In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
A ailable online 3 Feb ua y 2024
0360-3199/© 2024 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/).
Pe o mance o NiAl
2
O
4
spinel de i ed ca alys +dolomi e in he so p ion
enhanced s eam e o ming (SESR) o aw bio-oil in cyclic ope a ion
Lei e Landa
*
, Ainge u Remi o, Jos´
e Valecillos, Bea iz Valle, Ja ie Bilbao, Ana G. Gayubo
Depa men o Chemical Enginee ing, Uni e si y o he Basque Coun y (UPV/EHU), P.O. Box 644, Bilbao, 48080, Spain
ARTICLE INFO
Handling Edi o : D A Bha naga
Keywo ds:
SESR
Bio-oil
Hyd ogen
Ni spinel
Coke
Reac ion- egene a ion cycles
ABSTRACT
The p oduc ion o H
2
om aw bio-oil wi h high yield and pu i y equi es he de elopmen o e o ming ech-
nologies wi h low ene gy equi emen s, minimized CO
2
emissions, and s able and egene able ca alys s. This
wo k s udies he pe o mance (ac i i y, selec i i y, s abili y and egene abili y) in he so p ion enhanced s eam
e o ming (SESR) o aw bio-oil o a ca alys p epa ed by educ ion o a NiAl
2
O
4
spinel oge he wi h dolomi e as
CO
2
so ben . The eac ion uns we e ca ied ou in a luidized-bed eac o unde he ollowing condi ions:
550–700 ◦C; space ime, 0.15 and 0.30 g
ca alys
⋅h/g
oxygena es
; dolomi e/ca alys mass a io, 10 and 20; s eam/
ca bon (S/C) mola a io, 3.4; ime on s eam, 50 and 300 min. The highes H
2
yield (>92 %) and pu i y (>99 %)
in he CO
2
cap u e pe iod a e ob ained in he 600–650 ◦C ange and wi h a dolomi e/ca alys mass a io o 10,
due o syne gy be ween ca alys and so ben ac i i y. The ca alys /so ben sys em can be egene a ed (4 h in ai
a 850 ◦C and subsequen educ ion a 900 ◦C) and used in he successi e eac ion- egene a ion cycles. The
esul s a e o ele an in e es o p og ess owa ds scale-up o his p ocess, which combines sus ainable p o-
duc ion o high pu i y H
2
om biomass wi h CO
2
cap u e.
1. In oduc ion
The p oduc ion o enewable hyd ogen is a p io i y goal in he global
economic and poli ical s a egies o ca bon neu ali y and o a apid
ansi ion o clean ene gy [1]. Hyd ogen is a clean ene gy ca ie , wi h a
g ea po en ial o mee g owing demand in he anspo , indus ial and
powe gene a ion sec o s. I is also a aw ma e ial in g owing demand
o e ilize p oduc ion, and i s a ailabili y is key o he easibili y o
p oducing uels and syn he ic commodi ies (such as me hanol, dime hyl
e he , ole ins, a oma ics o uels) om CO
2
[2,3]. I is also impo an as a
bu e o non-dispa chable enewable ene gy (wind, sola ) [4].
The es ablishmen o he H
2
economy h ough elec olysis echnol-
ogies equi es a huge mul idisciplina y e o , including he applica ion
o ad anced echnologies, he de elopmen o s able and egene able
ca alys s and ad anced ma e ials (nanos uc u ed and composi es) o
H
2
p oduc ion [5] and s o age [6–10], combined wi h he e icien use o
enewable ene gy [11]. In he ansi ion pe iod be o e he a ailabili y o
g een H
2,
he p oduc ion o H
2
om lignocellulosic biomass is o in e es
[12,13]. The cos o H
2
p oduc ion om biomass wi h di e en ech-
nologies is highe han ha o me hane s eam e o ming [14], bu
emission axes and CO
2
cap u e and s o age cos s a e lowe [13]. Among
he di e en echnologies, s eam e o ming (SR) o bio-oil is an al e -
na i e o py olysis/gasi ica ion o biomass and in-line SR o ola iles [15,
16]. So p ion Enhanced S eam Re o ming (SESR) o bio-oil has g ea
appeal o a oiding CO
2
emissions h ough i s in si u cap u e [17–19]
and p oduces H
2
o highe pu i y han in SR. Bio-oil is ob ained om as
py olysis o lignocellulosic biomass wi h a high yield, i s p oduc ion can
be ca ied ou o -si e, wi h simple echnologies and educed en i on-
men al impac [20] and i s anspo o he alo iza ion uni ( o i s
la ge-scale e o ming) is easie han ha o he o iginal biomass. The SR
o SESR o bio-oil a oids he cos ly elimina ion o he high wa e con en
equi ed o i s use as a uel o o hyd oca bon p oduc ion [21], and he
by-p oduc handling equi ed in he sepa a ion o some o i s oxygena es
compounds [22]. The o e all eac ion o SR o SESR o bio-oil in ol es
he e o ming eac ion o p oduce (CO +H
2
) (Eq. (1)) and he subse-
quen wa e -gas-shi (WGS) eac ion (Eq. (2)):
CnHmOk+ (n-k)H2O → nCO +(n+m
2-k)H2(1)
CO +H2O ↔ CO2+H2(2)
Thus, he o e all equa ion is:
* Co esponding au ho .
E-mail add ess: [email p o ec ed] (L. Landa).
Con en s lis s a ailable a ScienceDi ec
In e na ional Jou nal o Hyd ogen Ene gy
jou nal homepage: www.else ie .com/loca e/he
h ps://doi.o g/10.1016/j.ijhydene.2024.01.228
Recei ed 18 Oc obe 2023; Recei ed in e ised o m 3 Janua y 2024; Accep ed 20 Janua y 2024
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1527
CnHmOk+ (2n-k)H2O → nCO2+ (2n +m
2-k)H2(3)
H
2
yield is also a ec ed by seconda y eac ions such as decomposi-
ion/c acking (Eq. (4)), e o ming o decomposi ion p oduc s (CH
4
and
hyd oca bons, Eqs. (5) and (6), espec i ely), and in e con e sion o
oxygena es (Eq. (7)).
CnHmOk→ CxHyOz+gas (CO,CH4,CO2,CaHb,H2,…) + coke (4)
CH4+H2O ↔ CO +3H2(5)
CaHb+aH2O → aCO + (a+b/2)H2(6)
CnHmOk→ CxHyOz(7)
Mo eo e , he eac ions o coke (C) o ma ion and gasi ica ion (Eqs.
(8)–(10)) should be conside ed as hey a ec ca alys s abili y.
Me hane decomposi ion, CH
4
→ 2H
2
+C (8)
Boudoua d eac ion, 2CO ↔ C +CO
2
(9)
Coke gasi ica ion, C +H
2
O → CO +H
2
(10)
Using a so ben ( aking CaO as example), he CO
2
is e ained as:
CaO +CO2↔ CaCO3(11)
The exis ence o his eac ion (Eq. (11)) in he SESR p ocess, in
addi ion o a oiding CO
2
emissions (wi h he consequen educ ion o
CO
2
sepa a ion and s o age cos s compa ed o he SR p ocess), o e s he
ollowing ad an ages o e he SR p ocess: i) imp o es he H
2
yield and
selec i i y by shi ing he equilib ium o he WGS eac ion (Eq. (2)),
p oducing H
2
o highe pu i y; ii) educes he ene gy equi emen (since
he deca bona ion eac ion is exo he mic); iii) dec eases he eac ion
empe a u e, hus educing ca alys sin e ing (a c i ical ac o in i s se-
lec ion). The elease o high pu i y CO
2
in he egene a ion o he so -
ben by calcina ion ( e e se o Eq. (11)) will acili a e i s cap u e and
subsequen alo iza ion. A de ailed e iew o so ben s has been ca ied
ou by Dou e al. [23], wi h calcium-based so ben s being he mos
widely used [24]. The high CO
2
so p ion capaci y a high empe a u es
and he low cos o CaO (easily p epa ed om limes one) make i s use
mo e in e es ing han ha o alkaline ce amic so ben s (such as
Li
2
Z O
3
). The p oblem o losing CaO so p ion capaci y du ing cyclic
ca bona ion/deca bona ion ope a ion has been add essed by inco po-
a ing CaO in o o he suppo ma e ials and eac i a ing i wi h di e en
he mal and chemical ea men s [25].
SESR has been s udied mainly o me hane (SE-SMR p ocess)
[24–27], which is conside ed o be a mo e e icien echnology han he
combina ion o s eam e o ming o me hane (SRM) and ca bon cap u e
and s o age (CCS) uni s, comme cially de eloped o blue H
2
p oduc ion
[28]. Using Ni and CaO based solids, SE-SMR is p e e en ially ope a ed
a ound 650 ◦C, a a mosphe ic p essu e and wi h he S/C (s eam/-
ca bon) mola a io in he ange o 3–4 [29]. Unde hese condi ions, H
2
pu i y o o e 95 % (d y basis) is achie ed, compa ed o 76 % in SR. The
scale-up o he p ocess is s udied by simula ion wi h a dual luidized-bed
sys em combining a e o me and a calcine wi h ca alys ci cula ion
be ween he wo uni s [30]. Thus, he ac i i y o he so ben is eco e ed
in he calcine and he homogeneous empe a u e in each uni (highe in
he calcine uni ) a o s i s con ol and ha o he hea ans e o he
ca alys /so ben s eam (key o minimize he ex e nal ene gy inpu ).
Due o he complexi y o bio-oil (a mix u e o oxygena es o di e en
na u e) and i s handling, SESR has gene ally been s udied o pu e
model compounds, such as ace ic acid [31–35], e hanol [36,37], phenol
[38,39], glyce ol [40,41] and oluene [42], and o mix u es o oxy-
gena es [32,43]. In gene al, Ni-based ca alys s ha e been used wi h
di e en ma e ials con aining CaO as CO
2
so ben , and also bi unc ional
composi es (usually based on Ni o e CaO and wi h di e en modi i-
ca ions) ha e been used [36,44–47]. The objec i es o hese wo ks ha e
ocused on he op imiza ion o he ca alys composi ion and he eac ion
condi ions, al hough he esul s a e di e en and he app op ia e con-
di ions o ob ain a H
2
pu i y abo e 95 % depends on he na u e o he
oxygena es. Thus, his na u e has a g ea in luence on he s abili y o he
ca alys s, which has also been es ablished in he SR o oxygena es in
bio-oil [48].
Expe imen al s udies on he SESR o he aqueous ac ion o bio-oil
a e e y sca ce [49–51]. Remi o e al. [50], using a Ni/La
2
O
3
–Al
2
O
3
ca alys and calcined dolomi e as so ben , obse ed a signi ican ole o
he ca alys /so ben mass a io due o he ac i e ole o dolomi e o
oxygena es c acking and also o e o ming eac ions. A sui able balance
was s icken a 600 ◦C o ca alys /dolomi e mass a ios ≥0.17 wi h a H
2
yield a ound 99 %. The SESR o aw bio-oil o e s a highe yield e e ed
o biomass, as es ablished by he modynamic s udies [52] and expe i-
men ally [35,53]. Landa e al. [53] ob ain high H
2
pu i y using Ni/Al
2
O
3
(ob ained by educ ion o a NiAl
2
O
4
spinel) as ca alys and dolomi e as
so ben , wi h be e pe o mance han o he ca alys +so ben sys ems,
such as Ni/CeO
2
+dolomi e o Ni/Al
2
O
3
+CaO/mayeni e. Iliu a e al.
[54] also de e mine he lowe ca alys deac i a ion in SESR o bio-oil
compa ed o SR by simula ing he p ocess in a packed-bed eac o ,
using he kine ic model o Gayubo e al. [55] o he eac ions in ol ed
in SR o bio-oil (on a Ni/La
2
O
3
-
α
Al
2
O
3
ca alys ) and he kine ics o CO
2
adso p ion o e Li
2
CuO
2
.
Conside ing he in e es o ac i i y eco e y by ca alys egene a ion
o he iabili y o he SESR p ocess, Acha e al. [43] ha e e i ied he
impo ance o he Ni ca alys suppo and he composi ion o he dolo-
mi e used as so ben o he ac i i y eco e y o bo h ma e ials used in
he SESR o ace ic acid and a mix u e o oxygena es. The esul s high-
ligh he impo ance o a suppo ha minimizes sin e ing and coke
deposi ion, while dolomi es lose hei cap u e capaci y upon eac-
i a ion. Li e al. [35] in es iga ed he s abili y o a Ni/Ce
1⋅2
Z
1
Ca
5
bi unc ional ca alys -so ben (Ce/Ca a io o 0.24) in he SESR o ace ic
acid and popla sawdus py olysis oil in a packed-bed eac o o e 15
eac ion- egene a ion cycles. While an excellen ca aly ic pe o mance
was epo ed a e 15 consecu i e cycles o he SESR o ace ic acid
(550 ◦C, S/C o 4 and LHSV o 0.48 ml/g
ca alys
⋅h), a dec ease in CO
2
adso p ion capaci y and H
2
yield was obse ed in he SESR o bio-oil,
which was explained by he sin e ing o CaO pa icles and he o ma-
ion o coke and sin e ing o Ni. These phenomena we e mo e impo an
in he bio-oil SESR, wi h he o ma ion o mo e condensed coke, whose
emo al by gasi ica ion (Eq. (10)) is no comple e a he eac i a ion
condi ions (750 ◦C).
Conside ing he li e a u e backg ound and ou abo e-men ioned
p e ious esul s, he a en ion o his manusc ip has been ocused on
he pe o mance o he ca alys /so ben sys em o (Ni/Al
2
O
3
ca alys
de i ed om NiAl
2
O
4
spinel) +dolomi e in he SESR o aw bio-oil
ope a ing in eac ion- egene a ion cycles. The p ocess add esses he
gaps in H
2
p oduc ion om biomass using bio-oil wi hou p e- ea men ,
p oducing high pu i y H
2
, no CO
2
emissions and lowe ene gy e-
qui emen s han o he e o ming p ocesses. The ca alys was selec ed
due o i s comple e egene abili y in he SR o aw bio-oil unde sui able
condi ions [56]. Likewise, dolomi e, besides being a low-cos ma e ial,
has a good pe o mance as a CO
2
so ben , wi h a posi i e syne gy wi h
Ni/Al
2
O
3
ca alys de i ed om NiAl
2
O
4
spinel in he SESR o bio-oil and
as gua d-bed, e aining a ac ion o he deposi ed coke [57]. Fu he -
mo e, he alo iza ion o aw bio-oil (wi h he a ac ion o high H
2
yield) and he low empe a u e equi ed o a o he ex en o CO
2
adso p ion ha e he d awback o high coke deposi ion, leading o apid
ca alys deac i a ion, which is a challenge o ca alys and so ben
eac i a ion. Consequen ly, gi en he impo ance o deac i a ion, spe-
cial a en ion has been paid o he s udy o i s causes and mechanism,
using se e al cha ac e iza ion echniques (Tempe a u e P og ammed
Oxida ion (TPO), adso p ion-deso p ion o N
2
, Tempe a u e P o-
g ammed Reduc ion (TPR), X-Ray Di ac ion (XRD), Scanning Elec on
Mic oscopy (SEM) and X-Ray Pho oelec on Spec oscopy (XPS)).
The esul s in his manusc ip a e explained by conside ing he e ec
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1528
o he eac ion condi ions on he ex en o each eac ion in ol ed in he
SESR p ocess and on he con en and na u e o he coke deposi ed on he
ca alys . These esul s show ha SESR o aw bio-oil is an a ac i e
ou e o he p oduc ion o H
2
om biomass, a oiding he CO
2
emissions
associa ed wi h o he e o ming echnologies. Fu he mo e, he p og ess
in knowledge o he p ocess co esponds o key objec i es o i s
iabili y, such as he use o aw bio-oil as eeds ock, he p oposal o a
highly ac i e ca alys wi h mode a e deac i a ion by coke and egen-
e able, and he es ablishmen o op imal condi ions ( eac ion empe a-
u e, so ben /ca alys mass a io, space- ime) o maximizing H
2
yield
and pu i y and o s able ope a ion in SESR o bio-oil in successi e
eac ion- egene a ion cycles. The use o a low-cos so ben such as
dolomi e is ano he ac o a o ing he easibili y o he p ocess. In
addi ion, he esul s a e based on expe imen s in a luidized-bed eac o ,
which a o s i s use in scale-up whe e his ype o eac o is likely o be
equi ed.
2. Expe imen al
2.1. Bio-oil
The aw bio-oil was supplied by BTG Bioliquids BV (The
Ne he lands) and was syn he ized by as py olysis o pine sawdus in a
plan wi h a capaci y o ope a e con inuously wi h 5 on/h o biomass
(bio-oil yield o 70 %) and wi h a conical o a o y eac o . The physico-
chemical p ope ies o he bio-oil a e as ollows: wa e con en , de e -
mined by Ka l-Fische i a ion (KF-Ti ino Plus 870), 24 w %; densi y,
1.201 g/ml; iscosi y a 40 ◦C, 250 cP (B ook ield DV2T Ame ek); pH,
2.5–3.5, and; empi ical o mula, ob ained by CHO analysis, using a Leco
CHN-932 analyze (wa e - ee basis), esul ed in C
4⋅6
H
6⋅2
O
2.4
. The
de ailed composi ion o he aw bio-oil was de e mined using a Shi-
madzu QP2010S gas ch oma og aphy/mass spec ome e (GC/MS)
equipped wi h a BPX-5 column (50 m ×0.22 mm ×0.25
μ
m) and mass-
selec i e de ec o . The main compounds a e ke ones, acids, phenols
(guaiacol), es e s, saccha ides (le oglucosan), aldehydes, u ans/ u -
anones, alcohols, and e he s (Table 1).
2.2. P epa a ion o ca alys and so ben
The NiAl
2
O
4
spinel (33 w % nominal Ni con en ) was syn hesized by
he co-p ecipi a ion me hod [58], mixing a 25 ◦C aqueous solu ions o
hexahyd a ed nickel ni a e (Ni(NO
3
)
2
⋅6H
2
O, Scha lau, pu i y o 98 %)
and nona-hyd a ed alumina ni a e (Al(NO
3
)
3
⋅9H
2
O, Honeywell Fluka,
pu i y o 98 %) and adding d opwise a 0.6 M solu ion o ammonium
hyd oxide (NH
4
OH 5 M, Honeywell Fluka) as a p ecipi a ing agen un il
a pH o 8 is eached. The p ecipi a e was hen eco e ed by il a ion and
washed wi h dis illed wa e o emo e he emaining ammonium ions,
d ied o e nigh a 110 ◦C, calcined (850 ◦C o 4 h, wi h a hea ing amp
o 10 ◦C/min), c ushed and sie ed o ob ain pa icle sizes in he ange o
150–250
μ
m. The NiAl
2
O
4
spinel p ecu so was educed (condi ions
indica ed in Sec ion 2.4) in o de o ob ain he Ni/Al
2
O
3
ca alys ac i e
o he e o ming eac ions. A schema ic o he ca alys syn hesis p ocess
is shown in Fig. S1.
Calcined dolomi e, composed o calcium and magnesium oxide (CaO
and MgO), is ob ained om dolomi ic mine als consis ing o calcium
and magnesium ca bona e (CaMg(CO
3
)
2
) wi h Fe
2
O
3
impu i ies (sup-
plied by Calcino S.A. (Can ab ia, Spain)) h ough a calcina ion p ocess
(Eq. (12)).
CaMg(CO3)2+hea ↔ CaO +MgO +2CO2(12)
The p epa a ion o he so ben consis s o a i s g inding o adjus
he pa icle size be ween 90 and 125
μ
m. Then, he na u al dolomi e is
d ied a 110 ◦C o 24 h. Finally, i is calcined a 850 ◦C o 5 h wi h a
hea ing a e o 10 ◦C/min, so ha he na u al so ben is he mally
decomposed in o CaO ( he ac i e phase o CO
2
adso p ion) and MgO
[59,60]. A e wa ds, i is sie ed again o ensu e a pa icle size be ween
90 and 125
μ
m, since du ing he calcina ion s ep he pa icles dec ease
in size due o dec epi a ion caused by he loss o CO
2
om he ca bon-
a es. A schema ic o he calcined dolomi e p epa a ion p ocedu e is
shown in Fig. S2. No e ha he pa icle size o he dolomi e is di e en
om ha o he ca alys o acili a e hei sepa a ion by sie ing a e
each eac ion, so ha he ca alys used can be cha ac e ized sepa a ely
om he so ben (and also om he ine solid used in he luidized-bed
eac o , as desc ibed in Sec ion 2.4).
2.3. Cha ac e iza ion o he ca alys
The physical p ope ies o he esh- educed and used ca alys s (BET
su ace a ea, po e olume and mean po e diame e ), we e cha ac e ized
by adso p ion-deso p ion o N
2
in a Mic ome i ics ASAP 2010. Tem-
pe a u e P og ammed Reduc ion (TPR) analysis was ca ied ou in a
Mic ome i ics Au oChem II 2920 appa a us o de e mining he educ-
ibili y o he me al species. The TPR measu emen was also used o es-
ima e he eal Ni con en in he spinel and a e ca alys egene a ion
using a he mobalance (TA Ins umen s SDT-2960) o de e mine he O
a oms emo ed as H
2
O. The s uc u al p ope ies o esh- educed and
used ca alys we e analyzed by X-Ray Di ac ion (XRD), measu ed on a
B uke D8 Ad ance di ac ome e wi h CuK
α
1 adia ion, in o de o
calcula e he sin e ing dynamics o he Ni c ys als (using he Sche e
equa ion) and he c ys alline s a e o he coke deposi s. The amoun and
na u e o coke deposi ed on used ca alys samples has been de e mined
by Tempe a u e P og ammed Oxida ion (TPO) in a TA-Ins umen s
TGA-Q5000IR he mobalance, coupled in line wi h a mass spec om-
e e (The mos a Balze s ins umen ) o moni o he CO
2
signal. The
coke con en has been quan i ied om he CO
2
spec oscopic signal,
since he oxida ion o Ni c ys als du ing he combus ion p ocess could
mask he he mog a ime ic signal. The su ace mass con en o he
di e en elemen s on he ca alys su ace and he possible mig a ion o
Ni du ing he deac i a ion o egene a ion s ages has been de e mined
by X-Ray Pho oelec on Spec oscopy (XPS) analysis in a SPECS sys em
equipped wi h a Phoibos 150 1D-DLD analyze , Al K
α
monoch oma ic
adia ion (1486.6 eV), an X- ay exci ing sou ce and a hemisphe ical
elec on analyze . The binding ene gy o he C1s ca bon peak was se a
284.6 eV o co ec he ma e ial cha ging, and he analysis was ca ied
ou using an elec on ake o angle o 90◦. The scanning elec on mi-
c oscopy (SEM) images o he used ca alys s we e ob ained in a Hi achi
S-4800 N ield emission gun scanning elec on mic oscope wi h an
accele a ing ol age o 5 kV and a seconda y elec on de ec o (SE -
SEM).
Table 1
Composi ion (w %) o aw bio-oil ob ained om
as py olysis o pine sawdus .
Compound w %
Acids 19.5
Ace ic acid 16.6
Ke ones 21.4
Linea 17.1
Ace one 5.2
Ace ol 9.4
Cyclic 4.3
Es e s 11.3
Fu ans/Fu anones 5.0
Alcohols 3.2
Aldehydes 6.8
E he s 0.8
Saccha ides 13.7
Le oglucosan 11.1
Phenols 18.4
Alkylphenol 1.4
Guaiacol 11.1
Ca echol 0.9
Sy ingol 0.5
O he s 4.5
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1529
2.4. Reac ion equipmen and ope a ing condi ions
The kine ic uns ha e been pe o med in an au oma ed eac ion
sys em (Mic oAc i i y-Re e ence, PID Eng & Tech) p o ided wi h wo
uni s ( he mal s ep and ca aly ic s ep) in se ies (Fig. S3 in Supplemen-
a y Ma e ial). The i s uni ( he mal ea men ) consis s o a U-shaped
s eel ube (inne diame e =0.75 in) a 500 ◦C o con olled apo -
iza ion o bio-oil and epolime iza ion o some oxygena es (mainly
phenolic compounds) which a e e ained as a ca bonaceous solid called
py oly ic lignin (PL) [50]. In his way, he subsequen ca alys deac i-
a ion in he SESR and SR p ocesses is a enua ed, wi h a mode a e
dec ease in he low a e o oxygena es en e ing he ca aly ic eac o
(uni 2). The ola ile s eam lea ing he he mal ea men uni is con-
e ed (by ca aly ic SESR o SR) in a luidized-bed eac o (s ainless
s eel, wi h 22 mm o in e nal diame e and o al leng h o 460 mm). In
his eac o , he ca aly ic bed composed o NiAl
2
O
4
spinel (wi h a pa -
icle size o 150–250
μ
m) and so ben (calcined dolomi e, wi h a pa icle
size be ween 90 and 125
μ
m) is mixed wi h ine solid (SiC, wi h a
pa icle size o 75
μ
m) in o de o imp o e he luid dynamics o he bed.
The ine na u e o SiC in he SR o bio-oil has been demons a ed in a
p e ious wo k [53].
An injec ion pump (Ha a d Appa a us 22) was used o eed he bio-
oil (0.06 ml/min) and a 307 Gilson pump o co- eed he addi ional wa e
equi ed acco ding o he desi ed s eam o ca bon (S/C) mola a io. The
eac ion p oduc s we e analyzed in a Mic o GC (Va ian CP-490) con-
nec ed in-line o he eac o h ough an insula ed line (130 ◦C) o a oid
condensa ion o he p oduc s. The gas ch oma og aph is equipped wi h
h ee analy ical channels: molecula sie e MS5 o quan i ica ion o H
2
,
O
2
, N
2
, CH
4
and CO; PPQ column o ligh hyd oca bons (C
2
–C
4
), CO
2
and wa e ; and S abilwax o oxygena ed compounds (C
2+
) and wa e .
P io o each eac ion, he Ni spinel p ecu so is educed in si u unde
H
2
–N
2
(7 ol% H
2
) a 850 ◦C o 4 h o ob ain he ac i e Ni/Al
2
O
3
ca alys wi h he Ni me allic phase well-dispe sed on he alumina sup-
po [58]. The SESR uns o aw bio-oil we e ca ied ou in he ca aly ic
eac o unde he ollowing condi ions: 550–700 ◦C; bio-oil low a e,
0.06 ml/min; space ime, 0.15 and 0.30 g
ca alys
⋅h/g
oxygena es
(co e-
sponding o 0.5 g and 1 g o ca alys , espec i ely); so ben /ca alys
mass a io, 10 and 20; s eam/ca bon (S/C) mola a io a he eac o
inle , 3.4; ime on s eam (TOS), 50 and 300 min. Join egene a ion
condi ions o ca alys and dolomi e we e: 850 ◦C in ai in an ex e nal
o en, o 4 h.
2.5. Reac ion indices
The ollowing eac ion indices we e used o quan i y he esul s:
The con e sion o oxygena es in he ea ed bio-oil ( ha is, he
ola ile oxygena es exi ing he Uni 1 used o he mal ea men ) is
exp essed as he ca bon uni s con e ed in o gas:
X=Fou ,gas
Fin
(13)
whe e F
ou ,gas
is he ca bon-mola low a e o he o al ca bon in gaseous
p oduc (CO
2
, CO, CH
4
and ligh hyd oca bons) a he eac o ou le , and
F
in
is he ca bon-mola low a e o he oxygena es a he eac o inle .
Eq. (13) is sui able o quan i ying he o al con e sion o oxygena es in
he e o ming eac o because he yield o solid ca bon (coke) is low
unde he condi ions s udied (below 5 % as de ined in he ollowing Eq.
(15)).
H2yield is calcula ed as :YH2=FH2
Fo
H2
(14)
whe e F
H2
is he H
2
mola low a e in he p oduc s eam and Fo
H2 is he
s oichiome ic mola low a e, calcula ed as (2n +m/2 – k)/n F
in
, ac-
co ding o he global s oichiome y o bio-oil (C
n
H
m
O
k
) s eam e o m-
ing (including he WGS eac ion) (Eq. (3)).
Yield o he ca bon p oduc s :Yi=Fi
Fin
(i=CO,CO2,CH4and HC)(15)
whe e F
i
is he ca bon-based mola low a e o he i-p oduc (CO, CO
2
,
CH
4
and HC) in he e luen (ou ) s eam o he second uni (ca aly ic
eac o ).
3. Resul s
3.1. Beha io o ca alys /so ben unde di e en condi ions
This sec ion shows he esul s o he e ec o di e en ope a ing
a iables on he eac ion indices in he SESR o bio-oil, bo h in he CO
2
cap u e pe iod and a e dolomi e sa u a ion, in o de o analyze he
ac i i y and s abili y o he ca alys /so ben sys em. The ocus is on he
e ec o empe a u e, bu he e ec o so ben /ca alys mass a io and
space ime a e also add essed, conside ing he ela ionship o he e ec
o hese a iables. The esul s shown in his sec ion a e mean alues
ob ained om epea ed expe imen s (wi h an a e age mean e o below
5 %).
Fig. 1 compa es he e olu ion wi h ime on s eam (TOS) o he e-
ac ion indices (con e sion and p oduc yield) o di e en empe a u es
in he 550–700 ◦C ange, wi h a so ben /ca alys mass a io o 10 and a
S/C a io o 3.4. A low space ime (0.15 g
ca alys
⋅h/g
oxygena es
) was used o
be e app ecia e he e ec o empe a u e on ca alys s abili y in he
pos -sa u a ion pe iod. The esul s ob ained in uns wi hou so ben (SR
es s) and wi h a so ben /ca alys mass a io o 20 a e shown in Figs. S4
and S5, espec i ely (Supplemen a y Ma e ial).
As obse ed in Fig. 1 and Fig. S5, he SESR p ocess can be di ided
in o h ee egions: p e-b eak h ough (wi h CO
2
cap u e, g een zone in
Fig. 1c), b eak h ough (when dolomi e sa u a ion begins o be no ice-
able, blue zone in Fig. 1c) and pos -b eak h ough (a e comple e sa u-
a ion o he dolomi e). Unde he condi ions o Fig. 1 (wi h a dolomi e/
ca alys mass a io o 10), he du a ion o he p e-b eak h ough pe iod is
abou 30 min abo e 600 ◦C and is sligh ly longe a 550 ◦C (Fig. 1c). The
du a ion o his pe iod is doubled when wice he so ben /ca alys a io
is used (Fig. S5). In he cap u e pe iod, he highe H
2
yield and lowe CO
and CH
4
yields compa ed o he SR es (Fig. S4) a e explained by he
shi o he WGS eac ion equilib ium by CO
2
cap u e (Eq. (2)) and, as a
consequence, he shi o he SRM eac ion equilib ium (Eq. (5)). The
b eak h ough pe iod in ol es he deple ion o he ca bona ion eac ion
e iciency, and he e o e, a sudden inc ease in he yield o CO
2
(Fig. 1c
and S5c) and a dec ease in he yield o H
2
(Fig. 1b and S5b) is obse ed.
In he pos -b eak h ough pe iod, CO
2
emo al is no longe e ec i e and
SR and WGS eac ions occu oge he wi h o he seconda y eac ions
esponsible o coke o ma ion. Consequen ly, in his pe iod, he eac-
ion indices e ol e wi h ime on s eam due o he deac i a ion o he
ca alys and so ben .
Fig. 2 compa es he p oduc dis ibu ion esul s in he CO
2
cap u e
pe iod o SESR eac ions ( o so ben /ca alys mass a io alues o 10
and 20) and SR eac ions (so ben /ca alys mass a io o 0) a di e en
empe a u es. The alues o SESR eac ions a e mean alues du ing he
CO
2
cap u e pe iod, whe eas hose o SR (wi h a no iceable deac i a ion,
Fig. S4) co espond o ze o ime on s eam. Ligh hyd oca bons yield is
no shown because i is insigni ican o all he ope a ing condi ions
s udied. In he SR eac ions (solid lines in Fig. 2), he H
2
yield inc eases
wi h inc easing empe a u e ( om 58 % a 550 ◦C o 81 % a 700 ◦C)
because he e o ming o oxygena es and CH
4
is a o ed. Consequen ly,
he yield o CH
4
is negligible a 700 ◦C in Fig. 2. The CO yield inc eases
wi h inc easing empe a u e as a consequence o a o ing he e e se-
WGS eac ion. In he SESR eac ions (dashed and do ed lines in
Fig. 2), empe a u e also a ec s he CO
2
adso p ion equilib ium, which
becomes un a o able wi h inc easing eac ion empe a u e. In he ange
s udied, he inc ease in empe a u e a o s he o ma ion o H
2
, whose
yield inc eases om 74 % o 92 % be ween 550 and 600 ◦C o a so ben /
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1530
ca alys mass a io o 10 (Fig. 2a). I is no ewo hy ha a 600 ◦C he
e o ming o CH
4
is almos comple e wi h a yield o less han 1 %. Abo e
his empe a u e, he yield o H
2
inc eases only sligh ly despi e he
inc ease in oxygena e con e sion (Fig. S4a). This esul is explained by
he ac ha CO
2
cap u e by dolomi e is un a o able abo e 600 ◦C [49].
Thus, a 650 and 700 ◦C, he so ben cap u e capaci y dec eases sligh ly,
Fig. 1. E ec o empe a u e on he e olu ion wi h ime on s eam o con e sion (a) and yields o H
2
(b), CO
2
(c), CO (d), CH
4
(e) and C
2
–C
4
hyd oca bons (HCs) ( ).
Reac ion condi ions: space ime, 0.15 g
ca alys
⋅h/g
oxygena es
; so ben /ca alys mass a io, 10; S/C, 3.4. Colo ed zones in g aph c: CO
2
cap u e pe iod (g een shade) and
b eak h ough pe iod (blue shade). (Fo in e p e a ion o he e e ences o colo in his igu e legend, he eade is e e ed o he Web e sion o his a icle.)
Fig. 2. E ec o empe a u e on p oduc yield dis ibu ion (H
2
, CO
2
, CO and CH
4
) (a) and H
2
pu i y (b) in he CO
2
cap u e pe iod o di e en so ben /ca alys mass
a io (0 (SR), 10, 20). Reac ion condi ions: space ime, 0.15 g
ca alys
⋅h/g
oxygena es
; S/C, 3.4.
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1531
as obse ed in he p e-b eak h ough pe iod in Fig. 1c. Al hough dolo-
mi e p omo es he o ma ion o CH
4
by oxygena es c ack-
ing/decomposi ion eac ion (Eq. (4)), he yield o CH
4
emains almos
null in he CO
2
cap u e pe iod abo e 600 ◦C due o he high ac i i y o
he Ni/Al
2
O
3
ca alys o i s SR a hese empe a u es [61]. Ne e heless,
he CO yield (which is less han 1 % a 600 ◦C) sligh ly inc eases o 6 %
a 700 ◦C wi h a so ben /ca alys mass a io o 10 due o he p omo ion
o he e e se-WGS eac ion wi h inc easing empe a u e, which leads o
a dec ease in H
2
pu i y a high empe a u es (Fig. 2b). Fo a so ben /-
ca alys mass a io o 20 (do ed lines in Fig. 2), he p oduc yields in he
CO
2
cap u e pe iod a e almos equal o hose ob ained wi h a so -
ben /ca alys mass a io o 10, excep a 550 ◦C. A his low empe a u e,
he H
2
yield o a so ben /ca alys mass a io o 20 is only 64 % (sligh ly
highe han ha ob ained in he SR es ), which can be a ibu ed o he
signi ican deac i a ion o he ca alys along he CO
2
cap u e pe iod in
he SESR un wi h high dolomi e loading, as obse ed in Figs. S5a–c.
This ca alys deac i a ion is due o he ac i i y o dolomi e o decom-
posi ion/c acking o oxygena es [59], which is a o ed by he low ex en
o SR eac ions a his low empe a u e o his low space ime.
As men ioned abo e, he e olu ion o he p oduc dis ibu ion wi h
ime on s eam in he pos -b eak h ough pe iod (a e CO
2
cap u e) in
Fig. 1 is a consequence o ca alys deac i a ion. Thus, he yields o H
2
and CO
2
dec ease due o he a enua ion o he SR and WGS eac ions,
which in ol es an inc ease in he yields o CO (Fig. 1d), CH
4
(Fig. 1e)
and hyd oca bons (Fig. 1 ). To explain he e ec o empe a u e on hese
ends, i is necessa y o ake in o accoun he di e en esidual ac i i y
o he ca alys a he beginning o his pe iod. Mo eo e , he ela ionship
be ween he e ec s o empe a u e and oxygena e concen a ion on
deac i a ion [62,63] and he ole o dolomi e on deac i a ion should be
conside ed [50,53,57]. Thus, as men ioned abo e, a low empe a u e
(550 ◦C) he ca alys has unde gone signi ican deac i a ion du ing he
CO
2
cap u e pe iod because he space ime was low o con e all he
oxygena es, which explains he subsequen slow dec ease in con e sion
(Fig. 1a) and yields o H
2
(Fig. 1b) and CO
2
(Fig. 1c), and he slow in-
c ease in CO yield (Fig. 1d). These ends indica e ha he ca a-
lys /so ben sys em has eached a quasi-s eady s a e a his empe a u e.
Con e sely, abo e 600 ◦C, he ca alys /so ben sys em has a highe
emaining ac i i y a e he CO
2
cap u e pe iod and i s deac i a ion in
he pos -b eak h ough pe iod is sligh ly mo e p onounced wi h
inc easing empe a u e, as obse ed by he as e dec ease in H
2
yield
wi h ime on s eam (Fig. 1b). The inc ease in he ex en o decom-
posi ion/c acking eac ions o oxygena es o e dolomi e (Eq. (4)) wi h
inc easing empe a u e con ibu es o he inc ease in he yield o
ca bonaceous p oduc s [59,60], which explains why he decay o oxy-
gena es con e sion wi h ime (Fig. 1a) is less p onounced han he
dec ease in H
2
yield (Fig. 1b) a high empe a u e.
Compa ing he esul s o Fig. 1 (co esponding o SESR wi h a so -
ben /ca alys mass a io o 10), Fig. S4 (SR) and Fig. S5 (SESR wi h a
so ben /ca alys mass a io o 20), i is obse ed ha he e ec o em-
pe a u e on ca alys s abili y depends on he p esence and con en o
dolomi e wi h he ca alys . Thus, a low empe a u es (550–600 ◦C), he
p esence and con en o dolomi e has a nega i e e ec , leading o a
as e ca alys deac i a ion. A 650 ◦C, he p esence and con en o
dolomi e p olongs he s abili y pe iod o he ca alys , bu a e his
pe iod, he deac i a ion is as e han in he SR p ocess, wi h a apid
dec ease in he eac ion indices. Consequen ly, he H
2
yields a 300 min
on s eam a his empe a u e a e 33 %, 20 % and 17 % o SR (Fig. S4b),
so ben /ca alys mass a ios o 10 (Figs. 1b) and 20 (Fig. S5b), espec-
i ely. On he con a y, a 700 ◦C, he mode a e p esence o dolomi e
(so ben /ca alys mass a io o 10) has a posi i e e ec , wi h he eac-
ion indices dec easing mo e slowly wi h ime on s eam. This esul is
explained by he ac i i y o dolomi e o SR eac ions a his empe a-
u e, and also because dolomi e deac i a es mo e slowly han he ca a-
lys [60]. Howe e , o a so ben /ca alys mass a io o 20, he e is a
as e ca alys deac i a ion. In ac , he use o a high dolomi e/ca alys
a io o 20 (Fig. S5) esul s in as e ca alys deac i a ion a all
empe a u es.
In addi ion o he a iables s udied abo e ( eac ion empe a u e and
so ben /ca alys mass a io), he e ec o space ime has been s udied, as
he amoun o a ailable ac i e ca alys si es in luences he ex en o
e o ming and WGS eac ions, and consequen ly he dis ibu ion o
p oduc s and ca alys deac i a ion. Fig. 3 shows he e olu ion o he
eac ion indices wi h ime on s eam o a high space ime o 0.3
g
ca alys
⋅h/g
oxygena es
( wice ha o Fig. 1). The esul s co espond o
600 ◦C and a so ben /ca alys mass a io o 10 ( he condi ions leading o
he highes H
2
pu i y acco ding o Fig. 2b). The du a ion o he CO
2
cap u e pe iod in Fig. 3 (abou 50 min) is longe han ha ob ained in
Fig. 1 a he same empe a u e (abou 30 min, ed do s) due o he highe
con en o dolomi e in he bed (10 g, compa ed o 5 g in Fig. 1). I should
be no ed ha his dolomi e con en is he same as ha used in he uns
wi h a so ben /ca alys mass a io o 20 (Fig. S5), bu now wi h wice he
amoun o ca alys . Compa ing he p oduc yields in he CO
2
cap u e
pe iod in Fig. 3 and he ed cu es in Fig. 1 ( hose co esponding o
600 ◦C wi h hal he space ime as in Fig. 3), a signi ican imp o emen
in H
2
yield is obse ed wi h inc easing space ime, wi h a alue o 95 %
in Fig. 3 and almos he same H
2
pu i y (a ound 99 %) in bo h uns. The
combined e ec o inc easing he CO
2
cap u e capaci y by inc easing he
amoun o dolomi e and inc easing he ex en o e o ming and WGS
eac ions by inc easing he amoun o ca alys con ibu es o his esul .
Fu he mo e, a e he CO
2
cap u e pe iod, i is obse ed ha he s a-
bili y o he ca alys has inc eased wi h inc easing space ime alue. As a
esul , he H
2
yield emains high (abou 72 %) a e 5 h o ope a ion,
compa ed o abou 25 w % unde he same condi ions bu wi h hal he
space ime ( ed cu e in Fig. 1b). This highe s abili y is consis en wi h
p e ious s udies on bio-oil SR, ha show an imp o emen in ca alys
s abili y wi h inc easing space ime alues due o he lowe concen a-
ion o oxygena es in he eac ion medium, as a esul o he highe
ex en o oxygena e e o ming eac ions [63,64].
3.2. Cha ac e iza ion o ca alys and so ben used unde di e en
condi ions
The esul s in sec ion 3.1 show ha he s abili y o he ca alys de-
pends on he eac ion condi ions. In o de o iden i y he causes
esponsible o he deac i a ion and i s dynamics du ing he CO
2
cap u e
pe iods in Fig. 1, samples o he ca alys used a e 50 min on s eam
(end o he CO
2
cap u e pe iod) and a e 300 min on s eam (end o he
uns in Fig. 1, comp ising he pos -cap u e pe iod) we e analyzed using
di e en echniques (TPO, TPR, XRD, N
2
adso p ion-deso p ion and
SEM). A e each eac ion, he h ee componen s o he ca aly ic bed
(ca alys /so ben /ine solid) we e sepa a ed by sie ing, in o de o
Fig. 3. E olu ion wi h ime on s eam o ca bon con e sion and p oduc yields.
Reac ion condi ion: 600
◦C; space ime, 0.3 g
ca alys
⋅h/g
oxygena es
; dolomi e/
ca alys mass a io, 10; S/C, 3.4. Colo ed zones: CO
2
cap u e pe iod (g een
shade) and b eak h ough pe iod (blue shade). (Fo in e p e a ion o he e e -
ences o colo in his igu e legend, he eade is e e ed o he Web e sion o
his a icle.)
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1532
cha ac e ize he used ca alys sepa a ed om he so ben o ine solids.
3.2.1. Analysis o coke deposi s
The Tempe a u e P og ammed Oxida ion (TPO) o he used ca alys
p o ides in o ma ion on he o al con en o coke deposi ed, and he
na u e and/o loca ion o he coke on he ca alys [63,65]. Fig. 4 shows
he TPO p o iles and he co esponding coke con en (w %) o he
ca alys used a di e en empe a u es. I should be no ed ha he scale
o he wo g aphs is di e en . G aph (a) co esponds o he uns pe -
o med up o he b eak h ough egime wi h CO
2
cap u e (50 min ime on
s eam), whe eas g aph (b) co esponds o he ca alys used in he 300
min du a ion es s (co esponding o he eac ions in Fig. 1). Fo com-
pa ison, he TPO p o iles o he ca alys s used in he SR es s (wi hou
so ben ) a e plo ed in Fig. S6. In line wi h p e ious esul s in he
li e a u e on coke deposi ion on suppo ed Ni-based ca alys s used in he
SR o oxygena es and bio-oil [58,63,65], he TPO p o iles in Fig. 4 show
wo di e en combus ion domains, co esponding o coke bu ning
below 500 ◦C o abo e 500 ◦C. These ac ions can be a ibu ed o he
di e en na u e o he coke: amo phous ca bon (bu ning a low em-
pe a u e) and ca bon ilamen s o g aphi ic s uc u es (bu ning a high
empe a u e) [66,67]. I has also been claimed ha he loca ion o he
combus ion peaks depends on he loca ion o he coke, wi h coke
deposi ed on he me al su ace bu ning a a lowe empe a u e han coke
deposi ed on he suppo [61,65].
The esul s in Fig. 4 e eal a ema kable e ec o he e o ming
empe a u e on he coke con en and na u e in he wo g oups o uns.
Fo he ca alys s used only in he CO
2
cap u e pe iod (Fig. 4a), he coke
bu ning below 500 ◦C (p esumably amo phous coke deposi ed on he Ni
si es) is he only coke ac ion a a e o ming empe a u e o 550 ◦C. As
he eac ion empe a u e inc eases, he o al coke con en dec eases and
so does his amo phous coke ac ion, whe eas a ac ion bu ning abo e
500 ◦C appea s a 600 ◦C and becomes he p edominan ac ion abo e
650 ◦C. These ends indica e ha he deposi ion o coke is a enua ed,
bu he deposi ed coke is mo e e ac o y as he e o ming empe a u e
inc eases. Compa ing he TPO p o iles in Fig. 4a and b, i can be seen
ha a a e o ming empe a u e o 550 ◦C he na u e o he deposi ed
coke is simila o bo h alues o ime on s eam, al hough he coke
con en inc eases o he highe ime on s eam (Fig. 4b). Fo he
e o ming empe a u es o 650 and 700 ◦C, he TPO p o iles o he coke
deposi ed du ing he CO
2
cap u e pe iod (Fig. 4a) and a 300 min e-
ac ion ime (Fig. 4b) a e also simila , especially o he eac ions a
700 ◦C. Bu in e es ingly, he TPO p o iles o he ca alys samples used a
600 ◦C in he CO
2
cap u e pe iod and a he end o he eac ion (a e
sa u a ion o he dolomi e) di e signi ican ly, so ha he s uc u ed
coke bu ning a high empe a u e is he majo coke ac ion o he la e
sample (used o longe eac ion ime a e dolomi e sa u a ion).
Mo eo e , he TPO p o iles o he ca alys used a 600, 650 and 700 ◦C
o 300 min in he p esence o sa u a ed dolomi e (Fig. 4b) di e
signi ican ly om hose in Fig. S6, co esponding o he SR (wi hou
dolomi e) o he bio-oil. All hese esul s show a ema kable ole o
dolomi e in coke deposi ion, wi h he ole o dolomi e a a e o ming
empe a u e o 600 ◦C being di e en om ha a highe e o ming
empe a u es.
The esul s on he na u e o he coke (amo phous o ilamen ous)
ob ained by TPO analysis ha e been co obo a ed by SEM images o he
ca alys used a selec ed ope a ing condi ions, shown in Fig. S7. Thus, in
he SEM images o he ca alys pa icles used a 600 ◦C and wi h a
ca alys /mass a io o 10 (Figs. S7c and d), a high p esence o ca bon
ilamen s is obse ed. Con e sely, he e a e no ca bon ilamen s in he
ca alys used a 550 ◦C wi h a so ben /ca alys mass a io o 10
(Figs. S7a and b), and he p esence o ca bon ilamen s is signi ican ly
lowe han a 600 ◦C in he ca alys pa icles used a 700 ◦C wi h a
so ben /ca alys mass a io o 20 (Figs. S7e and ).
Conce ning he e ec o space ime on coke deposi ion, he com-
pa ison o he TPO p o iles o wo space ime alues a 600 ◦C
(con inuous and dashed ed lines in Fig. 4b) con i ms he well-known
ac ha coke deposi ion in bio-oil e o ming is signi ican ly a enu-
a ed wi h dec easing oxygena e concen a ion (p ecu so s o amo phous
coke) [61,64]. In his case, bo h amo phous and ilamen ous coke o -
ma ion a e no ably a enua ed by doubling he space ime.
3.2.2. De e io a ion o he used ca alys p ope ies
To assess he sin e ing o Ni c ys als and he c ys alline s a e o he
Fig. 4. TPO p o iles and coke con en (w %) o he ca alys used a di e en empe a u es in he CO
2
cap u e pe iod (50 min on s eam) (a), and a e 300 min on
s eam (b). Reac ion condi ions: space ime, 0.15 g
ca alys
⋅h/g
oxygena es;
dolomi e/ca alys mass a io, 10; S/C, 3.4. *space ime, 0.30 g
ca alys
⋅h/g
oxygena es
o dashed ed
line in g aph b. (Fo in e p e a ion o he e e ences o colo in his igu e legend, he eade is e e ed o he Web e sion o his a icle.)
Fig. 5. XRD pa e ns o esh- educed and used ca alys samples in he SESR
uns a di e en empe a u es. Reac ion condi ions: space ime, 0.15 g
ca alys
⋅h/
g
oxygena es
(*0.30 g
ca alys
⋅h/g
oxygena es
); so ben /ca alys mass a io, 10; S/C, 3.4.
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1533
coke deposi s, Fig. 5 shows he XRD pa e ns o he esh- educed and
used ca alys samples a di e en empe a u es unde he condi ions o
Fig. 1 (so ben /ca alys mass a io o 10). XRD esul s co esponding o a
so ben /ca alys mass a io o 20 a e shown in Fig. S8. Peaks co e-
sponding o Ni
0
(di ac ion angle a 44.5◦in (111) plane, 51.8◦in (200)
plane and 75.5◦in (110) plane) and Al
2
O
3
c ys als (37.3◦, 45.6◦and
66.8◦) a e iden i ied in he esh- educed ca alys . The same di ac ion
peaks a e obse ed in he ca alys used in he SESR o bio-oil. Simila ly,
he p esence o NiO is no de ec ed, con i ming he high educing ca-
paci y o he eac ion medium o main ain he ac i e me al in a educed
s a e. The a e age size o he Ni c ys als o he esh- educed and used
ca alys s (Table 2) was calcula ed om he di ac ion peak a 2θ =51.8◦
using he Sche e equa ion. The es ima ed alue is 15.0 nm o he
esh- educed ca alys and inc eases sligh ly in he ange o 15–20 nm
o he used ca alys samples. Based on hese esul s, he e is no e idence
o a signi ican sin e ing phenomenon o Ni c ys als.
The XRD pa e ns o he used ca alys samples also p o ide in o -
ma ion abou he coke deposi s. The p esence o a b oad peak a a
di ac ion angle 2θ =26◦is obse ed in he ca alys used a 600 ◦C
(mo e p onounced a he lowes space ime). This peak co esponds o
high c ys allini y coke (g aphi ic ca bon), usually iden i ied in ca alys s
used in he SR o hyd oca bons [68] and pu e oxygena e compounds,
such as e hanol and phenol [69], ace one [70], ace ic acid [71]. This
esul is cohe en wi h he TPO p o iles in Fig. 4b and he SEM images in
Figs. S7c and d, and con i ms he s uc u ed na u e o he main com-
bus ion peak in he coke deposi ed a 600 ◦C, whose amoun dec eases
signi ican ly wi h inc easing space ime. The absence o he di ac ion
peak a 26◦in he coke deposi ed a 550 ◦C co obo a es i s amo phous
na u e. Fo he ca alys s used in he SESR abo e 650 ◦C, al hough some
s uc u ed coke is deposi ed acco ding o he TPO p o iles (Fig. 4b), i s
small amoun does no allow i o be de ec ed in he co esponding XRD
pa e ns in Fig. 5.
In o de o e alua e he possible de e io a ion o he po ous s uc u e
o he ca alys and i s con ibu ion o he deac i a ion, N
2
adso p ion-
deso p ion iso he ms ha e been ob ained om esh- educed and used
samples (Fig. S9), om which he physical p ope ies (BET su ace a ea,
mean po e diame e and po e olume) ha e been de e mined (Table 2).
All samples in Fig. S9 show a ype IV iso he m, cha ac e is ic o meso-
po ous ma e ials, associa ed wi h capilla y (po e) condensa ion aking
place in mesopo es and he occu ence o mul ilaye adso p ion. The
iso he ms o esh- educed ca alys and samples used a he lowe
empe a u e (Fig. S9a) p esen a hys e esis o he ype H2, which is
a ibu ed o a di e ence in mechanism be ween condensa ion and
e apo a ion p ocesses occu ing in po es wi h na ow neck and wide
bodies. Ma e ials wi h H2 hys e esis a e o en diso de ed and he dis-
ibu ion o po e size and shape is no well de ined. Con e sely, a H3-
ype hys e esis cycle is obse ed in he iso he ms o he ca alys s used
abo e 600 ◦C (Figs. S9c,e,g), which shows no limi ing adso p ion a high
P/P
0
, and is associa ed o agg ega ed pla e-like pa icles ha gi e ise o
sli -shaped po es [72].
O e all, he e is an e olu ion o mesopo osi y wi h a sligh p o-
g essi e dec ease om he highes empe a u e o 550 ◦C. Fo mos o
he used samples, he o al olume adso bed a high p essu es (P/P
0
≈1)
is lowe han ha o he esh- educed sample, demons a ing he pa ial
blockage o he mesopo es, which is mo e p onounced a he lowes
e o ming empe a u e. Fo all e o ming empe a u es s udied, he BET
su ace a ea o he used ca alys dec eases wi h inc easing dolomi e
p esence, e idencing a pa ial blockage o he mesopo ous s uc u e o
he ca alys s, which is p omo ed a highe so ben loadings, p obably
due o he high deposi ion o a low po osi y coke. On he o he hand, he
inc ease in BET su ace a ea obse ed o he used samples a 600 ◦C
(mo e p onounced a high space ime) compa ed o he esh- educed
ca alys , sugges s ha he e is an addi ional po osi y c ea ed by he
ilamen ous coke deposi ed on hese samples. O e all, he a e age po e
diame e and po e olume dec ease compa ed o he esh- educed
sample (mo e no iceable a low empe a u es, 550 and 600 ◦C), p ob-
ably due o he blockage o pa o he po ous su ace. To sum up, he
di e en BET su ace a ea alues (highe o lowe han hose o he esh
ca alys , depending on he eac ion condi ions) indica e he o ma ion o
bo h: i) ilamen ous coke (a 600 ◦C), which is p obably s acked on he
su ace o he ca alys , causing an inc ease in he BET su ace a ea, and
ii) coke clogging he po ous s uc u e, causing a dec ease in he BET
su ace a ea.
3.3. Cyclic ope a ion wi h join egene a ion o he ca alys and dolomi e
The e iciency o he join eac i a ion o ca alys and so ben has
been s udied unning 7 eac ion- egene a ion cycles. The condi ions in
he eac ion s ep (es ablished by he esul s in sec ion 3.1) a e 600 ◦C,
space ime o 0.30 g
ca alys
⋅h/g
oxygena es
(co esponding o 1 g o ca alys ),
so ben /ca alys mass a io o 10, S/C mola a io o 3.4, and ime on
s eam o 120 min. These condi ions allow o ope a e wi h a su icien ly
high CO
2
cap u e ime (abou 60 min), wi h he co esponding so ben /
ca alys mass a io o 10. The egene a ion s ep consis ed o coke com-
bus ion and CO
2
emo al in an ex e nal o en a 850 ◦C o 4 h in ai
a mosphe e (su icien ime o comple e emo al o coke deposi s and
CO
2
). A e each egene a ion s ep, he ca alys +so ben we e sub-
jec ed o a educ ion s ep (a 850 ◦C o 4 h in H
2
–N
2
(7 ol% H
2
)) o
ob ain he ac i e Ni
0
pa icles. Fig. 6 shows he e olu ion wi h ime o
he eac ion indices in he 1s eac ion s ep ( esh ma e ial) (g aph a)
and in he 7 h eac ion s ep (g aph b), and Fig. 7 shows he e olu ion o
he H
2
yield in he CO
2
cap u e pe iod in he successi e eac ion s eps.
When compa ing he 1s and 7 h eac ion s eps, a a he simila pe -
o mance o he ca alys s/so ben bed is obse ed, indica ing a good
o e all eco e y in he egene a ion s ep o bo h solids, he ca alys
ac i i y and he CO
2
so p ion capaci y o he dolomi e. Ne e heless, he
H
2
yield in he 7 h eac ion s ep is sligh ly lowe han ha ob ained wi h
he esh ca alys in he 1s eac ion, and a small CH
4
yield is obse ed in
he CO
2
cap u e pe iod.
As obse ed in Fig. 7, his sligh dec ease in H
2
yield in he CO
2
cap u e pe iod occu s in he i s wo eac ion- egene a ion cycles, bu
he ca alys /so ben bed eaches a s able pe o mance om he 3 d hi d
eac ion s ep onwa ds. The possible causes o he dec ease in H
2
yield in
he CO
2
cap u e pe iod obse ed in he i s wo cycles a e he incom-
ple e eco e y o he CO
2
so p ion capaci y o he so ben and/o o he
ac i i y o he e o ming and WGS eac ions o he ca alys . I is in e -
es ing o no e in Fig. 6 ha he du a ion o he cap u e pe iod in he 7 h
eac ion s ep is e en sligh ly longe han in he 1s s ep, sugges ing ha
he dolomi e has ully eco e ed i s CO
2
cap u e capaci y. This esul can
be obse ed mo e clea ly in Fig. 8, which shows he e olu ion o e ime
o he CO
2
yield in di e en eac ion s eps (1s , 2nd, 3 d and 7 h). As
obse ed, he du a ion o he CO
2
cap u e pe iod is sho e in he i s
cycles and emains cons an and longe a e he hi d cycle.
Table 2
Physico-chemical p ope ies (a e age Ni
0
c ys al size (d
Ni
), BET su ace a ea
(S
BET
), po e olume (V
po e
) and mean po e diame e (d
po e
) o esh- educed
ca alys and used in he SESR o aw bio-oil a di e en empe a u es and so -
ben /ca alys mass a io. Reac ion condi ions: space ime, 0.15 g
ca alys
⋅h/
g
oxygena es
; S/C, 3.4.
Ca alys Tempe a u e
(◦C)
d
Ni
(nm)
S
BET
(m
2
/g)
V
po e
(cm
3
/g)
d
po e
(nm)
F esh- educed – 15 65.1 0.24 13.1
Used wi h so ben /
ca alys mass a io
o 10
550 15 62.3 0.11 7.1
600 16 69.9 0.18 13.4
600
a
20 79.6 0.25 13.4
650 16 60.6 0.20 14.4
700 16 65.0 0.26 15.4
Used wi h so ben /
ca alys mass a io
o 20
550 16 58.9 0.13 8.3
600 16 52.1 0.13 10.8
650 16 55.5 0.16 12.0
700 17 57.9 0.21 13.1
a
Space ime, 0.30 g
ca alys
⋅h/g
oxygena es
.
L. Landa e al.
In e na ional Jou nal o Hyd ogen Ene gy 58 (2024) 1526–1540
1534
In e es ingly, he mesopo osi y o he dolomi e dec eases along he
eac ion- egene a ion cycles, as shown by he esul s in Table S1. The
BET su ace a ea and po e olume o he sa u a ed dolomi e a e he 7 h
eac ion dec ease signi ican ly compa ed o he esh-calcined dolomi e
sample, while he mean po e diame e inc eases, indica ing comple e
so ben sa u a ion due o CO
2
cap u e wi h a selec i e blockage o he
na owe po es. A e he egene a ion o he dolomi e by deca bon-
a ion, he mesopo osi y is only pa ially es o ed, so ha he su ace a ea
and he mean po e olume a e abou hal o hose o he esh-calcined
sample, which shows he pa ial blockage o he so ben mesopo es a e
he egene a ion pe iod, no being able o ully eco e he ini ial
s uc u e o he esh-calcined dolomi e. Ne e heless, he dolomi e e-
ains i s CO
2
cap u e capaci y. Consequen ly, he dec ease in H
2
yield
du ing he CO
2
cap u e pe iod in he i s cycles (Fig. 7) should be
a ibu ed o he i e e sible deac i a ion o he ca alys in he i s wo
eac ion- egene a ion cycles and no o he de e io a ion o he CO
2
cap u e capaci y o he dolomi e, as s udied in sec ion 3.4.
3.4. Cha ac e iza ion o ca alys used in he cyclic ope a ion
The ca alys used in some eac ion s eps, and also he ca alys e-
gene a ed a e he 7 h eac ion, ha e been cha ac e ized by se e al
echniques (TPO, TPR, XRD, N
2
adso p ion-deso p ion and XPS) (sec ion
2.2) in o de o analyze he possible changes in he physico-chemical
p ope ies o he ca alys along he cyclic ope a ion and o co ela e
hem wi h he kine ic pe o mance obse ed in sec ion 3.3.
3.4.1. Coke deposi ion
Fig. 9 depic s he TPO p o iles o he ca alys used a e he 1s and
7 h eac ion s eps. Fo bo h samples, he e is a majo combus ion peak
bu ning abo e 500 ◦C co esponding o ilamen ous and s uc u ed coke,
wi h a low amoun o amo phous coke bu ning below 500 ◦C. The coke
con en dec eases in he successi e eac ion s eps ( om 17.2 w % in he
i s eac ion o 12.8 w % in he se en h eac ion), especially he ila-
men ous coke, whe eas he amo phous ca bon emains ai ly cons an .
The di e ence in coke deposi ion in Fig. 9 is consis en wi h he sligh ly
lowe ca alys ac i i y obse ed a e he i s wo eac ion- egene a ion
Fig. 6. E olu ion wi h ime on s eam o ca bon con e sion and p oduc s yields in he 1s (a) and 7 h (b) eac ion s eps (wi h in e media e egene a ions). Reac ion
condi ions: 600
◦C; space ime, 0.3 g
ca alys
⋅h/g
oxygena es
; so ben /ca alys mass a io, 10; S/C, 3.4. Regene a ion condi ions: calcina ion wi h ai in ex e nal o en a
850 ◦C o 4 h. Colo ed zones: CO
2
cap u e pe iod (g een shade) and b eak h ough pe iod (blue shade). (Fo in e p e a ion o he e e ences o colo in his igu e
legend, he eade is e e ed o he Web e sion o his a icle.)
Fig. 7. E olu ion o H
2
yield in he CO
2
cap u e pe iod in successi e eac ion
s eps (wi h in e media e egene a ions). Reac ion and egene a ion condi ions
o Fig. 6.
Fig. 8. CO
2
yield e olu ion wi h ime on s eam in successi e eac ion s eps
(wi h in e media e egene a ion). Reac ion and egene a ion condi ions
o Fig. 6.
Fig. 9. Compa ison o he TPO p o iles and coke con en (w %) deposi ed on
samples used in he 1s and 7 h eac ion s eps. Reac ion and egene a ion
condi ions o Fig. 6.
L. Landa e al.
