H
2
- ich syngas p oduc ion om biogas e o ming:
O e coming coking and sin e ing using bime allic
Ni-based ca alys s
S. Ca asco-Ruiz
a
, Q. Zhang
b
,J.G

anda a-Loe
a
, L. Pas o -P
e ez
a
,
J.A. Od iozola
a,b
, T.R. Reina
a,b,**
, L.F. Bobadilla
a,*
a
Ino ganic Chemis y Depa men and Ma e ials Science Ins i u e, Uni e si y o Se ille-CSIC, 41092, Se illa, Spain
b
Depa men o Chemical and P ocess Enginee ing, Uni e si y o Su ey, Guild o d, GU2 7XH, UK
highligh s g aphical abs ac
H
2
- ich syngas can be ob ained
e icien ly om biogas e o ming.
Ni-based ca alys s a e deac i a ed
by me al sin e ing and coke
deposi ion.
Doping wi h low Rh loadings
deac i a ion can be success ully
p e en ed.
a icle in o
A icle his o y:
Recei ed 23 Augus 2022
Recei ed in e ised o m
11 Feb ua y 2023
Accep ed 20 Ma ch 2023
A ailable online 19 Ap il 2023
Keywo ds:
D y e o ming
H
2
- ich syngas
NieRh ca alys
Coking esis ance
abs ac
D y e o ming o me hane is a e y appealing ca aly ic ou e biogas (mainly composed by
g eenhouse gases: ca bon dioxide and me hane) con e sion in o added alue syngas,
which could be u he upg aded o p oduce liquid uels and added alue chemicals.
Howe e , he majo culp i s o his eac ion a e coking and ac i e phase sin e ing ha
esul in ca alys s deac i a ion. He ein we ha e de eloped a highly s able bime allic NieRh
ca alys suppo ed on mixed CeO
2
eAl
2
O
3
oxide using low-noble me al loadings. The
addi ion o small amoun s o hodium o nickel ca alys s p e en s coke o ma ion and
imp o es sin e ing esis ance, achie ing high con e sions o e ex ended eac ion imes
hence esul ing in p omising ca alys s o biogas upg ading.
©2023 The Au ho (s). Published by Else ie L d on behal o Hyd ogen Ene gy Publica ions
LLC. This is an open access a icle unde he CC BY-NC-ND license (h p://
c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
*Co esponding au ho .
** Co esponding au ho . Ino ganic Chemis y Depa men and Ma e ials Science Ins i u e, Uni e si y o Se ille-CSIC, 41092, Se illa,
Spain.
E-mail add esses: . ami ez eina@su ey.ac.uk (T.R. Reina), [email protected] (L.F. Bobadilla).
A ailable online a www.sciencedi ec .com
ScienceDi ec
jou nal homepage: www.else ie .com/loca e/he
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e27917
h ps://doi.o g/10.1016/j.ijhydene.2023.03.301
0360-3199/©2023 The Au ho (s). Published by Else ie L d on behal o Hyd ogen Ene gy Publica ions LLC. This is an open access a icle unde he CC BY-NC-
ND license (h p://c ea i ecommons.o g/licenses/by-nc-nd/4.0/).
In oduc ion
Figh ing global wa ming ale o mi iga e he nega i e human
impac on he en i onmen is one o he u gen p io i ies o
he scien i ic communi y aking op as Global Challenge p i-
o i y o a sus ainable de elopmen . The g eenhouse e ec
caused by he emission o ha m ul gases in o he a mosphe e
is p o oking an inc ease in global empe a u e and he e o e
o changing clima es causing d ough s, de o es a ion, and
dese i ica ion o he plane .
The ou main g eenhouse gases (GHG) a ac ing se ious
global a en ion oday a e CO
2
,CH
4,
SO
2
, and N
2
O. Ca bon di-
oxide is by a he mos emi ed gas in o he a mosphe e, ac-
coun ing o 76% o o al emissions in 2019 [1]. Acco ding o
Uni ed in Science 2020 [2], a epo compiled by he Wo ld
Me eo ological O ganiza ion (WMO) unde he di ec ion o he
Uni ed Na ions Sec e a y-Gene al, lockdown- ela ed all in
emissions caused by COVID-19 pandemic will no educe
su icien ly he CO
2
concen a ions in he a mosphe e. The
g owing end o CO
2
and CH
4
emissions d i en by he in-
c ease o global ene gy consump ion makes manda o y he
sea ch o al e na i es o achie e GHG alo isa ion [3].
The use o enewable ene gy sou ces ep esen s “mus do”
ac ion o a ou he ansi ion owa ds a low-ca bon econ-
omy. He ein, biogas can play an impo an ole in he de el-
opmen o he enewable ene gy ma ke nowadays, as i has a
wide a ie y o applica ions compa ed o o he enewable
ene gy sou ces [4]. Biogas is p oduced by anae obic diges ion
o biodeg adable was es and is mainly compound by ca bon
dioxide and me hane [5,6]. A e i s p oduc ion, he e a e h ee
main ou es o biogas u ilisa ion (Fig. 1b).
Biome hane is a nea ly pu e me hane s eam ha can be
ob ained by CO
2
sepa a ion o biogas. A p esen , bio-
me hane has been widely used as an engine ehicle uel in
many coun ies and has b oad de elopmen p ospec s [7,8].
Biogas is a aw ma e ial o p oduce syngas (H
2
/CO gas
mix u e) o indus ial syn heses o ene gy pu poses [7].
The e a e se e al ways o e o m biogas in o syngas such
as D y Re o ming o Me hane o Bi-Re o ming o Me hane.
G een elec ici y can be p oduced om biogas by Com-
bined Hea &Powe sys em (CHP) [9].
Syngas is a combus ible gas ha can be used o p oduce
elec ical ene gy in u bines and uel cells (Fig. 1c), bu he
cu en ma ke demand unco e s he po en ial o syngas as a
eeds ock o p oduce uels like diesel, naph ha, and gasoline,
as well as o high alue-added chemical inpu s ia Fische -
T opsch syn hesis (FTS) [10].
Focusing on biogas upg ading, he e se e al he mal ca a-
ly ic ou es o syngas p oduc ion om biogas namely: Oxy-
Re o ming o Me hane (ORM), Bi-Re o ming o Me hane
(BRM) and D y Re o ming o Me hane (DRM) [11]:
ORM :3CH4þCO2þO2#4CO þ6H2(Eq.1)
DH0¼58 kJmol
DG0¼1kJmol
BRM :3CH4þCO2þ2H2O#4CO þ8H2(Eq.2)
DH0¼220 kJmol
DG0¼151 kJmol
DRM :CH4þCO2#2CO þ2H2(Eq.3)
DH0¼247 kJmol
DG0¼170 kJmol
Fig. 1 eBiogas li e cycle: a) Biogas p oduc ion; b) Biogas u ilisa ion; c) Syngas u ilisa ion.
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e2791727908
E en hough he ORM (Eq. (1)) is p esumably he mos
a ac i e op ion gi en i s au o he mic na u e which esul s in
signi ican ene gy sa ings, sa e y conce ns associa ed wi h
oxygen use limi i s in e es o indus ial and la ge-scale ap-
plica ions [12]. The mos popula echnologies a e he BRM
(Eq. (2)) and he DRM (Eq. (3)). The i s one leads o a syngas
H
2
/CO mix u e a io o a maximum o 2, meanwhile he las
one can achie e a mola a io o 1 [13]. Howe e , in he con ex
o a ci cula economy when a CO
2
u ilisa ion ou e is pu sued,
CO
2
con e sion may p e ail o e H
2
/CO a io as selec ion
c i e ia o he op imal ou e. He ein, DRM akes he edge o e
BRM as epo ed elsewhe e [14].
In any case, he condi ions used o e o ming also a ou
o he side eac ions (Eqs. (4)e(7)), in which ca bon deposi s
(C*) a e o med as ollows [15]:
Me hane decomposi ion :CH4#C*þ2H2(Eq.4)
DH0¼75 kJmol
Boudoua d eac ion :2CO #C*þCO2(Eq.5)
DH0¼172 kJmol
CO2hyd ogena ion :CO2þ2H2#C*þ2H2O(Eq.6)
DH0¼90 kJmol
CO hyd ogena ion :CO þH2#C*þH2O(Eq.7)
DH0¼131 kJmol
The o ma ion o ca bon deposi s leads o ca alys s deac-
i a ion, so i is impo an o ind ma e ials capable o mini-
mizing ca bon o ma ion.
Besides hese eac ions ha can deac i a e he ca alys ,
he e is also a side eac ion ha would dec ease he H
2
yield,
he Re e se Wa e Gas Shi (RWGS):
RWGS eac ion :CO2þH2#CO þH2O(Eq.8)
DH0¼46 kJmol
When i comes o ca alys s selec ion, Ni-based ca alys s
ha e shown a good ac i i y and con e sion o biogas in o
syngas [16e18], bu hey a e ex book example o coking and
sin e ing [19,20]. I has also been epo ed ha noble me al-
based ca alys s ha e a g ea e ac i i y and con e sion, bu in
addi ion, hey ha e mo e s abili y and highe coking esis-
ance ye a e expensi e and less a ailable [21]. Fo hese e y
easons bime allic alloys comp ised o Ni and low amoun s o
noble me als (i.e. Pd, Rh, P , Ru) a e in e es ing in e o ming
p ocesses.
I espec i ely o he selec ed o mula ion, i is clea ha
ca alys s s abili y ( esis ance o he di e en deac i a ion
phenomena) is c ucial o an op imal p ocess design.
F equen ly in he e ogeneous ca alysis, deac i a ion issues a e
esol ed by ca alys egene a ion. Fo ins ance, he mal
ea men s o emo e ca bon deposi s o ge id o po en ial
sul ides [22].
The sequence o ac i i y o he DRM using M/MgO
(M≡I ,Ni,Pd,P ,Rh,Ru) ca alys s is Ru >Rh >Ni »I ,P ,Pd
conside ing he p esence o sul u compounds [23].
Compa ing Ru and Rh, he egene a ion o Ru ca alys s is
much lowe han ha o Rh, so i seems ha he bes op ion
o ou objec i e is a bime allic NieRh ca alys [24].
Beyond he ac i e phase, he suppo choice is also
essen ial o ensu e an adequa e pe o mance. Fo e o ming
eac ions gi en he demanding p ocess condi ions (high
empe a u es) he mal and mechanical s abili y a e key ac-
o s o conside . In his ega d, alumina suppo s (Al
2
O
3
) o e s
an excellen balance he mal/mechanical p ope y while also
displaying high speci ic a ea which enhances me allic
dispe sion. Howe e , he acidic si es o his kind o suppo s
can inc ease coke o ma ion. Tha is why ce ia (CeO
2
)isan
in e es ing edox p omo e . Indeed, ce ia allows a lexible
unning o he acid/base p ope ies o he suppo and p o-
ides excellen oxygen mobili y, p e en ing ca bon deposi ion
ia oxida ion o he coke p ecu so s [13].
Se e al au ho s ha e s udied bime allic NieRh ca alys s o
he DRM eac ion [25,26], ye he speci ic ole o Rh and how i
bene i s ca aly ic pe o mance is no ully add essed. Pa ic-
ula ly, sho - e m s abili y es and insu icien o pos -
eac ion analysis in he cu en epo s pose some ques ions
ega ding he undamen al ole o Rh in he bime allic
o mula ion. So, unde hese p emises, his wo k ocuses on
he de elopmen o an ad anced mul icomponen ca alys s
NieRh/CeO
2
eAl
2
O
3
whose pe o mance is compa ed o a ba e
monome allic Ni/CeO
2
eAl2O
3
sys em. Ca aly ic ac i i y and
s abili y esul s as well as p e- and pos - eac ion cha ac e i-
sa ion will be discussed o unde s and he bene icial e ec o
noble me al addi ion.
Expe imen al
Ca alys s p epa a ion
The suppo used was a comme cial ce ia-alumina oxidic
suppo wi h 20 w % o CeO
2
(Pu alox, SASOL).
The monome allic ca alys was p epa ed by we imp eg-
na ion, whe e he suppo was i s imp egna ed wi h
Ni(NO
3
)
2
·6H
2
O (Sigma-Ald ich) dilu ed in dis illed wa e ,
e apo a ed a educed p essu e in a o a apo , d ied o e -
nigh a 100 C and calcined a 550 C o 4 h.
In a simila p ocedu e, he bime allic ca alys was p epa ed
by we co-imp egna ion, whe e he suppo was imp egna ed
wi h Ni(NO
3
)
2
·6H
2
O (Sigma-Ald ich) and RhCl
3
(Sigma-Ald ich)
dilu ed in dis illed wa e , e apo a ed a educed p essu e in a
o a apo , d ied o e nigh a 100 C and calcined a 550 C o
4h.
In all cases he NiO con en is calcula ed o be 10 w % and
0.5 w % Rh o he bime allic sample. These a ios we e cho-
sen based on p e ious wo ks [27e29].
Cha ac e isa ion echniques
The ex u al p ope ies o he samples we e e alua ed om
ni ogen adso p ion-deso p ion iso he ms a liquid ni ogen
empe a u e in a Mic ome i ics T is a II appa a us. Be o e
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e27917 27909
analysis, he samples we e degassed a 150 C o 8 h in
acuum.
On one hand, he speci ic su ace a ea (S
BET
) was de e -
mined by he B unaue -Emme -Telle (BET) me hod [30] and
co esponds o he sum o he inne su ace o he po e plus
he ou e su ace o he g ains. Addi ionally, he Ba e -
Joyne -Halenda (BJH) me hod was used o de e mining he
po e size dis ibu ion using he deso p ion iso he m. Po e
olume ep esen s he inne and ou e g anula olume, and
he a e age po e size was calcula ed as he a io o he po e
olume and he speci ic su ace a ea and no malized using a
coe icien ha depends on he po es shape [31].
Scanning elec on mic oscopy (SEM) analysis was pe -
o med on he calcined samples in a acuum, using a JEOL
5400 mic oscope equipped wi h an EDS analyse (Ox o d Link).
X- ay di ac ion measu emen s we e ca ied ou in a
X'Pe P o PANaly ic ins umen . The di ac ion pa e ns we e
eco ded a 40 mA and 45 kV using Cu-Ka adia ion
(l¼0.154 nm). The 2qangle was inc eased using a s ep size o
0.05and a s ep ime o 300 s in a ange o 10 o 90.
Tempe a u e p og ammed educ ion (TPR) measu emen s
we e conduc ed in a con en ional U-shaped qua z eac o
connec ed wi h a he mal conduc i i y de ec o (TCD), pass-
ing a low o 50 mL/min o 5% H
2
dilu ed in A . TPR expe i-
men s we e pe o med using app oxima ely 50 mg o each
ca alys a a hea ing a e o 10 C/min om oom empe a u e
(RT ¼25 C) o 900 C. A mix u e con aining ace one and d y
ice was used as a cold ap o emo e he wa e o med
h oughou he p ocedu e.
Tempe a u e p og ammed oxida ion (TPO) expe imen s
we e ca ied ou a e he long- e m s abili y es s o in es i-
ga e he ca bon species deposi ed on he ca alys su ace. TPO
analysis we e conduc ed in a U-shaped qua z eac o coupled
o a PFEIFFER Vacuum P ismaPlus mass spec ome e . 25 mg
o bo h samples we e hea ed up o 900 C a a a e o 10 C/min
in a calib a ed low o 50 mL/min (5% O
2
, 95% He).
Ca aly ic ac i i y and s abili y
Ca aly ic pe o mance was e alua ed in a ixed-bed con in-
uous- low eac o desc ibed elsewhe e [32] in which 100 mg o
undilu ed ca alys we e loaded o each un. P io o he e-
ac ion, he samples we e in si u educed in a low o 10% H
2
/N
2
a 850 C o 1 h. The eac ion was pe o med a a mosphe ic
p essu e passing a low o 100 mL/min o CO
2
/CH
4
/N
2
(mola
a io o 1:1:6) and dec easing successi ely he empe a u e
om 850 C down o 550 C un il achie ing he s eady s a e
each 50 C. The WHSV (Weigh Hou ly Space Veloci y) was
ixed a 60 L/g
ca
$h. All he gases in ol ed in he expe imen s
we e moni o ed by u ilising an on-line gas analyse (ABB-
AO2020) which was equipped wi h bo h IR and TCD de ec o s.
The ca aly ic s abili y o bo h mono and bime allic ma e-
ials was also s udied. Fo his pu pose, 100 mg o each ca a-
lys we e educed unde he same condi ions as o he
ca aly ic ac i i y es s, and hen he eac ion condi ions we e
e alua ed a a mosphe ic p essu e passing he same low
condi ions as abo e a 650 C o 48 h. A second s abili y s udy
was ca ied ou a 850 C o 48 h.
The ca aly ic esul s ob ained ha e been exp essed in
e ms o con e sions o bo h eac an gases and he H
2
/CO
mola a io. The equa ions used o es ima e hese pa ame e s
a e as ollows:
CH4con e sion ð%Þ¼nCH4in nCH4ou
nCH4in
$100 (Eq.9)
CO2con e sion ð%Þ¼nCO2in nCO2ou
nCO2in
$100 (Eq.10)
H2=CO ¼nH2
nCO
$100 (Eq.11)
Being n he mola low o CH
4
,CO
2
,H
2
, and CO espec i ely
and he subsc ip s in o ou co espond o he inle o he ou le
eac o low.
Resul s and discussion
Cha ac e isa ion o solid ca alys s
The ex u al p ope ies (S
BET
, and po e size and olume) o he
ca alys s and he comme cial suppo a e summa ized in
Table 1. As can be obse ed, he speci ic su ace a eas o he
ca alys s a e qui e simila . This indica es a good dispe sion o
he me als on he su ace o he suppo , which will ha e a
posi i e e ec on he ac i i y o he ma e ials. Likewise, bo h
po e olume (V
po e
) and po e wid h (D
po e
) a e almos iden i-
cally, showing ha no agglome a ion o me al pa icles has
occu ed on he ca alys and ha he po es suppo ha e no
been blocked.
Besides ha , he ex u al p ope ies ob ained o he ce ia-
p omo ed alumina suppo sugges ha bo h Ni and Rh a e
in oduced in o he po es o he suppo , so ha he speci ic
su ace a ea, po e olume and po e wid h dec ease.
In addi ion, he ni ogen adso p ion/deso p ion iso he ms
a 77 K ha e been ob ained o bo h ca alys s as shown in
Figu e S1. Acco ding o IUPAC classi ica ion [33], ou iso he m
ypes a e usually ound in ca alys cha ac e isa ion, and each
iso he m shape depends di ec ly on he solid po ous ex u e.
In ou case, bo h iso he ms co espond o a ype IV iso he m,
which is ela ed o a mesopo ous ma e ial (po e size be ween
2 nm and 50 nm). Besides, he e is also a classi ica ion o he
ype o hys e esis, being he ype H1 ound in ou iso he ms.
This kind o hys e esis is cha ac e is ic o solids consis ing o
pa icles a e sed by quasi-cylind ical channels o consis ing
o agg ega es (consolida ed) o agglome a es (unconsolida ed)
o sphe oidal pa icles, and he po es may be uni o m in size
and shape.
Figu e S2 shows he po e size dis ibu ion ob ained by BJH
me hod o bo h ca aly ic ma e ials. As can be no ed, bo h
po e dis ibu ion cu es a e e y simila con i ming ha he e
Table 1 eTex u al p ope ies o he suppo and p epa ed
ca alys s.
Sample S
BET
(m
2
/g) V
po e
(cm
3
/g) D
po e
(
A)
CeO
2
eAl
2
O
3
159 0.39 71.9
Ni/CeO
2
eAl
2
O
3
132 0.31 67.7
NieRh/CeO
2
eAl
2
O
3
138 0.32 66.6
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e2791727910
a e no agglome a ion o me al pa icles and he ma e ial
ca i ies ha e no been blocked.
The mo phology o he calcined ca alys s is shown in he
SEM mic og aphs p o ided in Fig. 2. Analysing he dis ibu ion
o elemen s in he di e en mappings, we can con i m ha
bo h Ni and Rh a e well dispe sed and homogeneously
dis ibu ed on he su ace o he ca alys s, co obo a ing a
success ul syn hesis o bo h samples.
The s uc u e o bo h p epa ed ca alys s was analysed by
means o X-Ray Di ac ion (XRD). Bo h samples we e i s ly
analysed a e calcina ion a 550 C(Fig. 3a). F om his igu e, i
can be obse ed ha all samples con ain he ypical di ac-
ion lines co esponding o he (111), (200), (220) and (311)
c ys alline planes expec ed o he CeO
2
luo i e-like phase
while he peaks co esponding o he (311) and (440) c ys al-
line planes a e asc ibed o he g-Al
2
O
3
phase [34].
Fig. 2 eSEM mic og aphs o bo h calcined ca alys s: (I)Ni/CeO
2
eAl
2
O
3
; (II)NieRh/CeO
2
eAl
2
O
3
; whe e (a) gene al iew; (b) Ce
mapping; (c) Ni mapping; (d) Al mapping; (e) Rh mapping.
Fig. 3 eXRD o : a) calcined samples (550 C); b) educed samples (850 C).
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e27917 27911

Fig. 3b shows he XRD pa e ns o he samples a e
educ ion in a low o 10% H
2
/N
2
a 850 C o 1 h. Hence, he
di ac ion lines ela ed o me allic Ni can be clea ly seen,
al hough he me allic Rh peaks we e ha dly obse ed due o
he high dispe sion and/o low loading o he noble me al.
An o e all idea o he edox beha iou and he me al-
suppo in e ac ions was ga he ed by Tempe a u e P o-
g ammed Reduc ion (TPR) analysis. The TPR p o iles ob ained
o bo h ca alys s a e p esen ed in Fig. 4. No ably h ee
di e en egions a e dis inguished. Fi s , aRegion (150e350 C)
shows he educ ion o su ace ce ia [35e37], which is
enhanced on he bime allic ca alys due o he spill-o e
phenomenon [38]. The bRegion (450e650 C) exhibi s he
educ ion o NiO wi h low o medium s eng h o in e ac ion
wi h he suppo . Finally, he educ ion o NiO wi h high
s eng h o in e ac ion wi h he suppo occu s a empe a-
u es abo e 700 C(gRegion)[39].
Compa ing bo h educ ion p o iles, we can see ha he
addi ion o hodium imp o es no ably he educibili y o Ni,
acili a ing i s educ ion a lowe empe a u es. This highe
educibili y implies be e edox p ope ies, and hese edox
cha ac e is ics may enhance he ca aly ic pe o mance du ing
he DRM eac ion.
DRM ca aly ic ac i i y
The educed ca alys s we e es ed in he DRM eac ion a a
empe a u e ange o 550e850 C and a a mosphe ic p essu e
wi h a CO
2
/CH
4
mola a io 1:1. As we can see in Fig. 5.a, CH
4
con e sions we e lowe han CO
2
con e sions, which may
sugges ha he CH
4
ac i a ion is mo e di icul and equi es
highe empe a u es in good ag eemen wi h DFT esul s [40].
The ac i a ion and clea age o CeH bonds is mo e ene ge ic
han CO
2
dissocia ion.
On he o he hand, compa ing he esul s be ween bo h
ca alys s, no signi ican di e ences a e obse ed in he ca bon
dioxide and me hane con e sions, wi h simila con e sion
le els o he monome allic ca alys and he bime allic one.
In e ms o he H
2
/CO mola a io (Fig. 5b), bo h ca alys s
also showed simila esul s. Howe e , his mola a io s ays
o e 1 h oughou he eac ions, so we can conclude ha H
2
concen a ion is highe han CO concen a ion. This is an
in e es ing o biogas e o ming ia DRM since ypical H
2
/CO
a e sligh ly lowe ( ypically close o 1). The main eason o his
obse a ion is he p esence o mul iple pa allel eac ions
desc ibed abo e in he in oduc ion sec ion ei he consuming
CO o p oducing H
2
,i.e. he RWGS/WGS p ocess which is
e ec i ely ca alysed by NieCeO
2
sys ems [41].
Fig. 6 shows he XRD o he spen ca alys s. Apa om he
di ac ion peaks o he me allic and suppo phases, we can
obse e a highe in ensi y in he 20-402qzone o he
monome allic ca alys han o he bime allic one. This es-
i ies he o ma ion o ca bon deposi s as will be well dis-
cussed u he below.
DRM ca aly ic s abili y
S abili y es s show how e icien ca alys s a e o e long e-
ac ion uns and commonly s abili y is mo e impo an han
ac i i y in comme cial ca alys s design since i de e mines he
equency o s a -up/shu downs ope a ions and he ca alys s
egene a ion cos .
Fig. 4 eH
2
-TPR p o iles o bo h samples.
Fig. 5 eDRM ca aly ic ac i i y: a) CO
2
&CH
4
con e sions; b) H
2
/CO mola a io.
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e2791727912
Hence, we ha e pe o med s abili y s udies a wo di e en
empe a u es. This way, we will be able o s udy he ca bon
deposi ion ole ance and me al sin e ing esis ance o bo h
ca alys s in mo e in-dep h manne . On he hand we will assess
he coking ole ance a wo di e en ope a ion egimes and on
he o he hand we will e alua e he sin e ing esis ance when
inc easing he he mal s ess on he ca alys 's pa icles.
The i s s abili y es was pe o med a 650 C. As shown in
Fig. 7, bo h me hane and ca bon dioxide con e sions o he
Ni/CeO
2
eAl
2
O
3
sample dec ease wi h ime-on-s eam, while
con e sions o he NieRh/CeO
2
eAl
2
O
3
sample emain p ac-
ically cons an o e he 48 h o he es . This esul clea ly
e idences he supe io i y o he bime allic o mula ion
showcasing he posi i e impac o small amoun s o Rh.
In o de o s udy he ca bonaceous deposi s o e bo h
ca alys s, TPO and XRD analyses we e pe o med on he spen
samples a e he s abili y es .
The peaks obse ed in he TPO diag am (Fig. 8a) ep esen
he gene a ion o CO
2
by he oxida ion o he ca bon deposi s
o med, so i can be a i med ha he ca bon deposi ed in he
NieRh ca alys is minimal compa ed o ha o he e e ence
sample. The empe a u e a which he CO
2
maximums appea
p o ides in o ma ion on he na u e o he ca bonaceous spe-
cies deposi ed on he ca alys su ace. Gene ally speaking,
ca bonaceous deposi s ha e di e en s uc u e o de s and
mo phologies, and, in ou case, we can dis inguish be ween
h ee ypes [42e50]:
The i s peak appea s a 300e400 C and co esponds o a
e y labile amo phous ca bon (C
a
) which can be emo ed a
low empe a u es. The second peak appea s a ~500 C and
co esponds o g aphi ic ca bon (C
b
) which is e y he mally
s able. The C
b
is ha dly emo ed by gasi ica ion so i leads o
se e e ca aly ic deac i a ion. Finally, he C
g
species we e
asc ibed o ca bon nano ubes, hey appea a ~650 C.
As we can see in Fig. 8.a, pos -s abili y (650 C) TPO analysis
show mo e C
b
deposi s on he monome allic ca alys han on
he bime allic one. This jus i ies he con inuous dec ease o
CH
4
and CO
2
con e sions o he Ni/CeO
2
eAl
2
O
3
which is
ela ed o ca bon poisoning and e idences he poo e pe o -
mance o he e e ence ma e ial compa ed o he ad anced
bime allic o mula ion. In addi ion, he XRD analyses o bo h
spen samples (Fig. 8b) also show a clea di e ence since he e
is a peak a 26app oxima ely in he monome allic sample
ha i is no seen in he bime allic sample. This peak co e-
sponds o g aphi ic ca bon deposi s because i has a c ys al-
line s uc u e de e mined [51], hus con i ming ha he e a e
mo e g aphi ic ca bon deposi s on he monome allic ca alys
in ai ag eemen wi h he TPO analyses.
The second s abili y es was pe o med a 850 C(Fig. 9). In
his ins ance, he CO
2
con e sions dec ease as he same way
o bo h ca alys s, going om almos 100%e95% a he end o
he expe imen . None heless, he me hane con e sion o he
Ni/CeO
2
eAl
2
O
3
sample dec eases om 97% down o 75% while
he CH
4
con e sion o he NieRh/CeO
2
eAl
2
O
3
sample de-
clines less, going om almos 95%e80%. E en so, hese con-
e sions a e highe han hose achie ed du ing he s abili y
es a 650 C gi en he endo he mic na u e o he eac ion.
As in he p e ious case, he TPO analyses o he pos -
s abili y samples a 850 C also ep esen h ee e y dis inc
peaks co esponding o he h ee ypes o ca bon deposi s
men ioned abo e (Fig. 10a). In his case, he a ea unde he
cu e o NieRh/CeO
2
eAl
2
O
3
sample is signi ican ly lowe han
he a ea unde he cu e o Ni/CeO
2
eAl
2
O
3
sample, which
econ i ms ha he addi ion o Rh o he Ni ca alys imp o es
coking esis ance. In ac , wi h a iew o acili a e i s isual-
isa ion, he signal o he o me has been mul iplied by a
ac o o 4, as i was di icul o sepa a e he wo peaks on a
no mal scale. I mus be emphasised he ca bon deposi ion is
he modynamically mo e a ou ed a lowe empe a u es
be ween 600 and 750 C. In any case ou bime allic sample
ou pe o med he e e ence ca alys s in bo h empe a u e
egimes.
Howe e , g aphi ic ca bon o ma ion was lowe on he
monome allic ca alys in his es , as coke o ma ion is mo e
a ou ed a lowe empe a u es. A pa icula i y o his s a-
bili y es is ha i is possible o obse e how he sin e ing
phenomenon a ec s he monome allic ca alys wi h he XRD
Fig. 6 eXRD o spen samples.
Fig. 7 e650 ºC-S abili y es esul s.
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e27917 27913
o hese samples (Fig. 10b). The peak o Ni loca ed a 51.8was
used o es ima e he c ys alli e size by means o Sche e
equa ion. I was ound ha he c ys alli e size is highe in he
monome allic ca alys . To make a compa ison wi h he Ni
c ys alli e size be o e eac ion, he same p ocedu e was pe -
o med wi h he decon olu ed Ni peak o bo h educed ca a-
lys s. In he case o he monome allic sample, he Ni
c ys alli e size inc eased om 12 nm o 17.5 nm, while in he
bime allic ca alys his alue inc eased om 11.7 nm o 13 nm.
This is e idence o how he addi ion o a bime allic ac i e
phase p e en s he sin e ing deac i a ion o he ca alys s
since he seg ega ion o Ni species om s able bime allic
o mula ions is somewha less ene ge ically a ou able in
compa ison o pu e Ni pa icles [52,53].
Fig. 8 ePos cha ac e isa ion a e s abili y es a 650 C: a) TPO; b) XRD.
Fig. 9 e850 ºC-S abili y es esul s.
Fig. 10 ePos cha ac e isa ion a e s abili y es a 850 C: a) TPO; b) XRD.
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e2791727914
Compa ing he esul s ob ained in he s abili y es s a
650 C and 850 C, we can s a e ha he deac i a ion a 650 C
is mainly caused by he o ma ion o coke, while a 850 C he
deac i a ion by sin e ing p e ails, as less ca bon o ma ion is
obse ed. In bo h si ua ions, he bime allic ca alys s ands ou
as a p omising al e na i e.
A 650 C he p edominan seconda y eac ion is me hane
decomposi ion (Eq. (4)) whe e ca bon deposi s and hyd ogen
a e o med om me hane. This explains he low me hane
yields and ha he H
2
/CO a io emains abo e 1 a his em-
pe a u e. A 850 C, howe e , he e e se eac ion o he Bou-
doua d eac ion (Eq. (5)) s a s o occu , whe e he ca bon
o med wi h CO
2
is con e ed o CO. This could dec ease he H
2
/
CO a io, bu C
b
is e y he mallys able,and heCO o ma ionis
no la ge enough o coun e ac he excessi e H
2
o ma ion.
Conclusions
Biogas con e sion o H
2
- ich syngas ep esen s a s aigh o -
wa d app oach o g eenhouses alo isa ion in he con ex o
a ci cula economy. He ein, a mul icomponen ca alys based
on NieRh bime allic ac i e phase suppo ed on a ce ia-
p omo ed alumina (NieRh/CeO
2
eAl
2
O
3
) has been p epa ed
and es ed in biogas upg ading ia DRM. The pe o mance o
his ad anced sys em has been compa ed o ha o a e e -
ence monome allic nickel homologue (Ni/CeO
2
eAl
2
O
3
).
Di e en p e- eac ion and pos - eac ion cha ac e isa ion
es s ha e been ca ied ou , as well as a eac ion moni o ing
based on he CH
4
and CO
2
con e sions and he H
2
/CO mola
a io ob ained. The s a ing hypo hesis sugges ed be e pe -
o mances o he bime allic ca alys han o he mono-
me allic one.
TPR expe imen s showed a highe educibili y o he
bime allic sample compa ed o he monome allic one, which
implies be e edox p ope ies, enhancing he ca aly ic pe -
o mance du ing he DRM eac ion.
In he ca aly ic ac i i y es , he di e ences obse ed a e
negligible howe e key disc epancies a e obse ed in he
ca alys s’ s abili y. The esul s ob ained in he di e en s a-
bili y es s show ha he e is indeed a syne gic NieRh e ec
ha imp o es ca aly ic s abili y, hus he con e sions on he
bime allic ca alys emain almos cons an o e long- e m
uns while he monome allic ca alys su e s a signi ican
ac i i y d op showcasing i s deac i a ion.
Pos -s abili y TPO expe imen s showed he o ma ion o
wo di e en ca bonaceous species, an amo phous ca bon
and a c ys alline ca bon, he la e being he cause o deac i-
a ion as i is e y he mally s able. Ou esul s demons a e
ha he monome allic sys ems nuclea e g ea e concen a-
ion o ca bon han he ad anced bime allic sample being
ca bon deposi ion a ou ed in he low- empe a u e ange (i.e.
650 C).
Wi h he s abili y es a 850 C he sin e ing o nickel
pa icles can be also obse ed, wi h a ema kable inc ease on
he Ni clus e s size on he monome allic ca alys compa ed o
he bime allic which demons a es a g ea e ole ance o-
wa ds ac i e phase agglome a ion.
O e all, his pape showcases he key ole played by he -
e ogeneous ca alysis wi hin CO
2
con e sion echnologies and
he esea ch s ill needed o keep e ining ca aly ic o mula-
ions o a oid deac i a ion seeking o an op imal ope a ion.
In pa icula , his wo k has demons a ed ha he addi ion o
low amoun s o Rh signi ican ly imp o es he ca aly ic pe -
o mance, educing coke o ma ion and sin e ing. A no el
angle o his wo k is he p o en coking esis ance o ou
mul icomponen ca alys s in he low- empe a u e DRM win-
dow. This is a ema kable esul beyond he ca alys s op i-
miza ion opening a new esea ch a enue o he design o
low- empe a u e DRM uni s esul ing in a po en ial angible
educ ion in ene gy consump ion and o e all p ocess ope a-
ion cos s.
In any case, he design o ad anced mul icomponen ca -
alys s as he one p esen ed he ein will con ibu e o acili a e
he ansi ion owa ds g eene indus ial p ocesses o p oduce
low-ca bon uels and added alue chemicals.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing
inancial in e es s o pe sonal ela ionships ha could ha e
appea ed o in luence he wo k epo ed in his pape .
Acknowledgemen s
Financial suppo o his wo k was ga he om Spanish
Minis y o Science and Spanish Minis y o Science and
Inno a ion h ough he p ojec s PLEC2021-008086, RYC2018-
024387-I as well as he p ojec PID2019-108502RJ-I00. This
wo k was also pa ially unded by he Uni e si y o Se ille ia
he VI PPIT g an scheme o alen ed esea che s. Se gio
Ca asco would also like o acknowledge he Sociedad Espa-
~
nola de Ca 
alisis (SECAT) o his ellowship. Funding om he
Eu opean Commission h ough he BIOALL p ojec (G an
Ag eemen : 101008058) SASOL is kindly acknowledged o
p o iding he ca alys s'suppo s.
Appendix A. Supplemen a y da a
Supplemen a y da a o his a icle can be ound online a
h ps://doi.o g/10.1016/j.ijhydene.2023.03.301.
e e ences
[1] Yo o KO, Da amola MO. CO2 emission sou ces, g eenhouse
gases, and he global wa ming e ec . In: Ad ances in ca bon
cap u e. Else ie ; 2020. p. 3e28. h ps://doi.o g/10.1016/b978-
0-12-819657-1.00001-3.
[2] Kappelle M. Uni ed in science 2020 - UN clima e epo . 2020.
h ps://doi.o g/10.13140/RG.2.2.12801.28004.
[3] Bouche O, F iedlings ein P, Collins B, Shine KP. The indi ec
global wa ming po en ial and global empe a u e change
po en ial due o me hane oxida ion. En i on Res Le 2009;4.
h ps://doi.o g/10.1088/1748-9326/4/4/044007.
[4] Ullah Khan I, Ha iz Dza an O hman M, Hashim H,
Ma suu a T, Ismail AF, Rezaei-Dash A zhandi M, Wan
in e na ional jou nal o hyd ogen ene gy 48 (2023) 27907e27917 27915