Polyol hydrogenolysis on supported Pt catalysts: Comparison between glycerol and 1,2-propanediol

Article abstract of DOI:10.1016/j.catcom.2013.09.021

Pt-based catalysts supported on TiO₂ and SIRAL 20 (Al ₂O₃-20 wt.%SiO₂) were prepared and characterized by H₂ chemisorption, FTIR of adsorbed pyridine and 3,3-dimethyl-1-butene isomerization. The catalysts were evaluated for the transformation under aqueous phase of glycerol and 1,2-propanediol at 210 C, under 60 bar as total pressure (H₂ atmosphere). Under similar conditions, 1,2-propanediol is easier converted than glycerol, indicating that in the glycerol transformation process in aqueous phase, the 1,2-propanediol reactivity is inhibited by the presence of glycerol. This behavior is explained by a strong adsorption of glycerol compared to 1,2-propanediol on the catalyst surface.

Full text of DOI:10.1016/j.catcom.2013.09.021

Catalysis Communications 43 (2014) 107–111
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Polyol hydrogenolysis on supported Pt catalysts: Comparison between
glycerol and 1,2-propanediol
Séverine Noe Delgado, Laurence Vivier ⁎, Catherine Especel
IC2MP (Institut de Chimie des Milieux et des Matériaux de Poitiers), Université de Poitiers, UMR 7285 CNRS, 4 rue Michel Brunet, 86022 Poitiers Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 May 2013
Received in revised form 19 July 2013
Accepted 18 September 2013
Available online 2 October 2013
Pt-based catalysts supported on TiO and SIRAL 20 (Al O –20 wt.%SiO ) were prepared and characterized by H
2
chemisorption, FTIR of adsorbed pyridine and 3,3-dimethyl-1-butene isomerization. The catalysts were evaluat-
ed for the transformation under aqueous phase of glycerol and 1,2-propanediol at 210 °C, under 60 bar as total
2
2
3
2
2
pressure (H atmosphere). Under similar conditions, 1,2-propanediol is easier converted than glycerol, indicating
that in the glycerol transformation process in aqueous phase, the 1,2-propanediol reactivity is inhibited by the
presence of glycerol. This behavior is explained by a strong adsorption of glycerol compared to 1,2-propanediol
on the catalyst surface.
Keywords:
Glycerol hydrogenolysis
Propanediol
© 2013 Elsevier B.V. All rights reserved.
Aqueous phase process
Platinum catalysts
1
. Introduction
correspond to selective C\O hydrogenolysis leading to C
compared to the C\C bond cleavages and reforming producing C
compounds (such as ethylene glycol, ethanol, methanol, ethane,
methane), as well as H and CO [12–15].
In a previous paper, we have studied the influence of the nature of
the support (alumina, alumina-silica, titania) on the catalytic properties
of Pt-based catalysts for the glycerol hydrogenolysis performed at
3
products,
2
and
The conversions of renewable feedstock into chemicals and fuels are
C
1
important in the present scenario, as the availability of fossil fuels is lim-
ited. In this context, the development of processes for the transforma-
tion of the lignocellulosic biomass is one of the major challenges
facing researchers today. The use of the lignocellulosic biomass is attrac-
tive because of its low cost and no competitiveness with the food plants.
Due to the polar nature of carbohydrates and other oxygenates issued
from lignocellulose, liquid processing in water is a particularly attractive
approach. One example of such process is the APP (Aqueous Phase Pro-
cess) method developed in the last 10 years, consisting of the transfor-
2
2
210 °C, under 60 bar of total pressure (H
batch reactor [16]. Among the various studied systems, Pt/TiO
appeared to be the most efficient one, leading to the highest glycerol
conversion and to the highest selectivity towards C products. These
products were mainly constituted of di-ols and mono-ols (valuable
2
or N
2
atmosphere) in a
2
sample
3
C
3
mation of sugars or polyols from biomass into products such as H
2
,
intermediary oxygenated compounds) with an extremely limited for-
mation of propane. Nevertheless, under the performed experimental
conditions, the glycerol conversion remained limited. Consequently,
the present paper is the continuation of this study, with the aim to com-
pare the reactivity of glycerol and 1,2-PD (1,2-PD: intermediary product
during glycerol hydrogenolysis) during the aqueous phase process
performed under similar reaction conditions. Two catalysts previously
alkanes or valuable oxygenated molecules [1–5]. Glycerol, used as
model molecule, has been identified as one of the top ten building
blocks in the bio refinery feed stocks [6]. It is a highly functionalized
molecule and a variety of value added chemicals could be derived
from glycerol, such as propanediols, acrolein and lactic acid [7]. The
transformation of glycerol by APP process results in competitive C\C
and C\O cleavages, in which the selectivity of the reaction can be di-
rected by the experimental conditions [2,8–10] and the nature of the
catalyst used [2,8,9,11]. In the presence of supported metallic catalysts,
glycerol can be dehydrated to acetol or 3-hydroxypropanal, and then
hydrogenated to propanediols (Fig. 1). By the same mechanism,
propanediols can be converted into 1- or 2-propanol and then to
propane without cleavage of C\C bonds. These consecutive reactions
tested for the glycerol transformation are used: Pt/S20 (S20: Al
2 3
O –
20 wt.%SiO ) and Pt/TiO , systems among all the tested ones that lead
2
2
respectively to the lowest (22%) and highest (62%) selectivity to 1,2-PD.
2. Experimental
2.1. Catalyst preparation
(Degussa P25, surface area = 55 m · g−1) and Al
2
O
–SiO
(SASOL, surface
area = 420 m · g ) were used as supports. They were calcined in
TiO
2
2
3
2
SIRAL 20, labeled S20, containing 20 wt.% SiO
2
⁎
Corresponding author. Tel.: +33 5 49 45 34 79; fax: +33 5 49 45 37 41.
2
−1
E-mail address: laurence.vivier@univ-poitiers.fr (L. Vivier).
1566-7367/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.catcom.2013.09.021

108
S.N. Delgado et al. / Catalysis Communications 43 (2014) 107–111
Fig. 1. Reaction scheme for aqueous phase transformation of glycerol.
flowing air for 4 h at 500 °C. Prior to use, TiO
retain particles with sizes between 0.04 and 0.10 mm. 3.0 wt.% Pt cata-
lysts were prepared by ion exchange method using H PtCl · 6H
Alfa Aesar) as precursor salt and water as solvent, according to the pro-
cedure described in a previous study [16]. The catalysts were dried over-
night in an oven at 120 °C, calcined under an artificial air flow (80%
2
was ground and sieved to
exposure to air before introduction into the autoclave. The autoclave
was flushed with N and after with H . The temperature of the reac-
2
2
2
6
2
O
tor was raised to 210 °C under 20 bar. Then, reactant solution
(10 mL) was loaded into the autoclave and the pressure was adjust-
ed to 60 bar. Zero time was taken when stirring was switched on.
At different times, liquid samples were manually collected during the
run and analyzed by HPLC, using sulphuric acid in water as elute
(
N
2
+ 20% O
2
) at 450 °C for 2 h, and finally reduced for 2 h in flowing
support).
−
1
−1
H
2
at 500 °C (S20 support) or 300 °C (TiO
2
(0.004 mol · L ) with a flow rate of 0.7 mL · min . The column
was BIO-RAD Aminex HPX 87H and it was used at 30 °C with both refrac-
tive index (RI) and ultraviolet (UV) detectors. Gaseous products were an-
alyzed online by a CP-3800 RGA system with 3 channels (2 TCD and 1
FID).
2
2
.2. H chemisorption
Pt dispersion was determined by H
ambient temperature, considering that the stoichiometry between hy-
drogen and a surface Pt atom (H/Pt ) is equal to 1. The conditions of
the experiments are described elsewhere [16].
2
chemisorption performed at
Conversion and selectivity values were calculated on the basis of the
following equations:
s
Conversion ð%Þ
2
.3. Isomerization of 3,3-dimethyl-1-butene
¼ ðmoles of reactant consumed=moles of reactant initially chargedÞ
ꢀ
100
The skeletal isomerization of 3,3-dimethyl-1-butene (33DMB1) was
used as model reaction to characterize the Brønsted acid sites. Indeed, as
demonstrated by Kemball and colleagues [17], the Lewis centers are not
involved in this reaction and the 33DMB1 isomerization is likely to
occur through a pure protonic mechanism leading to 2,3-dimethyl-1-
butene and 2,3-dimethyl-2-butene isomers [18,19]. The conditions of
the reaction are detailed in [16].
Selectivity ð%Þ
¼ ðmoles in specific products=moles in all detected productsÞ
ꢀ100:
3. Results and discussion
2
.4. FTIR of adsorbed pyridine
3.1. Catalyst characterization
The Fourier transformed infra-red spectroscopy (FTIR) of adsorbed
The Pt dispersion of Pt/S20 and Pt/TiO catalysts was measured be-
2
pyridine was used to determine the number of Lewis acid sites on the
metallic catalysts [20]. The procedure describing the experiments was
given in detail previously [16].
fore (fresh) and after (used) the polyol (glycerol or 1,2-propanediol)
transformation (Fig. 2). On the fresh catalysts, the Pt dispersion value
is around two times higher on the alumina-silica support than on titania
(
68% against 35% respectively). After the polyol transformation
2
.5. Aqueous-phase transformation of polyol
(glycerol or 1,2-propanediol), the evolution of the Pt dispersion de-
pends highly on the support used. In fact, no significant modification
The aqueous phase reaction was carried out in a 300 mL stirred
2
of the dispersion values is observed with the TiO support, taking
batch autoclave (Autoclave Engineers) under 60 bar total pressure (H
atmosphere), at 210 °C for 6 h. The batch reactor was fitted with a
system for liquid and gas sampling.
Glycerol or 1,2-propanediol (4.5 wt.%) was loaded with 10 mL of
ultra pure water into a feeder and the catalyst (1 g) was loaded with
2
into account the experimental errors generally admitted for this
measurement (± 10%). But in the case of the S20 support, a large de-
crease is observed after the aqueous-phase glycerol transformation
(by a factor superior to 3). The decrease is more moderate (by approx-
imately a 1.5 factor) when glycerol was replaced by 1,2-propanediol.
This loss of dispersion is explained by a sintering of the Pt particles dur-
ing the catalytic test, as already described in recent studies devoted to
1
40 mL of ultra-pure water into the reactor. The catalysts were pre-
reduced and immersed in the solvent (ultra-pure water) without

S.N. Delgado et al. / Catalysis Communications 43 (2014) 107–111
109
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
then extrapolated for both catalysts at a same temperature (300 °C).
The histograms indicate a total acidity of Pt-supported catalyst higher
on alumina-silica than on titania support, with a twice higher number
of both Lewis acid sites (Fig. 3A) and Ln(A) value (Fig. 3B).
Fresh
Glycerol
1,2-PD
3.2. Aqueous phase transformation of polyol
Fig. 4 shows the conversion of glycerol and 1,2-propanediol (1,2-PD)
on both catalysts as a function of reaction time. Whatever the catalysts,
under the same reaction conditions (210 °C, 60 bar of total pressure
under H
with more or less a discrepancy. Indeed, the 1,2-PD is converted to 49%
and 61% after 6 h of reaction in the presence respectively of Pt/TiO and
2
), the conversion of 1,2-PD is higher than that of glycerol,
Pt/TiO2
Pt/S20
2
Fig. 2. Dispersion of Pt atoms on the catalysts before (fresh) and after hydrogenolysis of
glycerol or 1,2-propanediol (1,2-PD).
Pt/S20 catalysts, while the conversions of the glycerol reach only 23%
and 10% on these same respective solids. It must be noted that the Pt
metallic phases present different dispersion states according to the nature
of the catalyst and the reaction medium. In fact, Fig. 2 has shown that Pt
particles supported on TiO2 are more stable than these on S20 during
catalytic tests, leading to the following order of the dispersion values for
2 3 2 3 2
the stability of Pt catalysts supported on Al O or Al O –SiO during
aqueous phase hydrogenolysis of polyol [21–23]. Generally, according
to literature, the stability of supported catalysts in aqueous phase is
linked to several parameters, such as the nature of the support, the ini-
tial oxidation state of the metal, the pH of the solution, the presence of
chloride anions in the solution and the nature of the gas atmosphere
the used catalysts: Pt/S20 (glycerol as substrate) ≪ Pt/TiO (glycerol as
2
substrate) b Pt/TiO (1,2-PD as substrate) b Pt/S20 (1,2-PD as substrate),
2
the same order of conversion observed in Fig. 4.
Fig. 5 presents selectivities obtained in the main products according
to reaction time during the transformation of the glycerol and 1,2-PD on
both catalysts.
[24–27]. Consequently, during glycerol or 1,2-PD transformation, the
composition of the aqueous phase (nature and respective quantities of
reactant and oxygenated intermediates in the medium) can play a
nonnegligible role on the sintering phenomenon. The final pH values
measured in the reactional mixtures are acidic, but always a little
more with glycerol as substrate (3.0 against 3.6 with 1,2-PD). This dif-
ference in pH may explain the difference in Pt dispersion of the Pt/S20
catalysts after the tests with glycerol or 1,2-PD used as substrate. Never-
2
theless, the present results show that the use of TiO support allows
preparing Pt-based catalyst presenting a metallic phase in a quite mod-
erate dispersion state but more stable under reaction conditions.
Fig. 3 gives the proportions of Lewis and Brønsted acid sites on the
Pt-based catalysts determined by FTIR of adsorbed pyridine and isomer-
ization of 33DMB1 respectively. From FTIR experiments of adsorbed
pyridine, the quantity of Lewis acid sites is determined from the inte-
−
1
grated area of the band located at 1450 cm , and the strength of
these sites being a function of the temperature at which the spectra
are recorded (between 150 °C and 350 °C in the present study). For
the isomerization of 33DMB1, the reaction was performed at a given
temperature comprised between 150 °C and 300 °C, in order to obtain
conversions lower than 20%. Considering the activation energy value
−
1
of this reaction equal to 95 kJ · mol [19], the value of activity A was
Fig. 3. (A) Total Lewis acid sites numbers (μmol g− ) determined by FTIR of adsorbed
1
pyridine and (B) Brønsted acid sites activity (A per μmol · m− · h ) for 33DMB1
2
−1
Fig. 4. Polyol conversion (▲: glycerol, ■: 1,2-PD) as a function of reaction time on Pt-based
isomerization at 300 °C of the Pt-based catalysts.
2 2
catalysts: (A) Pt/TiO and (B) Pt/S20 [T = 210 °C, P = 60 bar under H ].

110
S.N. Delgado et al. / Catalysis Communications 43 (2014) 107–111
Fig. 5. Selectivity as function of reaction time during polyol transformation (A: glycerol, B: 1,2-PD) on Pt-based catalysts: (a) Pt/TiO
under H ]. ● acetol, × 1,2-PD, ○ 1P, * EG, ■ EtOH, Δ MeOH, ▲ saturated HC (propane, ethane and methane).
2
and (b) Pt/S20 [T = 210 °C, P = 60 bar
2
In both cases, 3-hydroxypropanal (3-HPA) and glyceraldehyde (GLA),
presented in Fig. 1 as potential initial intermediary products when
starting from glycerol, are not observed in a quantitative way, whereas
acetol, primary dehydration product of glycerol, is detected at the be-
ginning of the reaction. Nevertheless, the stabilities of 3-HPA and acetol
are known to be different, 3-HPA being rather unstable in aqueous me-
dium and nearly instantaneously converted in subsequent reac-
tions [13,28]. With 1,2-PD as reactant, acetol is also initially formed in
a little extent resulting from a dehydrogenation reaction despite the
(ii) Whatever the catalyst, the dehydration/hydrogenation step
converting 1,2-PD to C mono-alcohols (1P and 2P) does not ap-
3
pear as a favored way. In fact, with glycerol as initial reactant, the
formation of these mono-alcohols (mainly 1-propanol, 1P) re-
mains very low (selectivity inferior to 10%, Fig. 5A), and starting
with 1,2-PD, the selective cleavage of a C\C bond leads mainly
at first to the formation of EtOH and MeOH (Fig. 5B). Moreover,
as no ethylene glycol was detected in the medium in this latter
case, the C\C cleavage is exclusively located between the two
functionalized carbons.
H
2
reductive atmosphere.
A comparison of Fig. 5A and B leads to the following information about
the mechanism of transformation of these two polyols in aqueous phase:
(iii) The presence of EG observed during the glycerol transformation
results then from a C\C bond cleavage of the initial reactant
[
31], and not from one of the 1,2-PD intermediary.
(
i) On Pt/TiO catalyst, glycerol is mainly transformed by a dehydra-
2
(
iv) The formation of propane (main saturated hydrocarbon HC
observed) resulting from selective successive dehydration/
hydrogenation steps remains very limited, even when 1,2-PD
is used as started molecule. Though, in this case, the presence
of 1-propanol is noted, admittedly in a small quantity at first
but increasing during the catalytic test, only one dehydration/
hydrogenation step is then needed from 1-propanol to obtain
propane.
tion/hydrogenation step in 1,2-PD (with acetol as intermediary
compound) (Fig. 5Aa). On the Pt/S20 catalyst, the same pathway
exists but in competition with C\C bond cleavage since rather
3 2
comparable proportions of 1,2-PD and EG (C and C compounds
respectively) are observed initially (Fig. 5Ab). This behavior is
consistent with the strongest acidity revealed previously on the
Pt/S20 sample (Fig. 3), that can accentuate acidic cracking mech-
anisms, instead of favoring only the dehydration phenomenon
[29,30]. Moreover, one should point out that cracking becomes
Previously, the Pt dispersions were observed to evolve between
more predominant with time on Pt/S20 (Fig. 5Ab).
the fresh and used states, especially with the Pt/S20 catalyst, and

S.N. Delgado et al. / Catalysis Communications 43 (2014) 107–111
111
this differs according to the nature of the polyol introduced initial-
Acknowledgment
ly. The presence of glycerol in the media, and/or particular inter-
mediaries as EG, could contribute to explain the higher sintering
phenomenon highlighted on the Pt/S20 sample during glycerol
transformation.
We are grateful to Region Poitou-Charentes for the financial support
of this project.
Consequently, the present results show that during the transfor-
mation of glycerol the 1,2-propanediol reactivity is inhibited by the
presence of glycerol, 1,2-PD as reactant being easier converted than
glycerol under similar conditions (Fig. 4). This behavior can be
explained by a strong adsorption of glycerol compared to 1,2-PD on
the catalyst surface, as already reported in the literature over other
kinds of catalysts [32–34]. Lahr and Shanks [32] developed a kinetic
model to describe glycerol hydrogenolysis and noted the importance
of competitive adsorption between oxygenated compounds on Ru/C
catalyst. Zhou et al. [33] reported that 1,2-PD is the most weakly
adsorbed species among glycerol, acetol and 1,2-PD during the
References
[
[
1] R.D. Cortright, R.R. Davda, J.A. Dumesic, Nature 418 (2002) 964–967.
2] G.W. Huber, R.D. Cortright, J.A. Dumesic, Angew. Chem. Int. Ed. 43 (2004) 1549–1551.
[3] J.W. Shabaker, G.W. Huber, J.A. Dumesic, J. Catal. 222 (2004) 180–191.
[4] B. Zhang, X. Tang, Y. Li, Y. Xu, W. Shen, Int. J. Hydrogen Energy 32 (2007) 2367–2373.
[
5] S. Adhikari, S.D. Fernando, S.D. Filip To, R.M. Bricka, P.H. Steele, A. Haryanto, Energy
Fuel 22 (2008) 1220–1226.
[
6] S. Fernando, S. Adhikari, C. Chandrapal, N. Murali, Energy Fuel 20 (2006) 1727–1737.
[7] J.C. Serrano-Ruiz, R. Luque, A. Sepúlveda-Escribano, Chem. Soc. Rev. 40 (2011)
266–5281.
8] R.R. Davda, J.W. Shabaker, G.W. Huber, R.D. Cortright, J.A. Dumesic, Appl. Catal. B 56
2005) 171–186.
5
[
(
[9] A. Iriondo, J.F. Cambra, V.L. Barrio, M.B. Guemez, P.L. Arias, M.C. Sanchez-Sanchez,
R.M. Navarro, J.L.G. Fierro, Appl. Catal. B 106 (2011) 83–93.
2 3
hydrogenolysis of glycerol over Cu–ZnO–Al O . Vasiliadou and
[
10] F. Auneau, S. Noël, G. Aubert, M. Besson, L. Djakovitch, C. Pinel, Catal. Commun. 16
Lemonidou [34] suggested that glycerol adsorption on copper sur-
face is strong and the active metal surface is almost saturated with
glycerol adsorbed species.
(2011) 144–149.
[11] M. Checa, F. Auneau, J. Hidalgo-Carillo, A. Marinas, J.M. Marinas, C. Pinel, F.J. Urbano,
Catal. Today 196 (2012) 91–100.
[
[
[
12] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Appl. Catal. A 318 (2007) 244–251.
13] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
14] A. Ciftci, B. Peng, A. Jentys, J.A. Lercher, E.J.M. Hensen, Appl. Catal. A 431–432 (2012)
1
13–119.
4
. Conclusion
[
15] M.E. Doukkali, A. Iriondo, J.F. Cambra, L. Jalowiecki-Duhamel, A.S. Mamede, F.
Dumeignil, P.L. Arias, J. Mol. Catal. A 368–369 (2013) 125–136.
[16] S. Noe Delgado, D. Yap, L. Vivier, C. Especel, J. Mol. Catal. A 367 (2013) 89–98.
[17] C.S. John, C. Kemball, R.A. Rajadhayarsha, J. Catal. 57 (1979) 264–271.
The performances of Pt/TiO
evaluated for the glycerol and 1,2-propanediol (1,2-PD) transfor-
mation in aqueous phase, at 210 °C, under 60 bar as total pressure
2
and Pt/Al
2
O
3
–SiO
2
catalysts were
[
[
[
18] H.J. Pines, J. Catal. 78 (1982) 1–16.
19] D. Martin, D. Duprez, J. Mol. Catal. A 118 (1997) 113–128.
20] M. Guisnet, P. Ayrault, C. Coutanceau, M.F. Alvarez, J. Datka, J. Chem. Soc. Faraday
Trans. 93 (1997) 1661–1665.
2
(H ), in a batch reactor. On alumina-silica support, a higher total
acidity (Lewis and Brønsted acidity) is exhibited. During polyol
transformation, the acidic sites are involved in the dehydration re-
actions (leading to C\O bond cleavage). Both catalysts were
subjected to a sintering of Pt particles during the test under hydrother-
mal conditions, this phenomenon being however less important with
[
[
[
[
21] R.M. Ravenelle, J.R. Copeland, W.G. Kim, J.C. Crittenden, C. Sievers, ACS Catal. 1
(2011) 552–561.
22] L. Vilcocq, A. Cabiac, C. Especel, S. Lacombe, D. Duprez, Catal. Commun. 15 (2011)
18–22.
23] R.M. Ravenelle, J.R. Copeland, A.D. Van Pelt, J.C. Crittenden, C. Sievers, Top. Catal. 55
(2012) 162–174.
2
Pt-supported on TiO .
24] P. Granger, J.M. Dumas, C. Montassier, J. Barbier, J. Chim. Phys. 92 (1995) 1557–1575.
Systematically, under the same experimental conditions, the
conversion of 1,2-PD as reactant is much more important than
that of glycerol, suggesting an inhibited effect of the presence of
glycerol on the reactivity of the 1,2-PD intermediary. A strong ad-
sorption of glycerol compared to 1,2-PD is then considered on the
catalyst surface. Moreover, starting from 1,2-PD, an important for-
mation of ethanol and methanol is observed on both catalysts,
resulting from C\C bond cleavages. Thus, to limit these degradation
reactions and favor selectively the C\O bond cleavages, an improve-
ment of the balance between acid phase and metallic phase is therefore
still necessary.
[25] A. Douidah, P. Marecot, S. Labruquere, J. Barbier, Appl. Catal. A 210 (1–2) (2001) 111–120.
[
[
26] A. Douidah, P. Marecot, J. Barbier, Appl. Catal. A 225 (1–2) (2002) 11–19.
27] F. Epron, C. Especel, G. Lafaye, P. Marécot, in: D. Astruc (Ed.), Nanoparticles and Ca-
talysis, Wiley-VCH Verlag, 2008, pp. 279–302.
[28] I. Gandarias, P.L. Arias, J. Requies, M.B. Güemez, J.L.G. Fierro, Appl. Catal. B 97 (2010)
48–256.
2
[
[
29] G.M. Kramer, G.B. McVicker, J.J. Ziemiak, J. Catal. 92 (1985) 355–363.
30] I. Gandarias, P.L. Arias, J. Requies, M. El Doukkali, M.B. Güemez, J. Catal. 282 (2011)
237–247.
[31] S. Zhu, Y. Qiu, Y. Zhu, S. Hao, H. Zheng, Y. Li, Catal. Today 212 (2013) 120–126.
[
[
32] D.G. Lahr, B.H. Shanks, Ind. Eng. Chem. Res. 42 (2003) 5467–5472.
33] Z. Zhou, X. Li, T. Zeng, W. Hong, Z. Cheng, W. Yuan, Chin. J. Chem. Eng. 18 (3) (2010)
384–390.
[34] E.S. Vasiliadou, A.A. Lemonidou, Chem. Eng. J. 231 (2013) 103–112.