In ensi ying glyce ol s eam e o ming on a monoli h ca alys :
a eac ion kine ic model
L.F. Bobadilla1, V. Blay2, A. Ál a ez 1,†, M.I. Domínguez1, F. Rome o-Sa ia1, M.A.
Cen eno1,* and J.A. Od iozola1
1 Ins i u o de Ciencia de Ma e iales de Se illa, Uni e sidad de Se illa-Consejo Supe io
de In es igaciones Cien í icas, A enida Amé ico Vespucio s/n, 41092 Se illa, España
2 Ins i u o de Tecnología Química (UPV-CSIC), Uni e si a Poli ècnica de València,
Consejo Supe io de In es igaciones Cien í icas, A enida de los Na anjos s/n, 46022
Valencia, España
AUTHOR INFORMATION
Co esponding Au ho
To whom co espondence should be add essed:
* D . Miguel A. Cen eno ([email p o ec ed])
P esen Add esses
† Ins i u e o Chemical Resea ch o Ca alonia (ICIQ), The Ba celona Ins i u e o Science
and Technology, A . Països Ca alans 16, 43007 Ta agona, Spain
ABSTRACT. In his wo k, a s uc u ed monoli hic ca alys has been es ed unde a
wide ange o condi ions (pa ial p essu e, esidence ime, empe a u e and ime-on-
s eam), wi h he aim o modeling i s kine ic beha io and assessing i s economic and
upscaling po en ial. We ha e de eloped a sequen ial model o help us in e p e bo h
main ends and salien ea u es. Unexpec ed beha io was ound o ce ain pa ame e
alues, which led us o conside kine ic pa asi ic e ec s such as mass o hea ans e
limi a ions. By independen ly in oking hese e ec s, a concilia o y iew o he esul s
obse ed could no be eached. A combined explana ion may p o e success ul, al hough
o e i ing could no be uled ou a his poin . Mo e impo an ly, howe e , he obse ed
salien ea u es o his s able and selec i e monoli h ca alys may hold po en ial o
p ocess in ensi ica ion o glyce ol s eam e o ming, hus con ibu ing o a mo e
sus ainable indus y.
Keywo ds: Ni monoli h ca alys , kine ic modeling, mass and hea ans e
1. In oduc ion
In ecen yea s, biodiesel has a ac ed a conside able a en ion wo ldwide as an
al e na i e enewable uel o eplace adi ional pe oleum diesel p oduc because o
limi ed ese es o adi ional ossil esou ces, he ins abili ies o c ude oil p ice and he
conce ns o e g eenhouse gases emissions [1, 2]. Du ing he p oduc ion o biodiesel by
ca aly ic anses e i ica ion eac ion om ege able oils o animal a s, a la ge amoun
o glyce ol is o med as byp oduc . As a esul o he inc eased biodiesel p oduc ion
du ing he las yea s, a sa u a ed ma ke o glyce ol has p o oked an expec ed all in
glyce ol p ices; he e o e, inding e ec i e and economical ways o con e sion o
glyce ol in o use ul p oduc s is necessa y. Recen s udies ha e sugges ed ha an
a ac i e idea would be o p oduce hyd ogen om glyce ol ia ca aly ic e o ming
p ocesses, hus adding alue o he glyce ol su plus [3, 4].
Glyce ol s eam e o ming (SR) has been demons a ed o be an e ec i e me hod o
hyd ogen p oduc ion wi h high selec i i y, and i can be pe o med in gas o aqueous
phase depending on he eac ion condi ions, mainly empe a u e and p essu e [5]. S eam
e o ming, al hough highly endo he mic, is p e e able since he low p essu e a ou s
selec i i y o hyd ogen. Nickel is he mos used ac i e me al o glyce ol s eam
e o ming because o i s good ac i i y o C-C, C-O and C-H bond clea age, as well as
o i s abili y o emo e he adso bed CO by wa e gas shi (WGS) eac ion [6-17]. In
addi ion o hei op imal pe o mance, Ni-based ca alys s a e p e e ed due o he low
cos and high a ailabili y. Howe e , he supp ession o coke deposi ion o enhance
ca aly ic s abili y s ill emains a majo challenge. Di e en s udies ha e sugges ed ha
he combina ion o basic p omo e s (Mg, Ce) and a g oup IV alloying elemen such as
Sn a o s coke- esis ance o Ni ca alys s in he SR o hyd oca bons [18-21]. Typically
hese addi i es help o a oid ca bon deposi ion and enhance hei ca aly ic s abili y.
In a p e ious pape [22], we demons a ed ha a Ni-based monoli h ca alys is e y
ac i e and s able o hyd ogen p oduc ion by glyce ol s eam e o ming. Coke o ma ion
was no obse ed in he monoli h ca alys hanks o he s ong in e ac ion be ween he
ca alys pa icles and he alumina laye in he monoli h. The u iliza ion o me allic
monoli h ca alys s in p ac ical applica ions is e y impo an o con ol he hea and
mass anspo p ope ies since he majo i y o ca aly ic eac ions depend on hea
ans e , luid dynamics, and su ace eac ion kine ics. In addi ion o his, me allic
monoli hs a e excellen models o he ini ial s udies o mic ochannel eac o s in which
coupled endo- and exo he mic eac ions a e used o con olling he p ocess selec i i y
[23].
Conce ning he eac o modeling, se e al s udies on he kine ics and he eac ion
mechanism o glyce ol s eam e o ming ha e been ca ied ou in he las en yea s [14,
24-27]. Howe e , mos o he epo ed kine ics a e based on powe -law models and
e e o powde ed ca alys s. To he bes o ou knowledge, he e a e only wo wo ks
published e y ecen ly ha desc ibe a kine ic model o glyce ol s eam e o ming using
a wall-coa ed ca aly ic mic ochannel [28] and a kine ic s udy o au o he mal e o ming
o glyce ol in a dual laye monoli h ca alys [29], espec i ely. Liu e al. ound ha he
eac ion a e o glyce ol e o ming was no limi ed by mass ans e wi hin he ca alys
washcoa ed laye and he su ace eac ion was he a e con olling s ep Langmui -
Hinshelwood kine ics conside ing non-dissocia i e adso p ion o glyce ol and
dissocia i e adso p ion o s eam we e p oposed o desc ibe he au o he mal e o ming o
glyce ol in his dual laye monoli h ca alys [29].
This wo k in ol es a de ailed s udy wi h he aim o unde s anding he kine ic
beha io o he well-pe o ming Ni-based monoli h ca alys unde a wide ange o
condi ions (pa ial p essu e, esidence ime, empe a u e and ime-on-s eam), looking
ou o po en ial upscaling. We an icipa e ha unusual e ec s we e ound in his s udy,
which we a emp ed o a ionalize based on he da a a hand and p io knowledge. These
esul s may well dese e addi ional wo k, as hey hold p omise o imp o ed economics
a comme cial scale.
2. Expe imen al
2.1. Syn hesis o NiSn/CeO2-MgO/Al2O3 ca alys
NiSn/CeO2-MgO/Al2O3 ca alys was p epa ed by he imp egna ion me hod using a
syn hesis p ocedu e p e iously epo ed [19, 22]. Sphe ical pelle s o γ-Al2O3
(Sphe ali e SCS505) wi h 2.5 mm diame e we e milled in a high-ene gy ball mill in
o de o ob ain γ-Al2O3 wi h a pa icle size o 7–8 μm, which was used as suppo o
p epa e he ca alys used in his s udy. All he p ecu so s we e imp egna ed
simul aneously on he alumina suppo wi h an aqueous solu ion con aining app op ia e
quan i ies o Mg(NO3)2·6H2O (Sigma-Ald ich), Ce(NO3)3·6H2O (Fluka),
Ni(NO3)2·6H2O (Pan eac) and SnCl2 anhyd ous (Fluka), ollowed by d ying o e nigh
a 120°C and inal calcina ion in ai a 700°C o 12 h. The nominal me al loading was
26 w .% wi h a Ni- o-Sn a omic a io equal o 6.
2.2. Me allic monoli h manu ac u e
Me allic shee s o AluCh om YH ® wi h 50 μm hickness (Good ellow) we e used as
subs a es o p epa a ion o he me allic monoli h. Cylind ical monoli hs we e p epa ed
by olling la and co uga ed oils al e na i ely a ound a spindle (L = 30 mm, d = 16
mm, V = 6 cm3, cell densi y 170 cells/cm2), ollowing he s eps p esen ed in Fig. 1. This
geome y was op imized p e iously o ensu e a mo e e icien hea and mass ans e
[30-32], as will be discussed below. The me allic monoli h was he mally p e ea ed in
ai a 900 ºC o 22 h in o de o gene a e an adhe en α-Al2O3 laye [33]. This ea men
ensu es a good adhe ence o he ca alys o he me allic subs a e.
2.3.Washcoa ing p ocess
The i s equi emen o an e ec i e washcoa ing p ocess is ob aining s able slu ies.
In o de o achie e a uni o m coa ing o he me allic subs a e i is necessa y o p epa e
he slu y con olling he pa ame e s a ec ing i s s abili y, such as pa icle size o he
ma e ial o be dispe sed, solid con en o he suspension, pH and iscosi y. Fu he mo e,
he use o addi i es in he slu y o mula ion a emp s o imp o e he ca alys adhe ence
and he washcoa d ying p ocess. Fo ins ance, he addi ion o colloidal alumina
(Nyacol, 20 w % Al2O3), ha p esen s a na ow pa icle size dis ibu ion, enhances
ema kably he ca alys adhe ence owing o he smalle pa icles in e locking wi h he
la ge ones [34]. A e se e al ials o slu y o mula ion o washcoa ing, we achie ed
a s able aqueous slu y o he ca alys p e iously p epa ed as desc ibed in sec ion 2.1.
The ca alys was milled in an aga e mo a achie ing an a e age pa icle size o 12.7
μm. Since he isoelec ic poin (IEP) o he ca alys is equal o 7.8, he pH was ixed a

3.5 using ace ic acid o ensu e high alues o ze a po en ial, which ansla es in o high
epulsions be ween he pa icles and imp o ed s abili y o he slu ies. The ollowing
p opo ions o ca alys and colloidal alumina we e selec ed: 61 w % ca alys con en and
39 w % colloidal alumina, being he o al solid con en o he aqueous suspension ca. 24
w %. The me allic monoli h subs a e was dipped in o he slu y o 60 s, wi hd awn a a
cons an speed o 3 cm min-1 and hen he excess o suspension in he mic ochannels
was emo ed by cen i uga ion a 500 pm o 5 min. This p ocedu e was epea ed
se e al imes wi h in e media e d ying s eps a 120 C o 30 min be ween coa ings un il
ca. 200 mg o he ca alys was deposi ed. Finally, he coa ed s uc u ed suppo was
calcined a 700°C o 12 h. This p ocedu e esul ed in a ca alys loading on he monoli h
o 0.56 mg cm−2 a e ou washcoa ing s eps. The adhe ence o he ca aly ic laye was
e alua ed wi h he ul asound es [35] achie ing a alue be e han 97%
2.4. Ca aly ic pe o mance: s abili y es
The ca aly ic measu emen s we e ca ied ou in a compu e ized comme cial
Mic oac i i y Re e ence ca aly ic eac o (PID Eng&Tech), employing a Has elloy C-
276 ubula eac o (Au ocla e Enginee s) wi h 17 mm in e nal diame e , which
con ained he monoli h loaded wi h 200 mg o ca alys . P io o eac ion, he ca alys
was educed in si u a 750°C o 1 h wi h 100 mL min−1 o H2 (50%, / in ine ). The
expe imen al uns we e conduc ed a 750 °C and a mosphe ic p essu e wi h a s eam- o-
ca bon a io o 4 (ps eam ≈ 0.32 a m; pG ≈ 0.027 a m) and 100 NL g−1 h−1 con ac ime in
o de o ensu e an op imal ca aly ic ac i i y [19]. A he eac o ou le a Pel ie gas–
liquid sepa a o was i ed allowing he analysis o gas and liquid phase p oduc s
sepa a ely. Gas p oduc s we e analyzed on line using a mic oGC (Va ian 4900)
equipped wi h Po apak Q and MS-5A columns. Liquid p oduc s we e analyzed by
HPLC (Va ian 356-LC) equipped wi h a e ac i e index de ec o and a Hi-Plex H
column wi h deionized wa e as eluen .
2.5. Kine ic measu emen s
Kine ic da a o he glyce ol s eam e o ming eac ion we e collec ed a a mosphe ic
p essu e and a eac ion empe a u e o 600 ºC in o de o achie e in e media e
con e sions. Glyce ol-wa e mix u es wi h composi ions anging om 30 w .% o 45
w .% (co esponding o s eam- o-ca bon a ios o 1.7 o 4) we e ed o ensu e a
s oichiome ic excess o s eam. This s eam was dilu ed in o lowing N2 o yield s eam
pa ial p essu es in he ange ps eam = 0.11-0.32 a m. The esul ing con ac imes
employed in he expe imen al uns a e hus be ween 100 and 230 NL g-1 h-1. Thanks o
he high cell densi y o he monoli h designed, hese ansla e in o gas hou ly space
eloci y (GHSV) alues o 3500-7500 h-1. High GHSV igu es a e in e es ing om an
indus ial poin o iew, as hey would allow p ocess in ensi ica ion (PI) wi h mo e
compac eac o designs. Resul s p esen ed in his wo k co espond o a ime-on-s eam
o 7 min, al hough he ends we e ound consis en wi h esul s a e 1 h, and e en
much longe . A he indus ial scale a highe empe a u e would be used o ensu e a
maximal con e sion and ca aly ic s abili y such as 750 ºC (sec ion 2.4).
2.6. Model simula ion
As will be explained below, models used o help us in a ionalizing he da a comp ise
coupled di e en ial and algeb aic equa ions, namely p ope y balances along wi h a e
equa ions, which need o be sol ed simul aneously. To his end, models we e
implemen ed on Ma lab 7.11.0 as cus om made sc ip s. To sol e hem, we used he
buil -in sol e ode15s. I is a a iable-o de sol e based on nume ical di e en ia ion
o mulas. I was chosen because o i s abili y o deal wi h s i p oblems, which we
an icipa ed ha migh a ise i some o he pa hways in he p oposed eac ion scheme
lose ele ance as op imiza ion p oceeds. On op o ha , o pa ame e op imiza ion, a
high-le el mul is a p ocedu e was used, epea ing he op imiza ion algo i hm s a ing
om di e en combina ions o seed pa ame e alues esul ing om a loga i hmic-
ac o ial di ision o he pa ame e space. As o he op imiza ion algo i hm i sel , we
op ed o he minsea ch ou ine in Ma lab, which is a inely coded e sion o he
classical Nelde -Mead algo i hm. The objec i e unc ion (O.F.) in such a minimiza ion
was designed o be:
𝑂.𝐹.=∑[(𝐹𝐶𝑂
𝑒𝑥𝑝 −𝐹𝐶𝑂
𝑐𝑎𝑙𝑐)2+(𝐹𝐻2
𝑒𝑥𝑝 −𝐹𝐻2
𝑐𝑎𝑙𝑐)2] (Eq. 1)
whe e he summa ion ex ends o all expe imen s pe o med. This was chosen on he
basis ha : 1) he es o main p oduc s, namely H2O and CO2 we e ound o be
s oichiome ically ela ed o CO and H2 based on he wo main eac ions p esen ed
below, and 2) bo h H2 and CO a e majo p oduc s de ec ed by he same GC de ec o ,
hus maximizing epea abili y o measu emen s.
3. Resul s and discussion
3.1. Ca aly ic pe o mance
In o de o in es iga e he pe o mance o he monoli h ca alys o glyce ol s eam
e o ming, an ini ial expe imen was pe o med a 750 °C employing a s eam- o-ca bon
a io o 4 and a space eloci y o 100 NL g−1 h−1. Fig. 2 shows ha he monoli hic
ca alys exhibi ed good ac i i y and s abili y in e ms o hyd ogen yield o e 55 hou s o
ime-on-s eam. Only non-condensable p oduc s such as H2, CO2, CH4 and CO we e
o med by glyce ol s eam e o ming (Reac ion (1)), glyce ol decomposi ion (Reac ion
(2)) and WGS (Reac ion (3)). Fo ma ion o CH4 esul s om me hana ion eac ions
(Reac ions (4) and (5)) by coupling o H2 and CO o CO2 o med om glyce ol
decomposi ion.
C3H8O3 + 3H2O → 3CO2 + 7H2 (1)
C3H8O3 → 3CO + 4H2 (2)
CO + H2O ↔ CO2 + H2 (3)
CO + 3H2 ↔ CH4 + H2O (4)
CO2 + 4H2 ↔ CH4 + 2H2O (5)
2CO ↔ CO2 + C (6)
As can be seen in Fig. 2, he CO- o-CO2 mola a io emains p ac ically unal e ed.
This ag ees wi h he absence o ca bon deposi s ha could eac wi h he CO2 o med
acco ding o he Boudoua d eac ion (Reac ion (6)) and al e he CO- o-CO2 mola a io
as he eac ion p oceeds. In a p e ious epo [22], we ha e in es iga ed
comp ehensi ely he phenomena o deac i a ion by ca bon deposi ion as a unc ion o
he ca alys shape o glyce ol s eam e o ming eac ion and we demons a ed ha he
o ma ion o coke on he monoli hic ca alys is minimized. These cha ac e is ics make
his monoli hic ca alys app op ia e o an addi ional s udy o eac ion kine ics.
3.2. Reac ion kine ic model
Fig. 3 ep esen s ini ial glyce ol con e sion o non-condensable p oduc s as a unc ion
o space ime in he monoli hic ca alys . Space ime in his igu e has been e e ed o
he eac ion empe a u e (600 °C) o be mo e sugges i e o he sho esidence imes
ac ually used. Da a poin s a e labelled wi h pa ial p essu es o he eac an s in he eed
whe e Kp is he equilib ium cons an in eac ion (3), which can be es ima ed by
Callaghan’s equa ion [38] neglec ing he ugaci y coe icien s o being close o uni y:
𝑙𝑜𝑔10𝐾𝑃= −2.4198+0.0003855 𝑇+ 2180.6 𝑇−1 (Eq. 18)
In Fig. 7, he app oach o equilib ium o all da a poin s in Fig. 1 3 is ep esen ed as a
unc ion o glyce ol con e sion o e o ming p oduc s. No ice ha as glyce ol is ini ially
e o med on he ca alys i does no yield a p oduc mix u e in equilib ium. Howe e , in
Fig. 7 a clea ly e ol ing end can be disce ned, which would be in acco dance wi h
some in ol emen o he ca alys in he WGS eac ion.
3.3. Hea and mass ans e limi a ions
Rega ding Fig. 3b, one may hink o o he explana ions o hese ends, such as he
possibili y o a he mal unaway occu ing a longe space imes. In his case, as he
con e sion exceeds some h eshold alue, he hea eleased by he eac ion may no be
eadily exchanged h ough he monoli h walls, hus leading o hea ing and sel -
accele a ion o he eac ion. In spi e o he mode a e eac ion en halpy (-265 kJ mol-1),
gi en he high dilu ion o he eac an , good he mal conduc i i y (18 W m-1 K-1) and
high cell densi y (170 cells cm-2) o he me allic monoli h suppo , an unno iced ho spo
is unlikely o be a p oblem. Mo eo e , ansien esul s du ing eac o s a up do no
show an inc ease o con e sion wi h TOS o any change o he u nace empe a u e o
i s wa age.
Ne e heless, con en ional a e equa ions do no su ice o explain he obse ed
beha io , as was seen in Fig. 5. Ano he possibili y is ha he eac an mix u e does no
en e he monoli h su icien ly ho , and hus pa o he monoli h beha es as a hea

exchange , ope a ing a a limi ed eac ion a e. Only owa ds i s end he gas mix u e
would each he ac ual monoli h empe a u e, po en ially leading o a double
imp o emen as con ac ime is inc eased. To inspec his possibili y we de eloped an
al e na i e model in which hea ans e will be assumed p opo ional o he empe a u e
di e ence be ween he monoli h and he luid:
𝑑𝑇𝑓𝑙𝑢𝑖𝑑
𝑑𝑉 =𝜅·( 𝑇𝑚𝑜𝑛𝑜𝑙𝑖𝑡ℎ − 𝑇) (Eq. 19)
No ice ha in his case he p opo ionali y cons an would depend no ably on he
low, al hough i can be aken as a cons an o simula e an app oxima ed p o ile unde
gi en expe imen al condi ions. In his line, o educe he numbe o pa ame e s
depending on empe a u e, glyce ol consump ion a e was simpli ied down o a i s
o de eac ion, whose kine ic cons an depends exponen ially on empe a u e:
𝑑𝐹𝐺
𝑑𝑉 = −𝑘𝑑𝐺,873 ·𝑒𝑥𝑝(−𝐸𝑎𝑑𝐺
𝑅(1
𝑇− 1
873.15))·𝑝𝐺 (Eq. 20)
Fo his simula ion we assumed a ela i ely high ac i a ion ene gy o 110 kJ mol-1 as
a wo s -case scena io. Addi ionally, he equilib ium cons an o he WGS is also highly
empe a u e dependen (Eq. 18). One can obse e ha hese kine ics lead o a ela i ely
so e olu ion o he empe a u e and concen a ion p o iles (Fig. 8) and, consequen ly,
hey alone canno explain he ab up ends obse ed in Fig. 3b. In e es ingly, by
compa ing he CO- o-CO2 a io in Figs. 5 and 8 one can no ice, howe e , ha he e a e
di e ences: i is e en possible o obse e a maximum in he CO yield along he ca alys
in he la e igu e due o he e e sibili y o he WGS eac ion, hus emphasizing he
impo ance o empe a u e con ol i hyd ogen yields a e o be maximized, mo e so in
comme cial scale ope a ions whe e such he mal e ec s a e easie o occu
inad e en ly.
Ano he e ec ha dese es conside a ion is he hyd odynamic beha io o he
sys em. So a , we ha e based ou discussion on pseudohomogeneous kine ics, i.e. by
analogy o eac ions in homogeneous media. Howe e , he ca alys is he e ogeneous,
and he e o e i s ac i i y is limi ed o he su ace o i s channels. In pa icula , eac an s
ha e o di use o he monoli h su ace, and p oduc s need o coun e di use back o he
bulk gas phase (Fig. 9). The ease o his anspo is a ec ed by he bulk eloci y o he
luid s eam, a ec ing he e ec i e hickness o he bounda y laye nex o he ca alys .
On a mo e mac oscopic scale, Reynolds numbe , impinging on axial and adial
dispe sion, will also a ec he gas eloci y p o ile o some ex en . Howe e , hese
anspo coe icien s a e ela ed o he luid eloci y wi h powe -laws, whose exponen s
usually ake alues lowe han 1 [39]. In addi ion, no ice ha he ange o lows s udied
is a sho e han an o de o magni ude and s ill ma ked di e ences ha e been
obse ed. Gi en ha empe a u e di e ences, which (app oxima ely) ollow a law o
di ec p opo ionali y (i.e. a powe o he uni y, Eq. 19) and which a ec exponen ially
he eac ion a e cons an s canno explain some o he ab up changes obse ed, i is
unlikely ha mass anspo limi a ions could p o ide mo e ma ked dependencies.
As has been discussed, in spi e o he ela i ely limi ed pa ame ic space swep in his
s udy, a non-mono onic end has al eady been obse ed. This makes da a ha d o
in e p e e en when physical e ec s a e accoun ed o (di usion, dispe sion,
empe a u e). A combina ion o e ec s migh ce ainly o e an explana ion, al hough
would equi e mo e ex ensi e s udies o a oid o e i ing. Al e na i ely, i may well be
ha an addi ional, unaccoun ed e ec is in luencing he esul s. In ei he case, hese
ends a e ce ainly wo h u he s udy since, as illus a ed in Fig. 3b, could enable
d ama ic imp o emen s in e ms o ca alys ac i i y, allowing in e es ing op ions o
p ocess in ensi ica ion. I is expec ed ha pos -mo em cha ac e iza ion a e each
eac ion cycle along wi h ope ando s udies o he ca alys su ace will allow iden i ying
he pa icula ly ac i e ca aly ic cen e o he glyce ol s eam e o ming eac ion and
unde s anding he in insic eac ion mechanism, ye main aining he ema kable
selec i i y and s abili y o he p esen monoli h, os e ing in his way he indus ial
ealiza ion o his impo an p ocess.
4. Conclusions
We ha e de eloped models o in e p e he beha io o a Ni-based monoli hic ca alys
in he glyce ol s eam e o ming eac ion. A he e ogeneous kine ic model based on a
sequen ial eac ion scheme p esen ed easonable adhe ence o he expe imen al esul s.
Howe e , an unexpec ed beha io was encoun e ed o ce ain pa ame e alues, which
canno be explained by con en ional kine ic pa asi ic e ec s such as ca alys su ace
sa u a ion o he mal unaway. This beha io dese es u he in es iga ion as i may
un a el be e -pe o ming ca alys s o glyce ol s eam e o ming, a p ocess holding
g ea indus ial in e es .
NOMENCLATURE AND ABBREVIATIONS
Va iables, pa ame e s and cons an s
d Monoli h ex e nal diame e / cm
Ea Ac i a ion ene gy / kJ mol-1
F Mola low / mol s-1
GHSV Gas hou ly space eloci y ( o al low, 0 °C, 1 a m) / h-1
k Kine ic cons an / mol s-1 m-3 a m-1 o mol s-1 m-3 a m-2
KP Equilib ium cons an / -
L Monoli h leng h / cm
p Pa ial p essu e / a m
P To al p essu e / a m
Reac ion a e / mol s-1 m-3
R Uni e sal gases cons an / J mol-1 K-1
T Tempe a u e / °C
κ Modi ied global hea ans e coe icien / cm-3
V Monoli h olume / cm3
Γ App oach o equilib ium / -
τ Space ime / s
Subsc ip s
j j h componen
0 Ini ial
A Lump A
B Lump B
d The mal decomposi ion
G Glyce ol
S eam e o ming
WGS Wa e gas shi
ACKNOWLEDGMENTS
The au ho s g a e ully acknowledge he inancial assis ance om he Spanish
Minis e io de Economía y Compe i i idad – MINECO (ENE2013-47880-C3-2-R and
ENE2015-66975-C3-2-R) co- inanced by FEDER unds om he Eu opean Union. L.
F. Bobadilla hanks he “Jun a de Andalucía” o his ellowship wi h he p ojec POG-
TEP01965. V. Blay hanks he suppo om he Valencian Minis y o Educa ion.
REFERENCES
[1] G. Baska , R. Aiswa ya, T ends in ca aly ic p oduc ion o biodiesel om a ious eeds ocks,
Renew. Sus ain. Ene gy Re . 57 (2016) 496-504.
[2] D.M. Ma inko ić, M.V. S anko ić, A.V. Veličko ić, J.M. A amo ić, M.R. Miladino ić,
O.O. S amenko ić, V.B. Veljko ić, D.M. Jo ano ić, Calcium oxide as a p omising
he e ogeneous ca alys o biodiesel p oduc ion: Cu en s a e and pe spec i es, Renew. Sus ain.
Ene gy Re . 56 (2016) 1387-1408.
[3] J.M. Sil a, M.A. So ia, L.M. Madei a, Challenges and s a egies o op imiza ion o glyce ol
s eam e o ming p ocess, Renew. Sus ain. Ene gy Re . 42 (2015) 1187-1213.
[4] B. Dou, Y. Song, C. Wang, H. Chen, Y. Xu, Hyd ogen p oduc ion om ca aly ic s eam
e o ming o biodiesel byp oduc glyce ol: Issues and challenges, Renew. Sus ain. Ene gy Re .
30 (2014) 950-960.
[5] P.R.d.l. Piscina, N. Homs, Use o bio uels o p oduce hyd ogen ( e o ma ion p ocesses),
Chem. Soc. Re . 37 (2008) 2459-2467.
[6] S. Shao, A.-W. Shi, C.-L. Liu, R.-Z. Yang, W.-S. Dong, Hyd ogen p oduc ion om s eam
e o ming o glyce ol o e Ni/CeZ O ca alys s, Fuel P ocess. Technol. 125 (2014) 1-7.
[7] E. Gallegos-Suá ez, F.R. Ga cía-Ga cía, I.D. González-Jiménez, I. Rod íguez-Ramos, A.
Gue eo-Ruiz, K. Li, Ce amic hollow ib es ca aly ic enhanced eac o s o glyce ol s eam
e o ming, Ca al. Today 233 (2014) 21-30.
[8] E.A. Sanchez, R.A. Comelli, Hyd ogen p oduc ion by glyce ol s eam- e o ming o e nickel
and nickel-cobal imp egna ed on alumina, In . J. Hyd ogen Ene gy 39 (2014) 8650-8655.
[9] K. Seung-hoon, J. Jae-sun, Y. Eun-hyeok, L. Kwan-Young, M. Dong Ju, Hyd ogen
p oduc ion by s eam e o ming o biomass-de i ed glyce ol o e Ni-based ca alys s, Ca al.
Today 228 (2014) 145-151.

[10] M.L. Dieuzeide, V. Iannibelli, M. Jobbagy, N. Amadeo, S eam e o ming o glyce ol o e
Ni/Mg/γ-Al2O3 ca alys s. E ec o calcina ion empe a u es, In . J. Hyd ogen Ene gy 37 (2012)
14926-14930.
[11] M.L. Dieuzeide, M. Jobbagy, N. Amadeo, Glyce ol s eam e o ming o e Ni/γ-Al2O3
ca alys s, modi ied wi h Mg(II). E ec o Mg (II) con en , Ca al. Today 213 (2013) 50-57.
[12] E.A. Sanchez, R.A. Comelli, Hyd ogen by glyce ol s eam e o ming on a nickel–alumina
ca alys : Deac i a ion p ocesses and egene a ion, In . J. Hyd ogen Ene gy 37 (2012) 14740-
14746.
[13] V.V. Thyssen, T.A. Maia, E.M. Assa , Ni suppo ed on La2O3–SiO2 used o ca alyze
glyce ol s eam e o ming, Fuel 105 (2013) 358-363.
[14] C.K. Cheng, S.Y. Foo, A.A. Adesina, S eam e o ming o glyce ol o e Ni/Al2O3 ca alys ,
Ca al. Today 178 (2011) 25-33.
[15] V. Nichele, M. Signo e o, F. Menegazzo, A. Gallo, V. Dal San o, G. C uciani, G. Ce a o,
Glyce ol s eam e o ming o hyd ogen p oduc ion: Design o Ni suppo ed ca alys s, Appl.
Ca al. B En i on. 111–112 (2012) 225-232.
[16] F. Pompeo, G.F. San o i, N.N. Nichio, Hyd ogen p oduc ion by glyce ol s eam e o ming
wi h P /SiO2 and Ni/SiO2 ca alys s, Ca al. Today 172 (2011) 183-188.
[17] I.N. Bu oni, F. Pompeo, G.F. San o i, N.N. Nichio, Nickel ca alys s applied in s eam
e o ming o glyce ol o hyd ogen p oduc ion, Ca al. Commun. 10 (2009) 1656-1660.
[18] A. I iondo, V.L. Ba io, J.F. Camb a, P.L. A ias, M.B. Guemez, M.C. Sanchez-Sanchez,
R.M. Na a o, J.L.G. Fie o, Glyce ol s eam e o ming o e Ni ca alys s suppo ed on ce ia and
ce ia-p omo ed alumina, In . J. Hyd ogen Ene gy 35 (2010) 11622-11633.
[19] L.F. Bobadilla, A. Penko a, A. Ál a ez, M.I. Domínguez, F. Rome o-Sa ia, M.A.
Cen eno, J.A. Od iozola, Glyce ol s eam e o ming on bime allic NiSn/CeO2–MgO–Al2O3
ca alys s: In luence o he suppo , eac ion pa ame e s and deac i a ion/ egene a ion p ocesses,
Appl. Ca al. A Gen. 492 (2015) 38-47.
[20] L.F. Bobadilla, A. Penko a, F. Rome o-Sa ia, M.A. Cen eno, J.A. Od iozola, In luence o
he acid–base p ope ies o e NiSn/MgO–Al2O3 ca alys s in he hyd ogen p oduc ion om
glyce ol s eam e o ming, In . J. Hyd ogen Ene gy 39 (2014) 5704-5712.
[21] J.W. Shabake , G.W. Hube , J.A. Dumesic, Aqueous-phase e o ming o oxygena ed
hyd oca bons o e Sn-modi ied Ni ca alys s, J. Ca al. 222 (2004) 180-191.
[22] L.F. Bobadilla, A. Ál a ez, M.I. Domínguez, F. Rome o-Sa ia, M.A. Cen eno, M. Mon es,
J.A. Od iozola, In luence o he shape o Ni ca alys s in he glyce ol s eam e o ming, Appl.
Ca al. B En i on. 123–124 (2012) 379-390.
[23] G. A zamendi, P.M. Diéguez, M. Mon es, J.A. Od iozola, E. Falabella Sousa-Aguia , L.M.
Gandía, Compu a ional luid dynamics s udy o hea ans e in a mic ochannel eac o o low-
empe a u e Fische –T opsch syn hesis, Chem. Eng. J. 160 (2010) 915-922.
[24] C.K. Cheng, S.Y. Foo, A.A. Adesina, Ca bon deposi ion on bime allic Co–Ni/Al2O3
ca alys du ing s eam e o ming o glyce ol, Ca al. Today 164 (2011) 268-274.
[25] G.S. Go, H.J. Lee, D.J. Moon, Y.C. Kim, Glyce ol s eam e o ming o e Ni–Fe–Ce/Al2O3
ca alys o hyd ogen p oduc ion, Res. Chem. In e med. 42 (2015) 289-304.
[26] S. Adhika i, S.D. Fe nando, A. Ha yan o, Kine ics and Reac o Modeling o Hyd ogen
P oduc ion om Glyce ol ia S eam Re o ming P ocess o e Ni/CeO2 Ca alys s, Chem. Eng.
Technol. 32 (2009) 541-547.
[27] R. Sunda i, P.D. Vaidya, Reac ion Kine ics o Glyce ol S eam Re o ming Using a
Ru/Al2O3 Ca alys , Ene gy & Fuels 26 (2012) 4195-4204.
[28] J.M. Sil a, M.A. So ia, L.M. Madei a, S eam e o ming o glyce ol o hyd ogen
p oduc ion: Modeling s udy, In . J. Hyd ogen Ene gy 41 (2016) 1408-1418.
[29] Y. Liu, A. Lawal, Kine ic s udy o au o he mal e o ming o glyce ol in a dual laye
monoli h ca alys , Chem. Eng. P ocess. 95 (2015) 276-283.
[30] L.C. Almeida, F.J. Echa e, O. Sanz, M.A. Cen eno, G. A zamendi, L.M. Gandía, E.F.
Sousa-Aguia , J.A. Od iozola, M. Mon es, Fische –T opsch syn hesis in mic ochannels, Chem.
Eng. J. 167 (2011) 536-544.
[31] L.C. Almeida, F.J. Echa e, O. Sanz, M.A. Cen eno, J.A. Od iozola, M. Mon es,
Washcoa ing o me allic monoli hs and mic ochannel eac o s, in: E.M. Gaigneaux, M.
De ille s, S. He mans, P.A. Jacobs, J.A. Ma ens, P. Ruiz (Eds.) S ud. Su . Sci. Ca al. Else ie
2010, pp. 25-33.
[32] O. Sanz, F.J. Echa e, F. Rome o-Sa ia, J.A. Od iozola, M. Mon es, Chap e 9 - Ad ances
in s uc u ed and mic os uc u ed ca aly ic eac o s o hyd ogen p oduc ion - Gandía, Luis M,
in: G. A zamendi, P.M. Diéguez (Eds.) Renew. Hyd ogen Technol. Else ie , Ams e dam, 2013,
pp. 201-224.
[33] D.M. F ías, S. Nousi , I. Ba io, M. Mon es, L.M. Ma ínez T, M.A. Cen eno, J.A.
Od iozola, Nuclea ion and g ow h o manganese oxides on me allic su aces as a ool o p epa e
me allic monoli hs, Appl. Ca al. A Gen. 325 (2007) 205-212.
[34] T.A. Nijhuis, A.E.W. Bee s, T. Ve guns , I. Hoek, F. Kap eijn, J.A. Moulijn, P epa a ion o
monoli hic ca alys s, Ca al. Re . 43 (2001) 345-380.
[35] K.Ohkubo, Y.Y.S. Yasaki, Uni ed S a es Pa en , 1993. Pa en numbe 5 208 206
[36] E.E. Iojoiu, M.E. Domine, T. Da idian, N. Guilhaume, C. Mi oda os, Hyd ogen p oduc ion
by sequen ial c acking o biomass-de i ed py olysis oil o e noble me al ca alys s suppo ed on
ce ia-zi conia, Appl. Ca al. A Gen. 323 (2007) 147-161.
[37] R.M. de Deugd, R.B. Chougule, M.T. K eu ze , F.M. Meeuse, J. G ie ink, F. Kap eijn,
J.A. Moulijn, Is a monoli hic loop eac o a iable op ion o Fische –T opsch syn hesis?,
Chem. Eng. Sci. 58 (2003) 583-591.
[38] C.A. Callaghan, Kine ics and Ca alysis o he Wa e -Gas-Shi Reac ion: A Mic okine ic
and G aph Theo e ic App oach Depa men o Chemical Enginee ing Wo ces e Poly echnic
Ins i u e 2006.
[39] W.E.S. R. By on Bi d, Edwin N. Ligh oo T anspo Phenomena, Re ised Second Edi ion
ed., Wiley 2012.
LIST OF FIGURES
Figu e 1. Manu ac u e and p epa a ion o cylind ical me allic monoli h
Figu e 7. App oach o equilib ium in he WGS eac ion as a unc ion o he ex en o
glyce ol s eam e o ming.

Figu e 8. E ec o limi ed p ehea ing on pseudohomogeneous i s -o de kine ics
wi h espec o glyce ol: (a) empe a u e p o ile, (b) main p oduc dis ibu ion (kdG
(873.15 K) = 200 mol s-1 m-3 a m-1, EadG 110 kJ mol-1, kWGS = 432 mol s-1 m-3 a m-2, pG0
= 0.034 a m, κ = 1 cm-3, T0 = 500 °C)
Figu e 9. Di usion and coun e di usion o and om he monoli h su ace