Tuning the Facet Proportion of Co2C Nanoprisms for Fischer-Tropsch Synthesis to Olefins

Article abstract of DOI:10.1002/cctc.201902273

Cobalt carbide (Co₂C) exhibits strong facet effect for Fischer-Tropsch to olefins (FTO) reaction. Herein, we report that the facet proportion of Co₂C nanostructures can be tuned effectively by incorporating and altering the Mn content in the CoMn composite oxide as catalyst precursor. With the addition of Mn promoter, the Co₂C nanoprisms with exposed (020) and (101) facets are generated under reaction conditions. In addition, the facet proportion of Co₂C(020) facet can be effectively improved by enhancing the Mn/Co ratio. With the increase of facet proportion of Co₂C(020)/Co₂C(101) ratio, the as-obtained Co₂C nanoprisms exhibit higher intrinsic activity and lower methane selectivity during the syngas conversion process. Kinetic experiments also demonstrate that the apparent activation energy (E_a) for CO conversion is significantly reduced as increasing the facet proportion of Co₂C(020). This work provides a simple and feasible way to tune the exposed facet proportion for the rational design of Co₂C nanocatalyst for FTO reaction.

Full text of DOI:10.1002/cctc.201902273

Accepted Article
Title: Tuning the facet proportion of Co2C nanoprisms for Fischer-
Tropsch synthesis to olefins
Authors: Liangshu Zhong, Yunlei An, Tiejun Lin, Kun Gong, Xinxing
Wang, Hui Wang, and Yuhan Sun
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.201902273
Link to VoR: http://dx.doi.org/10.1002/cctc.201902273
A Journal of
www.chemcatchem.org

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
Tuning the facet proportion of Co2C nanoprisms for Fischer-
Tropsch synthesis to olefins
Yunlei An,[a,b] Tiejun Lin,[a] Kun Gong,[a, c] Xinxing Wang, [a] Liangshu Zhong,*[a,d] Hui Wang, [a] Yuhan
Sun*[a,d]
[a]
CAS Key Laboratory of Low-Carbon Conversion Science and Engineering, Shanghai Advanced Research Institute, Chinese Academy of Sciences,
Shanghai 201203, PR China
E-mail for L. S. Zhong: zhongls@sari.ac.cn; E-mail for Y. H. Sun: sunyh@sari.ac.cn
[b]
[c]
[c]
University of the Chinese Academy of Sciences, Beijing 100049, PR China
College of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, PR China
School of Physical Science and Technology, ShanghaiTech University, Shanghai 201203, PR China
Supporting information for this article is given via a link at the end of the document.
Abstract: Cobalt carbide (Co2C) exhibits strong facet effect for
Fischer-Tropsch to olefins (FTO) reaction. Herein, we report that the
facet proportion of Co2C nanostructures can be tuned effectively by
incorporating and altering the Mn content in the CoMn composite
oxide as catalyst precursor. With the addition of Mn promoter, the
Co2C nanoprisms with exposed (020) and (101) facets are
generated under reaction conditions. In addition, the facet proportion
of Co2C(020) facet can be effectively improved by enhancing the
Mn/Co ratio. With the increase of facet proportion of
Co2C(020)/Co2C(101) ratio, the as-obtained Co2C nanoprisms
exhibit higher intrinsic activity and lower methane selectivity during
the syngas conversion process. Kinetic experiments also
demonstrate that the apparent activation energy (Ea) for CO
conversion is significantly reduced as increasing the facet proportion
of Co2C(020). This work provides a simple and feasible way to tune
the exposed facet proportion for the rational design of Co2C
nanocatalyst for FTO reaction.
theory study, HCP-Co was belonged to the D3h point group and
it exhibited a dihedral-like shape with two closed packed (0001)
facets, while FCC-Co had an octahedron-like shape composed
of (100), (110), (310) and (111) facets. In addition, the reaction
rates for (11–21), (10–11), (10–12), and (11–20) facets of HCP-
Co were higher than that for the exposed facets of FCC-Co.[5]
During the FT reaction process, many products including
paraffins, olefins and oxygenates are generated simultaneously.
For the traditional FT catalysts, paraffins are the main products
with ASF distribution. In order to make the FT technology more
economical, it is necessary to increase the selectivity to value-
added chemicals. Lower olefins (C2-4=), which are key building
blocks in chemical industry, are traditionally produced from
catalytic cracking of hydrocarbon feedstocks from petroleum
reserves [6]. Fischer-Tropsch to olefins (FTO) is a promising
route for direct production of olefins from syngas.[6-7] Iron carbide
is recognized as the active phase for Fe-based FTO reaction
and the effect of various promoters on catalytic performance are
widely investigated.[8] It was found that the addition of Na and S
played key role to improve the catalytic performance for FTO.[9]
In addition, Zn as promoter was also often used to modify the
nature of Fe-based FTO catalysts. Zhai et al prepared FeZn-Na
catalysts via co-precipitation method and proposed that Zn
promoter could change the size of Fe phase as it acted as a
structure promoter.[10] Liu et al found that Mn promoter could
accelerate the formation of θ-Fe3C as the active phase with high
selectivity toward lower olefins.[11] In addition, the C1 formation
was significantly suppressed and the formation of C5+ product
was favored after incorporation of Mn into Fe-based catalysts for
syngas conversion.[12]
Introduction
Fischer-Tropsch synthesis (FTS) is widely applied to produce
fuels and value-added chemicals from syngas (H2 and CO)
derived from various carbon-containing resources including coal,
natural gas and biomass.[1] The goal in the traditional FTS
reaction using Fe, Co and Ru based catalysts is to develop a
suitable catalyst with high activity and high selectivity toward C5+
hydrocarbons.[2] It is generally believed that Fischer-Tropsch
synthesis is a surface catalyzed structure-sensitive reaction, and
the catalytic performance is strongly influenced by the
morphology and exposed facets of the active phase.[3] For Fe-
based catalysts, iron carbide (FeCx) is easily formed under FT
reaction conditions, and different exposed facets of FeCx
possess different energy barriers for CO or H2 activation. It was
found that direct CO dissociation was the preferred pathway for
CO activation on the terrace-like χ-Fe5C2 (510) facet, while H-
assisted CO dissociation was the preferred one on the step-like
χ-Fe5C2 (010) and (001) surfaces.[4] Similarly, Co0 as the active
phase for Co-based FT reaction also possesses strong facet
effect. Co0 may exist in the form of hexagonal close-packed
(HCP) phase or face-centered cubic (FCC) phase. Based on
In our previous work, a new kind of Co2C-based catalyst was
explored to exhibit high activity with high olefins selectivity and
low CH4 selectivity for syngas conversion at mild reaction
conditions.[13] Although the formation of Co2C was often
responsible for the deactivation of cobalt-based FTS reaction,[14]
we found that Co2C nanoprisms with exposed facets of (020)
and (101) were very promising for FTO reaction. Furthermore,
we found that Na promoter could improve the formation rate of
Co2C.[15] In addition, it was found that Mn promoter was
important in modifying the active metal properties, and the
addition of Mn could increase the reaction rate and enhance the
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
olefins selectivity and olefin/paraffin (O/P) ratio.[16] We also
investigated the effect of the Mn promoter and found that Mn
could combine with Co element to form CoxMn1-xO spinel, which
was the essential precursor for the formation of Co2C
nanoprisms under FTO reaction conditions.[16] In addition,
density functional theory (DFT) calculations indicated that Co2C
possessed strong facet effect in syngas conversion, and
different facets of Co2C possessed different catalytic
performance. For example, the Co2C(020) facet was highly
active for CO dissociation and possessed the highest effective
barrier for CH4 formation. Compared with Co2C(020) facet, the
barrier for CO dissociative adsorption was higher on the
Co2C(101) surface, and the Co2C(101) was more active for CH4
formation [17]. However, further experimental results are still
needed to reveal the facet effect of Co2C in syngas conversion.
Co2C nanoprisms are composed of Co2C(020) and Co2C(101)
facets, it seems feasible to tune the catalytic performance via
control the preferential facet of Co2C nanoprisms. Herein, we
investigated the role of Mn in controlling the proportion of
Co2C(020) facet in Co2C nanoprisms in detail. The
corresponding changes in the catalytic performance for FTO
reaction as well as kinetic parameters were correlated with Co2C
nanostructure.
The XRD patterns of various CoMn samples after reduction and
reaction were shown in Figure 1. It was found that the Co
species existed in different phases at different stages. For the
reduced Co3O4 catalyst without Mn addition, the peaks were
mainly attributed to Co0 and CoO (Figure 1a). Besides, the
results for the spent Co3O4 sample revealed that only Co2C
phase at 2θ of 37.0°, 41.3°, 42.6°, 45.7° and 56.6° could be
detected (Figure 1b), suggesting that Co species were all
transformed into Co2C. However, all the reduced CoMn samples
showed the characteristic peak of CoxMn1-xO phase (Figure 1a),
which was the precursor for the formation of Co2C
nanoprisms.[16] For the spent Co3Mn1, Co2Mn1 and Co1Mn1
catalysts, both MnCO3 and Co2C phase co-existed. However,
MnCO3, CoxMn1-xO, MnO and Co2C phases were observed in
the spent Co1Mn2 and Co1Mn3 catalysts, and the peak intensity
of MnCO3 phase for the spent Co1Mn3 catalyst was very weak.
In addition, the Co2C phase showed different crystal planes
intensities for various spent samples.
Results and Discussion
Structure characterization
ICP and BET
The actual metal contents of Co and Mn for various catalysts
with different Mn/Co ratios were determined by ICP-OES as
summarized in Table 1. The molar ratio of Mn/Co was close to
the nominal composition. For all of the studied catalysts, a small
quantity of Na existed in various CoMn catalysts, which could
accelerate the formation of Co2C.[15]
Figure 1. X-ray diffraction patterns for various catalysts with different Mn/Co
ratio after reduction (a) and reaction (b). The reduction process was under 10
vol % CO/N2-300°C. The catalytic test was carried out at a H2/CO ratio of 0.5,
250 ℃, 5 bar and WHSV of 2000 ml·g-1·h -1.
The contents and average crystallite size of Co2C after reaction
were calculated as shown in Table 2, Figure S1 and Figure S2.
The content of Co2C showed the following sequence: Co3O4＞
Co3Mn1＞Co2Mn1＞Co1Mn1＞Co1Mn2＞Co1Mn3. As for the
crystallite sizes of Co2C, the maximum size of 31.6 nm could be
found for the spent Co3O4 catalyst. With doping of Mn promoter,
the crystallite size of Co2C decreased obviously to 18.5 nm
(Co3Mn1), then the size seemed to decline slightly to 15.3 nm
with further increasing of Mn content. Obviously, the addition of
Mn promoter could enhance the dispersion of Co species and
increase the nucleation number of Co2C. The proportion of (020)
facet in Co2C was also calculated in Table S1, Table S2 and
Table 2. The specific surface area proportion of Co2C(020) facet
was calculated based on the model method, as shown in Figure
S2. The change of surface area proportion of Co2C(020) showed
an increasing tendency with the increasing of Mn/Co ratio. For
example, the surface area proportion of Co2C(020) facet raised
from 18.2% to 25.1% as improving the Mn/Co ratio from 0.33 to
3. Furthermore, the probability of exposed Co2C(101) and (020)
facets was also calculated according to the means proposed by
Huo et al [18] based on the XRD results. As shown in Table 2 and
Table S2, the relative value (I/Lp) of Co2C(101) facet decreased
gradually with the increasing of Mn/Co ratio. However, the
relative value (I/Lp) for Co2C(020) facet increased. And the
The BET surface area of the calcined and reduced samples was
measured by N2 physisorption as shown in Table 1. For the
calcined catalysts, the specific surface area increased with
increasing of Mn content as Mn could improve the dispersion of
Co. For the reduced catalysts, once the Mn promoter was added,
the BET surface area was raised from 18.7 m2/g (Co3O4) to 98.3
m2/g (Co3Mn1), then decreased to 43.4 m2/g with further
increasing of Mn content.
Table 1. Elements analysis and surface properties of various catalysts.
ICP (wt%)
BET (m2/g)
Calcined
Catalyst
Co
Mn
Mn/Co (M/M)
Reduced
Co3O4
78.4
57.3
/
0
76.0
18.7
98.3
Co3Mn1
16.0
0.30
130.0
Co2Mn1
Co1Mn1
48.7
34.5
22.2
31.3
0.49
0.97
136.6
158.2
91.1
45.7
Co1Mn2
Co1Mn3
22.2
16.8
40.4
45.5
1.96
2.94
168.6
160.6
43.4
51.1
XRD
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
relative value for (020)/(101) increased from 1.7 for Co3O4 to
13.2 for Co1Mn3.
Co3Mn1, Co2Mn1, Co1Mn1, Co1Mn2 and Co1Mn3. The lattice
fringe of 2.18 Å was ascribed to Co2C(020) facet and 2.12 Å was
matched with Co2C(111) plane. In addition, the Co2C(020) facet
presented in Figure 3 seemed to grow larger with increasing of
Mn content, which was consistent with the XRD results.
Table 2 Content of Co2C, crystallite size of Co2C, proportion of Co2C(020) and
relative value (I/Lp) for (020)/(101) for various spent CoMn catalysts with
different Mn/Co ratio.
Crystallite
Size of Co2C
(nm)
Proportion of
Co2C(020)
facet (%)
(020)/(101)
{Relative
(I/Lp)}.
1.7
Content of
Co2C (%)
Catalysts
Co3O4
100
54
48
36
29
20
31.6
Co3Mn1
Co2Mn1
Co1Mn1
Co1Mn2
Co1Mn3
18.5
18.2
18.9
21.7
24.0
25.1
2.0
17.7
3.1
18.0
5.0
17.9
10.0
15.3
13.2
H2-TPR
The reduction behavior of various CoMn samples with different
Mn/Co ratio was studied by H2-TPR analysis (Figure 2). The
unpromoted Co3O4 catalyst showed two step reduction pattern:
Co3O4→CoO at 158-270 °C, and CoO→Co0 at the temperature
above 270 °C . With the increasing of Mn content, the H2
consumption peaks significantly shifted toward higher
temperature, suggesting that the addition of Mn hindered the
reduction of Co species. In previous work, Guse and Salazar-
Contreras et al also proposed that the formation of a stable Co-
Mn spinel hindered the formation of metallic Co during the
reduction process, even at a low content of Mn.[19] In addition,
there were only two peaks appeared in Co3O4 catalyst. However,
a new peak located at 200 °C- 300 °C appeared in the TPR
profiles was observed with Mn addition, which could be
attributed to the reduction process of CoxMn1-xO.[17]
Figure 3. TEM (a-f) and HRTEM (g-l) images of various spent CoMn catalysts
with different Mn/Co ratio. (a, g: Co3O4; b, h: Co3Mn1; c, i: Co2Mn1; d, j:
Co1Mn1; e, k: Co1Mn2; f, l: Co1Mn3). Reaction conditions: 250 ℃, 5 bar,
H2/CO=0.5, and WHSV of 2000 ml·g-1·h -1.
Catalytic performance
The catalytic performance of various CoMn catalysts at steady
state was presented in Table 3. The presence of Mn promoter
greatly altered the catalytic activity and product selectivity during
FTO process. Once introducing the Mn promoter, CO
conversion greatly increased from 6.1 C% for Co3O4 to 32.9 C%
for Co3Mn1. The CTY raised from 0.4 * 10-6 mol CO/ gCo2C·s to
7.6 * 10-6 mol CO/ gCo2C·s, and reaction rate increased from 0.33
* 10-7 mol CO/ m2Co2C·s to 1.01 * 10-7 mol CO/ m2Co2C·s.
Furthermore, CH4 selectivity decreased from 33.7 C% for Co3O4
to 11.9 C% for Co3Mn1. The above results agreed well with our
previous conclusion that the Co2C nanoprisms possessed higher
activity and lower CH4 selectivity than Co2C nanospheres.[17] In
addition, the C5+ + Oxy. selectivity increased from 23.9 C% to
68.9 C% as increasing the Mn/Co ratio from 0.33 to 3. The total
Figure 2. H2-TPR profiles for various catalysts with different Mn/Co ratio.
TEM
Transmission electron microscopy (TEM) images of the spent
samples with different Mn/Co ratio were displayed in Figure 3.
Co2C nanospheres were found in the spent Co3O4 catalyst and
the lattice distance of 1.98 Å was assigned to the Co2C(210)
facet. In addition, serious sintering occurred for the spent Co3O4
catalysts. A large amount of Co2C nanoprisms with exposed
(101) and (020) facets were found for the spent samples of
=
olefins selectivity and C5+ proportion also increased with the
increasing of Mn contents (Figure S3), which suggested that Mn
promoted the chain growth probability and benefited for the
formation of long chain products, especially for high carbon
olefins. The CO2 selectivities for various CoMn catalysts were all
above 40 C%, which suggested Co2C exhibited high WGS
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
activity and it was suitable for the conversion of syngas derived
from coal or biomass with low H2/CO ratio. Figure 4 showed the
change trends of activity and CH4 selectivity with the proportion
of Co2C(020) facet for various CoMn catalysts. It was clear that
lower Co2C(020) facet proportion would result in lower specific
activity and higher CH4 selectivity. So we can draw conclusion
that the Mn content can tune the catalyst activity and CH4
selectivity via changing the proportion of Co2C(020) facet of
Co2C nanostructures effectively.
chain growth factor (α) could be derived from the linear part of
the plot and obtained by the component with the carbon number
of ≥ 3, as C1 and C2 dropped dramatically from the ideal
predicted line. For the Co3O4 catalyst, an α value of 0.51 was
found and it decreased to 0.33 for Co3Mn1. However, the α
value increased from 0.33 to 0.46 as Mn/Co ratio improved from
0.33 to 3, which suggested that Mn can enhance the chain
growth ability for Co-based FTO reaction.
25
Co3O4
As shown in Table 3 and Figure 5, the O/P ratio in the C2-C4
fraction increased from 1 to 18 as the Mn/Co ratio was raised
from 0 to 3, suggesting that the Mn promoter could suppress the
second hydrogenation reaction of olefins. The maximum olefins
component was C3 fraction for various CoMn catalysts and C2
fraction was the lowest as readsorption and second
hydrogenation occurred easily for ethylene.
Co3Mn1
20
15
10
5
Co2Mn1
Co1Mn1
Co1Mn2
Co1Mn3
13
12
11
10
9
1.5
1.4
1.3
1.2
1.1
1.0
Co3Mn1
0
C3
C4
C2
8
Carbon number
7
Co2Mn1
6
Figure 5. Olefin to paraffin ratio (O/P) within C2-C4 product for various
catalysts. Reaction conditions: 250 ℃, 5 bar, H2/CO=0.5, and WHSV of 2000
ml·g-1·h -1.
5
Co1Mn1
4
3
Co1Mn2
Co1Mn3
2
1
0
18
19
20
21
22
23
24
25
Proportion of (020) facet (%)
Figure 4. Changes of CH4 selectivity and reaction rate with proportion of (020)
facet for various CoMn catalysts. Reaction conditions: 250 ℃ ,
H2/CO=0.5, and WHSV of 2000 ml·g-1·h -1.
5 bar,
Figure 6 plotted the hydrocarbon distribution for various CoMn
catalysts. For the Co3O4 catalyst, the liner fitting of ASF
distribution matched well with the conventional ASF distribution
trend. As Mn promoter was added into Co-based catalyst, a
remarkable deviation from the typical ASF distribution was
observed, and the CH4 value was far below the predicated one.
In addition, the degree of deviation from the classical ASF plot
increased with increasing of Mn/Co ratio, as the ability for
methane formation was greatly suppressed. The hydrocarbon
Figure 6. Hydrocarbons plot ((ln(Wn/n) versus n) for various CoMn catalysts
with different Mn/Co ratio. a: Co3O4; b: Co3Mn1; c: Co2Mn1; d: Co1Mn1; e:
Co1Mn2; f: Co1Mn3. Reaction conditions: 250 ℃, 5 bar, H2/CO=0.5, and
WHSV of 2000 ml·g-1·h -1.
Table 3 Catalytic performance of various CoMn catalysts with different Mn/Co ratio. Co2C time yield (CTY) represents moles of CO converted to products per
gram of Co2C per second. Reaction rate represents moles of CO converted to products per square meter of Co2C per second. Reaction conditions: 250 °C, 5bar,
H2/CO= 0.5, 2000 ml·gcat-1·h -1.
Product selectivity
o
C2-4=/C2-4
CO
CO2
(C%, CO2-free)
CTY
Reaction rate
(10-7 mol CO/ m2Co2C·s)
Catalyst
Conversion
(C%)
Selectivity
(C%)
(10-6 mol CO/ gCo2C·s)
C5++
=
o
CH4
C2-4
C2-4
Oxy.
12.4
23.9
62.4
65.6
66.0
68.9
6.1
1
Co3O4
37.6
49.7
46.7
47.9
40.3
45.7
33.7
11.9
5.8
33.4
48.0
29.0
28.4
30.6
28.5
20.4
16.2
2.6
0.4
0.33
1.01
1.11
1.24
1.36
1.38
32.9
28.3
20.4
16.3
9.2
3
Co3Mn1
Co2Mn1
Co1Mn1
Co1Mn2
Co1Mn3
7.6
11
14
17
16
8.2
3.9
2.0
11.0
12.9
14.2
1.8
1.7
1.0
1.6
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
Kinetic study
results showed that the Mn promoter would hinder the
reducibility of Co species because of the formation of a stable
Co-Mn spinel,[19] and the XRD results also indicated that only
CoxMn1-xO was obtained at the selected reduction temperature.
In addition, the crystallite size of Co2C nanoprisms decreased
with increasing Mn/Co ratio due to the confinement effect of Mn.
The as-obtained results illustrated that the specific surface area
proportion of (020) facet in the total exposed facets increased
rapidly, and it could be fine-tuned by controlling the Mn content.
Based on DFT calculation results, direct (CO → C + O) CO
dissociation was easier occurred on Co2C. For the flat- and
stepped−Co2C(020) surfaces, the calculated CO direct
dissociation barriers were 1.76 and 1.34 eV,[17] respectively. In
addition, the values were 2.16 and 2.18 eV for the flat- and
stepped-Co2C(101), and 2.18 and 2.16 eV for the flat- and
stepped-Co2C(111) surfaces [20]. It was suggested that CO was
much easier dissociated directly on Co2C(020). Besides, for the
CH4 formation, the values of Eeff,CH4 were 2.88, 3.44 eV on the
flat- and stepped-Co2C(020), which was higher than that on the
Co2C(101) and Co2C(111) surfaces (1.06 and 2.11 eV for the
flat- and stepped-Co2C(101), 2.13 and 0.62 eV for the flat- and
stepped-Co2C(111)). Obviously, CH4 was more difficult to be
formed on the Co2C(020) facet than Co2C(101) and Co2C(111)
facet.[17] Therefore, the CoMn catalyst with a larger surface area
proportion of (020) facet exhibited higher intrinsic activity for CO
conversion and lower methane selectivity, as illustrated in Figure
4. The kinetic experiment further elucidates the underlying
reason for the changes in intrinsic activity. It was found that the
apparent activation energy of CO hydrogenation continuously
decreased from 149 kJ mol-1 to 113 kJ mol-1 as Mn/Co ratio
increased from 0.33 to 3. The variation of apparent activation
energy reflected the CO activation ability of Co2C, and it can be
further ascribed to the difference in the surface area proportion
of (020) facet.
According to the above results, the catalyst structure and
catalytic performance evolved greatly for various catalysts with
different Mn content. Thus, it was also necessary to explore the
effect of Mn content on kinetic behaviors. As shown in Figure 7,
the apparent activation energy (Ea) for CO hydrogenation over
the Co3O4 catalysts was 224 kJ mol-1, which was obviously
higher than other Mn promoted catalysts. Based on the TEM
characterization and our previous work, Co2C nanospheres were
obtained for the Co3O4 sample, while Co2C nanoprisms were
observed for all of the Mn-promoted catalysts. Co2C nanoprisms
possessed higher activity due to its lower CO dissociation
energy barrier.[17] With increasing Mn/Co ratio from 0.33 to 3, the
apparent activation energy for CO hydrogenation decreased
gradually from 149 kJ mol-1 to 113 kJ mol-1, which suggested
that the increase of the proportion of Co2C (020) facet could
decrease the CO dissociation energy barrier.
Conclusion
The influence of Mn promoter on tuning the exposed facet
proportion of Co2C as well as the corresponding FTO
performance were investigated in detail. Co2C nanospheres with
(111) as dominant facet were found for the Co3O4 sample. The
addition of Mn into the Co-based catalysts caused a significant
change in catalyst structure and catalytic performance. CoxMn1-
xO spinel phase instead of MnO and Co0 was observed for
CoMn samples after reduction, and Co2C nanoprisms with (020)
and (101) exposed facets gradually evolved under FTO reaction
conditions. Besides, the specific surface area proportion of
Co2C(020) facet in the total exposed facet as well as the relative
intensity ratio of (020)/(101) increased with increasing of Mn/Co
ratio. With the increase of specific surface area proportion of
Figure 7. Arrhenius plots for the reaction rate constants (activation energies,
Ea) on catalysts with different Mn/Co ratio (a: Co3O4, b: Co3Mn1; c: Co2Mn1;
d: Co1Mn1; e: Co1Mn2; f: Co1Mn3).
According to the above study, we found that the addition of Mn
could lead to a significant change in catalyst structure and FTO
performance. As for pure Co3O4, Co0 and CoO phases were
observed after reduction, while CoxMn1-xO could be obtained
after introducing Mn into Co-based catalysts. Our previous work
has already demonstrated that the CoxMn1-xO spinel phase
played a vital role for the formation of Co2C nanoprisms with
exposed (101) and (020) facet. The absence of Mn promoter for
Co3O4 caused the formation of Co2C nanospheres with (111) as
dominant facet via carburization of Co0 and CoO route, which
restrained the CO dissociation and favored the production of
methane. In addition, the intrinsic activity was the lowest when
Co2C nanospheres with (111) facet were formed. The H2-TPR
Co2C(020) in nanoprisms,
the intrinsic activity for CO
conversion also increased and the methane formation was
greatly suppressed. Kinetic experiments further verified that the
apparent activation energy (Ea) for CO conversion decreased
from 149 kJ mol-1 to 113 kJ mol-1 with increasing surface area
proportion of Co2C(020) facet. Above all, this work demonstrated
that the Co2C morphology and the preferential exposed (020)
facet proportion can be effectively tuned by adjusting the Mn/Co
ratio, and it may provide a feasible way for the rational design of
Co2C nanocatalysts for FTO reaction.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
used to estimate the relative numbers of exposed facets, where I is the
diffraction intensity and Lp is the diffraction angle factor, which are
Experimental Section
described as following [18]
:
Catalyst preparation
Lp = (1+cos22θ)/sin2θcosθ
CoMn catalysts with different molar Co/Mn ratio (CoxMny, x/y denoted
the molar of Co/Mn ratio and it was controlled with 0, 0.33, 0.5, 1, 2 and 3,
respectively) were synthesized via co-precipitation method using
aqueous solutions of cobalt nitrate ((Co(NO3)2.6H2O), manganese nitrate
(50 wt % Mn(NO3)2) as the precursor salts, and sodium carbonate
(Na2CO3) as the precipitate, and they were all purchased from
Sinopharm Chemical Reagent Co, Ltd. The solution containing 2 M total
metal ions (Co2+ and Mn2+), and 2 M Na2CO3 were added simultaneously
into the mother solution (100 ml deionized water) under strong stirring
with rotational speed 260 rpm. The pH of the solution was fixed at about
8, and the temperature of the water bath was kept at about 30 oC. After
co-precipitation, the as-prepared mixture was aged at 30 oC for 2 h under
continuous stirring, then washed for several times until the pH value
reached ca. 7.0. The as-obtained samples were dried at 120 oC for 10 h
and then calcined at 330 oC for 3 h at atmospheric pressure. The
calcined catalysts were reduced under flowing atmosphere with 10 vol %
CO/N2 (200 ml/min) at 300 oC for 5 h. The reduced samples were
passivated with 1 vol % O2/Ar for 70 minutes at room temperature for
structure characterization. The spent samples were removed from the
reactor after similar passivation process for structure characterization
Transmission electron microscopy (TEM) characterization was carried
out on a JEOL JEM 2011 electron microscope, equipped with 200 kV
accelerating voltage. Samples for TEM test were prepared by dispersing
the powder in ethanol followed by ultrasonication. One droplet of the
suspension was dripped onto carbon-coated copper grids for
measurement. Spatial drift was corrected with a simultaneous image
collector.
Catalytic evaluation
Syngas conversion was carried out on a continuous flow fix bed reactor.
Generally, 1.5 g catalyst of 40-60 mesh mixed physically with 3 g silica
sand with the same size were all loaded into a stainless steel reactor with
inner diameter of 9 mm. Prior to the reaction, the catalyst was reduced in
situ 10 vol % CO/N2 at 300 °C for 5 h under atmosphere pressure. After
reduction, the temperature then dropped to 250 °C in He (99.999 vol %)
flow to purge the residual reduction gas. Subsequently, the He was
switched to a mixture of 97 vol % syngas (H2/CO=0.5, v/v) and 3 vol %
N2 as internal standard in the reactor as feed gas. A system pressure of 5
bar and a WHSV of 2000 ml/h·gcat were adopted for catalytic reaction.
The outlet gas, after passing through the hot trap (120 oC) and cold trap
(0 oC), was immediately analyzed online by gas chromatograph (GC,
Agilent 7890B). H2, N2, CO, CH4, and CO2 were analyzed through a
carbon molecular sieves column (TDX-1) with a thermal conductivity
detector (TCD), and the He was used as the carrier gas. C1 to C10
hydrocarbons were analyzed through a KCl modified alumina capillary
column (19095P-K25) with Ar carrier and hydrogen flame ionization
detector (FID). The CO conversion was calculated according to the
internal standard method, assuming that the N2 amount remained
constant after reaction. The aqueous products, oil products and wax
products, collected from cold trap and hot trap, were analyzed off-line by
Shimadzu GC. The activation energy was measured via changing the
reaction temperature at atmosphere pressure. In the process of studying
activation energy (Ea), we chose the suitable velocity to adjust the
catalyst activity and avoided the diffuse effects.
Catalyst characterization
The actual compositions of the catalysts were determined with a Varian
710-ES Inductively Coupled Plasma Atomic Emission Spectrometer
(ICP-OES, Optima 8000, PerkinElmer).
Nitrogen adsorption measurements were performed on a Micromerictics
TriStar 3020 instrument at −196 oC. 50 mg sample was prepared to load
in a tube and then degassed at 200 °C in vacuum for 6 h. Brunauer–
Emmett–Teller (BET) method was used to calculate the specific area.
The total pore volume and the average pore size were determined using
Barrett–Joyner–Halenda (BJH) method.
H2-temperature programmed reduction (TPR) profiles were collected on a
tp-5080 equipment with 30 mg sample loaded in a U-tube. During the
TPR process, 5 vol % H2/Ar with flow of 30 mL/min was introduced at the
temperature ranged from 50~800 oC with a ramping of 10 oC /min.
The catalytic activity and products selectivity were calculated according
to the following formula:
X-ray diffraction (XRD) patterns were recorded on a Rigaku Ultima IV X-
ray powder diffractometer using Cu Kα radiation that the wavelength (λ)
was 1.54056 Å, operated at 40 kV and 40 mA. The scanning angle (2θ)
was ranging from 30 - 80°. The average crystallite size was estimated
using the Scherrer equation with the strongest diffraction peaks. The
phases were identified by comparing the observed patterns with JCPDS
standard files.
CO conversion was calculated according to eq 1:
(1)
COinlet CO
COinlet
COConversion
outlet
CO2 selectivity was calculated according to eq 2:
CO
2 outlet

The content of different phases was calculated by means of Rietveld
refinement using the GSAS-II program [21]. Structural data for the Co2C,
MnO, MnCO3 and CoxMn1-xO phases were taken from the Inorganic
Crystal Structure Database (ICSD, accession numbers 16895, 643195,
28556 and 9865, respectively). Lattice parameters for Co2C and MnO
were predetermined in a separate refinement and kept fixed during the
refinements. For all specimens, the scale factors, profile shape and
broadening parameters, asymmetry and corrections for the preferred
orientation were refined during the quantitative analysis. Besides, the
parameters of Co2C were calculated by the MDI Jade 6 software.
According to half-width of the diffraction peak in the XRD, similarly to the
crystallite sizes, the edge length of the Co2C nanoprism particles were
deduced from the Scherrer equation. The strongest diffraction peak at 2θ
= 42.57o was used to calculate the crystal size based on the scherrer
equation. Edge length in the model as showed in Figure S2 was
calculated by peaks at 2θ = 41.28o and 37.01o. The relative (I/Lp) was
(2)
CO
2
Selectivity

COinlet COoutle
CTY (Co2C time yield) calculation was based on mass of the active sites
according to eq 3:
nCO
Conversion
(3)
CTY =
100%
mactive
sites
Reaction rates were normalized by surface area of the active phase
according to eq 4:
CTY
Sactive
(4)
Reaction rate 
100%
sites
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
The selectivity of the individual product (hydrocarbon or oxygenate) CnHm
(CO2-free) was obtained according to eq 5:
[11] Y. Liu, J. F. Chen, J. Bao, Y. Zhang, ACS Catal. 2015, 5, 3905-3909.
[12] a) G. L. Bezemer, P. B. Radstake, U. Falke, H. Oosterbeek, H. P. C. E.
Kuipers, A. J. V. Dillen, K. P. de Jong, J. Catal. 2006, 237, 152-161; b) J. H. D.
Otter, K. P. de Jong, Top. Catal. 2014, 57, 445-450; c) G. J. Hutchings, R. G.
Copperthwaite, M. V. D. Riet, Top. Catal. 1995, 2, 163-172; d) R. G.
Copperthwaite, G. J. Hutchings, M. V. D. Riet, J. Woodhouse, Ind. Eng. Chem.
Res. 1987, 26, 18-19; e) T. E. Feltes, Y. Zhao, R. F. Klie, R. J. Meyer, J. R.
Regalbuto, Chemcatchem 2010, 2, 1065-1068; f) M. Mansouri, H. Atashi, F. F.
Tabrizi, A. A. Mirzaei, G. Mansouri, Ind. Eng. Chem. 2013, 19, 1177-1183; g)
F. Morales, E. D. Smit, F. M. F. D. Groot, T. Visser, B. M. Weckhuysen, J.
Catal. 2007, 246, 91-99; h) J. Ming, N. Koizumi, T. Ozaki, M. Yamada, Appl.
Catal. A Gen. 2001, 209, 59-70; i) E. Iglesia, Appl. Catal. A Gen. 1997, 161,
59-78.
nCn Hm
outlet
(5)
Cn Hm Selectivity =
100%
COinlet  COoutlet  CO2
outlet
The C5+ represents hydrocarbons with 5 or more carbons and Oxy.
denotes oxygenates and it was obtained according to eq 6:
(C5+ + Oxy.) Selectivity = 100%- CO2 Selectivity – C1-4 Selectivity
(6)
Where COinlet and COoutlet represented moles of CO at the inlet and outlet,
CO2 outlet and CnHm outlet represented moles C of CO2 and hydrocarbons at
the outlet, and Oxy. refers to oxygenate products (mainly aldehydes and
[13] L. Zhong, F. Yu, Y. An, Y. Zhao, Y. Sun, Z. Li, T. Lin, Y. Lin, X. Qi, Y.
Dai, L. Gu, J. Hu, S. Jin, Q. Shen, H. Wang, Nature 2016, 538, 84-87.
[14] a) J. C. Mohandas, M. K. Gnanamani, G. Jacobs, W. Ma, Y. Ji, S.
Khalid, B. H. Davis, ACS Catal. 2011, 1, 1581-1588; b). Cheng, P. Hu, J. Phys.
Chem. C 2010, 114, 1085-1093; c) H. Karaca, O. V. Safonova, S. Chambrey,
P. Fongarland, P. Roussel, A. Griboval-Constant, M. Lacroix, A. Y. Khodakov,
J. Catal. 2011, 277, 14-26; d) M. Claeys, M. E. Dry, E. V. Steen, E. D. Plessis,
P. J. V. Berge, A. M. Saib, D. J. Moodley, J. Catal. 2014, 318, 193-202; e) G.
Kwak, H. W. Min, S. C. Kang, H. G. Park, Y. J. Lee, K. W. Jun, K. S. Ha, J.
Catal. 2013, 307, 27-36.
alcohols). m
was calculated based on the quantitative Rietveld
active sites
refinement of XRD. S
was calculated based on the Scherrer
active sites
equation and nanostructure of active phase. The detailed analysis
procedure could be found in our previous literatures.[13]
Acknowledgements
This work was supported by the Ministry of Science and
Technology of China (2016YFA0202802), Natural Science
Foundation of China (21573271, 91545112, 21703278, and
21776296), Key Research Program of Frontier Sciences, CAS
(Grant No. QYZDB-SSW-SLH035), the “Transformational
Technologies for Clean Energy and Demonstration”, Strategic
Priority Research Program of the Chinese Academy of Sciences
(Grant No. XDA21020600) and the Youth Innovation Promotion
Association of CAS.
[15] Z. Li, L. Zhong, F. Yu, Y. An, Y. Dai, Y. Yang, T. J. Lin, S. Li, H. Wang,
P. Gao, ACS Catal. 2017, 7, 3622-3631
[16] Z. Li, T. Lin, F. Yu, Y. An, Y. Dai, S. Li, L. Zhong, H. Wang, P. Gao, Y.
Sun, Acs Catalysis 2017, 7, 8023-8032.
[17] Y. An, F. Yu, T. Lin, Y. Lu, S. Li, Z. Li, Y. Dai, X. Wang, H. Wang, L.
Zhong, Y. Sun, J. Catal. 2018, 366, 289-299.
[18] B. H. Toby, R. B. V. Dreele, J. Appl. Crystal. 2013, 46, 544-549.
[19] C. Huo, B. Wu, P. Gao, Y. Yang, Y. Li, H. Jiao, Angew. Chem., Int. Ed.
2011, 50, 7403-7406.
[20] a) K. Guse, H. Papp, Fresenius J. Anal. Chem. 1993, 346, 84-91; b) H.
G. Salazar-Contreras, A. Martínez-Hernández, A. A. Boix, G. A. Fuentes, E.
Torres-García, Appl. Catal. B Environ. 2019, 244, 414-426.
[21] Y. Pei, J. Liu, Y. Zhao, Y. Ding, T. Liu, W. Dong, H. Zhu, H. Su, L. Yan,
J. Li, W. Li, ACS Catal. 2015, 5, 3620-3624.
Keywords: Syngas conversion 1, Fischer-Tropsch synthesis 2,
Fischer-Tropsch to olefins 3, Facet effect 4, Co2C 5
[1]
a) D. I. Foustoukos, W. E. Seyfried, Science 2004, 304, 1002-1005; b)
J. Horita, M. E. Berndt, Science 1999, 285, 1055-1057; c) Y. Pei, Y. Ding, H.
Zhu, D. Hong, Chinese J. Catal. 2015, 36, 355-361.
[2]
a) S. Abelló, D. Montané, Chemsuschem 2011, 4, 1538-1556; b) X. P.
Fu, Q. K. Shen, D. Shi, K. Wu, Z. Jin, X. Wang, R. Si, Q. S. Song, C. J. Jia, C.
H. Yan, Appl. Catal. B Environ. 2017, 211, 176-187; c) S. Zarubova, S. Rane,
J. Yang, Y. Yu, Y. Zhu, D. Chen, A. Holmen, Chemsuschem 2011, 4, 935 –
942; d). W. Niemantsverdriet, A. M. V. D. Kraan, J. Catal.1981, 72, 385-388.
[3]
W. Chen, T. Lin, Y. Dai, Y. An, F. Yu, L. Zhong, S. Li, Y. Sun, Catal.
Today 2018, 311, 8-22.
[4]
T. Herranz, S. Rojas, F. J. Pérez-Alonso, M. Ojeda, P. Terreros, J. L. G.
Fierro, J. Catal. 2006, 243, 199-211.
[5]
J. X. Liu, H. Y. Su, D. P. Sun, B. Y. Zhang, W. X. Li, J. Am. Chem. Soc.
2013, 135, 16284-16287.
[6]
a) F. Jiao, J. Li, X. Pan, J. Xiao, H. Li, H. Ma, M. Wei, Y. Pan, Z. Zhou,
M. Li, S. Miao, J. Li, Y. Zhu, D. Xiao, T. He, J. Yang, F. Qi, Q. Fu, X. Bao,
Science 2016, 351, 1065-1068; b) K. Cheng, B. Gu, X. Liu, J. Kang, Q. Zhang,
Y. Wang, Angew. Chem., Int. Ed. 2016, 55, 1-5.
[7]
a) Y. Cheng, J. Lin, K. Xu, H. Wang, X. Yao, Y. Pei, S. Yan, M. Qiao, B.
Zong, Acs Catalysis 2016, 6; b) B. Gu, A. Y. Khodakov, V. V. Ordomsky,
Chem. Commun. 2018, 54, 2345; c) E. Ø. Pedersen, I. H. Svenum, E. A.
Blekkan, J. Catal. 2018, 361, 23-32.
[8]
Dugulan, K. P. de Jong, Science 2012, 335, 835-838.
[9] H. M. Torres Galvis, A. C. J. Koeken, J. H. Bitter, T. Davidian, M.
H. M. Torres Galvis, J. H. Bitter, C. B. Khare, M. Ruitenbeek, A. I.
Ruitenbeek, A. I. Dugulan, K. P. de Jong, J. Catal. 2013, 303, 22-30.
[10] P. Zhai, C. Xu, R. Gao, X. Liu, M. Li, W. Li, X. Fu, C. Jia, J. Xie, M.
Zhao, Angew. Chem., Int. Ed. 2016, 55, 9902.
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201902273
ChemCatChem
FULL PAPER
Entry for the Table of Contents
The facet proportion of Co2C nanosprisms was tuned effectively for direct production of lower olefins from syngas.
8
This article is protected by copyright. All rights reserved.