Design and testing of a microchannel reactor for the PROX reaction

Design and es ing o a mic ochannel eac o o he PROX eac ion S. C uz1, O. Sanz1, R. Poya o1, O.H. Laguna1*, F. J. Echa e2, L.C. Almeida2, M.A. Cen eno1, G. A zamendi3, L.M. Gandia3, E.F. Souza-Aguia 4, M. Mon es2, J.A. Od iozola1 1 Ins . o Ma e ial Sciences o Se ille, CSIC – Uni e sidad de Se illa. A da. Amé ico Vespuccio 49, 41092 Se ille, Spain 2 Dep . Appl. Chemis y, Uni e si iy o he Basque Coun y, Paseo Manuel La dizábal 3, 20018, San Sebas ián, Spain 3Public Uni e si y o Na a e, Campus de A osadía, s/n, E-31006 Pamplona, Spain 4Pe ob as, Rio de Janei o, B azil *Co esponding au ho : Osca He nando Laguna Espi ia e-mail add ess: [email p o ec ed] Telephone: +34 954489501 Ex . 9221 Fax: +34 954460665 Abs ac The di e en s eps o manu ac u ing a mic ochannel eac o o he PROX eac ion a e discussed. T ansien Liquid Phase bonding (TLP) using a Ni-B-Si amo phous mel spun is used o joining mic omilled Al-alloyed e i ic s ainless s eel pla es ollowed by ec ys alliza ion a 1200ºC o 5 hou s. A CuOx-CeO2 ca alys syn hesized by he cop ecipi a ion me hod was washcoa ed on he mic ochannel block esul ing in a homogenous 20-30 µm hick laye . The ca aly ic ac i i y o CO-PROX eac ion is simila in bo h he powde ca alys and he mic ochannel coa ed eac o bu he selec i i y is highe in he mic ochannel eac o . Keywo ds 1) Mic ochannel eac o 2) CO-PROX 3) CuOx-CeO2 4) T ansien Liquid Phase bonding 5) Al-alloyed e i ic s ainless s eel 1. In oduc ion The widesp ead use o po able elec ic and elec onic de ices inc eases he need o e icien au onomous powe supplies ha eplace he cu en ly p edominan ba e y echnology. The use o common uels/chemicals (hyd oca bons o alcohols) as ene gy sou ces is a p omising al e na i e when combined wi h he ecen de elopmen s in mic ochannel eac o echnology. The high powe densi y, apid s a -up ime, and low- empe a u e ope a ion o PEMFC make hese de ices as he mos p omising o powe ing up elec ic o elec onic de ices. Howe e , an essen ial equi emen o he e o ma e- ed PEMFC is he emo al o CO om he H2 s eam a e he e o ming and wa e – gas shi (WGS) eac ions. A e he WGS uni s he CO con en o he hyd ogen s eams may a y be ween ca. 0.5 and 2 ol.% being, he e o e, manda o y clean-up uni s o he emo al o CO o ace le el. P e e en ial CO oxida ion (CO-PROX) is among he p e e ed echnologies o small-scale uel p ocesso applica ions [1]. CO-PROX eac ions allow educing CO con en s om 1 % in he eed o he ppm le el. When using PEMFCs o po able o au omo i e applica ions packed-bed eac o s ha e se e al d awbacks such as p essu e d op wi hin ca alys laye , empe a u e g adien s, and ho spo due o he high exo he mici y o he oxida ions o CO and H2. Mic o eac o s ha e he ad an ages o as esponse ime, easy in eg a ion, and small oo p in , which a e ideal o po able powe sys ems. In addi ion, enhanced mass and hea anspo p ope ies a e also widely ecognized as ad an ages o mic o eac o s [2-6]. In addi ion o his, ecen s udies o simula ed CO-PROX eac ions ha e shown ha he e e se wa e -gas-shi side eac ion is a ou ed in he case o packed bed eac o s wi h espec o hin ca aly ic ilms deposi ed on mic ochannel walls [4]. The e o e, CO-PROX uni s based on mic ochannel eac o s ha e been epo ed as pa o in eg a ed uel p ocesso s in a wide powe ange [6-9]. Di e en ca aly ic sys ems and eac o designs ha e been p oposed o he CO- PROX eac ion. In a ecen s udy Bion e al. [10] e iew he pe o mances o noble me als (including P , Ru, Rh, Pd and Au) and ansi ion me al oxides ca alys s and compa e he ad an ages and d awbacks o each ype o ca alys s in e ms o ac i i y and selec i i y including he in luence o he p esence o CO2 and H2O in he eac an s low. They conclude ha CuO–CeO2 ca alys s a e e y a ac i e o indus ial applica ions since hei excellen pe o mances, s abili y and low cos compa ed o noble me al-based ca alys s. Dud ield e al. [11, 12] ope a ed a compac in hea -exchange eac o con aining 2.5% P -Ru ca alys in he CO-PROX eac ion. Using a wo-s age con igu a ion his uni allowed educ ions in he CO concen a ions down o 15 ppm. Silicon wa e s [13] o aus eni ic s ainless s eel pla es [14, 15] we e used o building mic ochannel eac o s o he PROX eac ion. Hwang e al. [16] s udied he eac ion o e a silicon-based mic o eac o coa ed wi h a P /Al2O3 ca alys , yielding 99.4% CO con e sion. Kim e al. [15, 17] ob ained CO ou le concen a ion below 50 ppm a O2/CO a io o 2.5 by using a 13-laye ed mic o- channel eac o buil by s acking chemically e ched s ainless s eel pla es coa ed wi h a P –Co/Al2O3 ca alys . Despi e mos o he wo k on CO-PROX in mic o eac o s has been ca ied ou using noble me al based ca alys s a ew insigh s on coppe based ca alys s ha e been epo ed. Sny niko e al. [18] compa ed a 5 w .% Cu/CeO2 ca alys in bo h ixed bed and mic ochannel eac o s wi h he la e exhibi ing highe con e sion and selec i i y. This ca alys allowed he educ ion o he CO concen a ion om 1 ol.% o 10 ppm. Kim e al. [15] compa ed a CuO/CeO2 coa ed mic o eac o wi h a P -Co coa ed mic o eac o showing be e selec i i ies in he CO-PROX eac ion o he coppe -based ca alys . In he p esen wo k, we p o ide a desc ip ion o he di e en s eps equi ed o manu ac u ing a mic ochannel eac o o he CO-PROX eac ion. The eac o is es ed using con en ional CuOx/CeO2 ca alys s and he esul s compa ed wi h hose ob ained o powde ed ca alys s. 2. Ma e ials and me hods 2.1. Ca alys p epa a ion The CuOx-CeO2 ca alys was p epa ed by he cop ecipi a ion me hod. The amoun necessa y o p epa ing a 0,5 M solu ion o Cu(NO3)2·3H2O and Ce(NO3)3·6H2O we e mixed, unde igo ous s i ing, o ge a 9:1 Ce(OH)3:Cu(OH)2 weigh a io. A e homogenizing he sys em, a NaOH solu ion (2 M) was added d opwise un il a s able pH o 9. The p ecipi a e was il e ed and washed wi h dis illed wa e in o de o emo e he Na+ and NO3- ions. Finally, he ob ained solid was d ied o e nigh a 60°C and inally, calcined 2h a 300 ºC. 2.2. Ca alys s cha ac e iza ion BET speci ic su ace a eas we e measu ed by ni ogen adso p ion a liquid ni ogen empe a u e in a Mic ome i ics ASAP 2000 appa a us. Be o e analysis, he samples we e degassed 2 h a 150ºC in acuum. The ce ium and coppe con en s o he samples we e de e mined by X- ay luo escence spec ome y (XRF) in a Panaly ical AXIOS PW4400 sequen ial spec opho ome e wi h a hodium ube as he sou ce o adia ion. X- ay di ac ion (XRD) analysis was pe o med on a Siemens D 500 di ac ome e . Di ac ion pa e ns we e eco ded wi h Cu Ka adia ion (40 mA, 40 kV) o e a 10-80º 2θ ange using a posi ion-sensi i e de ec o wi h a s ep size o 0.01º and a s ep ime o 7 s. The Raman spec a we e eco ded in a dispe si e Ho i a Jobin Y on LabRam HR800 mic oscope, wi h a 20 mWHe-Ne g een lase (532,14 nm), wi hou il e , and wi h a 600 g·mm-1 g a ing. The mic oscope used a 50x objec i e and a con ocal pinhole o 100 μm. The Raman spec ome e is calib a ed using a silicon wa e . The educibili y s udies we e ca ied ou on a pilo plan buil by PID Eng&Tech in TPR mode equipped wi h a VINCI he mal conduc i i y de ec o . The expe imen s we e done by a he mo-p og ammed educ ion (TPR) o 50 mg o ca alys , in a H2 low o 5 % dilu ed in A ( o al low = 50 mL/min), using a empe a u e amp o 10 °C/min om oom empe a u e o 900 °C. The Ze a Po en ial was measu ed by using a MALVERN Ze asize 2000 ins umen . The solids we e dispe sed in an aqueous solu ion o 0.003 M NaCl. The pHs o he solu ions we e adjus ed wi h HNO3 o NaOH solu ions. Rheological p ope ies o he slu ies we e measu ed in a o a ional iscosime e HAAKE, model VT 500, geome y NV. The adhe ence o he ca aly ic laye deposi ed on o he subs a es was e alua ed using an ul asonic echnique. The weigh loss caused by he exposu e o he sample o ul asound is measu ed. The s uc u ed suppo s imme sed in pe oleum e he we e submi ed o an ul asonic ea men o 30 min a oom empe a u e. A e ha , he samples we e d ied and calcined. The weigh loss was de e mined by he di e ence in he mass o he samples be o e and a e he ul asonic es . The esul s a e p esen ed in e ms o he e ained amoun o coa ing on he monoli h, exp essed as pe cen age. 2.3. Ca aly ic ac i i y measu emen s The CO-PROX eac ion was ca ied ou a a mosphe ic p essu e in a PID Eng&Tech Mic oac i i y se -up, employing a s ainless s eel ubula eac o wi h in e nal diame e o 9 mm and a cons an eed s eam low a e o 100 cm3·min-1 (STP). The ca alys (100 mg, pa icle size in he 100–200 mm ange) was dilu ed wi h c ushed glass pa icles in he same pa icle size ange o ming a bed o abou 5 mm in leng h. The expe imen al uns we e ca ied ou in a low con aining 50 ol.% H2, 2 ol. % CO and 1 ol. % O2 concen a ions using N2 as balance. The eac ion empe a u e was inc eased om 50–60 o 190–250 ºC in s eps o 10 ºC. Fo each s ep he empe a u e was s abilized and da a we e eco ded a s eady-s a e condi ions. F esh ca alys was loaded in o he eac o a e each comple e un. Some expe imen s ca ied ou a he same space- ime bu a a ying gas linea eloci ies con i med he absence o ex e nal mass ans e e ec s. On-line analyses o he eed and p oduc s s eams we e pe o med on an Agilen 7890 gas ch oma og aph equipped wi h a Po apak Q, wo Molecula Sie e 5A and wo Hayesep Q columns and wo TCD de ec o s and a FID de ec o . The same compu e ized PID Eng&Tech Mic oac i i y se -up was used o s udy he eac ion in he mic ochannel eac o , eplacing he ubula s ainless s eel eac o o he buil mic ochannel eac o he same composi ions we e chosen bu 300 cm3·min-1 (STP) eed s eam low a e since 300 mg ca alys was loaded in he mic ochannel block. To s udy he e ec o CO2 and H2O, a se ies o expe imen s keeping cons an he H2 con en o he eed s eam se a 50 ol.%, and bo h he CO and O2 concen a ions ixed a 1 ol.% we e ca ied ou . The CO2 and H2O concen a ions in he eed we e a ied wi hin he 2–10 ol.% and 0–20 ol.% anges, espec i ely, using N2 as balance. The mic oac i i y e e ence ho box con olled he eac ion empe a u e. Fo each s ep he empe a u e was s abilized and da a we e eco ded a s eady-s a e condi ions. 3. Mic ochannel eac o The me allic mic ochannel eac o was manu ac u ed using Al-alloyed e i ic s ainless s eel ( o ins ance, Fec alloy®), since e i ic alloys con aining 3–5% o aluminium p oduce by he mal ea men an Al2O3 laye ha a ou s he in e ac ion wi h he ca aly ic coa ing [19-22]. The Fec alloy consis ed o C 22%, Al 4.8%, Si 0.3%, Y 0.3%, and Fe balance [21]. Cha ac e iza ion o he joined s eel pla es we e ca ied ou by op ical and elec on mic oscopy. Specimens we e ex ac ed om he samples, g inded wi h SiC pape o #240, #400, #600 and #1200 g ain size and mi o polished wi h 0,3 and 0,1 µm Al2O3 powde . E ching wi h Vilella’s eagen de eloped he mic os uc u e. Sample obse a ion in he as polished an e ched condi ions was done in a Leica-DM-IRM op ical mic oscope equipped wi h a digi al came a (Leica DC300). In o de o iden i y me allic and non-me allic inclusions as well as o de e mine local composi ions, he as-polished samples we e also obse ed by SEM using a JEOL 5400 sys em equipped wi h seconda y and backsca e ed elec on and x- ay de ec o s. Mic ochannels we e ab ica ed by mic o-milling 1 mm hick e i ic s ainless s eel pla es. Each pla e has 10 squa e channels o 750 µm sepa a ed be ween hem 300 µm machined in a 20x20 mm2 pla e. This p ocess esul s in channels wi h 56 mm2 geome ic su ace and 700 µm hyd aulic adius. In o de o minimize he high p essu es and p ocessing imes equi ed o solid s a e bonding he e i ic s ainless s eel pla es we e joined oge he using he ansien liquid phase (TLP) bonding p ocess. The TLP bonding p ocess uses in e laye s, which ei he con ain mel ing poin dep essan s (e.g. B, Si o P) o o m an eu ec ic wi h he pa en me al being bonded. The join is held a he bonding empe a u e un il he mel ing poin dep essan s a e los om he liquid in e laye by di usion and he liquid in e laye solidi ies iso he mally due o he change in composi ion o he bond. This echnique has been p e iously used o join di e en alloy sys ems including duplex s ainless s eels [23-29]. Figu e 1 is a schema ic pic u e o he s eel pla es-me allic glass assembly used in his wo k. Fo TLP bonding, a nickel-based in e laye wi h a composi ion o Ni-14B-7Si (w %, Good ellow) was used. This in e laye was an amo phous mel spun oil wi h a hickness o 25 µm. The bonding p ocess was pe o med in acuum using a es machine de eloped by Mic o es ha allows he con ol o he applied o ce, empe a u e and ime allowing apid hea ing up o he bonding empe a u e. Va ious ials we e conduc ed o es ablish he op imum bonding pa ame e s, a ying he empe a u e, applied o ce and ime. A bonding empe a u e o 850°C wi h an applied o ce o 2,8 kN (~7 MPa) was selec ed. Low empe a u e and ligh p essu e applied is equi ed o keeping he mechanical in eg i y o he machined pla e since he low c eep s eng h o he selec ed alloy [30]. Figu e 2 shows he mac oscopic aspec s o joined pla es a di e en applied o ces clea ly shown ha o applied p essu es abo e 15 MPa (6,0 kN) and 900ºC c eep esul s in a s ong de o ma ion o he machined mic ochannels. The TLP bonded samples we e inally ea ed a 1200ºC o 5 h. A e he joining p ocedu e he wo e i ic s eel pla es a e sepa a ed by a me allic alloy ca. 25 µm hick whose composi ion mainly co esponds o ha o he Ni-based in e laye . The mic os uc u e o he joined a ea is cha ac e ized by he exis ence o wo e i ic laye s in which he p ecipi a ion o ch omium ca bides is e iden sepa a ed by he Ni-based in e laye . A s ong ch omium ca bide p ecipi a ion occu s a he s ainless s eel-in e laye in e aces, igu e 3A. Upon pos -p ocessing a 1200ºC a single phase ec ys allizes. EDX analysis ac oss he joined a ea show ha he Fe, C , Al and Al line p o iles a e almos la indica ing in e di usion o he Ni-based in e laye and he e i ic s eel, excep o some ch omium maximum and i on minimum co esponding o he p esence o ch omium ca bide p ecipi a es. Using nickel as a ma ke , i should be no ed ha his elemen is absen in he e i ic s eel his di usion is clea ly seen. Figu e 3D shows he Gaussian- i ed Ni line p o ile ac oss he joined a ea, he FWHM o he Gaussian dis ibu ion, ca. 400 µm, is conside able highe han he hickness o he used in e laye demons a ing he o ma ion o a single phase upon he designed joining p ocedu e. Mic omachining and joining o he s eel pla es esul ed in a mic oblock ha was housed in Al-alloyed e i ic s eels cases designed using CFD algo i hms o ensu ing homogeneous low h ough all he channels, he housing was join oge he using g aphi e seals, igu e 4. Finally, he mic ochannel block was i ed wi h he mocouples moni o ing inle and ou le empe a u es as well as empe a u e g adien s wi hin he mic oblock. Mani olding and ins umen a ion o he mic o eac o was implemen ed in a compu e ized Mic oac i i y Re e ence Ca aly ic Reac o om PID Eng&Tech ha con ols he eac ion empe a u e in he eac o h ough he empe a u e con ol o he ho box. P io o he assembly o he mic o eac o he mic ochannel block was coa ed wi h he ca alys . Fo imp o ing he in e ac ion be ween he washcoa ed laye and he me allic suppo , he su ace o he mic ochannel block was modi ied gene a ing an oxide scale ha enhances adhesion o he ca aly ic laye bo h mechanically, h ough he gene a ed oughness, and chemically h ough he in e ac ion be ween his oxide scale and he ca aly ic ma e ial. Modi ied su aces a e ob ained upon hea ing in ai a ele a ed empe a u es being he esul ing oxide scale an excellen subs a e o adhe e ca alys s [19- 22]. The op imal ea men pa ame e s emain usually undisclosed. Upon hea ing a 900 ºC o 22 h in ai he mic ochannel block a homogeneous su ace laye o α-Al2O3 is o med as con i med by DRX. SEM mic og aphs, igu e 5, clea ly show he needle-like s uc u e o he o med whiske s ha ing hicknesses anging be ween 3 and 6 µm. Washcoa ing was selec ed o coa ing he mic ochannels wi h he CuOx/CeO2 ca alys . The i s s ep o washcoa ing a me allic subs a e is o p epa e s able slu ies o he ca alys o be deposi ed. Pa icle size, solid con en o he suspension and pH o he ca alys slu y a e pa ame e s ha in luence he slu y s abili y. The pa icle size o ou ca alys , d90=0.5μm, is well below 10µm he uppe ecommend limi o p epa ing s able slu ies [31] he e o e he usual ball milling p ocess o educing pa icle size was a oided. The isoelec ic poin (IEP) o he ca alys is ca. 7 and he e o e he pH was ixed a 4 o ensu ing high alues o ze a po en ial and hen high epulsions be ween he pa icles, a ou ing he s abili y o he slu ies [32]. The use o addi i es o he slu y o mula ion a emp s o imp o e he ca alys adhe ence and he washcoa ing d ying p ocess. The addi ion o colloidal alumina, ha p esen s a na ow pa icle size dis ibu ion, imp o es he ca alys adhe ence [31, 33] acco ding o he model p e iously p oposed by Nijhuis e al. [34] a bimodal pa icle size dis ibu ion inc ease adhe ence since he smalle pa icles a e loca ed be ween he bigge ones. The use o poly inyl alcohol (PVOH) helps in p e en ing c ack o ma ion du ing he d ying p ocess imp o ing he we ing p ope ies o he ca aly ic laye . A e se e al ials o slu y o mula ion o washcoa ing he ollowing p opo ions o ca alys and addi i es was selec ed: 76% ca alys con en , 7% PVOH (w/w), 17% colloidal alumina (w/w) and pH o he suspension adjus ed o 4 wi h dilu ed HNO3. Assuming he Eins ein model o he diluen s dispe sion o ha d sphe es, he iscosi y o ideal wa e slu ies only depends on he solid con en , being highe as he solid con en inc eases [35]. In he case o he CuOx/CeO2 ca alys he iscosi y anges om 20 cps o 15% solid con en o 200 cps o a dispe sion con aining 25% solids. The iscosi y o he ca aly ic slu y is he key pa ame e in con olling he coa ing p ocess; a de ailed desc ip ion o he di e en ac o s in luencing he washcoa ing p ocess is gi en elsewhe e [31, 33]. Slu y wi h 18% solid concen a ion p o ides an excellen comp omise be ween he slu y we ing p ope ies and ca alys loading. Figu e 6 p esen s he amoun o loaded ca alys o e he subs a es as a unc ion o he numbe o coa ings. The speci ic load inc eases almos linea ly wi h he numbe o coa ings. The washcoa ing me hod gi es addi i e and homogenous esul s. Low iscosi ies allow o ob ain highly adhe en and homogeneous coa ings bu wi h low speci ic loads. Thus o ob aining he a ge loading nume ous coa ing a e equi ed. On he con a y, high iscosi y will allow high speci ic load pe coa ing al hough wi h lowe homogenei ies (i.e. accumula ions, channel blocking) esul ing in less adhe en coa ings [33]. The slu y p epa ed in hese condi ions was s able o 96 hou s. Once he mic ochannels we e imme sed in he slu y he elimina ion o he excess was done by ai blowing o mic ochannels (2L·min-1). Finally, he mic ochannel block was d ied a 120ºC o 30 minu es be ween coa ings and a e he las coa ing p ocedu e he mic ochannel block was calcined a 300ºC o 3 hou s (1ºC/min). This p ocedu e esul ed in a mic ochannel block loading o 5,46 mg·cm-2 ca alys a e eigh washcoa ing p ocesses. The adhe ence o [23]T.I.Khan,E.R.Wallach,T ansien liquidphasedi usionbondingandassocia ed ec ys alliza ionphenomenonwhenjoiningODS e i icsupe alloys,Jou nalo Ma e ials Science,31(1996)2937‐2943. [24]J.R.Askew,J.F.Wilde,T.I.Khan,T ansien liquidphasebondingo 2124aluminiumme al ma ixcomposi e,Ma e ialsScienceandTechnology,14(1998)920‐924. [25]A.Ek ami,T.I.Khan,T ansien liquidphasedi usionbondingo oxidedispe sion s eng henednickelalloyMA758,Ma e ialsScienceandTechnology,15(1999)946‐950. [26]Y.Guan,W.F.Gale,T ansien liquidphasebondingo aH bea ingsinglec ys alnickel aluminide oMM247supe alloy,Ma e ialsScienceandTechnology,15(1999)207‐212. [27]T.I.Khan,N.O han,M.E oglu,T ansien liquidphasebondingo amic oduplexs ainless s eelusingamo phousin e laye s,Ma e ialsScienceandTechnology,18(2002)396‐400. [28]T.I.Khan,M.J.Kabi ,R.Bulpe ,E ec o  ansien liquid‐phasebonding a iableson he p ope ieso amic o‐duplexs ainlesss eel,Ma e ialsScienceandEnginee ingA,372(2004) 290‐295. [29]R.K.Saha,S.Wei,T.I.Khan,Acompa isono mic os uc u alde elopmen sinTLP di usionbondsmadeusingODSNialloy,Ma e ialsScienceandEnginee ingA,406(2005)319‐ 327. [30]S.R.J.Saunde s,H.E.E ans,M.Li,D.D.Gohil,S.Osge by,Oxida iong ow hs essesinan alumina‐ o ming e i ics eelmeasu edbyc eepde lec ion,Oxida iono Me als,48(1997) 189‐200. [31]L.C.Almeida,F.J.Echa e,M.A.Sanz,M.A.Cen eno,G.A zamendi,L.M.Gandía,E.F.Souza‐ Aguia ,J.A.Od iozola,M.Mon es,Fische ‐T opschSyn hesisinMic ochannels,11 h In e na ionalCon e enceonMic o eac ionTechnology,11 hIn e na ionalCon e enceon Mic o eac ionTechnology,Ma ch8‐10,2010,Kyo oResea chPa k,Kyo o,Japan.Submi ed o ChemicalEnginee ingJou nal. [32]S.Valla ,D.Houi e ,J.ElFallah,D.Ke adec,J.M.Haussonne,Oxideslu iess abili yand powde sdispe sion:Op imiza ionwi hze apo en ialand heologicalmeasu emen s,Jou nal o  heEu opeanCe amicSocie y,19(1999)1017‐1021. [33]L.C.Almeida,F.J.Echa e,S.O,M.A.Cen eno,J.A.Od iozola,M.Mon es,Washcoa ingo  me allicmonoli hsandmic ochannel eac o s,S udiesinSu aceScienceandCa alysis, Accep ed,(2010). [34]T.A.Nijhuis,A.E.W.Bee s,T.Ve guns ,I.Hoek,F.Kap eijn,J.A.Moulijn,P epa a iono  monoli hicca alys s,Ca alysisRe iews‐ScienceandEnginee ing,43(2001)345‐380. [35]T.C.Pa on,Pain FlowandPigmen Dispe sion,2nded.JohnWiley&Sons,NewYo k, (1979). [36]W.Y.He nández,M.A.Cen eno,F.Rome o‐Sa ia,J.A.Od iozola,Syn hesisand cha ac e iza iono Ce1‐xEuxO2‐x/2mixedo xidesand hei ca aly icac i i ies o COoxida ion, Jou nalo PhysicalChemis yC,113(2009)5629‐5635. [37]O.H.Laguna,M.A.Cen eno,G.A zamendi,L.M.Gandía,F.Rome o‐Sa ia,J.A.Od iozola, I on‐modi iedce iaandAu/ce iaca alys s o To alandP e e en ialOxida iono CO(TOXand PROX),Ca alysisToday,InP ess,Co ec edP oo . [38]Z.Liu,R.Zhou,X.Zheng,Compa a i es udyo di e en me hodso p epa ingCuO‐CeO2 ca alys s o p e e en ialoxida iono COinexcesshyd ogen,Jou nalo Molecula Ca alysisA: Chemical,267(2007)137‐142. [39]W.Y.He nández,M.A.Cen eno,F.Rome o‐Sa ia,J.A.Od iozola,Syn hesisand cha ac e iza iono Ce1‐xEuxO2‐x/2mixedoxidesand hei ca aly icac i i ies o COoxida ion, Jou nalo PhysicalChemis yC,113(2009)5629‐5635. [40]S.Damyano a,B.Pawelec,K.A ish i o a,M.V.M.Hue a,J.L.G.Fie o,S udyo  he su aceand edoxp ope ieso ce ia‐zi coniaoxides,AppliedCa alysisA:Gene al,337(2008) 86‐96. [41]T.Masui,Y.Peng,K.‐i.Machida,G.‐y.Adachi,Reduc ionBeha io o CeO2‐Z O2Solid Solu ionP epa ed omCe iumZi conylOxala e,Chemis yo Ma e ials,10(1998)4005‐4009. [42]G.A gou opoulos,M.Manzoli,F.Boccuzzi,T.Tabako a,J.Papa asiliou,T.Ioannides,V. Idakie ,Ca aly icpe o manceandcha ac e iza iono Au/doped‐ce iaca alys s o  he p e e en ialCOoxida ion eac ion,Jou nalo Ca alysis,256(2008)237‐247. [43]H.Bao,X.Chen,J.Fang,Z.Jiang,W.Huang,S uc u e‐ac i i yRela iono Fe2O3–CeO2 Composi eCa alys sinCOOxida ion,Ca alysisLe e s,125(2008)160‐167. [44]P.J.Gellings,H.J.M.Bouwmees e ,Ionandmixedconduc ingoxidesasca alys s,Ca alysis Today,12(1992)1‐101. [45]F.Ma iño,G.Ba one i,M.Labo de,N.Bion,A.LeValan ,F.Ep on,D.Dup ez,Op imized CuO‐CeO2ca alys s o COPROX eac ion,In e na ionalJou nalo Hyd ogenEne gy,33(2008) 1345‐1353. [46]G.J.Milla ,C.H.Roches e ,S.Bailey,K.C.Waugh,Acombined empe a u e‐p og ammed eac ionspec oscopyandFou ie ‐ ans o min a edspec oscopys udyo CO2‐H2andCO‐ CO2‐H2in e ac ionswi hmodelZnO/SiO2,Cu/SiO2andCu/ZnO/SiO2me hanol‐syn hesis ca alys s,Jou nalo  heChemicalSocie y,Fa adayT ansac ions,88(1992)2085‐2093. [47]G.A gou opoulos,T.Ioannides,H.K.Ma alis,J.Ba is a,S.Hoce a ,CuO‐CeO2mixed oxideca alys s o  heselec i eoxida iono ca bonmonoxideinexcesshyd ogen,Ca alysis Le e s,73(2001)33‐40. [48]G.A gou opoulos,T.Ioannides,C.Papadopoulou,J.Ba is a,S.Hoce a ,H.K.Ma alis,A compa a i es udyo P /‐Al2O3,Au/‐Fe2O3andCuO‐CeO2ca alys s o  heselec i e oxida iono ca bonmonoxideinexcesshyd ogen,Ca alysisToday,75(2002)157‐167. [49]G.A zamendi,I.U iz,P.M.Diéguez,O.H.Laguna,W.Y.He nández,A.Ál a ez,M.A. Cen eno,J.A.Od iozola,M.Mon es,L.M.Gandía,Selec i eCO emo alo e Au/CeFeand CeCuca alys sinmic o eac o ss udied h oughkine icanalysisandCFDsimula ions,11 h In e na ionalCon e enceonMic o eac ionTechnology,Ma ch8‐10,2010,Kyo oResea ch Pa k,Kyo o,Japan.Submi ed oChemicalEnginee ingJou nal. [50]G.G.Je nigan,G.A.Somo jai,Ca bonMonoxideOxida iono e Th eeDi e en Oxida ion S a eso Coppe :Me allicCoppe ,Coppe (I)Oxide,andCoppe (II)Oxide‐ASu aceScience andKine icS udy,Jou nalo Ca alysis,147(1994)567‐577. [51]D.Gama a,G.Munue a,A.B.Hung ía,M.Fe nández‐Ga cía,J.C.Conesa,P.A.Midgley, X.Q.Wang,J.C.Hanson,J.A.Rod íguez,A.Ma ínez‐A ias,S uc u e‐ac i i y ela ionshipin nanos uc u edcoppe ‐ce ia‐basedp e e en ialCOoxida ionca alys s,Jou nalo Physical Chemis yC,111(2007)11026‐11038. [52]T.Capu o,L.Lisi,R.Pi one,G.Russo,Kine icso  hep e e en ialoxida iono COo e  CuO/CeO2ca alys sinH2‐ ichgases,Indus ialandEnginee ingChemis yResea ch,46(2007) 6793‐6800. [53]W.Liu,M.Fly zanis ephanopoulos,To alOxida iono Ca bonMonoxideandMe hane o e T ansi ionMe alFluo i eOxideComposi eCa alys s:I.Ca alys Composi ionandAc i i y, Jou nalo Ca alysis,153(1995)304‐316.  TABLE AND FIGURE CAPTIONS Table 1. Tex u al p ope ies o he ca alys s Table 2. Hyd ogen consump ion du ing he TPR analysis o he CuOx-CeO2 solid Table 3. Pa ame e s chosen o welding he mic ochannel pla es Figu e 1: Schema ic pic u e o he s eel pla es-me allic glass assembly in c oss- low con igu a ion. Figu e 2. Joined Al-alloyed e i ic s ainless s eel pla es a 900ºC as a unc ion o he applied uniaxial p essu e. A) 10 MPa and B) 15 MPa Figu e 3. Mic os uc u es o he joined a ea. A) SEM mic ig aph be o e he ec ys alliza ion p ocess; B) Op ical mic og aph a e ec ys allizing he mic o eac o o 5h a 1200ºC; C) EDX line p o ile ac oss he joined a e a e ec ys alliza ion; D) Gaussian i o he Ni EDX line p o ile ac oss he joined a e a e ec ys alliza ion. The p o ile is aken along he line ma ked in pannel B. Figu e 4. Mic ochannel block and mic ochannel eac o o he CO-PROX eac ion. Figu e 5. SEM mic og aphs o he α-Al2O3 o med upon hea ing a 900 ºC o 22 h in ai he mic ochannel block Figu e 6. E olu ion o he ca alys loading a e he d ying p ocess as a unc ion o he numbe o coa ings pe o med. The amoun loaded upon calcina ion is indica ed by an a ow and he speci ic loading is indica ed a e 4, 6 o 8 coa ings Figu e 7. XRD pa e n o he CuOx-CeO2 solid Figu e 8. Raman spec a o he CuOx-CeO2 solid a di e en empe a u es Figu e 9. TPR p o ile o he p epa ed CuOx-CeO2 solid Figu e 10. CO con e sion (A) and O2 selec i i y o CO2 (B) o he CO-PROX eac ion o e CuOx-CeO2 ca alys s. Open ci cles syn hesized ca alys ; Full ci cles a e leaching in HNO3 a pH=2 Figu e 11. CO con e sion (A) and O2 selec i i y o CO2 (B) o he CO-PROX eac ion o e CuOx-CeO2 ca alys s in a a mosphe e wi h CO:O2:H2:N2 a ios o 2:1:50:47. F(powde ) = 100 Ncm3·min-1; F(mic ochannel) = 300 Ncm3·min-1; W(powde ) = 100 mg; W(mic ochannel) = 300 mg Figu e 12. CO con e sion o he CO-PROX eac ion o e CuOx-CeO2 ca alys s in he mic ochannel block as a unc ion o he CO2 and H2O concen a ions in a a mosphe e wi h CO:O2:H2:N2 a ios o 2:1:50:balance. F = 300 Ncm3·min-1; W(mic ochannel) = 300 mg Table 1 Ca alys BET Su ace A ea m2/g Po e Volume cm3/g A e age po e diame e Å CuOx-CeO2 75.9 0.136 74.1 CuOx-CeO2 Slu y 91.6 0.200 100.0 Table 2 Reduc on Zone H2 consump ion (mol) Chemical composi ion (mol) A 1.21x10-4 Cu Ce B 3.57x10-5 1.00x10-4 2.63x10-4 Table 3 Joining Pos -p ocessing Applied o ce (kN) 2.8 --- Tempe a u e (ºC) 850 1200 Hea ing a e (ºC·min-1) 100 Time (minu es) 40 300 A mosphe e Vacuum Ine gas Figu e 1 Figu e 2 Figu e 3 Figu e 10 Figu e 11 Figu e 12