Selective Production of Oxygenates from Carbon Dioxide Hydrogenation over a Mesoporous-Silica-Supported Copper-Gallium Nanocomposite Catalyst

Article abstract of DOI:10.1002/cctc.201701679

The hydrogenation of CO₂ to oxygenates (methanol and dimethyl ether; DME) was investigated over bifunctional supported Cu catalysts promoted with Ga. The supported Cu-Ga nanocomposite catalysts were characterized by using XRD, TEM with energy dispersive X-ray spectroscopy, X-ray photoelectron spectroscopy, and H₂ temperature-programmed reduction. In comparison with Cu-SBA-15-based catalysts, the Ga-promoted catalyst prepared by the urea deposition method (Cu-Ga/SBA-15-UDP) was more active and selective for CO₂ hydrogenation to oxygenates. The use of Ga as the promoter led to increased acidic sites, which was confirmed by using NH₃ temperature-programmed deposition and IR spectroscopy with pyridine and 2,6-lutidine as probe molecules. The favorable effect of Ga on the CO₂ conversion and selectivity to oxygenates may come from the strong interaction of Ga with silica, which is responsible for the enhanced metal surface area, the formation of the nanocomposite, and the metal dispersion. Notably, the incorporation of Ga into Cu/SiO₂ led to a several-fold higher rate for methanol formation (13.12 μmol g_Cu ^?1 s^?1) with a reasonable rate for DME formation (2.15 μmol g_Cu ^?1 s^?1) compared to Cu/SiO₂ catalysts.

Full text of DOI:10.1002/cctc.201701679

Accepted Article
Title: Selective production of oxygenates from CO2 hydrogenation over
mesoporous silica supported Cu-Ga nanocomposite catalyst
Authors: Kuo-Wei Huang, Amol Hengne, Kushal D. Bhatte, Samy
Ould-Chikh, Youssef Saih, and Jean-Marie Basset
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201701679
Link to VoR: http://dx.doi.org/10.1002/cctc.201701679
A Journal of
www.chemcatchem.org

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
Selective production of oxygenates from CO2 hydrogenation over
mesoporous silica supported Cu-Ga nanocomposite catalyst
Amol M. Hengne, Kushal D. Bhatte, Samy Ould-Chikh, Youssef Saih, Jean Marie Basset*
and Kuo-Wei Huang*
Abstract: Carbon dioxide hydrogenation to oxygenates
(methanol and dimethyl ether (DME)) was investigated over
bifunctional supported copper catalysts promoted with gallium
(Ga). Supported Cu-Ga nanocomposite catalysts were
characterized by X-ray diffraction, transmission electron
microscopy with energy dispersive X-ray spectroscopy, X-ray
photoelectron spectroscopy and H2 temperature programmed
reduction. In comparison with Cu-SBA-15 based catalysts, Ga
promoted catalysts prepared by the urea deposition method
(CuGa/SBA-15-UDP) was found active and selective for CO2
hydrogenation to oxygenates. The use of Ga as the promoter
showed increased acidic sites as confirmed by the NH3-TPD,
Pyridine-IR and 2,6-lutidine-IR studies. The favorable effect of
Ga on CO2 conversion and selectivity to oxygenate may come
from the strong interaction of Ga with silica, which is
responsible for the enhanced metal surface area, formation of
nanocomposite and metal dispersion. Notably, incorporation
of Ga to Cu/SiO2 showed a several-fold higher rate for
methanol formation (13.12 mol/gCu·sec) with a reasonable
rate for the DME formation (2.15 mol/gCu·sec) as compared
to those of Cu/SiO2 catalysts.
a single-step process which can be conducted in one reactor
over bi-functional catalysts having both hydrogenating
(metallic) and dehydrating (acid) functions.[8-10]
The synthesis of methanol through hydrogenation of CO or
CO2 were mainly investigated over copper based catalysts in
the presence of various modifiers (Zn, Zr, Ce, Al, Si, V, Ti, Ga,
B, Cr, etc.).11-19 Conversion of CO2 to methanol and DME or
in situ conversion of methanol to DME in one step may help to
displace the equilibrium for CO2 hydrogenation towards
methanol production.[20] As such, a great deal of research was
devoted to the study of combinations of the well-known tri-
component (CuZnAl) methanol synthesis catalyst with acidic
zeolite (e.g. ZSM-5) as catalyst for this reaction.[21] However,
under typical hydrogenation conditions, this combination
usually yields unstable catalysts as the microporous nature of
ZSM-5 imposes severe diffusion limitations. Indeed, the use
of acid catalysts during dehydration steps often leads to the
formation of coke on the catalyst surface, which limits access
to active sites and hence leads to a decrease in activity with
time on stream.[22] There is a need to develop efficient catalyst
systems having desirable properties, e.g. high specific surface
area, highly active site, dispersion, appropriate acid/basic
sites distribution and smaller particle size in order to increase
the activity and selectivity to oxygenates during CO2
hydrogenation.[23] Utilization of a suitable support could lead to
the formation and stabilization of the active phase of the
catalyst as well as tuning the interactions between the major
component and the promoter.[24]
Introduction
Methanol and dimethyl ether (DME) production from the
selective hydrogenation of carbon dioxide could be a useful
method for the utilization of less hazardous CO2 as an
alternative to toxic and reactive CO, which is extensively used
nowadays at the industrial scale for the production of several
chemicals and fuels.[1] Moreover, selective conversion of CO2
to value added products could be a green and economically
viable strategy towards the mitigation of global warming.[2]
Both methanol and DME are currently considered as suitable
starting components for various downstream processes. In
particular, DME has been recently considered as a potential
clean oxygenated additive to overcome the emission
problems of the diesel engine. Furthermore, it is nontoxic and
can be used to replace LPG for cooking, heating, electrical
power generation, etc.[3-5] Despite numerous studies devoted
to the catalytic synthesis of DME, there are still various
Because of the high surface area and wide pore volume of
SBA-15, one would expect a better catalytic performance of
supported Cu-Ga mixed oxides during the production of
oxygenates from CO2 hydrogenation. Herein, we present the
results dealing with the preparation, characterization, and
catalytic activity of Cu-Ga mixed oxides supported on
mesoporous silica (SBA-15) catalysts.
Results and Discussion
The results of surface area, pore volume and pore size
determined using N2 adsorption at (-196 C) over supported
Cu-Ga catalysts are summarized in Table 1. The specific
surface area of SBA-15 is 722 m2 /g, with an average pore
size of 6.2 nm. After the deposition of Cu and/or Cu-Ga, the
BET surface area decreased significantly. This was further
confirmed with the linear decrease in the pore volume. All the
catalysts showed type (IV) N2 adsorption and desorption
isotherms, and the results were consisted with the H2
hysteresis loop, indicating that they all possess cylindrical
pores having modified pore structures.[25] The isotherms of Cu
and/or Cu-Ga incorporated catalysts showed slightly sharp
and destructed H2 hysteresis loop as compared to the parent
SBA-15, presumably because of the partially deformed
ordered structures, as seen from the space between the pores
challenges to overcome before practical industrial
[6,7]
applications
Two main routes are envisaged for the
production of DME from CO2, i) a two-step process, whereby
methanol produced on a metallic catalyst in the first step is
subsequently dehydrated over solid acid catalyst to make
DME, and ii)
King Abdullah University of Science and Technology, Division
of Physical Science and Engineering, KAUST Catalysis
Center, Thuwal, Kingdom of Saudi Arabia. *corresponding
author
email:
hkw@kaust.edu.sa,
(K.-W.H)
jeanmariebasset@kaust.edu.sa (J.M.B)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
of silica wall (Figure S1a-b). The obtained differences in pore
size distribution by using N2 desorption method could come
from the high concentration of Cu-Ga in the channels of silica,
but there is no obvious change found in the pore size
distribution in the recovered catalysts. These observations
suggest that the mesoporous nature of silica support was only
altered to slightly smaller pore sizes with the UDP method
(Figure S2a-b).
XPS of 16Cu10Ga/SBA-15-IMP and 10Cu5Ga/SBA-15-UDP
catalyst samples after reaction exhibit an intense doublet at
932.8 and 952.4 eV (Figures S3 and S4), which can be
attributed to the electron ejection form the Cu2p3/2 and Cu2p1/2
core shells of metallic Cu, respectively. The disappearance of
satellite peaks, characteristic of Cu2+ species, suggests that
most of the surface CuO has been reduced to the metallic form
after the reaction.
Table 1. Textural properties of supported Cu-Ga catalysts.
Catalyst
Surface
Area
(m2/g)
Pore
Pore
Size
(nm)
Particle
volume
(cc/g)
Size (nm)*
CuO
a
SBA-15-
722
502
1.1
6.2
6.2
6.1
N/A
15
10Cu5Ga/SBA-15-UDP
Calcined
9Cu/SBA-15-
IMP calcined
0.79
0.66
16Cu2Ga/SBA- 359
15-IMP
20
16Cu10Ga/SBA-15-IMP
Calcined
10Cu5Ga/SBA- 387
15-UDP
0.58
0.56
6.1
28
Calcined
10Cu5Ga/SBA- 343
15-UDP
6.07
N/A
20
30
40
50
60
70
80
2
Recovered
*Particle size (nm) calculated by N2O method, N/A: Not
applicable.
XRD patterns of calcined CuGa catalysts prepared by
impregnation and urea deposition methods are shown in
Figure 1a. The appearance of indexed diffraction lines at 2=
35.7°, 38.8° 48.1°, and 64.1° can be attributed to the presence
of CuO crystalline phase [JCPDS file no. 80-1268]. The XRD
patterns of all used 16Cu10Ga/SBA-15-IMP and
10Cu5Ga/SBA-15-UDP catalysts exhibited diffraction peaks
characteristic of the inert carborundum, which is used to dilute
the catalyst during the reaction (Figure 1b). Those catalyst
samples recovered after the reaction showed peaks at 2=
43.5° (111), 50.6° (200) and 74.3° (220), indicative of the
presence of metallic Cu in all used catalysts [JCPDS file no.
04-0836]. In all the samples no characteristic peaks of Ga
oxides were observed indicating either amorphous nature of
Ga oxides and/or being diffused in the bulk matrix of the silica
support. The presence of a small peak at 2= 36.50° is likely
due to the formation of Cu2O. The latter might be formed via
partial oxidation of metallic Cu while exposing the samples to
air before XRD or during the preparation procedures as
discussed in the XPS study. The estimated particle sizes of
metallic Cu were found to be in the range of 10-20 nm based
on the calculation from the Scherer equation using the intense
peak at 2= 43.5, allowing us to correlate with the particle
sizes obtained from the HRTEM study.
Cuo
b
CuO
Cu2O
inert -C
Carborundum
10Cu5Ga/SBA-15-UDP
16Cu10Ga/SBA15-IMP
20
30
40
50
60
70
80
2
Figure 1. XRD patterns for: (a) Calcined 16Cu10Ga/SBA-15-IMP and
10Cu5Ga/SBA-15-UDP (b) Carborundum, Recovered
16Cu10Ga/SBA-15-IMP and 10Cu5Ga/SBA-15-UDP.
This observation is consistent with the TPR results (Figure 3),
clearly indicating the almost complete reduction of CuO in all
the catalysts under the hydrogenation conditions.
Distinguishing Cu0 from Cu+ species based solely on their Cu
2p binding energies is not unambiguous. Nevertheless,
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
judging from the asymmetric shape of Cu LMM Auger spectra,
one could tentatively conclude that both Cu+ and Cu0 are
present on the catalyst surface, as observed in
stack in the pores of SBA-15 (Figure 3 b). The Cu-Ga particles
observed in 10Cu5Ga/SBA-15-UDP showed lattice fringes,
which is mainly corresponding to Cu {111} and {220} planes,
in agreement with XRD results. The elemental mapping
demonstrate that Ga and Cu atoms tends to segregate (Figure
3 c). While the copper atoms tends to form nanoparticles
(Figure 3 d), gallium atoms seems to form an amorphous
phase inside the SBA-15 pores.
10Cu5Ga/SBA-15-UDP
16Cu10Ga/SBA-15-IMP.[26]
compare
to
impregnation
Similarly, the intense doublets at 1118.4, 1118.16 and 1145.2
eV represent the binding energy of Ga2p3/2 and Ga2p1/2
indicative of the Ga3+ oxidation state. HRTEM images of the
10Cu5Ga/SBA-15-UDP catalyst before and after reaction are
shown in (Fig. 2). The SBA-15 support shows the
characteristic hexagonal pore structure with particle sizes in
the range of 100-300 nm (Figure S4a). Using the impregnation
method to load Cu and Ga (16Cu10Ga/SBA-15-IMP) species
appeared to cause partial destruction of the ordered
mesoporous structure of the SBA-15 support.
Figure 3. Dark field STEM image of Cu5Ga/SBA-15-UDP activated
sample (a and b), c) segregation of Ga and Cu atoms highlighted by
elemental mapping from STEM-EELS. Yellow square on image c)
show the area where the EELS spectrum image was recorded. The
inset in c) is a colored overlay of the EELS maps corresponding to Cu
and Ga L-edge. Image d) shows metallic Cu nanoparticles left outside
the SBA-15 support. The inset in image d) is a fast Fourier transform
applied to the whole picture which show characteristic d-spacings
attributed to {220} and {111} planes of Cu metal.
The metallic Cu is generally the active species (or at least
involved in the activation of H2) in the catalytic hydrogenation
of CO2 and one can thus expect a better activity to be
observed when highly dispersed Cu species are prepared (at
lower temperatures). Accordingly, H2-TPR experiments were
performed in order to investigate the reducibility of Cu for all
prepared catalysts (Figure 4). All samples exhibited a sharp
reduction feature centered between 160 and 280 C. These
peaks are attributed to the consumption of H2 during the
reduction of CuO to the metallic Cu.[27] In the case of
9Cu/SBA-15-IMP catalyst, such H2 consumption reached its
maximum at 267 C, but the maximum H2 consumption
occurred at a lower temperature in the presence of Ga. These
Figure 2. HRTEM images 10Cu5Ga/SBA-15-UDP a) Calcined, b-c)
Activated d) Reaction recovered.
In this case, the Cu-Ga nanocomposite particles have an
average size between 20-40 nm and are present both on and
outside the surface of the SBA-15 support (Figures S4b-d).
Meanwhile, the UDP method yielded better dispersed
(10Cu5Ga/SBA-15-UDP) supported Cu-Ga species with a
relatively smaller average particle size between 5 to 20 nm
(Fig. 2b-c), in good agreement with the value of 15 nm
estimated from the XRD patterns of the used catalysts (vide
supra). Moreover, the activated sample 10Cu5Ga/SBA15-
UDP has been further explored by using DF-STEM imaging
and mapping of the elements by STEM-EELS (Figure 3). The
images suggested that the Cu-Ga particles are well dispersed
in the mesoporosity of SBA 15 support (Figure 3 a). The high
magnification image showed several Cu-Ga nanocomposites
observations suggest that Ga may activate H2 and hence
40]
improve the Cu reducibility.[28,
Addition of gallium is
expected to lead to a better reducibility and higher dispersion
of supported copper oxide species (Cf. the results of XRD and
HRTEM). The reduction of well dispersed Cu surface in
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
10Cu5GA/SBA-15-UDP indicated the sequential reduction
pattern for copper oxide (Cu+2) to metallic Cu (Cuo) via Cupric
oxide (Cu+).
250
10Cu5Ga/SBA-15-UDP
16Cu10Ga/SBA-15-IMP
9Cu/SBA-15-IMP
261
267
Figure 6. 2,6-lutidine desorption on 10Cu5Ga/SBA15-UDP; a)
activated sample b) 200 °C c) 300 °C d) 400 °C.
100
200
300
400
500
600
700
800
900
Temp (C)
In order to optimization of acidic sites and strength, the acidic
characters of the catalysts were further established by IR
using both pyridine and 2,6-lutidine as probe molecules
(Figures S6-S11). The peaks at 1455 cm-1 and 1545 cm-1 are
corresponding to the nature of Lewis and Bronsted acid sites,
respectively. Pyridine adsorption followed by FTIR further
Figure 4. H2 TPR study of copper and Cu-Ga oxide catalysts.
2000
confirmed the presence of Lewis acid sites on all catalysts.[29-
9Cu/SBA-15-IMP
16Cu10Ga/SBA-15-IMP
30]
10Cu5Ga/SBA-15-UDP
1500
On the other hand, 2,6-lutidine was used to identified weak
and moderate Bronsted acid sites.[31] SBA-15 supported
monometallic Cu catalysts did not show any peak of Bronsted
acid sites, but peaks at 1618 and 1583 cm-1, corresponding to
the weak Lewis acid sites using 2,6-lutidine. However,
incorporation of Ga and Cu-Ga to silica exhibited Bronsted
acid sites peaks at 1625 and 1645cm-1 along with
characteristic peaks of Lewis acid sites using both probe
molecules (Figure 6). Estimation of acid sites were conducted
at different temperatures from 100 to 400 °C, Cu (9Cu/SBA-
15-IMP) and Cu-Ga (10Cu5Ga/SBA-15-UDP) containing
catalysts did not showed measurable stability of acid sites
above 300 °C, while some of acid sites were observed in
5Ga/SBA-15.
The combination of supported Cu-Ga (10Cu5Ga/SBA-15-
UDP) catalyst showed both Lewis and Bronsted acid sites with
a strong acid strength (178 µmol/g) at 200 °C, three times
more than those of monometallic 9Cu/SBA15-IMP catalysts
(200 °C) (Table 2 and Figures S6-S11). These observations
clearly suggest that the higher acidic strength of SBA-15
supported Cu-Ga play an important role in the dehydration of
methanol to DME.
1000
500
0
0
100
200
300
400
500
600
700
800
Temp (oC)
Figure 5. NH3 TPD profiles of supported copper catalysts.
Acidity measurements (NH3-TPD) were also carried out
(Figure 5). Monometallic Cu catalysts showed ammonia
desorption peak at temperatures from 100-250 C, indicative
of the presence of weak acid sites and similar results were
observed with 16Cu10Ga/SBA15-IMP. In contrast, Ga
incorporated 10Cu5Ga/SBA15-UDP catalysts showed two
NH3 desorption peaks in both temperature ranges from 100-
300 C and 350-500 C, suggesting that addition of Ga
increases the acid strength compared to those of the
monometallic Cu catalyst and Cu-Ga prepared by the
impregnation method.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
Table 2 Acidic Strength of Cu, Ga and Cu-Ga catalysts using (pyridine and
2,6-lutidine) IR study
impregnation method, showed almost 99% selectivity to
methanol. Monometallic 10Cu/SBA-15 catalyst showed very
low conversion (around 0.05 %) with no formation of DME
whereas
(16Cu10Ga/SBA-15Imp)
introduction
of
Ga
using
impregnation
improves the
Catalyst
Lewis acid Bronsted
Total Acid
strength
significantly
strength
acid
conversion to reach 1 %. Moreover, we detect the formation
of DME with selectivity 22%. When using urea deposition-
precipitation method to add Ga (10Cu5Ga/SBA-15UDP)
showed increase in conversion up to 3% with both methanol
and DME selectivity 71% and 29% respectively. Under the
present operating conditions, the conversion of CO2 is
generally low. This could be due to the low residence times
used in the present investigation. The effect of contact time on
product activity and selectivity is further discussed below.
The results of the catalytic hydrogenation of CO2 using
different copper catalysts at 250 C, 25 bar and a weight
hourly space velocity around 30000 h-1 are summarized in
Figure 7. The major products were methanol (MeOH) and
DME. Only traces of carbon monoxide (CO) could be detected
in the outlet stream. All catalytic data were taken when the
catalyst performance reached a steady state. The effect of Ga
on the catalytic performance was assessed in terms of the
formation rate of oxygenates. The monometallic Cu catalyst
(Cu-SBA-15) showed a low methanol formation rate of 0.69
mol/gCu·sec without any detectable DME formation.
Interestingly, Ga promoted Cu catalysts prepared by
impregnation (16Cu10Ga/SBA-15-IMP) and urea deposition
method (10Cu5Ga/SBA-15-UDP) showed eight to ten times
increase in the methanol formation rate (7.19-13.12
mol/gCu·sec), with the formation rate of DME between 0.084
and 2.15 mol/gCu·sec. The lower activitiy of
mol/g*)
strength
mol/g#)
(mol/g*#)
9Cu/SBA-15- IMP 60
calcined
<0.01
60
5Ga/SBA-15- UDP 85
Calcined
63
148
178
10Cu5Ga/SBA-15-
UDP Calcined
140
38
*Acid strength calculated from pyridine as a probe molecule measured
at 200 °C
#Acid strength calculated from 2,6-lutidine as a probe molecule
measured at 200 °C
The results of the catalytic hydrogenation of CO2 using
different copper catalysts at 250 °C, 25 bar and a weight
hourly space velocity around 30000 h-1 are summarized in
Table 2. The major products were methanol (MeOH) and
dimethyl ether (DME). Only traces of carbon monoxide (CO)
could be detected in the outlet stream under the present
operating conditions. All catalytic data were taken when the
catalyst reached a steady state.
Table 3 Catalytic performance for CO2 hydrogenation over supported
16Cu10Ga/SBA-15-IMP
compared
to
those
of
Catalyst
Carbon Conversion,
Selectivity, %
10Cu5Ga/SBA-15-UDP are likely a result of the increase in
particle sizes and agglomeration of particles during the
preparation of Cu-Ga nanocomposites by wet impregnation
method (Figure S5).
Balance,
%
%
CH3OH
CO
DME
10Cu5Ga/
SBA-
99.95
99.95
3
2
71
73
<0.01
29
27
14
15UDP
4Cu2Ga/S
BA-
CH3OH
DME
12
<0.01
10
8
15UDP
10Cu/SBA
-15IMP
100
0.045
1
99.9
78
<0.01
<0.01
<0.01
22
6
16Cu10Ga
/SBA-
99.96
4
15IMP
2
17Cu13Zn
/SBA-
99.98
1
99.9
<0.01
<0.01
0
9Cu/SBA-15-IMP
16Cu10Ga/SBA-15-IMP 10Cu5Ga/SBA-15-UDP
Catalysts
15IMP
Figure 7. Formation rate of methanol and DME. Reaction conditions:
T= 250 °C, P = 25 bar, Gas Flow (CO2/Ar/H2)= 5/5/15 ml/min, GHSV
30000 h-1, CO2/H2 = 1:3, Time, 24h.
copper catalysts
Reaction conditions: T= 250 °C, P = 25 bar, Gas Flow (CO2/Ar/H2)
= 5/5/15 ml/min, GHSV 30000 h-1, CO2/H2 = 1:3, Time, 24h
Initially, copper in combination with Zn catalyst
(17Cu13Zn/SBA-15 Imp), which is prepared by using
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
The weak basic urea used as the precipitation agent in the
homogeneous phase helped to avoid agglomeration of
primary particles to offer uniform sizes and shapes
(TableT1).[32] Addition of Ga to Cu/SBA-15 catalysts seemed
to promote the catalytic formation of methanol and DME. The
formation of DME upon addition of Ga (Figure 7), can be
attributed to the dehydration of methanol in the presence of
Bronsted acid sites as demonstrated in the FT-IR study
(Figures S7 and S9). Indeed, the introduction of Ga via UDP
yielded an even more active catalyst. The results of XRD, H2-
TPR, N2O titration and HRTEM are consistent with the fact
that the introduction of Ga via UDP preparation method would
improve both reducibility and dispersion of the supported Cu
species. Accordingly, one can ascribe the better catalytic
performance of 10Cu5Ga/SBA-15-UDP to better reducibility
and higher dispersion of supported Cu species. While the
exact mechanism of the Ga promotion is still unclear, literature
studies on analogous Pd/β-Ga2O3 catalysts suggested that
the enhancement of the activity may come from gallia surface,
where carbon dioxide is chemisorbed to form carbonate
species along with the dissociative absorption of H2 on
gallium.[33] In addition, it has also been shown that Ga allows
the formation of Bronsted and Lewis acid sites in combination
of silica,[34] which subsequently dehydrate the formed
methanol to dimethyl ether.
10Cu5Ga/SBA-15-UDP
0.012
0.010
0.008
0.006
0.004
0.002
0.000
b
16Cu10Ga/SBA-15-IMP
9Cu/SBA15-IMP
14
16
18
20
22
24
26
28
30
Particle size (nm)
Figure 8. the Relationship between TOF of CH3OH Vs a] Cu surface
area, b] Cu- particle size. Reaction conditions: T= 250 °C, P = 25
bar, Gas Flow (CO2/Ar/H2)= 5/5/15 ml/min, GHSV 30000 h-1, CO2/H2
= 1:3, Time, 24h.
Interestingly, the TOF of methanol decreases linearly with
increasing the average Cu particle sizes, while increasing Cu
metal surface area favors the increase in TOF of methanol.
These results suggest that the addition of Ga may promote
the formation of smaller Cu particles and thus increase the
It has been demonstrated from previous studies that the
intrinsic activity of Cu, measured as turnover frequency (TOF),
depends on the Cu particle sizes. To get further insight into
the effect of Ga addition on the intrinsic activity of supported
Cu metallic species during the hydrogenation of CO2 at 250
°C and 25 bar, the TOF (s-1) vs. Cu particle size (nm) as well
as TOF (s-1) vs. Cu surface area (m2/g) were determined
(Figure 8).
metal surface area for a higher TOF for methanol formation.[35-
36]
We hypothesize that the Ga present on the surface could act
as a strong anchoring site for the supported Cu species
leading to higher dispersion upon reduction. Therefore, the
excellent catalytic performance of our Cu-Ga catalysts
supported on mesoporous silica can be rationalized by the
high surface area of support and uniform pore size
distribution.
10Cu5Ga/SBA-15-UDP
0.012
a
0.010
Effect of temperature
16Cu10Ga/SBA-15-IMP
0.008
0.006
0.004
Figure 9 shows the CO2 conversion obtained under different
temperatures ranging from 200 to 250 °C with a constant
pressure (25 bar) and the same contact time (WHSV 30000 h-
1). As expected, the CO2 conversion increased from 0.75 to
3% with increasing temperature from 200 to 250 °C.
Meanwhile, the selectivity to methanol was decreased from 87
to 71% with a concomitant increase in the selectivity to DME.
If one assumes that the formation of DME proceeds through a
sequential mechanism whereby hydrogenation of CO2 leads
first to the formation of methanol, one can conclude that the
dehydration of methanol to DME has a much higher activation
energy, and hence a lower selectivity is observed at lower
temperatures.
9Cu/SBA15-IMP
0.002
0.000
20
25
30
35
40
45
50
Cu S.A. (m2/g)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
in the selectivity of MeOH with a concomitant increase in the
selectivity of DME. These observations suggest that the
reaction may proceed via a sequential mechanism whereby
CO2 is first hydrogenated to MeOH and dehydrated later to
DME. Under the present operating conditions, no CO
formation was detected. The catalyst was further examined at
a higher catalyst loading (five times) at 200-250 °C (Figures
S11&S12). It was found that the conversion of CO2 was
increased from 2% to 11% with increasing the reaction
temperature from 200 to 250 °C.
However, the selectivity to oxygenates (methanol and DME)
was lowered from 78% to 45% with CO formation from 22-
55% from temperature 200 to 250 oC. It was clearly
demonstrated that with a low catalyst loading and contact time
highly selective production of oxygenates can be achieved
(99%).
30
25
20
15
88
86
84
82
80
78
76
74
72
70
CO2 conversion, %
10
5
DME
CH3OH
0
200
210
220
230
240
250
Temperature, oC
Effect of total (CO2 and H2) pressure
Figure 9. Effect of temperature on CO2 hydrogenation and oxygenate
selectivity over 10Cu5Ga/SBA-15-UDP. Reaction conditions: T= 200-
250 °C, P = 25 bar, Gas Flow (CO2/Ar/H2)= 5/5/15 ml/min, GHSV
30000 h-1, CO2/H2 = 1:3, Time, 24h.
As stated in the literature, high pressures are necessary for
selective hydrogenation of CO2 to methanol.[38-39] The CO2
conversion was about ≤0.5% at the reaction pressure of 1 bar
(Figure 11).
Under the present operating conditions, no CO formation was
detected. This is expected due to the endothermic nature of
the RWGS reaction, which is usually thermodynamically
limited at low temperatures.[37] In fact, when the temperature
was further increased to 300 C (25 bar and 30000 h-1), CO
was formed as a main product (selectivity around 60%;
Figures S10).
86
30
84
82
25
80
CO2 Conversion, %
20
DME
78
Effect of contact time
CH3OH
76
15
74
10
72
70
68
5
0
0
5
10
15
20
25
Reaction Pressure (bar)
Figure 11. Effect of H2 pressure on CO2 hydrogenation and oxygenate
selectivity over 10Cu5Ga/SBA-15-UDP.Reaction conditions: T= 250
°C, P = 1, 10 and 25 bar, Gas Flow (CO2/Ar/H2)= 5/5/15 ml/min, GHSV
30000 h-1, CO2/H2 = 1:3, Time, 24h.
When the pressure was increased to 25 bar, an increase in
CO2 conversion to 0.5-3% was observed. The improvement of
selectivity to DME from 15 to 29% was also achieved with
increasing pressure from 1 to 25 bar.
Figure 10. Effect of contact time on CO2 hydrogenation and
oxygenate
selectivity
over
10Cu5Ga/SBA-15-UDP.Reaction
conditions: T= 250 °C, P = 25 bar, Gas Flow (CO2/Ar/H2)= 5/5/15
ml/min, CO2/H2 = 1:3, Time, 24h.
Time on stream activity
The hydrogenation of CO2 over 10Cu5Ga/SBA-15-UDP was
studied by varying the contact time using the catalyst form the
same batch with various feed flows (Figure 10). The
conversion was found to increase linearly when increasing the
contact time, Using low contact time, and the selectivity to
methanol is high whereas the selectivity to DME is relatively
low. Further increasing the contact time leads to a decrease
Considering the interesting results obtained with the
10Cu5Ga/SBA-15-UDP catalyst, a time on stream test was
conducted while keeping all other parameters constant
(Figure 6). The catalyst maintained a very high productivity for
the methanol formation at 13.12 mol/g·sec along with a DME
formation rate of 2.15 mol/g·sec over 50 h (Figure 12). Our
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
catalysts showed stable performance during time on stream
stability test presumably due to the high surface area of the
SBA-15 support with availability of the pores and the formation
of stable Cu-Ga nanocomposite to effectively avoid the
thermal agglomeration under the hydrogenation conditions.
Ga acts as a promoter and stabilizer, and can create the
appropriate acidic sites and enhance the acidic strength in
combination with silica, to favor the dehydration of methanol
to DME.
orthosilicate (TEOS) and Urea were purchased from Aldrich
and used directly.
Catalyst preparation
SBA15
Mesoporous silica SBA-15 was prepared according to the
literature.[40] 2.0g of amphiphilic triblock copolymer (P123) was
dissolved in 15.0 ml of water and mixed with 60.0 g of 2M
HCl(aq), and to this solution, 4.250 g of tetraethyl ortho-silicate
(TEOS) were slowly added under constant stirring. The
resulting mixture was stirred for additional 24h at 40 °C and
then aged in a Teflon bottle at 100 °C for 48h. The solid
precipitate was filtered and washed with deionized water until
a neutral pH was reached. The solid was dried in open air and
finally calcined in static air at 550 °C for 24 h to decompose
the triblock copolymer.
3.0
14
2.5
12
10
8
2.0
1.5
1.0
0.5
0.0
Impregnation method IMP-A
CH3OH
DME
6
Supported Cu, Ga and Zn catalysts were prepared by the wet
impregnation method. The synthesis was performed by
suspending 0.50 g of SBA-15 in the aqueous medium with
specific amounts of the respective metal precursors such as
Cu nitrate, Zn nitrate and Ga nitrate. The suspension was
stirred for 8 h at room temperature and filtered. The resulting
solid was dried at 120 °C for 12 h and calcined at 450 °C for 8
h.
4
2
0
0
10
20
30
40
50
Time (h)
Figure 12. Time on stream activity for formation rate of oxygenate
over 10Cu5Ga/SBA-15-UDP. Reaction conditions: T= 250 °C, P = 25
bar, Gas Flow (CO2/Ar/H2)= 5/5/15 ml/min, GHSV 30000 h-1, CO2/H2
= 1:3, Time, 50h.
Urea Deposition Co-precipitation UDP-B
To an aqueous suspension of specific amount of the Cu nitrate,
Ga nitrate, urea, 0.50 g of calcined SBA-15 was slowly added
and then the mixture was stirred and heated at 90 °C for 8 h.
The resulting precipitate was filtered, dried in static air at
120 °C for 12 h, and finally calcined at 450 °C for 8 h.
As a result, Cu-Ga nanocomposite supported on SBA-15
plays a vital role to maintain the consistent activity and
productivity to oxygenates during time on stream.
Conclusion
Catalyst testing
In the single step oxygenates synthesis from CO2
hydrogenation, the Cu-Ga catalysts are highly active and
selective for the methanol and DME production because
of increasing acidic sites (Bronsted and Lewis), i.e. metal
(Cu) surface area and Cu particle size. The presence of
Ga indeed enhances both the rate of methanol formation
and the product selectivity of DME (30-40%) as compared
to those of supported Cu catalysts. These Cu-Ga
catalysts also showed excellent performance in the
stability test for up to 50 h with constant selectivity for the
formation of oxygenates (99%).
All catalytic tests were conducted in gas phase using a fixed
bed isothermal flow reactor purchased from Process Integral
Development and engineering Technology (PI&D Eng. &
Tech.). In a typical experiment, 0.50-0.250g of the as prepared
catalysts (vide supra) diluted with inert carborundum (0.950-
0.750g) were loaded into a stainless-steel (SS-316) reactor
(length 30 cm and 9 mm i.d.). The SS tube is equipped with a
20 m mesh size porous plate made from Hastelloy C, which
is used to hold the catalyst inside the reactor. The catalyst
was then reduced at 250 °C in a flow of 5% H2/Ar (5 mL/min
@ NTP) for 12 h and later cooled to room temperature. The
reactor was then heated to 250 °C (50 °C/min) under a mixture
of gases (5mL/min@NTP) CO2, (15mL/min@NTP) H2 and
(5mL/min@NTP) Ar (1:3:1) having total feed flow (25 mL/min
@ NTP). The pressure (1 to 25 barg) inside the reactor was
controlled using an automated micrometric valve.
Experimental Section
Materials
High purity gases (Hydrogen 99.9995%, CO2, argon
99.9999%) were purchased from Specialty Gases Center of
Abdullah Hashim Industrial Gases & Equipment Co. Ltd.
(Saudi Arabia). Copper nitrate, gallium nitrate, zinc nitrate,
amphiphilic triblock copolymer [poly(ethylene glycol)-block-
poly(propylene glycol)-block-poly(ethylene glycol)], tetraethyl
On-line gas analysis of the products was performed on a
Varian 450 GC gas chromatograph. A sample from the reactor
outlet stream was automatically injected on three parallel
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
channels referred to here as channel A, channel B and
channel C. In Channel A, the sample (1ml @ STP) was
injected on a set of three packed columns, “Hayesep” Q
(CP81073), “Hayesep” T (CP81072), and “Molsieve” 13X
(CP81073) connected in series. A set of 10-way and 6-way
Valco valves were used to allow automatic injection of the
sample, backflushing of Hayesep T, and bypassing of
Molsieve 13X columns. This channel was equipped with a
TCD detector (He as reference gas) and used to monitor the
amounts of CO and CO2. Channel B uses a set of two capillary
columns, CP-Wax 52 CB (CP8553) as a pre-column and Al2O3
MAPD (CP7432) connected in series. This channel was used
to monitor methane, ethane, and ethylene. Meanwhile,
channel C uses a CP-wax 52CB column (CP7668) to separate
oxygenates. Each one of these two channels was equipped
with a Flame Ionization Detector (FID). A 10-way Valco valve
was used to allow simultaneous injection of the gas samples
on both Channels A (500 l @ STP was injected) and B (250
l @ STP was injected). Estimation of CO2 conversion was
determined directly from the carbon balance, based on the
product concentrations for which the sensitivity is higher. The
formation rate of methanol is defined in micromoles of
methanol produced per gram of Cu in the catalyst per second.
Turn over frequency (TOF) is expressed as the number of
methanol molecules produced per surface copper atom of the
reduced catalyst per second.
A special care was taken with the definition of the energy
window and background for Ga L-edges in order to avoid the
overlap with the Cu L1-edge.H2 Temperature programed
measurements (TPR) were carried out using an Altamira
Instrument (AMI-200Ip) equipped with a TCD detector to
measure the H2 uptake. The catalyst sample (about 0.0250g)
was kept in a U-shaped quartz tube and pretreated under Ar
flow (50 ml/min) for 2h at 250 °C in order to remove moisture
and other surface impurities present on the sample. Further,
the sample was allowed to cool down at room temperature
under flow of Ar and then the gas was replaced by a mixture
of 5% H2/Ar for the TPR experiment. Finally, the temperature
was raised from 50 °C to 750 °C (10 °C/min) under a flow of
5% H2/Ar flow (50 ml/min). The X-ray photoelectron
spectroscopy (XPS) of catalyst samples were carried out
using Kratos AMICUS/ESCA 3400 spectrometer, using a
monochromatic Mg Kα X-ray source operating at 120W and a
pass energy of 20 eV under an operating pressure of 10-6
mbar. The binding energies of Cu2p and Ga2p core levels are
referenced to the C 1s core level at 285.6 eV.
Copper surface area and dispersion were measured by a
nitrous oxide decomposition method using Altamira
Instrument (AMI-200Ip). In a typical experiment, 0.030g of the
catalyst sample were first reduced under 5% H2/Ar at 250 °C
for 2 h followed by purging with He for 30 min and cooling to
60 °C. The catalysts were then exposed to 10%N2O/He for 1h,
in order to oxidize surface copper atoms to Cu2O. Finally, the
samples were cooled to room temperature and subjected to
temperature programed reduction (TPR) from 50 °C to 350 °C
(10 °C/min) using 5% H2/Ar flow to reduce Cu2O back to
metallic Cu. Cu surface area and dispersion were calculated
from Cu surface area and dispersion were calculated from the
amount of H2 consumed during the TPR step by assuming that
copper crystallites are spherical.[41-42] Equation 1 is used to
calculate Cu surface area
Characterization methods
Surface area measurements were carried out using N2
adsorption/desorption isotherms at 77 K on a Micromeritics@
ASAP2420. Prior to measurements, the catalyst samples
were degassed in vacuum for 2 h at 250 °C and then surface
areas of the samples were analyzed by using the multipoint
BET analysis method in the P/P0 = 0.05–0.30 pressure
range.X-ray diffraction (XRD) patterns of all samples were
collected using Bruker D8 Advanced A25 diffractometer with
Cu K radiations, which is operated at 40 kV and 40 mA. The
data sets were acquired in the step scan mode in the range of
20–80, using a step interval of 0.05 and a counting time of
10o/min.
The TEM specimens were prepared by placing 1 μL of solution
on carbon coated copper grids. The grids were then plasma
cleaned for 50 seconds to reduce hydrocarbon contamination
(Solarus, model 950). HRTEM images were obtained with a
Tecnai T12 operated at 120 kV. Dark-field scanning
transmission electron microscopy (DF-STEM) was performed
on a Titan G 60–300 ST electron microscope at an
accelerating voltage of 300 kV using a Gatan STEM
detector (model 806). The elemental compositions of the
samples were characterized based on the acquisition of
spectra from electron energy loss spectroscopy (EELS). A
small camera length of 38 mm was selected for imaging and
elemental mapping in order to enhance the signal to noise
ratio of the EELS spectra. Maps were recorded by using a
beam current of 0.5 nA and a dwell time of 50 μs/pixel. The
Cu and Ga L-edges were selected to build the chemical maps.
풎ퟐ
푪풖 푺. 푨. ( ) =
품
ퟏퟎퟎ 풎풐풍 푯2 푺푭 푵푨
[
(
)( )( )]
( )
ퟏ
[( )( )]
푺푫푪풖 푾푪풖
Where mol H2 = amount of H2 consumed during the TPR step
per unit mass of the catalyst (mol H2/gcat), SF = stoichiometric
factor =2, NA = Avogadro’s number = 6.022 × 1023 atoms/mol,
SDCu = copper surface density =1.47 × 1019 atoms/m2, WCu
=
Cu concentration evaluated from elemental analysis (wt. %).
Copper dispersion is the surface copper to the total copper
concentration in the catalyst. Cu dispersion (%) was estimated
using equation 2.
%푫푪풖 = [ퟏퟎퟒ (풎풐풍 푯ퟐ )(푺푭)(푨풕. 푾풕.푪풖 )]/푾푪풖
)
(ퟐ)
Where At.Wt.Cu = Atomic weight of Cu = 63.54.
The surface acidity of different catalysts were assessed using
Temperature Programmed Desorption of NH3 (NH3-TPD). All
experiments were performed using the same Altamira
instrument described above. The catalyst sample (0.025g)
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
was placed in a U-shaped quartz tube and the whole system
was purged with Ar (25 mL/min @STP) for 15 minutes. Prior
to each NH3-TPD experiment, all catalyst samples were
heated under a flow of H2 (50 mL/min) up to 250 °C and then
kept at the same temperature for another 1h. After cooling to
50 °C, the catalyst was then exposed to a stream of 1%
NH3/He (25 mL/min @ STP) for 1h. Weakly (or “reversibly”)
adsorbed ammonia was subsequently removed by purging
the catalyst under He for ca. 1h. The purging step was
We are grateful for the financial support from King Abdullah
University of Science and Technology and a SABIC Chair
Professorship to K.-W.H.
Keywords: CO2 hydrogenation • Copper gallium •
Mesoporus silica • Bronsted acid sites • Oxygenates
[1]
[2]
A. Goeppert, M. Czaun, J.-P. Jones, G.K. Surya Prakash, G.A. Olah,
Chem. Soc. Rev., 2014, 7995-8048.
monitored by
a
Hiden HPR 20 Quadrupole Mass
G.A. Olah, G.K.S. Prakash, A. Goeppert, J. Am. Chem. Soc., 2011,
133, 12881-12898.
Spectrometer (QMS), and the amount of “reversibly” adsorbed
ammonia was generally observed to decrease with time to
finally reach negligible levels within 1 hour. Finally, TPD of
NH3 was carried out by heating the sample from 50 to 750 °C
at a rate of 10 °C/min under a flow of He. The ammonia
desorption was monitored by the QMS using the NH+ fragment
(m/e=15) of NH3. The response of the QMS was calibrated
using pulses of known volume of NH3.
[3]
J. Sun, G. Yang, Y. Yoneyama, N. Tsubaki, ACS Catal. 2014, 4,
3346-3356.
[4]
P. Cheung, A. Bhan, G.J. Sunley, E. Iglesia, Angew. Chem. Int. Ed.,
2006, 45, 1617-1620.
[5]
K. Faungnawakij, Y. Tanaka, N. Shimoda, T. Fukunaga, S.
Kawashima, R. Kikuchi, K. Eguchi, Appl. Catal. A, 2006, 304, 40-48.
X. San, Y. Zhang, W. Shen, N. Tsubaki, Energy Fuels, 2009, 23,
2843-2844.
[6]
[7]
A.A. Rownaghi, F. Rezaei, M. Stante, J. Hedlund, Appl. Catal. B,
2012, 119–120, 56-61.
The acid sites and strength of Cu-Ga catalytic system were
also investigated by FTIR using two different probe molecules
namely pyridine and 2,6-lutidine. Pyridine is commonly used
as a basic probe molecule with a pKb = 7.4 for the
determination of Lewis and Bronsted acid sites. It does not
interact with the weak acid sites due to the relatively lower
basicity. The use of 2,6-lutidine, with a pKb = 8.8), is thus used
to detect the weak and moderate Bronsted acid sites. Infrared
[8]
Y. Zhu, S. Wang, X. Ge, Q. Liu, Z. Luo, K. Cen, Fuel Process.
Technol., 2010, 91, 424-429.
[9]
L.A. Pellegrini, G. Soave, S. Gamba, S. Langè, Appl. Energy, 2011,
88, 4891-4897.
[10]
[11]
[12]
[13]
S.P. Naik, H. Du, H. Wan, V. Bui, J.D. Miller, W.W. Zmierczak, Ind.
Eng. Chem.Res., 2008, 47, 9791-9794.
J. Toyir, P.R.r. de la Piscina, J.L.G. Fierro, N.s. Homs, Appl. Catal.
B, 2001, 29, 207-215.
spectra were recorded with
a
Nicolet Magna FTIR
spectrometer (4 cm-1 resolution and 32 scans). The catalyst
samples were activated under vacuum (10 µbar) at 400 °C for
6 hours. The activated catalysts were then molded on a self-
supported wafer of 0.020g, prepared under a hydraulic pellet
machine with a thickness of 0.5 mm. The catalyst-molded
wafer was mounted on a Pyrex sample holder in an IR cell
followed by the introduction of 2,6-lutidine or pyridine over the
sample cell to allow the adsorption on the wafer under vacuum.
Both probe molecules were degassed by three cycles of
freeze-pump-thaw process. The catalyst wafer was evacuated
at 110 °C for 1.5 h to remove the physiosorbed probe
molecule after adsorption. The cell was then evacuated at
200 °C, 300 °C and 400 °C for 1.5 h, respectively, and the cell
was cooled down to room temperature and the spectra were
taken. Quantification of the acid sites was done by using the
modified Beer Lambert’s law (eqn. 3) using the literature
methods.[43-44]
J. Toyir, P. Ramırez de la Piscina, J.L.G. Fierro, N.s. Homs, Appl.
́
Catal. B, 2001, 34, 255-266.
J. Liu, J. Shi, D. He, Q. Zhang, X. Wu, Y. Liang, Q. Zhu, Appl. Catal.
A, 2001, 218, 113-119.
[14]
[15]
X.M. Liu, G.Q. Lu, Z. F. Yan, Appl. Catal. A, 2005, 279, 241-245.
J. Słoczyński, R. Grabowski, P. Olszewski, A. Kozłowska, J. Stoch,
M. Lachowska, J. Skrzypek, Appl. Catal. A, 2006, 310, 127-137.
R. Raudaskoski, M.V. Niemelä, R.L. Keiski, Top. Catal, 2007, 45,
57-60.
[16]
[17]
[18]
[19]
[20]
[21]
[22]
X. Guo, D. Mao, S. Wang, G. Wu, G. Lu, Catal. Commun., 2009, 10,
1661-1664.
C.V. Ovesen, B.S. Clausen, J. Schiøtz, P. Stoltze, H. Topsøe, J.K.
Nørskov, J. Catal., 1997, 168, 133-142.
S.I. Fujita, S. Moribe, Y. Kanamori, M. Kakudate, N. Takezawa, Appl.
Catal. A, 2001, 207 (1–2), 121-128.
L.R. Clausen, B. Elmegaard, J. Ahrenfeldt, U. Henriksen, Energy,
2011, 36, 5805-5814.
D. Mao, W. Yang, J. Xia, B. Zhang, Q. Song, Q. Chen, J. Catal.,
2005, 230, 140-149.
F.S. Ramos, A.M.D. Farias, L.E.P. Borges, J.L. Monteiro, M.A.
Fraga, E.F. Sousa-Aguiar, L.G. Appel, Catal. Today, 2005, 101, 39-
44.
푨푺
[23]
[24]
[25]
[26]
R.-w. Liu, Z.-z. Qin, H.-b. Ji, T.-m. Su, Ind. Eng. Chem. Res., 2013,
52, 16648-16655.
푪풔풘 =
(ퟑ)
풎풆
G. Yang, D. Wang, Y. Yoneyama, Y. Tan, N. Tsubaki, Chem.
Comm., 2012, 48, 1263-1265.
Where, Csw (mol/g) is the concentration of Lewis or Bronsted
acid sites in reference to a unit weight of the dry sample, A
(cm-1) is the integrated absorbance, S (cm2) is cross sectional
area of the sample wafer, m(g) is the weight of the dry sample
and e is the integrated molar extinction coefficient.
Z. Luan, M. Hartmann, D. Zhao, W. Zhou, L. Kevan, Chem. Mater.,
1999, 11, 1621-1627.
X. Dong, F. Li, N. Zhao, F. Xiao, J. Wang, Y. Tan, Appl. Catal. B,
2016, 191, 8-17.
[27]
[28]
P.H. Matter, D.J. Braden, U.S. Ozkan, J. Catal., 2004, 223, 340-351.
S. Natesakhawat, J.W. Lekse, J.P. Baltrus, P.R. Ohodnicki, B.H.
Howard, X. Deng, C. Matranga, ACS Catal., 2012, 2, 1667-1676.
Acknowledgements
This article is protected by copyright. All rights reserved.

10.1002/cctc.201701679
ChemCatChem
FULL PAPER
[29]
O. Oyola-Rivera, M.A. Baltanas, N. Cardona-Martinez, J. of CO2
[37]
[38]
M.M.J. Li, Z. Zeng, F. Liao, X. Hong, S.C.E. Tsang, J. Catal., 2015,
5, 3260-3270.
Util., 2015, 9, 8-15.
[30]
[31]
G. Connell, J.A. Dumesic, J. Catal., 1987, 105, 285-298.
A. Bansode, B. Tidona, P.R. von Rohr, A. Urakawa, Catal. Sci.
Technol., 2013, 3, 767-778.
G. Turnes Palomino, J. José Cuart Pascual, M. Rodrıguez Delgado,
́
J. Bernardo Parra, C. Otero Areán, Mater. Chem. and phys., 2004,
[39]
[40]
A. Bansode, A. Urakawa, J. Catal., 2014, 309, 66-70.
Z. Luan, J.A. Fournier, J.B. Wooten, D.E. Miser, Micropor. Mesopor.
Mater., 2005, 83, 150-158.
85, 145-150.
[32]
[33]
[34]
L. Zhang, Y. Zhang, S. Chen, Appl. Catal. A, 2012, 415–416, 118-
123.
[41]
[42]
[43]
[44]
K.V.R. Chary, G.V. Sagar, C.S. Srikanth, V.V. Rao, J. Phys. Chem.
B, 2007, 111, 543-550.
S.E. Collins, M.A. Baltanas and A.L. Bonivardi, J. Catal. 2004, 226,
410-421.
M. Balaraju, V. Rekha, P.S. Sai Prasad, R.B.N. Prasad, N. Lingaiah,
Catal. Lett., 2008, 126, 119-124.
C. O. Arean, M. R. Delgado, V. Montouillout, J. C. Lavalley, C.
Fernandez, J. J. C. Pascual, J. B. Parra, Micropor. Mesopor. Mater.,
2004, 67, 259–264.
N. Brodu, M.-H. Manero, C. Andriantsiferana, J.-S. Pic, H. Valdés,
Chem. Engine. Jour., 2013, 231, 281-286.
[35]
[36]
X. Guo, D. Mao, G. Lu, S. Wang, G. Wu, J. Catal., 2013, 140–141,
16-24.
T. Onfroy, G. Clet, M. Houalla, Micropor. Mesopor. Mater., 2005, 82,
99-104.
J. Schumann, M. Eichelbaum, T. Lunkenbein, N. Thomas, M.C.
Á lvarez Galván, R. Schlögl, M. Behrens, ACS Catal., 2010, 271,
178-185.
This article is protected by copyright. All rights reserved.