Gallium oxide nanorods: Novel, template-free synthesis and high catalytic activity in epoxidation reactions

Article abstract of DOI:10.1002/anie.201308384

Gallium oxide nanorods with unprecedented small dimensions (20-80 nm length and 3-5 nm width) were prepared using a novel, template-free synthesis method. This nanomaterial is an excellent heterogeneous catalyst for the sustainable epoxidation of alkenes

Full text of DOI:10.1002/anie.201308384

Angewandte
Chemie
DOI: 10.1002/anie.201308384
Heterogeneous Catalysis
Gallium Oxide Nanorods: Novel, Template-Free Synthesis and High
Catalytic Activity in Epoxidation Reactions**
Warunee Lueangchaichaweng, Neil R. Brooks, Sonia Fiorilli, Elena Gobechiya, Kaifeng Lin,
Li Li, Sonia Parres-Esclapez, Elsa Javon, Sara Bals, Gustaaf Van Tendeloo, Johan A. Martens,
Christine E. A. Kirschhock, Pierre A. Jacobs, and Paolo P. Pescarmona*
Abstract: Gallium oxide nanorods with unprecedented small
dimensions (20–80 nm length and 3–5 nm width) were pre-
pared using a novel, template-free synthesis method. This
nanomaterial is an excellent heterogeneous catalyst for the
sustainable epoxidation of alkenes with H2O2, rivaling the
industrial benchmark microporous titanosilicate TS-1 with
linear alkenes and being much superior with bulkier substrates.
A thorough characterization study elucidated the correlation
between the physicochemical properties of the gallium oxide
nanorods and their catalytic performance, and underlined the
importance of the nanorod morphology for generating a mate-
rial with high specific surface area and a high number of
accessible acid sites.
Additionally, the surface of one-dimensional nanomaterials is
inherently rich in coordinatively unsaturated sites that can
play an active role in catalytic reactions. Solution-phase
techniques have been shown to be a very advantageous and
viable approach for the preparation of metal-oxide nano-
materials.[2] However, these methods typically require the use
of templates or other additives to direct the growth of the
material towards a specific morphology.
Herein, we present a novel and straightforward method
for the fabrication of gallium oxide nanorods with unprece-
dented small dimensions (length ꢀ 80 nm). This nanomaterial
displays excellent properties as heterogeneous catalyst for the
epoxidation of alkenes using the environmentally friendly
hydrogen peroxide as the oxidant.
O
ne-dimensional nanomaterials such as nanorods and
Gallium oxide nanorods (Ga2O3-NR) were prepared
using a precipitation method involving solvolysis of GaCl3
with 2-butanol, followed by hydrolysis and condensation of
the formed species (Figure 1A). The method is reliable and
accessible, does not require any expensive template or
additive, is carried out at very mild temperature and no
energy-consuming calcination step is needed prior to catalytic
application (see the Supporting Information for the detailed
synthesis procedure). The material presents rod-like mor-
phology with a length varying from 20 to 80 nm and a width of
3–5 nm (Figure 1B,C). The X-ray diffractogram of the nano-
rods displays two broad peaks (Figure 1F), which are slightly
shifted compared to those of a g-Ga2O3 sample prepared as
a reference (g-Ga2O3-lit).[3] Although the broadened diffrac-
tion peaks of Ga2O3-NR suggest a low crystallinity, a well-
defined local structural order was evidenced by high-reso-
lution transmission electron microscopy (HR-TEM) and
selected area electron diffraction (SAED) (Figure 1D,E):
the observed planes correspond to the XRD reflections and
match well the d spacings and the orientation of the (100) and
(105) planes of the seldom reported e-Ga2O3 polymorph.[4]
The nanorod morphology of Ga2O3-NR results in the
highest specific surface area (192 m2 gꢁ1) ever reported for
a gallium oxide.[3,5] The adsorption isotherm of type II with H3
hysteresis loop (Figure S1A) indicates the presence of slit-
shaped interparticle pores originating from packing of the
nanorods.
nanowires are drawing growing attention for the specific
physical properties that they display compared to their bulk
counterparts.[1] For a surface-related application such as
heterogeneous catalysis, a key advantage of nanomaterials is
provided by the increased surface-to-volume ratio that
accompanies the decrease in the size of the catalyst particles.
[*] W. Lueangchaichaweng, Dr. E. Gobechiya, L. Li,
Dr. S. Parres-Esclapez, Prof. J. A. Martens, Prof. C. E. A. Kirschhock,
Prof. P. A. Jacobs, Prof. P. P. Pescarmona
Centre for Surface Chemistry and Catalysis, University of Leuven
Kasteelpark Arenberg 23, Heverlee, 3001 (Belgium)
E-mail: paolo.pescarmona@biw.kuleuven.be
Dr. N. R. Brooks
Department of Chemistry, University of Leuven
Celestijnenlaan 200F, Heverlee, 3001 (Belgium)
Dr. S. Fiorilli
DISAT, Politecnico di Torino
Corso Duca degli Abruzzi 24, Torino, 10129 (Italy)
Prof. K. Lin
Natural Science Research Centre, Academy of Fundamental and
Interdisciplinary Science, Harbin Institute of Technology
box-3026, 150080 Harbin (China)
Dr. E. Javon, Prof. S. Bals, Prof. G. Van Tendeloo
EMAT, University of Antwerp
Groenenborgerlaan 171, 2020 Antwerp (Belgium)
[**] This work was supported by START1, Methusalem, Prodex, IAP-PAI,
and the ERC (grant number 24691—COUNTATOMS and grant
number 335078—COLOURATOM) projects. The authors acknowl-
edge Dr. K. Houthoofd, G. Vanbutsele, Dr. C. Klaysom, Prof. J. W.
Seo, Dr. T. Korꢀnyi, and Prof. K. Binnemans for their support in the
characterizations, and Dr. C. ꢁzdilek for useful scientific discus-
sions.
The generation of nanorods is ascribed to an anisotropic
structural feature of the intermediate species, most likely
gallium oxyhydroxides, which are formed in the synthesis
mixture. Indeed, gallium oxyhydroxides and their better
known aluminum counterparts are characterized by ortho-
rhombic unit cells, in which one cell parameter is much larger
than the other two.[6] The conversion of gallium oxyhydroxide
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201308384.
Angew. Chem. Int. Ed. 2014, 53, 1585 –1589
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1585

.
Angewandte
Communications
Figure 1. A) Scheme of the formation of Ga2O3-NR (BuOH is 2-butanol). B,C) TEM images of Ga2O3-NR with 40–80 nm length (B) and with 20-
50 nm length (C). The size of the nanorods can be tuned by adjustments in the protocol of addition of H2O (see the Supporting Information),
without affecting the catalytic activity of the material. D) HR-TEM picture of a selected nanorod (d105 =1.5 ꢂ, d100 =2.5 ꢂ, and angle of 588).
E) Selected area electron diffraction pattern of Ga2O3-NR. F) XRD pattern of: g-Ga2O3-lit (a) and Ga2O3-NR (b). The vertical lines correspond to
reflections of the (100) and (105) planes of e-Ga2O3. G) 71Ga MAS NMR spectra of Ga2O3-NR (gray) and g-Ga2O3-lit (violet), showing Ga in
tetrahedral (ca. 150 ppm) and octahedral (ca. 0 ppm) coordination.
Table 1: Physicochemical properties of Ga2O3-NR and g-Ga2O3-lit.
Surface area [m2 gꢁ1 [a]
]
Gatetra:Gaocta
nH nmꢁ2 [c]
Acid sites [mmolgꢁ1 [d]
]
L:B ratio[d]
Density of acid sites (sitesnmꢁ2
)
[b]
[d]
Catalyst
2O
Ga2O3-NR
g-Ga2O3-lit
192
98
1:10
1:5
26
36
0.16
0.06
18:1
only L
0.63
0.38
[a] Measured by N2 physisorption. [b] Determined by 71Ga MAS NMR spectroscopy. [c] Estimated by TGA. [d] Based on FTIR spectroscopy of adsorbed
pyridine at 1508C. Prior to the IR measurement the samples were subjected to evacuation at 4008C, implying that the measured ratios of Lewis to
Brønsted acid sites are expected to be higher than in the materials used as catalysts. The epoxidation activity of Ga2O3-NR calcined at 4008C does not
differ significantly from that of the as-synthesized material (see Table 2).
into oxide is expected to proceed through condensation steps
followed by partial migration of Ga atoms from octahedral to
tetrahedral sites.[7] Accordingly, solid-state 71Ga NMR spec-
troscopy of Ga2O3-NR shows the presence of Ga atoms in
both tetrahedral and octahedral coordination (Figure 1G and
Table 1), as encountered in e-Ga2O3 and in other gallium
oxide polymorphs,[4] but in contrast
the epoxide yield achieved with Ga2O3-NR in the conversion
of a linear alkene such as 1-octene is comparable to that of
TS-1, whereas the catalytic performance is much superior
with bulky cyclic alkenes such as cyclohexene and cyclo-
octene, which are too large to diffuse through the micropores
of TS-1. The almost complete epoxide selectivity achieved
to oxyhydroxide structures in which
Table 2: Catalytic epoxidation of selected alkenes using 50 wt% aqueous H2O2 (Y=yield and
S=selectivity).
all metal atoms are octahedrally
coordinated.[8]
Alkene
Catalyst
Yepoxide [%]
Sepoxide [%]
TON[a]
Productivity[b]
Ga2O3-NR was tested as heter-
ogeneous catalyst for the epoxida-
tion of alkenes with aqueous H2O2
(Table 2). This is the preferential
path to produce epoxides, which are
a family of versatile compounds
used in numerous industrial synthe-
ses of fine chemicals.[9] Ethyl ace-
tate was employed as environmen-
tally acceptable and inexpensive
solvent. The catalytic performance
of Ga2O3-NR was compared both to
g-Ga2O3-lit and to titanium silica-
lite 1 (TS-1, see Figure S2), which
was chosen as benchmark because
this zeolite is considered the opti-
mum heterogeneous catalyst for the
cyclooctene
Ga2O3-NR
g-Ga2O3-lit
TS-1
84
14
0
>99
>99
0
>99
>99
98
266
109
–
–
250
55
5.3
0.9
0
blank
1
–
Ga2O3-NR4008C
Ga2O3-NR
g-Ga2O3-lit
TS-1
Ga2O3-NR
g-Ga2O3-lit
TS-1
Ga2O3-NR
g-Ga2O3-lit
TS-1
79
17
3
21
45
13
0
20
7
6
5.0
1.1
0.2
1.3
2.2
0.6
0
1.2
0.4
0.4
1-octene
cyclohexene
styrene
>99
>99
>99
>99
0
24
70
142
104
–
109
109
87
58[c]
44[c]
23[c]
Conditions: 2 mmol alkene, 4 mmol H2O2, 40 mg of catalyst, 2.0 g of ethyl acetate as solvent
(acetonitrile for the reactions with TS-1, see the Supporting Information), 4 h at 808C. [a] TON
(molalkene converted/molactive sites) was calculated using the number of acid sites for Ga2O3-NR and g-Ga2O3-lit
(Table 1) and the number of Ti atoms for TS-1. [b] Productivity is defined as gepoxide/gcatalyst. [c] The
epoxidation of small and linear incomplete epoxide selectivity observed with styrene is due to the formation of benzaldehyde, which
originates from the oxidative cleavage of the vinyl group.[12]
alkenes with H2O2.[10] Remarkably,
1586
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1585 –1589

Angewandte
Chemie
with Ga2O3-NR for these alkenes is also noteworthy, partic-
ularly in the case of cyclohexene, which is rather prone to the
formation of the diol and other side-products.[11] In the
epoxidation of styrene, Ga2O3-NR gave higher conversion
and higher epoxide selectivity compared to TS-1. The crucial
effect of the synthesis method on the catalytic performance of
gallium oxides is demonstrated by the strikingly superior
activity of Ga2O3-NR compared to g-Ga2O3-lit (Table 2), in
spite of having the same chemical composition.
negligible formation of products of allylic oxidation in the
reaction of cyclohexene (Table 2).[15]
The higher surface area of the nanorods compared to g-
Ga2O3-lit (Table 1) allows a better contact with the reagents,
and its open structure prevents water entrapment that, on the
other hand, can occur in the ink-bottle-shaped pores of g-
Ga2O3-lit (Figure S1B). These features result in the lower
hydrophilicity of Ga2O3-NR, as demonstrated by the lower
number of H2O molecules adsorbed per square nanometer of
surface (Table 1). Too high hydrophilicity is detrimental for
the epoxidation as it would hinder the approach of the apolar
alkene to the catalyst surface.[16] The morphology of the
nanorods is also favorable for the dispersion of the catalyst
particles in solution,[6b,17] in comparison with the large
aggregates of particles lacking a specific morphology of g-
Ga2O3-lit (Figure S4). Although important, these surface and
morphological properties cannot account alone for the
remarkable superiority of the activity of Ga2O3-NR compared
to g-Ga2O3-lit.
The correlation between the catalytic activity and the
number, type, and strength of the surface acid sites on Ga2O3-
NR and g-Ga2O3-lit was studied by temperature programmed
desorption (TPD) of adsorbed pyridine monitored by FTIR
spectroscopy (Figure 2C and Table 1).[5a,18] Lewis acid sites in
gallium oxides are generally attributed to coordinatively
unsaturated Ga3+ ions mainly located in tetrahedral sites on
the surface,[5a] whereas surface Ga-OH groups are responsible
for mild Brønsted acidity. These Brønsted acid sites can
convert into Lewis acid sites by surface dehydration upon
treatment at high temperature and tend to reconvert to
Brønsted acid sites by rehydration at room temperature.[19]
Ga2O3-NR displays a much higher number of acid sites per
gram of catalyst compared to g-Ga2O3-lit (Table 1). This
significant difference stems from the combination of two
parameters: the higher surface area of Ga2O3-NR, and its
higher surface density of acid sites (Table 1). Both these
The origin of the excellent catalytic properties of Ga2O3-
NR was investigated through a combination of physicochem-
ical characterization techniques. FTIR and Raman spectros-
copy unequivocally demonstrate the activation of H2O2 by
Ga2O3-NR: an IR signal at 835 cmꢁ1 and a clear Raman
shoulder at higher wavenumber compared to the signal of
physisorbed H2O2 at 877 cmꢁ1 appear in the spectra of H2O2 in
contact with Ga2O3-NR, while these signals are not observed
in the absence of H2O2 or if the inactive SiO2 is impregnated
with H2O2 (Figure 2A,B and Figure S3). We propose that the
two signals stem from bidentate (h2) Ga-hydroperoxide and
Ga-peroxide species that can be interconverted upon water
removal,[13] which occurs in the Raman cell upon laser heating
(see the Supporting Information). This assignment is sup-
ported by the similarity between the position and stability of
the IR signal at 835 cmꢁ1 (Figure 2A) and the band generated
by H2O2 activation on TS-1 (837 cmꢁ1), which was attributed
2
[14]
ꢁ
to the O O stretching of an h -hydroperoxide.
The
proposal is also in line with the higher epoxide yield obtained
over Ga2O3-NR with internal alkenes compared to terminal
linear alkenes, evidencing a pathway in which H2O2 is
activated to form
a
metal–hydroperoxide intermediate,
ꢁ
followed by heterolytic cleavage of the O O bond and
transfer of an oxygen atom to the alkene (the double bond in
internal alkenes is more electron-rich and thus more prone to
undergo electrophilic attack by one of the O atoms).[15]
Moreover, a homolytic radical pathway is ruled out by the
Figure 2. A) FTIR difference spectra of: a) Ga2O3-NR impregnated with 30 wt% aqueous H2O2, collected at nominal temperature of ꢁ1738C;
b) after 10 h evacuation at RT; c) Ga2O3-NR impregnated with H2O. The spectrum of pure Ga2O3-NR evacuated at RT was subtracted from all
spectra. The shoulder at 877 cmꢁ1 in (a) is ascribed to residual physisorbed H2O2, which is removed upon evacuation and heating from ꢁ1738C
to RT. In agreement with the observed signal in the IR spectrum, Ga2O3-NR is able to activate H2O2 towards the epoxidation of cyclooctene also
at 258C (though with lower yield, 5% after 24 h). B) Raman spectra of: a) Ga2O3-NR impregnated with 50 wt% aqueous H2O2; b) SiO2
impregnated with 50 wt% aqueous H2O2; c) Ga2O3-NR impregnated with H2O. The arrow indicates the shoulder appearing in the Ga2O3-NR/
H2O2/H2O spectrum. C) FTIR spectra of pyridine chemisorbed on Ga2O3-NR and on g-Ga2O3-lit (measured at 1508C). L and B indicate pyridine
adsorbed on Lewis and Brønsted acid sites, respectively. See Figure S5 for the evolution of the spectrum on Ga2O3-NR as a function of
temperature.
Angew. Chem. Int. Ed. 2014, 53, 1585 –1589
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1587

.
Angewandte
Communications
features originate from the nanorod morphology of the
catalyst: the higher surface area is a consequence of the
higher surface-to-volume ratio of nanosized materials,
whereas the higher surface density of acid sites is attributed
to the intrinsically defective nature of the nanorods and the
consequent higher density of coordinatively unsaturated sites
on their surface compared to materials consisting of larger
particles. Moreover, the Lewis acid sites in Ga2O3-NR are
stronger than those in g-Ga2O3-lit, as proven by the shift to
higher wavenumbers of the peaks originating from pyridine
adsorbed on this type of acid sites (Figure 2C).[5a,18] This is in
line with the higher chemical shift of the 71Ga NMR signal of
tetrahedral Ga sites in Ga2O3-NR compared to g-Ga2O3-lit
(Figure 1G).
which is competitive against the industrial benchmark TS-
1 with a terminal linear alkene as 1-octene but is much more
versatile as its nanorod open structure allows the efficient
epoxidation of larger substrates.
The truly heterogeneous character of Ga2O3-NR was
demonstrated by a hot-filtration test, which ruled out leaching
of catalytically active species during the epoxidation. No
acetic acid was detected by gas chromatography (GC) at the
end of the catalytic tests, thus excluding hydrolysis of ethyl
acetate, which could have contributed to the epoxidation
activity as acetic acid can act as homogeneous catalyst
through the formation of a peracid upon reaction with
H2O2.[21] Ga2O3-NR retained most of its activity upon
recycling in five consecutive catalytic runs (Figure S6). This
represents a remarkable improvement compared to other
transition-metal-free epoxidation catalysts, which suffer from
relatively rapid deactivation.[16,22] Ga2O3-NR was also more
efficient in the use of H2O2 compared to other transition-
metal-free oxides and to TS-1 (see the Supporting Informa-
tion).[16,22]
On the basis of these characterization data, a catalytic
mechanism for the epoxidation of alkenes over Ga2O3-NR
can be proposed (Scheme 1). Both coordinatively unsaturated
In summary, we introduced a template-free, straightfor-
ward method for the synthesis of gallium oxide nanorods with
very high specific surface area and uniquely small dimensions
compared to previously reported one-dimensional gallium
oxides.[1] Ga2O3-NR displays excellent catalytic performance
in the sustainable epoxidation of alkenes with H2O2, achiev-
ing high efficiency in terms of activity, selectivity, oxidant
consumption, substrate versatility, and recyclability. The
attractive features of Ga2O3-NR are very promising for
a broad range of other relevant applications of gallium
oxide that could benefit from a nanorod morphology, both
within catalysis and as functional material with semiconduct-
ing, optoelectronic or sensing properties.[23]
Scheme 1. Proposed catalytic cycle for the epoxidation of alkenes with
aqueous H2O2 over Ga2O3-NR.
Received: September 25, 2013
Revised: November 14, 2013
Published online: January 22, 2014
Ga atoms acting as Lewis acid sites and mainly tetrahedral
Ga-OH groups acting as mild Brønsted acid sites located at
the surface of the nanorods can be considered as catalytic sites
for the epoxidation reaction. Lewis acid sites that are present
in the as-synthesized catalyst are expected to convert in
contact with either H2O or H2O2 during the epoxidation
reaction (0!1 and 0!2). These Ga sites are able to activate
hydrogen peroxide (2), as indicated by FTIR and Raman
spectroscopy. The observed stronger acidity of the Ga centers
in Ga2O3-NR compared to those in g-Ga2O3-lit enhances the
Keywords: epoxidation · gallium oxide ·
heterogeneous catalysis · hydrogen peroxide · nanorods
.
[1] a) G. R. Patzke, F. Krumeich, R. Nesper, Angew. Chem. 2002,
114, 2554; Angew. Chem. Int. Ed. 2002, 41, 2446; b) Y. Jun, J.
Choi, J. Cheon, Angew. Chem. 2006, 118, 3492; Angew. Chem.
Int. Ed. 2006, 45, 3414.
[2] G. R. Patzke, Y. Zhou, R. Kontic, F. Conrad, Angew. Chem.
2011, 123, 852; Angew. Chem. Int. Ed. 2011, 50, 826.
[3] C. Otero Areꢀn, A. L. Bellan, M. P. Mentruit, M. R. Delgado,
G. T. Palomino, Microporous Mesoporous Mater. 2000, 40, 35.
[4] H. Y. Playford, A. C. Hannon, E. R. Barney, R. I. Walton, Chem.
Eur. J. 2013, 19, 2803.
[5] a) A. Vimont, J. Lavalley, A. Sahibed-Dine, C. O. Areꢀn, M. R.
Delgado, M. Daturi, J. Phys. Chem. B 2005, 109, 9656; b) C.
Aprile, E. Gobechiya, J. A. Martens, P. P. Pescarmona, Chem.
Commun. 2010, 46, 7712.
[6] a) For example, the cell parameters of boehmite, g-AlO(OH),
are a = 3.700, b = 12.227, c = 2.868 ꢁ, [00-021-1307]; b) P. A.
Buining, C. Pathmamanoharan, J. B. H. Jansen, H. N. W. Lek-
kerkerker, J. Am. Ceram. Soc. 1991, 74, 1303.
ꢁ
polarization of the O O bond of the hydroperoxide, thus
favoring the electrophilic attack on the double bond of the
alkene (3),[20] which leads to the transfer of an oxygen atom
with formation of the desired epoxide product.
The turnover numbers (TON) over Ga2O3-NR were
calculated based on the number of acid sites determined by
TPD-FTIR (Table 2). High TONs were reached in the
epoxidation of all tested alkenes, with a maximum value for
cyclooctene (266). Moreover, the productivity of Ga2O3-NR
(Table 2) is more than double compared to previously
reported transition-metal-free oxide catalysts.[16b] These data
confirm the outstanding catalytic properties of Ga2O3-NR,
[7] X. Krokidis, P. Raybaud, A. E. Gobichon, B. Rebours, P. Euzen,
H. Toulhoat, J. Phys. Chem. B 2001, 105, 5121.
1588
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1585 –1589

Angewandte
Chemie
ˇ
´
[8] D. Tunega, H. Pasalic, M. Gerzabek, H. Lischka, J. Phys.
[17] C. ꢄzdilek, K. Kazimierczak, S. J. Picken, Polymer 2005, 46,
Condens. Matter 2011, 23, 404201.
6025 – 6034.
[9] a) S. T. Oyama, Mechanisms in homogeneous and heterogeneous
epoxidation catalysis, Elsevier Science, Amsterdam, 2008;
b) B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457.
[10] a) F. Cavani, J. H. Teles, ChemSusChem 2009, 2, 508; b) W. Fan,
R. G. Duan, T. Yokoi, P. Wu, Y. Kubota, T. Tatsumi, J. Am.
Chem. Soc. 2008, 130, 10150.
[18] G. Busca, Phys. Chem. Chem. Phys. 1999, 1, 723.
[19] P. Atkins, T. Overton, J. Rourke, M. Weller, F. Armstrong,
Inorganic chemistry,4th ed., Oxford University, Oxford, 2006,
p. 695.
[20] K. Neimann, R. Neumann, Org. Lett. 2000, 2, 2861.
[21] H. Yao, D. E. Richardson, J. Am. Chem. Soc. 2000, 122, 3220.
[22] a) D. Mandelli, M. C. A. van Vliet, R. A. Sheldon, U. Schu-
chardt, Appl. Catal. A 2001, 219, 209; b) P. P. Pescarmona,
K. P. F. Janssen, P. A. Jacobs, Chem. Eur. J. 2007, 13, 6562.
[23] a) V. R. Choudhary, S. K. Jana, B. P. Kiran, J. Catal. 2000, 192,
257; b) K. Nakagawa, M. Okamura, N. Ikenaga, T. Suzuki, T.
Kobayashi, Chem. Commun. 1998, 9, 1025; c) X. Wang, Q. Xu,
M. Li, S. Shen, X. Wang, Y. Wang, Z. Feng, J. Shi, H. Han, C. Li,
Angew. Chem. 2012, 124, 13266; Angew. Chem. Int. Ed. 2012, 51,
13089.
[11] K. Lin, P. P. Pescarmona, K. Houthoofd, D. Liang, G. Van Ten-
deloo, P. A. Jacobs, J. Catal. 2009, 263, 75.
[12] V. Hulea, E. Dumitriu, Appl. Catal. A 2004, 277, 99.
[13] a) S. Bordiga, A. Damin, F. Bonino, G. Ricchiardi, C. Lamberti,
A. Zecchina, Angew. Chem. 2002, 114, 4928; Angew. Chem. Int.
Ed. 2002, 41, 4734; b) S. Bordiga, F. Bonino, A. Damin, C.
Lamberti, Phys. Chem. Chem. Phys. 2008, 10, 4854.
[14] W. Lin, H. Frei, J. Am. Chem. Soc. 2002, 124, 9292.
[15] I. Arends, R. Sheldon, Top. Catal. 2002, 19, 133.
[16] a) R. Rinaldi, U. Schuchardt, J. Catal. 2004, 227, 109; b) G.
Stoica, M. Santiago, P. A. Jacobs, J. Pꢂrez-Ramꢃrez, P. P. Pescar-
mona, Appl. Catal. A 2009, 371, 43.
Angew. Chem. Int. Ed. 2014, 53, 1585 –1589
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1589