Influence of different palladium precursors on the properties of solution-combustion-synthesized palladium/ceria catalysts for methane combustion

In luence o Di e en Palladium P ecu so s on he P ope ies o Solu ion-Combus ion-Syn hesized Palladium/ Ce ia Ca alys s o Me hane Combus ion Sa a Colussi,* [a] A up Gayen, [b] Ma a Boa o, [a] Jo di Llo ca, [c] and Alessand o T o a elli [a] A se ies o Pd/CeO 2 ca alys s was p epa ed by solu ion com- bus ion syn hesis (SCS) using di e en Pd p ecu so s. The pow- de s we e cha ac e ized by complemen a y echniques such as BET su ace a ea measu emen s, X-Ray di ac ion analysis, X- Ray pho oelec on spec oscopy, empe a u e-p og ammed e- duc ion, empe a u e-p og ammed oxida ion, and high- esolu- ion TEM. The esul s ob ained e idenced he o ma ion o a Pd-Ce solid solu ion on all SCS samples. This solid solu ion is in he o m o an o de ed supe cell s uc u e only in he SCS ca alys p epa ed om palladium ni a e. This was co ela ed o he hea o eac ion be ween Pd(NO 3 ) 2 and he uel. The samples we e es ed o me hane ca aly ic combus ion, and he eac ion a es on all SCS samples we e app oxima ely wice ha o he imp egna ed coun e pa , i espec i e o he p ecu so . This was a ibu ed o he p esence o he Pd-O-Ce solid solu ion, which ga e ise o s ong Pd–ce ia in e ac ions. In oduc ion Solu ion combus ion syn hesis (SCS) is an a ac i e me hod o he p epa a ion o ex emely dispe sed ca alys nanopow- de s in a single, cos -e ec i e s ep. [1] This echnique is based on he di ec eac ion o ca alys p ecu so s (oxidize s) wi h a uel ( educing agen ). All eac an s a e dissol ed in wa e , and he solu ion is hea ed in a u nace in which he combus- ion akes place wi h he o ma ion o a solid p oduc . The gene al scheme o he SCS o a me al oxide using oxalyl dihy- d azide (ODH, C 2 H 6 N 4 O 2 ) as a uel is shown in Equa ion (1): [2] nanopa icles compa ed o con en ional me hods. The e a e se e al examples ha demons a e how, wi h his echnique, i is possible o p epa e s uc u ed ca alys s (i.e. , monoli hs) by loading he ca alys powde di ec ly on o he su ace o he ce- amic o me al suppo . [3] Mo eo e , some pa en s desc ibe he possibili y o he indus ial scale-up o his syn hesis ap- p oach. [4] As a esul o he in iguing cha ac e is ics o he powde s and he simple p epa a ion me hod, many e o s ha e been de o ed o s udy he pe o mances o SCS ma e ials o di e en applica ions, which ange om h ee-way ca alys s o au omo i e exhaus s a e ea men [3a, 5] o composi e ma e- M ð NO 3 Þ2 ð aq Þ þ C 2 H 6 N 4 O 2 ð aq Þ ! MO s Þ þ 2 CO þ 3N þ3H O ð 2ðgÞ 2ðgÞ 2 ð g Þ ð 1 Þ ials o ion ba e ies, [6] ce amic powde s o solid oxide uel cells, [7] me al-based ca alys s o e o ming eac ions, [8] lumines- cen ma e ials, [9] bioma e ials o medicine use, [10] and ca alys s The powde s ob ained by his me hod show in e es ing p ope ies, such as small pa icle size and ine dispe sion, be- cause o he e olu ion o gases du ing eac ion ha can de- s oy la ge agglome a es and lead o he o ma ion o smalle o oxida ion eac ions. [11] Wi hin he las opic, we in es iga ed he pe o mances o Pd/CeO 2 ca alys s o me hane combus- ion p epa ed by SCS and compa ed hei ca aly ic ac i i y wi h ha o he co esponding imp egna ed samples. [12] We obse ed a signi ican inc ease in me hane combus ion a es [a] D . S. Colussi, P o . M. Boa o, P o . A. T o a elli Depa men o Chemis y, Physics and En i onmen Uni e si y o Udine Via del Co oni icio 108, IT-33100 Udine (I aly) E-mail : sa a.coluss[email p o ec ed] [b] D . A. Gayen Depa men o Chemis y Jada pu Uni e si y Kolka a 700032 (India) [c] P o . J. Llo ca Insi u de Tecniques Ene ge iques and Cen e o Resea ch in Nanoenginee ing Uni e si a Poli ecnica de Ca alunya, 08028 Ba celona (Spain) on SCS samples wi h espec o he imp egna ed ones. SCS samples p esen ed a peculia Pd-O-Ce supe cell s uc u e in which al e na e Ce a oms we e subs i u ed by Pd a oms, which led o he o ma ion o o de ed a ays o oxygen acan- cies oge he wi h highly unde coo dina ed O a oms. DFT cal- cula ions pe o med by Maye nick and Janik [13] p o ed heo e i- cally ha his supe s uc u e is ex emely ac i e o me hane combus ion, in ag eemen wi h ou expe imen al esul s. O he au ho s indica ed ha he inco po a ion o me al ions in o an oxide la ice, which implies he o ma ion o oxide ion acan- cies, is esponsible o he highe ca aly ic ac i i y in combus- ion eac ions ha ypically in ol e oxygen and uel dissocia- ion on he ca alys su ace. [11b] SCS samples p esen o he cha ac e is ics ha could be in- ol ed in he highe ac i i ies eco ded o hese ca alys s. Fo Table 1. Pd p ecu so sal s, Pd loading, su ace a eas, and la ice pa ame e s o he ca alys s conside ed in his wo k. Sample Pd p ecu so Pd loading Su ace a ea R wp R F2 c 2 Cell pa ame e Composi ion [b] [w %][a] [m 2 g - 1 ] [%] IWI Pd(NO 3 ) 2 1.74 12.2 6.8 4.7 2.41 5.4110 (1) – SCS2-N Pd(NO 3 ) 2 1.85 6.3 7.7 4.8 1.69 5.4103 (1) Ce 0.97 Pd 0.03 O 2 - d SCS2-Cl PdCl 2 1.84 5.7 6.4 4.9 1.16 5.4108 (1) Ce 0.99 Pd 0.01 O 2 - d SCS2-Ac Pd(O 2 CCH 3 ) 2 1.41 8.9 6.2 4.5 1.23 5.4105 (1) Ce 0.98 Pd 0.02 O 2 - d CeO 2 SCS – – 16.2 6.9 7.1 3.10 5.4113 (1) – [a] As measu ed by induc i ely coupled plasma a omic emission spec oscopy (ICP-AES). [b] Calcula ed acco d- ing o Vega d’s law and assuming a Ce 4+ ionic adius o 0.97 Ç and a Pd 2+ ionic adius o 0.86 Ç. R wp = weigh ed 2 = s uc u e R ac o . c 2 = ( R / R ) 2 , whe e R is he expec ed R ac o example Specchia e al. epo ed he p esence o dispe sed Pd 2+ ions oge he wi h Pd and PdO on Pd-ce ia-zi conia sam- ples p epa ed by SCS and a ibu ed he high ac i i y obse ed o a syne gis ic e ec be ween he di e en Pd species. [11a] We ha e epo ed a simila si ua ion o he ca aly ic combus ion o p opane and dime hyl e he on Pd/CeO 2 ca alys s, in which he coexis ence o highly ac i e Pd II cen e s wi h Pd and PdO species was conside ed o be esponsible o he highe eac- ion a es o SCS samples wi h espec o imp egna ed ones. [14] To ob ain he desi ed p ope ies and maximize he po en ial o SCS ca alys s, one should be able o inely une he syn he- sis p ocedu e. F om a p ac ical poin o iew, SCS is a e y simple echnique and he heo e ical backg ound is clea ; ne - e heless, he in luence o indi idual pa ame e s (i.e., ole o uel, p ecu so s, combus ion empe a u e, e c.) on he syn he- sized ma e ials a e complex o s udy because o he high exo- he mici y o he eac ion and i s as e olu ion. Recen ly, some s udies ha e ocused on he ole o he uel- o-oxidize a io and on eac ion he modynamics modeling, [10b, 15] bu many as- pec s s ill need o be in es iga ed in dep h. In his wo k, we y o add ess his issue by cha ac e izing SCS Pd/CeO 2 samples p epa ed om di e en Pd p ecu so s o e alua e he e ec o his pa ame e on he o ma ion o he supe cell s uc u e. Mo eo e , we s udy he ca aly ic beha io o he ma e ials o he combus ion o me hane by eac ion a e measu emen s pe o med in a ecycle eac o , wi h he aim o gi e mo e in- sigh s in o hei peculia ca aly ic ac i i y. Resul s and Discussion The Pd p ecu so , Pd loading, and su ace a ea o he samples conside ed in his wo k a e epo ed in Table 1. The measu ed Pd amoun is close o he nominal loading o 2 w %, and he loading on SCS2-Ac is sligh ly lowe han ha o he o he samples. In he ollowing, all he quan i a i e esul s will be calcula ed on a pe g am o Pd basis. In gene al all ca alys s p esen low su ace a eas, i espec i e o he p epa a ion me hod and he Pd p ecu so . The addi ion o Pd sal du ing SCS, e en i in e y low amoun , dec eases he su ace a ea o he ca alys s signi ican ly, as e idenced by compa ing Pd/CeO 2 wi h he CeO2 suppo p epa ed by he solu ion combus ion o ce ic ammonium ni a e wi h ODH. Figu e 1. XRD p o iles o he SCS samples. Inse : Pd me al peak (2 q = 40.28). XRD p o iles o he SCS samples a e shown in Figu e 1. The XRD pa e ns o he ce ia suppo s a e almos iden ical o all samples, whe eas a di e ence can be obse ed o he Pd me al peaks (2 q = 40.28). The p esence o Pd me al on SCS2-N can be seen clea ly in he inse o Figu e 1, and he Pd me al peak has a lowe in ensi y o SCS2-Cl and SCS2-Ac. No peaks om palladium oxide (2 q = 33.98) can be dis inguished on any o he SCS samples. The mean c ys al size o ce ia calcula ed by he Sche e equa ion is be ween 30 and 45 nm o all ca a- lys s. On he sample p epa ed by incipien we ness imp egna- ion (IWI ; Figu e S1) no Pd me al is de ec ed, bu small PdO peaks a 2 q = 33.9 and 42.38 a e obse ed. The calcula ed c ys- al size o ce ia in he IWI ca alys is 16 nm. Rie eld analysis o he SCS ca alys s indica es ha he s uc- u al ea u es a e compa ible wi h he p esence o a solid solu- ion in which Pd pa ially en e s he ce ia la ice (Table 1). The amoun o Pd in solid solu ion, calcula ed acco ding o Ve- ga d’s law, a ies sligh ly wi h he Pd p ecu so . The o ma ion o a Ce1-xPdxO2-d solid solu ion on 1 a % Pd/CeO2 samples p e- pa ed by SCS s a ing om a PdCl 2 p ecu so was epo ed by P iolka and co-wo ke s. [16] Acco ding o hei wo k, his solid solu ion comp ises Pd 2+ ions s abilized in Ce 4+ si es. Fo he IWI sample, he a ia ion in he la ice pa ame e is oo small o assume he inclusion o Pd in o he ce ia la ice. The X- ay pho oelec on spec oscopy (XPS) esul s o he as-p epa ed and spen ca alys s a e summa ized in Table 2 (see Figu e S7 o he Suppo ing In o ma ion o XPS spec a). By using XPS, i has been possible o iden i y di e en Pd species ha ha e been asc ibed o Pd0 me al (binding ene gy (BE) = 335.6 eV), Pd II species (BE = 337.6 eV), and nons oichiome ic palladium oxide wi h an in e - media e oxida ion s a e (BE = 336.5 eV). [17] I can be obse ed ha , in ag eemen wi h he XPS spec a o esh samples, Pd me al is de ec ed only on he SCS samples, whe eas IWI shows p o ile R ac o . R F wp exp exp he p esence o oxidized Pd spe- Table 2. XPS analysis o all samples. Sample a omic a io Pd 0 PdO x Pd II Pd/Ce [%] [%] [%] IWI 0.078 0 22.6 77.4 SCS2-N 0.053 10.6 81.1 8.3 SCS2-Cl 0.107 20.9 79.1 0 SCS2-Ac 0.046 24.7 66.5 8.8 IWI [a] 0.063 – 13.7 86.3 SCS2-N [a] 0.067 – 19.6 80.4 SCS2-Cl [a] 0.067 – 28.5 71.5 SCS2-Ac [a] 0.058 – 17.0 82.9 [a] Used ca alys s, ha is, samples a e one hea ing/cooling cycle du ing me hane combus ion. cies only. Nons oichiome ic PdO x species a e p esen in a con- side ably lowe amoun compa ed o ha on he SCS ca alys s, which indica es ha bulk PdO is he p edominan phase on he imp egna ed sample. Howe e , he ca alys s p epa ed by SCS comp ise mainly PdO x , and he amoun o Pd me al de- c eases in he o de SCS2-Ac > SCS2-Cl > SCS2-N. PdO is equally p esen on SCS2-N and SCS2-Ac and absen on SCS2-Cl. The amoun o nons oichiome ic Pd oxide shows an in e se end wi h espec o he Pd me al con en (SCS2-N > SCS2-Cl > SCS2-Ac). The use o di e en Pd p ecu so s has an in luence on he dis ibu ion o Pd species in he esh samples, and om XPS esul s i is clea ha SCS leads o he o ma ion o less oxidized ca alys s. A di e en pic u e is e ealed i we look a he spec a o he used ca alys s, collec ed a e one hea ing/cooling cycle o me hane combus ion. The Pd me al has disappea ed, and oxi- dized PdII species domina e he su ace composi ion. This is e iden pa icula ly o SCS ca alys s, bu i is ue also o IWI, which emains sligh ly iche in PdO wi h espec o SCS sam- ples. High- esolu ion ansmission elec on mic oscopy (HRTEM) images o he SCS samples a e shown in Figu es 2–4. In gene - al, all ca alys s show well-de ined ce ia c ys alli es wi h pa icle sizes be ween 20 and 50 nm, in good ag eemen wi h XRD da a. The sample p epa ed om palladium ni a e (SCS2-N) e- eals he p esence o he supe cell s uc u e iden i ied in ou p e ious wo k. [12] A cha ac e is ic image o his ca alys is shown in Figu e 2. In he Fou ie ans o m (FT) image, c ys al- log aphic spacings a 2.71 and 3.13 Ç a e eco ded, which co - espond o he (2 0 0) and (111) c ys allog aphic planes o CeO 2 . In he o he pa o he c ys al (Figu e 2, igh ), la ice inges a 5.7 Ç co espond o he o bidden {1 0 0} di ac ion, which becomes isible in he o de ed (2 x 1)(11 0) supe s uc- u e in which one Pd a om subs i u es one Ce a om e e y wo in he c ys al s uc u e. [12] HRTEM images o SCS2-Cl and SCS-Ac a e shown in Figu es 3 and 4, espec i ely. Bo h hese samples show no e idence o he o ma ion o he o de ed supe cell s uc u e. Fo SCS2-Cl, only CeO 2 c ys alli es a e iden i ied and no (2 2 2) la ice inges ha a e a ibu ed o he Pd-O-Ce supe s uc u e a e obse ed in he HRTEM image. The analysis o SCS2-Ac e eals ha his sample is cons i u ed by well-de ined CeO 2 c ys alli es. A high- Figu e 2. HRTEM images o SCS2-N ( op) and hei associa ed FT pa e ns (bo om). Figu e 3. HRTEM images o SCS2-Cl. a and b co espond o di e en pa icles o he sample. magni ica ion HRTEM image ha shows he p esence o he (111) c ys allog aphic planes o CeO 2 is shown in Figu e 4 b. In addi ion o CeO 2 c ys alli es, elec on-dense pa icles a e also iden i ied in he image shown in Figu e 4 a. These a e oo small ( ð 1 nm in diame e ) o an accu a e la ice inge analy- sis, bu hei elec on-dense na u e sugges s ha hey co e- spond o me allic Pd. The esul s o XPS (Table 2) and XRD, in which a small peak a 2 q = 40.28 is obse ed (inse o Fig- u e 1 a), suppo his iden i ica ion. F om he HRTEM analysis, i Figu e 4. HRTEM images o SCS2-Ac. a and b co espond o di e en pa i- cles o he sample. is possible o conclude ha he o de ed supe cell s uc u e de- sc ibed in ou p e ious wo k [12] is p esen only on SCS2-N, which sugges s he ole o he Pd p ecu so in he o ma ion o his pa icula s uc u e. Fo he o he samples, he subs i u- ion o Pd 2+ ions in o he ce ia la ice, as in e ed om XRD, is no obse ed by HRTEM possibly because i is no o de ed and/o i is limi ed o a e y hin laye . This supe cell s uc u e can be ob ained only by he dissolu ion o Pd 2+ species wi h he o de ing o he oxygen acancies; he p esence o a Pd- Ce-O solid solu ion does no necessa ily imply he o ma ion o a supe s uc u e. The HRTEM image o IWI does no e eal any pa icula ea- u es o his sample. Spo s a 2.1 Ç in he FT image co espond o (11 0) planes o PdO, which indica es ha Pd is p esen in he o m o PdO, in ag eemen wi h XPS and XRD indings. The palladium oxide pa icle size on IWI is be ween 5 and 10 nm (Figu e S2). We in es iga ed he spen SCS2-N ca alys , ha is, he sample a e one me hane combus ion cycle, by HRTEM. A la - ice inge image eco ded a a magni ica ion o 10 6 is shown in Figu e 5. The p esence o a supe cell s uc u e is e iden be- cause he e a e spo s no pe mi ed in CeO 2 . This supe cell s uc u e is again gene a ed by he pa ial subs i u ion o Ce4+ Figu e 5. HRTEM image o spen SCS2-N. by Pd2+ in CeO2 wi h he concomi an c ea ion o o de ed oxygen acancies. To in es iga e he ole o he Pd p ecu so on he p ope ies o he ca alys s p epa ed by SCS, we s udied he e ec o he di e en sal s on he en halpy o eac ion D H eac o each SCS sample, and he esul s a e p esen ed in Table 3. The calcula- Table 3. D H eac calcula ed o he syn hesis o each SCS sample. Sample Pd p ecu so DH eac [kcal mol-1] Pd sal +CAN+ODH Pd sal +ODH SCS2-N Pd(NO 3 ) 2 - 748 - 291 SCS2-Cl PdCl 2 - 737 4 SCS2-Ac Pd(O 2 CCH 3 ) 2 - 740 - 85 ion o D H eac was pe o med o bo h he global combus ion eac ion [acco ding o Equa ions (2)–(4) in he Expe imen al Sec ion] and he eac ion o he Pd sal wi h ODH alone. This la e si ua ion can occu i some Pd sal is no dissol ed com- ple ely o in he immedia e su oundings o Pd sal molecules ha eac di ec ly wi h he uel. I can be obse ed ha he e- ac ion pe o med wi h he ni a e p ecu so has a highe D H eac (absolu e alue), and he di e ence is mo e e iden i we exclude he con ibu ion o ce ic ammonium ni a e. The highe exo he mici y obse ed o he palladium ni a e p e- cu so migh play a ole in he o ma ion o he o de ed supe - cell s uc u e obse ed only o he SCS2-N sample. In gene al, he he modynamic p ope ies o a solid solu ion change ac- co ding o he mu ual posi ion o he a oms, [18] so i is eason- able o assume ha di e en en halpies o eac ion can esul in he di e en o de ing o Pd, O, and Ce a oms. Tempe a u e-p og ammed educ ion (TPR) and empe a u e- p og ammed oxida ion (TPO) we e pe o med o in es iga e he edox beha io o he ca alys s. The TPR p o iles o he IWI and SCS ca alys s a e shown in Figu e 6a and b, espec i ely. Gene ally, all samples show h ee hyd ogen-up ake egions : a low- empe a u e (LT) egion be ween 233 and 423 K, a medium- empe a u e (MT) egion be ween 573 and 873 K, and a high- empe a u e (HT) egion abo e 923 K. F om a quali- a i e poin o iew, he TPR p o iles o SCS and IWI ca alys s a e clea ly di e en in he LT egion. In e es ingly, IWI p esen s a single, sha p peak o hyd ogen up ake, whe eas he SCS samples show a complex p o ile ha includes a combina ion o hyd ogen-up ake and - elease peaks in a wide empe a u e ange. The di e ence be ween SCS ca alys s and IWI sugges s ha on he imp egna ed sample Pd is p esen in a mo e ho- mogeneous o m han ha on he SCS ones ; his e ec migh be ela ed o di e en pa icle size dis ibu ion and/o o he p esence o di e en PdO x species, acco ding o ha de ec ed by XPS analysis. The peaks in he LT egion can be a ibu ed o he educ- ion o Pd- ela ed species as epo ed widely in he li e a- u e ; [19] whe eas he peaks in he MT and HT egions can be a - ibu ed o he educ ion o su ace and bulk ce ia species, e- spec i ely. [20] On he SCS samples, H 2 up ake s a s ð 50 K lowe han ha on IWI, which indica es ha he SCS ca alys s Figu e 6. TPR p o iles o a) IWI and b) SCS samples. a e educed mo e easily han IWI. This e idence ma ches well wi h he XPS esul s, which indica e ha IWI is he mos oxi- dized ca alys . MT peaks a e mo e p onounced and co e a b oade empe a u e ange o SCS2-Ac and SCS2-Cl, whe eas o SCS2-N hey a e simila o hose o IWI. In he HT egion, all he samples show a simila end o hyd ogen up ake. SCS2-N and SCS2-Ac show a simila beha io , wi h he p es- ence o wo hyd ogen-consump ion peaks ollowed by a hyd o- gen- elease peak caused by he decomposi ion o b- ype palla- dium hyd ides. [19] In he case o SCS2-Cl, he peak caused by hyd ogen elease appea s in he middle o he wo hyd ogen- up ake ea u es and i s e olu ion is clea ly s opped by he ap- pea ance o he second peak. The gene al iden i ica ion o e e y peak o he LT egion is no s aigh o wa d. Acco ding o Luo and Zheng, [21] he spli ing o he hyd ogen-consump- ion peak a low empe a u e o e a PdO/Ce 0.5 Z 0.5 O 2 ca alys is ela ed o he condi ions o Pd oxida ion and can be caused by he nonuni o m size dis ibu ion o PdO and/o o he inclu- sion o some PdO in o he suppo a e edox cycles. In his case, he inclusion o PdO in he ce ia la ice is obse ed o all SCS samples ; mo eo e , he pa icula cha ac e is ics o SCS ( as e olu ion, exo he mici y, di e en local empe a u es du ing combus ion) migh gi e ise o nonuni o m Pd pa icle size and di e en Pd oxida ion s a es. F om XPS analysis, i is obse ed ha he p edominan phase on hese ca alys s is a nons oichiome ic palladium oxide ha is likely o eac wi h hyd ogen in a mo e complex way han bulk PdO, which p e- domina es on IWI. I is also known ha noble me als de elop s ong in e ac ions on ce ia unde educing condi ions ha can esul , o example, in he deco a ion o Pd pa icles by CeO 2 . [22] Unde he condi ions o SCS, s ongly educing en i- onmen s migh be es ablished locally du ing eac ion leading o he o ma ion o such deco a ed pa icles and/o o o he ypes o Pd-Ce alloys [23] beside he o de ed supe cell s uc u e obse ed on SCS2-N. F om he quan i a i e analysis epo ed in Table 4, i appea s ha he e is an amoun o hy- d ogen spillo e on o he sup- po o SCS2-Ac as he heo e i- cal pe cen age o PdO educ ion exceeds 100 %. The same is ue o IWI because he eco ded hy- d ogen consump ion is ð 60 % highe han he s oichiome ic amoun calcula ed o PdO e- duc ion. Fo SCS2-N and SCS2- Cl, a deg ee o educ ion close (SCS2-Cl) o sligh ly abo e uni y (SCS2-N) likely includes a educ- ion o he ce ia suppo as hese samples p esen a ce ain amoun o me allic Pd (seen om XRD, XPS, and HRTEM e- sul s). The spillo e o H 2 on o a educible suppo in he p es- ence o a noble me al is a well- known phenomenon. [24] Table 4. Quan i a i e analysis o TPR expe imen s (H 2 up ake/ elease cal- cula ed up o 423 K). Sample H 2 up ake H 2 elease PdO ed PdH dec [mL] [mL] [%] [a] [%] [b] IWI 0.39 – 164 – SCS2-N 0.28 0.06 107 22 SCS2-Cl 0.23 – 90 – SCS2-Ac 0.27 0.02 142 10 [a] % o PdO educed du ing TPR, calcula ed on he basis o he ac ual Pd loading. [b] % o PdH decomposed du ing TPR, calcula ed on he basis o he ac ual Pd loading. The hyd ogen consump ion in he MT egion is e y small o IWI and SCS2-N and app oxima ely h ee o ou imes highe o SCS2-Cl and SCS2-Ac, espec i ely. This hyd ogen up ake is smalle han ha obse ed on pu e CeO 2 p epa ed by SCS (Suppo ing In o ma ion). This is consis en wi h he spillo e obse ed in he LT egion: in he p esence o he noble me al, ce ia ends o be educed a a lowe empe a u e. The di e ences be ween he samples p epa ed om di e en p ecu so s a e di icul o explain in he MT egion, bu hese migh be ela ed o he di e en e olu ion o ca bon-con ain- ing species ha a e s ill p esen on he ca alys su ace. TPO expe imen s add some o he de ails o he edox p op- e ies o he ca alys s. The oxygen- elease and -up ake p o iles o SCS samples du ing he hi d TPO cycle a e shown in Figu e 7. The hi d cycle is chosen as ep esen a i e as no u - he modi ica ion o TPO p o iles can be de ec ed om his cycle onwa ds; mo eo e , his cycle is somehow ep esen a i e o he s a e o he ca alys s du ing he combus ion cycles. F om he cooling pa o he cycles, i can be obse ed ha he empe a u e and he dynamics o Pd eoxida ion a e simi- Table 5. Quan i a i e analysis o TPO expe imen s . Sample Oxygen elease Oxygen up ake PdO decomposi ion Pd oxida ion [x 10-3 mmol gPd-1] [x 10-3 mmolgPd-1] [%] [%] IWI 3.76 3.79 80 81 SCS2-N 3.34 3.28 71 69 SCS2-Cl 3.42 3.40 73 72 SCS2-Ac 4.06 4.29 86 91 Table 6. Reac ion a es o all samples calcula ed a 573 K. Sample Reac ion a es mmol gPd-1 s-1 x102 mmol m-2 s-1 IWI 10.0 1.4 SCS2-N 23.6 6.9 SCS2-Cl 23.4 7.6 SCS2-Ac 25.2 4.0 2PdIW_CeSCS 9.7 1.3 2Pd/Al 2 O 3 23.9 0.5 Figu e 7. TPO p o iles o he hi d cycle o all SCS samples (solid line : hea - ing; do ed line: cooling). la o all ca alys s, exhibi ing a single peak wi h a minimum a ð 920 K. A di e en pic u e appea s i we look a he oxygen- elease p o ile (hea ing amp): in his case a clea di e ence is obse ed be ween SCS2-N and SCS2-Ac on one side and SCS2- Cl on he o he . PdO decomposi ion akes place mos ly below 1073 K on SCS2-Cl, whe eas a signi ican pa o PdO is decom- posed abo e 1073 K o SCS2-N and SCS2-Ac. Fo alumina-sup- po ed Pd, he ac ion o PdO ha decomposes a a highe empe a u e has been a ibu ed en a i ely o su ace PdO, o PdO ha in e ac s s ongly wi h he suppo ,[17] and o he de- composi ion o a co e o PdO co e ed by a shell o Pd me al.[25] On hese Pd-ce ia sys ems, we can hypo hesize a simi- la scena io in which he occu ence o sepa a e s eps du ing PdO decomposi ion migh co espond o di e en PdO species o o PdO in di e en en i onmen s (bound mo e s ongly o he suppo , co e ed by Pd me al e c.). In gene al, hese s eps a e mo e e iden on he SCS samples han on he IWI sample (Figu e S3) in ag eemen wi h he esul s om TPR and wi h XPS analysis, which shows he p edominance o PdOx on he SCS ca alys s. Quan i a i e analysis pe o med du ing TPO indica es clea ly ha a highe amoun o Pd unde goes oxida ion-decomposi- ion on SCS2-Ac (Table 5). The amoun o Pd cycling is in he o de SCS2-Ac > IWI > SCS2-Cl SCS2-N. The esul s o eac ion a e measu emen s du ing me hane combus ion pe o med in he ecycle eac o a e epo ed in Table 6. The use o he ecycle eac o ensu es he di e en ial condi ions necessa y o ob ain consis en a e alues. Mo e- o e , he measu emen o he a e du ing he second ca aly ic cycle a oids he p oblems ela ed o sin e ing and edispe sion o Pd pa icles du ing he eac ion [26] and ensu es s able ca a- ly ic beha io because he hi d cycle is almos iden ical o he second one. The eac ion a es measu ed o SCS samples based on noble me al loading a e mo e han wice he a es measu ed on IWI. The di e ence is e en highe i we compa e he esul s based on he su ace a ea as he su ace a ea o he SCS samples is much lowe han ha o IWI. This e ec is al- eady known o SCS2-N o he combus ion o me hane, [12] p opane, and dime hyl e he . [14] To e alua e whe he his e ec could be ela ed o a pa icula s uc u e o he ce ia suppo made by SCS ha can in e ac di e en ly wi h Pd, a sample ha consis ed o Pd imp egna ed on CeO 2 made by SCS has been p epa ed and es ed. This ca alys , 2PdIW_CeSCS, showed a eac ion a e e y close o ha o IWI, which hus ex- cludes an e ec o he SCS ce ia suppo in he enhancemen o he ca aly ic pe o mance. In addi ion, we measu ed he e- ac ion a e on a 2 w % Pd/Al 2 O 3 ca alys p epa ed by incipien we ness imp egna ion wi h a simila p ocedu e o ha epo - ed o he IWI sample. The esul s indica e ha , al hough he su ace a ea is much highe han ha o he SCS samples (93.5 m 2 g - 1 ), he eac ion a e a 573 K is simila . As widely known om he li e a u e, [27] he alumina-based ca alys shows a much highe deac i a ion du ing he cooling b anch o he ligh -o cu e (Figu e S7) wi h espec o ce ia-based ma e ials, which makes he ce ia-based ma e ials e y in e es ing om he poin o iew o p ac ical applica ions. The alues o he eac ion a es measu ed on SCS samples indica e ha he e is no signi ican e ec o he di e en Pd p ecu so s on ca aly ic ac i i y, al hough i is clea ha he syn- hesis ou e plays a majo ole. XPS analysis o esh samples e idences a highe deg ee o oxida ion o IWI han he SCS ca alys s; e en so his di e ence almos disappea s a e one me hane combus ion cycle as e ealed by XPS o he spen ca alys s. These e idences end o exclude ha di - e en Pd oxida ion s a es a e esponsible o he highe eac ion a es eco ded on all SCS samples. Howe e , he p esence o he o de ed Pd-O-Ce su- pe s uc u e on SCS2-N a e eac ion de ec ed by HRTEM suppo s he ole o his s uc u e, and ha o i s nono de ed coun e pa , in he highe me hane combus ion ac i i y. On he basis o hese esul s, i is easonable o conclude ha he p esence o highly dispe sed Pd 2+ ions in he ce ia la ice, o - de ed o no , is key o achie e supe io me hane combus ion ac i i y. This explana ion is in ag eemen wi h he wo k o Ca gnello e al. in which a e y high ac i i y o me hane com- bus ion on Pd/ce ia ca alys was eco ded, which was ela ed o he p esence o unde coo dina ed Pd a oms in close in e ac- ion wi h CeO 2 . [28] P iolka and co-wo ke s a ibu ed he highe 0:03PdðNO3Þ2þ0:97CANþ2:358ODH ! 0:03PdOþ0:97CeO2 þ 4:716 CO 2 þ 10:954 H 2 O þ 8:626 N 2 0:03 PdCl2 þ0:97 CANþ2:328 ODH ! 0:03 PdOþ0:97 CeO2 þ 4:641 CO 2 þ 10:864 H 2 O þ 8:536 N 2 þ 0:03 Cl 2 þ 0:135 C ð 2 Þ ð 3 Þ ac i i y obse ed o CO oxida ion and NO educ ion o he o ma ion o a Pd-O-Ce solid solu ion. [16] Mo eo e , he heo e - ical wo k pe o med by Maye nick and Janik indica es ha he p esence o subs i u ional Pd ions in he ce ia la ice lowe s no only he ene gy ba ie o he o ma ion o oxygen acan- cies bu also he eac ion ene gy o me hane adso p ion, ha is, he i s s ep o ca aly ic combus ion. [13] Conclusions We cha ac e ized di e en Pd-ce ia ca alys s p epa ed by solu- ion combus ion syn hesis (SCS) s a ing om di e en Pd p e- cu so s and es ed o me hane ca aly ic combus ion. Cha ac- e iza ion esul s indica e he p esence on all SCS samples o a Pd-Ce solid solu ion, which appea s as an o de ed supe cell s uc u e on SCS2-N. The occu ence o his o de ed s uc u e has been ela ed en a i ely o he en halpy o eac ion DH eac be ween palladium ni a e and he uel (oxalyl dihyd azide) du ing SCS. The alue o DH eac o Pd(NO 3 ) 2 is much mo e exo he mic han ha o Pd(O 2 CCH 3 ) 2 and PdCl 2 (sligh ly endo- he mic). The esul s ob ained in his s udy explain he highe ca aly ic ac i i y o me hane combus ion eco ded on all SCS samples in ela ion o hei peculia s uc u e : on one side his pa icula Pd–ce ia in e ac ion is e iden om s uc u al and mo phological cha ac e iza ion (X- ay pho oelec on spec os- copy, XRD, high- esolu ion TEM), on he o he i is unde lined by he edox beha io o hese samples ( empe a u e-p o- g ammed educ ion and oxida ion), which is clea ly di e en and mo e complex wi h espec o ha o he imp egna ed coun e pa , ega dless o he Pd p ecu so . I seems hen plau- sible o asc ibe he supe io ac i i y o me hane combus ion eco ded on SCS samples o he subs i u ion o Pd 2+ ions in o he ce ia la ice, which is andom o SCS2-Cl and SCS2-Ac and o de ed o SCS2-N. Ou esul s indica e ha hese ca aly ic sys ems show p omise in he ield o me hane aba emen o he con ol o g eenhouse gas emissions in ligh o hei com- pa ison wi h s a e-o - he-a Pd-alumina ca alys s. Expe imen al Sec ion A se ies o 2 w % Pd/CeO 2 ca alys s was p epa ed by SCS. In his echnique Pd and Ce p ecu so sal s a e mixed in a sui able amoun o wa e o which a educe , o uel, is added. In his wo k, (NH 4 ) 2 Ce(NO 3 ) 6 (ce ic ammonium ni a e, CAN) was used as he CeO 2 p ecu so , and Pd(NO 3 ) 2 , PdCl 2 , and Pd(O 2 CCH 3 ) 2 we e used as Pd p ecu so s. The uel was ODH (C 2 H 6 N 4 O 2 ). Fuel and p ecu so s in a s oichiome ic amoun we e dissol ed in wa e and placed in a p ehea ed u nace (T ð 623 K) o he combus ion o ake place. A e comple e combus ion, he samples we e emo ed om he u nace and cooled o RT. The o e all heo e ical combus ion eac- ions a e shown in Equa ions (2)–(4): 0:03 Pd ð O 2 CCH 3 Þ2 þ 0:97 CAN þ 2:328 ODH ! 0:03 PdO þ 0:97 CeO 2 þ 4:656 CO 2 þ 10:954 H 2 O þ 8:536 N 2 þ 0:12 C ð 4 Þ The amoun o uel o each eac ion was calcula ed o balance he ne oxidizing alence o he sal s in ol ed by conside ing only ni- a e compounds as he oxidize . [2] The IWI ca alys ha consis ed o 2 w % Pd on comme cial CeO 2 (G ace Da ison) was p epa ed by adi ional incipien we ness imp egna ion o compa ison wi h SCS samples om a solu ion o 10 w % Pd(NO 3 ) 2 (Ald ich, 99.999 %). The imp egna ed sample was d ied o e nigh a 393 K and hen calcined a 1173 K in a u nace wi h ci cula ing ai o 4 h. Cha ac e iza ion me hods a e epo ed in he Suppo ing In o ma- ion. Ca aly ic es s and eac ion a e measu emen s we e pe - o med in a ecycle eac o [14] wi h a o al low a e o 40 mL min - 1 and a ecycle a io o 35. The eed gas composi ion was 0.5 % CH 4 and 2% O 2 in He. The sample (100 mg, p essed and sie ed) was loaded in o a qua z mic o eac o (i.d. 4 mm, leng h 400 mm) on a qua z wool bed, and he empe a u e was moni o ed by using a he mocouple posi ioned in he ca alys bed. The e luen gases we e moni o ed by using an Omnis a Quad upole Mass Spec om- e e and a Va ian CP 4900 mic o gas ch oma og aph equipped wi h molecula sie e and Po aplo Q columns. Fo eac ion a e measu emen s, a i s hea ing/cooling cycle was pe o med up o 1173 K, du ing which he ligh -o cu es o me hane combus ion we e eco ded (Suppo ing In o ma ion), hen he empe a u e was inc eased o 573 K a a hea ing a e o 10 K min - 1 . The a e was cal- cula ed using he con e sion alue measu ed a e 2 h o holding ime a 573 K; du ing his pe iod no signi ican a ia ion in con e - sion alues was obse ed. Acknowledgemen s We hank egione FVG o suppo . J.L. is a Se a Hffln e Fellow and is g a e ul o he ICREA Academia p og am and MINECO g an ENE2012-36368. Keywo ds: ce ium · he e ogeneous ca alysis · hyd oca bons · palladium · educ ion [1] a) S. T. A una, A. S. Mukasyan, Cu . Opin. Solid S a e Ma e . Sci. 2008, 12 , 44 –50; b) S. L. Gonz llez-Co ƒs, F. E. Imbe , Appl. Ca al. A 2013, 452 , 117– 131. [2] K. Su esh, K. C. Pa il in Pe spec i es in Solid S a e Chemis y (Ed.: K. J. Rao), Na osa Publishing House, New Delhi, 1995, pp. 376 –388. [3] a) S. Sha ma, M. S. Hegde, R. N. Das, M. Pandey, Appl. Ca al. A 2008, 337 , 130 –137; b) A. Vi a, G. C is iano, C. I aliano, S. Specchia, F. Cipi i, V. Specchia, In . J. Hyd ogen Ene gy 2014, 39 , 18592– 18603. [4] a) P. Dinka, A. Mukasyan, WO2007019332A1, 2007; b) Q. Weng, D. Sun, S. Lin, J. Chen, S. Xie, R. Huang, L. Zheng, CN101503188 (A), 2009. [5] P. Be a, K. C. Pa il, M. S. Hegde, Phys. Chem. Chem. Phys. 2000, 2 , 3715 – 3719. [6] a) B. T. Zhao, X. Yu, R. Cai, R. Ran, H. T. Wang, Z. P. Shao, J. Ma e . Chem. 2012, 22 , 2900 –2907; b) X. Li, Q. Xiao, B. Liu, H. C. Lin, J. B. Zhao, J. Powe Sou ces 2015, 273 , 128 –135. [7] a) E. Thomas, D. H. Lee, M. Y. Yoon, S. H. Eh man, H. J. Hwang, Ce am. In . 2011, 37 , 2269 –2274; b) C. Y. Zhu, A. Nobu a, Y. W. Ju, T. Ishiha a, T. Akiyama, In . J. Hyd ogen Ene gy 2013, 38 , 13419 –13426. [8] a) A. Kuma , A. S. Mukasyan, E. E. Wol , Appl. Ca al. A 2011, 401 , 20– 28; b) A. Vi a, G. C is iano, C. I aliano, L. Pino, S. Specchia, Appl. Ca al. B 2015, 162 , 551 –563. [9] a) G. Ramak ishna, H. Nagabhushana, D. V. Suni ha, S. C. P ashan ha, S. C. Sha ma, B. M. Nagabhushana, Spec ochim. Ac a Pa A 2014, 127 , 177 –184 ; b) Y. S. Vidya, K. S. Anan ha aju, H. Nagabhushana, S. C. Sha ma, H. P. Nagaswa upa, S. C. P ashan ha, C. Shi akuma a, Dani hku- ma , Spec ochim. Ac a Pa A 2015, 135 , 241 –251. [10] a) S. K. Ghosh, A. P akash, S. Da a, S. K. Roy, D. Basu, Bull. Ma e . Sci. 2010, 33 , 7 –16; b) M. A. Aghayan, M. A. Rod iguez, Ma e . Sci. Eng. C 2012, 32 , 2464 –2468. [11] a) S. Specchia, E. Finocchio, G. Busca, P. Palmisano, V. Specchia, J. Ca al. 2009, 263 , 134 –145 ; b) V. M. Shinde, G. Mad as, Ca al. Today 2012, 198 , 270 –279 ; c) R. Mis i, M. Rahaman, J. Llo ca, K. R. P iolka , S. Colussi, B. C. Ray, A. Gayen, J. Mol. Ca al. A 2014, 390 , 187 –197; d) M. Piume i, D. Fino, N. Russo, Appl. Ca al. B 2015, 163 , 277 –287. [12] S. Colussi, A. Gayen, M. F. Camellone, M. Boa o, J. Llo ca, S. Fab is, A. T o a elli, Angew. Chem. In . Ed. 2009, 48 , 8481 –8484 ; Angew. Chem. 2009, 121 , 8633 –8636. [13] a) A. D. Maye nick, M. J. Janik, J. Phys. Chem. C 2008, 112 , 14955 –14964; b) A. D. Maye nick, M. J. Janik, J. Ca al. 2011, 278 , 16– 25. [14] S. Colussi, A. Gayen, J. Llo ca, C. de Lei enbu g, G. Dolce i, A. T o a elli, Ind. Eng. Chem. Res. 2012, 51 , 7510 –7517. [15] a) H. Nasi i, E. B. Mo lagh, J. V. Khaki, S. M. Zeba jad, Ma e . Res. Bull. 2012, 47 , 3676 –3680 ; b) A. B. Salunkhe, V. M. Kho , M. R. Phada a e, S. H. Pawa , J. Alloys Compd. 2012, 514 , 91– 96 ; c) S. Haja pou , K. Ghei- sa i, A. H. Raou , J. Magn. Magn. Ma e . 2013, 329 , 165 –169 ; d) R. Ianos¸, R. Babu a, R. Lazau, Ce am. In . 2014, 40 , 12207– 12211. [16] K. R. P iolka , P. Be a, P. R. Sa ode, M. S. Hegde, S. Emu a, R. Kumashi o, N. P. Lalla, Chem. Ma e . 2002, 14 , 2120 –2128. [17] S. Colussi, A. T o a elli, E. Vesselli, A. Ba aldi, G. Comelli, G. G oppi, J. Llo ca, Appl. Ca al. A 2010, 390 , 1 –10. [18] a) Y. R. Wu, W. Y. Hu, S. C. Han, J. Alloys Compd. 2006, 420 , 83– 93; b) V. L. Vinog ad, M. H. F. Slui e , B. Winkle , Phys. Re . B 2009, 79 , 104201. [19] a) G. Chen, W.-T. Chou, C. T. Yeh, Appl. Ca al. 1983, 8 , 389 –397; b) T. C. Chang, J. J. Chen, C. T. Yeh, J. Ca al. 1985, 96 , 51– 57; c) H. Lieske, J. Vol e , J. Phys. Chem. 1985, 89 , 1841 –1842. [20] A. T o a elli, Ca al. Re . 1996, 38 , 439 –520. [21] M. F. Luo, X. M. Zheng, Appl. Ca al. A 1999, 189 , 15– 21. [22] S. Be nal, J. J. Cal ino, M. A. Cauqui, J. M. Ga ica, C. La ese, J. A. P. Omil, J. M. Pin ado, Ca al. Today 1999, 50 , 175 –206. [23] L. Kepin´ ski, M. Wolcy z, J. Okal, J. Chem. Soc. Fa aday T ans. 1995, 91 , 507 –515. [24] W. C. Conne , J. L. Falcone , Chem. Re . 1995, 95 , 759 –788. [25] X. Y. Chen, J. W. Schwank, G. B. Fishe , Y. S. Cheng, M. Jagne , R. W. McCabe, M. B. Ka z, G. W. G aham, X. Q. Pan, Appl. Ca al. A 2014, 475 , 420 –426. [26] O. Demoulin, G. Rupp ech e , I. Seunie , B. Le Cle , M. Na ez, P. Ruiz, J. Phys. Chem. B 2005, 109 , 20454– 20462. [27] a) R. J. Fa au o, M. C. Hobson, T. Kennelly, E. M. Wa e man, Appl. Ca al. A 1992, 81 , 227 –237; b) J. G. McCa y, Ca al. Today 1995, 26 , 283 –293. [28] M. Ca gnello, J. J. D. Jaen, J. C. H. Ga ido, K. Bakhmu sky, T. Mon ini, J. J. C. Gamez, R. J. Go e, P. Fo nasie o, Science 2012, 337 , 713 –717.