The effect of impregnation strategy on structural characters and CO 2 methanation properties over MgO modified Ni/SiO2 catalysts

Article abstract of DOI:10.1016/j.catcom.2014.05.022

Mg-modified Ni/SiO₂ catalysts with different MgO contents were prepared by two impregnation methods. The catalysts prepared by co-impregnation method could show better activity and stability than those prepared by sequential impregnation method

Full text of DOI:10.1016/j.catcom.2014.05.022

Catalysis Communications 54 (2014) 55–60
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short Communication
The effect of impregnation strategy on structural characters and CO2
methanation properties over MgO modified Ni/SiO2 catalysts
Meng Guo a,b, Gongxuan Lu a,
⁎
a
State Key Laboratory for Oxo Synthesis and Selective Oxidation, Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, Lanzhou 730000, PR China
University of Chinese Academy of Sciences, Beijing 100049, PR China
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Mg-modified Ni/SiO2 catalysts with different MgO contents were prepared by two impregnation methods. The
catalysts prepared by co-impregnation method could show better activity and stability than those prepared by
sequential impregnation method. The modification of MgO acted as a key factor in enhancing the capacity of
CO2 adsorption and accelerating the activation of CO2. Moreover, modified MgO could also increase Ni species
dispersion and suppress the metallic Ni sintering and oxidation.
Received 3 February 2014
Received in revised form 14 May 2014
Accepted 23 May 2014
Available online 2 June 2014
© 2014 Elsevier B.V. All rights reserved.
Keywords:
Methanation
Carbon dioxide
Supported catalyst
Modification
Stability
1. Introduction
2. Experimental
Methane as one of the renewable energy carriers can be produced
from the Sabatier reaction. If hydrogen in this reaction is provided by re-
newable energy, the sustainable cycle of carbon can be achieved. For the
above purposes, extensive studies have been carried out on group VIII
metal catalysts [1–5]. However, the industrialization of CO2 methana-
tion remains a great challenge due to the lack of the efficient and stable
catalysts [2].
For this process, the adsorption and activation of CO2 are especially
critical. Park et al. [6] have revealed that MgO can initiate the reaction
by binding a CO2 molecule via forming a magnesium carbonate species
on the surface. And then, a supply of atomic H is essential for further hy-
drogenation of magnesium carbonate to methane. Obviously, the addi-
tion of MgO can improve the catalytic performance due to its strong
basicity, which can enhance the capacity of CO2 adsorption, alter the
acid–base property of catalysts and improve the dispersion of active
phase [7,8]. In addition, different impregnation methods can affect sig-
nificantly catalyst surface characteristics and catalytic behavior [7,9,10].
Therefore, a series of Mg-modified Ni/SiO2 materials were prepared by
co-impregnation and sequential impregnation methods. The obtained
samples with good thermal stability were investigated as the catalysts
of CO2 methanation. The influence of MgO loadings and impregnation se-
quences on the performance of Mg-modified Ni/SiO2 catalysts was stud-
ied in order to find the catalysts with better activity and stability.
2.1. Catalyst preparation
Using Ni(NO3)2·6 H2O as a metal precursor, 10 wt.% Ni/SiO2 and
10 wt.% Ni/MgO (denoted as Ni/Si and Ni/Mg) catalysts were prepared
by the wet impregnation method followed by spontaneous dispersion
upon calcination. Prior to the impregnation, SiO2 and MgO supports
were calcined 773 K for 6 h. After the stabilization, their specific areas
were 438 and 8.1 m2/g, respectively. After the impregnation, the cata-
lysts were dried at room temperature for 24 h, dried further at 383 K
for another 24 h, and finally calcined at 773 K for 6 h.
10 wt.% Ni-xMgO/SiO2 (x = 1, 2 and 4 wt.%) catalysts were denoted
as co-Ni/xMg/Si catalysts and prepared by co-impregnation method
using the mixed solution containing both Ni(NO3)2 and Mg(NO3)2.
The drying and calcination procedures were the same as for Ni/Si
catalyst.
10 wt.% Ni/xMgO/SiO2 (x = 1, 2 and 4 wt.%) catalysts were denoted
as se-Ni/xMg/Si catalysts and prepared by sequential impregnation
method, in which impregnation of magnesium nitrate aqueous solution
was followed by heat treatment and then by impregnation of nickel ni-
trate aqueous solution. The two drying and calcination procedures were
the same as for Ni/Si catalyst.
2.2. Catalyst characterization
Hydrogen-temperature programmed reduction (H2-TPR) measure-
ments were carried out by heating a sample (50 mg) at 10 K min−1 in
⁎
Corresponding author. Tel.: +86 931 4968178; fax: +86 931 4968178.
E-mail address: gxlu@lzb.ac.cn (G. Lu).
http://dx.doi.org/10.1016/j.catcom.2014.05.022
1566-7367/© 2014 Elsevier B.V. All rights reserved.

56
M. Guo, G. Lu / Catalysis Communications 54 (2014) 55–60
(60 ml min−1). The products were collected after half an hour of
steady-state operation and analyzed on two on-line chromatographs
equipped with thermal-conductivity detectors (TCD). Nitrogen was
used as a carrier gas and internal standard for gas analysis. CO2 conver-
sion (XCO ), CH4 selectivity (SCH ) and CO selectivity (SCO) were de-
4
scribed as2follows:
XCO ð%Þ ¼ 100 ꢀ ðFCO −FCO2outÞ=FCO
ð1Þ
ð2Þ
ð3Þ
2
2in
2in
SCH ð%Þ ¼ 100 ꢀ FCH =ðFCH þ FCO
Þ
4
4out
4out
out
SCOð%Þ ¼ 100 ꢀ FCO =ðFCH þ FCO Þ:
out
4out
out
3. Results and discussion
3.1. Characterization of catalysts
3.1.1. H2-TPR analysis
Profiles of H2-TPR for the bare and modified Ni/SiO2 samples were
presented in Fig. 1 and the quantitative data of Ni species were given
in Table 1. TPR profile of Ni/Si catalyst was composed of a sharp peak
and a shoulder peak. This sharp peak was constituted by two maxima
at 668 and 697 K, respectively. The two peaks were attributed to the re-
duction of some inhomogeneous NiO phases that interacted weakly
with the silica [11]. The shoulder peak at 757 K could be attributed to
the stronger interactions between the bulk NiO phase and the support
[12].
Fig. 1. H2-TPR profiles of fresh co-impregnation (a) and sequential impregnation
(b) catalysts: A, Ni/Si; B, co-Ni–1Mg/Si; C, co-Ni–2Mg/Si; D, co-Ni–4Mg/Si; E, se-Ni/
1Mg/Si; F, se-Ni/2Mg/Si; G, se-Ni/4Mg/Si. The data of Ni/Si were plotted for comparison.
The reduction peaks of Mg-modified Ni/SiO2 materials gradually mi-
grated toward higher temperatures with the increase of the MgO con-
tents. However, the impregnation strategies had less influence on the
reduction behavior of the same loading catalysts due to their similar
properties in shapes and peak positions (Fig. 1 and Table 1). For co-
Ni–1Mg/Si sample, the δ, ε and ζ peaks were attributed to the surface,
subsurface and bulk phase reduction of Ni2+ ions in the outer-like
layers of MgO [13–15]. The η peak was attributed to the reduction of
Ni2+ ions located in the subsurface layers of MgO lattice [14]. With
the increase of the MgO loadings, the reduction peaks of co-Ni–2Mg/Si
and co-Ni–4Mg/Si samples were shifted significantly to higher tempera-
tures. The reduction behavior of the Ni oxides over Mg-modified Ni/SiO2
catalysts prepared by sequential impregnation and co-impregnation
methods was similar. For se-Ni/4Mg/Si sample, the H2-uptake fraction
of total area of the ζ peak at 907 K was up to 97%, which indicated
that MgO could be dispersed highly to the SiO2 support. The lower
H2-uptakes indicated that less metallic Ni species could be reduced.
In summary, compared with co-Ni–xMg/Si catalysts, se-Ni/xMg/Si
catalysts exhibited high temperature H2-uptakes, which indicated
the stronger surface interaction. In addition, the addition of higher
amount of MgO (2 or 4 wt.%) could result in more uniform NiO spe-
cies formed over Mg-modified Ni/SiO2 catalysts.
a flow 5 vol.% H2/Ar gas mixture (40 ml min−1). The amount of con-
sumed H2 was measured by a TCD.
X-ray diffraction (XRD) was performed on selected samples using
Cu Kα radiation (Philips X'pert MPD instrument) at a scattering rate
of 4°/min at 40 mA and 50 kV.
Chemical states of the atoms on the catalyst surface were investigat-
ed by X-ray photoelectron spectroscopy (XPS) on a VG ESCALAB 210
Electron Spectrometer (Mg Kα radiation; hν = 1253.6 eV). XPS data
were calibrated using the binding energy of the Si 2p (103.4 eV) as
the standard.
2.3. Catalytic performance
Catalytic performance tests were carried out in a fixed bed continuous
flow quartz reactor (i.d. 8 mm) at atmospheric pressure and temperatures
ranging from 473 to 773 K using a mixture of H2 (40 ml min−1) and CO2
(10 ml min−1) at the molar ratio of 4 balanced with N2 (30 ml min−1).
Typically, 0.2 g of catalyst was used in each turn. Previous to the reaction
tests, the calcined catalysts were reduced in situ at 723 K for 3 h and then
cooled down to room temperature under a 50 vol.% H2/N2 mixture
Table 1
H2-TPR quantitative data of fresh catalysts.
Samples
Tm (K)
Fraction of total area (%)
α
β
γ
δ
ε
ζ
η
α
β
γ
δ
ε
ζ
η
Ni/Si
668
–
697
–
757
750
771
–
–
–
–
–
27
–
43
–
30
9
11
–
–
–
–
–
2
4
1
2
3
2
co-Ni–1Mg/Si
co-Ni–2Mg/Si
co-Ni–4Mg/Si
se-Ni/1Mg/Si
se-Ni/2Mg/Si
se-Ni/4Mg/Si
810
821
790
797
811
–
870
862
842
863
863
–
917
905
904
906
905
907
1007
1014
1023
1009
1004
1012
42
22
13
30
25
–
31
13
12
26
10
–
15
50
74
16
53
97
–
–
–
–
–
–
–
–
–
–
740
751
729
–
–
25
10
1
–
–
–
–
–
–
–
–

M. Guo, G. Lu / Catalysis Communications 54 (2014) 55–60
57
3.1.2. XPS analysis
shown in Table 3. In addition, the metallic Ni peak intensity ratios of
Ni/Si, co-Ni–1Mg/Si, co-Ni–2Mg/Si and co-Ni–4Mg/Si catalysts were
1:0.29:0.12:0.04. For se-Ni/xMg/Si catalysts, the changes of peak intensi-
ties were similar. It also demonstrated that Ni species over Mg-modified
Ni/SiO2 catalyst were more difficult to be reduced than those over Ni/Si
catalyst under the same condition. However, the formation of small crys-
tallites improved the metal dispersion of Ni species and then affected the
catalyst performances, which was discussed further in the following
content. In brief, both impregnation sequence and MgO loadings could in-
fluence the catalyst structure. The catalysts prepared by sequential im-
pregnation method exhibited higher NiO dispersion. In addition, the
catalysts with high MgO loadings possessed high NiO dispersion.
The patterns related with binding energies of Ni 2p3/2 and the de-
rived atomic composition in the different samples were analyzed and
summarized in Fig. 2 and Table 2, respectively. The binding energies of
the surface Ni 2p3/2 species were not affected by different preparation
methods and MgO loadings. But the relative contents of nickel species
changed significantly. For all the catalysts, the peaks of binding energies
of Ni 2p3/2 were mainly distributed around 853.8, 854.8 and 856.8 eV,
which were assigned to different sorts of NiO species [16]. But these
NiO species were not fully in agreement with those NiO species studied
in TPR experiment. The two higher binding energies of Ni 2p3/2 (around
859.5 and 863.0 eV) could be assigned to the shake-up satellite peaks of
NiO species. The XPS results of samples in Table 2 demonstrated that Ni
species increased on the modified catalyst surface with the increase of
MgO loadings, as observed by the higher surface Ni/Si.
3.2. Influence of SiO2 and MgO supports
In order to show the influence of SiO2 and MgO supports on CO2
methanation performances over monometallic Ni-based catalysts [19],
Ni/Si, Ni/Mg and a physical mixture of two monometallic catalysts (a
weight ratio of 1:1) were investigated and the results were compared
in Table 4. The nature and texture of supports could affect significantly
the distribution of active sites for CO2 hydrogenation. CO2 methanation
occurred mainly over Ni/Si catalyst but the reverse water gas shift reac-
tion occurred largely over Ni/Mg catalyst. Previous studies [20] had
shown that the central carbon atom of adsorbed CO2 (bent) could coordi-
nate with Ni metal and make system energy reduce. At this circumstance,
adsorbed CO2 reacted with activated H2 by progressive hydrogenation
and dehydration. The catalytic performances of a physical mixture of
two monometallic catalysts were very similar to the ones of Ni/Si catalyst,
which demonstrated that CO generated from Ni/Mg catalyst could be
adsorbed once again and then be hydrogenated to CH4.
3.1.3. XRD analysis
XRD patterns of bare and modified Ni/SiO2 catalysts reduced at 723 K
for 3 h were presented in Fig. 3. The typical broad diffraction peaks of SiO2
support could be observed. For Ni/Si catalyst, the diffraction peaks located
at 2θ = 44.5, 51.8 and 76.4° (JCPDC Card No. 87-0712), indicated the ex-
istence of the characteristics of Ni metal phase [7,17]. The NiO peaks were
located at 2θ = 37.2, 43.3 and 62.8° (JCPDC Card No. 89-5881), which
suggested that some NiO species existed in the bulk phase of catalysts
[3,18].
By comparison of Mg-modified Ni/SiO2 catalysts, it was found that
the XRD peaks could be attributed to Ni phase (near 44.5 and 51.8°),
NiO (near 37.2, 43.3 and 62.8°), and NiO–MgO mixed phases (near
37.2, 43.3 and 62.8°, JCPDC Card No. 24-0712) [8]. With the increase of
MgO loadings, both metallic Ni particle sizes and peak intensities de-
creased. The metallic Ni particle sizes of Ni–1Mg/Si, Ni–2Mg/Si and
Ni–4Mg/Si catalysts were 36.2, 31.7 and 21.1 nm, respectively, as
3.3. Catalytic activity over Mg-modified Ni/SiO2 catalysts
The catalytic performances of Mg-modified Ni/SiO2 catalysts were
investigated at different temperatures and the results were shown in
Fig. 4. As shown in Fig. 4a, CO2 conversions depended largely on reaction
temperatures [2–6]. Their highest catalytic activities at 673 K were
observed in the temperature range studied. Compared with Ni/Si cata-
lyst, the catalytic activity of co-Ni–1Mg/Si catalyst increased. However,
with the further increase of MgO loadings, the catalytic activities of
co-Ni–2Mg/Si and co-Ni–4Mg/Si catalysts decreased. Moreover, co-
impregnation of Ni and Mg and the suitable content of MgO were
more beneficial to improve the activity for this reaction. In addition, as
shown in Fig. 4c, CO selectivities were also related closely to the reaction
temperatures, especially in the high temperature region ranging from
673 to 773 K. Sharma et al. [21] indicated that the reverse water gas
shift reaction competed with CO2 methanation main reaction over
Ni-based catalysts. The high reaction temperature favored the reverse
water gas shift reaction according to the thermodynamics [22]. There-
fore, the produced CO from the reverse shift reaction increased with
the increase of the temperature [23]. At the same time, high tempera-
ture was not beneficial to CO2 methanation reaction due to the con-
straint of thermodynamics [24]. These two factors might account for
the reasons of the high selectivities to CO in high temperature ranges.
3.4. Stability tests
The several catalyst performances during 50 h were monitored by
means of CO2 conversions and CH4 selectivities. Ni/Si, co-Ni–1Mg/Si
and se-Ni/1Mg/Si catalysts were selected as the representative catalysts.
The results of catalyst stability and deactivation were listed in Fig. 5. As
shown in Fig. 5, Ni/Si, co-Ni–1Mg/Si and se-Ni/1Mg/Si catalysts exhibit-
ed similar initial CO2 conversions and CH4 selectivities. However, com-
pared with Ni/Si and se-Ni/1Mg/Si catalysts, co-Ni–1Mg/Si catalyst
exhibited higher CO2 conversions, higher CH4 selectivities and more
stable catalytic behavior in the whole 50 h time on stream [3,6]. As
Fig. 2. XPS spectra of Ni 2p3/2 over fresh co-impregnation (a) and sequential impregnation
(b) catalysts: A, Ni/Si; B, co-Ni–1Mg/Si; C, co-Ni–2Mg/Si; D, co-Ni–4Mg/Si; E, se-Ni/1Mg/Si; F,
se-Ni/2Mg/Si; G, se-Ni/4Mg/Si. The data of Ni/Si were plotted for comparison.

58
M. Guo, G. Lu / Catalysis Communications 54 (2014) 55–60
Table 2
XPS results of fresh catalysts.
Samples
B.E. (Ni 2p3/2) (eV)
Fraction of total area (%)
Ni/Sia
Mg/Sia
α
β
γ
δ
ε
α
β
γ
δ
ε
Ni/Si
859.5
859.5
858.5
858.3
860.9
860.3
859.5
863.0
862.8
862.8
862.5
862.7
863.1
862.9
7.1
3.9
2.9
2.8
23.1
10
49.3
31.6
33.1
37.2
28.3
31.7
49.6
9.4
31.3
24.0
28.5
32.2
18.0
16.7
30.5
0.015
0.017
0.019
0.022
0.022
0.026
0.034
0
co-Ni–1Mg/Si
co-Ni–2Mg/Si
co-Ni–4Mg/Si
se-Ni/1Mg/Si
se-Ni/2Mg/Si
se-Ni/4Mg/Si
17.4
25.5
22.8
17.6
23.8
8.2
0.066
0.071
0.111
0.065
0.085
0.120
853.8
854.8
856.8
1.7
6.1
11.5
11.4
6.7
24.6
16.4
2.9
a
Calculated by Ni (or Mg) atom%/Si atom% from XPS spectra.
mentioned above, the excellent catalytic activities of Mg-modified cata-
lysts were attributed to the merits of CO2 activation, which made the
chemical adsorption much easier than Ni/Si catalyst [6]. During stability
tests, a fatal drawback of CO2 methanation catalysts was their deactiva-
tion caused mainly by the sintering and oxidation of the metallic phase,
which could lead to a decrease of the active metal sites on the catalyst
surface and ultimately influence the catalytic stability.
The XRD patterns of the used catalysts were plotted comparatively
in Fig. 3c. For Ni/Si catalyst, the metallic Ni phase sintered and the me-
tallic Ni species on the catalyst surface could be oxidized with the reac-
tion process, which could lower the catalytic stability. However, for the
used co-Ni–1Mg/Si catalyst, there were minor changes of the diffraction
peaks of metallic Ni species. The structure of co-Ni–1Mg/Si catalyst was
stable, which was also in agreements with the results of the stability
test. It demonstrated that in the reaction process, modified MgO by
co-impregnation method could inhibit the sintering and oxidation of
metallic Ni on the catalyst surface.
In conclusion, the results of characterization and reaction indicated
that the introduction of MgO influenced the catalyst structure and activ-
ity. Although TPR and XRD results demonstrated that MgO doping en-
hanced the interaction of NiO with the support, Mg-modified Ni/SiO2
catalysts still exhibited good activities. In addition, the excessive addi-
tion of MgO provided too much basic sites, which could inhibit the cat-
alyst activity further [8,25]. Furthermore, XPS results demonstrated that
the surface Ni species on the catalysts prepared by sequential impreg-
nation were easier to enrich than those the catalysts prepared by co-
impregnation, which resulted in the significant Ni sintering of se-Ni–
1Mg/Si catalyst in stability test. However, the synergistic effect of Ni
and Mg species over inhibited significantly the Ni sintering of co-Ni–
1Mg/Si catalyst in stability test.
4. Conclusion
In summary, the introduction of 1 wt.% MgO as a promoter to the
10 wt.% Ni/SiO2 catalyst via co-impregnation method led to superior
catalytic activity and stability (near 100%) at 623 K for 50 h because of
the synergistic effect of Ni and Mg species. In addition, although Ni/Si
Table 3
XRD results of catalysts reduced at 723 K.
Samples
Diameter (nm)a
Ni
NiO or/and NiO-MgO
Ni/Si
50.9 (61.6)b
36.2 (31.7)b
31.7
12.2 (17.7)b
43.4 (34.6)b
28.1
co-Ni–1Mg/Si
co-Ni–2Mg/Si
co-Ni–4Mg/Si
se-Ni/1Mg/Si
se-Ni/2Mg/Si
se-Ni/4Mg/Si
21.1
21.0
36.1 (44.0)b
21.1
43.4 (45.1)b
34.6
19.6
31.6
Fig. 3. XRD patterns of reduced (a, co-impregnation; b, sequential impregnation) and used
(c) catalysts: A, Ni/Si; B, co-Ni–1Mg/Si; C, co-Ni–2Mg/Si; D, co-Ni–4Mg/Si; E, se-Ni/1Mg/Si; F,
se-Ni/2Mg/Si; G, se-Ni/4Mg/Si. The data of Ni/Si were plotted for comparison.
a
Calculated by the Scherrer formula from XRD spectra.
Obtained after 50 h.
b

M. Guo, G. Lu / Catalysis Communications 54 (2014) 55–60
59
Table 4
Catalytic performances of monometallic Ni-based catalysts at different temperatures. Re-
action conditions: GHSV = 15,000 ml h−1 g−1; H2/CO2 molar ratio = 4; 1 atm.
Samples
573 K
623 K
673 K
X
S
SCO
3.5
91.7
8.3
X
S
SCO
2.5
87.5
4.8
X
S
CH4
SCO
1.3
91.0
3.2
CO2
CH4
CO2
CH4
CO2
Ni/Si
Ni/Mg
Ni/Si + Ni/Mg
30.0
2.7
23.9
96.5
8.3
91.7
64.7
7.8
54.9
97.5
12.5
95.2
73.2
28.9
66.5
98.7
9.0
96.8
Fig. 5. Long-term stability tests at 623 K. Reaction conditions: GHSV = 15,000 ml h−1 g−1
H2/CO2 molar ratio = 4; 1 atm.
;
and se-Ni/1Mg/Si catalysts exhibited similar initial catalytic activities
compared to co-Ni–1Mg/Si catalyst, their stabilities decreased to 91.2
and 89.9% at 623 K for 50 h, respectively. Besides, the addition of higher
amount of MgO (2 or 4 wt.%) resulted in a slight decrease in catalytic ac-
tivities due to the blockage of active sites. Therefore, the improved sta-
bility and activity of MgO doped Ni/SiO2 catalysts for CO2 methanation
could be achieved by applying different impregnation strategies and
adding different amounts of MgO.
Acknowledgments
This work has been supported by the 973 Program and 863 Program
of the Department of Sciences and Technology China (Grant Nos.
2013CB632404 and 2012AA051501); by the National Natural Science
Foundation of China (Grant Nos. 21373245).
Appendix A. Supplementary material
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.inoche.2014.05.023. These data include MOL files
and InChiKeys of the most important compounds described in this
article.
References
[1] M.R. Gogate, R.J. Davis, Catal. Commun. 11 (2010) 901–906.
[2] F. Ocampo, B. Louis, L. Kiwi-Minsker, A.C. Roger, Appl. Catal. A 392 (2011) 36–44.
[3] G.J. Zhi, X.N. Guo, Y.Y. Wang, G.Q. Jin, X.Y. Guo, Catal. Commun. 16 (2011) 56–59.
[4] A. Karelovic, P. Ruiz, J. Catal. 301 (2013) 141–153.
[5] W. Cai, Q. Zhong, Y.X. Zhao, Catal. Commun. 39 (2013) 30–34.
[6] J.N. Park, E.W. McFarland, J. Catal. 266 (2009) 92–97.
[7] J.Q. Zhu, X.X. Peng, L. Yao, D.M. Tong, C.W. Hu, Catal. Sci. Technol. 2 (2012) 529–537.
[8] L.L. Xu, H.L. Song, L.J. Chou, Int. J. Hydrog. Energy 38 (2013) 7307–7325.
Fig. 4. Catalytic performances of Mg-modified Ni catalysts at various temperatures: A, Ni/Si; B,
co-Ni–1Mg/Si; C, co-Ni–2Mg/Si; D, co-Ni–4Mg/Si; E, se-Ni/1Mg/Si; F, se-Ni/2Mg/Si; G, se-Ni/
4Mg/Si. Reaction conditions: GHSV = 15,000 ml h−1 g−1; H2/CO2 molar ratio = 4; 1 atm.

60
M. Guo, G. Lu / Catalysis Communications 54 (2014) 55–60
[9] A.E. Aksoylu, M. Madalena, A. Freitas, J.L. Figueiredo, Appl. Catal. A 192 (2000)
29–42.
[18] S.F. Shi, Y.Q. Wang, J.C. Ma, L. Zheng, R.Q. Yao, L.F. Zhou, J. Mol. Catal. (China) 27
(2013) 539–547.
[10] R. Morales, L. Melo, A. Llanos, F. Zaera, J. Mol. Catal. A 228 (2005) 227–232.
[11] S. Tomiyama, R. Takahashi, S. Sato, T. Sodsawa, S. Yoshida, Appl. Catal. A 241 (2003)
349–361.
[12] Y.X. Pan, C.J. Liu, P. Shi, J. Power Sources 176 (2008) 46–53.
[13] M. Kong, Q. Yang, W. Lu, Z.Y. Fan, J.H. Fei, X.M. Zheng, T.D. Wheelock, Chin. J. Catal.
33 (2012) 1508–1516.
[14] F. Arena, F. Licciardello, A. Parmaliana, Catal. Lett. 6 (1990) 139–150.
[15] Y.H. Wang, H.M. Liu, B.Q. Xu, Chin. J. Catal. 26 (2005) 1117–1121.
[16] S.F. Moya, R.L. Martins, M. Schmal, Appl. Catal. A 396 (2011) 159–169.
[17] Y.X. Zhang, Y. Zhang, Y.X. Zhao, J. Mol. Catal. (China) 27 (2013) 349–355.
[19] N. Takezawa, H. Terunuma, M. Shimokawabe, H. Kobayashib, Appl. Catal. 23 (1986)
291–298.
[20] M. Aresta, C.F. Nobile, J. Chem. Soc. Dalton Trans. 7 (1977) 708–711.
[21] S. Sharma, Z.P. Hu, P. Zhang, E.W. McFarland, H. Metiu, J. Catal. 278 (2011) 297–309.
[22] S. Abello, C. Berrueco, D. Montane, Fuel 113 (2013) 598–609.
[23] Z.K. Li, X. Hu, L.J. Zhang, S.M. Liu, G.X. Lu, Appl. Catal. A 417 (2012) 281–289.
[24] F. Ocampo, B. Louis, A.C. Roger, Appl. Catal. A 369 (2009) 90–96.
[25] M.H. Amin, K. Mantri, J. Newnham, J. Tardio, S.K. Bhargava, Appl. Catal. B 119 (2012)
217–226.