A Facile and Efficient Modification of CNTs for Improved Fischer-Tropsch Performance on Iron Catalyst: Alkali Modification

Article abstract of DOI:10.1002/cctc.201501219

In this paper, a facile and environmentally benign method is used to pretreat carbon nanotubes (CNTs) by alkali modification, which decreases the wall thickness, increases the inner diameter, and improves the graphitization degree of CNTs compared to conventional acid treatment. The iron catalyst supported on alkali modified CNTs shows extraordinary Fischer-Tropsch synthesis (FTS) activity and CO conversion (1.78 and 2.41 times, respectively) compared with those supported on either the acid treated or pristine CNTs. Additionally, the diesel (C_10-20) selectivity on Fe/CNTs-Na is up to 60 %, which is much higher than that calculated from the Anderson-Schulz-Flory distribution (39 %). Similar results are also found by using activated carbon (AC) and N-doped multiwall carbon nanotubes (NCNTs) as supports for FTS, proving the universal applicability of alkali modification on carbon material. The positive relationship between the C₅₊ selectivity and C₂-C₄ olefin/paraffin ratio of iron catalyst can also be achieved irrespective of the support type.

Full text of DOI:10.1002/cctc.201501219

DOI: 10.1002/cctc.201501219
Communications
A Facile and Efficient Modification of CNTs for Improved
Fischer–Tropsch Performance on Iron Catalyst: Alkali
Modification
Renjie Liu, Yan Xu, Zhenhua Li,* and Xinbin Ma*[a]
In this paper, a facile and environmentally benign method is
used to pretreat carbon nanotubes (CNTs) by alkali modifica-
tion, which decreases the wall thickness, increases the inner di-
ameter, and improves the graphitization degree of CNTs com-
pared to conventional acid treatment. The iron catalyst sup-
ported on alkali modified CNTs shows extraordinary Fischer–
Tropsch synthesis (FTS) activity and CO conversion (1.78 and
2.41 times, respectively) compared with those supported on
either the acid treated or pristine CNTs. Additionally, the diesel
(C10-20) selectivity on Fe/CNTs-Na is up to 60%, which is much
higher than that calculated from the Anderson–Schulz–Flory
distribution (39%). Similar results are also found by using acti-
vated carbon (AC) and N-doped multiwall carbon nanotubes
(NCNTs) as supports for FTS, proving the universal applicability
of alkali modification on carbon material. The positive relation-
ship between the C5+ selectivity and C2ÀC4 olefin/paraffin ratio
of iron catalyst can also be achieved irrespective of the sup-
port type.
generated from coal or biomass. For these reasons, a supported
iron catalyst is considered as one of the most promising cata-
lysts for FTS process.[3] In addition, the support property could
affect the dispersion, distribution, reduction, carbonization,
and then the catalytic performance (activity and selectivity) of
the supported iron catalyst. Compared with conventional
oxide supports, carbon materials are ideal support candidates
owing to their unique properties, such as an inert surface, high
electronic conductivity, high mechanical and chemical stability,
as well as well-refined nanochannel structure.[4] The flexibility
of tuning the physicochemical properties of carbon materials
makes research of the intrinsic catalytic mechanism particularly
interesting. Acid treatment is widely used to functionalize
carbon nanotubes (CNTs), but this procedure is detrimental to
the environment. To date there are limited reports of using
alkali treatment on CNTs was reported.[5]
Here, we report a facile and environmental benign method
to modify CNTs by alkali treatment (see the Supporting Infor-
mation for more details) with a high degree of graphitization
and a large inner diameter. The modified material was subse-
quently used as a support for an iron-based catalyst, which
was applied to the FTS reaction. The CNTs-t represents the
CNTs after nitric acid treatment.[6] As seen in Figure 1, CNTs-Na
(Figure 1c) shows a larger inner diameter than pristine CNTs
(Figure 1a) and CNTs-t (Figure 1b). The distribution bar graph
of inner/outer diameters for the pristine CNTs and treated CNTs
have been shown in Figure S1 (Supporting Information). Nota-
bly, the tube wall thickness of CNTs-Na decreases compared
with pristine CNTs and CNTs-t. Both pristine CNTs and CNTs-Na
display more or less closed caps, but no closed caps are found
for CNTs-t. Clearly, the treatment CNTs with acid and alkali re-
sults in different mechanisms. As reported previously, the acid
treatment of CNTs could introduce more oxygen-containing
groups, remove the amorphous carbon and metallic impurities
and open the caps.[7] However, the alkali modification may
lead to not only the amorphous carbon removed, but also the
graphite layer of the CNTs exfoliated.
With increasing resource crisis and stringent environment
issues, seeking renewable and sustainable energy resources
and developing green technology is the focus point to solve
the problem restricting the development of society. Fischer–
Tropsch synthesis (FTS) is an alternative non-petroleum route
to produce super-clean fuel and build chemicals from the
syngas (H2 +CO) derived from coal, biomass, nature gas, or
shale gas.[1] High selectivity to desired products is still a chal-
lenging problem to avoid extensive recycling of undesired
light products to syngas or the subsequent hydrogenation
processing of FTS heavy hydrocarbon.[2] According to the An-
derson–Schulz–Flory (ASF) distribution of the FTS reaction, the
diesel (C10–20) selectivity is limited to 39%. Iron-based catalysts,
with the advantages of high activity and low price, have
aroused the increasing attention of industrial researchers and
academics. The high water-gas-shift (WGS) activity on iron cat-
alyst was beneficial for tuning the H2/CO ratio of the syngas
Thermogravimetric (TG) analysis of the supports and the cor-
responding catalysts were performed to study the influence of
treatment methods and iron loading on the thermal stability
of CNTs (Figure 2). Clearly, the initial temperature of CNTs-t is
lower than those of the other samples, owing to the presence
of oxygen-containing functional groups in the tube surface.[8]
Higher O content in CNTs-t from the XPS result (Table S1, Sup-
porting Information) also reflects the more oxygen-containing
functional groups, which implies more defects or structural
damages were produced along the tubes. It was reported that
[a] Dr. R. Liu, Assoc. Prof. Dr. Y. Xu, Prof. Dr. Z. Li, Prof. Dr. X. Ma
Key Lab for Green Chemical Technology of Ministry of Education
Collaborative Innovation Center of
Chemical Science and Engineering
School of Chemical Engineering and Technology
Tianjin University
Tianjin 300072 (P.R. China)
E-mail: zhenhua@tju.edu.cn
xbma@tju.edu.cn
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201501219.
ChemCatChem 2016, 8, 1454 – 1458
1454
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Figure 1. Representative TEM images of pristine a) CNTs, b) CNTs-t, c) CNTs-Na, as-prepared catalysts d) Fe/CNTs, e) Fe/CNTs-t, and Fe/CNTs-Na. Arrowed lines
indicate the closed caps.
Figure 2. TG profiles of the supports (a) and the corresponding catalysts after calcinations (b).
the defects and open caps contributed to high dispersion of
iron particles throughout the CNTs tubes, which was also con-
firmed by TEM results (Figure S3a).[9] Comparatively, the oxida-
tion temperature of the CNTs-Na is lower than those of the
other samples, suggesting that CNTs modified by the alkali are
more vulnerable to oxidative destruction, which may be relat-
ed to the decreased thickness of CNTs indicated by the TEM re-
sults (Figure 1c).[10] A similar result is observed for the alkali
modification of other carbon materials, such as activated
carbon (AC) and N-doped multiwall carbon nanotubes (NCNTs)
(Figure S2). Additionally, the weight of ash residue of CNTs-t
above 6008C decreases more heavily than that of CNTs-Na,
which indicates that acid treatment could remove more impur-
ity and amorphous carbon than alkali treatment. However, an
shift to low temperature is observed from the stability of iron
supported on CNTs, which is attributed to the catalysis effect
of iron on the CNTs oxidation.[8,11] The mass of remnants (a-
Fe2O3) for all the catalysts after calcination is near 27%, corre-
sponding to 18.9% Fe, which is close to the theoretical Fe
loading 20 wt% on the CNTs.
sponding to the disorder-induced band is an indicative of im-
purities or other symmetry-breaking defects. The G mode
around 1575 cmÀ1, corresponding to the tangential vibrations
of the carbon atoms is related to the ordered graphite in the
CNTs.[12] The intensity ratio of the D-band to that of G-band (ID/
IG) is used to evaluate the graphitization degree of carbon ma-
The Raman spectra of the pristine CNTs, CNTs-t, and CNTs-Na
were recorded (Figure 3). The D mode around 1340 cmÀ1 corre-
Figure 3. Raman spectra of the CNTs treated by different methods.
ChemCatChem 2016, 8, 1454 – 1458
www.chemcatchem.org
1455
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
terials.[13] As listed in Figure 3, the graphitization degree of
CNTs-t is the lowest. Notably, the amorphous carbon can be ef-
ficiently removed by acid treatment, which will inevitably intro-
duce more oxygen-containing groups and defects through the
tubes.[6] Superposition of these two factors makes the CNTs-t
having a lower graphitization degree than the pristine sample.
Comparatively, CNTs-Na displays the highest graphitization
degree reflected by the ID/IG. High graphitization degree of the
carbon support facilitates the electron transfer between the
active species and CO molecules. The activation of CO might
be enhanced, accompanied with an improved FTS reaction ac-
tivity.[14] Notably, the G-bond intensity of the CNTs is closely di-
ameter dependence of the tube.[15] The G-band intensity of
CNTs-Na is the lowest among these samples, which corre-
sponds to the smallest diameter of tubes. This observation is
consistence with the TEM and TG analysis (Figure 1c and
Figure 2).
phase transition from a-Fe2O3 to Fe3O4 is mainly related to the
autoreduction of iron oxides during the calcination in Ar. The
iron oxides located inside the tubes or attached on the defect
sites of the carbon material can be easily reduced, owing to
the weakened metal–oxygen bonding.[18] Additionally, the au-
toreduction degree is enhanced with the decrease of the inner
diameter of CNTs.[19] This also demonstrates that alkali modifi-
cation could increase the inner diameter of CNTs, which is in
good agreement with the above results.
The reduction properties of the as-prepared catalysts were
determined by H2-TPR (Figure S4). All the catalysts exhibit
three partly overlapping peaks: the first two peaks around
200–6008C can be assigned to the reduction of Fe2O3 or Fe3O4
to metallic Fe with FeO as the intermediate species[20] and the
tail peak over 6008C ascribed to the carbon gasification.[16] The
reduction degree of Fe/CNTs-Na is higher than that of Fe/CNTs-
t (Table S1). It is reported that iron carbides is active phase for
iron-based FTS catalyst[21] and the formation of iron carbides is
much related with the reduction degree of iron oxides.[22] So, it
is expected that the alkali modification can be an effective
route to modify the CNTs for the supported iron-based FTS cat-
alyst.
The physical properties of the supports and corresponding
catalysts were presented in Table S1. It is clear that either acid
or alkali treatment of CNTs increases the specific surface area
with bigger increment for CNTs-t. The CNTs-Na displays
a bigger average pore size than CNTs-t, which is in agreement
with the TEM results (Figure 1c). Furthermore, both the pore
volumes and average pore sizes of the three catalysts decrease
compared with the corresponding supports. The representative
TEM images of the as-prepared iron catalysts were shown in
Figure S3. More endohedral iron particles than the exposed
iron particles can be found on Fe/CNTs-t and Fe/CNTs-Na,
owing to the open cap and capillary force effect.[16] However,
almost no particles are located on the CNTs interior surface for
Fe/CNTs. In this study, the carbon material treated by alkali is
washed with deionized water until reaching a neutral pH, and
no residual Na on the catalyst is detected by EDS or XPS analy-
sis.
Figure 5a shows the variations of CO conversion and C5+ se-
lectivity with time on stream (TOS) for three catalysts. The FTS
results were summarized in Table 1 by averaging the values
from 15 h to 35 h when the reaction was achieved steady-
state. At the tested reaction condition, the FTS result on the
blank CNTs showed no activity, which indicated that the residu-
al metal on CNTs has no activity toward FTS, whereas CNTs
supported Fe catalyst showed significant FTS activity. As listed
in Table 1, both the acid and alkali modifications of CNTs can
improve the FTS performance of the supported iron catalysts
According to the XRD results (Figure 4), the peaks at 2q=
25.98 and 43.08 are assigned to (002) and (100) diffractions of
graphitic carbon.[17] Both Fe/CNTs and Fe/CNTs-Na show a mix-
ture of the crystal phase of a-Fe2O3 (JCPDS no.33-0664) and
Fe3O4 (JCPDS no.19-0629), whereas Fe/CNTs-t contains a single
Fe3O4 crystalline phase, which may be attributed to the differ-
ent support properties and the location of iron particle. The
Figure 5. a) CO conversion and C5+ selectivity VS time on stream. Reaction
conditions: T=2708C, P=2 MPa, H2/CO=1 and WHSV=4500 hÀ1. b) The
C5+ hydrocarbon distributions on different catalysts.
Figure 4. XRD patterns of the as-prepared iron catalysts.
ChemCatChem 2016, 8, 1454 – 1458
www.chemcatchem.org
1456
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
for chain-growth activity and the formation of C5+, which re-
sults in increased a-olefin re-adsorption capacity for chain
propagation.[28] Therefore, there is a positive correlation be-
tween C5+ selectivity and the C2ÀC4 olefin/paraffin ratio.
Herein, to explore the universal applicability of the alkali
modification on other carbon materials, AC and NCNTs are also
treated by above method and the as prepared supporting iron
catalysts were tested for FTS. The results confirm that the cata-
lysts supported on alkali modified carbon materials display en-
hanced FTS activity and C5+ selectivity, and that there is a gen-
eral applicability for the alkali modification of carbon material
as expected (Figure S7 and S8).
Table 1. Activities and hydrocarbon selectivities of different catalysts.[a]
[b]
Catalyst
XCO
[%]
SCO
[%]
Hydrocarbon selectivity [%]
C=/C
À
2
CH4
C2
C3
C4
C5
Fe/CNTs
Fe/CNTs-t
Fe/CNTs-Na 56.4 35.5
23.4 16.1 29.0
31.6 13.5 16.0
6.4
4.3
2.6
3.8
4.2
2.3
1.8
2.4
1.3
59.0
73.0
88.5
0.09
0.89
3.83
5.4
[a] Reaction conditions: T=2708C, P=2 MPa, n(H2)/n(CO)=1 and
WHSV=4500 hÀ1. [b] Mainly for C2ÀC4 hydrocarbons.
with the Fe/CNTs-Na showing a remarkable increase. The high
graphitization degree, high reduction degree and high iron dis-
persion of Fe/CNTs-Na are responsible for its superior FTS cata-
lytic performance. Additionally, the formation of CH4 and light
hydrocarbons are greatly suppressed for Fe/CNTs-Na, whereas
the desired heavy hydrocarbons selectivity increases a lot. The
high C5+ selectivity (88.5%) may be related to not only the in-
ternal location of the active iron phase in CNTs but also the
big inner diameter of support, which can benefit for the forma-
tion of long chain hydrocarbons.[16,23] The hydrocarbon distri-
butions of different catalysts are shown in Figure 5b. Com-
pared to Fe/CNTs, the Fe/CNTs-t and Fe/CNTs-Na show a trend
toward heavy hydrocarbons production. Moreover, high diesel
selectivity (about 60%) is achieved over Fe/CNTs-Na, which
breaks restriction of the ASF law (39%). Usually, high diesel-
fuel selectivity is difficult to achieve over iron catalysts without
the use of promoters. To the best of our knowledge, our cata-
lyst displays the highest diesel-fuel selectivity obtained by
a non-promoted iron catalyst, which realizes a great potential
for a new route to the synthesis of diesel fuel through the FTS
process.
This report establishes a facile and environmentally benign
method for the modification of CNTs by alkali treatment, which
can then be employed as supports for an iron-based FTS cata-
lyst. The alkali modification could decrease wall thickness, in-
crease inner diameter, and improve the graphitization degree
of CNTs. The Fe/CNTs-Na catalyst displays an excellent FTS per-
formance with high chain-growth probability. An especially
high diesel-fuel selectivity is achieved (60%) over Fe/CNTs-Na.
The universal applicability of the alkali modification on other
carbon materials for supported iron catalyst has also con-
firmed. We believe that this research will open up a new treat-
ment method for the CNTs functionalization and the diesel-fuel
production for industrial applications. In addition, the high
WGS activity on Fe/CNTs-Na catalyst makes the alkali modifica-
tion applicable to a broad range of fuel-processing reactions.
Further research continues into the synergistic effect of pore
size and graphitization degree of the carbon support of our
iron catalysts for FTS.
Experimental Section
Here, both Fe/CNTs-t and Fe/CNTs-Na display good stabilities
during 35 h reaction. As reported previously, the aggregation
and re-oxidation of active phase were main reasons for the de-
activation of the carbon supported Fe catalyst. The Ostwald
ripening process can result in the sintering of iron species sup-
ported on carbon material.[24] The inside location of active
phase in the tube of CNTs can anchor the active phase in
stable manner, owing to the confinement effect.[19,23] Fe3O4 has
been established as the active phase for the WGS reaction.[25]
The existence of iron oxides stabilize the supported iron car-
bides against irreversible deactivation from re-oxidation by
regulating the WGS reaction versus FTS.[26] Moreover, Fe/CNTs-
Na with the high WGS activity can effectively impede the re-
oxidation of iron carbide by the generated water. The XRD pat-
terns of used catalysts were shown in Figure S5. The peaks be-
longing to the Fe5C2 and Fe3O4 can be detected for all the
spent catalysts. Clearly, Fe/CNTs-Na shows more Fe5C2 than the
others, corresponding to a greater reduction and carbonization
degree and, hence an improved FTS performance.[6]
The alkali treatment was performed using the hydrothermal
method. After the treatment, the suspension is filtered and washed
with de-ionized water until a neutral pH is reached, followed by
drying at 110 8C overnight. The resulting sample is designated
CNTs-Na for further use as an iron catalyst support. The 20 wt% Fe
catalyst was prepared by an impregnation method, so are the Fe/
CNTs and Fe/CNTs-t. The samples were characterized by N2 adsorp-
tion, transmission electron microscopy, thermal gravimetric analy-
sis, X-ray diffraction, X-ray photoelectron spectroscopy, Raman
spectroscopy and H2 temperature-programmed reduction. More
details on the preparation methods, characterization, and activity
of these catalysts are provided in the supporting information.
Acknowledgements
We greatly appreciate the financial support from the National
Natural Science Foundation of China (U1462204) and Program of
Introducing Talents of Discipline to Universities (B06006).
In addition, the positive relationship between the C5+ and
C2ÀC4 olefin/paraffin ratio was established (Figure S6). The var-
iation in C2ÀC4 olefin/paraffin ratio may be associated with the
olefin hydrogenation activity.[27] The low hydrogenation activity
corresponds to a high C2ÀC4 olefin/paraffin ratio. It has been
suggested that a high C2ÀC4 olefin/paraffin ratio is beneficial
Keywords: carbon
heterogeneous catalysis · iron
·
Fischer–Tropsch
synthesis
·
[1] a) A. Y. Khodakov, W. Chu, P. Fongarland, Chem. Rev. 2007, 107, 1692–
1744; b) E. de Smit, B. M. Weckhuysen, Chem. Soc. Rev. 2008, 37, 2758–
2781.
ChemCatChem 2016, 8, 1454 – 1458
www.chemcatchem.org
1457
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
[2] a) J. Kang, K. Cheng, L. Zhang, Q. Zhang, J. Ding, W. Hua, Y. Lou, Q. Zhai,
Y. Wang, Angew. Chem. Int. Ed. 2011, 50, 5200–5203; Angew. Chem.
2011, 123, 5306–5309; b) X. Peng, K. Cheng, J. Kang, B. Gu, X. Yu, Q.
Zhang, Y. Wang, Angew. Chem. Int. Ed. 2015, 54, 4553–4556; Angew.
Chem. 2015, 127, 4636–4639.
[15] A. Jorio, M. A. Pimenta, A. G. Souza Filho, G. G. Samsonidze, A. K. Swan,
M. S. Ünlü, B. B. Goldberg, R. Saito, G. Dresselhaus, M. S. Dresselhaus,
Phys. Rev. Lett. 2003, 90, 107403.
[16] R. M. M. Abbaslou, A. Tavassoli, J. Soltan, A. K. Dalai, Appl. Catal. A 2009,
367, 47–52.
[3] M. Casavola, J. Hermannsdçrfer, N. de Jonge, A. I. Dugulan, K. P.
de Jong, Adv. Funct. Mater. 2015, 25, 5309–5319.
[17] J. Jang, J. H. Oh, G. D. Stucky, Angew. Chem. Int. Ed. 2002, 41, 4016–
4019; Angew. Chem. 2002, 114, 4188–4191.
[4] a) A. Saha, C. Jiang, A. A. Martí, Carbon 2014, 79, 1–18; b) Y. Yan, J.
Miao, Z. Yang, F. X. Xiao, H. B. Yang, B. Liu, Y. Yang, Chem. Soc. Rev.
2015, 44, 3295–3346.
[5] a) M. L. Toebes, J. M. P. van Heeswijk, J. H. Bitter, A. Jos van Dillen, K. P.
de Jong, Carbon 2004, 42, 307–315; b) G. Bezemer, A. Van Laak, A. Van
Dillen, K. P. de Jong, Stud. Surf. Sci. Catal. 2004, 147, 259–264.
[6] Z. Li, R. Liu, Y. Xu, X. Ma, Appl. Surf. Sci. 2015, 347, 643–650.
[7] H. Liu, T. Ye, C. Mao, Angew. Chem. Int. Ed. 2007, 46, 6473–6475; Angew.
Chem. 2007, 119, 6593–6595.
[18] W. Chen, X. Pan, M. G. Willinger, D. S. Su, X. Bao, J. Am. Chem. Soc. 2006,
128, 3136–3137.
[19] W. Chen, X. Pan, X. Bao, J. Am. Chem. Soc. 2007, 129, 7421–7426.
[20] R. M. M. Abbaslou, J. Soltan, A. K. Dalai, Appl. Catal. A 2010, 379, 129–
134.
[21] C. Yang, H. Zhao, Y. Hou, D. Ma, J. Am. Chem. Soc. 2012, 134, 15814–
15821.
[22] S. H. Kang, H. M. Koo, A. R. Kim, D. H. Lee, J. H. Ryu, Y. D. Yoo, J. W. Bae,
Fuel Process. Technol. 2013, 109, 141–149.
[8] G. S. McKee, K. S. Vecchio, J. Phys. Chem. B 2006, 110, 1179–1186.
[9] a) R. M. Malek Abbaslou, A. Tavasoli, A. K. Dalai, Appl. Catal. A 2009, 355,
33–41; b) T. Fu, R. Liu, J. Lv, Z. Li, Fuel Process. Technol. 2014, 122, 49–
57.
[23] W. Chen, Z. L. Fan, X. L. Pan, X. H. Bao, J. Am. Chem. Soc. 2008, 130,
9414–9419.
[24] S. H. Joo, S. J. Choi, I. Oh, J. Kwak, Z. Liu, O. Terasaki, R. Ryoo, Nature
2001, 412, 169–172.
[10] D. K. Singh, P. K. Iyer, P. K. Giri, J. Appl. Phys. 2010, 108, 084313.
[11] H. Xiong, M. Moyo, M. A. Motchelaho, Z. N. Tetana, S. M. A. Dube, L. L.
Jewell, N. J. Coville, J. Catal. 2014, 311, 80–87.
[12] D. Baskaran, J. W. Mays, M. S. Bratcher, Angew. Chem. Int. Ed. 2004, 43,
2138–2142; Angew. Chem. 2004, 116, 2190–2194.
[25] G. P. Van Der Laan, A. A. C. M. Beenackers, Catal. Rev. 1999, 41, 255–318.
[26] P. Thüne, P. Moodley, F. Scheijen, H. Fredriksson, R. Lancee, J. Kropf, J.
Miller, J. W. Niemantsverdriet, J. Phys. Chem. C 2012, 116, 7367–7373.
[27] E. Iglesia, S. C. Reyes, R. J. Madon, S. L. Soled, Adv. Catal. 1993, 39, 221–
302.
[13] a) M. S. Strano, C. A. Dyke, M. L. Usrey, P. W. Barone, M. J. Allen, H. Shan,
C. Kittrell, R. H. Hauge, J. M. Tour, R. E. Smalley, Science 2003, 301, 1519–
1522; b) K. Zhang, Q. Zhao, Z. Tao, J. Chen, Nano Res. 2013, 6, 38–46.
[14] J. Xiao, X. Pan, S. Guo, P. Ren, X. Bao, J. Am. Chem. Soc. 2015, 137, 477–
482.
[28] R. J. Madon, E. Iglesia, J. Catal. 1993, 139, 576–590.
Received: November 5, 2015
Published online on January 20, 2016
ChemCatChem 2016, 8, 1454 – 1458
www.chemcatchem.org
1458
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim