scieee Science in your language
[en] (orig)

Promoting effect of Sn on supported Ni catalyst during steam reforming of glycerol

Abstract

The promoting effect of Sn on the catalytic performance of supported Ni catalyst in the reaction of glycerol steam reforming was studied. The physico-chemical properties of the prepared samples were investigated by X-ray fluorescence (XRF), BET surface area, in situ X-ray diffraction (XRD), laser Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), scanning electron microscopy (SEM) and temperature-programmed oxidation (TPO) techniques. The characterization results of the samples after reduction treatment (in the same conditions than the activation before catalytic activity measurements) revealed the formation of Ni-Sn alloy. The Sn-doped catalyst exhibited a high activity and it was demonstrated that the Sn addition increase the catalyst stability and durability by decreasing the coke deposition.

Read accessible full text

Promoting effect of Sn on supported Ni catalyst during steam reforming of glycerol

Author: Bobadilla Baladrón, Luis Francisco; Romero Sarria, Francisca; Centeno Gallego, Miguel Ángel; Odriozola Gordón, José Antonio
Publisher: Pergamon Press
Year: 2016
DOI: 10.1016/j.ijhydene.2016.04.119
Source: https://idus.us.es/bitstreams/81c7f617-cdf6-4787-a309-552f76328f33/download
P omo ing e ec o Sn on suppo ed Ni ca alys du ing s eam
e o ming o glyce ol
L. F. Bobadilla,†* F. Rome o-Sa ia, M. A. Cen eno and J. A. Od iozola
Depa amen o de Química Ino gánica e Ins i u o de Ciencia de Ma e iales de Se illa
(ICMSE). Cen o mix o CSIC - Uni e sidad de Se illa, 49 A . Amé ico Vespucio, 41092
Se illa (Spain)
† Cu en add ess: Ins i u o de Tecnología Química (ITQ). Cen o mix o CSIC-
Uni e sidad Poli écnica de Valencia, s/n A . de los na anjos, 46022 Valencia (Spain)
* Co esponding au ho : [email p o ec ed]
Abs ac :
The p omo ing e ec o Sn on he ca aly ic pe o mance o suppo ed Ni ca alys in he
eac ion o glyce ol s eam e o ming was s udied. The physico-chemical p ope ies o
he p epa ed samples we e in es iga ed by X- ay luo escence (XRF), BET su ace a ea,
in si u X- ay di ac ion (XRD), lase Raman spec oscopy, X- ay pho oelec on
spec oscopy (XPS), scanning elec on mic oscopy (SEM) and empe a u e
p og ammed oxida ion (TPO) echniques. The cha ac e iza ion esul s o he samples
a e educ ion ea men (in he same condi ions han he ac i a ion be o e ca aly ic
ac i i y measu emen s) e ealed he o ma ion o Ni-Sn alloy. The Sn-doped ca alys
exhibi ed a high ac i i y and i was demons a ed ha he Sn addi ion inc ease he
ca alys s abili y and du abili y by dec easing he coke deposi ion.
Keywo ds: Hyd ogen p oduc ion; Sn addi ion; Coke deposi ion
1. In oduc ion
Du ing he las decades impo an e o s a e ongoing o educe ossil uel
dependency and g eenhouse gas emissions. Among he di e en possibili ies o ind a
iable op ion, he use o hyd ogen as an ene gy ca ie is conside ed o be an in e es ing
al e na i e o he u u e and could ha e a c ucial ole in educing a mosphe ic
pollu ion. Biomass is conside ed as one o he mos a ac i e enewable sou ce o
hyd ogen p oduc ion and nume ous s udies a e being di ec ed owa d he de elopmen
o no el echnologies o p oduce hyd ogen om biomass [1-4]. S eam e o ming o
glyce ol is an impo an app oach o hyd ogen p oduc ion om biomass. Glyce ol is
he main by-p oduc gene a ed in he biodiesel p oduc ion. Biodiesel is one enewable
bio uel ob ained by ca aly ic ans-es e i ica ion o iglyce ides wi h me hanol [5].
Abou 10 w .% o glyce ol can be p oduced du ing he con e sion o ege able oils in o
biodiesel. The use o glyce ol o hyd ogen gene a ion is a e y ad an ageous op ion
since i s use would dec ease he p ice o biodiesel making i mo e compe i i e [6].
The o e all eac ion o glyce ol s eam e o ming is gi en by he ollowing equa ion:
C3H8O3 + 3H2O → 3CO2 + 7H2 (1)
which can be exp essed as a combina ion o glyce ol decomposi ion (2) and he wa e -
gas shi eac ion (3):
C3H8O3 → 3CO + 4H2 (2)
CO + H2O ↔ CO2 + H2 (3)
Theo e ically, a maximum o 7 moles o H2 pe mol o glyce ol can be p oduced,
al hough his a io depends on he eac ion condi ions such as empe a u e, p essu e and
s eam- o-glyce ol s eam a io.
Al hough he glyce ol s eam e o ming p ocess is e y a ac i e and i could be
de eloped on an indus ial-scale, i has some challenges ha mus o be o e come in
o de o accomplish i s e ec i e comme cializa ion. Fo example, he p ocess is an
endo he mic eac ion and equi es high empe a u es inc easing he ope a ion cos s.
Besides, he ca alys deac i a ion by coke deposi ion is also an issue since i a ec s
hyd ogen yield and long e m ope a ion. Wi h he aim o o e coming hese challenges,
he de elopmen o ac i e, s able and inexpensi e ca aly ic ma e ials is manda o y.
Ca alys s con aining g oup 8-10 me als such as Ni [7-12], Co [13-16], P [17-21], Ru
[22-24], Rh [25], Pd [26] o I [27] on di e en oxides ha e been la gely in es iga ed as
ac i e ca alys s o glyce ol s eam e o ming. Ni-suppo ed is one o he mos p omising
ac i e me als o such an applica ion because o i s high ac i i y, low cos and wide
a ailabili y. Howe e , Ni-based ca alys s su e deac i a ion by coke deposi ion on he
ca alys su ace ha block ac i e si es and a ou side eac ions. Deac i a ion o nickel-
based ca alys s by sin e ing o nickel c ys alli es is ano he impo an d awback.
P omo ing nickel ca alys s wi h a second me al has been p o en o be one o he mos
p omising app oach o ob ain mo e s able and op imal ca alys s [28]. Bime allic
ca alys s comp ising Sn as a Ni p omo e ha e been p o ed o ou pe o m Ni
monome allic ca alys in s eam e o ming p ocesses [29]. The coke deposi ion can be
ma kedly educed by using Sn-doped ca alys s in he s eam e o ming eac ion o
gene a e hyd ogen om hyd oca bons. Sn alloyed wi h nickel p e en s he o ma ion o
nickel a om ensembles, which a e he esponsible o he coke o ma ion, and a oids he
di usion o ca bon o o m la ge coke agglome a es [30]. The e o e, NiSn-based
ca alys s ha e he po en ial o dec ease he ca alys deac i a ion caused by coking
main aining i s high speci ic ac i i y. Sn-doped Ni ca alys s ha e been epo ed o
aqueous phase e o ming [31, 32] and me hane s eam e o ming [33] whe e he
o ma ion NixSny alloys play a key ole o inhibi coke deposi ion. Pengpanich e al.
[34] epo ed a clea example o his posi i e e ec in he pa ial oxida ion o iso-
oc ane. These au ho s ound ha he addi ion o small amoun s o in dec eased by mo e
han 50% he o ma ion o ca bon deposi s wi hou changes in he con e sion. This
enhancemen was asc ibed o he abili y o Sn o educe he g ow h o ca bon ilamen
by e a ding ca bon solubili y in he Ni pa icles [28]. Mo eo e , Saadi e al. [35] ha e
demons a ed, using densi y unc ional heo y (DFT) calcula ions, he abili y o Ni-Sn o
inhibi g aphi e o ma ion du ing s eam e o ming eac ions. They demons a ed ha he
p esence o Sn inc eases he C-C bond o ma ion ba ie .
We ha e in es iga ed p e iously he e ec o he na u e o he suppo (acidi y,
basici y and edox p ope ies) in he ca aly ic pe o mance o Sn-doped Ni ca alys s
du ing he s eam e o ming o alcohols [36-38]. The objec i e o he p esen wo k is o
in es iga e he e ec o in on he ca aly ic pe o mance o Ni-suppo ed ca alys s in
e ms o ac i i y, selec i i y and du abili y in he s eam e o ming o glyce ol.
2. Expe imen al
2.1. Ca alys s p epa a ion
The alumina suppo was ob ained by ball milling sphe ical alumina pelle s (SASOL,
1.78 mm diame e ), by using a PM4 Re sch ins umen pa icle sizes in 7 – 8 μm ange
we e achie ed. A monome allic Ni-based ca alys was p epa ed by imp egna ing he
alumina wi h an aqueous solu ion o all he ino ganic p ecu so s wi h he desi ed
concen a ions o ce ium (III) ni a e hexahyd a e (Sigma-Ald ich), magnesium ni a e
hexahyd a e (Sigma-Ald ich), and nickel (II) ni a e hexahyd a e (Sigma-Ald ich). Then,
he sample was d ied o e nigh a 100ºC and calcined a 700ºC o 12 h in lowing ai
using a hea ing a e o 10 C min-1 o ob ain 20 w .%Ni, 12 w .%CeO2, 8 w .%MgO, and
60 w .%Al2O3. The bime allic NiSn ca alys was p epa ed ollowing he same p ocedu e
bu subs i u ing a ac ion o he nickel sal by he adequa e amoun o anhyd ous in
chlo ide (Fluka) in o de o ob ain 17 w .%Ni, 3 w .%Sn, 12 w .%CeO2, 8 w .%MgO,
and 60 w .%Al2O3. The esul ing ca alys s we e labelled as Ni/CeMgAl and Ni-
Sn/CeMgAl, espec i ely.
2.2. Ma e ials cha ac e iza ion
The chemical composi ion o he samples was de e mined by X- ay luo escence
(XRF) spec oscopy in a Panaly ical AXIOS PW4400 sequen ial spec ome e wi h Rh
ube as sou ce o adia ion. The analysis we e ca ied ou on o p essed wa e s
con aining 6 w .% o wax. The ex u al p ope ies (BET su ace, po e size and po e
olume) we e s udied by ni ogen physical adso p ion-deso p ion iso he ms a 77 K
wi h Mic ome i ics ASAP 2010 equipmen . Be o e each measu emen , he samples
we e ou gassed o 2 h a 150ºC in acuum o emo e he adso bed impu i ies. X- ay
di ac ion (XRD) analysis was pe o med on a Siemens D-500 di ac ome e using Cu
Kα adia ion (40 mA, 40 kV) and a posi ion-sensi i e de ec o using a s ep size o 0.05º
and a s ep ime o 1 s. The educibili y o he ca alys s was in es iga ed by in si u XRD
analysis using a high empe a u e came a An on Paa HTK 1200 coupled wi h an X´Pe
P o Philips di ac ome e . The sys em was equipped wi h X´Cele a o de ec o wi h a
s ep o 0.05º and an equi alen ime acquisi ion o 30 s. The di ac og ams we e
eco ded in he 25 – 900ºC empe a u e ange unde a low o 100 mL min-1 (5% H2 in
A ). Su ace cha ac e iza ion was ca ied ou by X- ay pho oelec on mic oscopy (XPS)
on a Leybold-He aeus spec ome e (LHS-10/20), wo king wi h a cons an pass ene gy
o 50 eV and acuum o 5 x 10-9 To . The sys em was equipped wi h an EA-200 MCD
hemisphe ical elec on analyse wi h a dual X- ay sou ce wo king wi h Al Kα adia ion
(1486.6 eV) a 120 W and 30 mA. Binding ene gies we e e e enced o he spu ious C
(1s) signal a 284.6 eV and he a omic composi ion was es ima ed by he elemen al
sensi i i y ac o o each a om. Be o e each analysis, he sample was educed in an
ex e nal ube u nace o 1 h a 750ºC unde a o al low o 100 mL min-1 (50% H2 in
N2). Then, he sample was cooled down o oom empe a u e and placed in he
p echambe o he XPS spec ome e . The mo phology o he ca alys s was e alua ed by
SEM using a JEOL 5400 ins umen equipped wi h an Ene gy Dispe si e X- ay
Spec ome e (OXFORD LINK TETRA 1128-231).
Raman spec a we e eco ded in a dispe si e Ho i a Jobin Y on LabRam HR800
Con ocal Raman mic oscope wi h a g een lase (532.05 nm) wo king a 5 mV powe
and using a 600 g oo es/mm g a ing. The mic oscope used a 50x objec i e wi h
con ocal pinhole o 1000 μm. The Raman spec ome e was calib a ed using a silicon
wa e . The empe a u e-p og ammed oxida ion (TPO) was ca ied ou in a U-shaped
eac o unde a o al low o 50 mL min-1 (5% O2 in He) om oom empe a u e o
900ºC wi h hea ing a e o 10 ºC min-1. The o med p oduc s we e analysed by mass
spec ome y (MS) in a P ei e acuum mass spec ome e .
2.3. Ca aly ic pe o mance
The glyce ol s eam e o ming eac ion was e alua ed as a unc ion o ime o
pe iods o 4, 12 and 24 h in iso he mal condi ions a 750ºC, wi h a s eam- o-ca bon
mola a io o 4 and 100000 mL g-1 h-1 space eloci y. P io o eac ion, he samples
we e p essed, c ushed, and sie ed o achie e pa icles in he 100-200 μm ange. In each
un, 100 mg o sample was dilu ed wi h he same amoun o c ushed qua z sie ed o
he same pa icle size and placed in o he eac o be ween wo qua z wool plugs. The
ca alys s we e ac i a ed unde a o al low o 100 mL min-1 (50% H2 in N2) a 750ºC o
1 h. The expe imen s we e ca ied ou a a mosphe ic p essu e in a comme cial
Mic oac i i y Re e ence Reac o made by PiD Eng&Tech using a Has elloy C-276
ubula eac o wi h 9 mm in e nal diame e . Gas p oduc s we e analysed on-line using a
mic oGC (Va ian 4900) equipped wi h Po apak Q and Molecula Sie e – 5A columns.
A he eac o ou le a Pel ie liquid coole was i ed allowing he analysis o
condensable p oduc s by HPLC (Va ian 356-LC) equipped wi h a e ac i e index
de ec o and a Hi-Plex H column wi h milli-Q wa e as eluen .
3. Resul s and discussion
3.1. Ca alys s cha ac e iza ion

Table 1 shows he chemical composi ions and ex u al p ope ies (BET su ace, po e
size and po e olume) o he p epa ed ca alys s. Fo compa ison, he suppo was also
measu ed. The expe imen al composi ions o all samples we e close o he nominal ones
con i ming he e ec i eness o he imp egna ion me hod employed o p epa e he
ca alys s. The pa en suppo p esen s supe io BET su ace a ea, which dec eases a e
Ni o Ni-Sn addi ion caused by he inco po a ion o me al pa icles in o he po es o he
suppo .
Figu e 1 shows he XRD pa e ns eco ded o he p epa ed ca alys s. The di ac ion
peaks a 2θ = 28.8º, 33.3º, 47.7º and 56.6º can be assigned o he (111), (200), (220) and
(311) planes o he cubic luo i e phase cha ac e is ic o ce ium oxide (JCPDS 34-
0394). I can also be obse ed he di ac ion lines co esponding o he MgAl2O4 spinel
phase (JCPDS 21-1152). As desc ibed in a p e ious wo k [38] he p esence o nickel
induces he ans o ma ion o he MgAl2O4 spinel s uc u e in o a con inuous MgNi
spinel laye al hough only a pa o nickel pa ially subs i u e Mg in he MgAl2O4 phase
o ming a new NixMgyAl2O4 phase. Di ac ion peaks cha ac e is ic o nickel oxide
(JCPDS 71-1179) we e obse ed a 37º and 43º. The e o e, i can be es ablished ha a
ac ion o Ni is well dispe sed as NiO pa icles wi h mode a e in e ac ion wi h he
suppo and ano he po ion o Ni is inco po a ed wi hin he non-s oichiome ic spinel-
like phase NixMgyAl2O4 [8] wi h s ong me al-suppo in e ac ion. XRD peaks
associa ed o Sn phases we e no obse ed in he Ni-Sn/CeMgAl ca alys likely due o
i s high dispe sion.
In o de o analyse he e olu ion and he changes in he c ys alline phases p oduced
unde educ i e condi ions, in si u X- ay di ac ion analyses we e pe o med.
Reduc ion was ca ied ou in an a mosphe e o hyd ogen dilu ed in A . The ob ained X-
ay di ac og ams as a unc ion o empe a u e a e shown in Fig. 2. In bo h cases, he
s uc u al modi ica ions a e only app eciable in he high empe a u e ange (700-900
ºC). Fo he Ni/CeMgAl ca alys , he NixMgyAl2O4 and NiO phases a e educed o
me allic nickel (JDPDS 45-1027), whe eas he educ ion o he NiSn/CeMgAl ca alys
leads o he o ma ion o NixSny alloys coexis ing wi h me allic Ni pa icles. I is known
ha nickel can be comple ely alloyed wi h in o ming Ni3Sn, Ni3Sn2 and/o Ni3Sn4
in e me allic compounds depending o he Ni/Sn mola a io [39]. In ou case, he
cha ac e is ic peaks o he Ni3Sn alloy (JCPDS 35-062) we e e idenced al hough he
p esence o o he NixSny in e me allic compounds canno be uled ou . In bo h cases,
me allic nickel was also p esen a e educ ion a 700 ºC. A u he inc ease o he
empe a u e o 900 ºC led o an inc ease o he Ni di ac ion lines, which indica e a
sin e ing o he nickel pa icles.
Su ace analysis measu emen s by XPS we e pe o med o e Ni/CeMgAl and Ni-
Sn/CeMgAl ca alys s o in es iga e he chemical su ace p ope ies and oxida ion s a es
o bo h me als. P io o XPS analysis, esh ca alys s we e educed o 1 h a 750ºC
unde lowing hyd ogen dilu ed in ine o ge in o ma ion abou he chemical s a e o
he ac i a ed ca alys s. Ni 2p and Sn 3d XPS spec a a e shown in Fig. 3. Bo h
Ni/CeMgAl and Ni-Sn/CeMgAl ca alys s exhibi ed wo decon olu ed Ni2p3/2 co e-le el
peaks wi h he co esponding sa elli e peaks posi ioned o ~6 eV, espec i ely; he peak
a 856 eV can be assigned o nickel alumina e phase, while ha he peak a 853 eV was
a ibu ed o me allic nickel [40, 41]. I was obse ed ha he ela i e amoun o nickel
species inco po a ed in he non-s oichiome ic spinel-like phase NixMgyAl2O4 is simila
in bo h ca alys s while ha he ela i e quan i y o me allic nickel is sligh ly supe io in
he Ni/CeMgAl ca alys s. This obse a ion is cohe en wi h he chemical composi ion o
he ca alys s sugges ing ha he addi ional nickel in he Ni/CeMgAl ca alys is highly
dispe sed o e he suppo as nickel oxide c ys alli es. F om he XPS esul s, i is
e iden ha only he c ys alline nickel oxide was educed and he nickel wi h s ong
in e ac ion wi h he suppo equi es highe empe a u es. Bo h Ni/CeMgAl and Ni-
Sn/CeMgAl ca alys s exhibi ed iden ical amoun o nickel alumina e species s ongly
bound o he suppo which we e no educed unde ou condi ions a 750ºC. The
educ ion o Ni2+ ions inco po a ed non-s oichiome ically in e ahed al o oc ahed al
si es o spinel phases equi es empe a u es as high as 800-850 ºC [42].
The Sn 3d egion spec a o he Ni-Sn/CeMgAl ca alys is shown in Fig. 3. The peak
a 485 eV can be assigned o Sn0 species while ha he highe binding ene gy Sn 3d5/2
peak a 487 eV is a ibu able o in oxides (Sn2+ o Sn4+) [32]. The educed species o
in is p obably a Ni-Sn alloy. Disc imina ing be ween Sn2+ and Sn4+ wi h XPS is
complica ed, bu he p esence o his species indica es ha he educ ion o in was no
comple e, and oxidized species a e exis en e en a e educ ion a 750 ºC, in ag eemen
wi h p e ious esul s on alumina-suppo ed ca alys s [43]. Alumina in e ac s s ongly
wi h he in and hinde s he edu ion o SnO [44]. The e e o e, XPS analysis o he Ni-
Sn/CeMgAl a e educ ion a 750ºC indica es he p esence o Ni and Sn sepa a e
phases. The ongoing educ ion o Ni and Sn can be concu en wi h he o ma ion o a
Ni-Sn alloy, howe e his alloys a e indis inguishable by XPS om educed Ni su aces
[32]. Consequen ly, i can be p oposed ha he Ni-Sn ca alys a e ac i a ion may be
ep esen ed by he o ma ion o bime allic pa icles wi h a cen al co e o nickel
su ounded by an ou e laye ich in in (Fig. 3). Du ing he educ ion phase, a pa o Sn
a oms mig a e o he me allic nickel pa icles su ace o o m a Sn laye ha is pa ially
educed o ming a Ni-Sn alloy co e ing he su ace o Ni pa icles. As can be obse ed
in Table 2, he su ace Ni/Sn de e mined by XPS was lowe han he bulk one
sugges ing ha Sn is loca ed mainly in he su ace laye s o he ca alys . Simila
obse a ions we e ound by o he au ho s in he li e a u e [34, 45].
3.2. Ac i i y measu emen s
The ca alys pe o mances we e es ed o he hyd ogen p oduc ion in he glyce ol
s eam e o ming. Figu e 4 shows he speci ic ac i i y o bo h Ni/CeMgAl and Ni-
Sn/CeMgAl ca alys s. I is no ewo hy ha he ca aly ic beha iou was di e en in bo h
cases. The ac i i y o Ni/CeMgAl ca alys was ound o be highe han ha o he Sn-
p omo ed one and, appa en ly, e y s able o he i s 12 hou s. Once his pe iod is
eached he ca aly ic ac i i y d as ically dec eases sugges ing a as deac i a ion
p ocess. The pa ial subs i u ion o Ni by Sn in he NiSn/CeMgAl ca alys educes he
ca aly ic ac i i y bu no ably enhances he s abili y o he ca alys . As can be obse ed,
a e an ini ial s abiliza ion pe iod o he i s 4 hou s, he ca aly ic ac i i y emains
s able. The obse ed ac i i y loss upon in addi ion migh be due o he ac ha he
nickel ac i e si es o he glyce ol s eam e o ming a e co e ed by Sn species as
con i med by XPS analysis. The Sn-en iched su ace p e en s he adso p ion o he
glyce ol molecules on he me allic nickel su ace and he subsequen e o ming
eac ions. Simila obse a ions we e ea ly epo ed in he li e a u e o he
NiSn/Ce0.75Z 0.25O2 [34] and P Sn/C [46] sys ems.
The same gaseous p oduc s we e de ec ed in bo h ca alys s (H2, CO, CO2, CH4 and
C2H4, Figu e 5), al hough wi h a di e en p oduc dis ibu ion. Condensable p oduc s
such as oxygena ed compounds we e no de ec ed a his eac ion empe a u e in good
ag eemen wi h p e ious esul s [36]. Hyd ogen was he main componen in bo h
ca alys s and emained in a ian du ing all ime-on-s eam. I can be no iced ha du ing
he i s 12 hou s, he Ni/CeMgAl ca alys p oduced he maximum hyd ogen mola
ac ion possible (70%) acco ding o he s oichiome y o he eac ion (1). A e 12 h o
eac ion changes in he mola composi ion we e obse ed. F om he e, he p oduc ion o
CO2 dec eases along wi h he inc ease o CO, CH4 and C2H4. Mo eo e , he change o
selec i i y appea s when ac i i y loss is obse ed. Fo he Ni-Sn/CeMgAl ca alys , an
inc ease o CO and dec ease o CO2 s a ed om he beginning o he eac ion while
ha CH4 and C2H4 we e p oduced con inuously.
The di e ences obse ed in bo h cases would be associa ed o he ca aly ic p ocesses
in ol ed in he glyce ol decomposi ion p ocess. The o e all eac ion o glyce ol s eam
e o ming (1) may be iewed as he combina ion o he glyce ol decomposi ion (2) and
he wa e gas shi (WGS) eac ion (3). Fu he mo e, i may also be accompanied by
o he he modynamically easible eac ions ha include:
2C3H8O3 (g) + H2 (g) ↔ 3CH4 (g) + 3CO (g) + 3H2O (g) (4)
CO (g) + 3H2 (g) ↔ CH4 (g) + H2O (g) (5)
CO2 (g) + 4H2 (g) ↔ CH4 (g) + 2H2O (g) (6)
CO2 (g) + CH4 (g) ↔ 2CO (g) + 2H2 (g) (7)
C(s) + H2O (g) → CO (g) + H2 (g) (8)
C(s) + CO2 (g) ↔ 2CO (g) (9)
C(s) + 2H2 (g) ↔ CH4 (g) (10)
The ini ial gas p oduc s dis ibu ion o Ni/CeMgAl ca alys indica es ha he WGS
eac ion was p edominan in he i s 12 h. This is consis en wi h a p oduc ion o
hyd ogen close o he equilib ium alue. A e 12 h o eac ion me hane and e hylene
a e p oduced. A aque e al. [13] sugges ed ha he o ma ion o me hane p oceeds
h ough ca bon monoxide me hana ion (5) while e hylene would be o igina ed by
hyd ogena ion/dehyd a ion o ca bonyl in e media es species. Cheng e al. [47]
p oposed, howe e , ha me hane is p oduced om he hyd ogenolysis o glyce ol (4)
uling ou he me hana ion eac ion as pa h o me hane p oduc ion. The s able
p oduc ion o me hane o Ni-Sn/CeMgAl ca alys sugges s ha me hane could be
o igina ed p edominan ly om he glyce ol hyd ogena ion and he ca bon
hyd ogena ion (10). Indeed, he s eam gasi ica ion o ca bon (8) and he Boudoua d
eac ion (9) may also accoun o he inc eased s abili y o Ni-Sn ca alys educing he
deposi ion o ca bonaceous species while he p oduc ion o CO and H2 keeps cons an .
CO o ma ion on Sn is he a ou ed eac ion om CO2 decomposi ion, especially o
[25] Chiodo V, F eni S, Gal agno A, Mondello N, F us e i F. Ca aly ic ea u es o Rh
and Ni suppo ed ca alys s in he s eam e o ming o glyce ol o p oduce hyd ogen.
Appl. Ca al. A Gen. 2010;381:1-7.
[26] Ebshish A, Yaakob Z, Tau iq-Yap YH, Bshish A, Shaibani A. Ca aly ic S eam
Re o ming o Glyce ol O e Ce ium and Palladium-Based Ca alys s o Hyd ogen
P oduc ion. J. Fuel Cell Sci. Technol. 2013;10.
[27] Huang X, Dang C, Yu H, Wang H, Peng F. Mo phology E ec o I /La2O2CO3
Nano ods wi h Selec i ely Exposed {110} Face s in Ca aly ic S eam Re o ming o
Glyce ol. ACS Ca al. 2015;5:1155-63.
[28] Dal San o V, Gallo A, Naldoni A, Guido i M, Psa o R. Bime allic he e ogeneous
ca alys s o hyd ogen p oduc ion. Ca al. Today. 2012;197:190-205.
[29] Nikolla E, Schwank J, Linic S. P omo ion o he long- e m s abili y o e o ming
Ni ca alys s by su ace alloying. J. Ca al. 2007;250:85-93.
[30] Kan H, Hyun SH, Shul Y-G, Lee H. Imp o ed solid oxide uel cell anodes o he
di ec u iliza ion o me hane using Sn-doped Ni/YSZ ca alys s. Ca al. Commun..
2009;11:180-3.
[31] Shabake JW, Hube GW, Dumesic JA. Aqueous-phase e o ming o oxygena ed
hyd oca bons o e Sn-modi ied Ni ca alys s. J. Ca al. 2004;222:180-91.
[32] Shabake JW, Simone i DA, Co igh RD, Dumesic JA. Sn-modi ied Ni ca alys s
o aqueous-phase e o ming: Cha ac e iza ion and deac i a ion s udies. J. Ca al.
2005;231:67-76.
[33] Fe ei a MaL, Nichio NN, Fe e i OA. A semiempi ical heo e ical s udy o Ni/α-
Al2O3 and NiSn/α-Al2O3 ca alys s o CH4 e o ming. J. Mol. Ca al. A Chem.
2003;202:197-213.
[34] Pengpanich S, Meeyoo V, Ri ksomboon T, Schwank J. iso-Oc ane pa ial oxida ion
o e Ni-Sn/Ce0.75Z 0.25O2 ca alys s. Ca al. Today. 2008;136:214-21.
[35] Saadi S, Hinnemann B, Hel eg S, Appel CC, Abild-Pede sen F, Nø sko JK. Fi s -
p inciples in es iga ions o he Ni3Sn alloy a s eam e o ming condi ions. Su . Sci.
2009;603:762-70.
[36] Bobadilla LF, Penko a A, Ál a ez A, Domínguez MI, Rome o-Sa ia F, Cen eno
MA, e al. Glyce ol s eam e o ming on bime allic NiSn/CeO2–MgO–Al2O3
ca alys s: In luence o he suppo , eac ion pa ame e s and deac i a ion/ egene a ion
p ocesses. Appl. Ca al. A Gen. 2015;492:38-47.

[37] Bobadilla LF, Penko a A, Rome o-Sa ia F, Cen eno MA, Od iozola JA. In luence
o he acid–base p ope ies o e NiSn/MgO–Al2O3 ca alys s in he hyd ogen
p oduc ion om glyce ol s eam e o ming. In . J. Hyd ogen Ene gy. 2014;39:5704-
12.
[38] Penko a A, Bobadilla L, I ano a S, Domínguez MI, Rome o-Sa ia F, Roge AC,
e al. Hyd ogen p oduc ion by me hanol s eam e o ming on NiSn/MgO–Al2O3
ca alys s: The ole o MgO addi ion. Appl. Ca al. A Gen. 2011;392:184-91.
[39] Onda A, Koma su T, Yashima T. P epa a ion and Ca aly ic P ope ies o Single-
Phase Ni–Sn In e me allic Compound Pa icles by CVD o Sn(CH3)4 on o Ni/Silica.
J. Ca al. 2001;201:13-21.
[40] Kha a AN, Pendle on P, Badalyan A, Abedini M, Amini MM. Decomposi ion o
Nickel Fo ma e on Sol–Gel Alumina and Cha ac e iza ion o P oduc by X-Ray
Pho oelec on and TOF-SIMS Spec oscopies. J. Ca al. 2002;205:7-15.
[41] Requies J, Ba io VL, Camb a JF, Güemez MB, A ias PL, La Pa ola V, e al. E ec
o edox addi i es o e Ni/Al2O3 ca alys s on syngas p oduc ion ia me hane
ca aly ic pa ial oxida ion. Fuel. 2008;87:3223-31.
[42] Pa cas F, Hönicke D. E ec o alkali doping on ca aly ic p ope ies o alumina-
suppo ed nickel oxide in he selec i e oxidehyd ogena ion o cyclohexane. Ca al.
Commun. 2005;6:23-7.
[43] Bacaud R, Bussiè e P, Figue as F. Mössbaue spec a in es iga ion o he ole o
in in pla inum- in e o ming ca alys s. J. Ca al. 1981;69:399-409.
[44] Se ano-Ruiz JC, Hube GW, Sánchez-Cas illo MA, Dumesic JA, Rod íguez-
Reinoso F, Sepúl eda-Esc ibano A. E ec o Sn addi ion o P /CeO2–Al2O3 and
P /Al2O3 ca alys s: An XPS, 119Sn Mössbaue and mic ocalo ime y s udy. J. Ca al.
2006;241:378-88.
[45] Nichio NN, Casella ML, San o i GF, Ponzi EN, Fe e i OA. S abili y p omo ion o
Ni/α-Al2O3 ca alys s by in added ia su ace o ganome allic chemis y on me als:
Applica ion in me hane e o ming p ocesses. Ca al. Today. 2000;62:231-40.
[46] Pas o -Pé ez L, Me lo A, Bui ago-Sie a R, Casella M, Sepúl eda-Esc ibano A.
Bime allic P Sn/C ca alys s ob ained ia SOMC/M o glyce ol s eam e o ming. J.
Colloid In e . Sci. 2015;459:160-6.
[47] Cheng CK, Foo SY, Adesina AA. Glyce ol S eam Re o ming o e Bime allic
Co−Ni/Al2O3. Ind. Eng. Chem. Res. 2010;49:10804-17.
[48] Vicen e J, E eña J, Mon e o C, Azkoi i MJ, Bilbao J, Gayubo AG. Reac ion
pa hway o e hanol s eam e o ming on a Ni/SiO2 ca alys including coke o ma ion.
In . J. Hyd ogen Ene gy. 2014;39:18820-34.
[49] Mon e o C, Ochoa A, Cas año P, Bilbao J, Gayubo AG. Moni o ing Ni0 and coke
e olu ion du ing he deac i a ion o a Ni/La2O3–αAl2O3 ca alys in e hanol s eam
e o ming in a luidized bed. J. Ca al. 2015;331:181-92.
[50] Reina TR, I ano a S, Cen eno MA, Od iozola JA. Boos ing he ac i i y o a
Au/CeO2/Al2O3 ca alys o he WGS eac ion. Ca al. Today. 2015;253:149-54.
[51] D esselhaus MS, D esselhaus G, Jo io A, Souza Filho AG, Sai o R. Raman
spec oscopy on isola ed single wall ca bon nano ubes. Ca bon. 2002;40:2043-61.
[52] Beyssac O, Go é B, Pe i e J-P, F oigneux E, Mo eau M, Rouzaud J-N. On he
cha ac e iza ion o diso de ed and he e ogeneous ca bonaceous ma e ials by Raman
spec oscopy. Spec ochim. Ac a A. 2003;59:2267-76.
[53] Suel es I, Láza o MJ, Moline R, Co bella BM, Palacios JM. Hyd ogen p oduc ion
by he mo ca aly ic decomposi ion o me hane on Ni-based ca alys s: in luence o
ope a ing condi ions on ca alys deac i a ion and ca bon cha ac e is ics. In J.
Hyd ogen Ene gy. 2005;30:1555-67.
[54] Tomishige K, Chen Y-g, Fujimo o K. S udies on Ca bon Deposi ion in CO2
Re o ming o CH4 o e Nickel–Magnesia Solid Solu ion Ca alys s. J. Ca al.
1999;181:91-103.
[55] Vicen e J, Mon e o C, E eña J, Azkoi i MJ, Bilbao J, Gayubo AG. Coke deac i a ion
o Ni and Co ca alys s in e hanol s eam e o ming a mild empe a u es in a luidized
bed eac o . In . J. Hyd ogen Ene gy. 2014;39:12586-96.
TABLES AND FIGURES
Table 1. Chemical composi ion and ex u al p ope ies o he p epa ed ca alys s
w .% composi iona
a .% composi ion
SBET
(m2 g-1)
Al2O3
MgO
CeO2
Ni
Sn
Ni
Sn
Ni/Sn
CeMgAl
77.9
(78)
9.6
(10)
12.5
(12)
-
-
-
-
-
106
Ni/CeMgAl
63.6
(60)
6.6
(7)
11.1
(12)
18.7
(20)
-
27.5
-
-
89
Ni-Sn/CeMgAl
60.1
(60)
6.8
(7)
11.9
(12)
17.3
(17)
3.7
(3)
25.6
2.7
9.4
86
a Nominal alues in pa en hesis
Table 2. Su ace composi ion ob ained by XPS analysis o he educed and spen
ca alys s
a .% composi ion
a
Al
Mg
Ce
O
Ni
Sn
C
Ni/Sn
Ni/CeMgAl educed
18.7
5.1
2.0
69.5
4.6
(27.5)
-
-
Ni/CeMgAl spen
-
-
0.15
4.35
-
-
95.5
Ni-Sn/CeMgAl educed
17.8
6.1
2.0
69.2
3.7
(25.6)
1.1
(2.7)
-
3.4
(9.4)
Ni-Sn/CeMgAl spen
3.5
2.1
0.4
21.7
0.5
0.1
71.8
5.0
a Bulk alues in pa en hesis
Fig. 1 XRD pa e ns o he p epa ed ca alys s
Fig. 2 In si u XRD pa e ns in educ i e a mosphe e o he p epa ed ca alys s: (a)
Ni/CeMgAl and (b) Ni-Sn/CeMgAl

Fig. 3 XPS spec a in he Ni 2p and Sn 3d le el and pic o ial ep esen a ion o educed
ca alys
Fig. 4 Ca aly ic ac i i y as a unc ion o ime o glyce ol s eam e o ming a 750ºC o
he p epa ed ca alys s
Fig. 5 P oduc dis ibu ion in glyce ol s eam e o ming a 750ºC o he s udied
ca alys s
Fig. 6 SEM mic og aph o he esh and used ca alys s, a e 4 h and 24 h on s eam in
glyce ol s eam e o ming a 750 ºC