Promotional effect of spherical alumina loading with manganese-based bimetallic oxides on nitric-oxide deep oxidation by ozone

Article abstract of DOI:10.1016/S1872-2067(17)62860-2

Nitric oxide (NO) deep oxidation to dinitrogen pentoxide (N₂O₅) by ozone together with wet scrubbing has become a promising technology for nitrogen-oxide (NO_x) removal in industrial boilers. Catalysts were introduced to enhance the N₂O₅ formation rate with less ozone injection and leakage. A series of monometallic catalysts (manganese, cobalt, cerium, iron, copper, and chromium) as prepared by the sol-gel method were tested. The manganese oxides achieved an almost 80% conversion efficiency at an ozone (O₃)/NO molar ratio of 2.0 in 0.12 s. The crystalline structure and porous parameters were determined. The thermodynamic reaction threshold of NO conversion to N₂O₅ is oxidation with an O₃/NO molar ratio of 1.5. Spherical alumina was selected as the support to achieve the threshold, which was believed to improve the catalytic activity by increasing the surface area and the gas-solid contact time. Based on the manganese oxides, cerium, iron, chromium, copper, and cobalt were introduced as promoters. Cerium and iron improved the deep-oxidation efficiency compared with manganese/spherical alumina, with less than 50 mg/m³ of outlet NO + nitrogen oxide, and less than 25 mg/m³ of residual ozone at an O₃/NO molar ratio of 1.5. The other three metal oxides inhibited catalytic activity. X-ray diffraction, nitrogen adsorption, hydrogen temperature-programmed reduction, and X-ray photoelectron spectroscopy results indicate that the catalytic activity is affected by the synergistic action of NO_x oxidation and ozone decomposition.

Full text of DOI:10.1016/S1872-2067(17)62860-2

ChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280
ꢀ
ꢀ ꢀ ꢀ
ꢀ ꢀ ꢀ ꢀ
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
Promotionalꢀeffectꢀofꢀsphericalꢀaluminaꢀloadingꢀwithꢀ ꢀ
manganese‐basedꢀbimetallicꢀoxidesꢀonꢀnitric‐oxideꢀdeepꢀoxidationꢀ
byꢀozoneꢀ
FaweiꢀLin,ꢀZhihuaꢀWangꢀ*,ꢀJiamingꢀShao,ꢀDingkunꢀYuan,ꢀYongꢀHe,ꢀYanqunꢀZhu,ꢀKefaꢀCen
ꢀ
StateꢀKeyꢀLaboratoryꢀofꢀCleanꢀEnergyꢀUtilization,ꢀZhejiangꢀUniversity,ꢀHangzhouꢀ310027,ꢀZhejiang,ꢀChinaꢀ
A R T I C L E ꢀ I N F O ꢀ
A B S T R A C T ꢀ
Articleꢀhistory:ꢀ
Nitricꢀoxideꢀ(NO)ꢀdeepꢀoxidationꢀtoꢀdinitrogenꢀpentoxideꢀ(N
2 5
O )ꢀbyꢀozoneꢀtogetherꢀwithꢀwetꢀscrub‐
Receivedꢀ14ꢀAprilꢀ2017ꢀ
Acceptedꢀ17ꢀMayꢀ2017ꢀ
Publishedꢀ5ꢀJulyꢀ2017ꢀ
bingꢀ hasꢀ becomeꢀ aꢀ promisingꢀ technologyꢀ forꢀ nitrogen‐oxideꢀ (NO
x
)ꢀ removalꢀ inꢀ industrialꢀ boilers.ꢀ
CatalystsꢀwereꢀintroducedꢀtoꢀenhanceꢀtheꢀN ꢀformationꢀrateꢀwithꢀlessꢀozoneꢀinjectionꢀandꢀleakage.ꢀ
O
2 5
Aꢀseriesꢀofꢀmonometallicꢀcatalystsꢀ(manganese,ꢀcobalt,ꢀcerium,ꢀiron,ꢀcopper,ꢀandꢀchromium)ꢀasꢀpre‐
paredꢀbyꢀtheꢀsol‐gelꢀmethodꢀwereꢀtested.ꢀTheꢀmanganeseꢀoxidesꢀachievedꢀanꢀalmostꢀ80%ꢀconver‐
sionꢀefficiencyꢀatꢀanꢀozoneꢀ(O
parametersꢀwereꢀdetermined.ꢀTheꢀthermodynamicꢀreactionꢀthresholdꢀofꢀNOꢀconversionꢀtoꢀN
oxidationꢀ withꢀ anꢀ O /NOꢀ molarꢀ ratioꢀ ofꢀ 1.5.ꢀ Sphericalꢀ aluminaꢀ wasꢀ selectedꢀ asꢀ theꢀ supportꢀ toꢀ
3
)/NOꢀmolarꢀratioꢀofꢀ2.0ꢀinꢀ0.12ꢀs.ꢀTheꢀcrystallineꢀstructureꢀandꢀporousꢀ
Keywords:ꢀ
2 5
O ꢀisꢀ
Nitricꢀoxideꢀ
Deepꢀoxidationꢀ
Catalystꢀ
Ozoneꢀ
Transitionꢀmetalꢀoxide
3
achieveꢀtheꢀthreshold,ꢀwhichꢀwasꢀbelievedꢀtoꢀimproveꢀtheꢀcatalyticꢀactivityꢀbyꢀincreasingꢀtheꢀsurfaceꢀ
areaꢀandꢀtheꢀgas‐solidꢀcontactꢀtime.ꢀBasedꢀonꢀtheꢀmanganeseꢀoxides,ꢀcerium,ꢀiron,ꢀchromium,ꢀcop‐
per,ꢀandꢀcobaltꢀwereꢀintroducedꢀasꢀpromoters.ꢀCeriumꢀandꢀironꢀimprovedꢀtheꢀdeep‐oxidationꢀeffi‐
3
ciencyꢀcomparedꢀwithꢀmanganese/sphericalꢀalumina,ꢀwithꢀlessꢀthanꢀ50ꢀmg/m ꢀofꢀoutletꢀNOꢀ+ꢀnitro‐
3
genꢀoxide,ꢀandꢀlessꢀthanꢀ25ꢀmg/m ꢀofꢀresidualꢀozoneꢀatꢀanꢀO
3
/NOꢀmolarꢀratioꢀofꢀ1.5.ꢀTheꢀotherꢀthreeꢀ
metalꢀoxidesꢀinhibitedꢀcatalyticꢀactivity.ꢀX‐rayꢀdiffraction,ꢀnitrogenꢀadsorption,ꢀhydrogenꢀtempera‐
ture‐programmedꢀreduction,ꢀandꢀX‐rayꢀphotoelectronꢀspectroscopyꢀresultsꢀindicateꢀthatꢀtheꢀcata‐
lyticꢀactivityꢀisꢀaffectedꢀbyꢀtheꢀsynergisticꢀactionꢀofꢀNO ꢀoxidationꢀandꢀozoneꢀdecomposition.ꢀ
x
©ꢀ2017,ꢀDalianꢀInstituteꢀofꢀChemicalꢀPhysics,ꢀChineseꢀAcademyꢀofꢀSciences.
PublishedꢀbyꢀElsevierꢀB.V.ꢀAllꢀrightsꢀreserved.
1
.ꢀ ꢀ Introductionꢀ
uneconomicꢀandꢀunableꢀtoꢀmatchꢀincreasinglyꢀstrictꢀemissionsꢀ
legislation,ꢀespeciallyꢀforꢀultra‐lowꢀChineseꢀemissionsꢀ(particleꢀ
3
3
3
Nitrogenꢀoxidesꢀ(NO
1,2]ꢀandꢀareꢀtoxicꢀtoꢀhumanꢀhealth.ꢀTheyꢀcontainꢀmainlyꢀnitricꢀ
oxideꢀ (NO)ꢀ andꢀ nitrogenꢀ oxideꢀ (NO ).ꢀ Theꢀ mainꢀ NO ꢀ exhaustꢀ
x
)ꢀareꢀoneꢀofꢀtheꢀmainꢀsourcesꢀofꢀhazeꢀ
2 x
matterꢀ<ꢀ5ꢀmg/m ,ꢀSO ꢀ<ꢀ35ꢀmg/m ,ꢀNO ꢀ<ꢀ50ꢀmg/m ,ꢀoxygenꢀ
[
(O )ꢀ=ꢀ6%).ꢀBecauseꢀofꢀtheirꢀlowꢀcapacityꢀandꢀwideꢀdistribution,ꢀ
2
2
x
theꢀ controlꢀofꢀemissionꢀ fromꢀ small‐scaleꢀindustryꢀboilersꢀisꢀ aꢀ
significantꢀ challengeꢀ [6].ꢀ Temperatureꢀ windowsꢀ ofꢀ typicalꢀ in‐
dustrialꢀboilersꢀthatꢀ canꢀbeꢀusedꢀ forꢀemissionsꢀtreatmentꢀareꢀ
usuallyꢀ lowerꢀ thanꢀ 200ꢀ °C,ꢀ whichꢀ isꢀ unsuitableꢀ forꢀ SCRꢀ andꢀ
non‐SCRꢀ technologies.ꢀ Extremelyꢀ complexꢀ flue‐gasꢀ composi‐
tionsꢀhinderꢀtheꢀSCRꢀapplicationꢀinꢀdifferentꢀindustriesꢀsuchꢀasꢀ
fromꢀ power‐plantꢀ boilers,ꢀ industrialꢀ boilers,ꢀ andꢀ automotiveꢀ
vehiclesꢀisꢀNO,ꢀwhichꢀaccountsꢀforꢀmoreꢀthanꢀ95%ꢀofꢀtheꢀtotalꢀ
NO
lyticꢀ reductionꢀ (SCR)ꢀ [4]ꢀ andꢀ non‐SCRꢀ withꢀ ammonia‐basedꢀ
reducingꢀagentsꢀ[5].ꢀHowever,ꢀtheseꢀtwoꢀtechnologiesꢀmayꢀbeꢀ
x
ꢀ[3].ꢀExhaustedꢀNO ꢀgasesꢀareꢀeliminatedꢀbyꢀselectiveꢀcata‐
x
*ꢀCorrespondingꢀauthor.ꢀTel:ꢀ+86‐571‐87953162;ꢀFax:ꢀ+86‐571‐87951616;ꢀE‐mail:ꢀwangzh@zju.edu.cnꢀ
Thisꢀ workꢀ wasꢀ supportedꢀ byꢀ theꢀ Nationalꢀ Naturalꢀ Scienceꢀ Foundationꢀ ofꢀ Chinaꢀ (51422605)ꢀ andꢀ theꢀ Provincialꢀ Naturalꢀ Scienceꢀ Foundationꢀ ofꢀ
Zhejiang,ꢀChinaꢀ(LR16E060001).ꢀ
DOI:ꢀ10.1016/S1872‐2067(17)62860‐2ꢀ|ꢀhttp://www.sciencedirect.com/science/journal/18722067ꢀ|ꢀChin.ꢀJ.ꢀCatal.,ꢀVol.ꢀ38,ꢀNo.ꢀ7,ꢀJulyꢀ2017ꢀ ꢀ ꢀ

ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
1271ꢀ
glass,ꢀ carbonꢀ black,ꢀ andꢀ cement.ꢀ Therefore,ꢀ aꢀ developmentꢀ ofꢀ
2.ꢀ ꢀ Experimentalꢀ
alternativeꢀ economicꢀ andꢀ effectiveꢀ newꢀ technologiesꢀ forꢀ NO
x
ꢀ
controlꢀinꢀcomplexꢀindustryꢀboilersꢀisꢀrequiredꢀurgently.ꢀ
2.1.ꢀ ꢀ Catalystꢀpreparationꢀ
NOꢀisꢀnearlyꢀinsolubleꢀinꢀwaterꢀwhereasꢀtheꢀotherꢀNO
NO ꢀandꢀdinitrogenꢀpentoxideꢀ(N ),ꢀhaveꢀaꢀhigherꢀsolubility.ꢀ
ꢀ canꢀ beꢀ removedꢀ easilyꢀ byꢀ waterꢀ sprayingꢀ withoutꢀ anyꢀ
additivesꢀ[7,8].ꢀNOꢀperoxidationꢀcanꢀachieveꢀtheꢀsimultaneousꢀ
removalꢀofꢀSO ꢀandꢀNO ꢀinꢀaꢀwetꢀflue‐gasꢀdesulfurizationꢀtower.ꢀ
x
ꢀgases,ꢀ
2
O
2 5
Monometallicꢀ metalꢀ oxidesꢀ wereꢀ preparedꢀ byꢀ theꢀ sol‐gelꢀ
method.ꢀ Desirableꢀ amountsꢀ ofꢀ metalꢀ nitratesꢀ (Mn(NO
,ꢀ
49.0%‒51.0%;ꢀ Ce(NO ·6H O,ꢀ ≥99.0%;ꢀ Fe(NO ·9H O,ꢀ
≥98.5%;ꢀ Cu(NO ·3H O,ꢀ ≥99.0%;ꢀ Cr(NO ·9H O,ꢀ ≥99.0%;ꢀ
Co(NO ·6H O,ꢀ ≥98.5%;ꢀ allꢀ fromꢀ Sinopharm)ꢀ andꢀ citrateꢀ
(C ·H O,ꢀSinopharm,ꢀ≥99.5%)ꢀwereꢀdissolvedꢀinꢀdeionizedꢀ
N
2
O
5
3 2
)
)
3 3
2
)
3 3
2
2
x
)
3 2
2
)
3 3
2
Low‐temperatureꢀ oxidationꢀ byꢀ ozoneꢀ combinedꢀ withꢀ wetꢀ
flue‐gasꢀ desulfurization,ꢀ whichꢀ hasꢀ beenꢀ studiedꢀ byꢀ research‐
3
)
2
2
H
6 8
O
7
2
ers,ꢀmayꢀbeꢀaꢀgoodꢀchoiceꢀtoꢀdealꢀwithꢀNO
casesꢀ[3,9‒12].ꢀPreviousꢀworkꢀ[13,14]ꢀhasꢀfocusedꢀonꢀtheꢀcon‐
versionꢀofꢀNOꢀtoꢀN ,ꢀwhichꢀisꢀdefinedꢀasꢀNOꢀdeepꢀoxidation.ꢀ
Accordingꢀtoꢀtheꢀmechanismꢀ[13],ꢀdeepꢀoxidationꢀisꢀaꢀslowꢀre‐
actionꢀthatꢀrequiresꢀmoreꢀthanꢀaꢀ3‒5‐sꢀresidenceꢀandꢀaꢀlargeꢀ
x
ꢀremovalꢀinꢀspecialꢀ
waterꢀtoꢀ achieveꢀaꢀ 0.5ꢀmol/Lꢀ solution.ꢀAꢀ citrateꢀsolutionꢀwasꢀ
pouredꢀ intoꢀ theꢀ metal‐nitrateꢀ solutionꢀ withꢀ aꢀ cit‐
rate/metal‐nitrateꢀ molarꢀ ratioꢀ ofꢀ 2.ꢀ Afterꢀ 48ꢀ hꢀ stirring,ꢀ theꢀ
mixedꢀsolutionꢀwasꢀtransferredꢀtoꢀanꢀovenꢀovernightꢀatꢀ110ꢀ°C.ꢀ
Theꢀresultantꢀfluffyꢀgelꢀwasꢀcalcinedꢀatꢀ400ꢀ°Cꢀforꢀ3ꢀhꢀinꢀaꢀtubeꢀ
furnaceꢀinꢀairꢀatꢀ5ꢀ°C/min.ꢀ
Hereafter,ꢀSAꢀ(Sinopharm,ꢀ2‒3ꢀmmꢀdiameter)ꢀwasꢀusedꢀasꢀ
theꢀsupport.ꢀBimetallicꢀoxidesꢀwereꢀloadedꢀonꢀSAꢀbyꢀcoimpreg‐
nation.ꢀ Manganeseꢀ acetate,ꢀ asꢀ aꢀ precursor,ꢀ exhibitedꢀ aꢀ betterꢀ
O
2 5
ozoneꢀ dosageꢀ (O
overallꢀreactionꢀforꢀN
theoretically,ꢀNOꢀcouldꢀbeꢀconvertedꢀcompletelyꢀtoꢀN
anꢀO /NOꢀmolarꢀratioꢀofꢀ1.5.ꢀInꢀtheꢀlatestꢀresearchꢀ[14],ꢀcatalystꢀ
3
/NOꢀ molarꢀ ratioꢀ >ꢀ 2.0).ꢀ Ifꢀ weꢀ considerꢀ theꢀ
ꢀformation,ꢀ2NOꢀ+ꢀ3O ꢀ=ꢀN ꢀ+ꢀ3O ,ꢀ
ꢀwithꢀ
O
2 5
3
2
O
5
2
2 5
O
3
introductionꢀ intoꢀ NOꢀ deepꢀ oxidationꢀ wasꢀ attempted.ꢀ Theꢀ NOꢀ
deep‐oxidationꢀefficiencyꢀcouldꢀbeꢀimprovedꢀsignificantlyꢀwithꢀ
aꢀshortꢀresidenceꢀtimeꢀandꢀaꢀlowerꢀozoneꢀdosageꢀwithꢀcatalystꢀ
introduction.ꢀ
Ozoneꢀisꢀaꢀstrongꢀoxidantꢀ[15]ꢀandꢀitꢀhasꢀbeenꢀusedꢀexten‐
sivelyꢀ inꢀ waterꢀ purificationꢀ [16],ꢀ volatile‐organic‐compoundsꢀ
treatmentꢀ [17],ꢀ andꢀ disinfectionꢀ [18].ꢀ Inꢀ theseꢀ applications,ꢀ
catalystsꢀ wereꢀ addedꢀ toꢀ improveꢀ theꢀ ozonationꢀ efficiencyꢀ
performanceꢀ
Mn(CH
COO)
thanꢀ
manganeseꢀ
nitrateꢀ
[25,28].ꢀ
3
2
·4H
2
Oꢀ(Aladdin,ꢀ≥ 99.0%)ꢀwasꢀdissolvedꢀinꢀ10ꢀmLꢀ
ofꢀdeionizedꢀwaterꢀwithꢀtheꢀdesiredꢀmetalꢀnitrate.ꢀThenꢀ10ꢀgꢀofꢀ
SAꢀwasꢀimmersedꢀinꢀtheꢀmixedꢀsolution,ꢀandꢀleftꢀunstirredꢀatꢀ
roomꢀtemperatureꢀforꢀ24ꢀh.ꢀTheꢀMnꢀloadingꢀwasꢀfixedꢀatꢀ5ꢀwt%,ꢀ
andꢀtheꢀdopedꢀmetalꢀwasꢀfixedꢀatꢀanꢀM/Mnꢀmolarꢀratioꢀofꢀ1/5,ꢀ
whereꢀMꢀrepresentsꢀtheꢀdopedꢀmetal.ꢀTheꢀcoatedꢀSAꢀwasꢀdriedꢀ
atꢀ 110ꢀ °Cꢀforꢀ 12ꢀhꢀ andꢀcalcinedꢀ atꢀ400ꢀ°Cꢀ forꢀ3ꢀhꢀinꢀairꢀ atꢀ5ꢀ
°C/min.ꢀCatalystꢀsamplesꢀwereꢀlabelledꢀM‐Mn/SA.ꢀ
[19‒21].ꢀ Transition‐metalꢀ oxidesꢀ displayedꢀ anꢀ excellentꢀ per‐
formanceꢀ forꢀ catalyticꢀ ozonationꢀ inꢀ previousꢀ studiesꢀ [22‒24].ꢀ
Manganeseꢀ oxidesꢀ exhibitedꢀ theꢀ bestꢀ activityꢀ amongꢀ theseꢀ
transition‐metalꢀ oxidesꢀ becauseꢀ ofꢀ theirꢀ variousꢀ oxidationꢀ
statesꢀandꢀhighꢀoxygen‐storageꢀcapacity.ꢀTheꢀtransitionꢀofꢀoxi‐
dationꢀ statesꢀ duringꢀtheꢀredoxꢀreactionꢀ cyclesꢀplaysꢀ aꢀcriticalꢀ
roleꢀinꢀcatalyticꢀozonation.ꢀ
2.2.ꢀ ꢀ Catalystꢀcharacterizationꢀ
Theꢀ X‐rayꢀ diffractionꢀ (XRD)ꢀ patternsꢀ ofꢀ theꢀ monometallicꢀ
catalystsꢀ andꢀ theꢀ manganese‐oxide‐basedꢀ catalystsꢀ wereꢀ rec‐
ordedꢀ onꢀ aꢀ Rigakuꢀ D/maxꢀ 2550PCꢀ diffractometerꢀ atꢀ 4°/minꢀ
Inꢀourꢀlatestꢀwork,ꢀmanganeseꢀoxidesꢀthatꢀwereꢀsupportedꢀ
onꢀsphericalꢀaluminaꢀ(SA)ꢀexhibitedꢀexcellentꢀperformanceꢀforꢀ
catalyticꢀ NOꢀ deepꢀ oxidationꢀ byꢀ ozone.ꢀ Anꢀ increaseꢀ inꢀ NOꢀ
deep‐oxidationꢀefficiencyꢀresultedꢀforꢀaꢀstoichiometricꢀratioꢀofꢀ
usingꢀCu‐K
Nitrogenꢀ (N
ordedꢀusingꢀMicromeriticsꢀASAPꢀ2020ꢀequipmentꢀatꢀaꢀliquidꢀN
temperatureꢀ(‒196ꢀ°C).ꢀTheꢀsamplesꢀwereꢀdegassedꢀatꢀ200ꢀ°Cꢀ
forꢀ5ꢀhꢀpriorꢀtoꢀanalysis.ꢀSpecificꢀsurfaceꢀareasꢀwereꢀcalculatedꢀ
byꢀtheꢀBrunauer‐Emmett‐Tellerꢀ(BET)ꢀmethod.ꢀTheꢀtotalꢀporeꢀ
volumeꢀ andꢀ averageꢀ poreꢀ diameterꢀ wereꢀ obtainedꢀ fromꢀ
pore‐sizeꢀ distributionsꢀ byꢀ usingꢀ theꢀ Barret‐Joyner‐Halendaꢀ
(BJH)ꢀ methodꢀ toꢀ desorbꢀ theꢀ cumulativeꢀ surfaceꢀ areaꢀ ofꢀ theꢀ
pores.ꢀ
α
ꢀradiation.ꢀ
2
)ꢀ adsorption‐desorptionꢀ isothermsꢀ wereꢀ rec‐
2
ꢀ
O
3
/NOꢀ=ꢀ1.5.ꢀCatalystꢀstabilitiesꢀandꢀtheꢀresistanceꢀtoꢀSO ꢀandꢀ
2
waterꢀ vapor,ꢀ whichꢀ coexistꢀ inꢀ flueꢀ gas,ꢀ needꢀ toꢀ beꢀ improved.ꢀ
Basedꢀ onꢀ previousꢀ catalytic‐ozonationꢀ studiesꢀ [14],ꢀ catalyticꢀ
activityꢀisꢀrelatedꢀtoꢀtheꢀadsorptionꢀabilityꢀofꢀreactantsꢀandꢀtheꢀ
ozoneꢀdecompositionꢀactivity.ꢀTherefore,ꢀtheꢀobjectiveꢀofꢀcata‐
x
lystꢀmodificationꢀisꢀtoꢀincreaseꢀtheꢀNO ꢀadsorptionꢀabilityꢀandꢀ
improveꢀozoneꢀdecomposition.ꢀ
Hydrogenꢀ temperature‐programmedꢀ reductionꢀ (H ‐TPR)ꢀ
2
Theꢀredoxꢀabilitiesꢀofꢀmanganeseꢀoxidesꢀcanꢀbeꢀenhancedꢀbyꢀ
combiningꢀthemꢀwithꢀotherꢀmetalꢀoxidesꢀ[25].ꢀTheꢀNO ꢀadsorp‐
analysisꢀ wasꢀ carriedꢀ outꢀ byꢀ usingꢀ anꢀ automaticꢀ tempera‐
ture‐programmedꢀ chemisorptionꢀ analyzerꢀ (Micromeriticsꢀ Au‐
toChemꢀIIꢀ2920).ꢀCatalystꢀsamplesꢀ(50ꢀmg)ꢀwereꢀpretreatedꢀatꢀ
x
tionꢀabilityꢀ[26]ꢀandꢀtheꢀozoneꢀdecompositionꢀactivityꢀ[27]ꢀcanꢀ
alsoꢀbeꢀimproved.ꢀThus,ꢀfiveꢀmetalꢀoxidesꢀwereꢀselectedꢀtoꢀloadꢀ
theꢀmanganeseꢀ(Mn)‐basedꢀcatalystꢀthatꢀwasꢀsupportedꢀonꢀSAꢀ
toꢀ investigateꢀ theꢀ catalyticꢀ activity.ꢀ Theꢀ catalyticꢀ activitiesꢀ ofꢀ
monometallicꢀmetalꢀoxidesꢀhaveꢀnotꢀbeenꢀdetermined.ꢀInꢀthisꢀ
work,ꢀoxidesꢀofꢀMn,ꢀcobaltꢀ(Co),ꢀceriumꢀ(Ce),ꢀironꢀ(Fe),ꢀcopperꢀ
200ꢀ°Cꢀinꢀheliumꢀforꢀ1ꢀhꢀpriorꢀtoꢀtesting.ꢀSubsequently,ꢀH
patternsꢀwereꢀobtainedꢀbyꢀmeasuringꢀtheꢀH ꢀconsumptionꢀfromꢀ
100ꢀtoꢀ800ꢀ°Cꢀatꢀ10ꢀ°C/minꢀinꢀ5%ꢀH ‐95%ꢀargonꢀ(20ꢀmL/min).ꢀ
2
‐TPRꢀ
2
2
X‐rayꢀ photoelectronꢀ spectroscopyꢀ (XPS)ꢀ spectraꢀ wereꢀ col‐
lectedꢀ originallyꢀ onꢀ aꢀ photoelectronꢀ spectrometerꢀ (Thermoꢀ
(Cu),ꢀandꢀchromiumꢀ(Cr)ꢀwereꢀpreparedꢀbyꢀtheꢀsol‐gelꢀmethodꢀ
ScientificꢀEscalabꢀ250Xi)ꢀwithꢀaꢀstandardꢀAl‐K ꢀsourceꢀ(1486.6ꢀ
α
andꢀwereꢀtestedꢀforꢀNOꢀdeepꢀoxidation.ꢀSubsequently,ꢀtheꢀNOꢀ
oxidationꢀabilityꢀofꢀtheꢀmetalꢀoxidesꢀthatꢀwereꢀsupportedꢀonꢀSAꢀ
wasꢀinvestigated.ꢀ
eV).ꢀ Allꢀ bindingꢀ energiesꢀ wereꢀ referencedꢀ toꢀ theꢀ Cꢀ 1sꢀ lineꢀ atꢀ
284.5ꢀeVꢀbeforeꢀdataꢀanalysis.ꢀTheseꢀresultsꢀwereꢀtreatedꢀafterꢀ
peak‐fitꢀprocessing.ꢀ

1272ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
2.3.ꢀ ꢀ Activityꢀtestsꢀ
100
Non-catalyst
Monometallic Mn
Monometallic Ce
Monometallic Co
Monometallic Fe
Monometallic Cu
Monometallic Cr
Aꢀdetailedꢀexperimentalꢀprocedureꢀandꢀcalculationꢀformulaꢀ
canꢀbeꢀfoundꢀinꢀourꢀpreviousꢀworkꢀ[14].ꢀNOꢀdeepꢀoxidationꢀwasꢀ
carriedꢀoutꢀinꢀaꢀfixed‐bedꢀreactorꢀwithꢀanꢀinternalꢀdiameterꢀofꢀ
80
5
.8ꢀmmꢀandꢀaꢀlengthꢀofꢀ120ꢀmm.ꢀTheꢀtotalꢀgasꢀflowꢀrateꢀwasꢀ2ꢀ
6
0
0
L/min,ꢀwithꢀanꢀoxygenꢀconcentrationꢀofꢀ2.5ꢀvol%,ꢀandꢀtheꢀcor‐
respondingꢀresidenceꢀtimeꢀinꢀtheꢀreactorꢀwasꢀ~0.12ꢀs.ꢀTheꢀini‐
3
tialꢀ NOꢀ concentrationꢀ wasꢀ fixedꢀ atꢀ 410ꢀ mg/m ꢀ withꢀ aꢀ smallꢀ
4
amountꢀofꢀresidualꢀNO
2
ꢀinꢀtheꢀcylinder.ꢀTheꢀO
3
‐O ꢀmixtureꢀthatꢀ
2
wasꢀ generatedꢀ fromꢀaꢀdielectric‐barrier‐dischargeꢀozoneꢀ gen‐
eratorꢀ(HTU‐500,ꢀ1ꢀg/h,ꢀAZCO,ꢀCanada)ꢀwasꢀinjectedꢀseparatelyꢀ
20
0
intoꢀtheꢀcatalystꢀbedꢀtoꢀavoidꢀaꢀreactionꢀbetweenꢀtheꢀNOꢀandꢀO
beforeꢀcatalystꢀcontact.ꢀTheꢀinletꢀandꢀresidualꢀozoneꢀconcentra‐
tionsꢀ wereꢀ detectedꢀ byꢀ aꢀ high‐ꢀ (BMT‐964ꢀ BT,ꢀ OSTIꢀ Inc.;ꢀ
3
ꢀ
0
.0
0.5
1.0
1.5
2.0
2.5
3.0
3
3
0
‒200000ꢀ mg/m ,ꢀ ±100ꢀ mg/m )ꢀ andꢀ aꢀ low‐concentrationꢀ
O
3
/NO
3
ozoneꢀanalyzerꢀ(Modelꢀ205,ꢀ2BꢀTechnologies;ꢀ0‒430ꢀmg/m ꢀ±ꢀ
.002ꢀ mg/m ),ꢀ respectively.ꢀ Theꢀ initialꢀ andꢀ outletꢀ concentra‐
3
Fig.ꢀ1.ꢀNOꢀdeep‐oxidationꢀefficienciesꢀvariedꢀwithꢀO /NOꢀmolarꢀratiosꢀ
byꢀmonometallicꢀcatalysts.ꢀ
3
0
tionsꢀofꢀtheꢀgasꢀcomponents,ꢀincludingꢀNO,ꢀNO
measuredꢀbyꢀusingꢀaꢀFourierꢀtransformꢀinfraredꢀspectroscopyꢀ
gasꢀanalyzerꢀ(Gasmet,ꢀO ꢀwasꢀmeasuredꢀbyꢀtheꢀexternalꢀoxygenꢀ
analyzer).ꢀBecauseꢀofꢀtheꢀshortꢀlifetimeꢀofꢀN ,ꢀnoꢀcommercialꢀ
instrumentꢀexistsꢀtoꢀmeasureꢀtheꢀquantitativeꢀN ꢀconcentra‐
tion.ꢀPreviousꢀworkꢀ[13,14]ꢀhasꢀindicatedꢀthatꢀN ꢀisꢀtheꢀonlyꢀ
remainingꢀproductꢀduringꢀdeepꢀoxidation.ꢀTheꢀcatalyticꢀactivityꢀ
wasꢀexpressedꢀbyꢀtheꢀresidualꢀconcentrationꢀofꢀNOꢀ+ꢀNO .ꢀTheꢀ
)ꢀ wasꢀ calculatedꢀ
2 2
,ꢀandꢀO ,ꢀwereꢀ
2
Therefore,ꢀ Mnꢀ displayedꢀ theꢀ bestꢀ performanceꢀ amongꢀ theseꢀ
transition‐metalꢀoxides,ꢀasꢀwasꢀexpected.ꢀ ꢀ
O
2 5
O
2 5
ItꢀhasꢀbeenꢀprovenꢀthatꢀNOꢀisꢀtransferredꢀmainlyꢀtoꢀNO
anꢀO /NOꢀmolarꢀratioꢀthatꢀisꢀlowerꢀthanꢀ1.0ꢀ[3].ꢀButꢀtheꢀconver‐
sionꢀefficiencyꢀofꢀNOꢀandꢀNO ꢀtoꢀN ꢀincreasedꢀgraduallyꢀfromꢀ
aꢀlowerꢀO /NOꢀmolarꢀratio.ꢀEspeciallyꢀforꢀmonometallicꢀMn,ꢀtheꢀ
conversionꢀefficiencyꢀreachedꢀnearlyꢀ15%ꢀatꢀanꢀO /NOꢀmolarꢀ
ratioꢀofꢀ0.9.ꢀUnderꢀthisꢀcondition,ꢀNOꢀwasꢀnotꢀconvertedꢀcom‐
pletelyꢀtoꢀNO .ꢀThisꢀresultꢀindicatesꢀthatꢀtheꢀNO ꢀfromꢀNOꢀoxi‐
dationꢀwasꢀoxidizedꢀ partlyꢀbyꢀO ꢀtoꢀpromoteꢀN ꢀformation,ꢀ
whichꢀ precedesꢀ theꢀ totalꢀ oxidizationꢀ ofꢀ NO.ꢀ Asꢀ mentionedꢀ
above,ꢀtheꢀactivationꢀenergyꢀforꢀNO ꢀoxidationꢀisꢀmuchꢀhigherꢀ
thanꢀthatꢀforꢀNOꢀoxidation.ꢀSoꢀN ꢀcannotꢀformꢀbeforeꢀNOꢀisꢀ
convertedꢀ completelyꢀ intoꢀ NO ꢀ inꢀ theꢀ homogeneousꢀ gaseousꢀ
reactorꢀwithoutꢀcatalysts.ꢀThisꢀphenomenonꢀindicatesꢀthatꢀtheꢀ
presenceꢀofꢀcatalystꢀreducesꢀtheꢀactivationꢀenergyꢀofꢀNO ꢀoxi‐
dation,ꢀandꢀresultsꢀinꢀNO ꢀandꢀNOꢀoxidation.ꢀ
TheꢀN ꢀformationꢀefficiencyꢀreachedꢀtheꢀhighestꢀlevelꢀafterꢀ
anꢀO /NOꢀmolarꢀratioꢀofꢀ2.0ꢀasꢀshownꢀinꢀFig.ꢀ1.ꢀAlthoughꢀN
2
ꢀwithꢀ
O
2 5
3
2
O
2 5
2
3
conversionꢀ efficiencyꢀ (NOꢀ andꢀ NO ꢀ toꢀ N O
2 2 5
3
fromꢀaꢀpreviousꢀequationꢀ[14].ꢀ
Firstly,ꢀ0.3ꢀgꢀofꢀmonometallicꢀmetalꢀoxidesꢀwereꢀarrangedꢀinꢀ
theꢀ deep‐oxidationꢀ reactorꢀ withoutꢀ anyꢀ dilution.ꢀ Theꢀ conver‐
2
2
3
O
2 5
sionꢀ efficienciesꢀ atꢀ 70ꢀ °Cꢀ variedꢀ withꢀ O /NOꢀ ratio.ꢀ Allꢀ resultsꢀ
3
wereꢀrecordedꢀafterꢀaꢀ10‐minꢀstabilizationꢀperiod.ꢀThen,ꢀ2.2ꢀgꢀ
ofꢀmanganeseꢀoxide‐basedꢀcatalystsꢀwereꢀarrangedꢀinꢀtheꢀsameꢀ
reactor,ꢀandꢀtheꢀcatalyticꢀreactionꢀtemperatureꢀwasꢀfixedꢀatꢀ100ꢀ
2
O
2 5
2
°
C,ꢀ unlessꢀ otherwiseꢀ noted.ꢀ Theꢀ ozoneꢀ injectionꢀ amountꢀ wasꢀ
3
fixedꢀatꢀ642ꢀmg/m ꢀtoꢀensureꢀanꢀO
3
/NOꢀmolarꢀratioꢀofꢀ1.5.ꢀBe‐
2
causeꢀ ofꢀ theꢀ sphericalꢀ design,ꢀ someꢀ spaceꢀ wasꢀ excludedꢀ be‐
tweenꢀtheseꢀcatalystꢀpellets.ꢀThisꢀresultedꢀinꢀaꢀlongerꢀcatalyticꢀ
reactionꢀtimeꢀ(0.12ꢀs)ꢀorꢀgas‐solidꢀcontactꢀtime,ꢀcomparedꢀwithꢀ
traditionalꢀpowderꢀcatalysts,ꢀwhichꢀareꢀbelievedꢀtoꢀplayꢀaꢀcriti‐
calꢀroleꢀtoꢀimproveꢀperformance.ꢀ
2
O
2 5
3
2 5
O ꢀ
formationꢀwasꢀpromotedꢀsignificantlyꢀcomparedꢀwithꢀtheꢀblankꢀ
results,ꢀ theꢀ ozoneꢀ usageꢀ wasꢀ stillꢀ asꢀ highꢀ asꢀ before.ꢀ Theꢀ re‐
quiredꢀ residenceꢀ timeꢀ wasꢀ reducedꢀ byꢀ theseꢀ catalysts.ꢀ Thisꢀ
3
.ꢀ ꢀ Resultsꢀandꢀdiscussionꢀ
resultꢀindicatesꢀthatꢀnotꢀonlyꢀtheꢀN
theꢀO ꢀdecompositionꢀrateꢀwereꢀacceleratedꢀbyꢀtheseꢀcatalysts,ꢀ
whichꢀ resultedꢀ inꢀ tooꢀ muchꢀ O ꢀ beingꢀ decomposedꢀ withoutꢀ aꢀ
2
O
5
ꢀformationꢀrate,ꢀbutꢀalsoꢀ
3
3.1.ꢀ ꢀ Catalyticꢀactivityꢀwithꢀmonometallicꢀcatalystsꢀ
3
participatingꢀ reaction.ꢀ Itꢀ isꢀ believedꢀ thatꢀ theseꢀ resultsꢀ canꢀ beꢀ
attributedꢀ toꢀ aꢀ lessꢀ specificꢀ surfaceꢀ areaꢀ andꢀ aꢀ compactꢀ ar‐
rangement.ꢀTheꢀformerꢀisꢀunfavorableꢀforꢀreactantꢀadsorptionꢀ
andꢀozoneꢀdecomposition.ꢀTheꢀlatterꢀwouldꢀleadꢀtoꢀanꢀaccumu‐
3
.1.1.ꢀ ꢀ Catalyticꢀdeepꢀoxidationꢀresultsꢀ
Theꢀcatalyticꢀactivitiesꢀofꢀmonometallicꢀcatalystsꢀwereꢀtest‐
edꢀ atꢀ 70ꢀ °Cꢀ asꢀ weꢀ haveꢀ foundꢀ thisꢀ reactionꢀ optimalꢀ forꢀ N
2 5
O ꢀ
formationꢀ[13].ꢀTheꢀresultsꢀareꢀshownꢀinꢀFig.ꢀ1.ꢀTheꢀpresenceꢀofꢀ
aꢀcatalystꢀenhancedꢀtheꢀNOꢀdeepꢀoxidationꢀcomparedꢀwithꢀtheꢀ
blankꢀresults.ꢀOverall,ꢀtheꢀcatalyticꢀactivityꢀdecreasedꢀasꢀMnꢀ>ꢀ
Feꢀ>ꢀCrꢀ>ꢀCoꢀ>ꢀCuꢀ>ꢀCe.ꢀMnꢀisꢀtheꢀbestꢀcatalystꢀforꢀozoneꢀde‐
compositionꢀ[18,29].ꢀManganeseꢀoxidesꢀshowedꢀexcellentꢀper‐
lationꢀofꢀreactionꢀintermediatesꢀandꢀaꢀrapidꢀO
whichꢀinhibitsꢀtheꢀcatalyticꢀreactionꢀprogressꢀ[14,21].ꢀAlthoughꢀ
theꢀ activationꢀ energyꢀ forꢀ NO ꢀ oxidationꢀ hasꢀ beenꢀ reducedꢀ toꢀ
someꢀextentꢀbyꢀcatalystꢀaddition,ꢀitꢀisꢀstillꢀhigherꢀthanꢀthatꢀofꢀ
NOꢀoxidation,ꢀwhichꢀcanꢀbeꢀverifiedꢀbyꢀtheꢀfactꢀthatꢀN ꢀcannotꢀ
beꢀdetectedꢀforꢀanꢀO /NOꢀmolarꢀratioꢀofꢀ0.5.ꢀ
Theꢀgoalꢀofꢀcatalystꢀintroductionꢀisꢀtoꢀachieveꢀaꢀtheoreticalꢀ
thresholdꢀatꢀanꢀO /NOꢀmolarꢀratioꢀofꢀ1.5.ꢀWeꢀintroducedꢀspher‐
3
ꢀdecomposition,ꢀ
2
O
2 5
formanceꢀ forꢀ NO
32].ꢀCatalyticꢀNOꢀdeepꢀoxidationꢀcanꢀbeꢀregardedꢀasꢀaꢀcombi‐
nationꢀ ofꢀ ozoneꢀ decomposition,ꢀ NO ꢀ storage,ꢀ andꢀ oxidation.ꢀ
x
ꢀ storageꢀ [30],ꢀ oxidationꢀ [31],ꢀ andꢀ reductionꢀ
3
[
x
3

ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
1273ꢀ
icalꢀarrangementsꢀtoꢀextendꢀtheꢀreactantꢀresidenceꢀtimeꢀonꢀtheꢀ
catalystꢀsurface,ꢀandꢀtoꢀallowꢀforꢀsufficientꢀtimeꢀtoꢀfacilitateꢀNO
2
ꢀ
oxidation.ꢀHereafter,ꢀSAꢀwasꢀusedꢀasꢀaꢀsupportꢀandꢀdopedꢀwithꢀ
otherꢀtransition‐metalꢀoxides.ꢀ
2 3
Cr O
3
.1.2.ꢀ ꢀ Crystalliteꢀstructureꢀofꢀmonometallicꢀcatalystsꢀ
TheꢀXRDꢀpatternsꢀofꢀtheꢀmonometallicꢀcatalystsꢀareꢀshownꢀ
CoO
CuO
inꢀFig.ꢀ2.ꢀForꢀeachꢀsample,ꢀonlyꢀoneꢀcrystallineꢀphaseꢀwasꢀde‐
tected.ꢀ CeO ,ꢀ Mn ,ꢀ Fe ,ꢀ CuO,ꢀ CoO,ꢀ andꢀ Cr ꢀ diffractionꢀ
peaksꢀareꢀvisibleꢀforꢀtheꢀmonometallicꢀCe,ꢀMn,ꢀFe,ꢀCu,ꢀCo,ꢀandꢀCrꢀ
catalysts,ꢀrespectively.ꢀ
2
O
2 3
O
2 3
O
2 3
2 3
Fe O
2 3
Mn O
3
.1.3.ꢀ ꢀ Porousꢀstructureꢀparametersꢀofꢀmonometallicꢀcatalystsꢀ
TheꢀN ꢀadsorption‐desorptionꢀisothermsꢀandꢀpore‐sizeꢀdis‐
tributionꢀcurvesꢀofꢀtheꢀmonometallicꢀcatalystsꢀareꢀshownꢀinꢀFig.ꢀ
.ꢀ Catalystsꢀ Mn,ꢀ Co,ꢀ Fe,ꢀ Cu,ꢀ andꢀ Crꢀ showedꢀ aꢀ characteristicꢀ
type‐IVꢀisothermꢀaccordingꢀtoꢀtheꢀIUPACꢀclassification,ꢀwithꢀtheꢀ
H3ꢀhysteresisꢀloopꢀatꢀP/P ꢀ=ꢀ0.8−1.0ꢀindicatingꢀaꢀmesoporousꢀ
2
CeO
2
3
20
30
40
50
60
70
80
2
0
Fig.ꢀ2.ꢀXRDꢀpatternsꢀofꢀmonometallicꢀcatalysts.ꢀ
structureꢀ [33,34].ꢀ Thisꢀ correspondsꢀ toꢀ theꢀ broaderꢀ pore‐sizeꢀ
distributionꢀandꢀslit‐shapedꢀporesꢀ[35].ꢀAꢀcharacteristicꢀtype‐IVꢀ
isothermꢀwithꢀanꢀH1ꢀhysteresisꢀloopꢀatꢀP/P
bleꢀ forꢀ theꢀ monometallicꢀ Ce,ꢀ whichꢀ exhibitedꢀ aꢀ narrowꢀ
pore‐sizeꢀdistribution.ꢀ
0
ꢀ=ꢀ0.4−1.0ꢀwasꢀvisi‐
Ceriaꢀ(CeO
anꢀexcellentꢀsupportꢀorꢀpromotor.ꢀHowever,ꢀliteratureꢀ[18,36]ꢀ
hasꢀshownꢀthatꢀCeO ꢀexhibitsꢀaꢀveryꢀlowꢀactivityꢀforꢀozoneꢀde‐
2
)ꢀhasꢀaꢀlargeꢀoxygenꢀcapacityꢀandꢀisꢀalwaysꢀusedꢀasꢀ
2
TheꢀBETꢀsurfaceꢀarea,ꢀporeꢀvolume,ꢀandꢀaverageꢀporeꢀdiam‐
eterꢀ areꢀ shownꢀ inꢀ Tableꢀ 1.ꢀ Monometallicꢀ Crꢀ hadꢀ theꢀ highestꢀ
composition.ꢀ Thus,ꢀ theꢀ surfaceꢀ areaꢀ ofꢀ monometallicꢀ Ceꢀ wasꢀ
similarꢀtoꢀMn,ꢀbutꢀitꢀwasꢀtheꢀworstꢀcatalystꢀforꢀNOꢀdeepꢀoxida‐
tion.ꢀ
2
3
surfaceꢀareaꢀ(38.1ꢀm /g)ꢀandꢀporeꢀvolumeꢀ(0.27ꢀcm /g).ꢀMon‐
2
ometallicꢀCuꢀhadꢀtheꢀlowestꢀsurfaceꢀareaꢀ(0.5ꢀm /g)ꢀandꢀporeꢀ
3
volumeꢀ (0.002ꢀ cm /g).ꢀTheꢀhigherꢀsurfaceꢀareaꢀandꢀporeꢀvol‐
3.2.ꢀ ꢀ CatalyticꢀactivityꢀwithꢀMn‐basedꢀbimetallicꢀcatalystsꢀ
umeꢀimproveꢀtheꢀadsorptionꢀability.ꢀThisꢀresultꢀcouldꢀexplainꢀ
whyꢀ monometallicꢀ Crꢀ displayedꢀ aꢀ betterꢀ catalyticꢀ activity,ꢀ
whereasꢀmonometallicꢀCuꢀshowedꢀaꢀveryꢀlowꢀcatalyticꢀactivity.ꢀ
3.2.1.ꢀ ꢀ Catalyticꢀdeep‐oxidationꢀresultsꢀ
Aluminaꢀhasꢀbeenꢀusedꢀwidelyꢀasꢀaꢀcatalystꢀsupportꢀbecauseꢀ
1
00
50
10
0.20
0.20
0.020
Mn
Ce
Co
0.15
0.15
0.015
4
0
8
6
4
2
0
7
5
2
5
0
5
0
0
.10
0
.10
0.010
0.005
0.05
30
0.05
0.00
0.00
0
10
20
30
40
50
60
70
0.000
0
10
20
30
Pore diameter (nm)
40
50
60
70
0
10
20
30
40
50
60
70
Pore diameter (nm)
Pore diameter (nm)
20
10
0
0
.0
0.2
0.4
0.6
0.8
1.0
0
.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P )
0
Relative pressure (P/P
0)
Relative pressure (P/P )
0
1
00
2.0
1.5
1.0
0.5
0.0
200
150
100
50
0
.20
0.8
0.6
Fe
Cu
Cr
0
.0020
0.15
0.0015
7
5
2
5
0
5
0
0.10
0.0010
0.4
0.2
0.0005
0.05
0.0000
0.00
0.0
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
0
10
20
30
40
50
60
70
Pore diameter (nm)
Pore diameter (nm)
Pore diameter (nm)
0
0.0
0
.0
0.2
0.4
0.6
0.8
1.0
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.4
0.6
0.8
1.0
Relative pressure (P/P0)
Relative pressure (P/P0)
Relative pressure (P/P0)
Fig.ꢀ3.ꢀN ꢀadsorption‐desorptionꢀisothermsꢀandꢀpore‐sizeꢀdistributionꢀcurvesꢀofꢀmonometallicꢀcatalysts.ꢀ
2

1274ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
Tableꢀ1ꢀ
6
5
4
3
2
1
0
0
0
0
0
0
0
Pore‐structureꢀparametersꢀofꢀmonometallicꢀcatalysts.ꢀ
Ce-Mn/SA
Fe-Mn/SA
Cu-Mn/SA
Cr-Mn/SA
Co-Mn/SA
a
BETꢀsurfaceꢀ ꢀ
areaꢀ(m /g)ꢀ
Poreꢀvolumeꢀ ꢀ
(cm /g)ꢀ
Averageꢀporeꢀ ꢀ
diameterꢀ ꢀ(nm)ꢀ
Catalystꢀ
2
3
b
Mnꢀ
Ceꢀ
Coꢀ
Feꢀ
Cuꢀ
15.4ꢀ
18.2ꢀ
ꢀ 1.6ꢀ
12.2ꢀ
ꢀ 0.5ꢀ
38.1ꢀ
0.13ꢀ
0.06ꢀ
0.01ꢀ
0.15ꢀ
ꢀ 0.002ꢀ
0.27ꢀ
27.0ꢀ
10.3ꢀ
38.8ꢀ
45.1ꢀ
33.7ꢀ
23.2ꢀ
Crꢀ
a
ꢀ
ꢀ
BJHꢀdesorptionꢀcumulativeꢀporeꢀvolume.ꢀ
BJHꢀdesorptionꢀaverageꢀporeꢀdiameter.ꢀ
b
ꢀ
ꢀ
ofꢀitsꢀhighꢀsurfaceꢀareaꢀ[17,37].ꢀBimetallicꢀcatalystsꢀthatꢀwereꢀ
supportedꢀ onꢀ SAꢀ wereꢀ preparedꢀ toꢀ conductꢀ catalyticꢀ
deep‐oxidationꢀ tests.ꢀ Theꢀ bimetallicꢀ catalyticꢀ activitiesꢀ areꢀ
shownꢀinꢀFig.ꢀ4.ꢀOzoneꢀwasꢀinjectedꢀatꢀtimeꢀzero.ꢀTheꢀconcen‐
0
20
40
60
80
100
120
Time (min)
Fig.ꢀ5.ꢀConcentrationꢀofꢀresidualꢀozoneꢀafterꢀcatalyticꢀNOꢀdeepꢀoxidationꢀ
forꢀMn‐basedꢀcatalystsꢀdopedꢀwithꢀdifferentꢀmetalꢀoxides.ꢀ
2
trationꢀofꢀNOꢀ+ꢀNO ꢀdecreasedꢀimmediatelyꢀafterꢀozoneꢀinjec‐
tion,ꢀwhichꢀindicatedꢀtheꢀformationꢀofꢀN
lysts,ꢀFe‐Mn/SAꢀandꢀCe‐Mn/SAꢀdisplayedꢀtheꢀhighestꢀcatalyticꢀ
activity.ꢀAfterꢀ20ꢀminꢀofꢀozoneꢀinjection,ꢀtheꢀconcentrationsꢀofꢀ
2 5
O
.ꢀAmongꢀtheseꢀcata‐
newlyꢀformedꢀN
wasꢀ desorbedꢀ slowlyꢀ untilꢀ saturationꢀ adsorption.ꢀ Theꢀ met‐
al‐atomꢀvalenceꢀstateꢀwasꢀincreasedꢀbyꢀozoneꢀinjectionꢀbeforeꢀ
reactingꢀwithꢀtheꢀreactantsꢀ(NO,ꢀNO
centrationsꢀofꢀNOꢀ+ꢀNO ꢀdecreasedꢀcontinuouslyꢀatꢀtheꢀbegin‐
ningꢀofꢀozoneꢀinjectionꢀandꢀthenꢀreachedꢀequilibrium.ꢀThisꢀalsoꢀ
explainsꢀ whyꢀ theꢀ concentrationsꢀ ofꢀ residualꢀ ozoneꢀ increasedꢀ
continuouslyꢀinitiallyꢀasꢀshownꢀinꢀFig.ꢀ5.ꢀ ꢀ
TheꢀconcentrationꢀofꢀresidualꢀozoneꢀinꢀFig.ꢀ5ꢀwasꢀnotꢀofꢀtheꢀ
sameꢀorderꢀasꢀtheꢀconcentrationꢀofꢀNOꢀ+ꢀNO ꢀasꢀshownꢀinꢀFig.ꢀ4.ꢀ
Catalystꢀ Cu‐Mn/SAꢀ exhibitedꢀ theꢀ highestꢀ concentrationꢀ ofꢀ re‐
sidualꢀ ozone,ꢀ whereasꢀ Co‐Mn/SAꢀ exhibitedꢀ theꢀ lowestꢀ ozoneꢀ
concentration.ꢀ Thisꢀ demonstratesꢀthatꢀtheꢀcatalyticꢀ activityꢀofꢀ
theꢀNOꢀdeepꢀoxidationꢀbyꢀozoneꢀwasꢀnotꢀonlyꢀaffectedꢀbyꢀtheꢀ
ozoneꢀdecompositionꢀabilityꢀofꢀtheꢀcatalyst.ꢀTheꢀresultsꢀinꢀFig.ꢀ5ꢀ
2
O ꢀwasꢀadsorbedꢀonꢀtheꢀcatalystꢀsurface,ꢀandꢀ
5
NOꢀ+ꢀNO
2
ꢀtendedꢀtoꢀstabilizeꢀatꢀaꢀconcentrationꢀlowerꢀthanꢀ20ꢀ
2
)ꢀ[14].ꢀTherefore,ꢀtheꢀcon‐
3
3
mg/m ꢀforꢀFe‐Mn/SAꢀandꢀ50ꢀmg/m ꢀforꢀCe‐Mn/SA.ꢀTheꢀcorre‐
spondingꢀ conversionꢀ efficienciesꢀ wereꢀ higherꢀ thanꢀ 95%ꢀ andꢀ
2
88%,ꢀrespectively,ꢀwhichꢀwereꢀallꢀhigherꢀthanꢀtheꢀmonometallicꢀ
catalystꢀMn/SAꢀ(83%).ꢀTheꢀultra‐lowꢀemissionꢀstandardꢀ(NO
x
ꢀ<ꢀ
3
5
0ꢀmg/m )ꢀhasꢀbeenꢀreached.ꢀThisꢀshowsꢀthatꢀcatalystsꢀthatꢀareꢀ
dopedꢀ withꢀ Feꢀ andꢀ Ceꢀ couldꢀ improveꢀ theꢀ NOꢀ deep‐oxidationꢀ
efficiency,ꢀandꢀcouldꢀbeꢀappliedꢀforꢀultra‐lowꢀemissionsꢀcontrol.ꢀ
However,ꢀforꢀtheꢀotherꢀthreeꢀcatalysts,ꢀtheꢀaverageꢀstableꢀcon‐
2
3
2
centrationꢀofꢀNOꢀ+ꢀNO ꢀwasꢀhigherꢀthanꢀ100ꢀmg/m .ꢀTheꢀaver‐
ageꢀ stableꢀ concentrationꢀ ofꢀ NOꢀ +ꢀ NO
2
ꢀ wasꢀ higherꢀ thanꢀ 200ꢀ
3
mg/m ꢀforꢀCo‐Mn/SA.ꢀThisꢀindicatesꢀthatꢀMn/SAꢀdopingꢀwithꢀ
Cu,ꢀCr,ꢀandꢀCoꢀinhibitedꢀtheꢀcatalyticꢀNOꢀdeep‐oxidationꢀactivity.ꢀ
AnotherꢀinterestingꢀphenomenonꢀinꢀFig.ꢀ4ꢀisꢀthatꢀtheꢀstableꢀ
resultsꢀ requiredꢀ aꢀ longꢀ time.ꢀ Especiallyꢀ forꢀ Cr‐Mn/SA,ꢀ
thatꢀrelateꢀtoꢀNO ꢀoxidationꢀdoꢀnotꢀrepresentꢀtheꢀcompleteꢀcat‐
x
alystꢀactivityꢀorderꢀofꢀtheꢀozoneꢀdecomposition.ꢀCo‐Mn/SAꢀdis‐
playedꢀ theꢀ lowestꢀ catalyticꢀ potentialꢀ forꢀ NOꢀ deepꢀ oxidation,ꢀ
withꢀtheꢀlowestꢀozoneꢀresidual.ꢀPerhapsꢀbecauseꢀofꢀthis,ꢀlargeꢀ
Cu‐Mn/SA,ꢀ andꢀ Co‐Mn/SA,ꢀ theꢀ concentrationꢀ ofꢀ NOꢀ +ꢀ NO
waysꢀ showedꢀ aꢀ certainꢀ degreeꢀ ofꢀ volatility.ꢀ Thisꢀ isꢀ becauseꢀ
2
ꢀal‐
amountsꢀofꢀNOꢀ+ꢀNO ꢀthatꢀwereꢀadsorbedꢀonꢀtheꢀcatalystꢀsur‐
2
faceꢀwouldꢀreduceꢀtheꢀoxidizedꢀatomꢀcontinuously,ꢀwhereasꢀtheꢀ
ozoneꢀ residualꢀ wasꢀ insufficientꢀ toꢀ recoverꢀ theꢀ reducedꢀ atom.ꢀ
4
3
3
2
2
1
1
00
50
00
50
00
50
00
Ce-Mn/SA
Fe-Mn/SA
Cu-Mn/SA
Cr-Mn/SA
Co-Mn/SA
2
Thus,ꢀtheꢀconcentrationꢀofꢀNOꢀ+ꢀNO ꢀforꢀCo‐Mn/SAꢀexhibitedꢀaꢀ
largeꢀvolatility.ꢀ
Theꢀresidualꢀozoneꢀconcentrationsꢀwereꢀlessꢀthanꢀ25ꢀmg/m3ꢀ
forꢀCe‐Mn/SAꢀandꢀFe‐Mn/SA.ꢀAsꢀmentionedꢀabove,ꢀtheꢀconcen‐
trationsꢀofꢀNOꢀ+ꢀNO
mg/m ꢀforꢀtheseꢀtwoꢀsamples,ꢀwhichꢀwereꢀcloseꢀtoꢀtwiceꢀtheꢀ
2
ꢀwithoutꢀdeepꢀoxidationꢀwereꢀlessꢀthanꢀ50ꢀ
3
residualꢀozone.ꢀTheꢀequivalenceꢀratioꢀofꢀNO
2
/O
3
ꢀforꢀN
2
O ꢀfor‐
5
mationꢀ wasꢀ 2.0.ꢀ Itꢀ canꢀ beꢀ concludedꢀ thatꢀ ozoneꢀ participatedꢀ
efficientlyꢀinꢀNOꢀdeepꢀoxidation.ꢀ
3
.2.2.ꢀ ꢀ Effectꢀofꢀreactionꢀtemperatureꢀonꢀcatalyticꢀactivityꢀ
TemperatureꢀisꢀcriticalꢀforꢀN ꢀformationꢀ[13]ꢀandꢀcatalystꢀ
activity.ꢀBecauseꢀofꢀtheꢀbetterꢀresistanceꢀtoꢀSO ꢀandꢀwaterꢀva‐
porꢀ [38],ꢀ Ce‐Mn/SAꢀ wasꢀ usedꢀ toꢀ studyꢀ theꢀ effectꢀ ofꢀ reactionꢀ
temperatureꢀ onꢀ catalyticꢀ activity.ꢀ Theꢀ concentrationꢀ ofꢀ NOꢀ +ꢀ
50
O
2 5
2
0
0
20
40
60
80
100
120
Time (min)
2
Fig.ꢀ 4.ꢀ Variationꢀ inꢀ NOꢀ +ꢀ NO ꢀ concentrationꢀ withꢀ reactionꢀ timeꢀ forꢀ
2
NO ꢀvariedꢀwithꢀtimeꢀatꢀdifferentꢀtemperaturesꢀasꢀshownꢀinꢀFig.ꢀ
Mn‐basedꢀcatalystsꢀdopedꢀwithꢀdifferentꢀmetalꢀoxides.ꢀ
6.ꢀ Thereꢀ wasꢀ noꢀ obviousꢀ differenceꢀ betweenꢀ theseꢀ fourꢀ tem‐

ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
1275ꢀ
400
300
200
100
0
Bohmite AlO(OH)
o
Gibbsite Al(OH)
3
7
1
1
0 C
Al2O3
o
00 C
30 C
160 C
Co-Mn/SA
o
o
Cr-Mn/SA
Cu-Mn/SA
Fe-Mn/SA
Ce-Mn/SA
Mn/SA
SA
0
20
40
60
80
100
120
10
20
30
40
50
60
70
80
Time (min)
2
2
Fig.ꢀ6.ꢀNOꢀ+ꢀNO ꢀconcentrationꢀvariationꢀwithꢀtimeꢀatꢀdifferentꢀreaction
Fig.ꢀ7.ꢀXRDꢀpatternsꢀofꢀMn‐basedꢀcatalystsꢀdopedꢀwithꢀdifferentꢀmetalꢀ
temperaturesꢀ(Ce‐Mn/SA).ꢀ
oxides.ꢀ
peraturesꢀ (70,ꢀ 100,ꢀ 130,ꢀ andꢀ 160ꢀ °C).ꢀ Whenꢀ theꢀ temperatureꢀ
increasedꢀfromꢀ100ꢀtoꢀ130ꢀ°C,ꢀtheꢀstableꢀconcentrationꢀofꢀNOꢀ+ꢀ
strateꢀ catalystsꢀ exhibitedꢀ IVꢀ isothermsꢀ withꢀ H1ꢀ hysteresisꢀ
loops.ꢀTheꢀbiggerꢀhysteresisꢀloopꢀatꢀP/P ꢀ=ꢀ0.4‒1.0ꢀindicatesꢀaꢀ
0
NO
2
ꢀwasꢀrelativelyꢀlowerꢀthanꢀthatꢀwithꢀtheꢀformerꢀtwoꢀtem‐
largerꢀsurfaceꢀareaꢀcomparedꢀwithꢀpreviousꢀmonometallicꢀcat‐
alysts.ꢀFig.ꢀ8ꢀandꢀTableꢀ2ꢀshowꢀthatꢀafterꢀimpregnation,ꢀtheꢀsur‐
faceꢀareaꢀdecreasedꢀforꢀallꢀsamples,ꢀexceptꢀforꢀCe‐Mn/SA.ꢀTheseꢀ
metalꢀoxidesꢀwereꢀimmersedꢀintoꢀtheꢀporesꢀofꢀtheꢀSA,ꢀwhichꢀledꢀ
toꢀ theꢀ occupationꢀ ofꢀ someꢀ pores.ꢀ Theꢀ averageꢀ poreꢀ diameterꢀ
increasedꢀ afterꢀ impregnation,ꢀ whichꢀ indicatesꢀ aꢀ decreaseꢀ inꢀ
poreꢀ numbers.ꢀ Theꢀ increaseꢀ inꢀ averageꢀ poreꢀ diameterꢀ mostꢀ
likelyꢀresultsꢀbecauseꢀOHꢀisꢀremovedꢀduringꢀimpregnation,ꢀasꢀisꢀ
shownꢀinꢀtheꢀXRDꢀresults.ꢀAsꢀisꢀshownꢀinꢀFig.ꢀ10ꢀandꢀTableꢀ3,ꢀ
theꢀratioꢀofꢀchemisorbedꢀoxygenꢀspeciesꢀdecreasedꢀsignificantlyꢀ
afterꢀ loadingꢀ withꢀ Mnꢀ (Mn/SAꢀ comparedꢀ withꢀ SA)ꢀ andꢀ de‐
creasedꢀ againꢀ forꢀ theꢀ bimetallicꢀ catalysts.ꢀ Thisꢀ couldꢀ provideꢀ
evidenceꢀ forꢀ theꢀ decreaseꢀ inꢀ OHꢀ duringꢀ impregnation.ꢀ Theꢀ
pore‐blockageꢀ positionꢀ byꢀ OHꢀ ofꢀ theꢀ SAꢀ wasꢀ recovered.ꢀ
peraturesꢀ(70ꢀandꢀ100ꢀ°C).ꢀThisꢀresultꢀcanꢀbeꢀattributedꢀtoꢀtheꢀ
increasingꢀ decompositionꢀ rateꢀ ofꢀ theꢀ intermediates,ꢀ whichꢀ
wouldꢀrecoverꢀmoreꢀactiveꢀsites.ꢀTherefore,ꢀtheꢀcatalyticꢀactivi‐
tyꢀcanꢀbeꢀenhancedꢀtoꢀsomeꢀextentꢀbyꢀanꢀincreaseꢀinꢀtempera‐
ture.ꢀ Theꢀ decreaseꢀ inꢀ concentrationꢀ atꢀ higherꢀ temperaturesꢀ
(
130ꢀ andꢀ 160ꢀ °C)ꢀ becameꢀ slowerꢀ comparedꢀ withꢀ theꢀ lowerꢀ
temperaturesꢀ(70ꢀandꢀ100ꢀ°C).ꢀMoreꢀthanꢀ80ꢀminꢀwasꢀrequiredꢀ
toꢀreachꢀaꢀstableꢀstateꢀatꢀ160ꢀ°C.ꢀAsꢀmentionedꢀabove,ꢀN ꢀhasꢀ
aꢀshortꢀlifeꢀtime,ꢀandꢀitsꢀdecompositionꢀrateꢀwillꢀbeꢀacceleratedꢀ
atꢀaꢀhigherꢀtemperature.ꢀTherefore,ꢀtheꢀN ꢀresidenceꢀtimeꢀonꢀ
theꢀ catalystꢀ surfaceꢀ wasꢀ longerꢀ thanꢀ 0.12ꢀ sꢀ beforeꢀ saturationꢀ
adsorptionꢀwasꢀreached,ꢀwhichꢀmayꢀleadꢀtoꢀN ꢀdecomposi‐
tionꢀtoꢀNO ꢀandꢀNOꢀbeforeꢀitꢀisꢀdesorbedꢀfromꢀtheꢀcatalystꢀsur‐
O
2 5
O
2 5
O
2 5
2
2
face.ꢀ Thisꢀ mayꢀ beꢀ anotherꢀ reasonꢀ forꢀ theꢀ smallꢀ higherꢀ stableꢀ
concentrationꢀcomparedꢀwithꢀthatꢀatꢀ130ꢀ°C.ꢀCatalyticꢀNOꢀdeepꢀ
oxidationꢀbyꢀozoneꢀoverꢀCe‐Mn/SAꢀcanꢀbeꢀappliedꢀinꢀindustrialꢀ
applicationsꢀatꢀaꢀreasonableꢀtemperature.ꢀ
Ce‐Mn/SAꢀ hadꢀ theꢀ highestꢀ surfaceꢀ areaꢀ (318.7ꢀ m /g),ꢀ whichꢀ
mayꢀ resultꢀ inꢀ anꢀ improvedꢀ catalyticꢀ activity.ꢀ Theꢀ increaseꢀ inꢀ
300
250
200
SA
Mn/SA
3.2.3.ꢀ ꢀ Crystalliteꢀstructuresꢀ
Ce-Mn/SA
Fe-Mn/SA
Cu-Mn/SA
Cr-Mn/SA
Co-Mn/SA
TheꢀXRDꢀpatternsꢀofꢀtheseꢀfiveꢀcatalystꢀsamplesꢀareꢀshownꢀ
inꢀFig.ꢀ7.ꢀOnlyꢀtheꢀaluminaꢀdiffractionꢀpeaksꢀareꢀvisibleꢀinꢀtheꢀ
samples,ꢀ whichꢀ indicatesꢀ thatꢀ theꢀ dopedꢀ metalsꢀ wereꢀ highlyꢀ
dispersedꢀ inꢀ theꢀ support.ꢀ Theꢀ diffractionꢀ peaksꢀ ofꢀ gibbsiteꢀ
3
Al(OH) ꢀ disappearedꢀ andꢀ theꢀ diffractionꢀ peaksꢀ ofꢀ bohmiteꢀ
150
AlO(OH)ꢀweakenedꢀafterꢀloadingꢀwithꢀMn.ꢀTheꢀdiffractionꢀpeaksꢀ
ofꢀ bohmiteꢀ AlO(OH)ꢀ almostꢀ disappearedꢀ afterꢀ loadingꢀ withꢀ
bimetallicꢀoxides.ꢀThisꢀindicatesꢀthatꢀimpregnationꢀresultedꢀinꢀ
theꢀremovalꢀofꢀOHꢀspecies.ꢀTheꢀdiffractionꢀpeaksꢀofꢀCu‐Mn/SAꢀ
wereꢀrelativelyꢀhigherꢀthanꢀthatꢀforꢀotherꢀsamples,ꢀwhichꢀindi‐
catedꢀthatꢀcrystallizationꢀoccurred.ꢀ
1
00
5
0
0
0
.0
0.2
0.4
0.6
0.8
1.0
3
.2.4.ꢀ ꢀ Physicochemicalꢀpropertiesꢀ
TheꢀN ꢀadsorption‐desorptionꢀisothermsꢀofꢀtheꢀSAꢀsupportꢀ
andꢀ Mn‐basedꢀ catalystsꢀ inꢀ Fig.ꢀ 8ꢀ showꢀ thatꢀ theꢀ aluminaꢀ sub‐
Relative pressure (P/P
0
)
2
Fig.ꢀ8.ꢀN ꢀadsorption‐desorptionꢀisothermsꢀofꢀMn‐basedꢀcatalysts.ꢀ
2

1276ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
9ꢀ shouldꢀ beꢀ attributedꢀ toꢀ transition‐metalꢀ oxides.ꢀ Ce‐Mn/SA,ꢀ
Tableꢀ2ꢀ
Pore‐structureꢀparametersꢀofꢀMn‐basedꢀcatalysts.ꢀ
Fe‐Mn/SA,ꢀ andꢀ Co‐Mn/SAꢀ exhibitedꢀ littleꢀ differenceꢀ fromꢀ
Mn/SA.ꢀWhenꢀtheseꢀthreeꢀprofilesꢀareꢀoverlapped,ꢀtheꢀreduc‐
tionꢀpeakꢀareaꢀdecreasesꢀslightlyꢀafterꢀdopingꢀwithꢀFe.ꢀThisꢀmayꢀ
beꢀrelatedꢀtoꢀtheꢀreducedꢀsurfaceꢀareaꢀofꢀFe‐Mn/SA.ꢀTheꢀfirstꢀ
reductionꢀ regionꢀ betweenꢀ 100ꢀ andꢀ 450ꢀ °Cꢀ wasꢀ attributedꢀ
a
BETꢀsurfaceꢀareaꢀ Poreꢀvolumeꢀ ꢀ
Averageꢀporeꢀ ꢀ
diameterꢀ ꢀ(nm)ꢀ
Catalystꢀ
2
3
b
(m /g)ꢀ
(cm /g)ꢀ
SAꢀ
Mn/SAꢀ
Ce‐Mn/SAꢀ
Fe‐Mn/SAꢀ
Cu‐Mn/SAꢀ
Cr‐Mn/SAꢀ
294.3ꢀ
268.3ꢀ
318.7ꢀ
228.9ꢀ
264.0ꢀ
281.8ꢀ
246.1ꢀ
0.37ꢀ
0.37ꢀ
0.43ꢀ
0.32ꢀ
0.36ꢀ
0.36ꢀ
0.34ꢀ
4.5ꢀ
5.0ꢀ
4.8ꢀ
5.0ꢀ
4.9ꢀ
4.7ꢀ
4.9ꢀ
mainlyꢀ toꢀ aꢀ two‐stepꢀ reductionꢀ ofꢀ MnO
whereasꢀtheꢀsecondꢀreductionꢀpeakꢀbetweenꢀ600ꢀandꢀ700ꢀ°C,ꢀ
whichꢀ representedꢀ theꢀ reductionꢀ ofꢀ MnO ꢀspeciesꢀwithꢀlargerꢀ
particlesꢀ[40],ꢀalmostꢀdisappeared.ꢀThisꢀresultꢀindicatesꢀthatꢀtheꢀ
amountsꢀofꢀlargerꢀparticlesꢀofꢀMnO ꢀspeciesꢀdecreasedꢀvisiblyꢀ
x 3 4
→Mn O →MnOꢀ [39],ꢀ
x
Co‐Mn/SAꢀ
a
ꢀ
ꢀ
BJHꢀdesorptionꢀcumulativeꢀvolumeꢀofꢀpores.ꢀ
BJHꢀdesorptionꢀaverageꢀporeꢀdiameter.ꢀ
x
b
afterꢀdopingꢀwithꢀtheseꢀmetalꢀoxides.ꢀForꢀtheꢀotherꢀtwoꢀcatalystꢀ
samples,ꢀCu‐Mn/SAꢀandꢀCr‐Mn/SA,ꢀtheꢀreductionꢀpeaksꢀbecameꢀ
sharperꢀ andꢀ theꢀ intensitiesꢀ increasedꢀ significantly.ꢀ Theꢀ firstꢀ
reductionꢀpeakꢀofꢀCu‐Mn/SAꢀoccurredꢀatꢀaꢀlowerꢀtemperatureꢀ
(220ꢀ °C).ꢀ Thisꢀ suggestsꢀ thatꢀ theꢀ introductionꢀ ofꢀ Cuꢀ enhancedꢀ
theꢀredoxꢀabilityꢀbyꢀaꢀstrongꢀsynergisticꢀeffectꢀbetweenꢀMnꢀandꢀ
Cu.ꢀCuꢀpossessedꢀaꢀhigherꢀreducibilityꢀthanꢀMn,ꢀandꢀthereforeꢀ
theꢀ firstꢀreductionꢀpeakꢀ atꢀ220ꢀ°Cꢀcorrespondsꢀmainlyꢀtoꢀtheꢀ
ꢀ
ꢀ
Tableꢀ3ꢀ
BindingꢀenergiesꢀandꢀdistributionꢀofꢀMnꢀandꢀoxygenꢀspecies.ꢀ
3
+
4+
Mn ꢀ
(
Mn ꢀ
O
(eV)ꢀ
α
ꢀ ꢀ
O
(eV)ꢀ
β
ꢀ ꢀ
4
+
Sampleꢀ
Mn /Mnꢀ
β
O /O
eV)ꢀ
—ꢀ
(eV)ꢀ
—ꢀ
SAꢀ
Mn/SAꢀ
—ꢀ
530.3ꢀ 531.3
530.2ꢀ 531.1
530.6ꢀ 531.6
530.4ꢀ 531.3
530.7ꢀ 531.7
530.7ꢀ 531.6
530.7ꢀ 531.6
0.80ꢀ
0.57ꢀ
0.43ꢀ
0.50ꢀ
0.39ꢀ
0.47ꢀ
0.52ꢀ
641.7ꢀ 644.5ꢀ
642.1ꢀ 643.6ꢀ
641.7ꢀ 643.8ꢀ
642.3ꢀ 645.2ꢀ
642.2ꢀ 644.3ꢀ
642.4ꢀ 644.8ꢀ
0.33ꢀ
0.55ꢀ
0.46ꢀ
0.26ꢀ
0.42ꢀ
0.35ꢀ
Ce‐Mn/SAꢀ
Fe‐Mn/SAꢀ
Cu‐Mn/SAꢀ
Cr‐Mn/SAꢀ
Co‐Mn/SAꢀ
2+
reductionꢀ ofꢀ Cu ꢀ [41,42].ꢀ Theꢀ reductionꢀ peakꢀ atꢀ 281ꢀ °Cꢀ forꢀ
Cr‐Mn/SAꢀmostꢀlikelyꢀresultedꢀbecauseꢀofꢀtheꢀreductionꢀofꢀCrO
toꢀCr ꢀ[43].ꢀHowever,ꢀthisꢀresultꢀdoesꢀnotꢀimplyꢀanꢀabsenceꢀofꢀ
reductionꢀ forꢀ MnO ꢀ species.ꢀ Theꢀ betterꢀ redoxꢀ abilitiesꢀ ofꢀ
6
ꢀ
O
2 3
x
Cu‐Mn/SAꢀ andꢀ Cr‐Mn/SAꢀ contributedꢀ toꢀ theꢀ betterꢀ catalyticꢀ
activity.ꢀ
surfaceꢀareaꢀafterꢀdopingꢀwithꢀCeꢀmayꢀresultꢀbecauseꢀofꢀtheꢀH1ꢀ
hysteresisꢀloopꢀofꢀtheꢀmonometallicꢀCeꢀsampleꢀasꢀshownꢀinꢀFig.ꢀ
3.2.5.ꢀ ꢀ Surfaceꢀpropertiesꢀ
3.ꢀAlthoughꢀtheꢀquantityꢀofꢀadsorbedꢀmonometallicꢀCeꢀsampleꢀ
Theꢀ Mnꢀ 2pꢀ spectraꢀ inꢀ Fig.ꢀ 10ꢀ showꢀ thatꢀ theꢀ asymmetricalꢀ
wasꢀlowerꢀthanꢀforꢀtheꢀotherꢀsamples,ꢀtheꢀhysteresisꢀloopꢀwasꢀ
broaderꢀthanꢀforꢀtheꢀotherꢀsamples.ꢀThisꢀmayꢀfavorꢀtheꢀinterac‐
tionꢀ betweenꢀ Ceꢀ andꢀ theꢀ supports.ꢀ Theꢀ BETꢀ surfaceꢀ areaꢀ forꢀ
theseꢀ fiveꢀ samplesꢀ decreasedꢀ asꢀ Ce‐Mn/SAꢀ >ꢀ Cr‐Mn/SAꢀ >ꢀ
Cu‐Mn/SAꢀ>ꢀCo‐Mn/SAꢀ>ꢀFe‐Mn/SA.ꢀSurfaceꢀareaꢀisꢀnotꢀtheꢀonlyꢀ
determiningꢀ factorꢀ forꢀ catalyticꢀ activity,ꢀ whichꢀ resultedꢀ inꢀ aꢀ
differentꢀorderꢀofꢀcatalyticꢀactivity.ꢀ
signalsꢀ ofꢀ allꢀ samplesꢀ includedꢀ aꢀ spinꢀ orbitꢀ doubletꢀ withꢀ Mnꢀ
3/2ꢀandꢀMnꢀ2p1/2ꢀatꢀBEꢀ=ꢀ642ꢀandꢀ653ꢀeV,ꢀrespectively.ꢀTheꢀ
Mnꢀ2p3/2ꢀregionꢀcouldꢀbeꢀresolvedꢀintoꢀtwoꢀmainꢀpeaksꢀthatꢀareꢀ
2p
3+
4+
ascribableꢀtoꢀtheꢀsurfaceꢀofꢀMn ꢀandꢀMn ꢀspeciesꢀ[44,45]ꢀusingꢀ
anꢀ optimumꢀ combinationꢀ ofꢀ theꢀ Gaussianꢀ peaksꢀ method.ꢀ Theꢀ
specificꢀbindingꢀenergyꢀofꢀtheseꢀMnꢀspeciesꢀandꢀtheꢀmolarꢀratioꢀ
4+
ofꢀ Mn ꢀ asꢀ calculatedꢀ byꢀ theꢀ quantitative‐areaꢀ integrationꢀ
methodꢀ areꢀ listedꢀ inꢀ Tableꢀ 3.ꢀ Ce‐Mn/SAꢀ andꢀ Fe‐Mn/SAꢀ pos‐
H
2
‐TPRꢀmeasurementsꢀwereꢀcarriedꢀoutꢀtoꢀcompareꢀtheꢀre‐
doxꢀabilitiesꢀofꢀtheseꢀcatalysts,ꢀandꢀtheꢀresultsꢀareꢀshownꢀinꢀFig.ꢀ
.ꢀBecauseꢀtheꢀpureꢀSAꢀsupportꢀ[14]ꢀshowedꢀalmostꢀnoꢀredoxꢀ
abilityꢀinꢀtheꢀtemperatureꢀrange,ꢀallꢀtheꢀreductionꢀpeaksꢀinꢀFig.ꢀ
4+
sessedꢀtheꢀhighestꢀmolarꢀratioꢀofꢀMn ꢀamongꢀtheseꢀsamples.ꢀItꢀ
hasꢀbeenꢀreportedꢀthatꢀMnꢀspeciesꢀwithꢀaꢀhigherꢀvalenceꢀcouldꢀ
enhanceꢀtheꢀoxidizingꢀpowerꢀandꢀbenefitꢀtheꢀNOꢀoxidationꢀre‐
9
4+
actionꢀ[46,47].ꢀTheꢀhigherꢀdistributionꢀratioꢀofꢀMn ꢀcanꢀinduceꢀ
4+
aꢀmoreꢀactiveꢀoxygenꢀbondingꢀwithꢀMn ꢀtoꢀstabilizeꢀtheꢀchem‐
220
4+
icalꢀstatesꢀandꢀstructureꢀ[48].ꢀTherefore,ꢀtheꢀhigherꢀMn ꢀratioꢀ
ofꢀtheseꢀtwoꢀcatalystsꢀresultedꢀinꢀaꢀbetterꢀcatalyticꢀactivityꢀforꢀ
NOꢀ deepꢀ oxidation.ꢀ However,ꢀ itꢀ hasꢀ beenꢀ reportedꢀ thatꢀ Mn3+ꢀ
speciesꢀ couldꢀ enhanceꢀ theꢀ formationꢀ ofꢀ oxygenꢀ vacanciesꢀ
281
Cu-Mn/SA
Cr-Mn/SA
[
27,29].ꢀTheꢀdensityꢀofꢀtheꢀoxygenꢀvacanciesꢀaffectsꢀtheꢀcatalyt‐
icꢀactivityꢀofꢀtheꢀozoneꢀdecompositionꢀsignificantlyꢀ[49].ꢀMn4+ꢀ
favorsꢀ NOꢀ oxidation,ꢀ whereasꢀ Mn ꢀ favorsꢀ ozoneꢀ decomposi‐
tion.ꢀTherefore,ꢀtheꢀrelativeꢀbalanceꢀdistributionꢀbetweenꢀMn ꢀ
andꢀMn ꢀisꢀbetterꢀforꢀtheꢀcatalyticꢀactivityꢀofꢀNOꢀdeepꢀoxidationꢀ
3+
Co-Mn/SA
Fe-Mn/SA
4+
3+
byꢀozone.ꢀ
Ce-Mn/SA
Mn/SA
TheꢀOꢀ1sꢀspectraꢀofꢀtheseꢀfiveꢀsamplesꢀcanꢀbeꢀdeconvolutedꢀ
intoꢀ twoꢀ peaksꢀ thatꢀ areꢀ ascribableꢀ toꢀ latticeꢀ oxygenꢀ O
α
ꢀ andꢀ
chemisorbedꢀ oxygenꢀ O ꢀ [49,50].ꢀ Theꢀ specificꢀ bindingꢀ energyꢀ
β
100
200
Fig.ꢀ9.ꢀH
300
400
500
600
700
800
andꢀtheꢀmolarꢀratiosꢀareꢀlistedꢀinꢀTableꢀ3.ꢀCo‐Mn/SAꢀexhibitedꢀ
theꢀ mostꢀ chemisorbedꢀ oxygenꢀ onꢀ theꢀ catalystꢀ surfaceꢀ amongꢀ
o
Temperature ( C)
3+
2
‐TPRꢀprofilesꢀforꢀMn‐basedꢀcatalysts.ꢀ
theseꢀ samples.ꢀ Theꢀ correspondingꢀ Mn ꢀ ratioꢀ wasꢀ high.ꢀ Thisꢀ

ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
1277ꢀ
Ce3+
Ce-Mn/SA
Ce-Mn/SA
Ce-Mn/SA
3+
Mn
Mn4+
O
α
O
β
4
+
Ce
660
655
650
645
640
635
535
534
533
532
531
530
529
528
527
930
920
910
900
890
880
870
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Fe3+
Fe-Mn/SA
Fe-Mn/SA
Fe-Mn/SA
Mn3+
Fe2+
O
α
Mn4+
O
β
660
655
650
645
640
635
535
534
533
532
531
530
529
528
527
740
730
720
710
700
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Cu-Mn/SA
Satellite
Cu-Mn/SA
Cu-Mn/SA
Mn3+
O
α
Mn4+
Cu 2p1/2
Cu 2p3/2
O
β
660
655
650
645
640
635
535
534
533
532
531
530
529
528
527
965
960
955
950
945
940
935
930
925
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Mn3+
Cr-Mn/SA
Cr-Mn/SA
Cr 2p3/2
Cr-Mn/SA
O
α
Cr6+
Mn4+
Cr3+
O
β
Cr 2p1/2
6
60
655
650
645
640
635
535
535
535
534
533
532
531
530
529
528
527
592
588
584
580
576
572
568
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Mn3+
Co-Mn/SA
O
Co-Mn/SA
Co 2p3/2
Co-Mn/SA
O
α
4
+
Co3+
Co2+
Mn
Co 2p1/2
β
Co3+
Satellite
Satellite
Co2+
660
655
650
645
640
635
534
533
532
531
530
529
528
527
812
808
804
800
796
792
788
784
780
776
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Mn3+
Mn/SA
SA
Mn/SA
O
β
Mn4+
O
α
O
α
O
β
6
60
655
650
645
640
635
534
533
532
531
530
529
528
527
535
534
533
532
531
530
529
528
527
Binding energy (eV)
Binding energy (eV)
Binding energy (eV)
Fig.ꢀ10.ꢀMnꢀ2p,ꢀOꢀ1s,ꢀandꢀdopedꢀmetalꢀatomicꢀXPSꢀspectraꢀofꢀMn‐basedꢀcatalysts.ꢀ
4
+
3+
probablyꢀresultedꢀinꢀtheꢀlowestꢀconcentrationꢀofꢀresidualꢀozoneꢀ
inꢀFig.ꢀ5.ꢀBut,ꢀasꢀtheꢀcatalytic‐activityꢀtestꢀresultsꢀshowꢀinꢀFig.ꢀ4,ꢀ
Co‐Mn/SAꢀdisplayedꢀtheꢀworstꢀactivity.ꢀItꢀcanꢀbeꢀconcludedꢀthatꢀ
theꢀ excellentꢀ ozoneꢀ decompositionꢀ activityꢀ ofꢀ Co‐Mn/SAꢀ re‐
sultedꢀinꢀinsufficientꢀozoneꢀforꢀNOꢀdeepꢀoxidation.ꢀTheꢀhigherꢀ
ratioꢀofꢀchemisorbedꢀoxygenꢀthatꢀisꢀassociatedꢀwithꢀtheꢀhigherꢀ
Mn ꢀ contributedꢀ toꢀ aꢀ betterꢀ NOꢀ oxidationꢀ activityꢀ [48].ꢀ
Ce‐Mn/SAꢀ andꢀ Fe‐Mn/SAꢀ possessedꢀ theseꢀ twoꢀ factorsꢀ andꢀ
therefore,ꢀtheyꢀdisplayedꢀexcellentꢀactivity.ꢀ
Ce ꢀandꢀCe ꢀofꢀceriumꢀoxidesꢀresultsꢀinꢀlargeꢀamountsꢀofꢀlabileꢀ
oxygenꢀvacanciesꢀandꢀoxygenꢀspeciesꢀ[52,53].ꢀTheꢀhigherꢀratioꢀ
3+
ofꢀCe ꢀcreatesꢀmoreꢀchemisorbedꢀoxygenꢀspeciesꢀonꢀtheꢀcata‐
lystꢀ surfaceꢀ becauseꢀ ofꢀ theꢀ chargeꢀ imbalance,ꢀ labileꢀ oxygenꢀ
vacancies,ꢀandꢀunsaturatedꢀchemicalꢀbondsꢀ[54,55].ꢀTherefore,ꢀ
abundantꢀCe ꢀspeciesꢀexistꢀonꢀtheꢀcatalystꢀsurface,ꢀwhichꢀareꢀ
criticalꢀforꢀtheꢀexcellentꢀcatalyticꢀactivity.ꢀ
TheꢀFeꢀ2pꢀspectraꢀwasꢀrelativelyꢀsimpleꢀcomparedꢀwithꢀtheꢀ
Ceꢀ 3dꢀspectra.ꢀOnlyꢀtwoꢀmainꢀpeaksꢀareꢀvisibleꢀfromꢀtheꢀXPSꢀ
signal,ꢀwhichꢀareꢀascribableꢀtoꢀFe ꢀatꢀaꢀlowerꢀbindingꢀenergyꢀ
andꢀ Fe ꢀ atꢀ aꢀ higherꢀ bindingꢀ energyꢀ [56].ꢀ Theꢀ correspondingꢀ
integratedꢀ areaꢀratioꢀofꢀFe /Fetotalꢀwasꢀ0.51.ꢀThisꢀresultꢀindi‐
catesꢀ thatꢀ Fe ꢀandꢀFe ꢀ hadꢀ aꢀ nearlyꢀ averageꢀ distributionꢀ onꢀ
theꢀcatalystꢀsurface.ꢀ
3
+
4+
3+
TheꢀCeꢀ3d,ꢀFeꢀ2p,ꢀCuꢀ2p,ꢀCrꢀ2p,ꢀandꢀCoꢀ2pꢀXPSꢀspectraꢀareꢀ
2+
givenꢀ andꢀ theꢀ chemicalꢀ statesꢀ areꢀ labelledꢀ inꢀ Fig.ꢀ 10.ꢀ Twoꢀ
4+
3+
chemicalꢀstatesꢀofꢀCeꢀspeciesꢀexistꢀonꢀtheꢀcatalystꢀsurface:ꢀCe ꢀ
3+
4+
3+
2+
andꢀ Ce ,ꢀ andꢀ theꢀ integratedꢀ areaꢀ ratioꢀ ofꢀ Ce /Ceꢀ wasꢀ 0.47,ꢀ
whichꢀ couldꢀ beꢀ attributedꢀ toꢀ aꢀ strongꢀ interactionꢀ betweenꢀ
manganeseꢀ andꢀ ceriumꢀ oxidesꢀ [51].ꢀ Theꢀ transitionꢀ betweenꢀ
TheꢀCuꢀ2pꢀspectraꢀincludedꢀtwoꢀmainꢀregionsꢀofꢀCuꢀ2p1/2ꢀatꢀ

1278ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
3
+
2+
BEꢀ=ꢀ947‒960ꢀeVꢀandꢀCuꢀ2p3/2ꢀatꢀBEꢀ=ꢀ930‒940ꢀeV,ꢀalongꢀwithꢀ
twoꢀsatelliteꢀpeaksꢀasꢀshownꢀinꢀFig.ꢀ10.ꢀTheꢀCuꢀ2p1/2ꢀsignalꢀcanꢀ
beꢀ deconvolutedꢀ intoꢀ twoꢀ peaksꢀ thatꢀ areꢀ ascribableꢀ toꢀ Cu+ꢀ
wasꢀ0.47.ꢀThisꢀresultꢀindicatesꢀthatꢀCo ꢀandꢀCo ꢀexistedꢀwithꢀaꢀ
nearlyꢀaverageꢀdistributionꢀonꢀtheꢀcatalystꢀsurface.ꢀ
2+
(932.2ꢀeV)ꢀ(stabilizedꢀbyꢀCu‐Mnꢀinteractions)ꢀandꢀCu ꢀ(934.6ꢀ
4.ꢀ ꢀ Conclusionsꢀ
eV)ꢀ[57‒59],ꢀrespectively.ꢀTheꢀinteractionꢀbetweenꢀCuꢀandꢀMnꢀ
2
+
3+
+
4+
species,ꢀ Cu ꢀ +ꢀ Mn ꢀ →ꢀ Cu ꢀ+ꢀMn ,ꢀfavorsꢀtheꢀcatalyticꢀreac‐
tionꢀ[60].ꢀElectronicꢀtransferꢀbetweenꢀtheseꢀtwoꢀspeciesꢀisꢀas‐
sociatedꢀ withꢀ oxygen‐speciesꢀ adsorptionꢀ [39].ꢀ However,ꢀ ac‐
Aꢀseriesꢀofꢀmonometallicꢀcatalystsꢀ(Mn,ꢀCo,ꢀCe,ꢀFe,ꢀCu,ꢀandꢀ
Cr)ꢀwasꢀpreparedꢀbyꢀtheꢀsol‐gelꢀmethodꢀtoꢀeffectꢀcatalyticꢀNOꢀ
deepꢀoxidationꢀbyꢀozone.ꢀTheꢀmanganeseꢀoxidesꢀdisplayedꢀtheꢀ
3+
2+
cordingꢀ toꢀ theꢀ XPSꢀ resultsꢀ ofꢀ Cu‐Mn/SA,ꢀ Mn ꢀ andꢀ Cu ꢀ areꢀ
dominantꢀ speciesꢀ ofꢀ Mnꢀ andꢀ Cu,ꢀ respectively.ꢀ Thisꢀ indicatesꢀ
thatꢀ theꢀ interactionꢀ betweenꢀ Cuꢀ andꢀ Mnꢀ inꢀ Cu‐Mn/SAꢀ wasꢀ
weak,ꢀandꢀresultedꢀinꢀlessꢀchemisorbedꢀoxygenꢀonꢀtheꢀcatalystꢀ
surface,ꢀwhichꢀmayꢀbeꢀaꢀreasonꢀforꢀtheꢀhighestꢀozoneꢀresidualꢀ
inꢀFig.ꢀ5ꢀforꢀCu‐Mn/SA.ꢀ
TheꢀtwoꢀmainꢀpeaksꢀinꢀtheꢀCrꢀ2pꢀspectraꢀrepresentꢀCrꢀ2p3/2
atꢀBEꢀ=ꢀ579.4ꢀeVꢀandꢀCrꢀ2p1/2ꢀatꢀBEꢀ=ꢀ587.5ꢀeV,ꢀwhichꢀcanꢀbeꢀ
assignedꢀ toꢀ CrO
areaꢀratioꢀofꢀCr /Crꢀwasꢀ0.72.ꢀTheꢀenrichmentꢀinꢀCr ꢀonꢀtheꢀ
catalystꢀsurfaceꢀcontributedꢀtoꢀitsꢀhigherꢀcatalyticꢀactivityꢀ[62].ꢀ
Becauseꢀofꢀtheꢀspin‐orbit‐splittingꢀenergy,ꢀtheꢀCoꢀ2pꢀspectraꢀ
exhibitedꢀtwoꢀmainꢀpeaksꢀthatꢀcorrespondꢀtoꢀCoꢀ2p3/2ꢀatꢀBEꢀ=ꢀ
81.3ꢀeVꢀandꢀCoꢀ2p1/2ꢀatꢀBEꢀ=ꢀ797.3ꢀeV,ꢀrespectively,ꢀalongꢀwithꢀ
twoꢀsatelliteꢀpeaks.ꢀAfterꢀpeakꢀfitting,ꢀtheꢀCoꢀ2p3/2ꢀandꢀCoꢀ2p1/2
regionsꢀwereꢀdecomposedꢀintoꢀtwoꢀkindsꢀofꢀpeaksꢀthatꢀareꢀas‐
highestꢀconversionꢀefficiencyꢀ(almostꢀ80%ꢀatꢀanꢀO /NOꢀmolarꢀ
3
ratioꢀ ofꢀ 2.0)ꢀ amongꢀ theꢀ samples.ꢀ Onlyꢀ oneꢀ typeꢀ ofꢀ crystallineꢀ
phaseꢀwasꢀdetectedꢀbyꢀXRDꢀmeasurementsꢀforꢀallꢀsamplesꢀandꢀ
theseꢀwereꢀCeO
2
,ꢀMn
2
O
3
,ꢀFe
2
O
3
,ꢀCuO,ꢀ CoO,ꢀ andꢀ Cr
2
O .ꢀTheꢀpo‐
3
rousꢀ structureꢀ parametersꢀ resultsꢀ showedꢀ thatꢀ theꢀ catalyticꢀ
activityꢀ isꢀ relatedꢀ toꢀ surfaceꢀ areaꢀ andꢀ poreꢀ parameters,ꢀ butꢀ
theseꢀareꢀnotꢀtheꢀonlyꢀdeterminingꢀfactors.ꢀAꢀsignificantꢀdiffer‐
enceꢀresultedꢀforꢀtheseꢀmonometallicꢀsamples:ꢀCrꢀpossessedꢀtheꢀ
ꢀ
2
3
ꢀ andꢀ Cr
2
O
3
,ꢀ respectivelyꢀ [61].ꢀ Theꢀ integratedꢀ
highestꢀsurfaceꢀareaꢀ(38.1ꢀm /g)ꢀandꢀCuꢀexhibitedꢀtheꢀlowestꢀ
6+
6+
2
surfaceꢀareaꢀ(0.5ꢀm /g).ꢀ
Theꢀ manganese‐oxide‐basedꢀ bimetallicꢀ catalystsꢀ thatꢀ wereꢀ
supportedꢀonꢀSAꢀwereꢀstudiedꢀbyꢀusingꢀtheꢀsameꢀprocedure.ꢀCe,ꢀ
Fe,ꢀ Cr,ꢀ Cu,ꢀ andꢀ Coꢀ wereꢀ selectedꢀ asꢀ promoters,ꢀ respectively.ꢀ
Ce‐Mn/SAꢀandꢀFe‐Mn/SAꢀshowedꢀtheꢀbestꢀperformanceꢀandꢀanꢀ
7
ꢀ
excellentꢀstabilityꢀatꢀaꢀmolarꢀratioꢀofꢀO
3
/NOꢀ=ꢀ1.5,ꢀwithꢀlessꢀthanꢀ
3
3
2
50ꢀmg/m ꢀofꢀNOꢀ+ꢀNO ,ꢀandꢀlessꢀthanꢀ25ꢀmg/m ꢀofꢀozoneꢀre‐
3
+
2+
cribedꢀtoꢀCo ꢀ(782.3ꢀandꢀ797.8ꢀeV)ꢀandꢀCo ꢀ(780.8ꢀandꢀ796.0ꢀ
eV)ꢀ[27,56,63].ꢀTheꢀintegratedꢀareaꢀratioꢀofꢀCo /(Co ꢀ+ꢀCo )ꢀ
sidual,ꢀwhichꢀhadꢀreachedꢀtheꢀultra‐lowꢀemissionꢀstandard.ꢀTheꢀ
otherꢀthreeꢀmetalꢀoxidesꢀinhibitedꢀtheꢀcatalyticꢀactivity.ꢀXRD,ꢀN ꢀ
3+
2+
3+
2
GraphicalꢀAbstractꢀ
Chin.ꢀJ.ꢀCatal.,ꢀ2017,ꢀ38:ꢀ1270–1280ꢀ ꢀ doi:ꢀ10.1016/S1872‐2067(17)62860‐2
Promotionalꢀeffectꢀofꢀsphericalꢀaluminaꢀloadingꢀwithꢀmanganese‐basedꢀbimetallicꢀoxidesꢀonꢀnitric‐oxideꢀdeepꢀoxidationꢀbyꢀ
ozoneꢀ
FaweiꢀLin,ꢀZhihuaꢀWangꢀ*,ꢀJiamingꢀShao,ꢀDingkunꢀYuan,ꢀYongꢀHe,ꢀYanqunꢀZhu,ꢀKefaꢀCenꢀ
ZhejiangꢀUniversityꢀ
4
00
Ce-Mn/SA
Fe-Mn/SA
Cu-Mn/SA
Cr-Mn/SA
Co-Mn/SA
O3
350
300
O2
2
2
50
00
NO
NO
O3
N O
O2
O3
2
5
150
100
O3
Longer reaction time: 3−5 s
50
O2
0
0
20
40
60
80
100
120
O3
O3
O3
Time (min)
6
5
0
0
Ce-Mn/SA
Fe-Mn/SA
Cu-Mn/SA
Cr-Mn/SA
Co-Mn/SA
O2
O2
O2
Bimetallic Mn based catalysts
supported on spherical alumina (SA)
40
NO
NO
3
2
1
0
0
0
0
N O
2
5
Shorter reaction time: ~0.12 s
O3
0
20
40
60
80
100
120
Time (min)
ꢀ
ThisꢀpaperꢀfocusedꢀonꢀcatalyticꢀNOꢀdeepꢀoxidationꢀtoꢀN
2
O
3
5
.ꢀTheꢀMnꢀoxidesꢀexhibitedꢀtheꢀbestꢀactivity.ꢀTheꢀlowestꢀleftꢀconcentrationꢀofꢀ
/NOꢀmolarꢀratioꢀofꢀ1.5.ꢀ
3
2
NO+NO ꢀwasꢀlessꢀthanꢀ20ꢀmg/m ꢀbyꢀFe‐Mn/SAꢀunderꢀtheꢀO

ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
1279ꢀ
adsorption,ꢀ H
2
‐TPR,ꢀ andꢀ XPSꢀ wereꢀ conductedꢀ toꢀ studyꢀ theꢀ
341–350.ꢀ
24] H.ꢀC.ꢀWang,ꢀH.ꢀS.ꢀLiang,ꢀM.ꢀB.ꢀChang,ꢀJ.ꢀHazard.ꢀMater.,ꢀ2011,ꢀ186,ꢀ
781–1787.ꢀ
25] M.ꢀS.ꢀJin,ꢀJ.ꢀH.ꢀKim,ꢀJ.ꢀM.ꢀKim,ꢀJ.ꢀK.ꢀJeon,ꢀJ.ꢀJurng,ꢀG.ꢀN.ꢀBae,ꢀY.ꢀK.ꢀPark,ꢀ
Catal.ꢀToday,ꢀ2013,ꢀ204,ꢀ108–113.ꢀ
[
[
[
crystallineꢀ structure,ꢀ physicochemical,ꢀ andꢀ surfaceꢀ propertiesꢀ
ofꢀ theseꢀ catalysts.ꢀ Aꢀ higherꢀ surfaceꢀ areaꢀ andꢀ betterꢀ poreꢀ pa‐
rametersꢀfavoredꢀreactantꢀadsorption.ꢀMoreꢀoxygenꢀvacanciesꢀ
benefitedꢀ ozoneꢀ adsorptionꢀ andꢀ decomposition.ꢀ Theꢀ relativeꢀ
balanceꢀdistributionꢀbetweenꢀMn ꢀandꢀMn ꢀwasꢀbetterꢀforꢀNOꢀ
oxidationꢀ andꢀ ozoneꢀ decomposition,ꢀ andꢀ thisꢀ favoredꢀ N
formation.ꢀTheꢀsyntheticꢀactionꢀofꢀsurfaceꢀarea,ꢀporeꢀparame‐
1
26] A.ꢀBueno‐López,ꢀD.ꢀLozano‐Castelló,ꢀJ.ꢀA.ꢀAnderson,ꢀAppl.ꢀCatal.ꢀB,ꢀ
4+
3+
2016,ꢀ198,ꢀ189–199.ꢀ
2 5
O ꢀ
[
27] J.ꢀZ.ꢀMa,ꢀC.ꢀX.ꢀWang,ꢀH.ꢀHe,ꢀAppl.ꢀCatal.ꢀB,ꢀ2017,ꢀ201,ꢀ503–510.ꢀ
28] J.ꢀH.ꢀKo,ꢀS.ꢀH.ꢀPark,ꢀJ.ꢀK.ꢀJeon,ꢀS.ꢀS.ꢀKim,ꢀS.ꢀC.ꢀKim,ꢀJ.ꢀM.ꢀKim,ꢀD.ꢀChang,ꢀ
Y.ꢀK.ꢀPark,ꢀCatal.ꢀToday,ꢀ2012,ꢀ185,ꢀ290–295.ꢀ
[
ters,ꢀatomicꢀ chemicalꢀ states,ꢀ andꢀoxygenꢀvacanciesꢀaffectsꢀtheꢀ
catalyticꢀactivity.ꢀ
[29] C.ꢀX.ꢀWang,ꢀJ.ꢀZ.ꢀMa,ꢀF.ꢀD.ꢀLiu,ꢀH.ꢀHe,ꢀR.ꢀD.ꢀZhang,ꢀJ.ꢀPhys.ꢀChem.ꢀC,ꢀ
015,ꢀ119,ꢀ23119–23126.ꢀ
2
[
[
[
[
30] Z.ꢀS.ꢀZhang,ꢀB.ꢀB.ꢀChen,ꢀX.ꢀK.ꢀWang,ꢀL.ꢀXu,ꢀC.ꢀAu,ꢀC.ꢀShi,ꢀM.ꢀCrocker,ꢀ
Referencesꢀ
Appl.ꢀCatal.ꢀB,ꢀ2015,ꢀ165,ꢀ232–244.ꢀ
31] F.ꢀW.ꢀLin,ꢀY.ꢀHe,ꢀZ.ꢀH.ꢀWang,ꢀQ.ꢀMa,ꢀR.ꢀWhiddon,ꢀY.ꢀQ.ꢀZhu,ꢀJ.ꢀZ.ꢀLiu,ꢀ
RSCꢀAdv.,ꢀ2016,ꢀ6,ꢀ31422–31430.ꢀ
32] S.ꢀJ.ꢀYang,ꢀF.ꢀH.ꢀQi,ꢀS.ꢀC.ꢀXiong,ꢀH.ꢀDang,ꢀY.ꢀLiao,ꢀP.ꢀK.ꢀWong,ꢀJ.ꢀH.ꢀLi,ꢀ
Appl.ꢀCatal.ꢀB,ꢀ2016,ꢀ181,ꢀ570–580.ꢀ
33] Y.ꢀXꢀWang,ꢀH.ꢀQ.ꢀSun,ꢀH.ꢀM.ꢀAng,ꢀM.ꢀO.ꢀTadé,ꢀS.ꢀB.ꢀWang,ꢀAppl.ꢀCatal.ꢀ
B,ꢀ2015,ꢀ164,ꢀ159–167.ꢀ
[
1] R.ꢀJ.ꢀHuang,ꢀY.ꢀL.ꢀZhang,ꢀC.ꢀBozzetti,ꢀK.ꢀF.ꢀHo,ꢀJ.ꢀJ.ꢀCao,ꢀY.ꢀM.ꢀHan,ꢀK.ꢀR.ꢀ
Daellenbach,ꢀJ.ꢀG.ꢀSlowik,ꢀS.ꢀM.ꢀPlatt,ꢀF.ꢀCanonaco,ꢀP.ꢀZotter,ꢀR.ꢀWolf,ꢀ
S.ꢀM.ꢀPieber,ꢀE.ꢀA.ꢀBruns,ꢀM.ꢀCrippa,ꢀG.ꢀCiarelli,ꢀA.ꢀPiazzalunga,ꢀM.ꢀ
Schwikowski,ꢀG.ꢀAbbaszade,ꢀJ.ꢀSchnelle‐Kreis,ꢀR.ꢀZimmermann,ꢀZ.ꢀ
S.ꢀAn,ꢀS.ꢀSzidat,ꢀU.ꢀBaltensperger,ꢀI.ꢀE.ꢀHaddad,ꢀA.ꢀS.ꢀH.ꢀPrévôt,ꢀNa‐
ture,ꢀ2014,ꢀ514,ꢀ218–222.ꢀ
[
34] B.ꢀY.ꢀBai,ꢀJ.ꢀH.ꢀLi,ꢀJ.ꢀM.ꢀHao,ꢀAppl.ꢀCatal.ꢀB,ꢀ2015,ꢀ164,ꢀ241–250.ꢀ
35] F.ꢀS.ꢀDing,ꢀA.ꢀF.ꢀZhang,ꢀM.ꢀLiu,ꢀY.ꢀZuo,ꢀK.ꢀY.ꢀLi,ꢀX.ꢀW.ꢀGuo,ꢀC.ꢀS.ꢀSong,ꢀ
Ind.ꢀEng.ꢀChem.ꢀRes.,ꢀ2014,ꢀ53,ꢀ17563–17569.ꢀ
[
[
[
[
[
[
[
[
2] Q.ꢀZhang,ꢀY.ꢀHuang,ꢀS.ꢀQ.ꢀPeng,ꢀY.ꢀF.ꢀZhang,ꢀZ.ꢀX.ꢀShenc,ꢀJ.ꢀJ.ꢀCao,ꢀW.ꢀ
K.ꢀHo,ꢀS.ꢀC.ꢀLee,ꢀD.ꢀY.ꢀH.ꢀPui,ꢀAppl.ꢀCatal.ꢀB,ꢀ2017,ꢀ204,ꢀ346–357.ꢀ
3] Z.ꢀH.ꢀWang,ꢀJ.ꢀH.ꢀZhou,ꢀY.ꢀQ.ꢀZhu,ꢀZ.ꢀC.ꢀWen,ꢀJ.ꢀZ.ꢀLiu,ꢀK.ꢀF.ꢀCen,ꢀFuelꢀ
Process.ꢀTechnol.,ꢀ2007,ꢀ88,ꢀ817–823.ꢀ
[
[
[
[
[
[
[
[
[
[
36] C.ꢀ Heisig,ꢀ W.ꢀ H.ꢀ Zhang,ꢀ S.ꢀ T.ꢀ Oyama,ꢀ Appl.ꢀ Catal.ꢀ B,ꢀ 1997,ꢀ 14,ꢀ
117–129.ꢀ
4] D.ꢀMráček,ꢀP.ꢀKočí,ꢀJ.ꢀS.ꢀChoi,ꢀW.ꢀP.ꢀPartridge,ꢀAppl.ꢀCatal.ꢀB,ꢀ2016,ꢀ
37] F.ꢀ Gunnarsson,ꢀ J.ꢀ A.ꢀ Pihl,ꢀ T.ꢀ J.ꢀ Toops,ꢀ M.ꢀ Skoglundh,ꢀ H.ꢀ Härelind,ꢀ
Appl.ꢀCatal.ꢀB,ꢀ2017,ꢀ202,ꢀ42–50.ꢀ
38] F.ꢀW.ꢀLin,ꢀZ.ꢀH.ꢀWang,ꢀJ.ꢀM.ꢀShao,ꢀD.ꢀK.ꢀYuan,ꢀY.ꢀHe,ꢀY.ꢀQ.ꢀZhu,ꢀK.ꢀF.ꢀ
Cen,ꢀRSCꢀAdv.,ꢀ2017,ꢀ7,ꢀ25132–25143.ꢀ
39] W.ꢀX.ꢀTang,ꢀX.ꢀF.ꢀWu,ꢀS.ꢀD.ꢀLi,ꢀX.ꢀShan,ꢀG.ꢀLiu,ꢀY.ꢀF.ꢀChen,ꢀAppl.ꢀCatal.ꢀ
B,ꢀ2015,ꢀ162,ꢀ110–121.ꢀ
40] V.ꢀA.ꢀdeꢀlaꢀPeñaꢀO'Shea,ꢀM.ꢀC.ꢀAlvarez‐Galvan,ꢀJ.ꢀL.ꢀG.ꢀFierro,ꢀP.ꢀL.ꢀ
Arias,ꢀAppl.ꢀCatal.ꢀB,ꢀ2005,ꢀ57,ꢀ191–199.ꢀ
182,ꢀ109–114.ꢀ
5] D.ꢀQ.ꢀDao,ꢀL.ꢀGasnot,ꢀK.ꢀMarschallek,ꢀA.ꢀElꢀBakali,ꢀJ.ꢀF.ꢀPauwels,ꢀEn‐
ergyꢀFuels,ꢀ2010,ꢀ24,ꢀ1696–1703.ꢀ
6] Q.ꢀMa,ꢀZ.ꢀH.ꢀWang,ꢀF.ꢀW.ꢀLin,ꢀM.ꢀKuang,ꢀR.ꢀWhiddon,ꢀY.ꢀHe,ꢀJ.ꢀZ.ꢀLiu,ꢀ
EnergyꢀFuels,ꢀ2016,ꢀ30,ꢀ2302–2308.ꢀ
7] W.ꢀL.ꢀChang,ꢀP.ꢀV.ꢀBhave,ꢀS.ꢀS.ꢀBrown,ꢀN.ꢀRiemer,ꢀJ.ꢀStutz,ꢀD.ꢀDabdub,ꢀ
AerosolꢀSci.ꢀTechnol.,ꢀ2011,ꢀ45,ꢀ665–685.ꢀ
8] K.ꢀSkalska,ꢀJ.ꢀS.ꢀMiller,ꢀS.ꢀLedakowicz,ꢀSci.ꢀTotalꢀEnviron.,ꢀ2010,ꢀ408,ꢀ
41] D.ꢀA.ꢀAguilera,ꢀA.ꢀPerez,ꢀR.ꢀMolina,ꢀS.ꢀMoreno,ꢀAppl.ꢀCatal.ꢀB,ꢀ2011,ꢀ
3976–3989.ꢀ
1
04,ꢀ144–150.ꢀ
42] M.ꢀR.ꢀMorales,ꢀB.ꢀP.ꢀBarbero,ꢀL.ꢀE.ꢀCadús,ꢀAppl.ꢀCatal.ꢀB,ꢀ2006,ꢀ67,ꢀ
29–236.ꢀ
43] R.ꢀH.ꢀMa,ꢀP.ꢀJ.ꢀHu,ꢀL.ꢀY.ꢀJin,ꢀY.ꢀJ.ꢀWang,ꢀJ.ꢀQ.ꢀLu,ꢀM.ꢀF.ꢀLuo,ꢀCatal.ꢀToday,ꢀ
011,ꢀ175,ꢀ598–602.ꢀ
44] L.ꢀ L.ꢀ Zhang,ꢀ J.ꢀ J.ꢀ Tu,ꢀ L.ꢀ Lyu,ꢀ C.ꢀ Hu,ꢀ Appl.ꢀ Catal.ꢀ B,ꢀ 2016,ꢀ 181,ꢀ
61–569.ꢀ
9] I.ꢀJogi,ꢀE.ꢀStamate,ꢀC.ꢀIrimiea,ꢀM.ꢀSchmidt,ꢀR.ꢀBrandenburg,ꢀM.ꢀHo‐
lub,ꢀM.ꢀBonislawski,ꢀT.ꢀJakubowski,ꢀM.ꢀL.ꢀKääriäinen,ꢀD.ꢀC.ꢀCamer‐
on,ꢀFuel,ꢀ2015,ꢀ144,ꢀ137–144.ꢀ
2
[
[
[
[
[
[
[
[
10] S.ꢀL.ꢀDing,ꢀQ.ꢀYu,ꢀY.ꢀZ.ꢀZhang,ꢀ Y.ꢀ Liu,ꢀC.ꢀ X.ꢀ Xie,ꢀ G.ꢀYu,ꢀJ.ꢀAdv.ꢀOxid.ꢀ
Technol.,ꢀ2015,ꢀ18,ꢀ114–122.ꢀ
11] K.ꢀLi,ꢀX.ꢀL.ꢀTang,ꢀH.ꢀH.ꢀYi,ꢀP.ꢀNing,ꢀD.ꢀJ.ꢀKang,ꢀC.ꢀWang,ꢀChem.ꢀEng.ꢀJ.,ꢀ
2
5
2
012,ꢀ192,ꢀ99–104.ꢀ
12] K.ꢀ Skalska,ꢀ J.ꢀ S.ꢀ Miller,ꢀ S.ꢀ Ledakowicz,ꢀ Chem.ꢀ Eng.ꢀ Sci.,ꢀ 2011,ꢀ 66,ꢀ
386–3391.ꢀ
[
45] D.ꢀDelimaris,ꢀT.ꢀIoannides,ꢀAppl.ꢀCatal.ꢀB,ꢀ2008,ꢀ84,ꢀ303–312.ꢀ
46] D.ꢀM.ꢀMeng,ꢀW.ꢀC.ꢀZhan,ꢀY.ꢀGuo,ꢀY.ꢀL.ꢀGuo,ꢀL.ꢀWang,ꢀG.ꢀZ.ꢀLu,ꢀACSꢀ
Catal.,ꢀ2015,ꢀ5,ꢀ5973–5983.ꢀ
[
3
13] F.ꢀW.ꢀLin,ꢀZ.ꢀH.ꢀWang,ꢀQ.ꢀMa,ꢀY.ꢀHe,ꢀR.ꢀWhiddon,ꢀY.ꢀQ.ꢀZhu,ꢀJ.ꢀZ.ꢀLiu,ꢀ
EnergyꢀFues.,ꢀ2016,ꢀ30,ꢀ5101–5107.ꢀ
14] F.ꢀW.ꢀLin,ꢀZ.ꢀH.ꢀWang,ꢀQ.ꢀMa,ꢀY.ꢀYang,ꢀR.ꢀWhiddon,ꢀY.ꢀQ.ꢀZhu,ꢀK.ꢀFꢀ
Cen,ꢀAppl.ꢀCatal.ꢀB,ꢀ2016,ꢀ198,ꢀ100–111.ꢀ
15] D.ꢀY.ꢀYuan,ꢀZ.ꢀH.ꢀWang,ꢀC.ꢀDing,ꢀY.ꢀHe,ꢀR.ꢀWhiddon,ꢀK.ꢀF.ꢀCen,ꢀJ.ꢀPhys.ꢀ
D,ꢀ2016,ꢀ49,ꢀ455203/1–455203/12.ꢀ
16] U.ꢀ Roland,ꢀ F.ꢀ Holzer,ꢀ F.ꢀ D.ꢀ Kopinke,ꢀ Appl.ꢀ Catal.ꢀ B,ꢀ 2005,ꢀ 58,ꢀ
[47] M.ꢀQ.ꢀShen,ꢀZ.ꢀZhao,ꢀJ.ꢀH.ꢀChen,ꢀY.ꢀG.ꢀSu,ꢀJ.ꢀWang,ꢀX.ꢀQ.ꢀWang,ꢀJ.ꢀRareꢀ
Earths,ꢀ2013,ꢀ31,ꢀ119–123.ꢀ
[48] J.ꢀ H.ꢀ Chen,ꢀ M.ꢀ Q.ꢀ Shen,ꢀ X.ꢀ Q.ꢀ Wang,ꢀ G.ꢀ S.ꢀ Qi,ꢀ J.ꢀ Wang,ꢀ W.ꢀ Li,ꢀ Appl.ꢀ
Catal.ꢀB,ꢀ2013,ꢀ134–135,ꢀ251–257.ꢀ
[
49] J.ꢀB.ꢀJia,ꢀP.ꢀY.ꢀZhang,ꢀL.ꢀChen,ꢀAppl.ꢀCatal.ꢀB,ꢀ2016,ꢀ189,ꢀ210–218.ꢀ
50] V.ꢀP.ꢀSantos,ꢀM.ꢀF.ꢀR.ꢀPereira,ꢀJ.ꢀJ.ꢀM.ꢀÓrfão,ꢀJ.ꢀL.ꢀFigueiredo,ꢀAppl.ꢀ
Catal.ꢀB,ꢀ2010,ꢀ99,ꢀ353–363.ꢀ
[
2
17–226.ꢀ
17] A.ꢀ Ikhlaq,ꢀ B.ꢀ Kasprzyk‐Hordern,ꢀ Appl.ꢀ Catal.ꢀ B.,ꢀ 2017,ꢀ 200,ꢀ
74–282.ꢀ
[51] P.ꢀSudarsanam,ꢀB.ꢀHillary,ꢀM.ꢀH.ꢀAmin,ꢀS.ꢀB.ꢀA.ꢀHamid,ꢀS.ꢀK.ꢀBharga‐
va,ꢀAppl.ꢀCatal.ꢀB,ꢀ2016,ꢀ185,ꢀ213–224.ꢀ
2
[52] X.ꢀ H.ꢀ Li,ꢀS.ꢀ L.ꢀZhang,ꢀ Y.ꢀJia,ꢀX.ꢀX.ꢀLiu,ꢀ Q.ꢀZhong,ꢀJ.ꢀNat.ꢀGasꢀChem.,ꢀ
[
[
[
18] B.ꢀDhandapani,ꢀS.ꢀT.ꢀOyama,ꢀAppl.ꢀCatal.ꢀB,ꢀ1997,ꢀ11,ꢀ129–166.ꢀ
19] J.ꢀNawrocki,ꢀB.ꢀKasprzyk‐Hordern,ꢀAppl.ꢀCatal.ꢀB,ꢀ2010,ꢀ99,ꢀ27–42.ꢀ
20] M.ꢀT.ꢀN.ꢀDinh,ꢀJ.ꢀM.ꢀGiraudon,ꢀA.ꢀM.ꢀVandenbroucke,ꢀR.ꢀMorent,ꢀN.ꢀ
DeꢀGeyter,ꢀJ.ꢀF.ꢀLamonier,ꢀAppl.ꢀCatal.ꢀB,ꢀ2015,ꢀ172–173,ꢀ65–72.ꢀ
21] E.ꢀ Rezaei,ꢀ J.ꢀ Soltan,ꢀ N.ꢀ Chen,ꢀ Appl.ꢀ Catal.ꢀ B,ꢀ 2013,ꢀ 136–137,ꢀ
2012,ꢀ21,ꢀ17–24.ꢀ
[53] G.ꢀS.ꢀQi,ꢀR.ꢀT.ꢀYang,ꢀJ.ꢀCatal.,ꢀ2003,ꢀ217,ꢀ434–441.ꢀ
[54] X.ꢀ Gao,ꢀ Y.ꢀ Jiang,ꢀ Y.ꢀ Zhong,ꢀ Z.ꢀ Y.ꢀ Luo,ꢀ K.ꢀ F.ꢀ Cen,ꢀ J.ꢀ Hazard.ꢀ Mater.,ꢀ
2010,ꢀ174,ꢀ734–739.ꢀ
[
[
[
[55] B.ꢀ Guan,ꢀ H.ꢀ Lin,ꢀ L.ꢀ Zhu,ꢀ Z.ꢀ Huang,ꢀ J.ꢀ Phys.ꢀ Chem.ꢀ C,ꢀ 2011,ꢀ 115,ꢀ
2
39–247.ꢀ
22] S.ꢀDimitrova,ꢀG.ꢀIvanov,ꢀD.ꢀMehandjiev,ꢀAppl.ꢀCatal.ꢀA,ꢀ2004,ꢀ266,ꢀ
1–87.ꢀ
23] P.ꢀC.ꢀC.ꢀFaria,ꢀJ.ꢀJ.ꢀM.ꢀÓrfão,ꢀM.ꢀF.ꢀR.ꢀPereira,ꢀAppl.ꢀCatal.ꢀB,ꢀ2009,ꢀ88,ꢀ
12850–12863.ꢀ
8
[56] R.ꢀ Tholkappiyan,ꢀ K.ꢀ Vishista,ꢀ Appl.ꢀ Surf.ꢀ Sci.,ꢀ 2015,ꢀ 351,ꢀ 1016–ꢀ
1024.ꢀ

1280ꢀ
FaweiꢀLinꢀetꢀal.ꢀ/ꢀChineseꢀJournalꢀofꢀCatalysisꢀ38ꢀ(2017)ꢀ1270–1280ꢀ
[
[
[
57] J.ꢀ L.ꢀ Ayastuy,ꢀ A.ꢀ Gurbani,ꢀ M.ꢀ P.ꢀ González‐Marcos,ꢀ M.ꢀ A.ꢀ Gutiér‐
Carley,ꢀS.ꢀH.ꢀTaylor,ꢀM.ꢀJ.ꢀRosseinsky,ꢀG.ꢀJ.ꢀHutchings,ꢀJ.ꢀCatal.,ꢀ2011,ꢀ
281,ꢀ279–289.ꢀ
[61] L.ꢀZhong,ꢀQ.ꢀZhong,ꢀW.ꢀCai,ꢀS.ꢀZhang,ꢀY.ꢀYu,ꢀM.ꢀOu,ꢀF.ꢀJ.ꢀSong,ꢀRSCꢀ
Adv.,ꢀ2016,ꢀ6,ꢀ21056–21066.ꢀ
[62] A.ꢀY.ꢀWang,ꢀY.ꢀL.ꢀGuo,ꢀF.ꢀGao,ꢀC.ꢀH.ꢀF.ꢀPeden,ꢀAppl.ꢀCatal.ꢀB,ꢀ2017,ꢀ
202,ꢀ706–714.ꢀ
rez‐Ortiz,ꢀInt.ꢀJ.ꢀHydrogenꢀEnergy,ꢀ2012,ꢀ37,ꢀ1993–2006.ꢀ
58] M.ꢀKonsolakis,ꢀS.ꢀA.ꢀCarabineiro,ꢀP.ꢀB.ꢀTavares,ꢀJ.ꢀL.ꢀFigueiredo,ꢀJ.ꢀ
Hazard.ꢀMater.,ꢀ2013,ꢀ261,ꢀ512–521.ꢀ
59] X.ꢀQ.ꢀQiu,ꢀM.ꢀMiyauchi,ꢀK.ꢀSunada,ꢀM.ꢀMinoshima,ꢀM.ꢀLiu,ꢀY.ꢀLu,ꢀD.ꢀLi,ꢀ
Y.ꢀ Shimodaira,ꢀ Y.ꢀ Hosogi,ꢀ Y.ꢀ Kuroda,ꢀ K.ꢀ Hashimoto,ꢀ ACSꢀ Nano,ꢀ
2
012,ꢀ6,ꢀ1609–1618.ꢀ
[63] L.ꢀY.ꢀWang,ꢀX.ꢀX.ꢀCheng,ꢀZ.ꢀQ.ꢀWang,ꢀC.ꢀY.ꢀMa,ꢀY.ꢀK.ꢀQin,ꢀAppl.ꢀCatal.ꢀB,ꢀ
2017,ꢀ201,ꢀ636–651.ꢀ
[
60] S.ꢀA.ꢀKondrat,ꢀT.ꢀE.ꢀDavies,ꢀZ.ꢀL.ꢀZu,ꢀP.ꢀBoldrin,ꢀJ.ꢀK.ꢀBartley,ꢀA.ꢀF.ꢀ
球形氧化铝负载锰基双金属氧化物催化剂催化臭氧深度氧化NO
*
林法伟, 王智化 , 邵嘉铭, 袁定琨, 何 勇, 朱燕群, 岑可法
浙江大学能源清洁利用国家重点实验室, 浙江杭州310027
摘要: 在工业锅炉烟气处理领域, 由于锅炉容量低, 烟气温度往往无法满足传统选择性催化还原(SCR)所需温度窗口. 工业
锅炉烟气成分的复杂性也给氮氧化物治理带来了严峻考验. 臭氧深度氧化NO结合湿法洗涤同时脱硫脱硝技术具有独特的
应用优势. 传统臭氧氧化技术中, NO被臭氧氧化为NO
低, 需要在脱硫浆液中加入添加剂提高脱硝效率, 造成运行成本增加. NO经臭氧深度氧化后, NO
2 5
的N , 传统脱硫石膏浆液即可实现高效吸收N , 从而有效提高氮氧化物吸收效率. 但由于N O
2
, 进而在脱硫塔中实现一体化脱硫脱硝. 但由于NO
进一步转化为溶解度高
生成反应速率低, 深度
2
溶解度相对较
2
2
O
5
2
O
5
氧化存在臭氧投入量大、反应时间长及臭氧残留多的缺点. 臭氧耦合催化剂深度氧化NO可有效解决以上问题.
首先, 本文采用溶胶-凝胶法合成一系列单金属氧化物(Mn, Co, Ce, Fe, Cu, Cr)作为臭氧深度氧化NO的催化剂. 结果发
o
现锰氧化物表现出最高的催化活性, 在70 C下, O
3
/NO摩尔比为2.0时经过0.12 s的反应时间催化剂即可实现80%以上的转
化效率. 但根据N O 生成的总包反应(2NO + 3O = N O + 3O )可以看出, O /NO摩尔比为1.5时即可实现N O 的完全转化.
2
5
3
2
5
2
3
2
5
由于催化臭氧氧化反应温度较低, 中间产物在催化剂表面聚集, 占据大量活性位, 进而导致无法实现1.5摩尔比的高效转化.
通过采用球形氧化铝作为载体, 避免粉末状催化剂紧凑型布置, 增加换热面积, 可有效降低催化剂表面中间产物聚集; 同
时延长了气体与催化剂的接触时间, 提高反应效率. 在球形氧化铝载体上负载锰基双金属氧化物(Ce-Mn, Fe-M, Cr-Mn,
3
o
3
Cu-Mn和Co-Mn), 在初始NO浓度为410 mg/m , 反应温度100 C, O /NO摩尔比1.5, 催化反应时间0.12 s的工况下, 催化剂最
3
3
终实现95% (Fe-Mn)和88% (Ce-Mn)的转化效率, 剩余NO和NO
2
的浓度分别低于20 mg/m (Fe-Mn)和50 mg/m (Ce-Mn), 臭
3
氧残留浓度低于25 mg/m . 同负载单一锰氧化物(83%转化率)相比, 双金属氧化物进一步提高了N O 生成效率. 因此, 臭氧
2
5
3
耦合催化剂深度氧化NO结合湿法吸收在工业锅炉超低排放(NO
通过XRD、氮气吸附、H -TPR和XPS等手段研究了催化剂的晶体结构、孔结构参数、氧化还原性能和表面原子价态.
催化臭氧深度氧化NO主要与催化剂对臭氧的分解性能和对NO的氧化性能有关. 较大的比表面积和孔容有利于催化剂的
x
< 50 mg/m )领域具有广泛应用前景.
2
4
+
3+
吸附. 氧空位有利于臭氧的吸附和分解. Mn 和Mn 的均衡分布既有利于NO的吸附氧化又有利于臭氧的吸附分解, 最终
提高了N 生成效率.
2
O
5
关键词: 一氧化氮; 深度氧化; 催化剂; 臭氧; 过渡金属氧化物
收稿日期: 2017-04-14. 接受日期: 2017-05-17. 出版日期: 2017-07-05.
*
通讯联系人. 电话: (0571)87953162; 传真: (0571)87951616; 电子信箱: wangzh@zju.edu.cn
基金来源: 国家自然科学基金(51422605); 浙江省自然科学基金(LR16E060001).
本文的英文电子版由Elsevier出版社在ScienceDirect上出版(http://www.sciencedirect.com/science/journal/18722067).