ
withꢀpreviousꢀfindingsꢀbyꢀPredoanaꢀetꢀal.ꢀ[44].ꢀItꢀisꢀdifficultꢀtoꢀ
2
2
3
discriminateꢀbetweenꢀCuOꢀnanoparticlesꢀandꢀtheꢀLaCoO ꢀsup‐
3
1
0
20
30
40
2
50
60
70
80
portꢀwhenꢀtheꢀCuꢀloadingꢀwasꢀlowꢀ(i.e.,ꢀ2ꢀwt%),ꢀwhichꢀindicatesꢀ
theꢀ uniformꢀ dispersionꢀ ofꢀ CuO.ꢀ Withꢀ furtherꢀ increaseꢀ inꢀ Cuꢀ
loading,ꢀ aꢀ smallꢀ amountꢀ ofꢀ aggregatedꢀ particlesꢀ withꢀ sizesꢀ

Fig.ꢀ 2.ꢀ XRDꢀ patternsꢀ ofꢀ 10Cu‐LaCoO
calcined,ꢀ(2)ꢀreducedꢀinꢀpureꢀH ,ꢀ(3)ꢀreducedꢀinꢀ5%H
ducedꢀinꢀ5%H ‐95%N ꢀandꢀusedꢀonce,ꢀandꢀ(5)ꢀreducedꢀinꢀ5%H
andꢀusedꢀfourꢀtimes.ꢀ
3
ꢀ catalystsꢀ atꢀ differentꢀ states.ꢀ (1)ꢀ
‐95%N ,ꢀ(4)ꢀre‐
‐95%N
aroundꢀ5–10ꢀnmꢀwereꢀobservedꢀinꢀ10CuO‐LaCoO ꢀandꢀtheꢀag‐
3
2
2
2
2
2
2
2
gregatedꢀparticlesꢀbecameꢀmoreꢀobviousꢀwhenꢀtheꢀCuꢀloadingꢀ
wasꢀ 20ꢀ wt%.ꢀ Suchꢀ aggregationꢀ couldꢀ beꢀ ascribedꢀ toꢀ CuO,ꢀ asꢀ

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1715ꢀ
(b)
(
a)
Cu 2p
（
3）
932.5
Co 2p
782.3
（3）
（2）
（2）
933.5
780.3
（1）
（1）
970 965 960 955 950 945 940 935 930 810
805
800
795
790
785
780
775
Binding energy (eV)
Binding energy (eV)
3
+
Cu 2p3/2
(b)-(2)
Co
796.6
Co 2p3/2
(a)-(2)
3
+
Co 2p1/2
Co
80.5
7
0
Cu
932.5
2
+
Cu 2p1/2
Co
82.3
7
2
+
Co
798.5
2
+
Cu
33.5
9
810
805
800
795
790
785
780
775
965
960
955
950
945
940
935
930
Binding energy (eV)
Binding energy (eV)
ꢀsamples,ꢀ(2)ꢀreducedꢀ10Cu‐LaCoO
Fig.ꢀ3.ꢀXPSꢀspectraꢀforꢀCuꢀ2pꢀ(a)ꢀandꢀCoꢀ2pꢀ(b)ꢀofꢀ(1)ꢀcalcinedꢀ10CuO‐LaCoO
reducedꢀ10Cu‐LaCoO ꢀsamplesꢀinꢀ5%H ‐95%N ꢀandꢀusedꢀonce.ꢀ
3
3 2 2
ꢀsamplesꢀinꢀ5%H ‐95%N ,ꢀandꢀ(3)ꢀ
3
2
2
evidencedꢀ fromꢀ XRDꢀ characterizationꢀ (Fig.ꢀ 1).ꢀ Thereꢀ wasꢀ noꢀ
obviousꢀchangeꢀinꢀtheꢀmorphologyꢀofꢀtheꢀ10Cu‐LaCoO ꢀcatalystꢀ
afterꢀreductionꢀinꢀ5%H ‐95%N ꢀatꢀ573ꢀKꢀ(Fig.ꢀ4(d)).ꢀHighꢀreso‐
3
2
2
lutionꢀtransmissionꢀelectronꢀmicroscopeꢀ(HRTEM)ꢀimagesꢀandꢀ
Energyꢀdispersiveꢀspectrdmeterꢀ(EDS)ꢀmappingꢀanalysisꢀofꢀtheꢀ
3
reducedꢀ10Cu‐LaCoO ꢀcatalystꢀindicateꢀthatꢀCu,ꢀCo,ꢀandꢀLaꢀwereꢀ
distributedꢀquiteꢀevenlyꢀandꢀLa
2
O
2
CO ,ꢀLaCoO ꢀandꢀCo ꢀcoex‐
3
3
3 4
O
istedꢀinꢀtheꢀcatalyst;ꢀtheseꢀobservationsꢀareꢀconsistentꢀwithꢀtheꢀ
XRDꢀ results.ꢀ However,ꢀ aꢀ numberꢀ ofꢀ darkꢀ particlesꢀ couldꢀ beꢀ
clearlyꢀseenꢀwhenꢀtheꢀcatalystꢀwasꢀusedꢀfourꢀtimes.ꢀTheseꢀdarkꢀ
particlesꢀ mayꢀ beꢀ associatedꢀ withꢀ theꢀ continuousꢀ sinteringꢀ ofꢀ
0
Cu ꢀparticlesꢀduringꢀtheꢀreactionꢀprocess.ꢀNoteꢀthatꢀtheꢀdepos‐
itedꢀcarbonꢀdueꢀtoꢀcokeꢀformationꢀcanꢀalsoꢀcauseꢀtheꢀparticlesꢀ
inꢀtheꢀusedꢀcatalystꢀtoꢀappearꢀaꢀlittleꢀbitꢀdarker.ꢀTheꢀformationꢀ
ofꢀcokeꢀinꢀtheꢀusedꢀcatalystꢀwasꢀsupportedꢀbyꢀaꢀclearꢀmassꢀlossꢀ
inꢀtheꢀTGꢀprofileꢀinꢀtheꢀrangeꢀofꢀ550–855ꢀK,ꢀasꢀwillꢀbeꢀshownꢀ
later.ꢀ ꢀ
TheꢀreducibilityꢀofꢀtheꢀcalcinedꢀsamplesꢀwasꢀstudiedꢀbyꢀTPRꢀ
(
Fig.ꢀ5).ꢀItꢀcanꢀbeꢀseenꢀthatꢀtheꢀLaCoO ꢀsupportꢀwasꢀreducedꢀinꢀ
3
Fig.ꢀ4.ꢀTEMꢀandꢀHRTEMꢀimagesꢀofꢀ(a)ꢀcalcinedꢀ2CuO‐LaCoO
cinedꢀ 10CuO‐LaCoO ,ꢀ (c)ꢀ calcinedꢀ 20CuO‐LaCoO ,ꢀ (d,ꢀ e)ꢀ reducedꢀ
0Cu‐LaCoO ꢀinꢀ5%H ‐95%N ,ꢀandꢀ(f)ꢀ10Cu‐LaCoO3ꢀusedꢀfourꢀtimes;ꢀ(g)ꢀ
High‐angleꢀAnnularꢀDarkꢀField‐scanningꢀtransmissionꢀelectronꢀmicro‐
scopeꢀ(HAADF‐STEM)ꢀimageꢀofꢀ10Cu‐LaCoO ꢀreducedꢀinꢀ5%H ‐95%N ,ꢀ
3
,ꢀ(b)ꢀcal‐
twoꢀmajorꢀregions.ꢀTheꢀfirstꢀoneꢀinꢀtheꢀrangeꢀofꢀ644–890ꢀKꢀandꢀ
centeredꢀatꢀaboutꢀ763ꢀKꢀmightꢀbeꢀassociatedꢀwithꢀtheꢀreductionꢀ
ofꢀCo ꢀtoꢀCo ꢀandꢀtheꢀsecondꢀoneꢀatꢀtemperaturesꢀaboveꢀ890ꢀKꢀ
mightꢀbeꢀattributedꢀtoꢀtheꢀreductionꢀofꢀCo ꢀowingꢀtoꢀinterac‐
tionsꢀwithꢀLa ꢀspeciesꢀtoꢀCo ꢀ[45].ꢀAnꢀobviousꢀshiftꢀinꢀtheꢀre‐
3
3
1
3
2
2
3+
2+
2+
3
2
2
3+
0
andꢀtheꢀcorrespondingꢀEDSꢀelementalꢀmappingsꢀofꢀCu,ꢀCo,ꢀLa,ꢀandꢀO.ꢀ

1716ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
20Cu-LaCoO3
20Cu-LaCoO3
10Cu-LaCoO3
15Cu-LaCoO3
2Cu-LaCoO3
1
0Cu-LaCoO3
LaCoO
3
4
00
500
600
700
800
900
1000 1100
5
Cu-LaCoO3
Cu-LaCoO3
Temperature (K)
Fig.ꢀ6.ꢀCO
2
‐TPDꢀprofilesꢀofꢀcalcinedꢀxCuO‐LaCoO
3
ꢀcatalysts.ꢀ
2
atedꢀwithꢀtheꢀformationꢀofꢀcarbonateꢀspeciesꢀwhenꢀtheꢀsampleꢀ
wasꢀ exposedꢀ toꢀ CO ꢀ[37].ꢀSuchꢀcarbonateꢀ speciesꢀwouldꢀ con‐
tributeꢀ toꢀ theꢀ strongꢀ basicityꢀ ofꢀ theꢀ catalysts.ꢀ Thus,ꢀ theꢀ
0Cu‐LaCoO ꢀ catalystꢀ withꢀ theꢀ largestꢀ CO ꢀ desorptionꢀ peakꢀ
2
LaCoO3
1
3
2
areaꢀatꢀtemperaturesꢀ aboveꢀ 850ꢀKꢀ wouldꢀ possessꢀ theꢀlargestꢀ
amountꢀofꢀstrongꢀbasicꢀsites.ꢀCharacterizationꢀofꢀtheꢀacidityꢀofꢀ
4
00 500 600 700 800 900 1000
Temperature (K)
1
0Cu‐LaCoO
3
ꢀ catalystꢀ byꢀ NH ‐TPDꢀ (Fig.ꢀ S1ꢀ inꢀ Supportingꢀ In‐
3
2
Fig.ꢀ5.ꢀH ‐TPRꢀprofilesꢀofꢀcalcinedꢀxCuO‐LaCoO3ꢀcatalysts.ꢀ
formation)ꢀindicatedꢀaꢀratherꢀlowꢀacidityꢀforꢀthisꢀcatalyst.ꢀ
Tableꢀ1ꢀshowsꢀtheꢀtexturalꢀpropertiesꢀofꢀtheꢀcalcinedꢀsam‐
ples.ꢀTheꢀBETꢀ surfaceꢀareasꢀ ofꢀ theꢀcalcinedꢀ CuO‐LaCoO ꢀsam‐
plesꢀincreasedꢀfromꢀ3.5ꢀm /gꢀtoꢀaꢀmaximumꢀofꢀ21.7ꢀm /gꢀasꢀtheꢀ
Cuꢀloadingꢀincreasedꢀfromꢀ0ꢀtoꢀ5ꢀwt%ꢀandꢀthenꢀdecreasedꢀtoꢀ
3
ductionꢀpeakꢀtowardsꢀlowerꢀtemperatureꢀregionsꢀwasꢀobservedꢀ
uponꢀtheꢀincorporationꢀofꢀ2ꢀwt%ꢀCu.ꢀTheꢀonsetꢀreductionꢀtem‐
2
2
3
peratureꢀdecreasedꢀnoticeablyꢀfromꢀ644ꢀKꢀ(LaCoO ꢀsupport)ꢀtoꢀ
2
1
.7ꢀandꢀ3.8ꢀm /gꢀatꢀCuꢀloadingsꢀofꢀ15ꢀwt%ꢀandꢀ20ꢀwt%,ꢀrespec‐
5
27ꢀKꢀ(2CuO‐LaCoO )ꢀandꢀtheꢀpeakꢀtemperatureꢀalsoꢀdecreasedꢀ
3
tively.ꢀ Theꢀ poreꢀ volumesꢀ ofꢀ theꢀ samplesꢀ wereꢀ belowꢀ 0.06ꢀ
toꢀ 653ꢀ Kꢀ forꢀ theꢀ latterꢀ sample.ꢀ Theꢀ majorꢀ peakꢀ inꢀ theꢀ
low‐temperatureꢀregionꢀforꢀtheꢀcatalystsꢀwithꢀ(5ꢀandꢀ10)ꢀwt%ꢀ
Cuꢀbroadenedꢀandꢀtheꢀpeakꢀshiftedꢀtoꢀaroundꢀ713ꢀK.ꢀFurtherꢀ
increaseꢀ inꢀ Cuꢀ loadingꢀ toꢀ 15%ꢀ andꢀ aboveꢀ ledꢀ toꢀ anꢀ obviousꢀ
broadeningꢀandꢀshiftingꢀofꢀtheꢀreductionꢀpeakꢀtowardsꢀhigherꢀ
temperatures.ꢀBecauseꢀCuOꢀcanꢀbeꢀreducedꢀatꢀlowerꢀtempera‐
3
cm /g.ꢀSuchꢀlowꢀsurfaceꢀareasꢀandꢀporeꢀvolumesꢀofꢀtheꢀsamplesꢀ
couldꢀ beꢀ ascribedꢀ toꢀ theꢀ perovskite‐typeꢀ structureꢀ andꢀ highꢀ
calcinationꢀtemperaturesꢀ[47,48].ꢀTheꢀratherꢀhighꢀBETꢀsurfaceꢀ
areaꢀ ofꢀ 5CuO‐LaCoO ꢀ mayꢀ beꢀ associatedꢀ withꢀ theꢀ negligibleꢀ
3
formationꢀofꢀaꢀperovskiteꢀstructureꢀinꢀthisꢀsample,ꢀasꢀrevealedꢀ
byꢀXRDꢀcharacterizationꢀ(Fig.ꢀ1).ꢀTheꢀaverageꢀporeꢀdiametersꢀofꢀ
turesꢀwhenꢀcomparedꢀtoꢀCoO ,ꢀtheꢀobviousꢀshiftꢀinꢀtheꢀreduc‐
x
theꢀCuO‐LaCoO ꢀsamplesꢀwereꢀinꢀtheꢀmesoporeꢀrangeꢀofꢀ10–25ꢀ
3
tionꢀ temperatureꢀ forꢀ low‐Cuꢀ loadingꢀ ((2–10)ꢀ wt%)ꢀ catalystsꢀ
mightꢀ beꢀ ascribedꢀ toꢀ theꢀ spilloverꢀ ofꢀ theꢀ adsorbedꢀ hydrogenꢀ
nm.ꢀ Theꢀ Cuꢀ andꢀ Coꢀ loadingsꢀ ofꢀ theꢀ samples,ꢀ asꢀ measuredꢀ byꢀ
XRF,ꢀwereꢀcloseꢀtoꢀtheirꢀactualꢀamountsꢀ(Tableꢀ1).ꢀComparingꢀ
0
fromꢀCu ꢀtoꢀcobaltꢀoxidesꢀ[10,46].ꢀThus,ꢀitꢀisꢀlikelyꢀthatꢀtheꢀcol‐
theꢀ characterizationꢀ resultsꢀ ofꢀ BETꢀ andꢀ CO ‐TPD,ꢀ itꢀ couldꢀ beꢀ
2
2
+
0
3+
2+
lectiveꢀ reductionꢀ ofꢀ Cu ꢀtoꢀCu ꢀandꢀCo ꢀtoꢀCo ꢀ asꢀ wellꢀ asꢀ aꢀ
inferredꢀthatꢀtheꢀcatalyticꢀperformanceꢀofꢀtheꢀcatalystꢀisꢀmoreꢀ
profoundlyꢀaffectedꢀbyꢀtheꢀreductionꢀpropertiesꢀofꢀtheꢀcatalystꢀ
ratherꢀthanꢀtheꢀBETꢀsurfaceꢀareaꢀandꢀbasicity.ꢀ
2+
0
smallꢀamountꢀofꢀCo ꢀtoꢀCo ꢀcontributesꢀtoꢀtheꢀmajorꢀpeakꢀatꢀ
temperaturesꢀbelowꢀ890ꢀKꢀforꢀsamplesꢀwithꢀCuꢀloadingꢀbelowꢀ
1
0ꢀ wt%.ꢀ Theꢀ shiftingꢀ ofꢀ theꢀ reductionꢀ temperatureꢀ toꢀ higherꢀ
regionsꢀ forꢀ samplesꢀ withꢀ highꢀ Cuꢀ loadingsꢀ (>10%)ꢀ couldꢀ beꢀ
attributedꢀtoꢀtheꢀformationꢀofꢀCuOꢀandꢀperovskite‐typeꢀLaCoO
3
ꢀ
Tableꢀ1ꢀ
withꢀlargeꢀcrystalliteꢀsizes,ꢀasꢀinferredꢀfromꢀtheꢀXRDꢀcharacter‐
izationꢀresultsꢀ(Fig.ꢀ1).ꢀ
Becauseꢀtheꢀsurfaceꢀbasicityꢀofꢀcatalystsꢀplaysꢀanꢀimportantꢀ
roleꢀinꢀtheꢀcatalyticꢀhydrogenolysisꢀofꢀFAꢀandꢀFFAꢀ[8,13,14],ꢀtheꢀ
TheꢀtexturalꢀpropertiesꢀofꢀcalcinedꢀxCuO‐LaCoO ꢀsamples.ꢀ
3
ꢀ
BETꢀarea Poreꢀvolume Averageꢀporeꢀ Cuꢀ
Coꢀ
Sampleꢀ
2
3
(m /g)
(cm /g)ꢀ diameterꢀ(nm)ꢀ (wt%) (wt%)
LaCoO3ꢀ
2CuO‐LaCoO
5CuO‐LaCoO
10CuO‐LaCoO
ꢀ 3.5ꢀ
ꢀ 6.6ꢀ
21.7ꢀ
ꢀ 5.8ꢀ
ꢀ 1.7ꢀ
ꢀ 3.8ꢀ
0.02ꢀ
0.02ꢀ
0.06ꢀ
0.03ꢀ
ꢀ 0.004ꢀ
0.02ꢀ
24.0ꢀ
13.8ꢀ
10.2ꢀ
18.3ꢀ
10.0ꢀ
24.8ꢀ
ꢀ 0.0ꢀ
ꢀ 1.8ꢀ
ꢀ 5.0ꢀ
10.2ꢀ
14.7ꢀ
20.5ꢀ
18.8ꢀ
18.2ꢀ
17.4ꢀ
17.5ꢀ
15.9ꢀ
15.9ꢀ
basicityꢀofꢀxCu‐LaCoO
Fig.ꢀ 6).ꢀ Aꢀ veryꢀ smallꢀ peakꢀ atꢀ aroundꢀ 598ꢀ Kꢀ wasꢀ detectedꢀ inꢀ
eachꢀsample,ꢀwhichꢀcouldꢀbeꢀattributedꢀtoꢀmoderateꢀbasicꢀsites.ꢀ
Itꢀ canꢀ alsoꢀ beꢀ seenꢀ thatꢀ allꢀ theꢀ samplesꢀ showedꢀ obviousꢀ CO
desorptionꢀatꢀtemperaturesꢀaboveꢀ850ꢀK,ꢀwhichꢀmayꢀbeꢀassoci‐
3
ꢀcatalystsꢀwasꢀcharacterizedꢀbyꢀCO
2
‐TPDꢀ
3
ꢀ
ꢀ
(
3
3
15CuO‐LaCoO
3
2
ꢀ
20CuO‐LaCoO
3

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1717ꢀ
ityꢀ >70%)ꢀ (entriesꢀ 9–11).ꢀ Theꢀ commercialꢀ CuCr
2 4
O ꢀ catalystꢀ
Tableꢀ2ꢀ
ConversionꢀandꢀselectivityꢀcorrespondingꢀtoꢀFFAꢀhydrogenolysisꢀoverꢀ
differentꢀcatalysts.ꢀ ꢀ
exhibitedꢀ aꢀ ratherꢀ lowꢀ FFAꢀ conversionꢀ (9.5%)ꢀ andꢀ aꢀ higherꢀ
1,2‐PeDꢀselectivityꢀ(27.3%)ꢀoverꢀ1,5‐PeDꢀ(14.2%)ꢀunderꢀsimi‐
larꢀreactionꢀconditionsꢀ(entryꢀ12).ꢀClearly,ꢀamongꢀtheꢀcatalystsꢀ
a
Selectivityꢀ(%)ꢀ
Conversionꢀ
Entryꢀ
Catalystꢀ
(%)ꢀ
,2‐PeDꢀ 1,5‐PeDꢀ THFA Othersꢀb
investigated,ꢀ10Cu‐LaCoO ꢀexhibitedꢀtheꢀbestꢀperformanceꢀforꢀ
3
1
LaCoO3ꢀcꢀ
2Cu‐LaCoO
5Cu‐LaCoO
10Cu‐LaCoO
15Cu‐LaCoO
20Cu‐LaCoO
theꢀproductionꢀofꢀPeDsꢀbyꢀFFAꢀhydrogenolysisꢀwithꢀbothꢀhighꢀ
FFAꢀ conversionꢀ andꢀ PeDꢀ selectivity.ꢀ Thus,ꢀ thisꢀ catalystꢀ wasꢀ
1
ꢀ
ꢀ ꢀ 2.2ꢀ
ꢀ 56.2ꢀ
ꢀ 88.4ꢀ
ꢀ 94.6ꢀ
ꢀ 53.8ꢀ
ꢀ 51.6ꢀ
ꢀ 0.0ꢀ
14.3ꢀ
14.1ꢀ
13.9ꢀ
12.7ꢀ
12.8ꢀ
15.2ꢀ
ꢀ 9.3ꢀ
12.1ꢀ
traceꢀ
ꢀ 1.2ꢀ
27.3ꢀ
ꢀ 3.3ꢀ
ꢀ 2.7ꢀ
ꢀ 0.0ꢀ
35.0ꢀ
38.1ꢀ
36.9ꢀ
29.3ꢀ
28.8ꢀ
40.3ꢀ
26.8ꢀ
ꢀ 6.4ꢀ
traceꢀ
ꢀ 3.4ꢀ
14.2ꢀ
ꢀ 6.2ꢀ
ꢀ 0.0ꢀ
34.6
32.1
34.1
33.7
35.5
36.5
28.7
44.5
72.9
>99ꢀ
94.0
11.1
15.1
ꢀ 9.9ꢀ
65.4ꢀ
18.6ꢀ
13.7ꢀ
15.5ꢀ
22.5ꢀ
21.9ꢀ
15.8ꢀ
19.4ꢀ
ꢀ 8.6ꢀ
ꢀ 0.0ꢀ
ꢀ 1.4ꢀ
47.4ꢀ
75.4ꢀ
83.4ꢀ
2
ꢀ
3
ꢀ
ꢀ
3
3
3
3
ꢀ
3
selectedꢀforꢀfurtherꢀstudies.ꢀ
4
ꢀ
ꢀ
ꢀ
ꢀ
0
ToꢀinvestigateꢀtheꢀroleꢀofꢀCu ꢀandꢀLaCoO
3
ꢀinꢀFFAꢀhydrogen‐
5
ꢀ
0
0
olysis,ꢀLaCoO
3
,ꢀCu ꢀ+ꢀLaCoO
3
ꢀphysicalꢀmixture,ꢀandꢀCu ꢀcatalystsꢀ
6
ꢀ
d
wereꢀ alsoꢀ testedꢀ (entryꢀ 1,ꢀ 13,ꢀ andꢀ 14).ꢀ Theꢀ LaCoO ꢀ supportꢀ
3
7
ꢀ
10Cu‐LaCoO
5Pt‐LaCoO
5Ru‐LaCoO
5Pd‐LaCoO
5Rh‐LaCoO
CuCr
3
ꢀ ꢀ 100.0ꢀ
8
ꢀ
3
ꢀ
ꢀ 74.0ꢀ
ꢀ 93.5ꢀ
ꢀ 80.7ꢀ
100.0ꢀ
ꢀ ꢀ 9.5ꢀ
showedꢀanꢀextremelyꢀlowꢀFFAꢀ conversionꢀ(2.2%)ꢀandꢀalmostꢀ
noꢀ activityꢀ forꢀ theꢀ cleavageꢀ ofꢀ theꢀ furanꢀ ringꢀ C–Oꢀ bondsꢀ andꢀ
9
ꢀ
3
ꢀ
ꢀ
10ꢀ
11ꢀ
12ꢀ
13ꢀ
14ꢀ
3
0
resultedꢀinꢀTHFAꢀasꢀtheꢀmainꢀproductꢀ(entryꢀ1).ꢀPureꢀCu ꢀex‐
3
ꢀ
hibitedꢀaꢀlowꢀ1,2‐PeDꢀselectivityꢀ(2.7%)ꢀatꢀaꢀslightlyꢀhigherꢀFFAꢀ
conversionꢀ(4.4%)ꢀasꢀcomparedꢀtoꢀtheꢀLaCoO3ꢀsupportꢀ(entryꢀ
2 4
O ꢀ
e
10Cu+LaCoO3ꢀ ꢀ ꢀ ꢀ 7.8ꢀ
PureꢀCuꢀ ꢀ ꢀ 4.4ꢀ
aꢀReactionꢀconditions:ꢀ0.11ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanol,ꢀ413ꢀK,ꢀ6ꢀ
0
14).ꢀ Interestingly,ꢀ theꢀ Cu ꢀ+ꢀLaCoO
3
ꢀ physicalꢀ mixtureꢀcatalystꢀ
exhibitedꢀnotꢀonlyꢀaꢀhigherꢀFFAꢀconversionꢀ(7.8%),ꢀbutꢀalsoꢀaꢀ
higherꢀselectivityꢀtowardsꢀPeDsꢀwithꢀ1,5‐PeDꢀbeingꢀtheꢀmajorꢀ
b
MPaꢀ H
2
,ꢀ 2ꢀ h;ꢀ ꢀ Othersꢀ includeꢀ n‐pentane,ꢀ 1,4‐PeD,ꢀ 2‐methylꢀ furan,ꢀ
2
‐methylꢀ tetrahydrofuran,ꢀ n/2‐pentanol,ꢀ andꢀ someꢀ undeterminedꢀ
0ꢀ
productꢀ (entryꢀ 13).ꢀ Theseꢀ findingsꢀ indicateꢀ thatꢀ Cu providesꢀ
cꢀ
d
e
products;ꢀ 1.1ꢀgꢀLaCoO ;ꢀ ꢀ0.15ꢀgꢀCu;ꢀ ꢀPhysicalꢀmixtureꢀofꢀpureꢀCuꢀandꢀ
3
theꢀ majorityꢀ ofꢀ theꢀ activeꢀ sitesꢀ forꢀ theꢀ generationꢀ ofꢀ 1,2‐PeDꢀ
3
LaCoO .ꢀ
ꢀ
0ꢀ
fromꢀFFA,ꢀwhileꢀtheꢀcoexistenceꢀofꢀCu andꢀLaCoO ꢀwithꢀstrongꢀ
3
basicityꢀ(Fig.ꢀ6)ꢀcouldꢀsteerꢀtheꢀcleavageꢀofꢀtheꢀfuranꢀringꢀC–Oꢀ
bondꢀtoꢀproduceꢀ1,5‐PeD.ꢀTheꢀremarkablyꢀhighꢀFFAꢀconversionꢀ
3.2.ꢀ ꢀ FFAꢀhydrogenolysisꢀoverꢀdifferentꢀcatalystsꢀ
3
andꢀPeDꢀselectivityꢀobtainedꢀwithꢀ10Cu‐LaCoO ꢀ(entryꢀ4)ꢀindi‐
catesꢀ theꢀ importanceꢀ ofꢀ cooperationꢀ betweenꢀ Cuꢀ andꢀ LaCoO
3
ꢀ
Tableꢀ2ꢀshowsꢀtheꢀconversionsꢀandꢀselectivitiesꢀofꢀFFAꢀhy‐
withꢀ highꢀ dispersionꢀ forꢀ theꢀ efficientꢀ hydrogenolysisꢀ ofꢀ FFAꢀ
drogenolysisꢀ overꢀ differentꢀ catalystsꢀ atꢀ 413ꢀ Kꢀ andꢀ 6ꢀ MPaꢀ H
Theꢀ conversionꢀ ofꢀ FFAꢀ forꢀ xCu‐LaCoO ꢀ catalystsꢀ increasedꢀ
sharplyꢀ fromꢀ 56.2%ꢀ (2Cu‐LaCoO )ꢀ toꢀ 94.6%ꢀ (10Cu‐LaCoO )ꢀ
2
.ꢀ
0
intoꢀPeDs.ꢀItꢀseemsꢀthatꢀtheꢀpartiallyꢀreducedꢀCu ‐CoOꢀbounda‐
3
ryꢀplayedꢀanꢀimportantꢀroleꢀinꢀtheꢀselectiveꢀhydrogenolysisꢀofꢀ
3
3
0
FFAꢀtoꢀ1,5‐PeDꢀ[10],ꢀwhileꢀCu ꢀwasꢀmainlyꢀresponsibleꢀforꢀtheꢀ
andꢀthenꢀdecreasedꢀgraduallyꢀtoꢀ51.6%ꢀwithꢀaꢀfurtherꢀincreaseꢀ
inꢀtheꢀCuꢀcontentꢀtoꢀ20ꢀwt%ꢀ(entriesꢀ2–6).ꢀTheꢀselectivitiesꢀforꢀ
productionꢀofꢀ1,2‐PeD.ꢀ
1
,5‐PeDꢀ wereꢀ inꢀ theꢀ rangeꢀ ofꢀ 35.0%–38.1%ꢀ forꢀ xCu‐LaCoO ꢀ
3
3
1
.3.ꢀ ꢀ EffectsꢀofꢀreactionꢀparametersꢀonꢀFFAꢀhydrogenolysisꢀtoꢀ
,2‐PeDꢀandꢀ1,5‐PeDꢀ
catalystsꢀ withꢀ lowꢀ Cuꢀ loadingsꢀ ((2–10)ꢀ wt%),ꢀ whichꢀ furtherꢀ
decreasedꢀtoꢀ~29.0%ꢀforꢀhighꢀCu‐loadingꢀ((15–20)ꢀwt%)ꢀcata‐
lysts.ꢀMeanwhile,ꢀ1,2‐PeDꢀselectivitiesꢀdroppedꢀfromꢀ~14.0%ꢀatꢀ
lowꢀCuꢀloadingsꢀ((2–10)ꢀwt%)ꢀtoꢀ~12.7%ꢀatꢀhighꢀCuꢀloadingsꢀ
Fig.ꢀ7ꢀdisplaysꢀtheꢀeffectꢀofꢀreactionꢀtemperatureꢀonꢀFFAꢀhy‐
drogenolysisꢀactivityꢀoverꢀ10Cu‐LaCoO3ꢀatꢀ6ꢀMPaꢀH ꢀinꢀtheꢀFFAꢀ
conversionꢀrangeꢀofꢀ15%–25%.ꢀTheꢀreactionꢀrateꢀofꢀPeDs,ꢀde‐
finedꢀasꢀmolesꢀofꢀPeDsꢀproducedꢀbyꢀaꢀmoleꢀofꢀtheꢀactiveꢀmetalꢀ
2
((15–20)ꢀ wt%).ꢀ Forꢀ theꢀ xCu‐LaCoO ꢀ catalysts,ꢀ THFAꢀ wasꢀ de‐
3
terminedꢀ toꢀ beꢀ theꢀ majorꢀ byproductꢀ withꢀ aꢀ selectivityꢀ inꢀ theꢀ
rangeꢀofꢀ32.1%–36.5%.ꢀWhenꢀtheꢀamountꢀofꢀtheꢀcatalystꢀisꢀin‐
creasedꢀ (0.15ꢀ gꢀ Cu ),ꢀtheꢀyieldꢀofꢀ1,5‐PeDꢀ(40.3%)ꢀisꢀcloseꢀtoꢀ
–
1
(Co+Cu)ꢀwithꢀrespectꢀtoꢀtime,ꢀincreasedꢀdrasticallyꢀfromꢀ0.1ꢀh ꢀ
0
–1
atꢀ393ꢀKꢀtoꢀ5.4ꢀh ꢀatꢀ453ꢀKꢀ(Fig.ꢀ7(a)),ꢀwithꢀanꢀactivationꢀener‐
gyꢀofꢀ47.8ꢀkJ/mol.ꢀTheꢀselectivityꢀforꢀ1,2‐PeDꢀandꢀ1,5‐PeDꢀfirstlyꢀ
increasedꢀfromꢀ8.8%ꢀandꢀ21.0%ꢀatꢀ393ꢀKꢀtoꢀtheꢀmaximumꢀval‐
uesꢀofꢀ13.7%ꢀandꢀ38.5%ꢀatꢀ433ꢀK,ꢀrespectively,ꢀandꢀthenꢀwereꢀ
almostꢀ constantꢀ asꢀ theꢀ temperatureꢀ increasedꢀ toꢀ 453ꢀ K.ꢀ Anꢀ
increaseꢀinꢀtheꢀtemperatureꢀresultedꢀinꢀaꢀsharpꢀdropꢀinꢀTHFAꢀ
selectivityꢀ fromꢀ 61.1%ꢀ atꢀ 393ꢀ Kꢀ toꢀ 19.8%ꢀ atꢀ 453ꢀ K,ꢀ whichꢀ
showsꢀ thatꢀ theꢀ increaseꢀ ofꢀ theꢀ relativeꢀ reactionꢀ rateꢀ forꢀ C=Cꢀ
bondꢀhydrogenationꢀisꢀlowerꢀthanꢀthatꢀofꢀC–Oꢀbondꢀhydrogen‐
olysisꢀ withꢀ theꢀ increaseꢀ ofꢀ temperature.ꢀ Theꢀ selectivityꢀ forꢀ
n‐pentanolꢀ andꢀ 2‐pentanolꢀ (referredꢀ toꢀ asꢀ pentanols),ꢀ whichꢀ
areꢀdehydrationꢀproductsꢀofꢀPeDs,ꢀincreasedꢀlinearlyꢀfromꢀ1.1%ꢀ
atꢀ393ꢀKꢀtoꢀ9.5%ꢀatꢀ453ꢀK.ꢀSimilarly,ꢀtheꢀselectivityꢀforꢀtheꢀC–OHꢀ
dehydrationꢀ productsꢀ ofꢀ FFA,ꢀ 2‐methylfuranꢀ (2‐MF)ꢀ andꢀ
thatꢀ obtainedꢀ withꢀ recentlyꢀ reportedꢀ Cu‐Co‐Alꢀ (44%ꢀ yieldꢀ ofꢀ
1
,5‐PeD)ꢀ[10]ꢀandꢀNi‐Y
lysts.ꢀ
Forꢀ comparison,ꢀ theꢀ catalyticꢀ performancesꢀ ofꢀ La‐
CoO ‐supportedꢀnobleꢀmetals,ꢀPt,ꢀRu,ꢀPd,ꢀandꢀRh,ꢀwithꢀ5ꢀwt%ꢀ
loadingꢀpreparedꢀbyꢀtheꢀsameꢀcitrate‐complexingꢀmethodꢀandꢀaꢀ
commercialꢀCuCr ꢀcatalystꢀwereꢀalsoꢀstudiedꢀ(entriesꢀ8–12).ꢀ
Pt‐LaCoO ꢀ exhibitedꢀ moderateꢀ butꢀ inferiorꢀ FFAꢀ conversionꢀ
andꢀPeDꢀselectivityꢀcomparedꢀtoꢀ5Cu‐LaCoO ꢀandꢀ10Cu‐LaCoO .ꢀ
Theꢀ higherꢀ selectivityꢀ forꢀ 1,5‐PeDꢀ (26.8%)ꢀ obtainedꢀ overꢀ
Pt‐LaCoO3ꢀ asꢀ comparedꢀ toꢀ 1,2‐PeDꢀ (9.3%)ꢀ isꢀ inꢀ lineꢀ withꢀ aꢀ
previousꢀstudyꢀonꢀtheꢀhydrogenolysisꢀofꢀFFAꢀoverꢀPt‐Co AlO
15].ꢀ Althoughꢀ highꢀ FFAꢀ conversionsꢀ (>80%)ꢀ wereꢀ attainedꢀ
2
O
3
ꢀ(41.9%ꢀyieldꢀofꢀ1,5‐PeD)ꢀ[11]ꢀcata‐
3
O
2 4
5
3
3
3
5
2
4
ꢀ
[
overꢀLaCoO ‐supportedꢀRu,ꢀRh,ꢀandꢀPdꢀcatalysts,ꢀtheseꢀcatalystsꢀ
3
2‐methyltetrahydrofuranꢀ(2‐MTHF),ꢀincreasedꢀuniformlyꢀfromꢀ
.2%ꢀatꢀ393ꢀKꢀtoꢀ14.9%ꢀatꢀ453ꢀK.ꢀ
showedꢀaꢀquiteꢀlowꢀselectivityꢀtowardsꢀPeDsꢀ(combinedꢀselec‐
tivityꢀ<20%)ꢀwithꢀTHFAꢀasꢀtheꢀpredominantꢀproductꢀ(selectiv‐
1

1718ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
6
5
4
3
2
1
0
70
(
a)
60
(b)
1
1
2
,2-PeD
,5-PeD
-MF+2MTHF
60
50
40
30
20
PeDs+Pentanols
THFA+2-MTHF
Pentanols
THFA
Rate
50
40
30
20
10
0
2-MF+2-MTHF+Pentanols
10
0
3
90
400
410
420
430
440
450
460
390
400
410
420
430
440
450
460
Reaction temperature (K)
Fig.ꢀ7.ꢀEffectꢀofꢀreactionꢀtemperatureꢀonꢀFFAꢀhydrogenolysisꢀoverꢀ10Cu‐LaCoO
ꢀMPaꢀH ,ꢀ1ꢀh.ꢀ
Reaction temperature (K)
.ꢀReactionꢀconditions:ꢀ0.003–0.129ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanol,ꢀ
3
6
2
BecauseꢀtheꢀhydrogenolysisꢀofꢀcyclicꢀetherꢀC–Oꢀbondsꢀpro‐
sisꢀofꢀFFAꢀtoꢀPeDs.ꢀ
ducesꢀPeDs,ꢀC–OHꢀbondꢀdehydrationꢀandꢀfuranꢀringꢀC=Cꢀbondꢀ
hydrogenationꢀ ofꢀ FFAꢀ areꢀ competitiveꢀ reactions;ꢀ atꢀ theꢀ sameꢀ
time,ꢀ furtherꢀ hydrogenolysisꢀ ofꢀ PeDsꢀ toꢀ pentanolsꢀ (dehydra‐
tion)ꢀ andꢀ hydrogenationꢀ ofꢀ 2‐MFꢀ (furanꢀ ringꢀ C=Cꢀ bonds)ꢀ toꢀ
Fig.ꢀ8ꢀshowsꢀtheꢀeffectꢀofꢀH ꢀpressureꢀonꢀFFAꢀhydrogenolysisꢀ
overꢀ10Cu‐LaCoO3ꢀatꢀaꢀcontrolledꢀFFAꢀconversionꢀofꢀ15%–25%.ꢀ
Theꢀ reactionꢀ rateꢀ forꢀ PeDꢀ productionꢀ rapidlyꢀ increasedꢀ fromꢀ
2
–1
–1
0.1ꢀh ꢀatꢀ2ꢀMPaꢀtoꢀ1.7ꢀh ꢀatꢀ8ꢀMPaꢀ(Fig.ꢀ8(a)).ꢀTheꢀselectivityꢀofꢀ
1,5‐PeDꢀfirstlyꢀincreasedꢀfromꢀ25.2%ꢀatꢀ2ꢀMPaꢀtoꢀaꢀmaximumꢀofꢀ
2‐MTHFꢀareꢀsecondaryꢀreactionsꢀinvolvedꢀinꢀFFAꢀhydrogenoly‐
sisꢀ[6,27].ꢀFigꢀ7(b)ꢀdepictsꢀaꢀclearerꢀpictureꢀseeingꢀtheꢀchangesꢀ
ofꢀtheꢀactivityꢀofꢀcyclicꢀetherꢀC–Oꢀbondsꢀhydrogenolysis,ꢀC–OHꢀ
bondꢀdehydrationꢀandꢀfuranꢀringꢀC=Cꢀbondsꢀhydrogenationꢀofꢀ
theꢀcatalystꢀwithꢀtheꢀchangeꢀofꢀreactionꢀtemperature.ꢀTheꢀselec‐
tivityꢀforꢀPeDsꢀandꢀpentanolsꢀ(cyclicꢀetherꢀC–Oꢀbondꢀhydrogen‐
olysisꢀproducts)ꢀincreasedꢀmonotonouslyꢀwithꢀincreasingꢀtem‐
perature;ꢀ meanwhile,ꢀ theꢀ selectivityꢀ forꢀ 2‐MF,ꢀ 2‐MTHF,ꢀ andꢀ
pentanolsꢀ(C–OHꢀdehydrationꢀproducts)ꢀalsoꢀincreasedꢀlinearly,ꢀ
suggestingꢀthatꢀhigherꢀtemperaturesꢀfavorꢀnotꢀonlyꢀtheꢀcleav‐
ageꢀ ofꢀ cyclicꢀ etherꢀ C–Oꢀ bonds,ꢀ butꢀ alsoꢀ theꢀ hydrogenolysisꢀ ofꢀ
C–OHꢀbonds.ꢀInꢀcontrast,ꢀtheꢀselectivityꢀforꢀfuranꢀringꢀC=Cꢀbondꢀ
hydrogenationꢀproducts,ꢀTHFAꢀandꢀ2‐MTHF,ꢀdecreasedꢀsharplyꢀ
fromꢀ61.1%ꢀatꢀ393ꢀKꢀtoꢀ21.7%ꢀatꢀ453ꢀKꢀwithꢀincreasingꢀtem‐
perature.ꢀClearly,ꢀhigherꢀtemperaturesꢀareꢀmoreꢀfavorableꢀforꢀ
34.3%ꢀatꢀ6ꢀMPaꢀandꢀthenꢀremainedꢀalmostꢀconstantꢀasꢀtheꢀH
pressureꢀincreasedꢀtoꢀ8ꢀMPa;ꢀ1,2‐PeDꢀselectivityꢀremainedꢀsta‐
bleꢀatꢀ12%ꢀwithꢀanꢀincreaseꢀinꢀH ꢀpressureꢀ(Fig.ꢀ8(a)).ꢀTheꢀse‐
lectivityꢀ ofꢀ THFAꢀ increasedꢀ steadilyꢀ fromꢀ 27.3%ꢀ atꢀ 2ꢀ MPaꢀ toꢀ
44.1%ꢀ atꢀ 8ꢀ MPaꢀ withꢀ increasingꢀ H2ꢀpressure.ꢀ Simultaneously,ꢀ
theꢀcombinedꢀselectivityꢀofꢀ2‐MFꢀandꢀ2‐MTHFꢀdecreasedꢀfromꢀ
14.6%ꢀtoꢀ 3.7%ꢀ duringꢀtheꢀsameꢀ process.ꢀInꢀ addition,ꢀaꢀ slightꢀ
decreaseꢀ inꢀ pentanolꢀ selectivityꢀ fromꢀ 5.5%ꢀ toꢀ 3.5%ꢀ wasꢀ ob‐
servedꢀwithꢀanꢀincreaseꢀinꢀpressure.ꢀTheꢀaboveꢀfindingsꢀsuggestꢀ
2
ꢀ
2
thatꢀaꢀhighꢀH ꢀpressureꢀfavorsꢀnotꢀonlyꢀtheꢀhydrogenolysisꢀofꢀ
2
cyclicꢀetherꢀC–Oꢀbondsꢀbutꢀalsoꢀtheꢀhydrogenationꢀofꢀfuranꢀringꢀ
C=Cꢀbonds;ꢀtheꢀreactionꢀrateꢀofꢀtheꢀlatterꢀincreasedꢀmoreꢀrap‐
idly,ꢀleadingꢀtoꢀaꢀslightꢀdecreaseꢀinꢀtheꢀproductsꢀofꢀtheꢀformerꢀ
reactionꢀatꢀrelativelyꢀhighꢀpressuresꢀofꢀ8ꢀMPa.ꢀTheꢀgradualꢀde‐
clineꢀofꢀtheꢀcombinedꢀselectivityꢀofꢀtheꢀC–OHꢀbondꢀdehydrationꢀ
productsꢀ ofꢀ pentanols,ꢀ 2‐MF,ꢀ andꢀ 2‐MTHFꢀ indicatesꢀ thatꢀ theꢀ
C–OHꢀ bondꢀ dehydrationꢀ reactionꢀ wasꢀ suppressedꢀ withꢀ in‐
theꢀhydrogenolysisꢀofꢀC–Oꢀoverꢀ10Cu‐LaCoO
ꢀthanꢀforꢀtheꢀhy‐
3
drogenationꢀ ofꢀ furanꢀ ringꢀ C=Cꢀ bonds.ꢀ Thus,ꢀ anꢀ appropriatelyꢀ
highꢀtemperatureꢀ(433ꢀK)ꢀisꢀfoundꢀtoꢀbenefitꢀtheꢀhydrogenoly‐
1.8
1.5
1.2
0.9
0.6
0.3
0.0
50
40
1
1
2
,2-PeD
,5-PeD
-MF+2-MTHF
(
a)
(b)
50
Pentanols
THFA
Rate
40
PeDs+Pentanols
THFA+2-MTHF
30
20
10
0
2-MF+2-MTHF+Pentanols
30
2
0
0
1
2
3
4
5
6
7
8
2
3
4
5
6
7
8
H pressure (MPa)
H pressure (MPa)
2
2
2
Fig.ꢀ8.ꢀEffectꢀofꢀH ꢀpressureꢀonꢀFFAꢀhydrogenolysisꢀreactivityꢀoverꢀtheꢀ10Cu‐LaCoO3ꢀcatalyst.ꢀReactionꢀconditions:ꢀ0.018–0.166ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀ
FFAꢀinꢀethanol,ꢀ413ꢀK,ꢀ1ꢀh.ꢀ

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1719ꢀ
creasingꢀH
2
ꢀpressureꢀ(Fig.ꢀ8(b)).ꢀClearly,ꢀanꢀappropriatelyꢀhighꢀ
drogenolysisꢀtoꢀstudyꢀitsꢀreusabilityꢀ(Fig.ꢀ10).ꢀTheꢀconversionꢀofꢀ
FFAꢀincreasedꢀfromꢀ93.7%ꢀtoꢀ99.3%ꢀafterꢀoneꢀcycleꢀandꢀthenꢀ
wasꢀalmostꢀconstantꢀforꢀthreeꢀmoreꢀcycles.ꢀNonetheless,ꢀaꢀcon‐
tinualꢀdecreaseꢀwasꢀobservedꢀinꢀtheꢀselectivityꢀofꢀ1,5‐PeDꢀ(fromꢀ
37.9%ꢀ toꢀ 28.7%)ꢀ andꢀ 1,2‐PeDꢀ (fromꢀ 13.0%ꢀ toꢀ 7.8%)ꢀ duringꢀ
fourꢀcycleꢀreactions.ꢀInꢀtheꢀmeantime,ꢀtheꢀselectivityꢀofꢀTHFAꢀ
increasedꢀslightlyꢀfromꢀ31.7%ꢀtoꢀ41.8%.ꢀTheꢀincreaseꢀinꢀTHFAꢀ
selectivityꢀ atꢀ theꢀ expenseꢀ ofꢀ PeDꢀ selectivityꢀ suggestsꢀ thatꢀ theꢀ
hydrogenationꢀactivityꢀofꢀfuranꢀringꢀC=Cꢀbondsꢀincreased,ꢀwhileꢀ
theꢀhydrogenolysisꢀactivityꢀofꢀcyclicꢀetherꢀC–Oꢀbondsꢀdeclined.ꢀ
Thus,ꢀ theꢀ increaseꢀ inꢀ FFAꢀ conversionꢀ afterꢀ recyclingꢀ canꢀ beꢀ
mainlyꢀ attributedꢀ toꢀ theꢀ enhancedꢀ hydrogenationꢀ activityꢀ ofꢀ
furanꢀ ringꢀ C=Cꢀ bonds,ꢀ leadingꢀ toꢀ theꢀ generationꢀ ofꢀ highꢀ
amountsꢀofꢀTHFA,ꢀwhichꢀcannotꢀbeꢀfurtherꢀconvertedꢀtoꢀPeDsꢀ
underꢀ theꢀ usedꢀ reactionꢀ conditions.ꢀ Similarꢀ trendsꢀ forꢀ FFAꢀ
conversionꢀandꢀproductꢀselectivitiesꢀwereꢀobservedꢀwhenꢀtheꢀ
recyclingꢀofꢀtheꢀcatalystꢀwasꢀstudiedꢀatꢀlowꢀinitialꢀFFAꢀconver‐
sionꢀ(30.5%,ꢀseeꢀFig.ꢀS2ꢀinꢀSI).ꢀ
H ꢀpressureꢀ(~6ꢀMPa)ꢀfavorsꢀtheꢀproductionꢀofꢀtheꢀtargetꢀPeDsꢀ
2
atꢀhighꢀyields.ꢀ
Theꢀ conversionsꢀ andꢀ selectivitiesꢀ ofꢀ 10Cu‐LaCoO3ꢀ duringꢀ
FFAꢀhydrogenolysisꢀasꢀfunctionsꢀofꢀreactionꢀtimeꢀatꢀ413ꢀKꢀandꢀ6ꢀ
MPaꢀH ꢀareꢀshownꢀinꢀFig.ꢀ9.ꢀTheꢀconversionꢀofꢀFFAꢀincreasedꢀ
2
drasticallyꢀfromꢀ25.3%ꢀatꢀ0.2ꢀhꢀtoꢀ94.6%ꢀatꢀ2ꢀhꢀandꢀfullꢀconver‐
sionꢀwasꢀachievedꢀafterꢀ4ꢀh.ꢀTheꢀselectivityꢀofꢀ1,5‐PeDꢀincreasedꢀ
fromꢀ25.9%ꢀatꢀ0.2ꢀhꢀtoꢀaꢀmaximumꢀofꢀ40.6%ꢀatꢀ4ꢀhꢀandꢀthenꢀ
remainedꢀalmostꢀconstantꢀasꢀtheꢀreactionꢀtimeꢀincreasedꢀtoꢀ8ꢀh.ꢀ
Theꢀselectivityꢀofꢀ1,2‐PeDꢀincreasedꢀslightlyꢀfromꢀ11.0%ꢀatꢀ0.2ꢀ
hꢀtoꢀ14.8%ꢀafterꢀ8ꢀh.ꢀSimultaneously,ꢀaꢀslightꢀdecreaseꢀinꢀtheꢀ
selectivityꢀofꢀTHFAꢀfromꢀaroundꢀ36.0%ꢀatꢀ0.2ꢀhꢀtoꢀ28.4%ꢀwasꢀ
observedꢀ withꢀ anꢀ increaseꢀ inꢀ theꢀ reactionꢀ timeꢀ toꢀ 8ꢀ h;ꢀ thisꢀ
mightꢀbeꢀaꢀresultꢀofꢀtheꢀmoreꢀfavoredꢀreactionꢀofꢀFFAꢀhydro‐
genolysisꢀtoꢀPeDs,ꢀwhichꢀwereꢀnotꢀfurtherꢀhydrogenolyzedꢀtoꢀ
pentanolsꢀorꢀ2‐MTHF.ꢀAꢀseparateꢀreactionꢀonꢀtheꢀdirectꢀhydro‐
genolysisꢀ ofꢀ THFAꢀ underꢀ theꢀ sameꢀ conditionsꢀ asꢀ thoseꢀ em‐
ployedꢀ forꢀ FFAꢀ hydrogenolysisꢀ resultedꢀ inꢀ noꢀ reactionꢀ prod‐
ucts,ꢀ supportingꢀ thatꢀ bothꢀ PeDsꢀ andꢀ 2‐MTHFꢀ areꢀ generatedꢀ
fromꢀFFAꢀbutꢀnotꢀTHFA,ꢀwhichꢀisꢀinꢀlineꢀwithꢀpreviousꢀfindingsꢀ
Inꢀ orderꢀ toꢀ evaluateꢀ theꢀ reasonsꢀ forꢀ theseꢀ changesꢀ inꢀ theꢀ
performanceꢀofꢀtheꢀcatalyst,ꢀitꢀwasꢀcharacterizedꢀbyꢀXRDꢀ(Fig.ꢀ
2)ꢀandꢀTEMꢀ(Fig.ꢀ4)ꢀatꢀthreeꢀdifferentꢀstates.ꢀTheꢀappearanceꢀofꢀ
0
[8,15,27].ꢀ Althoughꢀ 2‐MFꢀ couldꢀ beꢀ furtherꢀ hydrogenatedꢀ toꢀ
aꢀCu ꢀdiffractionꢀpeakꢀalongꢀwithꢀaꢀlargelyꢀdisappearedꢀCo
3 4
O ꢀ
2‐MTHFꢀandꢀpentanolsꢀ[15,27],ꢀtheꢀslightꢀincreaseꢀinꢀpentanolꢀ
diffractionꢀpeakꢀafterꢀcatalystꢀreductionꢀ(inꢀ5%H
2
‐95%N )ꢀandꢀ
2
selectivityꢀ(fromꢀ3.5%ꢀtoꢀ5.6%)ꢀandꢀsimultaneousꢀdeclineꢀinꢀtheꢀ
combinedꢀselectivityꢀofꢀ2‐MFꢀandꢀ2‐MTHFꢀ(fromꢀ6.0%ꢀtoꢀ3.5%)ꢀ
afterꢀ 8ꢀ hꢀ ofꢀ reactionꢀ indicatesꢀ thatꢀ furtherꢀ hydrogenationꢀ ofꢀ
usageꢀforꢀoneꢀtimeꢀ(Fig.ꢀ2((3)ꢀandꢀ(4)))ꢀgivesꢀusꢀtoꢀunderstandꢀ
thatꢀtheꢀactiveꢀsitesꢀforꢀFFAꢀhydrogenolysisꢀareꢀprobablyꢀasso‐
0
ciatedꢀwithꢀtheꢀpresenceꢀofꢀCu ꢀandꢀCoꢀwithꢀlowꢀvalencies.ꢀBe‐
0
2‐MFꢀtoꢀpentanolsꢀisꢀnotꢀtheꢀmainꢀcauseꢀforꢀtheꢀobservedꢀde‐
causeꢀ Cu ꢀ itselfꢀ (Tableꢀ 2,ꢀ entryꢀ 14)ꢀ andꢀ itsꢀ combinationꢀ withꢀ
creaseꢀ inꢀ theꢀ combinedꢀ selectivityꢀ ofꢀ 2‐MFꢀ andꢀ 2‐MTHF.ꢀ Theꢀ
moreꢀfavoredꢀreactionꢀofꢀFFAꢀhydrogenolysisꢀtoꢀPeDsꢀasꢀcom‐
paredꢀtoꢀFFAꢀhydrogenolysisꢀtoꢀ2‐MFꢀprobablyꢀaccountsꢀforꢀtheꢀ
declineꢀinꢀtheꢀcombinedꢀselectivityꢀofꢀ2‐MFꢀandꢀ2‐MTHF.ꢀThus,ꢀ
itꢀisꢀlikelyꢀthatꢀfurtherꢀhydrogenolysisꢀofꢀPeDsꢀcontributedꢀtoꢀ
theꢀincreaseꢀinꢀpentanolꢀselectivity.ꢀ
acidicꢀandꢀbasicꢀsupportsꢀfavorꢀtheꢀgenerationꢀofꢀ1,2‐PeDꢀoverꢀ
1,5‐PeDꢀ duringꢀ FFAꢀ hydrogenolysisꢀ [8,27,29],ꢀ theꢀ higherꢀ
1,5‐PeDꢀselectivityꢀobtainedꢀoverꢀCu‐LaCoO
3
ꢀcatalystsꢀasꢀwellꢀasꢀ
0
3
theꢀphysicalꢀmixtureꢀofꢀCu ꢀ+ꢀLaCoO ꢀindicatesꢀthatꢀCoꢀspeciesꢀ
playedꢀanꢀimportantꢀroleꢀinꢀtheꢀselectiveꢀcleavageꢀofꢀtheꢀsec‐
ondaryꢀC–OꢀbondꢀofꢀFFA.ꢀTheꢀpresenceꢀofꢀCoꢀspeciesꢀpreferablyꢀ
cleavedꢀsecondaryꢀC–OꢀbondsꢀbutꢀnotꢀtheꢀprimaryꢀC–Oꢀbondꢀofꢀ
FFA,ꢀwhichꢀisꢀinꢀlineꢀwithꢀpreviousꢀfindingsꢀ[10,15,30].ꢀItꢀwasꢀ
3.4.ꢀ ꢀ Recyclingꢀofꢀtheꢀ10Cu‐LaCoO
3
ꢀcatalystꢀusedꢀforꢀFFAꢀ ꢀ
2+
hydrogenolysisꢀ
reportedꢀthatꢀtheꢀpresenceꢀofꢀlargeꢀamountsꢀofꢀCo ꢀfacilitatesꢀ
theꢀadsorptionꢀofꢀFFAꢀinꢀaꢀtiltedꢀconformation,ꢀwhichꢀenhancesꢀ
theꢀcleavageꢀofꢀtheꢀsecondaryꢀC–OꢀbondsꢀofꢀFFAꢀ[10],ꢀleadingꢀtoꢀ
aꢀhighꢀselectivityꢀtowardꢀ1,5‐PeD.ꢀBecauseꢀCoOꢀisꢀmuchꢀharderꢀ
3
Theꢀ10Cu‐LaCoO ꢀcatalystꢀwasꢀrepeatedlyꢀusedꢀforꢀFFAꢀhy‐
toꢀbeꢀreducedꢀthanꢀCo
3
O
4
ꢀ[10,46],ꢀtheꢀpartialꢀreductionꢀofꢀCo
O ꢀ
3 4
100
Conversion
100
1
,2-PeD
,5-PeD
40
1
,5-PeD
80
60
40
20
0
1
1,2-PeD
THFA
80
60
40
20
0
THFA
-MF+2-MTHF
Pentanols
2
30
20
10
0
1
2
3
4
5
6
7
8
1
2
3
4
Number of cycles
Fig.ꢀ10.ꢀReusabilityꢀofꢀtheꢀ10Cu‐LaCoO ꢀcatalystꢀforꢀFFAꢀhydrogenolysis
Reactionꢀconditions:ꢀ0.11ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanol,ꢀ413ꢀK,ꢀ6ꢀ
MPaꢀH ,ꢀ2ꢀh.ꢀ
Reaction time (h)
3
Fig.ꢀ9.ꢀEffectꢀofꢀreactionꢀtimeꢀonꢀFFAꢀhydrogenolysisꢀoverꢀ10Cu‐LaCoO
Reactionꢀconditions:ꢀ0.11ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanol,ꢀ413ꢀK,ꢀ6
MPaꢀH .ꢀ
3
.
2
2

1720ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
toꢀhighlyꢀdispersedꢀCoOꢀinꢀaꢀhydrogenationꢀatmosphereꢀasꢀwellꢀ
asꢀtheꢀpresenceꢀofꢀanꢀalcoholicꢀsolventꢀmightꢀbeꢀresponsibleꢀforꢀ
Tableꢀ3ꢀ
Effectꢀofꢀpretreatmentꢀatmosphereꢀonꢀtheꢀconversionꢀofꢀfurfurylꢀalcoholꢀ
overꢀtheꢀ10Cu‐LaCoO
a
theꢀ largelyꢀ diminishedꢀ Co
0Cu‐LaCoO ꢀ catalystꢀ wasꢀ usedꢀ oneꢀ timeꢀ (Fig.ꢀ 2(4)).ꢀ Furtherꢀ
diminishingꢀofꢀtheꢀCo ꢀdiffractionꢀpeakꢀafterꢀtheꢀcatalystꢀwasꢀ
repeatedlyꢀusedꢀforꢀthreeꢀmoreꢀcyclesꢀindicatesꢀtheꢀsequentialꢀ
3 4
O ꢀ diffractionꢀ peakꢀ afterꢀ theꢀ
3
ꢀcatalyst.ꢀ ꢀ
1
3
Selectivityꢀ(%)ꢀ
1,2‐PeDꢀ 1,5‐PeDꢀ THFA Othersꢀb
Pretreatmentꢀ
atmosphereꢀ
Conversionꢀ
O
3 4
(%)ꢀ
PureꢀH
2
ꢀ
100.0ꢀ
ꢀ 94.3ꢀ
ꢀ 94.6ꢀ
ꢀ 92.2ꢀ
14.5ꢀ
14.4ꢀ
13.9ꢀ
15.4ꢀ
28.6ꢀ
30.5ꢀ
36.9ꢀ
37.4ꢀ
48.9ꢀ
44.0ꢀ
33.7ꢀ
27.8ꢀ
8.0ꢀ
0
50%H ‐50%N ꢀ
11.1ꢀ
15.5ꢀ
19.4ꢀ
reductionꢀofꢀCo
3
O
4ꢀtoꢀCoOꢀandꢀlaterꢀtoꢀCo ꢀ(Fig.ꢀ2(5)).ꢀTheꢀde‐
2
2
creaseꢀinꢀPeDꢀselectivity,ꢀespeciallyꢀ1,5‐PeDꢀselectivity,ꢀwithꢀtheꢀ
continuousꢀ useꢀ ofꢀ theꢀ catalystꢀ impliesꢀ aꢀ decreaseꢀ inꢀ theꢀ CoOꢀ
amountꢀinꢀtheꢀcatalyst.ꢀThus,ꢀanꢀincreaseꢀinꢀFFAꢀconversionꢀandꢀ
THFAꢀ selectivityꢀ mayꢀ beꢀ associatedꢀ withꢀ anꢀ increaseꢀ inꢀ theꢀ
2
5%H ‐95%N ꢀ
2
5
%H
2
‐95%Arꢀ
aꢀReactionꢀconditions:ꢀ0.11ꢀgꢀCu,ꢀ30ꢀgꢀofꢀ5ꢀwt%ꢀFFAꢀinꢀethanol,ꢀ413ꢀK,ꢀ6ꢀ
MPaꢀH
2
,ꢀ2ꢀhꢀ
b
ꢀ
Includingꢀn‐pentane,ꢀ1,4‐PeD,ꢀ2‐MF,ꢀ2‐MTHF,ꢀn/2‐pentanol,ꢀandꢀsomeꢀ
0
amountꢀ ofꢀ Co .ꢀ Noteꢀ thatꢀ theꢀ sinteringꢀ ofꢀ Cuꢀ particles,ꢀ asꢀ re‐
undeterminedꢀproductsꢀ
ꢀ
vealedꢀ byꢀ XRDꢀ (Fig.ꢀ 2((3)ꢀ andꢀ (4)))ꢀ andꢀ TEMꢀ (Fig.ꢀ 4((d)ꢀ andꢀ
(
f))),ꢀinꢀtheꢀusedꢀcatalystꢀwouldꢀalsoꢀcauseꢀaꢀdeclineꢀinꢀtheꢀhy‐
drogenolysisꢀactivityꢀofꢀcyclicꢀetherꢀC–Oꢀbondsꢀ[27].ꢀTheꢀsignif‐
icantꢀweakeningꢀofꢀallꢀtheꢀdiffractionꢀpeaksꢀofꢀtheꢀ10Cu‐LaCoO
catalystꢀafterꢀfourꢀcyclesꢀcanꢀbeꢀassociatedꢀwithꢀcarbonꢀdeposi‐
tion,ꢀasꢀevidencedꢀbyꢀtheꢀobviousꢀmassꢀlossꢀ(7.2%)ꢀinꢀbetweenꢀ
50–855ꢀKꢀinꢀtheꢀTGꢀprofileꢀofꢀtheꢀusedꢀ10Cu‐LaCoO ꢀcatalystꢀ
Fig.ꢀ11(2))ꢀ[49].ꢀTheꢀmassꢀlossꢀatꢀtemperaturesꢀaboveꢀ855ꢀKꢀ
forꢀbothꢀtheꢀfreshlyꢀreducedꢀcatalystꢀandꢀusedꢀcatalystꢀcanꢀbeꢀ
associatedꢀwithꢀtheꢀdecompositionꢀofꢀLa CO ꢀ[33].ꢀ
characterizationꢀresultsꢀ(Fig.ꢀ2(3)).ꢀPrereductionꢀofꢀtheꢀcatalystꢀ
withꢀ highꢀ concentrationsꢀ ofꢀ H ꢀ (i.e.,ꢀ 100%)ꢀ enhancedꢀ theꢀ re‐
2
3
ꢀ
ducibilityꢀofꢀtheꢀcatalystꢀasꢀrevealedꢀbyꢀtheꢀpresenceꢀofꢀtheꢀdif‐
0
fractionꢀ peakꢀ ofꢀ Cu ꢀ andꢀ simultaneousꢀ disappearanceꢀ ofꢀ theꢀ
diffractionꢀ peakꢀ ofꢀ Co
peaksꢀofꢀCo ꢀand/orꢀCoOꢀwereꢀobservedꢀafterꢀtheꢀcatalystꢀwasꢀ
3
O ꢀ (Fig.ꢀ 2(2)).ꢀ Althoughꢀ noꢀ diffractionꢀ
4
5
(
3
0
0
2
reducedꢀinꢀpureꢀH ,ꢀCo ꢀand/orꢀhighlyꢀdispersedꢀCoOꢀwereꢀde‐
ducedꢀtoꢀbeꢀpresentꢀinꢀtheꢀcatalyst.ꢀItꢀhasꢀbeenꢀreportedꢀthatꢀ
O
2 2
3
0
0
Co ꢀexhibitedꢀ aꢀhigherꢀ C=Cꢀ hydrogenationꢀreactivityꢀthanꢀ Cu ꢀ
[
0
50].ꢀ Thus,ꢀ aꢀ highꢀ amountꢀ ofꢀ Co ꢀ wouldꢀ promoteꢀ theꢀ hydro‐
3
.5.ꢀ ꢀ Effectꢀofꢀpretreatmentꢀatmosphereꢀonꢀtheꢀcatalyticꢀ ꢀ
genationꢀofꢀC=Cꢀinꢀtheꢀfuranꢀring,ꢀleadingꢀtoꢀtheꢀgenerationꢀofꢀaꢀ
largeꢀ amountꢀ ofꢀ aꢀ furan‐ringꢀ saturatedꢀ productꢀ (THFA);ꢀ thisꢀ
observationꢀwasꢀreportedꢀinꢀpreviousꢀstudiesꢀasꢀwellꢀ[10,27].ꢀ
Clearly,ꢀ theꢀ increaseꢀ inꢀ FFAꢀ conversionꢀ andꢀ THFAꢀ selectivityꢀ
performanceꢀofꢀ10Cu‐LaCoO
3
ꢀ
Inꢀ orderꢀ toꢀ furtherꢀ elucidateꢀ theꢀ activeꢀ speciesꢀ inꢀ theꢀ
Cu‐LaCoO ꢀcatalystꢀusedꢀforꢀFFAꢀhydrogenolysis,ꢀtheꢀeffectꢀofꢀ
reductionꢀatmosphereꢀonꢀtheꢀstructureꢀandꢀreactionꢀpropertiesꢀ
ofꢀtheꢀ10Cu‐LaCoO ꢀcatalystꢀwasꢀinvestigated.ꢀPrereductionꢀofꢀ
theꢀcatalystꢀwithꢀlowꢀconcentrationsꢀofꢀH ꢀ(e.g.,ꢀ5%)ꢀledꢀtoꢀhighꢀ
selectivityꢀ (>36%)ꢀ towardsꢀ 1,5‐PeDꢀ butꢀ slightlyꢀ lowerꢀ FFAꢀ
conversionsꢀ(Tableꢀ3).ꢀTheꢀselectivityꢀofꢀTHFAꢀincreasedꢀandꢀitꢀ
becameꢀtheꢀmajorꢀproductꢀatꢀtheꢀexpenseꢀofꢀ1,5‐PeDꢀwithꢀin‐
3
withꢀanꢀincreaseꢀinꢀtheꢀconcentrationꢀofꢀH
gasꢀ(Tableꢀ3)ꢀisꢀassociatedꢀwithꢀtheꢀincreaseꢀofꢀCo ꢀinꢀtheꢀcata‐
2
ꢀinꢀtheꢀprereductionꢀ
0
3
lyst.ꢀ Theꢀdeclineꢀinꢀ1,5‐PeDꢀselectivityꢀ withꢀincreasingꢀH ꢀgasꢀ
2
2
concentrationꢀcanꢀbeꢀrelatedꢀtoꢀtheꢀdecreasingꢀamountꢀofꢀpar‐
tiallyꢀ reducedꢀ CoOꢀ species,ꢀ dueꢀ toꢀ theirꢀ furtherꢀ reductionꢀ atꢀ
2
highꢀ H ꢀ concentrations.ꢀ Therefore,ꢀ studyingꢀ theꢀ effectꢀ ofꢀ theꢀ
prereductionꢀatmosphereꢀsupportedꢀtheꢀpreviouslyꢀmentionedꢀ
creasingꢀH
uponꢀ prereductionꢀ ofꢀ theꢀ catalystꢀ withꢀ pureꢀ H
‐TPRꢀresultsꢀ(Fig.ꢀ5),ꢀitꢀcanꢀbeꢀunderstoodꢀthatꢀitꢀisꢀhardꢀtoꢀ
reduceꢀtheꢀcalcinedꢀCuO‐LaCoO ꢀcatalystꢀatꢀaꢀlowꢀprereductionꢀ
temperatureꢀofꢀ573ꢀKꢀwithꢀ5%ꢀH ;ꢀthisꢀisꢀprovedꢀbyꢀtheꢀXRDꢀ
2
ꢀconcentration;ꢀfullꢀconversionꢀofꢀFFAꢀwasꢀobtainedꢀ
findingsꢀonꢀtheꢀrecyclingꢀofꢀ10Cu‐LaCoO
occurredꢀ betweenꢀ Cu ꢀ andꢀ CoO,ꢀ whichꢀ promotedꢀ theꢀ hydro‐
genolysisꢀofꢀFFAꢀtoꢀPeDs,ꢀespeciallyꢀtoꢀ1,5‐PeD,ꢀwhileꢀCo ꢀpro‐
3
;ꢀcooperativeꢀcatalysisꢀ
2
.ꢀ Fromꢀ theꢀ
0
H
2
0
3
motedꢀtheꢀhydrogenationꢀofꢀFFAꢀtoꢀTHFA.ꢀTakingꢀtheꢀcatalyticꢀ
2
3
performancesꢀofꢀCu‐LaCoO ꢀcatalystsꢀintoꢀconsiderationꢀ(Tableꢀ
2
),ꢀitꢀisꢀsuggestedꢀthatꢀtheꢀsynergeticꢀeffectꢀbetweenꢀbalancedꢀ
0
100
Cu ꢀandꢀCoOꢀsitesꢀplaysꢀaꢀcriticalꢀroleꢀinꢀachievingꢀaꢀhighꢀyieldꢀ
ofꢀ1,5‐PeDꢀfromꢀFFAꢀhydrogenolysis.ꢀ
(1)
9
9
8
8
6
2
8
4
4
.ꢀ ꢀ Conclusionsꢀ
(
2)
Inꢀ summary,ꢀ Cu‐LaCoO ꢀ withꢀ aꢀ perovskiteꢀ structureꢀ wasꢀ
3
employedꢀasꢀanꢀeffectiveꢀcatalystꢀforꢀtheꢀselectiveꢀhydrogenoly‐
sisꢀofꢀbiomass‐derivedꢀFFAꢀtoꢀ1,5‐PeDꢀandꢀ1,2‐PeD.ꢀTheꢀcata‐
lyticꢀperformancesꢀofꢀtheꢀcatalystsꢀwereꢀfoundꢀtoꢀdependꢀonꢀCuꢀ
loadingꢀandꢀprereductionꢀconditions,ꢀsuchꢀasꢀtheꢀH
tionꢀ ofꢀ theꢀ reducingꢀ gas.ꢀ Theꢀ reactionꢀ parameters,ꢀ includingꢀ
temperatureꢀandꢀH ꢀpressure,ꢀalsoꢀgreatlyꢀaffectedꢀtheꢀreactionꢀ
activityꢀ andꢀ productꢀ selectivity.ꢀ Itꢀ wasꢀ demonstratedꢀ thatꢀ co‐
operationꢀbetweenꢀpartiallyꢀreducedꢀCo
andꢀfullyꢀreducedꢀCuꢀspeciesꢀ(Cu )ꢀinducesꢀaꢀsynergeticꢀeffectꢀ
forꢀtheꢀselectiveꢀadsorptionꢀandꢀcatalyzedꢀopeningꢀofꢀtheꢀfuranꢀ
2
ꢀconcentra‐
2
4
00
500
600
700
800
900
1000
Temperature (K)
3
O ꢀ(probablyꢀasꢀCoO)ꢀ
4
Fig.ꢀ 11.ꢀ TGꢀ patternsꢀ ofꢀ 10Cu‐LaCoO3ꢀcatalystsꢀ atꢀ differentꢀ states.ꢀ (1)ꢀ
Reducedꢀinꢀ5%H ‐95%N ;ꢀ(2)ꢀReducedꢀinꢀ5%H ‐95%N2ꢀandꢀusedꢀfour
times.ꢀ
0
2
2
2

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1721ꢀ
ring,ꢀleadingꢀtoꢀ1,5‐PeDꢀwithꢀaꢀhighꢀyield.ꢀIncreasingꢀtheꢀreduc‐
ibilityꢀofꢀtheꢀcatalystꢀtoꢀenhanceꢀtheꢀreductionꢀofꢀCoꢀspeciesꢀtoꢀ
Co ꢀwouldꢀpromoteꢀtheꢀproductionꢀofꢀTHFA.ꢀByꢀoptimizingꢀtheꢀ
catalystꢀcompositionꢀandꢀprereductionꢀandꢀreactionꢀconditions,ꢀ
aꢀhighꢀFFAꢀconversionꢀofꢀ~100%ꢀandꢀselectivityꢀofꢀ55.5%ꢀforꢀ
[11] H.ꢀ W.ꢀ Wijaya,ꢀ T.ꢀ Kojima,ꢀ T.ꢀ Hara,ꢀ N.ꢀ Ichikuni,ꢀ S.ꢀ Shimazu,ꢀ Chem‐
CatChem,ꢀ2017,ꢀ9,ꢀ28692874.ꢀ
[12] R.ꢀF.ꢀMa,ꢀX.ꢀP.ꢀWu,ꢀT.ꢀTong,ꢀZ.ꢀJ.ꢀShao,ꢀY.ꢀQ.ꢀWang,ꢀX.ꢀH.ꢀLiu,ꢀQ.ꢀN.ꢀXia,ꢀ
X.ꢀQ.ꢀGong,ꢀACSꢀCatal.,ꢀ2017,ꢀ7,ꢀ333337.ꢀ
0
[
13] B.ꢀZhang,ꢀY.ꢀL.ꢀZhu,ꢀG.ꢀQ.ꢀDing,ꢀH.ꢀY.ꢀZheng,ꢀY.ꢀW.ꢀLi,ꢀGreenꢀChem.,ꢀ
012,ꢀ14,ꢀ34023409.ꢀ
2
1
,5‐PeDꢀ(40.3%)ꢀandꢀ1,2‐PeDꢀ(15.2%)ꢀwereꢀachievedꢀatꢀ413ꢀKꢀ
andꢀ6ꢀMPaꢀH ꢀoverꢀaꢀcatalystꢀcontainingꢀ10ꢀwt%ꢀCuꢀloadingꢀandꢀ
prereducedꢀ inꢀ 5%ꢀ H .ꢀ Theꢀ importantꢀ findingsꢀ ofꢀ thisꢀ workꢀ
[14] T.ꢀMizugaki,ꢀT.ꢀYamakawa,ꢀY.ꢀNagatsu,ꢀZ.ꢀMaeno,ꢀT.ꢀMitsudome,ꢀK.ꢀ
2
Jitsukawa,ꢀ K.ꢀ Kaneda,ꢀ ACSꢀ Sustainableꢀ Chem.ꢀ Eng.,ꢀ 2014,ꢀ 2,ꢀ
2
22432247.ꢀ
wouldꢀ shedꢀ lightꢀ onꢀ theꢀ developmentꢀ ofꢀ efficientꢀ non‐nobleꢀ
metalꢀcatalystsꢀforꢀtheꢀproductionꢀofꢀvaluableꢀPeDsꢀfromꢀbio‐
mass.ꢀ
[
15] W.ꢀJ.ꢀXu,ꢀH.ꢀF.ꢀWang,ꢀX.ꢀH.ꢀLiu,ꢀJ.ꢀW.ꢀRen,ꢀY.ꢀQ.ꢀWang,ꢀG.ꢀZ.ꢀLu,ꢀChem.ꢀ
Commun.,ꢀ2011,ꢀ47,ꢀ39243926.ꢀ
[16] M.ꢀChia,ꢀY.ꢀJ.ꢀPagan‐Torres,ꢀD.ꢀHibbitts,ꢀQ.ꢀH.ꢀTan,ꢀH.ꢀN.ꢀPham,ꢀA.ꢀK.ꢀ
Datye,ꢀ M.ꢀ Neurock,ꢀ R.ꢀ J.ꢀ Davis,ꢀ J.ꢀ A.ꢀ Dumesic,ꢀ J.ꢀ Am.ꢀ Chem.ꢀ Soc.,ꢀ
2
011,ꢀ133,ꢀ1267512689.ꢀ
Referencesꢀ
[
[
[
[
[
[
[
[
[
[
[
17] M.ꢀChatterjee,ꢀH.ꢀKawanami,ꢀT.ꢀIshizaka,ꢀM.ꢀSato,ꢀT.ꢀSuzuki,ꢀA.ꢀSu‐
zuki,ꢀCatal.ꢀSci.ꢀTechnol.,ꢀ2011,ꢀ1,ꢀ14661471.ꢀ
18] J.ꢀGuan,ꢀG.ꢀM.ꢀPeng,ꢀQ.ꢀCao,ꢀX.ꢀD.ꢀMu,ꢀJ.ꢀPhys.ꢀChem.ꢀC,ꢀ2014,ꢀ118,ꢀ
[
[
[
1] L.ꢀ T.ꢀ Mika,ꢀ E.ꢀ Csefalvay,ꢀ A.ꢀ Nemeth,ꢀ Chemꢀ Rev.,ꢀ 2018,ꢀ 118,ꢀ
05613.ꢀ
2] C.ꢀZ.ꢀLi,ꢀX.ꢀC.ꢀZhao,ꢀA.ꢀQ.ꢀWang,ꢀG.ꢀW.ꢀHuber,ꢀT.ꢀZhang,ꢀChemꢀRev.,ꢀ
015,ꢀ115,ꢀ1155911624.ꢀ ꢀ
3] R.ꢀMariscal,ꢀP.ꢀMaireles‐Torres,ꢀM.ꢀOjeda,ꢀI.ꢀSádaba,ꢀM.ꢀLópezꢀGra‐
nados,ꢀEnergyꢀEnviron.ꢀSci.,ꢀ2016,ꢀ9,ꢀ11441189.ꢀ
5
2555525566.ꢀ
19] S.ꢀKoso,ꢀN.ꢀUeda,ꢀY.ꢀShinmi,ꢀK.ꢀOkumura,ꢀT.ꢀKizuka,ꢀK.ꢀTomishige,ꢀJ.ꢀ
Catal.,ꢀ2009,ꢀ267,ꢀ8992.ꢀ
20] K.ꢀY.ꢀChen,ꢀS.ꢀKoso,ꢀT.ꢀKubota,ꢀY.ꢀNakagawa,ꢀK.ꢀTomishige,ꢀChem‐
CatChem,ꢀ2010,ꢀ2,ꢀ547555.ꢀ
21] S.ꢀ Koso,ꢀ I.ꢀ Furikado,ꢀ A.ꢀ Shimao,ꢀ T.ꢀ Miyazawa,ꢀ K.ꢀ Kunimori,ꢀ K.ꢀ
Tomishige,ꢀChem.ꢀCommun.,ꢀ2009,ꢀ20352037.ꢀ
22] K.ꢀ Y.ꢀ Chen,ꢀ K.ꢀ Mori,ꢀ H.ꢀ Watanabe,ꢀ Y.ꢀ Nakagawa,ꢀ K.ꢀ Tomishige,ꢀ J.ꢀ
Catal.,ꢀ2012,ꢀ294,ꢀ171183.ꢀ ꢀ
23] S.ꢀB.ꢀLiu,ꢀY.ꢀAmada,ꢀM.ꢀTamura,ꢀY.ꢀNakagawa,ꢀK.ꢀTomishige,ꢀGreenꢀ
Chem.,ꢀ2014,ꢀ16,ꢀ617626.ꢀ
24] S.ꢀB.ꢀLiu,ꢀY.ꢀAmada,ꢀM.ꢀTamura,ꢀY.ꢀNakagawa,ꢀK.ꢀTomishige,ꢀCatal.ꢀ
Sci.ꢀTechnol.,ꢀ2014,ꢀ4,ꢀ25352549.ꢀ
2
[4] M.ꢀBesson,ꢀP.ꢀGallezot,ꢀC.ꢀPinel,ꢀChemꢀRev.,ꢀ2014,ꢀ114,ꢀ18271870.ꢀ ꢀ
[5] M.ꢀJ.ꢀCliment,ꢀA.ꢀCorma,ꢀS.ꢀIborra,ꢀGreenꢀChem.,ꢀ2014,ꢀ16,ꢀ516547.ꢀ ꢀ
[6] Y.ꢀ Nakagawa,ꢀ M.ꢀ Tamura,ꢀ K.ꢀ Tomishige,ꢀ ACSꢀ Catal.,ꢀ 2013,ꢀ 3,ꢀ
26552668.ꢀ
[
7] J.ꢀY.ꢀHe,ꢀK.ꢀF.ꢀHuang,ꢀK.ꢀJ.ꢀBarnett,ꢀS.ꢀH.ꢀKrishna,ꢀD.ꢀM.ꢀAlonso,ꢀZ.ꢀJ.ꢀ
Brentzel,ꢀS.ꢀP.ꢀBurt,ꢀT.ꢀWalker,ꢀW.ꢀF.ꢀBanholzer,ꢀC.ꢀT.ꢀMaravelias,ꢀI.ꢀ
Hermans,ꢀJ.ꢀA.ꢀDumesic,ꢀG.ꢀW.ꢀHuber,ꢀFaradayꢀDiscuss.,ꢀ2017,ꢀ202,ꢀ
247267.ꢀ
[8] H.ꢀL.ꢀLiu,ꢀZ.ꢀW.ꢀHuang,ꢀF.ꢀZhao,ꢀF.ꢀCui,ꢀX.ꢀM.ꢀLi,ꢀC.ꢀG.ꢀXia,ꢀJ.ꢀChen,ꢀ
Catal.ꢀSci.ꢀTechnol.,ꢀ2016,ꢀ6,ꢀ668671.ꢀ
25] Y.ꢀNakagawa,ꢀM.ꢀTamura,ꢀK.ꢀTomishige,ꢀCatal.ꢀSurv.ꢀAsia,ꢀ2015,ꢀ19,ꢀ
2
49256.ꢀ
26] K.ꢀ Tomishige,ꢀ Y.ꢀ Nakagawa,ꢀ M.ꢀ Tamura,ꢀ Greenꢀ Chem.,ꢀ 2017,ꢀ 19,ꢀ
8762924.ꢀ
27] H.ꢀL.ꢀLiu,ꢀZ.ꢀW.ꢀHuang,ꢀH.ꢀX.ꢀKang,ꢀC.ꢀG.ꢀXia,ꢀJ.ꢀChen,ꢀChin.ꢀJ.ꢀCatal.,ꢀ
[9] Z.ꢀ J.ꢀ Brentzel,ꢀ K.ꢀ J.ꢀ Barnett,ꢀ K.ꢀ F.ꢀ Huang,ꢀ C.ꢀ T.ꢀ Maravelias,ꢀ J.ꢀ A.ꢀ
Dumesic,ꢀG.ꢀW.ꢀHuber,ꢀChemSusChem,ꢀ2017,ꢀ10,ꢀ13511355.ꢀ
2
[
10] T.ꢀP.ꢀSulmonetti,ꢀB.ꢀHu,ꢀS.ꢀLee,ꢀP.ꢀK.ꢀAgrawal,ꢀC.ꢀW.ꢀJones,ꢀACSꢀSus‐
tainableꢀChem.ꢀEng.,ꢀ2017,ꢀ5,ꢀ89598969.ꢀ
GraphicalꢀAbstractꢀ
Chin.ꢀJ.ꢀCatal.,ꢀ2018,ꢀ39:ꢀ1711–1723ꢀ ꢀ ꢀ doi:ꢀ10.1016/S1872‐2067(18)63110‐9
Selectiveꢀhydrogenolysisꢀofꢀfurfurylꢀalcoholꢀtoꢀ1,5‐ꢀandꢀ1,2‐pentanediolꢀoverꢀCu‐LaCoO
FangfangꢀGao,ꢀHailongꢀLiu,ꢀXunꢀHu,ꢀJingꢀChenꢀ*,ꢀZhiweiꢀHuangꢀ*,ꢀChunguꢀXiaꢀ
0
3
ꢀcatalystsꢀwithꢀbalancedꢀCu ‐CoOꢀsitesꢀ
LanzhouꢀInstituteꢀofꢀChemicalꢀPhysicsꢀ(LICP),ꢀChineseꢀAcademyꢀofꢀSciences;ꢀUniversityꢀofꢀChineseꢀAcademyꢀofꢀSciencesꢀ
ꢀ
0
TheꢀsynergismꢀbetweenꢀbalancedꢀCu ꢀandꢀCoOꢀsitesꢀplayedꢀaꢀcriticalꢀroleꢀinꢀachievingꢀaꢀhighꢀyieldꢀofꢀPeDsꢀwithꢀhighꢀ1,5‐/1,2‐pentanediolꢀ
selectivityꢀratioꢀduringꢀtheꢀhydrogenolysisꢀofꢀbiomass‐derivedꢀfurfurylꢀalcohol.ꢀ

1722ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
2
016,ꢀ37,ꢀ700710.ꢀ
[41] X.ꢀ H.ꢀ Guo,ꢀ Y.ꢀ Li,ꢀ W.ꢀ Song,ꢀ W.ꢀ J.ꢀ Shen,ꢀ Catal.ꢀ Lett.,ꢀ 2011,ꢀ 141,ꢀ
4581463.ꢀ
[
[
[
28] D.ꢀGötz,ꢀM.ꢀLucas,ꢀP.ꢀClaus,ꢀCatalysts,ꢀ2017,ꢀ7,ꢀ50/150/7.ꢀ
29] H.ꢀAdkins,ꢀR.ꢀConnor,ꢀJ.ꢀAm.ꢀChem.ꢀSoc.,ꢀ1931,ꢀ53,ꢀ10911095.ꢀ
30] J.ꢀLee,ꢀS.ꢀP.ꢀBurt,ꢀC.ꢀA.ꢀCarrero,ꢀA.ꢀC.ꢀAlba‐Rubio,ꢀI.ꢀRo,ꢀB.ꢀJ.ꢀO'Neill,ꢀH.ꢀ
J.ꢀKim,ꢀD.ꢀH.ꢀK.ꢀJackson,ꢀT.ꢀF.ꢀKuech,ꢀI.ꢀHermans,ꢀJ.ꢀA.ꢀDumesic,ꢀG.ꢀW.ꢀ
Huber,ꢀJ.ꢀCatal.,ꢀ2015,ꢀ330,ꢀ1927.ꢀ
1
[
[
[
[
42] X.ꢀW.ꢀDu,ꢀS.ꢀP.ꢀLuo,ꢀH.ꢀY.ꢀDu,ꢀM.ꢀTang,ꢀX.ꢀD.ꢀHuang,ꢀP.ꢀK.ꢀShen,ꢀJ.ꢀMa‐
ter.ꢀChem.ꢀA,ꢀ2016,ꢀ4,ꢀ15791585.ꢀ
43] Y.ꢀC.ꢀWei,ꢀX.ꢀRen,ꢀH.ꢀM.ꢀMa,ꢀX.ꢀSun,ꢀY.ꢀZhang,ꢀX.ꢀKuang,ꢀT.ꢀYan,ꢀH.ꢀX.ꢀ
Ju,ꢀD.ꢀWu,ꢀQ.ꢀWei,ꢀChem.ꢀCommun.,ꢀ2018,ꢀ54,ꢀ15331536.ꢀ
44] L.ꢀPredoana,ꢀB.ꢀMalic,ꢀM.ꢀKosec,ꢀM.ꢀCarata,ꢀM.ꢀCaldararu,ꢀM.ꢀZaha‐
rescu,ꢀJ.ꢀEur.ꢀCeram.ꢀSoc.,ꢀ2007,ꢀ27,ꢀ44074411.ꢀ
[
31] J.ꢀC.ꢀLee,ꢀY.ꢀXu,ꢀG.ꢀW.ꢀHuber,ꢀAppl.ꢀCatal.ꢀB,ꢀ2013,ꢀ140141,ꢀ98107.ꢀ ꢀ
32] M.ꢀ Y.ꢀ Chen,ꢀ C.ꢀ B.ꢀ Chen,ꢀ B.ꢀ Zada,ꢀ Y.ꢀ Fu,ꢀ Greenꢀ Chem.,ꢀ 2016,ꢀ 18,ꢀ
[
38583866.ꢀ ꢀ
45] L.ꢀZhao,ꢀT.ꢀHan,ꢀH.ꢀWang,ꢀL.ꢀH.ꢀZhang,ꢀY.ꢀLiu,ꢀAppl.ꢀCatal.ꢀB,ꢀ2016,ꢀ
187,ꢀ1929.ꢀ
[
[
[
33] B.ꢀH.ꢀZhao,ꢀB.ꢀH.ꢀYan,ꢀS.ꢀY.ꢀYao,ꢀZ.ꢀH.ꢀXie,ꢀQ.ꢀY.ꢀWu,ꢀR.ꢀRan,ꢀD.ꢀWeng,ꢀ
C.ꢀZhang,ꢀJ.ꢀG.ꢀChen,ꢀJ.ꢀCatal.,ꢀ2018,ꢀ358,ꢀ168178.ꢀ
34] X.ꢀ L.ꢀ Zhang,ꢀ Y.ꢀ D.ꢀ Gong,ꢀ S.ꢀ Q.ꢀ Li,ꢀ C.ꢀ W.ꢀ Sun,ꢀ ACSꢀ Catal.,ꢀ 2017,ꢀ 7,ꢀ
[46] Z.ꢀW.ꢀHuang,ꢀK.ꢀJ.ꢀBarnett,ꢀJ.ꢀP.ꢀChada,ꢀZ.ꢀJ.ꢀBrentzel,ꢀZ.ꢀR.ꢀXu,ꢀJ.ꢀA.ꢀ
77377747.ꢀ
Dumesic,ꢀG.ꢀW.ꢀHuber,ꢀACSꢀCatal.,ꢀ2017,ꢀ7,ꢀ84298440.ꢀ
35] J.ꢀHwang,ꢀR.ꢀR.ꢀRao,ꢀL.ꢀGiordano,ꢀY.ꢀKatayama,ꢀY.ꢀYu,ꢀY.ꢀShao‐Horn,ꢀ
Science,ꢀ2017,ꢀ358,ꢀ751–756.ꢀ
[
[
[
47] A.ꢀA.ꢀA.ꢀdaꢀSilva,ꢀM.ꢀC.ꢀRibeiro,ꢀD.ꢀC.ꢀCronauer,ꢀA.ꢀJ.ꢀKropf,ꢀC.ꢀL.ꢀMar‐
shall,ꢀP.ꢀGao,ꢀG.ꢀJacobs,ꢀB.ꢀH.ꢀDavis,ꢀF.ꢀB.ꢀNoronha,ꢀL.ꢀV.ꢀMattos,ꢀTop.ꢀ
Catal.,ꢀ2014,ꢀ57,ꢀ637655.ꢀ
[
36] F.ꢀPolo‐Garzon,ꢀZ.ꢀL.ꢀWu,ꢀJ.ꢀMater.ꢀChem.ꢀA,ꢀ2018,ꢀ6,ꢀ28772894.ꢀ
37] S.ꢀDama,ꢀS.ꢀR.ꢀGhodke,ꢀR.ꢀBobade,ꢀH.ꢀR.ꢀGurav,ꢀS.ꢀChilukuri,ꢀAppl.ꢀ
Catal.ꢀB,ꢀ2018,ꢀ224,ꢀ146158.ꢀ
[
48] C.ꢀZhou,ꢀX.ꢀLiu,ꢀC.ꢀZ.ꢀWu,ꢀY.ꢀW.ꢀWen,ꢀY.ꢀJ.ꢀXue,ꢀR.ꢀChen,ꢀZ.ꢀL.ꢀZhang,ꢀB.ꢀ
Shan,ꢀ H.ꢀ F.ꢀ Yin,ꢀ W.ꢀ G.ꢀ Wang,ꢀ Phys.ꢀ Chem.ꢀ Chem.ꢀ Phys.,ꢀ 2014,ꢀ 16,ꢀ
[
[
[
38] H.ꢀL.ꢀLiu,ꢀZ.ꢀW.ꢀHuang,ꢀZ.ꢀB.ꢀHan,ꢀK.ꢀL.ꢀDing,ꢀH.ꢀC.ꢀLiu,ꢀC.ꢀG.ꢀXia,ꢀJ.ꢀ
51065112.ꢀ
Chen,ꢀGreenꢀChem.,ꢀ2015,ꢀ17,ꢀ42814290.ꢀ ꢀ
49] D.ꢀP.ꢀLiu,ꢀX.ꢀY.ꢀQuek,ꢀH.ꢀH.ꢀA.ꢀWah,ꢀG.ꢀM.ꢀZeng,ꢀY.ꢀH.ꢀLi,ꢀY.ꢀYang,ꢀCatal.ꢀ
Today,ꢀ2009,ꢀ148,ꢀ243250.ꢀ
39] H.ꢀL.ꢀLiu,ꢀZ.ꢀW.ꢀHuang,ꢀH.ꢀX.ꢀKang,ꢀX.ꢀM.ꢀLi,ꢀC.ꢀG.ꢀXia,ꢀJ.ꢀChen,ꢀH.ꢀC.ꢀLiu,ꢀ
Appl.ꢀCatal.ꢀB,ꢀ2018,ꢀ220,ꢀ251263.ꢀ
40] F.ꢀMa,ꢀW.ꢀChu,ꢀL.ꢀH.ꢀHuang,ꢀX.ꢀP.ꢀYu,ꢀY.ꢀY.ꢀWu,ꢀChin.ꢀJ.ꢀCatal.,ꢀ2011,ꢀ
[50] E.ꢀA.ꢀCepeda,ꢀU.ꢀIriarte‐Velasco,ꢀB.ꢀCalvo,ꢀI.ꢀSierra,ꢀ ꢀJ.ꢀAm.ꢀOil.ꢀChem.ꢀ
Soc.,ꢀ2016,ꢀ93,ꢀ731741.
32,ꢀ970977.ꢀ
Cu-LaCoO 催化剂选择氢解生物质基糠醇制备1,5-和1,2-戊二醇
3
a,b
a
a
a,#
a,*
a
高芳芳 , 刘海龙 , 胡 勋 , 陈 静 , 黄志威 , 夏春谷
a
中国科学院兰州化学物理研究所, 羰基合成与选择氧化国家重点实验室, 苏州研究院, 甘肃兰州730000
b
中国科学院大学, 北京100049
摘要: 高效转化可再生生物质资源制备人类社会必需的燃料和化学品是当前关注和研究的热点之一. 生物质基糠醇来源
于玉米芯、甘蔗渣、秸秆等农林副产物, 价廉易得, 是选择氢解合成高附加值1,2-和1,5-戊二醇的理想原料. 目前生物质基
呋喃衍生物氢解制备二元醇的研究主要集中在Pt, Ru, Rh和Ir等贵金属催化剂, 对无Cr非贵金属催化剂的研究甚少. 基于纳
米Cu催化剂较高的C–O键氢解活性和较低的C–C键裂解活性, 以及碱性载体对反应物和反应中间体的稳定作用, 我们在前
3 2 3 3
期Cu-Mg AlO4.5和Cu-Al O 催化剂催化糠醇氢解研究基础上, 以具有一定碱性的ABO 结构的钙钛矿型化合物为载体负载
活性Cu开展糠醇氢解研究, 深入研究催化剂结构、组成和活性金属价态等对催化剂活性和选择性影响, 并研究了催化剂循
环使用稳定性.
首先我们采用柠檬酸一步络合法制备了一系列具有一定钙钛矿结构的不同Cu负载量(0–20 wt%)的Cu-LaCoO
3
催化剂
以及LaCoO 负载的5 wt% Pt, Ru, Rh和Pd催化剂并考察了它们的糠醇选择氢解制备戊二醇性能. 研究发现, 在相同活性金
3
属负载量(5 wt%)时, Cu-LaCoO
氢饱和. 考察不同Cu负载量的Cu-LaCoO
在10 wt% Cu负载量时达最高(94.6%), 戊二醇总选择性也随Cu负载量的增加先升高后降低, 在5 wt% Cu负载量时最高
3
催化剂具有较优异的呋喃环C–O键氢解活性, 而贵金属催化剂倾向于催化呋喃环C=C键加
3
催化剂催化糠醇氢解性能发现, 随着Cu负载量的增加, 糠醇转化率先升高后降低,
(
52.2%), 总体以10 wt% Cu负载量催化剂表现出最优异的性能.
3
接着我们考察了反应动力学条件如温度、压力和反应时间以及还原处理条件对10 wt%Cu-LaCoO 催化性能的影响. 研
究发现适当的高温(~433 K)和高压(6 MPa H )有利于Cu-LaCoO 催化糠醇氢解制戊二醇, 而低浓度氢气(5 vol%)还原有利于
2
3
1
5
,5-戊二醇的生成, 高氢气浓度(纯氢)还原有利于呋喃环加氢饱和的四氢糠醇生成. 10 wt% Cu负载量的催化剂经
%H -95%N 处理后, 在413 K和6 MPa H 条件下可取得100%的糠醇转化率以及55.5%的戊二醇总选择性 (其中1,5-戊二醇
2
2
2
和1,2-戊二醇的选择性之比接近3:1). 进一步考察了10 wt%Cu-LaCoO 催化剂的循环使用稳定性, 研究发现无论是在高初
3
始转化率(~93.7%)还是低初始转化率(~30.5%)条件下, 经多次循环使用后糠醇转化率先升高后基本保持不变, 而戊二醇总
选择性呈下降趋势, 四氢糠醇的选择性逐渐上升.
2 2 3 3
结合XRD, XPS, BET, H -TPR, CO -TPD, NH -TPD和HRTEM等多种表征技术对Cu-LaCoO 催化剂的结构及在糠醇氢
解反应中的活性位进行了表征, 发现高分散的活性物种、合适的碱性以及部分还原的活性组分均有利于提高催化剂的活性
0
与1,5-戊二醇的化学选择性, 高分散的Cu 与部分还原的Co
3
O
4
(很可能是CoO)之间的协同催化对于取得较优异的糠醇氢解

ꢀ
FangfangꢀGaoꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ39ꢀ(2018)ꢀ1711–1723ꢀ
1723ꢀ
性能, 尤其是较高的1,5-/1,2-戊二醇比例至关重要.
关键词: 糠醇; 选择性氢解; 戊二醇; 铜-钴酸镧催化剂; 钙钛矿结构
收稿日期: 2018-04-07. 接受日期: 2018-05-25. 出版日期: 2018-10-05.
*
通讯联系人. 电话: (0931)4968070; 传真: (0931)4968129; 电子信箱: zwhuang@licp.cas.cn
#
通讯联系人. 电话: (0931)4968068; 传真: (0931)4968129; 电子信箱: chenj@licp.cas.cn
基金来源: 国家自然科学基金(21473224); 中科院前沿科学重点研究项目(QYZDJ-SSW-SLH051); 中科院青年创新促进会
2016371); 苏州科技发展计划(SYG201626).
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).
(