Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
Please ci e his a icle as: U. I ia e-Velasco e al., Jou nal o Indus ial and Enginee ing Chemis y, h ps://doi.o g/10.1016/j.jiec.2023.12.003
A ailable online 10 Decembe 2023
1226-086X/© 2023 The Au ho s. Published by Else ie B.V. on behal o The Ko ean Socie y o Indus ial and Enginee ing Chemis y. 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/).
Ni suppo ed on bioapa i e o WGS: Imp o ing ca alys s abili y and H
2
selec i i y by P -doping and he mochemical ac i a ion o he suppo
U. I ia e-Velasco
a
, M.A. Gu i´
e ez-O iz
b
, A.J. Reynoso
b
,
*
, J.L. Ayas uy
b
a
Depa men o Chemical Enginee ing, Facul y o Pha macy, Uni e si y o he Basque Coun y UPV/EHU, Paseo de la Uni e sidad, 7 01006, Vi o ia, Spain
b
Depa men o Chemical Enginee ing, Facul y o Science and Technology, Uni e si y o he Basque Coun y UPV/EHU, Sa iena S/N 48940, Leioa, Spain
ARTICLE INFO
Keywo ds:
Bioapa i e
Nickel
WGS
K
2
CO
3
H
2
SO
4
ABSTRACT
Ca aly ic p ope ies o Ni and NiP ca alys s suppo ed on o na u ally de i ed hyd oxyapa i e (HAp) we e
in es iga ed o he Wa e -Gas Shi (WGS) eac ion in he 200–450 ◦C ange. HAp was ob ained by he mo-
chemical con e sion o was e animal bones which yield a po ous solid mainly composed by apa i e. A eed
composi ion ep esen a i e o eal e o me ou le s eam was used (CO/H
2
O/CO
2
/H
2
=5/46/4/31 mol %) a a
gas hou ly space eloci y o 120,000 h
−1
. The ca alys s we e ho oughly cha ac e ised by N
2
physiso p ion, ICP-
AES, H
2
-chemiso p ion, XPS, FTIR, SEM-EDX, XRD, H
2
-TPR, CO
2
-TPD, and NH
3
-TPD. F om he ligh -o ca aly ic
es s, i was ound ha all ca alys s eached he equilib ium CO con e sion in he 350–400 ◦C ange. P doping
in o p is ine Ni/HAp did no enhance nei he ca aly ic ac i i y no selec i i y o hyd ogen. In e es ingly, he -
mochemical ac i a ion (acid ea men ) aimed o educe he CH
4
o ma ion (@ 350 ◦C) by a ound wo- old (25 %
s. 9–14 %) wi h a concomi an inc ease in he hyd ogen yield. Mo eo e , ca aly ic s abili y was also imp o ed.
Fo ins ance, a e 30 h TOS CO con e sion d opped by 50 % o he p is ine Ni/HAp and only 7 % o NiP /SHAp
ca alys . Simila ly, he la e showed he highes and mos s able hyd ogen yield h oughou all he long- e m
es .
In oduc ion
Was e- o-Ene gy has become a leading app oach o sol e was e
disposal p oblems while ha nessing he ene gy con ained in i s lue gases
o elec ici y and hea p oduc ion. The e m “was e- o-ene gy” en-
compasses a la ge a ie y o biomass-based ma e ials ha can be p o-
cessed by di ec combus ion, gasi ica ion, py olysis, o anae obic
diges ion [1]. Bo h gasi ica ion and py olysis gene a e syn hesis gas
(syngas), a uel consis ing mainly o hyd ogen, CO and CO
2
, along wi h
ela i ely small amoun s o CH
4
, N
2
, and o he impu i ies [2]. Wa e -Gas
Shi (WGS) eac ion is essen ial o upg ade con e sion e iciency o his
he mochemical p ocesses, p oducing high-pu i y hyd ogen om was e-
de i ed syngas.
CO(g) + H2O(g)↔CO2(g) + H2(g)(1)
CO(g) + 3H2(g)→CH4(g) + H2O(g)(2)
CO2(g) + 4H2(g)→CH4(g) + 2H2O(g)(3)
C(s) + 2H2(g)↔CH4(g)(4)
2CO(g)↔C(s) + CO2(g)(5)
WGS eac ion (Equa ion (1) is an exo he mic and e e sible eac ion,
which is he modynamically a ou ed a low empe a u e, and kine i-
cally a ou ed a high empe a u e. Side eac ions can also ake place,
leading o he o ma ion o me hane (Equa ion 2–4) and ca bon ia
Boudoua d eac ion (Equa ion (5). Conside ing bo h he modynamic
and kine ic aspec s, he indus ial WGS eac ion occu s in wo s ages,
wi h di e en comme cial ca alys s in each one: high- empe a u e shi
(HTS) a 350–550 ◦C using Fe-C oxide ca alys s and low- empe a u e
shi (LTS) a 190–250 ◦C wi h Cu-ZnO o p ecious me al-based ca a-
lys s [3]. Due o he high concen a ion o CO in was e-de i ed syngas,
cu en ca alys s o he HTS eac ion a e no sui able [4]. Fu he mo e,
i is i al o ind al e na i es o eplace ch omium owing o i s high
human oxici y and nega i e en i onmen al impac [5,6].
Nickel-based ca alys s cons i u e an economically a ac i e al e -
na i e wi h almos compa able WGS ac i i y wi h p ecious me al-based
ca alys [7]. Nickel demons a es unique pe o mance in he WGS
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (A.J. Reynoso).
Con en s lis s a ailable a ScienceDi ec
Jou nal o Indus ial and Enginee ing Chemis y
jou nal homepage: www.else ie .com/loca e/jiec
h ps://doi.o g/10.1016/j.jiec.2023.12.003
Recei ed 30 Sep embe 2023; Recei ed in e ised o m 20 No embe 2023; Accep ed 2 Decembe 2023
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
2
eac ion, he e o e is commonly u ilized in p ocesses like S eam
Re o ming, whe e WGS eac ion plays a pi o al ole [8–10]. The nickel
ca alys s exhibi excellen ac i i y in he high- empe a u e WGS eac ion
and ou pe o m o he ansi ion me al ca alys s, including Co, Cu, and
Fe, by achie ing high CO con e sion and hyd ogen selec i i y [11,12].
Howe e , nickel-based ca alys s usually ha e as majo d awback o he
WGS p ocess he occu ence o me hana ion side eac ions which educe
selec i i y and yield o hyd ogen [13]. Mo eo e , when he eed o he
ca aly ic eac o con ains hyd ogen, as usually occu s in he e o me
ou le s eam, he p oduc ion o he undesi ed me hane is inc eased. A
high empe a u e, ca alys deac i a ion by sin e ing is ano he limi a-
ion o nickel-based ca alys s [14]. Se e al s a egies such as, size
con ol o he me al nanopa icles, con ol o he mo phology and po e
s uc u e, c ea ion o oxygen de ec s o uning su ace acidi y/basici y
ha e been used o he de elopmen o e icien , selec i e and s able
ca alys s in he WGS eac ion [15]. Doping wi h small concen a ion o
noble me als like o P o he o ma ion o me al–me al in e ace ac i e
si es can imp o e he esis an o coking and sin e ing [16,17].
The cha ac e is ics o he ca alys suppo play an impo an ole.
Indeed, i is known ha s ong suppo -me al in e ac ion can educe
sin e ing o he ac i e si es du ing high empe a u e WGS eac ion [18],
whe eas high basici y o suppo can a ou ably ac i a e wa e molecule
[19]. Mo eo e , he p e alence o weak acid and basic si es has been
ela ed o limi ed me hana ion [19–21]. Among he di e en s a egies
o su ace acid-base uning, a mo e acile and sus ainable app oach
could be he use as suppo o bone cha which con ains he e oa oms (Cl,
Mg, Na, e c.) [22]. Hyd oxyapa i e (HAp), eadily and cheaply a ailable
om solid animal was e, has ga ne ed a en ion as a ca alys suppo
owing o i s high he mal s abili y, abundan oxygen acancies, and
adjus able acid and basic si es [23]. I is a compac assemblage whe e
Ca
2+
ions and e ahed al [PO
4
]
3−
g oups delimi wo unconnec ed
channels. The i s one (2.5 Å in diame e ) con ains Ca
2+
(I) ions su -
ounded by oxygen a oms. The second ype has a diame e o 3.5 Å and
hos s he hyd oxyl g oups and Ca
2+
(II) ca ions. The s uc u e o apa i e
is e y lexible and can be eplaced by a la ge numbe o o he me al ions
h ough ion exchange, wi hou a ec ing he c ys alline s uc u e [24].
Se e al s udies ha e epo ed on he ou s anding pe o mance o WGS
on HAp-suppo ed ca alys s [25,26]. The sui abili y o HAp o WGS
eac ions may be c edi ed o he supe io ac i a ion o wa e molecules
on Lewis acidic calcium ca ion and H-bonding o basic oxygen a oms
om phospha e ion. In addi ion, HAp suppo ensu es ha ac i e si es
emain highly esis an o sin e ing [27].
In his wo k, a ca alys wi h Ni suppo ed on o hyd oxyapa i e o
na u al o igin he mochemically ac i a ed was employed. The ion ex-
change abili y o he HAp was employed o alle ia e he s abili y issue.
As a esul , de elopmen o an en i onmen ally sus ainable s a egy o
p omo e WGS ac i i y and selec i i y o hyd ogen is he main challenge
o his wo k.
The na u al apa i e was syn hesized by a one-s ep he mochemical
ac i a ion o was e animal bone, using ei he H
2
SO
4
o K
2
CO
3
, and hen
Ni was loaded by we imp egna ion. P was added as a second ac i e
me al o p omo e he WGS pe o mance. The p omo ion e ec o P
doping on he chemically ac i a ed and non-ac i a ed bone cha was
s udied in dep h. Thus, his s udy explo ed o he i s ime he po en ial
o he combina ion o Ni-P syne gy o e he mochemically ac i a ed
hyd oxyapa i e suppo o he e icien and clean p oduc ion o
hyd ogen.
Expe imen al
Ca alys syn hesis
The suppo was p epa ed om po k chop bones collec ed om a
local bu che shop. Fi s , bones we e cleaned om mea , cu in o pieces
o 2–5 cm and p ecalcined in ai low (120 NmL/min), hea ing a 5 ◦C/
min o 500 ◦C and hold o 2 h. The ob ained biocha was sie ed and
pa icles in he 0.09–0.25 mm size ange we e selec ed. This ma e ial
will be e e ed as p ecu so (HAp). The p ecu so was modi ied by
chemical ea men by ei he H
2
SO
4
(S) (a 0.2 mmol
H2SO4
/g) o K
2
CO
3
(K) (a 5 mmol
K2CO3
/g). Suppo s we e ea ed o e nigh , il e ed and
hen hea ed in ai low a 550 ◦C o 2 h. These ac i a ion condi ions
we e se based on p e ious expe ience [28].
Ac i a ed suppo s we e loaded wi h Ni and P (10 w % Ni; 1 w % P
loadings) by successi e imp egna ion me hod, using aqueous solu ions
o Ni(NO
3
)
2
⋅6H
2
O and P (NH
3
)
4
(NO
3
)
2
. The p ecu so was i s ly
imp egna ed wi h he solu ion o Ni, d ied o e nigh a 110 ◦C. Fo he
bime allic ca alys s, his p ocedu e was epea ed o he deposi ion o
he P . Finally, he solids we e calcined in ai low a 450 ◦C o 2 h. As a
e e ence, monome allic (Ni/HAp) and bime allic (NiP /HAp) was p e-
pa ed using p is ine HAp suppo . O e all ou ca alys s o mula ions
we e p epa ed (Ni/HAp, NiP /HAp, NiP /KHAp, NiP /SHAp). Fu he
in o ma ion on eagen chemicals and ca alys p epa a ion (Scheme S1)
is p o ided in he ESI.
Ca alys cha ac e isa ion
The bulk chemical composi ion o he solids was e alua ed by ICP-
AES (Ho iba Yobin Y on Ac i a). BET me hod was used o de e mine
he speci ic su ace a ea, while he a e age po e size, o al po e olume
and po e size dis ibu ion (PSD) we e es ima ed based on he BJH
me hod om he N
2
deso p ion b anch (analysis ca ied ou in a T is a
II 3020 equipmen ).
The c ys alline s uc u es and he a ia ion in he uni cell di-
mensions o he ca alys s was analysed by XRD (Philips PW1710
di ac ome e ) using he ICDD da abase o he iden i ica ion o he
phases. The Rie eld me hod was used o mic os uc u e analysis
(p o ile e inemen ) and o he es ima ion o he uni cell pa ame e s.
The c ys allini y o he solids was calcula ed om he ull wid h a hal
maximum o (002) e lec ion o hyd oxyapa i e [29].
The iden i ica ion o he educible species and hei in e ac ion wi h
suppo we e examined by empe a u e-p og ammed educ ion (H
2
-
TPR) (Mic ome i ics Au ochem 2920 appa a us). The exposed me al
su ace a ea was de e mined by H
2
chemiso p ion in a Mic ome i ics
ASAP 2020. Su ace unc ionali ies we e analysed by Fou ie ans o m
in a ed spec oscopy (FTIR) ansmission spec a using disks o samples
dilu ed in KB (Nicole P o ege 460). The oxida ion s a e o su ace
me al ac i e si es was s udied by X- ay pho oelec on spec oscopy
(XPS) (Phoibos 150 1D-DLD, monoch oma ized Al Ka, 1486.7 eV, X- ay
adia ion) in bo h calcined and educed o m o ca alys s. The su ace
acid and basic si e densi y was measu ed by ammonia/ca bon dioxide
pulse chemiso p ion and subsequen NH
3
-TPD and CO
2
-TPD (Mic o-
me i ics Au ochem 2920). The ca alys su ace mo phology and me al
dispe sion was analysed by Scanning elec on mic oscopy coupled wi h
ene gy dispe si e spec oscopy (FEG-SEM-EDX) on a JEOL JSM-7000F
appa a us. Mo e de ails on analy ical me hods a e gi en in ESI.
Ca aly ic pe o mance
The WGS pe o mance o he ca alys was s udied in a down low
ixed bed s ainless s eel eac o (D
i
=13.3 mm; leng h =305 mm)
(Scheme S2, ESI). P io o eac ion, he ca alys (0.1 g dilu ed in silicon
ca bide pa icles o he same size, o achie e 1 mL olume) was in-si u
educed unde 15 % H
2
/He low a 400 ◦C o 1 h. Then, he eac o
was cooled down o 200 ◦C unde He low and he eed s eam was
in oduced. Deionized wa e was supplied by HPLC pump (Gilson 307),
and apo ized a 150 ◦C be o e mixing wi h eac an gases. Ca alys
ac i i y es s we e ca ied ou a a mosphe ic p essu e in he 200–450 ◦C
ange, wi h a o al low a e o 200 mL/min STP (GHSV =120,000 h
−1
).
The eed composi ion was CO/H
2
O/CO
2
/H
2
/He =5/46/4/31/14,
ep esen a i e o a ealis ic e o me ou le s eam. A Pel ie de ice was
used o condense he un eac ed wa e om he eac o ou le s eam
be o e ou ing he non-condensable p oduc s o an
μ
GC (Agilen 490,
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
3
wi h wo molecula sie es wi h He and A ca ie s, espec i ely, and one
PPQ column wi h He ca ie ) equipped wi h a TCD de ec o . Samples
we e collec ed once s eady s a e a each empe a u e was a ained.
S abili y es s (du a ion 24 h) we e ca ied ou a isocon e sional con-
di ions (X
CO
=70–80 %, T =325 ◦C o p is ine ca alys s; 350 ◦C o he
chemically ea ed samples).
The CO con e sion and yields (Y
i
) o H
2
and CH
4
we e calcula ed
acco ding o equa ions (1)-(3):
XCO(%) = 100 ×FCO,in −FCO,ou
FCO,ou
(6)
YH2(%) = 100 ×FH2,ou
FCO,in
(7)
YCH4(%) = 100 ×FCH4,ou
FCO,in +FCO2,in
(8)
C selec i i y was de ined as:
Cselec i i y(%) = 100 ×FC,in −FC,ou
FC,in
(9)
Resul s and discussion
Chemical composi ion and ex u al p ope ies
Table 1 shows ICP da a and ex u al and mo phological esul s. The
Ni and P con en was close o heo e ical alue in he ba e suppo
whe eas i was sligh ly lowe in he chemically ea ed assays. The ba e
suppo (HAp) had a mola Ca/P a io o 1.64, sligh ly lowe han he
s oichiome ic alue o pu e hyd oxyapa i e (Ca/P =1.67) because o
he na u al appea ance o Ca-de icien phases such as β- icalcium
phospha e [30]. Acid- ea ed sample showed he highes de ia ion and
he lowes Ca/P a io o 1.61. I is well known he capaci y o apa i es o
eplace Ca
2+
by o he ca ions [31]. This esul sugges ed ha Ca ions in
he apa i e we e mo e eadily subs i u ed by o he ca ions and/o
leached ou a e acid ea men o bone cha . High acidi y could a ou
local su ace dissolu ion o he na u al apa i e, and hus enhance he
ca ionic exchange be ween Ca
2+
and Ni
2+
/P δ
+
.
The N
2
adso p ion–deso p ion iso he ms o all ca alys s (Figu e S1,
ESI) showed a simila ypology o iso he m and hys e esis loop (Type II,
H3) which sugges ed a pla e-like laye ed s uc u e wi h open po es on
he mesopo e and mac opo e scale. The dis ibu ion o po es ob ained
om he analysis o hese iso he ms a e epo ed in Table 1 and
Figu e S1b (ESI). P -doping in he e e ence ca alys (Ni/HAp) ha dly
modi ied he PSD wi h a esul an S
BET
o a ound 62 m
2
/g, a po e olume
o 0.28 cm
3
/g, and a e age po e size o 15.4 nm. The chemical ea men
o he suppo can signi ican ly a ec i s ex u al p ope ies [32]. As
obse ed, he S
BET
o he de eloped ca alys s dec eased (by 26–39 %),
and he main peak o po e size dis ibu ions signi ican ly upshi ed ( om
c.a. 15 nm o c.a. 26–29 nm), indica ing a po e widening in he meso-
po ous ange. In e es ingly, a e ca alys educ ion in hyd ogen low,
he S
BET
and po e olume (Table 1) inc eased by a ound 12–20 % o he
non-chemically ac i a ed ca alys s (Ni/HAp, NiP /HAp), howe e , i
inc eased by oughly 5 % o he chemically ac i a ed ca alys s (NiP /
KHAp and NiP /KSAp). These esul s showed ha he applied chemical
ea men con e ed mo phological s abili y o he de eloped ca alys s,
since ex u al p ope ies ha dly a ied upon educ ion in hyd ogen low.
S udy o he su ace mo phological cha ac e is ics
The c ys alline s uc u e o he suppo and Ni pa icles was in es-
iga ed using XRD. The di ac og ams o he eshly calcined, educed
and exhaus ed samples a e epo ed in Fig. 1. I can be obse ed ha
cha ac e is ic peaks o apa i e phase we e p ominen in he ba e suppo
(p ecu so ) and in all he p epa ed ca alys s a a ound 2θ o 26.0, 32.0,
and 33.0◦. Cha ac e is ic peaks o NiO a 37.2◦and 43.3◦we e obse ed
in he calcined o m o ca alys s. No peaks due o P o P Ox phases we e
de ec ed, likely due o low amoun s o P (0.4–0.8 w %) and below
de ec ion limi size o nanopa icles. P -doping in o he e e ence ca a-
lys did no a ec he posi ion o he HAp di ac ion peaks (Fig. 1a). Fo
Table 1
Chemical, ex u al and s uc u al p ope ies o he ca alys s.
Ca alys Me al loading
a
(w .%) Ca/P
b
S
BET c
(m
2
/g) V
P
(cm
3
/g) d
P
(nm) d
HAp
d
(nm) d
NiO
e
(nm) d
Ni
(nm) Spen ca alys s
P Ni d
NiO(111) g
(nm) d
Ni
o g
(nm) (111) (200)
Ni/HAp n.a. 9.6 1.64 61.8 (69.0) 0.289 (0.323) 16.1 (17.1) 9.2 (10.1) 7 4 5 13 5
NiP /HAp 0.8 9.1 1.64 62.1 (75.8) 0.275 (0.329) 15.4 (15.3) 10.3 (9.8) 8 3 7 6 9
NiP /KHAp 0.5 7.7 1.65 37.8 (39.2) 0.288 (0.302) 28.6 (31.1) 16.5 (20.7) 12 7 6 4 10
NiP /SHAp 0.8 7.9 1.61 45.5 (48.3) 0.321 (0.343) 26.1 (26.9) 14.3 (9.7) 46 10 12 5 13
In pa en hesis alues measu ed a e educ ion a 400 ◦C; n.a.: no analysed;
a
om ICP-AES;
b
mol/mol;
c
om ni ogen iso he ms;
d,e
om XRD (hkl);
om XRD, plane
(111) o he educed samples;
g
om XRD o used ca alys s in long e m WGS eac ion
Fig. 1. XRD pa e ns o bone cha suppo ed ca alys s (a) Calcined, (b)
Reduced, and (c) Spen ca alys .
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
4
ins ance, he peaks co esponding o he (112) and (300) e icula
planes, a 32.07◦and 33.00◦ o he p ecu so and ca alys s Ni/HAp and
NiP /HAp ha dly a ied.
The 2θ alues ob ained by he Rie eld e inemen (Table S1, ESI)
we e sligh ly highe han hose o s oichiome ic hyd oxyapa i e (ICDD
01–84-1997, Ca/P =1.67, a =b =9.4180, c =6.8840). Because o he
na u al o igin and he coexis ence o o he he e oa oms (Table S2, ESI)
in ou samples, which p oduced a calcium-de icien hyd oxyapa i e (Ca/
P =1.61–1.64), he la ice pa ame e s measu ed o ou na u ally
de i ed HAp (a =b =9.40809, c =6.8783) we e smalle han pu e
hyd oxyapa i e. This sugges ed ha smalle a oms (i.e. Na, Mg) we e
mo e eadily exchanged wi h Ca
2+
.
The la ice pa ame e s and cell olume emained simila a e
loading o P (Table S1, ESI). The e o e, i seems ha he P species
mainly sp ead on he HAp su ace wi h e y li le inco po a ion in o he
HAp s uc u e. A e educ ion a 400 ◦C, a sligh dec ease in he HAp
cell pa ame e s o Ni/HAp and NiP /HAp was obse ed which caused a
dec ease in he apa i e la ice olume o a ound 0.25 % ( om 528.7 o
527.5 Å
3
).
Rega ding he ca alys s suppo ed on he chemically ea ed apa i e,
hei XRD di ac og ams showed a clea displacemen in he posi ion o
2θ co esponding o (112) and (300) planes o HAp. Simila ly, he
cha ac e is ic di ac ion peaks o he NiO phase we e also a ec ed.
Mo eo e , la ice expansion was highe han ha obse ed o he
e e ence ca alys (Table S1, ESI), suppo ing he pa ial inco po a ion
o hese ca ions in o he apa i e amewo k [27,33]. This obse a ions
poin ou ha he chemical ea men modi ied he cell pa ame e s o
he na u ally ob ained apa i e, and also he in e ac ion be ween he
inco po a ed Ni, and likely P , species wi h he suppo . In e es ingly,
a e educ ion he cell pa ame e o he chemically ea ed suppo s
a ied o a lesse ex en as compa ed o he non- ea ed assays. The e-
o e, in line wi h obse a ion o he ex u al da a, i seemed ha
chemical ac i a ion enhanced he chemical s abili y o he na u al
apa i e suppo ed ca alys s.
The Sche e ’s equa ion was applied o XRD diag ams o es ima e
c ys al sizes o he suppo ed species. As can be deduced om da a in
Table 1, NiO c ys alli es we e de ec ed on he su ace o he e e ence
ca alys . No e ha hese c ys alli es we e longe in ho izon al g ow h
di ec ion han e ical (i.e. 7 nm s. 4 nm). Also, alues o ca alys NiP /
HAp e lec ed ha P doping did no a ec g owing o NiO c ys alli es.
Rega ding he chemically ea ed samples, o ins ance, ca alys NiP /
KHAp con ained NiO c ys alli es o la ge size (12 nm x 7 nm) as
compa ed o he e e ence ca alys Ni/HAp. In e es ingly, he la ges
NiO pa icles we e o med on he su ace o he acid ea ed sample
(NiP /SHAp), wi h a pa icula ly enhanced g owing in he ho izon al
plane (46 nm x 10 nm). A e educ ion in hyd ogen low, no cha ac-
e is ic XRD peaks o NiO we e obse ed and a new peak, cha ac e is ic
o me allic Ni, appea ed a a ound 2θ =43.5 ◦. The a e age c ys al size
o me allic nickel was o a ound 5–7 nm in ca alys s Ni/HAp, NiP /HAp
and NiP /KHAp. Howe e , ca alys NiP /SHAp con ained Ni
0
c ys alli es
o a ound 12 nm. Thus, i seemed ha he ea men o he na u al HAp
wi h acid a ec ed g owing geome y o deposi ed NiO pa icles, wi h
p e e en ial ho izon al g owing, which p oduced he la ges Ni
0
pa i-
cles upon educ ion (i.e. wo- old inc ease in size). This end sugges s
ha he la ge size o me allic Ni pa icles is mos p obably due o di -
e ences in he ype o in e ac ion Ni-P -HAp compa ed wi h ha in he
non- ea ed suppo . As will be discussed in he educibili y sec ion,
ca alys NiP /SHAp con ained a b oad peak a he highes empe a u e
due o deposi ion o educible Ni-P species wi h a ela i ely s ong
in e ac ion wi h suppo .
SEM images o he esh ca alys s (Figu e S2, ESI) showed ha he
mo phology a mic ome e scale consis s o a combina ion o smoo h
and ough su aces and olds appea on he su ace oge he wi h
g anula ma e ial. The dis ibu ion o he di e en elemen s on he
su ace o esh ca alys s was pe o med by EDX do mapping images
(Figu e S3, ESI), in which a highe concen a ion o species appea s
b igh e . This e ealed ha he me allic species we e well dis ibu ed on
he su ace o he bone cha suppo . Mo eo e , pa icles de ec ed by
EDX analysis o he bime allic ca alys demons a ed he p esence o Ni,
O and P en iched egions. In e es ingly, he coexis ence o Mg can be
also obse ed. The deposi ion o nickel pa icles could p omo e he
o ma ion o pa icles o he second ac i e me al. The supe imposi ion o
loaded me als on o hyd oxyapa i e was also obse ed by o he s [34].
Con a ily, as can be obse ed, he ac i e me als a e loca ed in Ca, P and
C lean a eas.
Reducibili y and me allic unc ion
The e e ence ca alys showed he highes amoun o exposed Ni
a oms on he su ace (Table 2). The me allic si es densi y dec eased by
a ound 25 % a e P doping, om 1.12 o 0.79 si es pe nm
2
. The sample
ac i a ed wi h he alkali showed a simila me allic si es densi y. How-
e e , a signi ican dec ease was obse ed in he acid ea ed ca alys (i.e.
0.41 me al si es pe nm
2
). This esul is ully consis en wi h he a -
ou ed g ow h o he me allic Ni nanopa icles measu ed by XRD o
ca alys NiP /SHAp.
Fig. 2 shows he H
2
-TPR p o iles o he calcined ca alys s. The
educ ion p o ile o he suppo showed no educ ion peak a below
500 ◦C, which was he maximum hea ing empe a u e used o ob ain he
na u al apa i e used as suppo . Hyd ogen consump ion a abo e 500 ◦C
could be asc ibed o apa i e dehyd oxyla ion eac ions [35]. The H
2
-TPR
p o ile o he e e ence ca alys Ni/HAP could be di e en ia ed in o six
hyd ogen consump ion peaks. The low empe a u e peaks cen ed a
185 ◦C and 285 ◦C can be assigned o educ ion o NiO pa icles wi h
weak in e ac ion wi h he suppo [36]. The p ominen peak cen ed a
360 ◦C was assigned o he educ ion o nickel species ha ing mild
in e ac ion wi h he suppo (i.e. NiO pa icles iden i ied by XRD). Upon
in eg a ion o hese h ee peaks a H
2
/Me mola a io o 0.57 was
measu ed (Table 3). Thus, in line wi h XRD and XPS da a, he pa ial
educ ion o he NiO species was e idenced a e educ ion in hyd ogen
low a 400 ◦C. No e ha he o e all hyd ogen consump ion up o 500 ◦C
( he suppo s abiliza ion empe a u e) was e y close o he s oichio-
me ic alue (H
2
/Me =0.997). The hyd ogen consump ion peaks a
highe empe a u es could no be unambiguously assigned. P e ious
epo s on nickel suppo ed on o hyd oxyapa i e sugges ed he educ-
ion o : (i) NiOx species wi h s ong in e ac ion wi h he suppo ; (ii) ion
exchanged Ni
2+
in he Ca
2+
si es o he apa i e; and (iii) he dehyd ox-
yla ion o he apa i ic suppo [37,38]. The abo e men ioned s oichio-
me ic consump ion, sugges ed ha he la e p ocess was he mos
p obable.
The p o ile o he P -doped ca alys (NiP /HAp) was compa able
wi h ha o he e e ence ca alys wi h wo main educ ion e en s, being
he mos p ominen ha a he highe empe a u e (320 ◦C). Howe e ,
he peaks appea ed shi ed o lowe empe a u es (i.e. 230/320 s. 285/
360 ◦C). The in eg a ion alues summa ised in Table 3, e ealed a H
2
/P
alue o 11, wha indica es he educ ion o all su ace P Ox species wi h
weak in e ac ion wi h he suppo . Boukha e al. [27] epo ed ha P
could be inco po a ed in o he HAp s uc u e, howe e , in ou na u ally
de i ed apa i e, his beha iou seemed unlikely. XRD da a showed no
modi ica ion o cell pa ame e s upon P loading. Mo eo e , he abo e
men ioned H
2
/P mole a io sugges ed ha educ ion o addi ional
species ook place. Se e al au ho s ha e epo ed ha he educ ion o
bulk nickel oxide can occu a lowe empe a u es in he p esence o P ,
as due o he spillo e on me allic P [39–41]. As obse ed om SEM
images, P and Ni we e deposi ed on ca alys su ace in close in e ac ion.
Thus, i seemed likely he occu ence o hyd ogen spillo e om P o
he adjacen Ni pa icles wha would acili a e i s educ ion. Fu he -
mo e, he educ ion o oxygen unc ionali ies on he suppo i sel could
no be disca ded based on he la ge han s oichiome ic hyd ogen
consump ion o hese ca alys s (H
2
/Me =1.12–1.52) [42].
The ac i a ion o he apa i ic suppo wi h acid did no signi ican ly
a ec he posi ion o he low empe a u e peak (c.a. 230 ◦C) as
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
5
compa ed o he non-chemically ea ed bime allic ca alys . Mo eo e ,
he amoun o species educible a T <300 ◦C emained simila (up ake
0.42 s. 0.50 mmol
H2
/g). The esul ing H
2
/P mole a io o he la e
samples was 7 and 11, espec i ely. Thus, he concomi an educ ion o
P and Ni species was likely o occu . Indeed, he dec ease in he
educ ion empe a u e o he nickel species could be asc ibed o P Ox-
NiO in e ac ions.
The educ ion p o ile o he sample ac i a ed wi h alkali (NiP /
KHAp) was he mos dissimila among he p epa ed ca alys s. I showed
an in ense peak a 185 ◦C, which e lec ed he exis ence o an inc eased
amoun o easily educible species on he ca alys su ace. As o he
p e ious ca alys s, he concomi an educ ion o P -Ni species is likely o
occu . Howe e , he H
2
/P mole a io (H
2
/P =24) was no ably
inc eased. This sugges ed an abundance o o he species, apa om he
loaded me als, exis ing on he ca alys su ace as a consequence o he
alkali ea men , ha we e also educed. Indeed, addi ional hyd ogen
spillo e om P o he suppo su ace could ake place. The o al H
2
/
Me mole a io was highes o his ca alys (1.52), wha suppo s his
scena io.
In summa y, om H
2
-TPR da a i can be concluded ha P p omo ion
signi ican ly inc eased he educibili y o Ni
δ+
species (low empe a u e
H
2
consump ion inc eased wo- old om 0.2 mmol
H2
/g in he e e ence
ca alys o 0.42–0.49 in NiP /HAP and NiP /SHAp, espec i ely). H
2
consump ion in ca alys NiP /KHAp occu ed a he lowes empe a u e
(185 ◦C), likely due o ca bona es deposi ion on ca alys su ace a e he
alkali ac i a ion. In e es ingly, he acid ea ed sample, NiP /SHAp,
showed a b oad mul ipeak ea u e in he 350–390 ◦C ange wha would
imply ha he NiO species in close in e ac ion wi h he suppo
exhibi ed a wide size dis ibu ion wi h a la ge a e age size, in acco -
dance wi h XRD da a.
S uc u e and coo dina ion s a e
The analysis by FTIR (Figu e S4, ESI) showed he cha ac e is ic
bands o phospha e ion a 1093, 1035, 961, 603, 563, and 473 cm
−1
.
Table S3 (ESI) lis s he wa e numbe s and hei co esponding ib a ion
modes o he di e en species ound in he samples. The bands a
a ound 3440 and 1637 cm
−1
we e assigned o he bending mode o he
adso bed wa e , whe eas he peaks a 3570 cm
−1
and 633 cm
−1
, co -
esponding o OH bands o HAp, we e clea ly isible o he ea ed
samples. The exis ence o ca bona es (1461, 1415, 1385, and 870 cm
−1
)
was con i med in all samples. An un esol ed weak double ep esen a-
i e o A- ype (hyd oxyl subs i u ed ca bona e ions) a 880 cm
−1
[43]
and B- ype (phospha e subs i u ed ca bona e ions) a 870 cm
−1
[44] in
he HAp s uc u e was obse ed in ou samples. Mo eo e , he alkali
ea ed sample showed he mos in ense signal a a ound 1385 cm
−1
which has been associa ed o he ans o ma ion o ee ca bona es in o
Table 2
Su ace chemical p ope ies and kine ic pa ame e s o he ca alys s.
Ca alys Accesible me allic si es
a
(a
Me
./g)(×10
-18
)
Acid si es densi y
b
(si es/nm
2
)
Basic si es densi y
b
(si es/nm
2
)
Basic/acid
si e a io
Su ace uniden a e
ca bona e o CO
3
2– c
T
50
d
(◦C)
CO e
(µmol
CO
/g/
s) @275 ◦C
TOF (s
−1
)
@275 ◦C
Ni/HAp 77.1 0.93 0.57 0.61 0.12 316 4.0 0.03
NiP /
HAp
59.8 1.17 0.51 0.43 0.11 300 6.7 0.07
NiP /
KHAp
54.2 0.37 0.82 2.19 0.27 325 4.0 0.04
NiP /
SHAp
19.7 0.90 0.48 0.53 0.08 335 3.2 0.10
a
H
2
pulse chemiso p ion;
b
Tempe a u e-p og ammed deso p ion (NH
3
-TPD; CO
2
-TPD).
c
om FTIR spec a peak in eg a ion a 1385 cm
−1
s 1415–1461 cm
−1
;
d
Tempe a u e o each 50 % con e sion o CO;
e
WGS eac ion a isocon e sion condi ions (75 % CO con e sion @325–350 ◦C).
Fig. 2. H
2
-TPR p o iles o esh calcined ca alys s.
Table 3
H
2
-TPR da a o he ca alys s.
Ca alys H
2
up ake (mmol
H2
/g) T (◦C)
T <500
a
(H
2
/Me) T <300
b
(H
2
/P ) (H
2
/Me)
Ni/HAp 1.63 (0.997) 0.20 (n.a.) 0.07 (0.04)
0.13
(0.08)0.73
(0.45)
185
285
360
NiP /HAp 2.18 (1.41) 0.50 (11) 0.50 (0.31)
1.28
(0.81)
230
320
NiP /
KHAp
2.00 (1.52) 0.93 (24) 0.93 (0.69)
0.77
(0.57)
185
360
NiP /SHAp 1.50 (1.12) 0.42 (7) 0.42 (0.30)
0.45
(0.33)0.56
(0.40)
230
350
390
c
Likely, includes in ense dehyd oxyla ion.
a
In b acke s, H
2
/Me a io. Calcula ed using T =500 ◦C in eg a ion cu -o
me hod by ini e inc emen s.
b
Cacula ed by in eg a ion o decon olu ion peaks up o T ◦C.
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
6
su ace uniden a e ca bona es [45]. F om he FTIR spec a, he amoun
o uniden a e su ace ca bona es was 2.3 imes highe o NiP /KHAp as
compa ed o he e e ence ca alys . I is likely ha he educ ion o such
species would con ibu e o he obse ed p ominen educ ion peak a
a ound 185 ◦C, discussed in H
2
-TPR.
XPS was used o s udy he chemical composi ion and he oxida ion
s a e o elemen s on he ca alys s su ace. The su ey spec um o he
calcined ca alys showed ha O, Ca, P and Ni we e he main elemen s.
The p esence o C was e idenced by peaks a BE alues cha ac e is ic o
alipha ic 284.6 eV (C–H, C–C) and a oma ic 289.1 eV (O-C =O)
ca bonaceous ma e ial (Table S4, ESI). I accoun ed o a ound 4 a .%,
excep o he alkali ea ed sample, which con ained a signi ican ly
la ge amoun o C (i.e. 8 a .%), mainly due o alipha ic ca bon. Con en
o K and Na was below 2 a .%. Some Mg 2 s could be iden i ied, hough
close o me hod de ec ion limi . P was iden i ied in he su ace o doped
samples a concen a ions o a ound 0.05 a .%. In he educed samples,
all same elemen s could be iden i ied on ca alys su ace, wi h he
excep ion o C, which was emo ed upon educ ion. The su ace Ca/P
Fig. 3. XPS spec a o Ni 2p and P 4 o (a,b) calcined and (c,d) educed ca alys s.
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
7
a omic a io (1.3–1.4, Table S4, ESI) was no ably lowe han he bulk
alue (1.6) wha sugges ed a lowe amoun o exposed Ca a oms in
su ace. These esul s sugges ed ha he loaded Ni and P could be
deposi ed p e e ably o e he su ace Ca si es.
The high- esolu ion XPS spec a we e used o analyse nickel and
pla inum specia ion. Fig. 3a shows he Ni 2p
3/2
spec a o he calcined
ca alys s. Fo he e e ence ca alys , he main pho oemission peak can be
decon olu ed in o wo peak a 853.3 and 855.3–856.2 eV and a b oad
sa elli e s uc u e appea ing a a ound 860 eV. In ag eemen wi h p e-
ious s udies hese a e consis en wi h he p esence o NiO pa icles wi h
weak and s ong in e ac ion wi h he suppo , espec i ely. The low BE
peak can be a ibu ed o he binding ene gy o Ni 2p
3/2
in NiO dispe sed
on he phospha e [46]. The second peak a a ound 855.7 eV was
asc ibed o NiO pa icles wi h a s onge in e ac ion wi h he suppo
[47]. On he o he hand, Boukha e al. [38] obse ed an analogous peak
a highe binding ene gies (856.9 ±0.3 eV) which was asc ibed o ion
exchanged Ni
2+
in he Ca
2+
(I) and Ca
2+
(II) si es o he apa i e. In he
spec a o he NiP /KHAp and NiP /SHAp ca alys s he highe BE peak
appea s a 856.0–856.2 which migh be associa ed o he pa ial ex-
change o he Ni
2+
a oms in he apa i e s uc u e.
Fig. 3b displays he P 4 co e le els spec a o he calcined ca alys s.
Fo he P p omo ed ca alys (NiP /HAp), he P 4
7/2
and P 4
5/2
a e
obse ed a a ound 73.7 and 77.1 eV, espec i ely. In he chemically
ac i a ed ca alys s (NiP /KHAp and NiP /SHAp) he P 4
7/2
peak
appea ed a highe BE alues (73.9–74.1 eV). Binding alues a a ound
72.4–72.8 eV o P 4
7/2
ha e been assigned o P
2+
species, whe eas
alues in he 73.7–74.3 eV ha e been asc ibed o P
4+
species deposi ed
on o syn he ic hyd oxyapa i e [48]. Thus, he ob ained XPS da a sug-
ges ed a mo e oxidized s a e o P
δ+
species (i.e. P O
2
species) in he
chemically ac i a ed samples. Mo eo e , he la ge FWHM alue co -
esponding o he P 4
7/2
peak o he acid ac i a ed ca alys s indica ed a
highe he e ogenei y in P en i onmen s. The coexis ence o ace
amoun s o me allic P should no be disca ded [49].
In he educed samples, he XPS spec a o Ni 2p
3/2
(Fig. 3c) showed
a new and in ense peak a a ound 852.4 eV, cha ac e is ic o me allic Ni,
and ano he wide band in he 856.4–857.1 eV ange, which deno es he
exis ence o Ni
δ+
species wi h s ong in e ac ion wi h he suppo o ion-
exchanged. These esul s e idenced ha Ni was pa ially educed du ing
he ac i a ion unde hyd ogen low. On one hand, i can be deduced ha
chemical ac i a ion did no induce signi ican modi ica ion on he
chemical en i onmen o he me allic Ni suppo ed on o he ca alys
su ace (simila BE alue o 852.4 eV o all ca alys s). On he o he
hand, he exis ence o some Ni
δ+
species in he educed ca alys could be
asc ibed o he exis ence o Ni pa icles in close in e ac ion wi h he
apa i e suppo which we e mo e di icul o educe. The nanome ic
size o hese pa icles would explain he absence in he XRD spec a. The
XPS analysis o he educed samples e idenced ha P was ully educed
o me allic P as all samples exhibi ed P 4
7/2
ea u es a a ound 70.5 eV
[50]. Mo eo e , P p omo ion did no o igina e signi ican al e a ion in
he oxida ion s a e o Ni
δ+
(857.1 eV o Ni/HAp and NiP /HAp sam-
ples). Howe e , in he chemically ac i a ed samples he ea u e o he Ni
species peaked a lowe BE alues compa ed o he non-chemically
ea ed samples (856.4–856.7), wha sugges ed changes in he oxida-
ion s a e o he Ni
δ+
deposi ed on o ca alys su ace.
Oxygen exis ed in wo o ms (Figu e S5, ESI), as la ice oxygen
(~530.3 eV) and hyd oxides o de ec oxide oxygen (~531.3 eV) [51].
In e es ingly, alues summa ised in Table S4 (ESI) indica ed ha
chemical ac i a ion modi ied he ela i e a omic amoun o oxygen
species, wi h a signi ican inc ease ( om 53 o 56 a .% o 63–64 a .%) in
oxygen de ec s.
Acid-base si es o he ca alys s
Table 2 assembles he da a ob ained o NH
3
and CO
2
pulse chemi-
so p ion o all he educed ca alys s. A e P -imp egna ion, he o al
numbe o acidic si es inc eased by a ound 25 % when compa ed o Ni/
HAp. The sample ea ed wi h K
2
CO
3
displayed he lowes o al su ace
acidi y, as could be expec ed. In e es ingly, despi e ha ing used a s ong
acid, he acid ac i a ed NiP /SHAp sample p ese ed acid si es densi y
close o ha o he e e ence sample (0.9 s 0.93 si es/nm
2
). The acidi y
pe uni mass o he NiP /SHAp sample compa ed o he e e ence
sample has ac ually dec eased due o acid imp egna ion (72.4 s 106.7
μ
mol
NH3
/g
ca
). This imp egna ion pa ially dissol ed he bioapa i e and
c ea ed a ca ion-de icien s uc u e wi h educed P-OH unc ionali ies,
which a e p ecu so s o acid cen es [32]. The su ace basici y dec eased
wi h P -doping. The change in he numbe o basic si es o he chemi-
cally ea ed samples anged be ween −15.8 % and +43.2 % when
compa ed o he e e ence sample. In gene al, he ca alys s ha e an
ampho e ic na u e wi h acid and basic cen es, wi h a sligh p edomi-
nance o acid cen es o e he basic ones in he un ea ed samples. Only
a e alkaline ea men his a io e e ed (basic/acid =2.19 s.
0.4–0.6) wha could be asc ibed o ca bona es deposi ed on he ca alys
su ace, as in e ed om TPR and FTIR esul s.
Ca aly ic ac i i y
Fig. 4 compa es he WGS pe o mance o he p epa ed ca alys s in
he 200–450 ◦C ange o a ypical e o me ou le s eam mix u e (i.e. in
he p esence o bo h H
2
and CO
2
). I can be obse ed ha he P -
p omo ed ca alys NiP /HAp showed he highes ac i i y, ollowed by
he e e ence ca alys Ni/HAp. The o me ca alys eached 32 % con-
e sion o CO a a ound 300 ◦C and 80 % a 325 ◦C (Fig. 4a). The
he modynamic WGS equilib ium was eached a a ound 350 ◦C by
hese wo ca alys s. Fo he chemically ea ed ca alys s (NiP /SHAp and
NiP /KHAp), he ligh -o cu e was sligh ly displaced o highe em-
pe a u es, eaching WGS equilib ium a 370 ◦C and 390 ◦C, espec i ely.
Fo ins ance, a 325 ◦C, he CO con e sion was 32 % and 51 %,
espec i ely. I he T
50
alues a e compa ed (Table 2), he WGS ac i i y
o he p epa ed ca alys s a ies as ollows: NiP /HAp >Ni/HAp >NiP /
KHAp >NiP /SHAp. The lowe speci ic su ace a ea o he la e ca a-
lys s (35–45 % lowe han hose non-chemically ac i a ed) could pa ly
explain his beha io .
The speci ic ac i i y (
CO
) was measu ed a 275 ◦C in o de o
app oach o di e en ial eac o condi ions (X
CO
<10 %). The mos
ac i e ca alys was NiP /HAp (Table 2, 6.7 µmol
CO
/g/s) which showed
68 % highe ac i i y han he e e ence ca alys (4.0 µmol
CO
/g/s). Da a
om Table 2 e eal ha NiP /HAp con ained he highes me allic si es
(59.8 ×10
18
a
Me al
/g) among he P -doped samples (NiP /HAp, NiP /
KHAp, NiP /SHAp). Howe e , i is wo h ou lining ha he e e ence
ca alys Ni/HAp con ained he highes amoun o exposed si es (77.1 ×
10
18
a
Me al
/g), hough, as p e iously no ed, a lowe speci ic ac i i y. As
e ealed by XPS analyses o he educed ca alys s (Table S4, ESI), NiP /
HAp con ained he maximum con en o me allic Ni on he su ace (39
a .%), and also, mo e su ace oxygen de ec s. I seemed ha he
enhanced educibili y o he P -p omo ed ca alys s caused some nano-
s uc u al modi ica ions (i.e. inc ease oxygen de ec s) in he su ace o
he na u al apa i e which p omo e WGS ac i i y [52]. I is likely ha he
Ni
2+
and P
1+
a oms exchanged in he apa i e amewo k, lea e he
ca ionic si es a e educ ion, and gene a e su ace acancies in he
suppo .
Rega ding he WGS selec i i y, i should be no ed ha me hane was
de ec ed o all ca alys s. The selec i i y o he eac ion o CH
4
and H
2
is
illus a ed in Fig. 4b,c. The me hane and hyd ogen o ma ion p o iles
wi h espec o eac ion empe a u e ob ained by he ca alys s suppo ed
on non-chemically ea ed HAp (Ni/HAp and NiP /HAp) was simila .
Me hane o ma ion began a a ound 275 ◦C and peaked a a ound 450 ◦C
wi h a me hane yield eaching a alue o 40 % by bo h ca alys s.
Hyd ogen yield showed nega i e alues o hese wo ca alys s, ha is,
unde he eac ion condi ions, whe e CO
2
and H
2
we e co- ed, hyd ogen
was a he consumed han gene a ed in he eac o . In he li e a u e, Ni-
based ca alys s we e well epo ed as e icien ca alys s o he e e se
WGS (RWGS) [53,54]. As can be obse ed, he hyd ogen p oduc ion
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
8
yield ollowed a descending end wi h empe a u e. Mo eo e , an
ab up change in slope (mo e nega i e) a abo e 325 ◦C occu ed. I
seemed ha P -doping o Ni ca alys suppo ed on o na u al apa i e did
no signi ican ly modi y he selec i i y o he WGS eac ion. We may
conclude ha P enhanced WGS ac i i y by inc easing he amoun o
accessible Ni
0
si es and, likely by a a ou ed wa e dissocia ion on su -
ace oxygen acancies on he na u al apa i e [55,56].
I is wo h no ing ha he eac ion selec i i y signi ican ly a ied o
he chemically ea ed assays. On one hand, ca alys s NiP /KHAp and
NiP /SHAp we e mo e selec i e o hyd ogen p oduc ion han he o me
ca alys s. The me hane yield was no ably educed by he chemically
ac i a ed ca alys s, wi h a maximum alue, a a ound 425 ◦C (Y
CH4
=20
%), which is 50 % lowe han ha measu ed o he non-chemically
ac i a ed samples (Y
CH4
=40 %). Likewise, by he chemically ea ed
ca alys s, he ne hyd ogen p oduc ion became posi i e in a wide em-
pe a u e ange (Fig. 3c), in con as o he nega i e hyd ogen p oduc ion
by he non- ea ed assays.
The u no e equency (TOF) no malized by su ace ac i e me al
si es (measu ed by H
2
chemiso p ion), a 275 ◦C, a e shown in Table 2.
The TOF was lowes o he e e ence ca alys (0.03 s
−1
) and inc eased
wi h P -doping (TOF =0.07 s
−1
), ha is, he me allic si es in he la e
ca alys a e wo- old mo e ac i e han in he o me . The chemical
ac i a ion o he suppo wi h he alkali caused he educ ion o he TOF
(0.04 s
−1
). In e es ingly, he acid ea ed NiP /SHAp ca alys ga e he
highes TOF alue (0.10 s
−1
), h ee- old mo e ac i e han he e e ence
ca alys . The ela i ely high ac i i y o he me allic si es sp ead on he
su ace on he na u al apa i e could be explained by he la ge pa icle
size and s onge in e ac ion wi h he suppo [57]. Mo eo e , ca alys
NiP /SHAp con ained mo e eac i e si es on he apa i e su ace and also
mo e OH
–
si es (as assessed by FTIR) and mo e oxygen de ec s (as
assessed by XPS). The ole o hyd oxyl g oups on ca alys su aces is
widely ecognised as signi ican o enhancing hyd ogen p oduc ion
e iciency, since hese g oups acili a e he adso p ion and dissocia ion
o wa e , p omo ing eac ion a e and ca alys s abili y [12,58]. The
NiP /KHAp ca alys , howe e , con ained highly dispe sed ca bona e
species. The abundance o such species would imply a close in e ac ion
wi h he Ni-P nanopa icles and likely lowe hei e iciency. This lowe
e iciency seemed o be compensa ed by he highe amoun o exposed
me al si es on he NiP /KHAp ca alys s, which could explain he
inc eased o e all ac i i y (4.0 µmol
CO
/g/s). The ca alys NiP /SHAp
con ained he lowes amoun o ca bona es. I s pe o mance is compa-
able o o he s epo ed in he li e a u e (Table S5, ESI). Thus, he ob-
ained alues e eal ha na u ally ob ained apa i e is a sui able ca aly ic
suppo o he WGS eac ion.
The abo e men ioned low CH
4
and high H
2
selec i i y and he
obse ed bene i in he TOF alue mus be due o he physicochemical
and s uc u al p ope ies o he suppo . As men ioned abo e, alkali and
acid ea men modi ied he p ope ies o he apa i e. As e ealed by he
XRD s udy, c ys al g owing o bo h he apa i e and he NiO phase we e
enhanced. Mo eo e , he g owing geome y o he NiO nanopa icles
was signi ican ly inc eased in he ho izon al plane by he acid ac i a-
ion. I seemed ha he me allic nickel de i ed om he educ ion o
ho izon al ace s migh ha e bene icial e ec s on he selec i i y o H
2
in
he WGS eac ion. Senanayake e al. epo ed a p e e en ial adso p ion
o physiso bed CO in he ho izon al plane wi h espec o he su ace
[59] wha could be ela ed o he abo e obse a ion.
S abili y s udy
The s abili y o he ca alys s was s udied a isocon e sional condi-
ions (ini ial CO con e sion in he 70–80 % ange). Based on he WGS
ligh -o cu es, a empe a u e o 325 ◦C o Ni/HAp and NiP /HAp and
350 ◦C o NiP /KHAp and NiP /SHAp we e chosen, and he WGS e-
ac ion was conduc ed o 30 h ime-on-s eam (TOS). As can be obse ed
in Fig. 5a, he e e ence ca alys showed a s able ca aly ic beha iou
du ing he i s 13 h o TOS. A e wa ds, i su e ed deac i a ion and CO
con e sion d opped-o by 52 % a e 30 h o TOS. A e P doping (NiP /
HAp) ini ial ac i i y d opped by a ound 64 %, hough i seemed o be
mo e esis an wi h an ac i i y loss o 15 % a e 30 h TOS. In e es ingly,
acid ea men did no a ec he CO con e sion end in he long- e m
un. As will be discussed la e , a di e en ou come was obse ed in
he selec i i y da a. Finally, ca alys NiP /KHAp showed a high ini ial
con e sion o 77 %, hough, a apid and signi ican loss o ac i i y
occu ed a e 3 h TOS, ending wi h a 33 % loss o ac i i y a e 30 h
TOS. A common ea u e o all he P -p omo ed ca alys s was he good
s abili y du ing he 30 h o eac ion, hus, i s posi i e e ec on he
apa i e suppo ed ca alys s can be highligh ed.
I is wo h ou lining he beha iou in he selec i i y o hese ca alys s
wi h TOS. Those suppo ed on o he ba e apa i e (Ni/HAp, NiP /HAp)
showed a compa able e olu ion o he me hane yield (Fig. 5b). They
showed an ini ial alue o Y
CH4
=14 % which p og essi ely dec eased
wi h TOS o a ound 10 % a e 30 h TOS. The chemically ac i a ed
ca alys s showed a dis inc e olu ion be ween each o he . The ini ial
me hane p oduc ion by NiP /KHAp and NiP /SHAp was a ound 9 %, in
line wi h da a om he ligh -o cu es. Fo he alkali- ea ed ca alys ,
he ini ial me hane yield sha ply inc eased du ing he i s 5 h TOS (Y
CH4
Fig. 4. Ca aly ic ac i i y and selec i i y in he WGS. (a) CO con e sion ligh -o cu e, (b) me hane yield, (c) Hyd ogen yield. (WGS eac ion condi ions: 5 % CO, 4 %
CO
2
, 31 % H
2
, 14 % He, 46 % H
2
O; GHSV o 120,000 h
−1
).
U. I ia e-Velasco e al.
Jou nal o Indus ial and Enginee ing Chemis y xxx (xxxx) xxx
9
om 9 o 15 %). The acid ea ed assay, con a ily, showed a quasi-
s able me hane o ma ion h oughou all he ca aly ic un (Y
CH4
=
7–8 %).
Fig. 5c shows he e olu ion o hyd ogen yield wi h TOS. The e e -
ence ca alys Ni/HAp showed an ini ial nega i e ne hyd ogen p oduc-
ion which became e en mo e nega i e a e 10 h o TOS, coinciding
wi h he decline in he ca alys ac i i y. P -doping dec eased he selec-
i i y o hyd ogen, as e ealed by Y
H2
o ca alys s NiP /HAp which
anged om −40 o −20 % h oughou all he un. The alkali- ea ed
ca alys s showed a high ini ial hyd ogen yield (a ound 20 %), in line
wi h p e ious ligh -o da a (Fig. 4c). Howe e , i de e io a ed a e 3 h
o TOS. The ca alys NiP /SHAp showed he mos p omising beha iou
wi h a posi i e ne hyd ogen in he whole ope a ion. I showed an ini ial
alue o a ound 10 % which e en inc eased wi h TOS ( o a ound 20 %,
a e 10 h TOS), coinciding wi h he abo e-men ioned decline in
me hana ion ac i i y. Zhou e al. [60] s a ed ha he ac i a ion mode o
hyd ogen on single a oms depends on he local coo dina ion en i on-
men , jus like he in e play be ween ligands and me al cen es in ho-
mogeneous o ganome allic chemis y. Indeed, he obse ed in e se
co ela ion is consequence o he occu ence o CO and/o CO
2
hyd o-
gena ion o gi e me hane. The H
2
o CH
4
mola a io (H
2
/CH
4
) was used
o analyse he eac ion selec i i y. The highes H
2
/CH
4
we e ob ained by
he acid ac i a ed assay (Fig. 5d). Fo his ca alys , H
2
/CH
4
s eadily
inc eased om 40 o c.a. 50 a e 30 h TOS. This a io was abou wo-
old la ge han ha eached by he es o ca alys s in he long- e m
un. As no ed abo e, ca alys NiP /KHAp showed a dis inc beha iou .
The ini ial dec ease in he H
2
/CH
4
mole a io could be de i ed om he
boos ing o he me hana ion ac i i y, which is a hyd ogen-consuming
eac ion. This beha iou migh be ela ed o i s highe basic cha ac e
(Table 2) which end o e ain he chemiso bed ca bonaceous species.
Some eoxida ion o he Ni pa icles could be deduced om XRD
di ac og am o exhaus ed ca alys s (Fig. 1c), basically in he non-
chemically ea ed ca alys s. I could be en isaged ha he chemical
ac i a ion could inhibi he agg ega ion o he Ni nanopa icles and
p omo e he ca alys s abili y e en unde he high empe a u e condi-
ions in he WGS eac ion. I is also wo h no ing he o ma ion o solid
ca bon, as deduced om he mass balance (Figu e S6, ESI), specially, in
he non-chemically ea ed assays. Fo ins ance, he solid ca bon o -
ma ion was low by he e e ence ca alys du ing he i s 10 h o TOS
(<5%), he ea e , ab up ly inc eased o a ound 25 %. Such al e a ion
can be linked o he chemical and ex u al ins abili y o his ca alys . I
seemed ha he occu ence o he Boudoua d eac ion (Equa ion (5)
migh be a ou ed [34]. This eac ion scheme would imply he simul-
aneous inc ease in CO
2
selec i i y. Howe e , as illus a ed in Figu e S6a
(ESI), he selec i i y o CO
2
o Ni/HAp dec eased a e 10 h TOS. The
ac ha he sha p change in he solid ca bon (inc ease), CO
2
(dec ease)
and H
2
(dec ease) selec i i y ook place simul aneously, sugges ed ha
all hese species con o med a complex eac ion mechanism. Tha is, he
CO
2
o med om he Boudoua d eac ion could be eadily consumed by
he RWGS. Unde he expe imen al condi ions used in his wo k, CO
o ma ion om RWGS (Equa ion (1) is possible as p edic ed by he -
modynamic calcula ions [61]. F om esul s in Figs. 4-5 i seemed ha
RWGS mechanism was likely o occu o he e e ence ca alys Ni/HAp.
No e ha hyd ogen and wa e we e also co- ed o he WGS eac ion
es s. As p e iously no ed, P -doping aimed o s abilize he ca aly ic
beha iou , and selec i i y o solid ca bon was a ound 7 % o NiP /HAp
du ing 30 h TOS ha las ed he un. The a ou able in e ac ion be ween
Ni-P alloys and he ca ie seemed o pose a clea e ec on such
beha iou and somewha limi ed he me hana ion capaci y. Howe e , i
seemed ha he Boudoua d eac ion and RWGS eac ion we e a ou ed,
Fig. 5. (a) CO con e sion, (b) me hane yield, (c) hyd ogen yield, and (d) H
2
/CH
4
a io in long- e m es s ob ained by di e en ca alys s.
U. I ia e-Velasco e al.
