Highly efficient polyoxometalate-based catalysts for clean-gasoline synthesis

Article abstract of DOI:10.1039/c5ta05344g

Poor selectivity for gasoline products is a critical issue for the Fischer-Tropsch synthesis (FTS). Herein, we report that the introduction of a polyoxometalate Cs_2.5H_0.5PW₁₂O₄₀ (CsPW) into a conventional FTS catalyst (Co/Al₂O₃) can create a highly efficient bifunctional catalyst, leading to 118% increase in the selectivity of gasoline. Furthermore, it was found that such a significant improvement is due to the effective hydrocracking of heavier hydrocarbon products at CsPW sites.

Full text of DOI:10.1039/c5ta05344g

Journal of
Materials Chemistry A
View Article Online
View Journal
COMMUNICATION
Highly efficient polyoxometalate-based catalysts
for clean-gasoline synthesis†
Cite this: DOI: 10.1039/c5ta05344g
Chunling Wang,a Cheng Liu,a Yunhang Hu,b Xianni Bu,c Tiejun Zhao,c Kuochih Choua
ad
*
and Qian Li
Received 15th July 2015
Accepted 30th September 2015
DOI: 10.1039/c5ta05344g
www.rsc.org/MaterialsA
Poor selectivity for gasoline products is a critical issue for the Fischer– FTS products. They possess suitable acidity,18–20 which would be
Tropsch synthesis (FTS). Herein, we report that the introduction of efficient for the hydrocracking of heavier FTS products.
a polyoxometalate Cs2.5H0.5PW12O40 (CsPW) into a conventional FTS Furthermore, their acidic properties could be easily tuned by
catalyst (Co/Al2O3) can create a highly efficient bifunctional catalyst, simple synthetic routes, such as embedding metal ions in their
leading to 118% increase in the selectivity of gasoline. Furthermore, it structures.21,22
was found that such a significant improvement is due to the effective
Tungstophosphoric acid (H3PW12O40) is one of the strong
acidic compounds which has been used as an efficient catalyst
in various reactions.23,24 Partial substitution of protons by Cs+
could adjust the insolubility and the amount of available
hydrocracking of heavier hydrocarbon products at CsPW sites.
surface acidic sites. The 2.5 Cs+ substituted sample, Cs2.5H0.5
-
High quality diesel and gasoline fuels are required to reduce air
pollution. The Fischer–Tropsch synthesis (FTS) is an important
industrial process to produce clean hydrocarbon fuels from
syngas obtained from non-petroleum carbon resources such as
natural gas, shale gas, coal, or biomass.1 Over conventional FTS
catalysts, hydrocarbon products generally follow the unselective
PW12O40 (CsPW), has been demonstrated to possess sufficiently
higher thermal-stability and acidity than the parent acid
H3PW12O40,25,26 which can serve as a heterogeneous catalyst for
hydrocracking/hydroisomerization reactions associated with
Fischer–Tropsch processes.27 So far, however, the direct utili-
zation of POMs for a FTS process has not been explored. Herein,
we report an efficient bifunctional FTS catalyst (P–F) consisting
of an insoluble POM (Cs2.5H0.5PW12O40) and a conventional FTS
catalyst (Co/Al2O3) for the FTS process, which exhibited a high
selectivity for C5–C12 gasoline.
CsPW (Fig. 1a) samples were synthesized using a solid
approach (followed by calcination for two hours) which can
easily control the molar ratio of precursors H3PW12O40 and
CsNO3 as described in the literature.28,29 Furthermore, the
samples were subjected to thermal and structure characteriza-
tion. In the thermogravimetric analysis, no obvious weight loss
was observed in the temperature range of 473 to 573 K (Fig. S1†).
Anderson–Schulz–Flory (ASF) distribution from C1 to C100.
2–4 To
obtain gasoline (C5–C12), the heavier products need to be
further hydrocracked and isomerized to light and/or branched
hydrocarbons, in which the acidity of the catalyst is crucial.5
Acidic zeolites are widely used as catalyst supports or co-cata-
lysts to increase the selectivity of gasoline or diesel in a one-step
process.6–11 However, the strong acidity of zeolites might also
cause over-cracking of hydrocarbon products.6
Polyoxometalates (POMs) comprise a large class of nano-
sized, polynuclear metal-oxo anions being formed by early
transition metal ions (V, Nb, Ta, Mo, and W) in high oxidation
states.12–14 As typical polyoxoanions with unique properties,15–17
POMs could meet the requirement for tuning the selectivity of
aState Key Laboratory of Advanced Special Steels, Shanghai University, Shanghai
200072, P. R. China. E-mail: shuliqian@shu.edu.cn
bDepartment of Materials Science and Engineering, Michigan Technological University,
Houghton, Michigan, 49931, USA
cCAS Key Laboratory of Low-Carbon Conversion Science and Engineering, Shanghai
Advanced Research Institute, Chinese Academy of Sciences, Shanghai 201210, P. R.
China
dInstitute of Genomic Material, Shanghai University, Shanghai 200444, P. R. China
Fig. 1 (a) The polyanion structure of CsPW represented by polyhedra.
Colour codes: W grey, P yellow, O red; (b) XRD patterns of CsPW from
473 K to 573 K.
† Electronic supplementary information (ESI) available: Experimental details and
the results of characterization techniques such as TG, IR, H2-TPR, XRD, and FTS
(Fig. S1–8). See DOI: 10.1039/c5ta05344g
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Communication
The range covers special temperatures, including 473 K (syngas
ow to catalyst), 493 K (the FTS reaction generally occurred), 523
K (produce gasoline), and 573 K (metal oxide Co3O4 could be
reduced to active metal Co). The IR spectra of the samples
showed four characterization absorption bands of the Keggin
unit at 1080, 989, 889, and 805 cmꢀ1, which could be assigned
to nas(P–O–W), nas(W–Ot), nas(W–Oc–W), and nas(W–Oe–W)
vibrations,30 respectively. The absorption bands remained
almost unchanged with increasing temperature up to 773 K
(except some decrease of the peak intensities) (Fig. S2†). This
indicates that the primary structure (Keggin unit structure) of
CsPW is stable at a temperature below 773 K. Furthermore, the
XRD patterns of CsPW samples (Fig. 1b) all exhibit reections
corresponding to a cubic Pn3m crystalline structure, similar to
other literature data.21,22 This conrms that the secondary
structure of the CsPW is stable in the temperature range from
473 to 573 K.
Fig. 2 (a) XRD patterns of CsPW, Co/Al2O3 and 0.20P–F; (b) NH3-TPD
profiles of bifunctional catalysts.
0.50P–F, the large reduction peaks at about 823 and 1043 K
indicate the reduction of some WVI species.
The acidity of the bifunctional catalysts was evaluated by
ammonia temperature-programmed desorption (NH3-TPD)
(Fig. 2b). The CsPW exhibits three desorption peaks in the NH3-
TPD prole, corresponding to the weak, medium-strong, and
strong acidic sites. For the CsPW, the higher temperature peak
(observed at about 843 K) can be attributed to the strong
Brønsted acidic sites. Two NH3 desorption peaks were observed
for the bifunctional catalysts at 343 and 523 K. For Co/Al2O3,
three peaks at 313–383 K, which are associated with weaker
acidic sites, were observed. This indicates that the acidity of Co/
Al2O3 is signicantly weaker than that of the bifunctional
catalyst. The strength of the acidic site increases with increasing
CsPW. As a result, for the bifunctional catalyst, the heavier
hydrocarbons can be hydrocracked at the strong acidic sites of
CsPW, while the active metal Co sites are responsible for the
syngas conversion to linear hydrocarbons (Scheme 1).
Bifunctional catalysts (P–F) with four CsPW contents (i.e. 5%,
10%, 20%, and 50%), which were prepared by ball milling, are
denoted as 0.05P–F, 0.10P–F, 0.20P–F, and 0.50P–F, respec-
tively. As shown in Table 1, one can see that the measured bulk
weights of the W element are close to their theoretical values.
Furthermore, XRD measurements showed that the bifunctional
catalyst consists of Co3O4 and CsPW (Fig. 2a and S3†). The
crystallite sizes of Co3O4, which were calculated from the
diffraction peak at 2q ¼ 59.5 using the Scherrer equation,
increased with increasing CsPW amount (Table 1). This indi-
cates that the size of Co3O4 in P–F modestly increased with
increasing amount of CsPW. The H2-TPR proles of the
bifunctional catalysts were obtained. As shown in Fig. S4,† one
can see three H2 consumption peaks for Co/Al2O3, 0.05P–F,
0.10P–F, and 0.20P–F samples and a more complicated TPR
curve for the 0.50P–F sample. The Co/Al2O3 catalyst exhibits two
peaks at 443 K and 513 K, which are associated with the
reduction of Co3O4 to CoO and the transformation of CoO to Co,
respectively. The peak at a higher temperature (about 733 K)
indicates that some Co oxide species are much more difficult to
reduce probably due to the strong metal–support interaction
between Co and Al2O3 in some phases (such as CoAl2O4). The
consumption of H2 at temperatures below 673 K decreased with
increasing CsPW, indicating that CsPW decreased the cobalt
oxide reducibility. This is probably due to the strong interaction
The pore-size distribution curves, which were obtained from
the N2 adsorption–desorption isotherms using the Barrett–Joy-
ner–Halenda (BJH) method, show the mesoporous structure of
the bifunctional catalysts (Fig. S5 and S6†). Compared with the
model FTS catalyst Co/Al2O3 (123.7 m2 gꢀ1), the bifunctional
catalysts exhibited an obvious decrease in surface area and pore
Scheme 1 The reaction mechanism of the bifunctional catalyst con-
between the Cs ions and the cobalt species.31,32 For the catalyst sisting of the FTS catalyst and polyoxometalates.
Table 1 Effect of CsPW amount on C5–C12 selectivity
Selectivity (%)
CsPW CsPW
SBET
Dpore Vpore
Particle size CO conv.
Catalyst
(wt%) contenta (wt%) (m2 gꢀ1
)
(nm) (cm3 gꢀ1
)
(nm) XRD
(%)
CH4 CO2 C2–C4 C5–C12
C13–C20 $C21
Co/Al2O3
0.05P–F
0.10P–F
0.20P–F
0.50P–F
—
5
10
20
50
—
123.7
118.5
110.2
110.4
73.5
12.1
8.8
9.7
10.3
10.5
0.41
0.26
0.25
0.28
0.18
7.5
7.6
7.9
8.3
8.7
84.0
85.1
85.3
82.7
77.9
12.6
0
8.2
16.9
14.4
13.7
10.0
21.3
34.2
42.4
46.4
39.3
27.2
22.2
20.6
19.1
23.7
30.7
11.3
3.1
2.8
7.0
4.97
10.15
19.93
50.32
13.3 2.1
15.7 3.7
14.5 3.7
16.5 3.5
a
Reaction conditions: H2/CO ¼ 2 : 1, SV ¼ 6000 hꢀ1, P ¼ 2.0 MPa, Co/Al2O3 catalyst 0.5 g, 523 K, 72 h. The CsPW content was calculated based on the
W element.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2015

View Article Online
Communication
Journal of Materials Chemistry A
Fig. 3 (a) The TEM image of the Co3O4 particles in the catalyst 0.2P–F;
(b) the TEM image shows the dispersed CsPW particles on the catalyst
0.2P–F.
volume, particularly for the sample 0.5P–F (remarkably
decrease to 73.5 m2 gꢀ1). This happened because CsPW parti-
cles entered the pores of Co/Al2O3, reducing the pore volume.
This was conrmed by the TEM images (Fig. 3), which clearly
showed that the CsPW particles were located in the mesopores
of the g-Al2O3 support and well dispersed in the 0.20P–F sample
with an average size of ca. 0.4 nm (Fig. 3b). However, the average
particle size of Co3O4 is 13.5 nm, which is much larger than the
particle size of CsPW. The mapping of the elements P, Cs and W
further revealed the high dispersion of CsPW (Fig. S7†). The
well-dispersed CsPW on the FTS catalyst created highly
dispersed acidic sites in the bifunctional catalyst. In other
words, ball milling constitutes an efficient approach to disperse
the CsPW on the FTS catalyst. The excellent contact between the
CsPW and the FTS catalyst (Co/Al2O3) ensures that heavy
hydrocarbons generated over Co/Al2O3 can be easily captured
and then hydrocracked by the CsPW. Such an in situ hydro-
cracking can be expected to increase the selectivity of gasoline
products. This was conrmed by the following catalytic
performance tests.
The performance of the bifunctional catalysts was tested for
the FTS using a xed-bed reactor and the results are listed in
Table 1. For all catalysts, the reaction reached a steady state
within 6 hours and then CO conversion remained unchanged.
For the FTS catalyst (Co/Al2O3) without CsPW, one can see the
typical ASF distribution of products, which is wide and unse-
lective from C1 to C50+ hydrocarbons (Fig. 4a and b). In contrast,
the introduction of CsPW into the FTS catalyst led to a remark-
able improvement of C5–C12 selectivity (Fig. 4a and b), namely,
the C5–C12 products increased from 21.3 to 46.4% by adding 20
wt% CsPW into the FTS catalyst, which was accompanied by
only a slight decrease in the CO conversion (the average CO
conversion of the 0.20P–F catalyst remained as high as 82.7%).
This indicates that the selectivity of gasoline (C5–C12) was
increased by 118%.
Fig. 4 (a) Hydrocarbon selectivity of Co/Al2O3 and the bifunctional
catalyst 0.2P–F; (b) ASF distribution of products for catalysts 0.2P–F
and Co/Al2O3; (c) effect of CsPW content on C5–C12 selectivity; (d)
effect of CsPW content on C21+ selectivity.
POMs is benecial to the selective production of gasoline.
+
However, the C21 selectivity went up to 7% when the CsPW
content increased to 50 wt%, leading to a relatively low CO
conversion (77.9%) and high CH4 selectivity (16.5%) over the
0.5P–F catalyst. It is generally recognized that the FTS activity is
mainly dependent on the number of exposed cobalt atoms. The
low conversion of CO over the 0.5P–F catalyst can be explained
as follows: since the CsPW content is large, some of the CsPW
possess the pores of Co/Al2O3 and thus cover some of the
exposed cobalt surface area, which prevented syngas molecules
from coming into contact with Co active metals. The selectivity
for CO2 over 0.10P–F, 0.20P–F, and 0.50P–F catalysts is 3.5–
3.7%, which is much larger than that over the Co/Al2O3 catalyst
without CsPW. This suggests that the acidity of CsPW enhanced
CO dissociation and promoted the water–gas shi reaction.
Finally, IR spectra revealed that the Keggin type structure of
CsPW was stable in the FTS reaction. As shown in Fig. S8,† one
can see four featured peaks (of Keggin type structure) at 804,
888, 985, and 1075 cmꢀ1 indicating that the primary structure of
CsPW remained in the used 0.20P–F-spent catalyst. However,
the other two main bands of the used catalyst at 2846 cmꢀ1 and
2907 cmꢀ1 are associated with the C–H and C–C band vibra-
tions. This is probably due to the hydrocarbon products
retained in the pores of the catalyst.
In conclusion, a highly efficient bifunctional catalyst was
successfully synthesized by introducing CsPW into the
The effect of CsPW content (from 0 to 50 wt%) on the C5–C12
+
conventional FTS catalyst Co/Al2O3 for gasoline (C5–C12
)
selectivity was examined at 523 K (Fig. 4c and Table 1). The C21
production from syngas via the FTS reaction. The C5–C12
production reaches about 44.6% over the bifunctional catalyst,
which is signicantly larger than the maximum value (ca.
21.3%) over a conventional FTS catalyst. The unique acidic
property of POMs plays an important role in the selectivity
enhancement. Compared to Fe-based high-temperature
Fischer–Tropsch catalysts (>573 K) for gasoline-range
products decreased signicantly from 30.7 to 2.8% with
increasing CsPW from 0 to 20 wt% (Fig. 4d). Such a decrease in
+
C21 products with the increase of C5–C12 selectivity over the
+
bifunctional catalyst indicates that C21 hydrocarbons were
hydrocracked by the acidic sites of CsPW. The results are better
than those reported for zeolite under similar reaction condi-
tions.6,11 This indicates that the relatively strong acidity of the
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Communication
¨
hydrocarbons, the POM based bifunctional catalyst consumes 12 M. T. Pope and A. Muller, Angew. Chem., Int. Ed., 1991, 30,
less energy and possesses relatively high CO conversion and less
34–48.
CO2 selectivity.33–35 But there are still two disadvantages of POM 13 U. Kortz and T. Liu, Eur. J. Inorg. Chem., 2013, 1559–1560.
based catalysts for the FTS process. Firstly, the gasoline selec- 14 A. Banerjee, B. S. Bassil, G.-V. Roeschenthaler and U. Kortz,
tivity still needs to be enhanced. In our resent research, we used
POM promoted Ru based FTS catalysts to obtain higher gasoline 15 N. Mizuno, C. Nozaki, I. Kiyoto and M. Misono, J. Am. Chem.
selectivity. Secondly, the isomerization of the C5–C12 over POM Soc., 1998, 120, 9267–9272.
Chem. Soc. Rev., 2012, 41, 7590–7604.
based catalysts was not as high as zeolite based catalysts. The 16 N. Mizuno and M. Misono, Chem. Rev., 1998, 98, 199–217.
elemental composition (such as added Al or Ga elements) and 17 M. Sun, J. Zhang, P. Putaj, V. Caps, F. Lefebvre, J. Pelletier
the type of POM (using the lacunary POMs or other types such as
and J.-M. Basset, Chem. Rev., 2014, 114, 981–1019.
Wells–Dawson type POMs) could be changed to enhance the 18 W. Knaeble, R. T. Carr and E. Iglesia, J. Catal., 2014, 319,
extent of isomerization.
283–296.
19 K. Inumaru, T. Ishihara, Y. Kamiya, T. Okuhara and
S. Yamanaka, Angew. Chem., Int. Ed., 2007, 46, 7625–7628.
20 W. Alharbi, E. Brown, E. F. Kozhevnikova and
I. V. Kozhevnikov, J. Catal., 2014, 319, 174–181.
21 V. Sasca, L. Avram, O. Verdes and A. Popa, Appl. Surf. Sci.,
2010, 256, 5533–5538.
22 K. Narasimharao, D. Brown, A. Lee, A. Newman, P. Siril,
S. Tavener and K. Wilson, J. Catal., 2007, 248, 226–234.
23 N. V. Maksimchuk, K. A. Kovalenko, S. S. Arzumanov,
Y. A. Chesalov, M. S. Melgunov, A. G. Stepanov, V. P. Fedin
and O. A. Kholdeeva, Inorg. Chem., 2010, 49, 2920–2930.
24 J. C. Yori, J. M. Grau, V. M. Benitez and J. Sepulveda, Appl.
Catal., A, 2005, 286, 71–78.
Acknowledgements
We gratefully acknowledge the nancial support from the
National Natural Science Foundation of China (No. 21201122,
No. 51222402) and “ShuGuang” project supported by Shanghai
Municipal Education Commission and Shanghai Education
Development Foundation (13SG39).
Notes and references
1 A. Steynberg and M. Dry, Fischer-Tropsch Technology, Elsevier,
2004.
2 G. P. van der Laan and A. Beenackers, Catal. Rev.: Sci. Eng.,
1999, 41, 255–318.
3 A. Y. Khodakov, W. Chu and P. Fongarland, Chem. Rev., 2007,
107, 1692–1744.
25 X. K. Yang, L. F. Chen, J. A. Wang, L. E. Norena and
O. Novaro, Catal. Today, 2009, 148, 160–168.
26 Y. Liu and M. Misono, Materials, 2009, 2, 2319–2336.
27 Y. Liu, K. Murata and K. Sakanishi, Fuel, 2011, 90, 3056–
3065.
4 Y. Liu, I. Florea, O. Ersen, C. Pham-Huu and C. Meny, Chem.
Commun., 2015, 51, 145–148.
28 L. Zhang, B. Yue, Y. Ren, X. Chen and H. He, Catal. Sci.
Technol., 2013, 3, 2113–2118.
5 J. R. Rostrup-Nielsen, Science, 2005, 308, 1421–1422.
6 J. Bao, J. He, Y. Zhang, Y. Yoneyama and N. Tsubaki, Angew.
Chem., Int. Ed., 2008, 47, 353–356.
7 Q. Zhang, J. Kang and Y. Wang, ChemCatChem, 2010, 2,
1030–1058.
8 J. Kang, K. Cheng, L. Zhang, Q. Zhang, J. Ding, W. Hua,
Y. Lou, Q. Zhai and Y. Wang, Angew. Chem., Int. Ed., 2011,
50, 5200–5203.
9 X. Peng, K. Cheng, J. Kang, B. Gu, X. Yu, Q. Zhang and
Y. Wang, Angew. Chem., Int. Ed., 2015, 54, 4553–4556.
10 C. Xing, G. Yang, M. Wu, R. Yang, L. Tan, P. Zhu, Q. Wei,
J. Li, J. Mao, Y. Yoneyama and N. Tsubaki, Fuel, 2015, 148,
48–57.
29 Y. Zou, B. Yue, B. Zhang and H. Y. He, Chem. Lett., 2006, 35,
202–203.
30 S. M. Choi, Y. Wang, Z. M. Nie, J. Liu and C. H. F. Peden,
Catal. Today, 2000, 55, 117–124.
31 A. H. Lillebo, E. Patanou, J. Yang, E. A. Blekkan and
A. Holmen, Catal. Today, 2013, 215, 60–66.
32 O. Borg, N. Hammer, B. C. Enger, R. Myrstad, O. A. Lindvag,
S. Eri, T. H. Skagseth and E. Rytter, J. Catal., 2011, 279, 163–
173.
33 R. M. M. Abbaslou, A. Tavassoli, J. Soltan and A. K. Dalai,
Appl. Catal., A, 2009, 367, 47–52.
34 Z. Yang, S. Guo, X. Pan, J. Wang and X. Bao, Energy Environ.
Sci., 2011, 4, 4500–4503.
11 C. Xing, W. Shen, G. Yang, R. Yang, P. Lu, J. Sun,
Y. Yoneyama and N. Tsubaki, Catal. Commun., 2014, 55,
53–56.
35 Q. Zhang, W. Deng and Y. Wang, J. Energy Chem., 2013, 22,
27–38.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2015