Effect of SiO2-coating of FeK/Al2O3 catalysts on their activity and selectivity for CO2 hydrogenation to hydrocarbons

Article abstract of DOI:10.1039/c3ra44485f

SiO₂-coated FeK/Al₂O₃ catalysts with different silica content were prepared and examined for the synthesis of hydrocarbons from CO₂ hydrogenation. It was found that SiO ₂ coating affects both the acti

Full text of DOI:10.1039/c3ra44485f

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Effect of SiO2-coating of FeK/Al2O3 catalysts on
their activity and selectivity for CO2 hydrogenation
to hydrocarbons†
Cite this: RSC Adv., 2014, 4, 8930
a
a
a
a
ab
*
*
Fanshu Ding, Anfeng Zhang, Min Liu, Xinwen Guo and Chunshan Song
SiO2-coated FeK/Al2O3 catalysts with different silica content were prepared and examined for the synthesis
of hydrocarbons from CO2 hydrogenation. It was found that SiO2 coating affects both the activity for CO2
conversion and the selectivity to higher hydrocarbons, depending on the loading level. The catalyst with 9
wt% SiO2 coating showed both high CO2 conversion (63%) and high selectivity toward C2+ hydrocarbons
(74%), while further higher coating led to suppression of catalyst activity. The analytical characterization of
SiO2-coated catalysts confirmed the interactions of SiO2–Fe and SiO2–Al2O3, which largely reduced the
metal–support interaction and decreased the reduction temperature of iron oxide. In addition, an
increase of adsorption capacity for CO was observed with desired SiO2 coating. It is likely that SiO2-
coating improved the catalyst hydrophobic property, thereby reducing the negative impact of water by-
product and facilitating the reverse water gas shift reaction and subsequent hydrogenation towards
higher CO2 conversion to higher hydrocarbons.
Received 22nd May 2013
Accepted 17th January 2014
DOI: 10.1039/c3ra44485f
www.rsc.org/advances
CO2 + H2 / CO + H2O RWGS
CO + 2H2 / (–CH2–) + H2O FTS
CO + H2O / CO2 + H2 WGS
(1)
(2)
(3)
Introduction
Carbon dioxide capture, storage and utilization technologies
(CCSU) are of great environmental importance in recent years
because of the global greenhouse effect which is related to the
rising level of CO2 in the atmosphere.1–4 Owing to the problem
of depleting fossil resources, in particular coal and petroleum,
the effort for developing sustainable energy technology in a
variety of directions is receiving global attention during the past
decade. From this point of view, various chemical approaches
for converting CO2, the most abundant carbon source available,
into value-added products have been attempted,5–8 and catalytic
hydrogenation of CO2 to important chemical products is a
potential approach.9
Many investigations have focused on the conversion of CO2
into hydrocarbons.10–14 CO2 hydrogenation over iron-based
catalyst has been shown to proceed via a two-step process, with
initial conversion of CO2 to CO via the reverse water gas shi
reaction (RWGS) followed by CO hydrogenation into hydrocar-
bons as in Fischer–Tropsch synthesis reaction (FTS) as
follows:15,16
The synthesis of value-added hydrocarbons from CO2 and
CO hydrogenation reactions has something in common; the
adsorption and activation of CO are of great importance.
However, due to principle of microscopic reversibility, the
intrinsic WGS and RWGS activity of iron-based catalyst should
be taken into consideration. For instance, adsorption and
activation of CO2 behave differently in CO and CO2 hydroge-
nation reactions, higher CO2 conversion is facilitated in the
former reaction, while more CO2 is obtained as main by-product
in sole FTS reaction. In order to enhance the dissociative
adsorption of CO2, alkali promoter can be added. Potassium is a
signicant promoter in CO2 hydrogenation reactions for
enhancing the CO2 adsorption and selectivity towards olens,
leading to a signicant increase in CO2 conversion.17 Riedel
et al. reported a calculated equilibrium conversion of 72% at
573 K over a potassium-promoted iron catalyst, with direct
hydrogenation of CO2 merely playing a minor role.16 Moreover,
Huo et al. demonstrated that potassium promoter (K2O) could
stabilize the high-index, more active facets, notably Fe (211) and
change the relative rate of crystal growth in different directions
to form small particles with a large percentage of more active
facets.18 To improve the CO2 hydrogenation performance, the
addition of structural promoters or support into iron-based
catalyst is benecial to suppress the agglomeration of active
aState Key Laboratory of Fine Chemicals, PSU-DUT Joint Centre for Energy Research,
School of Chemical Engineering, Dalian University of Technology, Dalian, P.R.
China. E-mail: guoxw@dlut.edu.cn; Fax: +86 0411 84986134; Tel: +86 0411 84986134
bDepartment of Energy and Mineral Engineering, EMS Energy Institute, PSU-DUT Joint
Centre for Energy Research, Pennsylvania State University, University Park, 16802,
USA. E-mail: csong@psu.edu; Fax: +1-814-863-4466; Tel: +1-814-865-3573
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c3ra44485f
8930 | RSC Adv., 2014, 4, 8930–8938
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
phase and to improve the mechanical properties of the cata-
lysts. Investigation of various modied iron catalysts showed
alumina to be the best support for CO2 hydrogenation, followed
by silica and titania.19 In spite of improvement of catalyst
stability and stabilization of active sites of iron catalyst, it was
found that the strong Fe–Al2O3 interaction inhibits the reduc-
tion and carbonization of iron catalyst, nally the FTS activity
decreased.20 Furthermore, although water has been identied
as an effective agent in removing carbon deposits from metal
surface, it is worth noting that the water generated during the
reaction also suppresses the reaction rate of RWGS and FTS. It is
postulated that an iron-based catalyst that facilitates the RWGS
reaction can also catalyse the reverse reaction, thus inhibition
of reaction rate by water could be more signicant in CO2
hydrogenation than in FTS reaction using syngas, since the
intermediate product CO could react with H2O for CO shi to
CO2, resulting in a decrease of CO2 conversion. For the meth-
anol synthesis from CO2 hydrogenation, Farsi and Jahanmiri
CO2 hydrogenation reaction
The catalytic hydrogenation of carbon dioxide was carried out in
a pressurized xed-bed ow reactor (inner diameter 8 mm) where
a weighed 1 g catalyst (20–40 mesh) was loaded for each test.
Prior to the reaction, the catalyst was pre-treated by reduction
with pure H2 at 673 K overnight. Aer the reduction, the feed gas
was changed to the mixture of carbon dioxide and hydrogen
under the reaction conditions of n(H2)/n(CO2) ¼ 3 (molar ratio); P
¼ 3 MPa; T ¼ 673 K and the space velocity was 1800 ml gꢀ1 hꢀ1
.
The products were analysed on-line by a gas chromatograph
(FULI GC 97). Carbon monoxide, carbon dioxide and methane
were analysed on a carbon molecular sieve column with a
thermal conductivity detector (TCD) while methane and C2–C8
hydrocarbons (C2+) were analysed with a ame ionization
detector (FID) with a HayeSep Q column. Chromatograms of
FID and TCD were correlated through methane and product
selectivity was obtained based on carbon. The conversion
percentage of CO2 was based on the fraction of CO2 that formed
carbon-containing products according to:
proposed
a
water vapor–permselective alumina–silica
composite membrane reactor to overcome the thermodynamic
equilibrium limitations and increasing catalyst activity.21
Although many approaches have been reported to improve
catalyst activity and C2+ hydrocarbons selectivity, little is known
about how surface silica coating affects the catalysts for CO2
hydrogenation to hydrocarbons.22 In the present work, a series
of SiO2-coated FeK/Al2O3 catalysts with different SiO2 contents
were prepared and examined for CO2 hydrogenation to hydro-
carbons. Catalyst activity and product selectivities were evalu-
ated in a x-bed reactor. The catalysts were characterized by N2
adsorption, X-ray diffraction (XRD), Fourier transform infrared
spectroscopy (FT-IR), H2 temperature-programmed reduction
(TPR), temperature-programmed desorption (TPD) of CO2, H2
and CO, and thermal gravimetric (TG) analysis.
P
niMi
CO2conversionð%Þ ¼
ꢁ 100
(4)
MCO
2
where ni is the number of carbon atoms in product i, Mi is the
percentage of product i and MCO is the percentage of carbon
CO2 in the mixed feed. The selectivity (S) to product i, is based
on the total number of carbon atoms in the product and is
therefore dened as:
2
niMi
P
Sið%Þ ¼
ꢁ 100
(5)
niMi
Catalyst characterization
The textural properties of the samples were determined by N2
adsorption on a Quantachrome AUTO-SORB-1-MP sorption
analyser at liquid nitrogen temperature. Prior to the measure-
ments, the samples were degassed at 623 K for 2 h. The specic
surface area was determined by the Brunauer–Emmett–Teller
(BET) method. The total pore volume was obtained from the
amount of vapour adsorbed at a relative pressure (P/P0) close to
unity, where P and P0 are the measured and equilibrium pres-
sure respectively. The pore size distribution was evaluated using
the Barrett–Joyner–Halenda (BJH) method.
Experimental
Preparation of Al2O3-supported catalysts
g-Al2O3 was used as the support material, which was obtained
aer calcining pseudo-boehmite in air at 773 K for 4 h. The
FeK/Al2O3 catalyst was prepared by the incipient wetness
impregnation method using aqueous solutions of iron and
potassium nitrates (Fe(NO3)3$9H2O and KNO3), to obtain 15 wt
% of Fe and 10 wt% of K loading, respectively. The fresh
catalyst (denoted as 0Si) was obtained aer drying at 393 K for
12 h followed by calcination in air at 773 K for 4 h to decom-
pose iron and potassium nitrates, for which the heating rate
XRD patterns of fresh and used catalysts were determined on
a Rigaku D/Max 2400 diffractometer with Cu Ka radiation (l ¼
ꢂ
ꢂ
˚
1.5406 A) source. The spectra were recorded from 2q ¼ 5 to 80
with 0.02ꢂ step size. The crystallite phases were identied by
comparing the diffraction patterns with the data of the Joint
Committee on Powder Diffraction Standards (JCPDS).
was 2 K minꢀ1
.
Preparation of the SiO2-coated catalysts
The FT-IR spectra were recorded with an EQUINOX55
A series of SiO2-coated catalysts were prepared by impregnating (Bruker) Fourier transform infrared spectrometer by means of
the calcined 0Si catalyst with tetraethyl orthosilicate (TEOS) the KBr pellet technique. All spectra were taken in the range
dissolved in cyclohexane at room temperature, dried at 393 K 4000–400 cmꢀ1 at a resolution of 4 cmꢀ1. The spectra of all
and then calcined at 773 K for 4 h. Calcination turned TEOS into samples were presented by subtracting the background spec-
SiO2. SiO2-coating contents of the 0Si catalyst were varied trum. The micro-IR imaging measurements were carried out
between 1 and 11 wt% and the corresponding samples were with the OMNIC Picta soware on a Nicolet iN 10 MX Infrared
denoted as xSi (x ¼ 1, 3, 5, 7, 9, 11).
Microscope (Thermo Scientic).
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 8930–8938 | 8931

View Article Online
RSC Advances
Paper
H2-TPR measurements were carried out with ChemBET
Pulsar TPR/TPD equipment (Quantachrome, USA) to analyse the
reducibility of the calcined catalysts. Prior to the reduction, the
calcined sample (0.10 g) was placed in a quartz tube reactor in
the interior of a controlled oven. The sample was ushed with
high purity argon at 573 K for 1 h to remove water and other
contaminants then cooled down to room temperature. A gas
mixture containing 5 vol% H2 in Ar was passed through the
sample at a total ow rate of 30 ml minꢀ1. The temperature and
detector signals were then continuously recorded while heating
at 10 K minꢀ1 up to 1173 K. A cooling trap was also placed
between the sample and the detector for removal of released
water formed during the reduction process.
H2-TPD measurements were conducted in the same equip-
ment as TPR with helium as the carrier gas. About 0.20 g of
catalyst was placed in the reactor. The sample was reduced with
5% H2/95% Ar (30 ml minꢀ1) at a temperature of 673 K for 2 h.
The catalyst was subsequently ushed at the same temperature
with helium gas (30 ml minꢀ1) for 30 min. Aer reduction, the
sample was cooled to 303 K, and then H2 ow (30 ml minꢀ1) was
continued for 30 min at 303 K. Aer adsorption, the system was
purged with helium gas (30 ml minꢀ1) for 30 min to remove
weakly adsorbed species. The H2-TPD prole was monitored
using TCD while the temperature was increased from 303 K to
1173 K at a rate of 10 K minꢀ1. CO2-TPD and CO-TPD
measurements were conducted the same as H2-TPD, and the
adsorption gas was pure CO2 and CO, respectively.
TG analysis was conducted on a TGA/SDTA851e thermo-
balance (Mettler Toledo). The thermal analysis data were
collected in the range of 308–1173 K in an air ow. The heating
rate was 10 K minꢀ1 at a ow rate of 25 ml minꢀ1 while the
sample weight was between 7 and 10 mg. TG analysis was used
to measure the weight change of the catalysts aer reaction.
Fig. 1 (a) Catalyst activity, product selectivity of C1 and C2+ with
0–11% SiO2-coating contents. (b) C2+ hydrocarbon distribution of 0Si
and 9Si catalysts. Reaction conditions: molar ratio of H2/CO2 ¼ 3/1,
GHSV ¼ 1800 ml gꢀ1
h
ꢀ1, P ¼ 3 MPa, T ¼ 673 K.
butylene and ethylene were the major products. It can be seen
that SiO2-coating could increase long chain hydrocarbons
selectivity signicantly, and for the 9% coating content catalyst,
C5+ hydrocarbons selectivity increased from 21.9% to 28.0%.
Results
Catalytic hydrogenation of CO2 into hydrocarbons
Textural properties
Fig. 1(a) presents the CO2 hydrogenation performances of the
FeK/Al2O3 catalysts with different SiO2-coating contents aer 8 Table 1 gives the BET surface area, pore volume and pore size of
h (solid line) on stream. Upon initial coating of the catalyst with the support and catalysts. BET surface area of the 0Si catalyst
1 wt% SiO2, CO2 conversion increased, along with a decrease of was about 138 m2 gꢀ1; SiO2-coating rst slightly increased the
CO selectivity. The catalysts with more SiO2-coating at 3 and 5 BET surface area of catalysts until the 7Si sample, while the
wt% behaved almost the same and they both give higher CO2 surface area increased slightly from 138 to 144 m2 gꢀ1, indi-
conversion than with 1Si, which is accompanied by a slight cating an effective deposition of silica either on the surface or in
increase of C2+ hydrocarbons selectivity. Upon further the pores of the catalysts. Fig. 2 shows the N2 adsorption–
increasing the SiO2 coating up to 7 and 9 wt%, the conversion of desorption isotherms of the 0Si, 7Si and 9Si samples (isotherms
CO2 increased further, together with the additional rise in of the 1–7Si samples are analogous, while 9Si and 11Si samples
selectivity of C2+ hydrocarbons. Moreover, selectivity of by- exhibit virtually the same, see ESI Fig. S1†). All samples
product CO decreased signicantly in the case of 7Si and 9Si exhibited typical type IV isotherms with hysteresis loop,
catalyst, while selectivity of methane remained almost the according to the IUPAC classication; the isotherm of the
same. However, further increasing SiO2-coating content from 9 catalyst coated with 7% SiO2 was almost the same as the 0Si
to 11 wt% suppressed the catalyst activity as well as the C2+ catalyst. However, a signicant decrease in the surface area to
hydrocarbons selectivity. Based on these observations 9 wt% 130 m2 gꢀ1 was observed, along with a decrease in the pore
was chosen for more study on SiO2-coating of FeK/Al2O3 cata- volume (see Table 1 and Fig. 2) when SiO2-coating content rose
lysts for CO2 hydrogenation.
to 9% or more. This apparent decrease in the surface area and
C2+ hydrocarbons selectivities over 0Si and 9Si catalysts are pore volume might be due to either the narrowing of the pores
shown in Fig. 1(b). Among all the C2+ hydrocarbons, propylene, or partial blockage of pore entrances by SiO2 coating at 9 wt% or
8932 | RSC Adv., 2014, 4, 8930–8938
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Table 1 Physical properties of support and catalysts as prepared
Theoretical content (g MxOy per g cat.)
BET surface area
Pore volume
Average pore
size (nm)
Samples
(m2 gꢀ1
)
(cm3 gꢀ1
)
SiO2
Fe2O3
K2O
Al2O3
g-Al2O3
283
138
137
140
142
143
144
130
130
0.42
0.21
0.21
0.21
0.21
0.21
0.21
0.19
0.19
4.3
3.9
3.8
3.1
3.4
3.4
3.4
3.4
3.4
—
—
—
0.01
0.03
0.05
0.07
0.09
0.11
—
—
1.000
0.750
0.750
0.743
0.728
0.713
0.698
0.683
0.668
0Si
0.160
0.160
0.158
0.155
0.152
0.149
0.146
0.142
0.090
0.090
0.089
0.087
0.086
0.084
0.082
0.080
0Si (cyclohexane)a
1Si
3Si
5Si
7Si
9Si
11Si
a
The 0Si (cyclohexane) sample was prepared by impregnating the 0Si sample in cyclohexane without TEOS.
reduction peak temperature (Fe2O3 / Fe3O4), corresponding
H2 consumption and relative degrees of reduction compared
with the 0Si sample are listed in Table 2. Gao et al. concluded
from their studies on Fe2O3/g-Al2O3 by TPR and XRD, FeO was
metastable at temperature below 840 K, which compared with
either Fe3O4 or a-Fe.23 Thus FeO could hardly be detected
during the reduction of Fe2O3 to Fe. As shown in Fig. 3 and
Table 2, coating the alumina supported catalysts with SiO2 had
complex effects on the reduction behavior of iron-based cata-
lysts. Coating a small amount of silica (1–3 wt%) both
decreased the reduction temperature and inhibited the
reduction degree of iron oxides. Aer increasing SiO2-coating
content to 5–7 wt%, reduction temperature of iron oxide
changed slightly, while reduction degree of iron oxide
increased from 71% to 78%. In the case of 9Si sample, the
reduction temperature dropped to 754 K, with a minimum
reduction degree. Further increasing the coating content to
11 wt% increased both the reduction temperature and reduc-
tion degree compared to the 9Si catalyst.
Fig. 2 N2 adsorption–desorption isotherm of the 0Si, 7Si and 9Si
catalysts.
higher, while the depositions of 1–7 wt% SiO2 were largely
dispersed on the surface inside and outside the pores.
Reduction behaviour of catalysts
Crystallite structure of catalysts
The reduction behavior of catalysts with different coating
content was studied by H2-TPR. As shown in Fig. 3, the rst
On the basis of catalyst activity in Fig. 1, 9 wt% is the optimal
coating level for CO2 hydrogenation to hydrocarbons. Among
all coated catalysts, BET and TPR results also indicate a
Table 2 Temperature of reduction peak and reduction degree of the
catalysts
Peak
H2 consumption Fe content Reduction
Samples maximum (K) degreea (%)
(%)
degreeb (%)
0Si
1Si
3Si
5Si
7Si
9Si
11Si
789
773
772
772
776
754
762
100.0
79.9
69.4
74.8
73.1
63.3
66.7
100.0
99.0
97.0
95.0
93.0
91.0
89.0
100
80.8
71.1
78.9
78.5
69.2
75.3
a
H2 consumption degree of the catalysts was obtained by comparing total
b
Fig. 3 H2 TPR profiles of catalysts with different levels of SiO2 coating
H2 consumption of 0Si catalyst. Reduction degree was obtained by
regulating H2 consumption degree with 100% Fe content.
on FeK/Al2O3.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 8930–8938 | 8933

View Article Online
RSC Advances
Paper
unique character of the 9Si sample. XRD was used to charac-
terize the iron-phase and aluminium-phase structures of the
catalysts. XRD patterns of the 0Si and 9Si catalysts as-prepared
and aer the reaction are presented in Fig. 4. For the fresh
catalysts, the XRD patterns of 0Si catalyst exhibited the
characteristic diffraction peaks of hematite (a-Fe2O3) and
alumina (g-Al2O3) while the patterns of SiO2-coated catalyst
changed slightly. Aer the CO2 hydrogenation reaction, XRD
patterns of both used catalysts exhibited the characteristic
peaks of magnetite (Fe3O4) and carbide (c-Fe5C2), and c-Fe5C2
is an active phase during the CO and CO2 hydrogenation.
Additionally, the SiO2-coated catalyst exhibited the charac-
teristic peaks of KFeSi3O8 (JCPDS 14-0617) and KAlSiO4
(JCPDS 11-0579) at 2q ¼ 27ꢂ and 28.6ꢂ respectively, which were
quite different from 0Si sample.
FT-IR
FT-IR spectra of the 0Si and 9Si samples before and aer the
reaction are shown in Fig. 5. In the spectrum of the fresh 0Si
catalyst, two broad bands around 780 and 540 cmꢀ1 were
observed, both of which are characteristic bands of Fe–O in the
a-Fe2O3 phase. With SiO2-coating, new bands appeared at 1100,
1000 cmꢀ1, which can be attributed to the asymmetric Si–O–Al
and Si–O–Fe stretching vibration respectively (base on IR
spectra in ESI Fig. S2†). Aer the CO2 hydrogenation, 0Si sample
exhibited a weak vibration band at 3000 cmꢀ1, assigned to the
surface carbon, while the 9Si catalyst showed essentially no
peak here. Meanwhile, for the 9Si catalyst, bands at 1100 and
1000 cmꢀ1 became stronger aer the CO2 hydrogenation, which
is apparently different from the 0Si catalyst.
Fig. 5 (a) The FT-IR spectra of the fresh 0Si and 9Si catalysts. (b) The
FT-IR spectra of the 0Si and 9Si catalysts after reaction.
Chemical images of the used catalysts (Fig. 6) were recorded
on a Nicolet iN10 MX Infrared Imaging Microscope, the colour
intensity from blue to red is proportional to the content of the
target molecule, according to the asymmetric vibration at
different wavenumbers. Micro-IR spectra were also obtained, as
shown in Fig. 6. SiO2-coated catalyst displayed red at 1000 cmꢀ1
in Fig. 6(d) corresponded to line 3 in Fig. 7(b), conrming the
formation of Si–O–Fe structure, while line 4 and 5 showed peaks
at 1400 and 1550 cmꢀ1 corresponding to the same structure of
line 1 and 2 in Fig. 7(a).
H2, CO2 and CO chemisorption
TPD measurement was used to investigate the effect of SiO2-
coating on the adsorption behaviour of the iron-based catalyst.
In Fig. 8(a), H2 desorption behaviour over SiO2-coated catalyst
was basically the same as the 0Si sample below 800 K. When
temperature was higher than 800 K, g-Al2O3 began to transform
into q-Al2O3 and a-Al2O3, which is accompanied by some weight
loss. Weight loss over the SiO2-coated catalyst was lower
compared to the 0Si catalyst, indicating less g-Al2O3 trans-
formations. Similar weight loss over 800 K could also be
observed in CO2 and H2-TPD proles as follows.
CO2-TPD proles of the catalysts are shown in Fig. 8(b).
The CO2 uptakes could be used to indicate the intensity of
the surface basicity. In general, there are three desorption
peaks in CO2-TPD proles. One peak at the lower temperatures
(350–400 K) corresponds to weak CO2 adsorption, while the
other two at higher temperatures (400–550 K, 550–770 K)
can be ascribed to the adsorption of CO2 that interacted
moderately with the surface basic site. It was found that
peak areas of 400–550 K and 550–770 K decreased aer coating
the catalyst with SiO2. Coating the catalyst with SiO2 resulted
in the decrease of iron, potassium and aluminium contents
over the entire catalyst (Table 1), while minor potassium
result in the decrease of catalyst surface basicity. Therefore
the SiO2-coated sample exhibited weaker CO2 adsorption
capacity.
Fig. 4 XRD patterns of fresh and used 0Si and 9Si catalysts. (C):
g-Al2O3; (A): a-Fe2O3; (:): Fe3O4; (;): c-Fe5C2; (P): KFeSi3O8; (O):
KAlSiO4.
8934 | RSC Adv., 2014, 4, 8930–8938
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
Fig. 6 (a–c) Micro-IR mappings of the 0Si and 9Si catalysts after reaction. (a) 0Si, 1000 cmꢀ1; (b): 0Si, 1400 cmꢀ1; (c): 0Si, 1550 cmꢀ1. (d–f) Micro-
IR mappings of the 0Si and 9Si catalysts after reaction. (d) 9Si, 1000 cmꢀ1; (e) 9Si, 1400 cmꢀ1; (f) 9Si, 1550 cmꢀ1
.
Fig. 8(c) presents the CO desorption behavior of the pre- subsequent weight loss around 573 K was attributed to the
reduced catalysts. Differing from the 0Si catalyst, a minor peak removal of crystal water. Weight loss for the 0Si sample was
appeared at 700–800 K in the CO-TPD prole of 9Si sample. obviously larger than that for the SiO2-coated below 673 K. The
Desorption temperature of molecular CO on Fe single-crystal subsequent gradual weight loss above 723 K could be attributed
planes were below 473 K, while the dissociative CO desorption to the decomposition of carbon deposit formed during the CO2
occurred at about 770 K. Therefore, adsorption capability of by- hydrogenation reaction; the SiO2-coated catalyst exhibited less
product CO corresponding to desorption temperature of 700– carbon deposit. Crystal change of alumina resulted in the weight
800 K was enhanced by SiO2 coating treatment.
loss above 800 K, which has been detected in TPD measurement.
TG analysis
Discussion
The typical TG/DTG curves of the used 0Si and 9Si catalysts are
shown in Fig. 9. Both samples displayed a weight loss around 373 Iron oxides have tested used as CO2 hydrogenation catalysts for
K due to the removal of physically adsorbed water. The many years.17,24,25 Dorner et al. prepared a K and Mn doped iron-
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 8930–8938 | 8935

View Article Online
RSC Advances
Paper
Fig. 7 Micro-IR spectra of the used 0Si and 9Si catalysts. (a) 0Si
catalysts. (b) 9Si catalyst.
based CO2 hydrogenation catalysts, the KAlH4 phase was
detected as part of the catalyst's active phase, and K8Mn12Fe17/
Al2O3 catalyst performed highest CO2 conversion (41.4%) and
C2–C5+ hydrocarbons selectivity (62.4%).17 Effect of the type of
zeolites on catalyst activity for synthesis of iso-alkanes from CO2
hydrogenation were investigated over Fe–Zn–Zr/zeolites, and
HY zeolite was the most effective one for iso-alkane synthesis
due to the presence of medium and strong acid sites.24 C2–C5+
olen production from CO2 hydrogenation was investigated
using ceria modied FeMnK catalysts, the results showed that
the larger ceria particle catalyst (FeMnKCeAl800) performed the
best, yielding a CO2 conversion level of 50.4%.25
Fig. 8 Desorption of probe gases from 0Si and 9Si catalysts. (a) H2-
TPD. (b) CO2-TPD. (c) CO-TPD.
Hydrophobicity has been considered a crucial property of
coating materials especially due to the recent demand for high
temperature performance in various applications, and the SiOx–
DLC (diamond-like carbon) lm from the plasma polymeriza-
tion of HMDSO was chosen to increase the surface hydropho-
bicity due to its low surface energy compared with a pure DLC
lm.26 Xu et al. prepared superhydrophobic SiO2 lms from
nano-husk ash by sol–gel method using hydroxy silicone oil
(HSO), hexamethyldisilazane (HMS), or methyl triethoxysilane
(MTS) as modiers, the results showed that the contact angle of
SiO2 lms was more than 160ꢂ when volume ratio of the
modiers to silicon-sodium solution (SSS) was 0.15.27 TEOS is
widely used as the Si source to alternate the material hydro-
phobic property. Jeong, Goo and Kim prepared a hydrophobic
Fig. 9 TG and DTG curves of the used catalysts: (solid line) 0Si;
SiO2 powder by surface modication of TEOS wet gel with 6 and (broken line) 9Si.
8936 | RSC Adv., 2014, 4, 8930–8938
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
RSC Advances
12 vol% of TMCS (trimethylchlorosilane).28 Mokhtarimehr et al. acceptor in CO structure; the dissociative adsorption of CO was
reported a vanadium doping TiO2–SiO2 thin lms exhibited easier on Fe species with an electron-decient state. Due to the
signicant effect on the hydrophilicity.29 Ghazzal modulated the formation of Si–O–Fe by SiO2-coating, the coated catalyst facil-
relative content of TEOS and (methyltriethoxysilane) MTES itates the CO adsorption, which is proved by CO-TPD
mixtures in order to modify the hydrophilic–hydrophobic measurement in Fig. 8(c).
balance of the mesoporous SiO2 layer.30
Apart from the CO2 conversion and product distribution,
Since the synthesis of hydrocarbons by CO2 hydrogenation SiO2-coating content also affects catalyst physical properties
over iron-based catalyst can be regarded as a combination of and reduction behavior. BET surface area in Table 1 presents a
RWGS and F–T synthesis reactions, it is of great importance to procedure of the transformation of SiO2-coating with different
enhance the activation of both CO2 and CO. Moreover, by- SiO2 content. Initially coating of the Al2O3-supported catalyst
product H2O (from CO2 hydrogenation) could inhibit CO2 increases the surface area from 138 to 144 m2 gꢀ1; however,
adsorption on surface, and decrease catalyst activity as well as when the coating content is 9% or higher, the surface area
C2+ hydrocarbons selectivity. For this reason reducing or elim- decreases to 130 m2 gꢀ1, accompanied by the reduced pore
inating the inhibition of water is of great importance, either in volume. It is clear that coating content can affect the catalyst
increasing the relative concentration of reactants and reaction physical properties. SiO2 stays as dispersed tiny particles at SiO2
intermediate or in overcoming thermodynamic constrains. content up to 9 wt%, while further increasing coating level
Improvement of the catalyst hydrophobic property by SiO2- enlarges the SiO2 particles, resulting in the partial blockage of
coating process has been proved by the characterization results. alumina pore entrances. Iron-oxide reduction temperature is
The formation of Si–O–Fe and Si–O–Al structures by SiO2- also affected by SiO2-coating, due to the weakened Fe–Al2O3
coating treatment can be found in XRD (Fig. 4) and FT-IR (Fig. 5) interaction. Adding a small amount of SiO2 (1–3 wt%) decreases
measurements. Moreover, TPD and TG results indicate a reduction temperature, but also inhibits the reduction of iron
decreasing Al–O–Al amount over the SiO2-coated catalyst, oxide. Metal–support interaction is weakened by silica inter-
according to minor weight loss over 800 K by Al2O3 crystal phase acting with iron and alumina, leading to a decline of reaction
transformation. SiO2-coating treatment alters the catalyst temperature. Such a small amount of SiO2 is not strong enough
hydrophobicity by decreasing content of Al–O–H and increasing to interact directly with iron and alumina to form Si–O–Fe and
Si–O–Al content. Removing water generated during the reaction Si–O–Al species, and the inhibiting effect of reduction degree is
promotes CO2 and CO hydrogenation to hydrocarbons, and this due to electron-decient state of Fe-species in silica-promoted
consideration is supported by catalyst activity data in Fig. 1; catalysts compared to the silica-free catalyst,31 Fe–O bonds are
coating the catalyst with SiO2 signicantly increases the CO2 strengthened and become difficult to cleave during the reduc-
conversion from 49% to 63% when SiO2 content is 9%, and tion. Upon increasing coating content to 5–7 wt%, reduction
selectivity toward C2+ hydrocarbons rises from 64% to 72%. temperature changes slightly while reduction degree increases
Since CO is the intermediate product in CO2 hydrogenation, from 71–78%. The excess silica interacts stronger with its
concentration of CO is determined by the relative rates of neighbouring Al atoms and iron oxides, metal–support inter-
RWGS, WGS and FTS reaction. However, directly reducing action become even weaker. Si–O–Fe and Si–O–Al take shape
content of CO could not be obtained by promoting RWGS and with about 9 wt% coating content, and SiO2 coating at 9 wt% or
FTS reactions simultaneously in ways of hydrophobicity higher level might narrowing the pores or become partial
improvement.
blockage of alumina pore entrances according to the BET
TPD results indicate that SiO2-coating alters the adsorption analysis; the suppression of metal–support interaction
of reactant CO2 and H2 and intermediate reactant CO. Too decreases the reduction temperature to 754 K; the Si–O–Fe
strong a hydrogen adsorption and activation capacity would species are hardly reducible, corresponding to decline of
result in methanation rather than C–C coupling; H2-TPD reduction degree. With a further increase of coating content to
proles in Fig. 8(a) illustrates that there is hardly any inuences 11%, some iron oxide are covered by surface silica, and inter-
on chemisorption capability of hydrogen, rationalising no actions between Si and Fe lead to an increase in reduction
change of methane selectivity, as can be seen in Fig. 1(a). Al, Fe temperature. Moreover, in order to well disperse on SiO2, the
and K elemental contents decreased aer SiO2-coating, while K crystallite size of Fe2O3 decreases, causing a higher probability
plays the role of basic promoter of the catalyst. CO2-TPD proles of contact with H2 during the reduction. Catalyst activity is
in Fig. 8(b) shows that SiO2-coating largely reduces the closely related to the interaction strength between metal–
adsorption of CO2, owing to a lower K content. Since sufficient support and also silica–iron. Weaker metal–support interaction
CO2 exists in the system, the process of CO2 activation is more benets the iron oxide reduction, more active sites are provided
likely the rate-determining step rather than that of CO2 for the adsorption and activation of the reactants. Stronger
adsorption. The promotion of RWGS by SiO2-coating on FeK/ silica–iron interaction prots the adsorption of CO, and the
Al2O3 catalyst is an important aspect, which can explain at least process of F–T synthesis is promoted. When the coating content
in part why the SiO2-coated catalyst with a lower CO2 adsorption is 9 wt%, the metal–support interaction is the least, and Si–O–Fe
capability exhibits higher activity for CO2 conversion. In the Si– structure is sufficient for CO adsorption compared with other
O–Fe structure, there is electron transfer from Fe to Si atoms, samples, according to the lowest reduction temperature and
showing an electron-decient state of Fe species.31 Due to the degree in TPR results. When the coating content is lower than
triple bond between O and C, carbon atom is the electron-pair 9%, the promotion of CO adsorption is limited due to weak
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 8930–8938 | 8937

View Article Online
RSC Advances
Paper
silica–iron interaction; when the coating content is higher than 10 R. W. Dorner, D. R. Hardy, F. W. Williams and
9%, there are not enough active sites for CO2 adsorption and H. D. Willauer, Energy Environ. Sci., 2010, 3, 884.
activation due to the excessive Fe–O–Si species. The optimal 11 R. W. Dorner, D. R. Hardy, F. W. Williams and
coating content is around 9%. H. D. Willauer, Catal. Commun., 2010, 11, 816–819.
It is likely that SiO2-coating improved the catalyst hydro- 12 G. Weatherbee, J. Catal., 1984, 87, 352–362.
phobic property, thereby reducing the negative impact of water 13 E. Vesselli, J. Schweicher, A. Bundhoo, A. Frennet and
by-product and facilitating the reverse water gas shi and
N. Kruse, J. Phys. Chem. C, 2011, 115, 1255–1260.
subsequent hydrogenation coupled with C–C coupling towards 14 M. K. Gnanamani, W. D. Shafer, D. E. Sparks and B. H. Davis,
higher CO2 conversion to long chain hydrocarbons.
Catal. Commun., 2011, 12, 936–939.
15 S.-H. Kang, J. W. Bae, J.-Y. Cheon, Y.-J. Lee, K.-S. Ha,
K.-W. Jun, D.-H. Lee and B.-W. Kim, Appl. Catal., B, 2011,
103, 169–180.
Conclusions
In this study, SiO2-coating treatment was used to modify the 16 T. Riedel, G. Schaub, K.-W. Jun and K.-W. Lee, Ind. Eng.
FeK/Al2O3 catalyst for CO2 hydrogenation to hydrocarbons. Chem. Res., 2001, 40, 1355–1363.
Coating the catalyst with SiO2 at desired loading levels largely 17 R. W. Dorner, D. R. Hardy, F. W. Williams and
improved the synthesis of hydrocarbons by CO2 hydrogenation, H. D. Willauer, Appl. Catal., A, 2010, 373, 112–121.
likely due to improved catalyst hydrophobic property for 18 C. F. Huo, B. S. Wu, P. Gao, Y. Yang, Y. W. Li and H. Jiao,
reducing the negative impact of water by-product and for Angew. Chem., Int. Ed., 2011, 50, 7403–7406.
restraining WGS. SiO2-coating also reduced the Fe–Al2O3 19 T. Riedel, M. Claeys, H. Schulz, G. Schaub, S. S. Nam,
interaction, resulting in a decrease of iron oxide reduction
temperature, along with an enhancement of CO adsorption.
K. W. Jun, M. J. Choi, G. Kishan and K. W. Lee, Appl.
Catal., A, 1999, 186, 201–213.
Based on the improvements of CO2 activation and CO adsorp- 20 H. Wan, B. Wu, C. Zhang, H. Xiang, Y. Li, B. Xu and F. Yi,
tion capacity, the optimal coating content is around 9 wt%, Catal. Commun., 2007, 8, 1538–1545.
where the CO2 conversion was increased from 49% to 63%, and 21 M. Farsi and A. Jahanmiri, Ind. Eng. Chem., 2012, 18, 1088–
selectivity of C2+ hydrocarbons shied from 63% to 74%.
1095.
22 F. Ding, B. Zheng, C. Song and X. Guo, Prepr. Pap. – Am.
Chem. Soc., Div. Fuel Chem., 2012, 57, 444.
23 X. Gao, J. Shen, Y. Hsia and Y. Chen, J. Chem. Soc., Faraday
Trans., 1993, 89, 1079.
24 R. Bai, Y. Tan and Y. Han, Fuel Process. Technol., 2004, 86,
293–301.
Notes and references
1 X. Ma, X. Wang and C. Song, J. Am. Chem. Soc., 2009, 131,
5777–5783.
2 P. Markewitz, W. Kuckshinrichs, W. Leitner, J. Linssen,
¨
P. Zapp, R. Bongartz, A. Schreiber and T. E. Muller, Energy 25 R. W. Dorner, D. R. Hardy, F. W. Williams and
Environ. Sci., 2012, 5, 7281. H. D. Willauer, Catal. Commun., 2011, 15, 88–92.
3 D. Wang, X. Ma, C. Sentorun-Shalaby and C. Song, Ind. Eng. 26 S.-C. Cha, E. K. Her, T.-J. Ko, S. J. Kim, H. Roh, K.-R. Lee,
Chem. Res., 2012, 51, 3048–3057.
4 Z.-Z. Yang, L.-N. He, J. Gao, A.-H. Liu and B. Yu, Energy
Environ. Sci., 2012, 5, 6602.
5 T. Inui, T. Yamamoto, M. Inoue, H. Hara, T. Takeguchi and
J. B. Kim, Appl. Catal., A, 1999, 186, 395–406.
6 W. Wang, S. Wang, X. Ma and J. Gong, Chem. Soc. Rev., 2011,
40, 3703–3727.
7 X. Nie, M. R. Esopi, M. J. Janik and A. Asthagiri, Angew.
Chem., Int. Ed., 2013, 52, 2459–2462.
K. H. Oh and M.-W. Moon, J. Colloid Interface Sci., 2013,
391, 152–157.
27 K. Xu, Q. Sun, Y. Guo and S. Dong, Appl. Surf. Sci., 2013, 276,
796–801.
28 A. Y. Jeong, S. M. Koo and D. P. Kim, J. Sol-Gel Sci. Technol.,
2000, 19, 483–487.
29 M. Mokhtarimehr, M. Pakshir, A. Eshaghi and M. H. Shariat,
Thin Solid Films, 2013, 532, 123–126.
30 M. N. Ghazzal, M. Joseph, H. Kebaili, J. De Coninck and
E. M. Gaigneaux, J. Mater. Chem., 2012, 22, 22526.
31 H. Suo, S. Wang, C. Zhang, J. Xu, B. Wu, Y. Yang, H. Xiang
and Y.-W. Li, J. Catal., 2012, 286, 111–123.
8 D.-A. Yang, H.-Y. Cho, J. Kim, S.-T. Yang and W.-S. Ahn,
Energy Environ. Sci., 2012, 5, 6465.
9 C. Song, Catal. Today, 2006, 115, 2–32.
8938 | RSC Adv., 2014, 4, 8930–8938
This journal is © The Royal Society of Chemistry 2014