Basic oxide-supported Ru catalysts for liquid phase glycerol hydrogenolysis in an additive-free system

Article abstract of DOI:10.1016/j.catcom.2013.11.031

Several basic oxide-supported Ru catalysts (Ru/CeO₂, Ru/La ₂O₃ and Ru/MgO) were prepared and evaluated for the hydrogenolysis of glycerol. The Ru catalysts were characterized by inductively coupled plasma-atomic emission spectroscopy, nitrogen adsorption, powder X-ray diffraction, X-ray photoelectron spectroscopy, and transmission electron microscopy. Ru/CeO₂ showed the best performance in the reaction, which is associated with its smaller metal particle size and the weak surface basicity feature of CeO₂. 1,2-Propanediol is obtained as the main product through a dehydrogenation-dehydration-hydrogenation mechanism. The oxidation product lactic acid can be formed by a Cannizzaro reaction from the pyruvaldehyde intermediate.

Full text of DOI:10.1016/j.catcom.2013.11.031

Catalysis Communications 46 (2014) 98–102
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
Basic oxide-supported Ru catalysts for liquid phase glycerol
hydrogenolysis in an additive-free system
a
b
b
a
c
Jian Feng a, , Wei Xiong , Bin Xu , Weidong Jiang , Jinbo Wang , Hua Chen
⁎
a
College of Chemistry and Chemical Engineering, Chongqing University of Science & Technology, Chongqing 401331, China
Key Laboratory of Green Catalysis of Sichuan Institutes of Higher Education, School of Chemistry and Pharmaceutical Engineering, Sichuan University of Science & Engineering,
Zigong 643000, China
b
c
Key Laboratory of Green Chemistry and Technology, Ministry of Education, College of Chemistry, Sichuan University, Chengdu 610064, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 October 2013
Received in revised form 25 November 2013
Accepted 29 November 2013
Available online 7 December 2013
Several basic oxide-supported Ru catalysts (Ru/CeO2, Ru/La2O3 and Ru/MgO) were prepared and evaluated for
the hydrogenolysis of glycerol. The Ru catalysts were characterized by inductively coupled plasma–atomic emis-
sion spectroscopy, nitrogen adsorption, powder X-ray diffraction, X-ray photoelectron spectroscopy, and trans-
mission electron microscopy. Ru/CeO2 showed the best performance in the reaction, which is associated with
its smaller metal particle size and the weak surface basicity feature of CeO2. 1,2-Propanediol is obtained as the
main product through a dehydrogenation–dehydration–hydrogenation mechanism. The oxidation product lactic
acid can be formed by a Cannizzaro reaction from the pyruvaldehyde intermediate.
Keywords:
Ruthenium
Basic oxide support
Hydrogenolysis
Glycerol
© 2013 Elsevier B.V. All rights reserved.
1,2-Propanediol
Lactic acid
1. Introduction
found that the addition of NaOH markedly increased the reactivity of
Cu, Ru or Pt catalysts, but a certain amount of lactic acid was formed
In recent years, because of the rapid development of biodiesel pro-
duction by transesterification of vegetable oils or animal fats, large
amounts of glycerol are available as a byproduct. How to utilize these
biomass-derived glycerol has attracted great attention. In particular,
the catalytic hydrogenolysis process is one of the most promising routes
in the chemical transformation of glycerol. The main product arising
from the hydrogenolysis of glycerol is 1,2-propanediol (1,2-PDO),
which is extensively used as a monomer for the production of polyester
resins. Undoubtedly, the production of 1,2-PDO from glycerol will be a
sustainable and feasible alternative to the current petroleum-based
1,2-PDO production method.
In general, the hydrogenolysis of glycerol is studied under hetero-
geneous conditions over various metal catalysts. Several recent re-
views have summarized relevant catalytic systems for the glycerol
hydrogenolysis reaction [1–3]. In order to obtain a good catalytic perfor-
mance, metal catalysts are usually used in combination with a certain
acid or base additive [2,3]. For example, H2WO4 [4], cation-exchange
resin [5–7] and Nb2O5 [8] are effective acid additives, which can en-
hance the reaction rate and the diol selectivity over Ru or Rh catalysts.
Regarding the base additives, Wang and Liu [9] and Maris et al. [10,11]
at high pH. In our previous works [12–16], we have developed a series
of Ru-based catalysts, which exhibited good performances in the
hydrogenolysis of glycerol to 1,2-PDO. Using LiOH as additive, both
the conversion of glycerol and the selectivity to 1,2-PDO can be high
up to 90% in the presence of Ru/TiO2 catalyst [12–14]. We have also
demonstrated that the base aids the initial dehydrogenation of glycerol
to glyceraldehyde and promotes the dehydration of glyceraldehyde to
2-hydroxyacrolein (see Scheme 1) [12]. This effect of the base additive
is responsible for the enhancement in the reaction rate and 1,2-PDO
selectivity.
Although the introduction of some additives can promote the cata-
lytic performance, it results in many troubles in practical manipulation,
such as the product separation and catalyst recycling. According to the
reaction pathway in Scheme 1, a catalyst that contains both metal
sites and base sites should also achieve good results in the glycerol
hydrogenolysis reaction. Solid-base supported metal catalysts will
make it possible for the successful combination of metal sites and base
sites. Such metal–base bifunctional catalysts bring the advantage that
the product separation or catalyst recycling will be more maneuverable
and more practicable. In fact, solid-base supported Pt catalysts [17] and
Cu catalysts [18–21] are indeed efficient for the selective hydrogenolysis
of glycerol to 1,2-PDO. As a typical example, the research group of Hou
[18–20] developed a series of Cu-based catalysts supported on the
⁎
Corresponding author. Tel.: +86 23 65023739; fax: +86 23 65023753.
E-mail address: fengjianscu@163.com (J. Feng).
1566-7367/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.catcom.2013.11.031

J. Feng et al. / Catalysis Communications 46 (2014) 98–102
99
OH
OH
OH
O
OH
OH
O
OH
+ H2
- H2O
- H2
OH
HO
-
Ru
-
OH
Ru, OH
Glyceraldehyde
1,2-PDO
2-Hydroxyacrolein
Scheme 1. Hydrogenolysis of glycerol to 1,2-PDO over Ru-based catalyst in basic aqueous solution.
layered double hydrotalcite-like compounds. These Cu-based catalysts
were highly selective for 1,2-PDO. However, the catalyst preparation
procedure seems a bit complicated and unmanageable.
In this work, we directly utilize the surface basicity of several easily
available supports to modify the Ru catalyst, with the intent of develop-
ing a simple and additive-free catalytic system. Several Ru catalysts
supported on basic oxides (CeO2, La2O3 and MgO) were prepared and ex-
amined for the glycerol hydrogenolysis reaction. Ru/CeO2 was found to be
an efficient catalyst for the hydrogenolysis of glycerol in an additive-free
aqueous solution.
than those on La2O3. Our previous studies [13,14] have shown that the
support material can significantly influence the metal particle size.
The bigger Ru particles on La2O3 may be related to the fact that Ru par-
ticles are easier to agglomerate in this support.
As presented in Table 1, the Ru0/(Ru0 + Ruδ+) refers to the surface
Ru atomic ratios obtained from XPS experiments, where Ru0 represents
the Ru atoms with metallic state and Ruδ+ represents the Ru species
with intermediate value between Ru0 and Ru3+. The calculation meth-
od is referred to our earlier work [13]. It can be seen from Table 1 that
the Ru species on Ru/La2O3 and Ru/CeO2 are completely reduced in
the reduction process. This indicates that the rare earth element is ben-
eficial to the reduction of Ru, which is in agreement with the findings of
Yu and co-workers [24]. A typical Ru 3d XPS spectrum of Ru/CeO2 is
shown in Fig. 1. The singlet in dotted line is attributed to C 1s (binding
energy = 284.9 eV). This is an unavoidable contamination signal of the
spectrometer [13]. There is only one doublet of Ru 3d (binding energy:
Ru 3d5/2 = 280.1 eV, Ru 3d3/2 = 284.3 eV), which is assigned to the
metallic ruthenium. For reference, the Ru 3d XPS spectra of Ru/MgO
and Ru/La2O3 are shown in Figs. S1 and S2 (Supplementary material),
respectively.
2. Experimental
2.1. Catalyst preparation
Ru catalysts (Ru/CeO2, Ru/La2O3 and Ru/MgO) with a metal loading
of 3 wt.% were prepared by impregnation technique. Weighed amounts
of the corresponding oxide were impregnated with the aqueous solu-
tion of RuCl3. After impregnation, the solvent was removed, and the
resulting powder was dried in vacuum at 110 °C for 10 h. All the cata-
lysts were reduced in an autoclave by hydrogen at 200 °C. For more de-
tails, see the Supplementary material.
The XRD patterns of the supported Ru catalysts are illustrated in
Fig. 2. The crystalline phases were identified by comparison with
JCPDS files. For Ru/CeO2, the clear diffraction peaks at 2θ = 28.6°, 33.1°,
47.5°, 56.4°, 59.1°, 69.4°, 76.7°, 79.1° and 88.4° are all attributed to
cerianite CeO2 (JCPDS Card #34-0394). In the case of Ru/MgO, the inten-
sive and sharp diffraction peaks at 2θ = 18.6°, 32.8°, 38.0°, 50.8°, 58.6°,
62.1°, 68.2°, 72.0° and 81.2° can be assigned to brucite Mg(OH)2 (JCPDS
Card #83-0114), indicating that MgO is converted to Mg(OH)2. This
hydration of MgO mainly results from the hydrothermal condition used
in the catalyst preparation process [25]. In the XRD pattern of Ru/La2O3,
a new phase of La(OH)3 (2θ = 15.7°, 27.4°, 31.6° and 48.4°; JCPDS
Card #75-1900) is observed accompanied with the phase of La2O3
(2θ = 27.9°, 39.6° and 49.5°; JCPDS Card #89-4016). The existence of
La(OH)3 could be also owing to the hydrothermal condition in the cata-
lyst preparation process [26]. Accordingly, CeO2 is comparatively stable
under the condition we used. As far as the Ru crystalline phase is con-
cerned, no obvious diffraction peaks can be observed in the XRD pat-
terns of Ru/CeO2 and Ru/MgO, suggesting that the Ru particles on
CeO2 and MgO are too small to be detected. In contrast, the diffraction
peaks of metallic Ru at 2θ = 38.4° and 42.2° (JCPDS Card #89-3942)
appeared in the Ru/La2O3 catalyst. The particle size of Ru is determined
2.2. Characterizations
The Ru catalysts were characterized by ICP-AES, BET, XRD, XPS, and
TEM. According to the references [22,23], the surface basicity features of
the oxide supports were determined by the benzoic acid titration method
using two Hammett indicators: bromothymol blue (pKa = +7.2) and
2,4-dinitroaniline (pKa = +15.0). The base strength (H_) was
expressed by the Hammett function that was scaled by the pKa values
of the indicators. The details are discussed in the Supplementary
material.
2.3. Catalytic performance testing
Hydrogenolysis of glycerol was carried out in a 30 mL stainless steel
autoclave. Because of the word count limit, more details are presented
in the Supplementary material.
3. Results and discussion
3.1. Characterization of catalysts and supports
Some characterization results are shown comparatively in Table 1.
ICP-AES results show that the Ru weight loadings are near to the stated
value, i.e., about 3 wt.%. The BET surface area of Ru/MgO is the highest,
while that of Ru/La2O3 and Ru/CeO2 is very close. The Ru particle sizes
are all determined by TEM due to weak diffraction peaks of Ru in the
XRD patterns. Obviously, Ru particles are smaller on MgO and CeO2
C1s
Table 1
Some characterization results of the Ru catalysts.
Catalyst
Ru loading
(wt.%)
BET surface area
(m2/g)
Ru particle size
(nm)a
Ru0/(Ru0 + Ruδ+
(%)
)
Ru/La2O3
Ru/MgO
Ru/CeO2
3.11
2.98
3.03
59
246
63
11.4
5.8
5.3
100
57.1
100
290
285
280
275
Binding Energy (eV)
a
Determined by TEM.
Fig. 1. Ru 3d XPS spectrum of Ru/CeO2 catalyst.

100
J. Feng et al. / Catalysis Communications 46 (2014) 98–102
Table 2
Base strength (H_) distribution of the supports.
Support
Basicity (mmol/g)
Ru
+7.2 ≤ H_ ≤ +15.0
H_ ≥ +15.0
Total
La2O3
MgO
CeO2
0.04
0.08
0.13
0.01
0.35
–
0.05
0.43
0.13
Ru/La O3
2
of +7.2 and +15.0 can be considered as weak base sites [22], whereas
those higher than +15.0 can be assigned as strong base sites [22,23].
Obviously, MgO is a strong base, while CeO2 and La2O3 are weak
bases. In addition, there are no strong base sites on the surface of CeO2.
Ru/MgO
Ru/CeO
2
10
20
30
40
50
60
70
80
90
3.2. Catalytic performance of the Ru catalysts
2 Thete (degree)
Catalytic performance results of the Ru/CeO2, Ru/La2O3 and Ru/MgO
catalysts in the glycerol hydrogenolysis reaction are presented in
Table 3. All the basic oxide-supported Ru catalysts were efficient for
the hydrogenolysis of glycerol to 1,2-PDO, ethylene glycol, lactic acid,
and monohydroxy-alcohols. The Ru/CeO2 catalyst exhibited markedly
high activity and high selectivity to 1,2-PDO, giving the highest yield
of 1,2-PDO (53.4%). In contrast, Ru/La2O3 showed the lowest activity,
which was about a third of the activity of Ru/CeO2. Moderate catalytic
activity was observed on the Ru/MgO catalyst, but it was not very selec-
tive for 1,2-PDO.
Combination of physical properties of the as-prepared Ru catalysts in
Table 1 reveals that the hydrogenolysis of glycerol is more active on
small metal particles. The catalytic performance of the Ru catalysts
seems independent of the BET surface area, but partly related to the ac-
tive metal surface area (Table S1 in the Supplementary material). These
results are in good agreement with those reported in published studies
Fig. 2. XRD patterns of the Ru catalysts.
to be about 11.7 nm, which is in accordance with the TEM result listed
in Table 1.
The TEM images of the as-prepared Ru catalysts are shown in Fig. 3.
The nanosized Ru particles are well dispersed on the surface of CeO2 and
MgO (Fig. 3a and b), giving an average particle size of Ru around 5.5 nm
(Table 1). As for the Ru/La2O3 catalyst, the dispersion of Ru is not very
good (Fig. 3c), and the average particle size of Ru is determined to be
about 11.4 nm (Table 1).
As mentioned above, we envisage using the oxide supports to pro-
vide base sites. Therefore, the surface basicity features of CeO2, La2O3
and MgO were determined by the benzoic acid titration method as de-
scribed in the Experimental section. Table 2 presents the base strength
distribution of the supports. The base sites between the pKa values
Fig. 3. TEM images of the Ru catalysts: (a) Ru/CeO2; (b) Ru/MgO; and (c) Ru/La2O3.

J. Feng et al. / Catalysis Communications 46 (2014) 98–102
101
Table 3
decomposition of glycerol to lactic acid, we have conducted blank reac-
tions using neat CeO2, La2O3 and MgO under the same reaction con-
ditions in Table 3. However, without Ru metal, almost no reaction
occurred, indicating that the basic oxide supports are unable to catalyze
the hydrothermal decomposition of glycerol under our reaction condi-
tions. Furthermore, it also suggests that the Ru metal is necessary for
the glycerol hydrogenolysis reaction, and lactic acid is not formed di-
rectly from glycerol.
In fact, lactic acid and 1,2-PDO can be formed through a same inter-
mediate: 2-hydroxyacrolein [10,28,29]. As we described above, the
hydrogenation of 2-hydroxyacrolein over the Ru sites gives 1,2-PDO
(see Scheme 1). Of course, 2-hydroxyacrolein can also turn into pyruv-
aldehyde by keto-enol tautomerism. The combination of experimental
results and density functional theory calculations shows that lactic
acid can be easily obtained by a Cannizzaro reaction from pyruvalde-
hyde [28,29]. Based on the known mechanism of Cannizzaro reaction,
the reaction pathway for the conversion of 2-hydroxyacrolein to lactic
acid is presented in Scheme 2. The hydroxyl ion can be generated
from the hydration of the basic oxide supports, which has been proved
by the XRD characterization results. This explanation can also account
for the high lactic acid selectivity of Ru/MgO, because MgO was all trans-
formed into Mg(OH)2 under hydrothermal condition.
Catalytic performance of the Ru catalysts for the hydrogenolysis of glycerol.a
Catalyst
Conversion (%)
Selectivity (%)b
1,2-PDO
EG
6.3
18.0
4.2
LA
8.5
21.3
3.1
Ru/La2O3
Ru/MgO
Ru/CeO2
31.5
60.4
85.2
52.8 (16.6)c
49.7 (30.0)
62.7 (53.4)
a
Reaction conditions: 3 mL of 20 wt.% glycerol aqueous solution, 180 °C, 5 MPa, 10 h,
50 mg catalyst. EG: ethylene glycol; LA: lactic acid.
b
Other products: 1-propanol, 2-propanol, ethanol, methanol and methane.
Yield in bracket.
c
[5–13]. Comparison of the Ru reduction degree, i.e., Ru0/(Ru0 + Ruδ+
)
in Table 1 indicates that the differences in the catalytic performance of
the three Ru catalysts are not mainly influenced by the Ru reduction
degree. For example, the Ru reduction degree of Ru/MgO was poor
than that of Ru/La2O3, but the catalytic activity of Ru/MgO was much
better. This is possibly because the active metal sites on the catalysts
are sufficient to catalyze the reaction even when the reduction degree
is low. In fact, increasing the reduction temperature of Ru/MgO to pro-
mote the reduction degree is disadvantageous to the hydrogenolysis
activity. This phenomenon has been systematically studied in our previ-
ous work [13], and it will not be discussed here.
Combination of the base strength distribution of the supports listed
in Table 2 shows that the catalyst support with more weak base sites are
more active for the hydrogenolysis of glycerol (in the case of Ru/CeO2). In
addition, it seems that the strong base sites in the catalyst support are
advantageous to the formation of lactic acid (in the case of Ru/MgO),
which is consistent with the results when the glycerol hydrogenolysis
reaction is performed over Ru/C in basic aqueous solution [10]. The
above-mentioned finding will be useful in selecting other appropriate
basic supports for the development of more efficient catalysts. Never-
theless, other factors such as the metal–support interaction may also in-
fluence the catalytic performance. Our earlier work has shown that the
metal–support interaction will lead to catalyst deactivation [13]. Fur-
ther research is needed to understand the exact effect between the
base sites of the support and the Ru sites, and it will be elucidated in
our following work.
In addition, it should be pointed out that if acidic supports [13] or
acidic additives [4–8] are employed, lactic acid will not be formed.
This is because the glycerol hydrogenolysis reaction will take place in
a different pathway, i.e., the dehydration–hydrogenation mechanism
[2,3].
4. Conclusions
Ruthenium supported on basic oxides (Ru/CeO2, Ru/La2O3 and
Ru/MgO) have been found to be effective bifunctional catalysts for the
hydrogenolysis of glycerol. Ru/CeO2 exhibited the best performance in
terms of the glycerol conversion and of the selectivity to 1,2-PDO. The
good performance of Ru/CeO2 is related to its smaller metal particle
size and the weak surface basicity feature of the support. The glycerol
hydrogenolysis reaction occurs via a dehydrogenation–dehydration–
hydrogenation mechanism. The oxidation product lactic acid is formed
by a Cannizzaro reaction from pyruvaldehyde.
3.3. The formation of lactic acid
Acknowledgments
It is worth noting that an oxidation product lactic acid can be formed
under high hydrogen pressure in reductive conditions. The conversion
of glycerol to lactic acid can be achieved by hydrothermal decomposi-
tion using NaOH as catalyst [27]. Nevertheless, high temperature up to
300 °C is needed to overcome the energy barrier, which is much higher
than the reaction temperature (180 °C) in our experiment. In order to
examine if the basic oxide supports can catalyze the hydrothermal
This work was supported by the National Natural Science Foundation
of China (No. 21302237), the Natural Science Foundation Project of CQ
CSTC (cstc2011jjA90004), the Opening Project of Key Laboratory of
Green Catalysis of Sichuan Institutes of Higher Education (LZJ1104),
and the Research Foundation of Chongqing University of Science &
Technology (CK2011B13, CK2013Z06).
H
-
OH
HC
O
C
HC
O
CH3
C
O
HO
C
O
C
O
CH3
CH2
-
OH
2-Hydroxyacrolein
Pyruvaldehyde
H
C
O
H3O+
-
O
C
O
CH CH3
OH
HO
C
O
CH CH3
OH
HO
C
CH3
-
O
Lactic acid
Scheme 2. Conversion of 2-hydroxyacrolein to lactic acid.

102
J. Feng et al. / Catalysis Communications 46 (2014) 98–102
[12] J. Feng, J.B. Wang, Y.F. Zhou, H.Y. Fu, H. Chen, X.J. Li, Chem. Lett. 36 (2007)
Appendix A. Supplementary data
1274–1275.
[13] J. Feng, H.Y. Fu, J.B. Wang, R.X. Li, H. Chen, X.J. Li, Catal. Commun. 9 (2008)
1458–1464.
[14] J. Feng, W. Xiong, Y. Jia, J.B. Wang, D.R. Liu, H. Chen, X.J. Li, Chin. J. Catal. 32 (2011)
1545–1549.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2013.11.031.
[15] J. Feng, B. Xu, W.D. Jiang, W. Xiong, J.B. Wang, Adv. Mater. Res. 549 (2012) 297–300.
[16] J. Feng, B. Xu, D.R. Liu, W. Xiong, J.B. Wang, Adv. Mater. Res. 791–793 (2013) 12–15.
[17] Z.L. Yuan, P. Wu, J. Gao, X.Y. Lu, Z.Y. Hou, X.M. Zheng, Catal. Lett. 130 (2009)
261–265.
[18] S.X. Xia, Z.L. Yuan, L.N. Wang, P. Chen, Z.Y. Hou, Appl. Catal. A Gen. 403 (2011)
173–182.
References
[1] A. Martin, U. Armbruster, I. Gandarias, P.L. Arias, Eur. J. Lipid Sci. Technol. 115 (2013)
9–27.
[2] Y. Nakagawa, K. Tomishige, Catal. Sci. Technol. 1 (2011) 179–190.
[3] J. Feng, M.L. Yuan, H. Chen, X.J. Li, Prog. Chem. 19 (2007) 651–658.
[4] J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel, C. Rosier, Green Chem. 6
(2004) 359–361.
[5] Y. Kusunoki, T. Miyazawa, K. Kunimori, K. Tomishige, Catal. Commun. 6 (2005)
645–649.
[6] T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomishige, J. Catal. 240 (2006) 213–221.
[7] T. Miyazawa, S. Koso, K. Kunimori, K. Tomishige, Appl. Catal. A Gen. 318 (2007)
244–251.
[8] M. Balaraju, V. Rekha, P.S.S. Prasad, B.L.A.P. Devi, R.B.N. Prasad, N. Lingaiah, Appl.
Catal. A Gen. 354 (2009) 82–87.
[9] S. Wang, H. Liu, Catal. Lett. 117 (2007) 62–67.
[10] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
[11] E.P. Maris, W.C. Ketchie, M. Murayama, R.J. Davis, J. Catal. 251 (2007) 281–294.
[19] Z.L. Yuan, L.N. Wang, J.H. Wang, S.X. Xia, P. Chen, Z.Y. Hou, X.M. Zheng, Appl. Catal. B
Environ. 101 (2011) 431–440.
[20] S.X. Xia, R.F. Nie, X.Y. Lu, L.N. Wang, P. Chen, Z.Y. Hou, J. Catal. 296 (2012) 1–11.
[21] S. Wang, Y. Li, H. Liu, Acta Chim. Sin. 70 (2012) 1897–1903.
[22] J.I. Take, N. Kikuchi, Y. Yoneda, J. Catal. 21 (1971) 164–170.
[23] P. Käβner, M. Baerns, Appl. Catal. A Gen. 139 (1996) 107–129.
[24] W.Q. Yu, J. Zhao, H. Ma, H. Miao, Q. Song, J. Xu, Appl. Catal. A Gen. 383 (2010) 73–78.
[25] Q.C. Xu, J.D. Lin, X.Z. Fu, D.W. Liao, Catal. Commun. 9 (2008) 1214–1218.
[26] C.M. Hung, Powder Technol. 196 (2009) 56–61.
[27] H. Kishida, F.M. Jin, Z.Y. Zhou, T. Moriya, H. Enomoto, Chem. Lett. 34 (2005) 1560–1561.
[28] J.T. Dam, F. Kapteijn, K. Djanashvili, U. Hanefeld, Catal. Commun. 13 (2011) 1–5.
[29] F. Auneau, C. Michel, F. Delbecq, C. Pinel, P. Sautet, Chem. Eur. J. 17 (2011)
14288–14299.