Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts

Article abstract of DOI:10.1016/j.cattod.2013.03.031

The hydrogenation of carbon dioxide to lower (C₂-C₄) olefins is an important reaction for the utilization of CO₂ as a carbon feedstock for the production of building-block chemicals. We found that an Fe/ZrO₂ cat

Full text of DOI:10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
Catalysis Today xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Synthesis of lower olefins by hydrogenation of carbon dioxide over
supported iron catalysts
∗
Jingjuan Wang, Zhenya You, Qinghong Zhang, Weiping Deng, Ye Wang
State Key Laboratory of Physical Chemistry of Solid Surfaces, National Engineering Laboratory for Green Chemical Productions of Alcohols, Ethers and
Esters, College of Chemistry and Chemical Engineering, Xiamen University, Xiamen 361005, China
a r t i c l e i n f o
a b s t r a c t
Article history:
The hydrogenation of carbon dioxide to lower (C –C ) olefins is an important reaction for the utilization of
2
4
Received 29 October 2012
Received in revised form 15 March 2013
Accepted 26 March 2013
Available online xxx
CO2 as a carbon feedstock for the production of building-block chemicals. We found that an Fe/ZrO2 cata-
lyst could catalyze the hydrogenation of CO2, but the main products were CH4 and lower (C2–C4) paraffins.
The modification of the Fe/ZrO2 catalyst by alkali metal ions except for Li significantly decreased the
+
selectivities to CH4 and lower paraffins and increased those to lower olefins and C5+ hydrocarbons, par-
+
+
+
ticularly C5+ olefins. The modification by Na , K , or Cs also increased the conversion of CO2. The best
Keywords:
+
performance for lower olefin synthesis was obtained over the K -modified Fe/ZrO2 catalyst with a proper
Carbon dioxide
Hydrogenation
Lower olefins
Fe/ZrO2 catalyst
Alkali metal ion promoter
+
K
content (0.5–1.0 wt%). Among several typical supports including SiO2, Al2O3, TiO2, ZrO2, mesoporous
carbon, and carbon nanotube, ZrO2 provided the highest selectivity and yield to lower olefins. Our char-
+
acterizations suggest that the modification by K accelerates the generation of ꢀ-Fe5C2 phase under the
reaction conditions. This together with the decreased hydrogenation ability in the presence of K+ has
been proposed to be responsible for the enhanced selectivity to lower olefins.
©
2013 Elsevier B.V. All rights reserved.
1
. Introduction
by steam-cracking of naphtha. Numerous studies have been con-
tributed to producing lower olefins through non-petroleum routes
such as the dehydrogenation of lower paraffins [15], the two-step
process from methane [16,17], and the conversion of synthesis gas,
which may be obtained from natural gas, coal or biomass, either
directly [18] or via methanol [19]. The production of lower olefins
The concerns on the emissions of CO , which is a major
2
anthropogenic greenhouse gas, have grown rapidly in recent years
because of the worsening global environment. This has driven
research activities in the capture and transformation of CO2 [1–5].
The diminishing of fossil resources has also increased the research
by the hydrogenation of CO , which is an exothermic reaction (Eq.
2
interests in the utilization of CO as an alternative carbon feedstock
(1)), would be highly desirable from the viewpoint of utilizing CO2
as a carbon feedstock for the production of building-block chemi-
cals.
2
for the production of chemicals and fuels [6–10]. CO2 possesses
advantages of being nontoxic, abundant, and renewable as a chem-
ical feedstock. However, the conversion of CO2 typically requires
external energy input such as photo or electric energy [11–13]
and/or high energy co-reactants such as H2 and reactive organic
compounds (e.g., epoxide) [7] because CO2 is a highly stable
molecule. Although there are only limited industrial processes
utilizing CO2 at this moment, many promising reactions for the
transformation of CO2 have been reported [6–10].
−1
nCO + 3nH2 → CnH2n + 2nH O ꢁrH573 K = −128 kJ mol
(1)
2
2
Iron-based catalysts have been studied for the hydrogenation of
CO2 to C2+ hydrocarbons [20–31], but only a few of these studies
have focused on the formation of lower olefins [24,25,31]. It is gen-
erally accepted that, instead of direct hydrogenation of CO , the
2
reverse water-gas shift reaction (Eq. (2)) proceeds over these cat-
alysts, followed by the hydrogenation of CO to hydrocarbons (Eq.
(3)) via the Fischer–Tropsch (FT) mechanism. Fe-based catalysts can
catalyze both the reverse water–gas shift reaction [32,33] and the
hydrogenation of CO to light olefins [18], and thus, are expected
Among various transformations of CO , the hydrogenation of
2
CO is a versatile route capable of producing various chemicals such
2
as methanol, higher alcohols, formic acid, CH , and C hydrocar-
4
2+
bons [10,14]. Lower (C –C ) olefins are important building-block
2
4
chemicals and are currently being produced mainly from petroleum
to show good performances for the hydrogenation of CO2 to C –C4
2
olefins.
CO + H2 → CO + H O ꢁrH573 K = 38 kJ mol 1
−
(2)
(3)
2
∗ Corresponding author. Tel.: +86 592 2186156; fax: +86 592 2183047.
E-mail address: wangye@xmu.edu.cn (Y. Wang).
nCO + 2nH2 → CnH2n + nH O ꢁrH573 K = −166 kJ mol 1
−
2
0
920-5861/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.03.031
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
2
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
However, the selectivity to C –C4 olefins is still low in the hydro-
by ThermoStar GSD 301 T2 mass spectrometer with the signal of
2
genation of CO2 over most of the reported catalysts. It can be
expected that both the support and the promoter may affect the
activity and the selectivity of Fe-based catalysts. However, lit-
tle information is available on how to increase the selectivity to
lower olefins. Undoubtedly, more fundamental studies are needed
to develop an efficient Fe-based catalyst for the selective produc-
m/e = 44.
2.3. Catalytic reaction
Catalytic reactions were performed on a high-pressure fixed-
bed flow reactor. The catalyst (typically 1.0 g) loaded in the reactor
was first reduced by H with a flow rate of 50 mL min at 673 K for
tion of C –C olefins from CO . This paper reports our recent studies
2
4
2
−1
2
on the effects of alkali metal ions on the hydrogenation of CO2 to
C2–C4 olefins over supported Fe catalysts.
5
h. After the catalyst was cooled down to 353 K, a reactant gas mix-
ture of (H + CO ) with a H /CO molar ratio of 3.0 and a flow rate
2
2
2
2
−
1
of 20 mL min , which contained 5% argon as an internal standard
for the calculation of CO2 conversions, was introduced to the reac-
tor. Then, the pressure and temperature were increased typically
to 2.0 MPa and 613 K, respectively. The products were analyzed by
online gas chromatography. The data at steady states, which were
obtained typically after 10 h of reaction, were used for discussion.
2
. Experimental
2.1. Catalyst preparation
Metal oxide supports including SiO , TiO , Al O , and ZrO
2
2
2
2
3
were purchased from Alfa Aesar (for ZrO ) or Sinopharm Chemical
2
Reagent Co. (for the other metal oxides). Carbon nanotubes (CNTs)
were synthesized by a method reported previously [34], and were
pretreated with a 68 wt% nitric acid at 393 K to remove the Ni cat-
alyst used for CNT synthesis, followed by washing and drying. The
supported Fe catalysts were prepared by the conventional impreg-
nation method. For example, for the preparation of the 10 wt%
Fe/ZrO2 catalyst, ZrO2 (4.5 g) was first added into an aqueous solu-
tion of Fe(NO3)3 (0.2 mol dm , 45 mL). The suspension was then
stirred for 8 h, followed by evaporation at 353 K to dryness. The
obtained powdery catalyst was further dried at 393 K for 12 h and
calcined in air at 773 K for 5 h. The alkali metal ion-modified sup-
ported Fe catalysts were prepared by a co-impregnation method
using a mixed aqueous solution containing certain amount of an
3
. Results and discussion
3.1. Catalytic behaviors of Fe-based catalysts for hydrogenation
of CO2
3.1.1. Effect of modification by alkali metal ions on catalytic
behaviors of Fe/ZrO catalysts
−3
2
It has been demonstrated that ZrO2 is a unique support for the
hydrogenation of CO to CH OH or CH when Cu, Ag or Ni is used as
the active metal [35]. However, few studies have used ZrO2 for the
hydrogenation of CO2 to hydrocarbons. Here, we first investigated
the catalytic behavior of the Fe/ZrO catalyst for the hydrogenation
of CO . The conversion of CO2 was 32% over the 10 wt% Fe/ZrO2
2
3
4
2
2
alkali metal salt (LiNO , NaNO , KNO , RbCl or CsCl) and Fe(NO )
3
3
3
3
3
under our reaction conditions (Table 1). The selectivity to CO was
with a similar procedure.
2
5% and that to hydrocarbons (CnHm) was 75% over this catalyst.
CH4 and C –C4 paraffins were the dominant products in hydrocar-
2
2.2. Catalyst characterization
bons. Thus, the Fe/ZrO catalyst without modification is not suitable
2
for the hydrogenation of CO2 to lower olefins.
X-ray diffraction (XRD) measurements were carried out on
We investigated the effect of the modification by alkali metal
a Panalytical X’pert Pro Super X-ray diffractometer with Cu-K␣
radiation (40 kV and 30 mA). For in situ XRD measurements, the
powdery sample was loaded into an XRK-900 cell, which was
directly attached with the X-ray diffractometer. The XRD pattern
ions on catalytic behaviors of the Fe/ZrO2 catalyst. Table 1 shows
+
that the addition of Li decreases the conversion of CO . The selec-
2
tivity to hydrocarbons also became lower while that to CO became
+
higher by the addition of Li to the Fe/ZrO catalyst. These indicate
2
for the fresh sample was first recorded. Then, a H gas flow with
that Li+ suppresses both the reverse water-gas shift reaction and
the hydrogenation of CO to hydrocarbons. The inhibiting roles of Li+
in the water-gas shift and FT reactions were reported in previous
2
−
1
a flow rate of 50 mL min was introduced into the XRK-900 cell
−
1
and the temperature was raised at a rate of 10 K min . When the
desired temperature was reached, the catalyst was kept at that tem-
perature for 5 min, and the XRD pattern was recorded. 25 min were
typically required for the collection of one pattern with 2ꢂ ranging
studies [36,37]. The distribution of hydrocarbons did not change
+
significantly after the modification by Li ; CH4 and C –C4 paraffins
2
were still the dominant products. This indicates that the presence
of Li+ in the Fe/ZrO2 does not affect the ability of catalyst for the
hydrogenation of CHx intermediates or olefins, which are believed
to be the primary products, but decreases the ability of catalyst
◦
◦
from 20 to 50 . Before the appearance of the characteristic peaks
◦
◦
ascribed to Fe species (32 –45 ), the catalyst had been kept at the
desired temperature for >15 min. The in situ XRD patterns under
the reactant gas mixture of (H + CO ) were also recorded at 618 K.
for the activation of CO and CO . On the other hand, the addition
2
2
2
For these measurements, the sample was first reduced in the in situ
of other alkali metal ions to the Fe/ZrO2 catalyst did not decrease
−
1
the conversion of CO . The modification by Na , K and Cs+ rather
+
+
XRK-900 cell with H2 gas flow with a rate of 50 mL min at 673 K.
After being cooled down to 618 K, the reactant gas mixture was
introduced to the in situ XRK-900 cell, and then the XRD pattern
was recorded after a certain time.
2
+
increased the conversion of CO . Particularly, K showed a signifi-
2
cant enhancing effect on the catalytic activity, while the promoting
effects of Rb+ and Cs+ were less significant on CO2 conversions.
This might be caused by the residual chlorine [38], since RbCl and
CsCl were used as the precursors. The selectivity to CO decreased
CO2 temperature-programmed desorption (CO -TPD) was per-
2
formed on a Micromeritics AutoChem 2920II instrument. Typically,
the sample loaded in a quartz reactor was first pretreated with
high-purity He at 623 K for 1 h. After the sample was cooled down
to 373 K, CO2 adsorption was performed by switching the He flow
to a CO –He (10 vol% CO ) gas flow and then keeping at 373 K for
+
+
+
+
from 25% to 15–21% after the modification by Na , K , Rb , or Cs .
Organic oxygenates were also formed with considerable selectiv-
ities (17–20%) in addition to hydrocarbons over these catalysts.
+
More significantly, the modification of the Fe/ZrO catalyst by Na ,
2
2
2
K , Rb or Cs+ remarkably changed the hydrocarbon distributions.
+
+
1
h. Then, the gas phase or the weakly adsorbed CO2 was purged
by high-purity He at the same temperature. CO -TPD was per-
The fractions of CH4 and C –C4 paraffins decreased significantly
2
2
formed in the He flow by raising the temperature to 1073 K at a
rate of 10 K min , and the desorbed CO2 molecules were detected
from 70% to 18–26% and from 29% to <10%, respectively. Simultane-
ously, the fractions of C –C olefins and C5+ hydrocarbons increased
−1
2
4
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
3
Table 1
a
Catalytic performances of alkali metal ion-modified 10 wt% Fe/ZrO2 catalysts for the hydrogenation of CO2
Alkali metalb
CO2 conv. (%)
Select.c (%)
CO
Hydrocarbon distributiond (%)
C2–4 yield (%)
CnHm
CH4
C2–4
C2–40
C5+
C5+0
Non
32
26
39
43
31
39
25
42
21
15
15
16
75
56
59
66
68
67
70
68
21
18
19
26
0.1
1.4
49
44
43
29
30
8.8
9.2
8.0
9.6
0.4
0.1
15
19
19
14
0.5
0.5
6.2
9.8
11
0
+
Li
0.20
11
13
9.1
11
+
Na
+
K
+
Rb
+
Cs
43
7.4
a
−1
Reaction conditions: W (catalyst) = 1.0 g, H2/CO2 = 3, T = 613 K, P = 2 MPa, F = 20 mL min , time on stream = 10 h.
The content of alkali metal ion was 1.0 wt%.
b
c
CnHm denotes hydrocarbons, and the other products were organic oxygenates.
d
C2–4 : C2–C4 olefins; C2–40: C2–C4 paraffins; C5+ : C5+ olefins; C5+0: C5+ paraffins.
+
+
dramatically from 0.1% to >40% and from 0.5% to >20%, respectively.
The ratios of olefins to paraffins in C –C hydrocarbons over these
3.1.3. Effect of K content on catalytic behaviors of K -modified
Fe/ZrO catalysts
2
4
2
modified catalysts were >4.4. Olefins also dominated the C5+ hydro-
carbons, and the ratio of olefins to paraffinsin C5+ hydrocarbonswas
Although an appropriate support is important for obtaining
higher activity and higher C –C4 olefin selectivity, the presence of
2
+
+
+
+
around 2 over the Na , K , Rb or Cs -modified Fe/ZrO2 catalyst.
These observations allow us to consider that the modification by
an alkali metal ion plays an essential role in the hydrogenation of
+
CO to C –C olefins. To gain further insights into the function of K ,
2
2
4
+
+
alkali metal ions except for Li can decrease the hydrogenation abil-
we investigated the effect of K content on catalytic performances
of the K -Fe/ZrO catalysts. Fig. 1A shows that the modification by
+
ity and increase the chain-growth probability. Among these alkali
2
+
+
metal ion-modified catalysts, the K -Fe/ZrO2 exhibited the highest
K with a proper content (0.1–5.0 wt%) can increase the conversion
+
yield to C –C olefins (13%).
of CO . When the K content exceeded 5.0 wt%, the conversion of
2
4
2
CO decreased again. The selectivities to CO and hydrocarbons both
2
+
decreased slightly with increasing K content to 0.5 wt% (Fig. 1B).
+
At the same time, the selectivity to oxygenates increased from
3
.1.2. Effect of support on catalytic behaviors of K -modified Fe
0
.4% to ∼20%. Inside the hydrocarbons, the fractions of CH and
4
catalysts
C –C paraffins decreased steeply from 70% and 29% to ∼20% and
2
4
We investigated the effect of support on catalytic performances
of the K -modified Fe catalysts. Table 2 shows that support can
+
+
∼10%, respectively, with an increase in the K content to 0.5 wt%
(Fig. 1C and D). At the same time, the fractions of C5+ hydrocarbons,
affect not only the activity but also the product selectivity. The use
of SiO2 as a support showed both a lower conversion of CO2 and a
lower selectivity to hydrocarbons, and thus was unsuitable for the
hydrogenation of CO2 to hydrocarbons. Iron catalysts loaded on
other typical metal oxide supports such as Al O , TiO , and ZrO
particularly C5+ olefins, and C –C4 olefins increased significantly
2
from almost 0 to ∼30% and ∼45%, respectively (Fig. 1C and D). Fur-
+
ther increases in the K content did not change the fractions of CH ,
4
C2–C paraffins, C –C4 olefins, and C5+ hydrocarbons. In a range
4
2
2
3
2
2
+
in the presence of K+ provided much better performances for the
hydrogenation of CO2 to lower olefins. Among these metal oxides,
ZrO2 was the most active and selective support for the conversion
of CO2 to C –C olefins. A carbon nanofiber (CNF)-supported Fe
of K content of 0.5–5.0 wt%, the yields of C2–C4 olefins kept at
+
1
2–13%. A further increase in the K content decreased the yield of
C2–C4 olefins because of the decreased conversion of CO2.
2
4
catalyst was recently demonstrated to exhibit excellent catalytic
performances for the hydrogenation of CO to lower olefins [18].
We examined the catalytic performances of modified Fe catalysts
loaded on two carbon materials, i.e., mesoporous carbon (meso-C)
.2. Characterizations of Fe/ZrO catalysts with and without K+
2
modification
3
To gain insights into the evolution of iron phases during the
reductive pretreatment, we carried out XRD measurements for the
10 wt% Fe/ZrO2 catalysts with and without K+ modification in H2
+
and CNT. The K -Fe/CNT catalyst showed a better performance for
the hydrogenation of CO2 to lower olefins, affording a C –C4 olefin
2
yield of 8.8%. The conversion of CO and the selectivity to lower
2
gas flow at different temperatures. The catalyst was loaded in the
+
olefins over the K -Fe/CNT catalyst were, however, both lower than
in situ XRD cell (XRK-900), and the XRD pattern of the fresh cat-
+
those over the K -Fe/ZrO2 catalyst.
alyst was first recorded. The fresh catalysts with and without K+
Table 2
Catalytic performances of K (1 wt%)-modified Fe (10 wt%) catalysts loaded on various supports for the hydrogenation of CO2a
+
Supportb
CO2 conv. (%)
Select.c (%)
CO
Hydrocarbon distributiond (%)
C2–4 yield (%)
CnHm
CH4
C2–4
C2−40
C5+
C5+0
SiO2
Al2O3
TiO2
ZrO2
Meso-C
CNT
7.1
33
21
42
33
92
17
55
15
31
12
8.1
76
38
64
58
72
17
22
20
32
26
18
36
43
46
30
34
8.6
6.0
7.2
8.2
26
0.1
1.3
13
9.8
8.8
10
0.10
9.0
3.4
12
5.7
28
18
18
2.0
19
35
74
10
11
8.8
a
−1
Reaction conditions: W(catalyst) = 1.0 g, H2/CO2 = 3, T = 613 K, P = 2 MPa, F = 20 mL min , time on stream = 10 h.
Meso-C denotes mesoporous carbon.
b
c
CnHm denotes hydrocarbons; the other products were CO and oxygenates.
d
C2–4 : C2–C4 olefins; C2–40: C2–C4 paraffins; C5+ : C5 olefins; C5+0: C5+ paraffins.
+
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
4
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
Fig. 1. Effect of K+ content on catalytic performances of the K -10 wt% Fe/ZrO2 catalysts for hydrogenation of CO2. (A) CO2 conversion, (B) selectivity, (C) and (D) distribution
+
−
1
of hydrocarbons. Reaction conditions: W(catalyst) = 1.0 g, H2/CO2 = 3, T = 613 K, P = 2 MPa, F = 20 mL min , time on stream = 10 h.
modification contained Fe O3 as the sole iron phase besides mon-
shows that Fe O4 was the main iron phase for the 10 wt% Fe/ZrO2
2
3
−
1
oclinic ZrO . Then, the H gas with a flow rate of 50 mL min was
catalyst under these conditions. The XRD pattern did not undergo
significant changes with time on stream up to 5 h. On the other
2
2
introduced into the cell, and the temperature was raised at a rate
−
1
+
of 10 K min . When the desired temperature was reached, the cat-
alyst was kept at that temperature for 5 min, and the XRD pattern
hand, for the catalyst with K modification, besides the diffraction
◦
◦
◦
peaks at 2ꢂ of 30.1 , 37.0 , and 43.0 belonging to Fe O , new peaks
3
4
+
◦ ◦
at 2ꢂ of 43.5 and 44.2 were observed after 2 h under the (H + CO )
2 2
was recorded. For the Fe/ZrO2 catalyst without K modification, no
significant changes were observed at temperatures ≤573 K (Fig. 2).
gas mixture at 618 K (Fig. 5). These two peaks could be attributed to
◦
At 673 K, the diffraction peak at 2ꢂ of 33.2 ascribed to Fe O dis-
2
3
◦
◦
◦
appeared, and new diffraction peaks at 2ꢂ of 30.1 , 37.0 , and 43.0
attributed to Fe O were observed, revealing the transformation of
3
4
◦
Fe O to Fe O at 673 K. Moreover, the peak at 2ꢂ of 44.9 shifted to
2
3
◦
3
4
4
4.6 and the intensity of this peak increased significantly. A further
increase in the temperature to 723 K led to the disappearance of the
peaks belonging to Fe O , and the further increase in the intensity
3
4
◦
of the peak at 2ꢂ of 44.5 . It should be noted that the peak at 2ꢂ of
◦
◦
◦
4
4.5 –44.9 may arise from both monoclinic ZrO (2ꢂ = 44.9 ) and
2
◦
metallic Fe (2ꢂ = 44.5 ). Because the positions and the intensities
of diffraction peaks belonging to ZrO2 alone did not change signif-
icantly with temperatures, the shift of the peak position slightly
to lower diffraction angles and the significant increase in the peak
intensity at higher temperatures (≥673 K) indicate the formation of
+
metallic Fe. For the K -Fe/ZrO catalyst under H2 flow at different
2
temperatures, similar XRD patterns were observed (Fig. 3). Fe O in
2
3
the fresh catalyst was completely reduced to Fe O at 673 K under
3
4
H flow. The formation of metallic Fe became more significant even
2
at 673 K from the significantly enhanced intensity of the diffraction
◦
peak at 2ꢂ of 44.5 .
We performed in situ XRD measurements for the Fe/ZrO and
2
+
the K -Fe/ZrO catalysts under gas mixture of (H + CO ) with a
2
2
2
H /CO2 ratio of 3. In these experiments, after reduction with H2
2
at 673 K, the temperature was cooled down to 618 K, and then the
reactant gas mixture was introduced to the in situ XRD cell. Fig. 4
Fig. 2. In situ XRD patterns for the Fe/ZrO2 catalyst under H2 gas flow at different
temperatures.
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
5
+
Fig. 5. In situ XRD patterns for the K –Fe/ZrO2 catalyst under (H2 + CO2) (H2/CO2 = 3)
gas flow at 618 K for different times.
+
Fig. 3. In situ XRD patterns for the K –Fe/ZrO2 catalyst under H2 gas flow at different
temperatures.
demonstrates that K+ accelerates the generation of ꢀ-Fe5C2 under
our reaction conditions.
ꢀ
-Fe5C phase. This suggests the formation of ꢀ-Fe5C species over
2 2
+
the K -modified Fe/ZrO catalyst under the (H + CO ) gas mixture.
2
2
2
We performed CO -TPD studies to gain information about the
2
It should be noted that the above in situ XRD measurements
were carried out under atmospheric pressure. We also measured
the XRD patterns for the catalysts after reactions under 2 MPa of
+
adsorption of CO2 on the Fe/ZrO2 catalysts with and without K
modification. The 10 wt% Fe/ZrO2 catalyst without K+ modifica-
tion exhibited very weak CO2 desorption peaks in the CO -TPD
2
(
H + CO ) at 613 K for 10 h. The results for the Fe/ZrO catalysts
2
2
2
+
profile (Fig. 8). The addition of K to the Fe/ZrO caused the appear-
2
modified by different alkali metal ions are displayed in Fig. 6. For
the Fe/ZrO and the Li –Fe/ZrO , the diffraction peaks at 2ꢂ of 43.5
and 44.2 attributable to ꢀ-Fe5C were quite weak. These two peaks
became significantly stronger for the Fe/ZrO catalysts containing
Na , K , Rb , and Cs modifiers. This allows us to propose that the
alkali metal ions of Na , K , Rb , and Cs are more effective for the
generation of ꢀ-Fe5C2 species under the reaction conditions. We
further investigated the effect of K content on the formation of
iron carbides. Fig. 7 shows the XRD patterns after 10 h of reaction
under 2 MPa (H + CO ) at 613 K for the K -10 wt% Fe/ZrO catalysts
with different K contents. The intensities of the peaks ascribed to
ꢀ
ance of two CO2 desorption peaks at lower (∼450 K) and higher
+
◦
2
2
(
900–965 K) temperatures. The intensities of the desorbed CO , par-
◦
2
2
ticularly the one at higher temperatures, increased with the content
2
+
+
of K . This observation indicates that the presence of K promotes
the adsorption of CO2 onto the catalyst surface. We have quanti-
fied the amounts of CO2 desorbed at the lower (peak at ∼450 K)
and the higher (peak at 900–965 K) temperatures. As displayed in
Table 3, the amount of CO2 desorbed at lower temperatures was
+
+
+
+
+
+
+
+
+
+
2
+
2
2
+
-Fe5C increased with the content of K up to 0.5 wt%. This further
2
Fig. 4. In situ XRD patterns for the Fe/ZrO2 catalyst under (H2 + CO2) (H2/CO2 = 3)
Fig. 6. XRD patterns for the Fe/ZrO2 catalysts with different alkali metal ions after
gas flow at 618 K for different times.
reactions. The reaction conditions were the same as those in Fig. 1.
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
6
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
3.3. The roles of alkali metal ions and supports and the nature of
the active Fe species
Our catalytic studies demonstrated that Fe/ZrO2 could catalyze
the hydrogenation of CO , but the main products were CH4 and
2
+
C –C paraffins. The modification by alkali metal ions except for Li
2
4
significantly changed the product distributions; the formations of
CH4 and C –C4 paraffins were dramatically suppressed, and C –C4
2
2
olefins and C5+ hydrocarbons became the main products. The ratios
of olefins to paraffins in C –C4 and C5+ hydrocarbons were >4.4
2
and ∼2.0 over theses alkali metal ion-modified Fe/ZrO catalyst.
2
The highest yield of C –C olefins (13%) was obtained over the
2
4
+
+
K -10 wt% Fe/ZrO2 catalysts with K contents of 0.5–1.0 wt%. The
modification by K+ with proper contents also enhanced CO2 con-
version.
These catalytic results have provided us information about the
+
+
roles of K . The increase in CO2 conversion after the addition of K
to the Fe/ZrO2 catalyst suggests that K+ may promote the activa-
tion of CO . It is generally accepted that the hydrogenation of CO2
2
to hydrocarbons over Fe-based catalysts proceeds via a two-step
mechanism or indirect mechanism, i.e., the conversion of CO to CO
2
via the reverse water–gas shift reaction and the subsequent hydro-
genation of CO [20,29]. Thus, the decrease in the selectivity to CO
and the increase in that to oxygenates in the presence of K (Table 1
+
+
+
Fig. 7. XRD patterns for the K –Fe/ZrO2 catalysts with different K contents after
and Fig. 1) may indicate that K+ enhances the chemisorption and
the subsequent conversion of CO. However, we cannot completely
exclude the possibility that the direct hydrogenation of CO2 might
also contribute to the formation of hydrocarbons, and this direct
mechanism may involve the dissociation of CO2 to carbon species,
reactions. The reaction conditions were the same as those in Fig. 1.
followed by hydrogenation of the adsorbed species [39]. In this case,
+
K
might enhance the direct activation of CO2 to adsorbed carbon
species.
Our CO -TPD studies have demonstrated that the introduction
2
+
of basic K enhanced the adsorption of acidic CO molecules (Fig. 8).
2
This may contribute to increasing the conversion of CO2 over the
+
K -modified Fe/ZrO2 catalysts. It is reasonable that the strongly
chemisorbed CO2 or carbonate species with a desorption peak at
9
00–965 K may be difficult to be hydrogenated under our reaction
conditions and may not contribute to the catalytic conversion of
CO . Thus, it is probable that the chemisorbed CO2 species, which
2
show a desorption peak at a lower temperature (∼450 K), mainly
+
account for the catalytic reaction. At higher K contents, the forma-
tion of larger amounts of strongly chemisorbed CO2 or carbonates
on catalyst surfaces (Fig. 8 and Table 3) may lead to the lower cat-
alytic activity (Fig. 1) because of the possible covering of the active
Fe species.
Concerning the product selectivity, the decrease in the fraction
of C –C4 paraffins and the increases in that of C –C4 olefins clearly
2
2
+
+
Fig. 8. CO2-TPD profiles for the K –Fe/ZrO2 catalysts with different K contents.
suggest the decrease in the hydrogenation ability of the catalyst
+
after the modification by K . The dramatic decrease in the fraction
of CH4 and the increase in that to C5+ hydrocarbons suggest the
increase in the chain-growth probability over the K -modified cat-
about 2.6 times higher than that at higher temperatures for the
+
+
+
catalyst with a K content of 0.2 wt%. The increase in the K content
caused the increases in the amounts of CO2 desorbed, particularly
that at higher temperatures.
alysts. The relatively higher fraction of olefins in C5+ hydrocarbons
+
also indicates that the presence of K decreases the hydrogenation
ability or inhibits the re-adsorption of olefins.
Our characterizations showed that the alkali metal ion pro-
moted the generation of ꢀ-Fe5C under the (H + CO ) gas mixture.
2
2
2
Table 3
The generation of ꢀ-Fe5C2 was confirmed during both the in situ
XRD measurements under the (H + CO ) gas flow (atmospheric
Amounts of CO2 desorption evaluated from CO2-TPD results.
2
2
Catalyst
AmountsofCO2 desorption(mmol g−1)
pressure) and the XRD measurements after reactions for the cata-
+
400–500 K
900–960 K
lysts with K modification. Our studies suggest that there is a good
correlation between the generation of ꢀ-Fe5C2 and the catalytic
1
0
0
1
5
0 wt% Fe/ZrO2
0
0
7.6
21
84
83
+
performance for the production of C –C olefins. Many recent stud-
.2 wt% K –10 wt% Fe/ZrO2
20
24
30
32
2 4
+
.5 wt% K –10 wt% Fe/ZrO2
ies have provided evidence that iron carbide species, particularly
-Fe5C , is an active phase for the production of C –C olefins or
+
.0 wt% K –10 wt% Fe/ZrO2
ꢀ
2
2
4
+
.0 wt% K –10 wt% Fe/ZrO2
C5+ hydrocarbons in the hydrogenation of CO, i.e., Fischer–Tropsch
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
7
+
Fig. 9. Changes of catalytic performances of K –Fe/ZrO2 catalyst for the hydrogena-
tion of CO2. Reaction conditions: W(catalyst) = 1.0 g, H2/CO2 = 3, T = 613 K, P = 2 MPa,
−1
F = 20 mL min
.
+
+
+
Fig. 10. XRD patterns for the K –Fe/ZrO2, K –Fe/SiO2, and K –Fe/CNT catalysts after
reactions. The reaction conditions were the same as those in Fig. 1.
synthesis [32,33,40–46]. For example, Pan and co-workers once
demonstrated that a higher fraction of iron carbide phase resulted
in a higher C5+ selectivity and a lower CH4 selectivity for the Fe
catalyst confined inside the CNTs [41]. Sun et al. found that the
increase in the concentration of ꢀ-Fe5C2 in a Raney Fe@HZSM-5
catalyst led to a higher C5+ selectivity [44]. Very recently, Yang et al.
Concerning the role of the alkali metal ion in the formation of
iron carbide species, through in situ EXAFS/XANES studies, Ribeiro
et al. recently showed that the alkali metal ions also accelerated
the carburization of iron over an Fe/silica catalyst, i.e., the forma-
reported a successful controlled synthesis of ꢀ-Fe5C nanoparticles
tion of iron carbides (mainly ꢀ-Fe5C ), in a CO/He mixture at 563 K
2
2
and found that the ꢀ-Fe5C2 nanoparticles exhibited excellent cat-
alytic performances in Fischer–Tropsch synthesis, providing higher
selectivities to C5+ hydrocarbons and C –C olefins as compared to
[47]. The rate of carburization was found to increase in the order of
unpromoted catalyst < Li-promoted catalyst < Na-promoted cata-
lyst < K-promoted catalyst ≈ Rb-promoted catalyst ≈ Cs-promoted
catalyst [47]. Many studies suggested that the presence of an alkali
2
4
the conventional H -reduced hematite catalyst [45]. As mentioned
2
+
above, the hydrogenation of CO over our catalysts mainly proceeds
metal ion, particularly K , could facilitate the dissociation of CO
2
via the indirect mechanism, i.e., the reverse water-gas shift of CO2
with H to CO and H O, and the subsequent hydrogenation of CO to
because of the electronic effect [33,37,47–50]. The changed elec-
tron density on Fe species in the presence of K may weaken the C-O
+
2
2
hydrocarbons, i.e., Fischer–Tropsch synthesis, although we cannot
completely exclude the possibility of the direct hydrogenation of
bond in CO molecules chemisorbed on Fe surfaces and strengthen
the Fe C bond, accelerating the generation of iron carbide species.
Such an electronic effect may also weaken the Fe-H bond, reducing
the hydrogenation ability of Fe catalysts [48–50].
CO . Actually, the reaction conditions adopted in the present work
2
were similar to those employed in the hydrogenation of CO to lower
olefins over Fe-based catalysts except for the ratio of H2 to CO2 or
CO [18]. Detailed analyses of the catalytic results for the hydro-
genation of CO [18] and CO2 (this work) showed similar product
distributions in hydrocarbons for the two reactions. Therefore, the
To understand the reasons for the differences in catalytic behav-
+
iors of K -modified Fe catalysts loaded on different supports, we
have performed XRD measurements for three typical catalysts, i.e.,
+
+
+
the K -Fe/SiO , K -Fe/CNT, and K -Fe/ZrO catalysts, after reac-
2
2
enhanced generation of ꢀ-Fe5C , which is an active phase for the
tions. The results displayed in Fig. 10 show that, besides the
2
hydrogenation of CO to C5+ hydrocarbons and C –C olefins, would
diffraction peaks ascribed to the support, the peaks attributed to ꢀ-
2
4
+
+
promote the formation of C –C4 olefins in the hydrogenation of
Fe5C2 and Fe O4 were observed for the K -Fe/ZrO2 and K -Fe/CNT
2
3
CO2.
catalysts. On the other hand, only diffraction peaks assignable to
+
Based on the results and discussion described above, we pro-
pose that the enhanced formation of ꢀ-Fe5C2 species after the
modification by an alkali metal ion except for Li+ contributes to
promoting the selectivities to C –C olefins and C5+ hydrocarbons
Fe O could be detected for the K -Fe/SiO2 catalyst. We specu-
3
4
+
late that absence of the ꢀ-Fe5C2 species for the K -Fe/SiO2 catalyst
may be responsible for its lower catalytic activity and selectivity
(Table 2). It was once reported that iron (II), which was gener-
ated during the reduction process, may interact strongly with SiO2,
forming iron silicate [51]. Iron silicate was difficult to be further
2
4
(
mainly olefins) during the hydrogenation of CO . The decreased
2
hydrogenation ability in the presence of an alkali metal ion would
also contribute to the increased fraction of C –C olefins. Further-
reduced, and this might hinder the formation of ꢀ-Fe5C . On the
other hand, such an iron zirconate cannot be formed although there
2
4
2
more, the presence of electron-donating K+ modifier on catalyst
surfaces may hinder the re-adsorption of olefins, which were also
electron donors, and this may also result in higher olefin selectiv-
may exist an interaction between Fe and ZrO [51]. The compari-
2
+
+
son of the K -Fe/CNT and K -Fe/ZrO catalysts, both of which could
2
ities. We further performed a relatively longer-term reaction for
generate ꢀ-Fe5C2 phase during the reaction, showed that the for-
mer catalyst exhibited relatively higher selectivities to CH4 and
C2–C4 paraffins. It is known that CNTs are excellent supports for
hydrogenation reactions because of their unique properties in H2
adsorption and spillover [52,53]. Such an excellent hydrogenation
+
the K -Fe/ZrO catalyst to investigate the stability of this catalyst
2
containing ꢀ-Fe5C . Fig. 9 demonstrates that both the conversion of
2
CO2 and the selectivity to C –C4 olefins do not undergo significant
2
decreases after a reaction at 623 K for ∼60 h.
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031

GModel
CATTOD-8412; No. of Pages8
ARTICLE IN PRESS
8
J. Wang et al. / Catalysis Today xxx (2013) xxx–xxx
ability of the CNT-based catalyst may lead to the relatively higher
selectivities to CH and C –C paraffins. Further studies are needed
[15] A.H. Tullo, Chemical and Engineering News 81 (50) (2003) 15–16.
[
16] J. He, T. Xu, Z. Wang, Q. Zhang, W. Deng, Y. Wang, Angewandte Chemie Inter-
national Edition 51 (2012) 2438–2442.
4
2
4
+
in the future to clarify why the ZrO -supported K -modified Fe cat-
2
[17] T. Xu, Q. Zhang, H. Song, Y. Wang, Journal of Catalysis 295 (2012) 232–241.
[18] H.M.T. Galvis, J.H. Bitter, C.B. Khare, M. Ruitenbeek, A.I. Dugulan, K.P. de Jong,
Science 335 (2012) 835–838.
alyst can afford better catalytic performances for the formation of
C2–C4 olefins.
[
19] U. Olsbye, S. Svelle, M. Bjørgen, P. Beato, T.V.W. Janssens, F. Joensen, S. Bordiga,
K.P. Lillerud, Angewandte Chemie International Edition 51 (2012) 5810–5831.
4
. Conclusions
[20] P.H. Choi, K.W. Jun, S.J. Lee, M.J. Choi, K.W. Lee, Catalysis Letters 40 (1996)
15–118.
21] Z.H. Suo, Y. Kou, J.Z. Niu, W.Z. Zhang, H.L. Wang, Applied Catalysis A: General
48 (1997) 301–303.
1
[
The Fe/ZrO2 catalyst modified by a proper alkali metal ion was
1
found to be an efficient catalyst for the hydrogenation of CO2 to
C2–C4 olefins. The Fe/ZrO2 without modification provided CH4 and
[22] H. Ando, Q. Xu, M. Fujiwara, Y. Matsumura, M. Tanaka, Y. Souma, Catalysis Today
45 (1998) 229–234.
[
23] L. Xu, Q. Wang, D. Liang, X. Wang, L. Lin, W. Cui, Y. Xu, Applied Catalysis A:
General 173 (1998) 19–25.
C2–C4 paraffins as the main products in the hydrogenation of CO .
2
+
The addition of an alkali metal ion except for Li significantly sup-
[24] S.S. Nam, H. Kim, G. Kishan, M.J. Choi, K.W. Lee, Applied Catalysis A: General
179 (1999) 155–163.
[25] M.L. Cubeiro, H. Morales, M.R. Goldwasser, M.J. Pérez-Zurita, F. González-
Jiménez, G. Urbina de N, Applied Catalysis A: General 189 (1999) 87–97.
pressed the formations of CH4 and C –C4 paraffins, and promoted
2
+
the formations of C –C olefins and C5+ hydrocarbons. K and ZrO2
2
4
were demonstrated to be the best modifier and support of iron cat-
[
26] T. Riedel, G. Schaub, K.W. Jun, K.W. Lee, Industrial and Engineering Chemistry
Research 40 (2001) 1355–1363.
alysts for the hydrogenation of CO2 to C –C4 olefins. The catalytic
2
+
+
[27] T. Riedel, H. Schulz, G. Schaub, K.W. Jun, J.S. Hwang, K.W. Lee, Topics in Catalysis
6 (2003) 41–54.
performance was also dependent on the content of K . Over the K -
2
Fe/ZrO2 catalyst with a proper K+ content (0.5–1 wt%), the yield of
[28] T. Herranz, S. Rojas, F.J. Pérez-Alonso, M. Ojeda, P. Terreros, J.L.G. Fierro, Applied
Catalysis A: General 311 (2006) 66–75.
C2–C4 olefins could reach 13%, with the fraction of C –C4 olefins in
2
[
29] P.S.S. Prasad, J.W. Bae, K.W. Jun, K.W. Lee, Catalysis Surveys from Asia 12 (2008)
70–183.
the hydrocarbons being ∼45%. The modification of the Fe/ZrO cat-
2
1
+
alyst by K accelerated the generation of ꢀ-Fe5C species under the
reaction conditions. The presence of K also enhanced the adsorp-
2
[30] R.W. Dorner, D.R. Hardy, F.W. Williams, B.H. Davis, H.D. Willauer, Energy Fuel
23 (2009) 4190–4195.
+
[
31] R.W. Dorner, D.R. Hardy, F.W. Williams, H.D. Willauer, Applied Catalysis A:
General 373 (2010) 112–121.
tion of CO2 and decreased the hydrogenation ability. These were
proposed to mainly contribute to the increase in the conversion of
[
32] E. de Smit, B.M. Weckhuysen, Chemical Society Reviews 37 (2008) 2758–2781.
CO and the change in the product distributions in the presence of
[33] Q. Zhang, J. Kang, Y. Wang, ChemCatChem 2 (2010) 1030–1058.
[34] P. Chen, H.B. Zhang, G.D. Lin, Q. Hong, K.R. Tsai, Carbon 35 (1997) 1495–1501.
2
+
K
modification.
[
35] J. Wambach, A. Baiker, A. Wokaun, Physical Chemistry Chemical Physics 1
1999) 5071–5080.
[36] P. Wang, J. Kang, Q. Zhang, Y. Wang, Catalysis Letters 114 (2007) 178–184.
(
Acknowledgement
[
37] W. Ngantsoue-Hoc, Y. Zhang, R.J. O’Brien, M. Luo, B.H. Davis, Applied Catalysis
A: General 236 (2002) 77–89.
This work was supported by the National Basic Research Pro-
gram of China (No. 2013CB933102), the Natural Science Foundation
of China (Nos. 21173174, 21161130522, 21033006, and 20923004),
and the Program for Changjiang Scholars and Innovative Research
Team in Chinese Universities (No. IRT1036).
[
38] D. Li, N. Ichikuni, S. Shimazu, T. Uematsu, Applied Catalysis A: General 172
(1998) 351–358.
[39] R.A. Fiato, E. Iglesia, G.W. Rice, S.L. Soled, Studies in Surface Science and Catalysis
07 (1998) 339–344.
1
[
[
[
40] M.D. Shroff, D.S. Kalallad, K.E. Coulter, S.D. Köhler, M.S. Harrington, N.B. Jackson,
A.G. Sault, A.K. Datye, Journal of Catalysis 156 (1995) 185–207.
41] W. Chen, Z. Fan, X. Pan, X. Bao, Journal of the American Chemical Society 130
(
2008) 9414–9419.
References
42] E. de Smit, F. Cinquini, A.M. Beale, O.V. Safonova, W. van Beek, P. Sautet,
B.M. Weckhuysen, Journal of the American Chemical Society 132 (2010)
[
[
[
[
1] C.S. Song, Catalysis Today 115 (2006) 2–32.
14928–14941.
2] G. Centi, S. Perathoner, Catalysis Today 148 (2009) 191–205.
3] A.J. Hunt, E.H.K. Sin, R. Marriott, J.H. Clark, ChemSusChem 3 (2010) 306–322.
4] Q. Wang, J. Luo, Z. Zhong, A. Borgna, Energy and Environmental Science 4 (2011)
[
[
[
43] H.M.T. Galvis, J.H. Bitter, T. Davidian, M. Ruitenbeek, A.I. Dugulan, K.P. de Jong,
Journal of the American Chemical Society 134 (2012) 16207–16215.
44] B. Sun, G. Yu, J. Lin, K. Xu, Y. Pei, S. Yan, M. Qiao, K. Fan, X. Zhang, B. Zong,
Catalysis Science & Technology 2 (2012) 1625–1629.
4
2–55.
[5] A. Goeppert, M. Czaun, G.K.S. Prakash, G.A. Olah, Energy and Environmental
45] C. Yang, H. Zhao, Y. Hou, D. Ma, Journal of the American Chemical Society 134
Science 5 (2012) 7833–7853.
(
2012) 15814–15821.
[
[
[
6] M. Aresta, A. Dibendetto, Dalton Transactions (2007) 2975–2992.
7] T. Sakakura, J.C. Choi, H. Yasuda, Chemical Reviews 107 (2007) 2365–2387.
8] J. Ma, N. Sun, X. Zhang, N. Zhao, F. Xiao, W. Wei, Y. Sun, Catalysis Today 148
[
[
46] Q. Zhang, W. Deng, Y. Wang, Journal of Energy Chemistry 22 (2013) 27–38.
47] M.C. Ribeiro, G. Jacobs, B.H. Davis, D.C. Cronauer, A.J. Kropf, C.L. Marshall, Journal
of Physical Chemistry C 114 (2010) 7895–7903.
(
2009) 221–231.
[
48] Y. Yang, H.W. Xiang, Y.Y. Xu, L. Bai, Y.W. Li, Applied Catalysis A: General 266
[9] M. Cokoja, C. Bruckmeier, B. Rieger, W.A. Herrmann, F.E. Kühn, Angewandte
(
2004) 181–194.
Chemie International Edition 50 (2011) 8510–8537.
[
[
49] N. Lohitharn, J.G. Goodwin Jr., Journal of Catalysis 260 (2008) 7–16.
50] R. Luque, A.R. de la Osa, J.M. Campelo, A.A. Romero, J.L. Valverde, P. Sanchez,
Energy and Environmental Science 5 (2012) 5186–5202.
[
10] W. Wang, S. Wang, X. Ma, J. Gong, Chemical Society Reviews 40 (2011)
703–3727.
11] V.P. Indrakanti, J.D. Kubicki, H.H. Schobert, Energy and Environmental Science
(2009) 745–758.
3
[
[
[
[
51] F.R. van den Berg, M.W.J. Crajé, P.J. Kooyman, A.M. van der Kraan, J.W. Geus,
Applied Catalysis A: General 235 (2002) 217–224.
52] H. Zhang, X. Dong, G. Lin, X. Liang, H. Li, Chemical Communications 40 (2005)
2
[
[
12] S.C. Roy, O.K. Varghese, M. Paulose, C.A. Grimes, ACS Nano 4 (2010) 1259–1278.
13] W. Fan, Q. Zhang, Y. Wang, Physical Chemistry Chemical Physics 15 (2013)
5
094–5096.
53] J. Kang, S. Zhang, Q. Zhang, Y. Wang, Angewandte Chemie International Edition
8 (2009) 2565–2568.
2
632–2649.
14] F. Ovampo, B. Louis, L. Kiwi-Minsker, A.C. Roger, Applied Catalysis A: General
92 (2011) 36–44.
[
4
3
Please cite this article in press as: J. Wang, et al., Synthesis of lower olefins by hydrogenation of carbon dioxide over supported iron catalysts,
Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.03.031