One-pot synthesis of menthol catalyzed by a highly diastereoselective Au/MgF2 catalyst

Article abstract of DOI:10.1002/anie.201002090

No toxic compounds such as KCN and no thermal activation is required for the preparation of Au/MgF₂ complexes by a simple and facile incipient wetness impregnation method in which hydrogen tetrachloroaurate (HAuCl ₄) is the gold precursor. One of the complexes prepared exhibits unique catalytic properties and serves as a heterogeneous catalyst for the highly diastereoselective one-pot synthesis of (±)-menthol from citronellal (see picture).

Full text of DOI:10.1002/anie.201002090

Communications
DOI: 10.1002/anie.201002090
Gold Catalysis
One-Pot Synthesis of Menthol Catalyzed by a Highly Diastereoselec-
tive Au/MgF Catalyst**
2
Alina Negoi, Stefan Wuttke, Erhard Kemnitz,* Dan Macovei, Vasile I. Parvulescu,
C. M. Teodorescu, and Simona M. Coman*
During the last two decades gold has attracted much attention
on account of its unique catalytic properties for numerous
[
1,2]
reactions.
It is generally accepted that the catalytic activity
of gold-based catalysts critically depends on the size of the
gold particles, the nature of the support material, the
[3–6]
preparation method, and the activation procedure.
Several
preparation methods have been developed that yield active
[7]
nanosized gold catalysts. Surprisingly, only little attention
has been paid to cationic gold species and their application in
catalytic reactions. Recently, it has been shown that isolated
Scheme 1. The synthesis of menthol from citronellal in two steps
(route 1A+2A) and in one pot (route B).
n+
and heterogenized Au ions (n = 1 or 3) are highly active in
reactions such as the selective hydrogenation of 1,3-buta-
[
8]
dienes to butenes, the homocoupling of phenylboronic
[
9]
acids, and the selective isomerization of epoxides to allylic
catalysts, they also exhibit excellent support properties (high
surface area, mesoporosity, high proportion of undercoordi-
nated sites ).
[10]
alcohols.
[
11]
More recently, we have showed that nanosized hydroxy-
lated fluoride materials (e.g., MgF2 (OH) ), which are
synthesized by a novel sol–gel synthesis
The synthesis of (Æ )-menthol in a single-step one-pot
Àx
x
11]
[
and exhibit
reaction (Scheme 1, route B) would be an attractive alter-
native, both from economic and environmental point of view,
to the industrial Takasago process, which involves the isomer-
ization of (+)-citronellal to (À)-isopulegol in the presence of
ZnBr2 dissolved in organic solvents, such as CH Cl and
interesting bi-acidic properties, can catalyze the cyclization
of citronellal to (Æ )-isopulegol (an intermediate in the
synthesis of (Æ )-menthol, Scheme 1, route 1A); the diaste-
reoselectivity of 91.7% is superior to that of most conven-
tional catalysts used for this reaction. This is a result of the
controlled introduction of defined amounts of OH groups on
the surface of nanoscopic metal fluorides; owing to the strong
electron-withdrawing effect of the dominating fluoride envi-
ronment, these Mg-OH groups are Brønsted acidic. By fine-
tuning the OH/F ratio, the catalytic properties can be
2
2
[
12]
benzene, or in an aqueous ZnBr solution, followed by its
2
hydrogenation to menthol in a separate second step on Raney
[14]
Ni (Scheme 1, steps 1A and 2A).
Even though several active and selective heterogeneous
catalysts for the cyclization of citronellal to isopulegols and
the one-pot synthesis of menthols from citronellal have been
reported, the diastereoselectivity in the synthesis of
(Æ)-isopulegol and (Æ )-menthol was generally lower than
[
11–13]
optimized for different catalytic syntheses.
nanoscopic metal fluorides are not only superior bi-acidic
These new
[
15–17]
that with homogeneous catalysts (52–76%).
There are
[18]
only a few exceptions. Corma and Renz showed that Sn-b-
zeolite catalyzes the cyclization of citronellal to (Æ )-isopule-
[
*] Dipl.-Chem. A. Negoi, Prof. Dr. V. I. Parvulescu,
Prof. Dr. S. M. Coman
Department Of Chemical Technology and Catalysis
Faculty of Chemistry, University of Bucharest
Bdul Regina Elisabeta 4-12, 030016 Bucharest (Romania)
Fax: (+40)214-100-241
[
19]
gol with a diastereoselectivity of 85%. Chuah et al. also
showed that over a Zr-b-zeolite catalyst, the cyclization of
citronellal occurs with 93% diastereoselectivity.
Therefore, we focused our attention on a simple synthesis
of a bifunctionalized metal fluoride catalyst that provides high
E-mail: simona.coman@g.unibuc.ro
Dr. S. Wuttke, Prof. Dr. E. Kemnitz
diastereoselectivity for the cyclization of citronellal to
Institut fur Chemie, Humboldt-Universitat zu Berlin
Brook-Taylor-Strasse 2, 12489 Berlin (Germany)
Fax: (+49)30-2093-7468
[12]
(
Æ)-isopulegol
and also catalyzes the hydrogenation of
(Æ)-isopulegol to (Æ )-menthol. We report herein the prep-
aration of a new gold/hydroxylated magnesium fluoride
catalyst (for simplicity the hydroxylated fluoride
E-mail: erhard.kemnitz@chemie.hu-berlin.de
Dr. D. Macovei, Dr. C. M. Teodorescu
National Institute of Materials Physics
Bucharest (Romania)
MgF2 (OH) will be referred to as simple fluoride, MgF )
Àx
x
2
by an “incipient wetness impregnation” method. This catalyst
provides unexpectedly active ionic gold species for the
selective hydrogenation of isopulegols to menthols. At the
same time, the catalytic features of fluoride responsible for
the diastereoselective isomerization of citronellal to
[
**] The authors thank the CNCSIS (PNCDI II 40/2007) for financial
support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002090.
8134
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 8134 –8138

Angewandte
Chemie
(
Æ)-isopulegol are preserved. In this way, the new bifunc-
tional catalyst can serve in the highly diastereoselective one-
pot synthesis of (Æ )-menthol from citronellal.
Iridium-based zeolites have been previously reported as
active
catalysts
for
the
one-pot
synthesis
of
[
17]
(Æ)-menthol. However, when metal fluorides were used
as supports, neither the deposition of iridium as a metal salt
nor its reduction generated active catalysts for the cyclization
of citronellal to (Æ )-isopulegol (cf. Table 2S in the Supporting
Information). The reduction of the iridium salt is accompa-
nied by extensive damage to the active acidic sites of the
metal fluoride support as a result of crystallization at the
3
+
temperature required for the reduction of Ir (4508C). Even
at lower temperatures (2008C) the amorphous fluorides
partially crystallize leading to a substantial decrease in
surface area and pore diameters. This is evidenced by XRD
and BET measurements (not shown here). Consequently, for
the synthesis of (Æ )-menthol the loss of acidic sites in
combination with the decrease in pore diameter resulted in a
drastic drop in catalytic activity and lowered diastereoselec-
tivity for
2
Figure 1. k -weighted EXAFS spectra of the Au catalysts and Au foil
À3
and the magnitude I (in ꢀ ) of the corresponding Fourier transforms.
contain chlorine, evinced by faint, but still visible, Cl 2p
photoemission peaks in the survey spectra (not shown here).
This prompted us to assign the maximum at 1.87 ꢀ to Cl
neighbors in preserved fragments of the precursor structure.
Further details regarding the fitting of the EXAFS spectra
can be found in the Supporting Information.
(
Æ)-isopulegol.
In order to preserve the high diastereoselectivity and
acidity of nanoscopic fluorides, it is necessary to either
optimize the reduction step in order to prevent crystallization
or to synthesize bifunctional catalysts that do not require
thermal activation. Consequently, we developed the synthesis
of gold-based nanoscopic fluoride catalysts.
The magnesium fluoride phases were prepared by the
reaction of magnesium methoxide with a methanolic hydro-
gen fluoride solution. By employing HF solutions with
different water content, it is possible to control the introduc-
The Au environment in the sample Au-100 (3.6 Cl atoms
at 2.281 ꢀ) closely resembles that in the tetrachloroauric acid
[
21]
structure (6 Cl atoms at 2.286 ꢀ). This indicates that the
structure of the precursor is preserved after thermal treat-
ment at 1008C. The reduced number of Cl neighbors in the
structure of the sample Au-100 indicates a Cl-defective
structure of the precursor, and/or also small precursor
particles, which effectively lower the coordination number
tion of OH groups. If the OH content in the MgF2 (OH)x
Àx
[
22]
phases is very low (x < 0.1), these OH groups are Brønsted
derived from EXAFS.
[
20]
acidic in nature. As a result, the bi-acidic catalysts obtained
Upon increasing the treatment temperature to 1508C, a
large fraction of gold is reduced to the metallic state. This was
already shown by the FT of the sample Au-150, which is a
combination of the FTs corresponding to the sample Au-100
and the metallic Au (Figure 1). EXAFS fitting indicates Cl
and Au around the Au atoms, corresponding to the remaining
precursor and the developing metallic phase, respectively. For
a phase mixture, the coordination numbers (CNs) specific to
each component are weighted in the EXAFS analysis by the
fraction of that component in the phase composition. In this
case, CNAu = 6 corresponds to about one-half of the Au atoms
in metallic state, while the decrease of CNCl from 3.6 to 1.3
would indicate about one-third of Au still belonging to the
precursor. It is very probable that this discrepancy results
from a greater underestimation of CNCl for the sample
Au-150, with smaller particles of the precursor remnants, than
the sample Au-100. The change of the chemical state of gold
with the treatment temperature is also shown by the “white
2
+
exhibited both Lewis (Mg ) and Brønsted (OH) sites.
This material was treated with the incipient wetness
impregnation method using hydrogen tetrachloroaurate as
the gold precursor. The catalyst precursors were calcined at
1
008C and 1508C, and the resulting materials were denoted
Au-100 and Au-150. As evidenced by inductively coupled
plasma/atom emission spectrometry (ICP-AES), complete
impregnation of the mesoporous MgF co-catalyst with the
2
gold compound was achieved without any loss of gold during
the preparation procedure; the final concentration was 4.0%
Au. Furthermore, partial hydrolysis of the tetrachloroauric
acid (evidenced by a blueshift in UV/Vis spectrum) results in
negatively charged gold complexes that are strongly absorbed
onto the positively charged MgF2 surface (see also the
Supporting Information).
2
The k -weighted EXAFS spectra of the Au catalysts and
Au foil, together with their Fourier transforms (FT) are shown
in Figure 1. For the Au foil, the main split maximum of FT
corresponds to the nearest neighbors (12 atoms at 2.884 ꢀ) in
the fcc structure of the metal. The FT of the sample Au-100
shows a main maximum at 1.87 ꢀ, in line with closer
neighbors of Au. The same maximum is visible for the
sample Au-150, together with the split maximum character-
istic of metallic gold. As indicated by XPS, both samples
line” (WL) of the Au L -edge absorption spectra (Figure 2S in
the Supporting Information).
3
The formation of high agglomerations of gold particles
was also observed in XRD diffraction spectra for the Au-150
sample. The same XRD technique showed that the amor-
phous state of the pure nanoscopic fluoride was successfully
preserved. However, the formation of agglomerations of gold
Angew. Chem. Int. Ed. 2010, 49, 8134 –8138
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8135

Communications
[
27]
particles was expected, since chloride ions promote the
mobility and agglomeration of gold species during thermal
catalysts generated higher amounts of these by-products.
An important effect of Au-100 catalyst is the complete
suppression of these by-products. This behavior probably
corresponds to the blockage of the strongly acidic sites during
the impregnation with the gold precursor. By comparing the
Au-100 gold catalyst with Ir-based catalysts, it was found
that the former is totally selective to menthols (no by-
products such as citronellol, 3,7-dimethyloctanol, and
[
23]
treatment. In accordance with the literature, such agglom-
erations of the metallic gold particles lead to low-index gold
[
24]
surfaces which are known to be inert and inactive towards
most molecules. Such behavior is also confirmed in the one-
pot synthesis of menthol from citronellal (Table 1).
[
17]
3
,7-dimethyloctanal were observed) suggesting a different
Table 1: Comparison of MgF and the gold-based catalysts in terms of
conversion (X), overall selectivity (S), and diastereoselectivity (ds).
2
[
a]
reaction mechanism.
The hydrogenation of the C=C bond of isopulegol
Sample
X [%]
Sisopulegols [%]
dsmenthols [%]
requires the activation of molecular hydrogen by the sup-
ported gold species (Scheme 2). In principle, this activation
[
b]
MgF2
95.0
99.0
0.5
99.0
99.0
87.0
57.0
0
39.2
7.5
–
43.0
0
60.8
92.5
Au-100
Au-150
Au-100
Au-100
[
[
c]
d]
[a] Reaction conditions: 100 mg catalyst, 1.0 mL (860 mg) citronellal,
5
mL toluene, 808C, 15 atm H , 22 h. [b] The cyclization of citronellal to
2
isopulegol: 100 mg catalyst, 1.0 mL citronellal, 5 mL toluene, 808C, 6 h.
[c] The second catalytic charge. [d] The third catalytic charge.
Interesting is the fine line between a highly active and
diastereoselective catalyst (Au-100) and a totally catalytically
inactive material (Au-150), imposed by a difference of only
5
08C in the calcination temperature (Table 1). Therefore, as
Table 1 lines 1 and 2 show, the ds of the fluoride support was
preserved after its impregnation with gold salt. The Au-100
catalyst is as active as the MgF sample in the isomerization of
2
[12]
citronellal to (Æ )-isopulegol; high conversion (99.0%) is
also reached after 6–7 h, but the hydrogenation of the formed
Scheme 2. Proposed catalytic cycle.
(
Æ )-isopulegols to (Æ )-menthol is much slower.
After reaction time of 22 h the selectivity to
Æ)-menthol reaches 43%, and the diastereoselectivity in
a
(
may occur by two routes: the homolytic activation or the
heterolytic cleavage of molecular hydrogen. Homolytic
activation is highly improbable because it is difficult for the
the formation of the isopulegols in the first step is preserved
(
Table 1, entry 2). The (Æ )-isopulegol/(Æ)-neo-isopulegol
1
V
ratio is 87.8:12.2, according to GC and H NMR analysis
metallic species to reach the oxidation state Au . Therefore,
(
Figures 4S and 5S in the Supporting Information). The
the heterolytic cleavage of H2 is more probable and in
agreement with the previous work of Comas-Vives et al.
1
[28]
H NMR data are in perfect agreement with the literature
[25]
data. When the catalyst was removed by filration, dried in
air at room temperature, and reintroduced for another 16 h,
the selectivity to menthol improved by another 17.8%,
reaching a level of 60.8%. This can be explained by the re-
oxidation of the gold particles to the Au oxidation state
upon contact with air; this has been confirmed by XPS and is
in agreement with previous results. Based on this observa-
tion, the reaction was stopped after 15 h and the catalyst was
exposed to air. As a consequence of re-oxidation the
selectivity to (Æ )-menthol increased to 92.5%. This proce-
dure was repeated and the selectivity to (Æ )-menthol was
maintained in the range of 91–93%.
These authors proved that the heterolytic cleavage of hydro-
gen requires a polar environment provided by the solvent.
We have found that a highly polar heterogeneous catalyst
can be used instead of homogeneous catalysis in a polar
solvent. In the presence of the Au-100 catalyst, the hydrolytic
3
+
activation of hydrogen is favored by the MgF support; the
2
[
26]
proton remains on the polar surface whereas the hydride is
bonded to the metal center. Thus this novel catalyst can be
used to perform such reactions in nonpolar solvents like
III
I
toluene. However, the reduction of Au to Au accompanies
I
the hydrogenation cycle. This reduction to the in active Au
species occurs through electron donation from the hydride
with the formation of hydrogen radical, which recombines to
form the molecular hydrogen (Scheme 2).
The pure MgF support catalyzed only the cyclization of
2
citronellal to isopulegol with a conversion of 95% and a
selectivity of 87% (Table 1). Several dehydration products
(
etherification of isopulegol to various isomeric isopulegol
ethers also occurred, and these underwent further dehydra-
tion and cracking (see the Supporting Information). How-
ever, the presence of stronger acidic sites than those in our
III
I
In order to regenerate the active Au species, Au must be
reoxidized. This oxidation can occur either through the
C H ) of isopulegol were identified. In addition, the
10 16
+
generation of H in the media or more effectively by air.
This mechanism is less probable when the Au-150 catalyst;
0
the Au sites present are not active since they cannot
participate in a spontaneous electron-transfer process.
8136
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 8134 –8138

Angewandte
Chemie
To compare the above results we used the same procedure
to prepare bifunctional catalysts with different loadings of
gold and with b-zeolite as the support. The catalysts obtained
displayed high activity (conversions higher than 91%) for the
cyclization of citronellal to (Æ )-isopulegol, which indicates
that the acidic properties of the b zeolite are conserved.
However, when these catalysts were used under conditions
similar to those applied with the Au/fluoride catalysts, the
selectivity to menthol was very low (maximum 31.0% for the
catalysts and a Au foil standard were recorded in transmission mode
at the CEMO beamline of the HASYLAB synchrotron radiation
facility (Hamburg, Germany). Further details regarding calculation of
the EXAFS spectra can be found in the Supporting Information. The
thermogravimetric analysis was conducted with a SDT Q600 instru-
ment supplied by TA Instruments. The samples were placed in an
À1
aluminum sample holder and heated at a rate of 108Cmin from
room temperature to 3008C under a N flow with a rate of
2
À1
100 mLmin . UV/Vis spectra were recorded using the diffuse
reflectance technique. The spectrometer was a Specord 250 (Analytik
Jena).
4
.0% Au/b-zeolitecatalyst).
Activity tests were carried out in pressurized reactors as
Such results support the unique behavior of nanoscopic
described in the following procedure: Racemic citronellal
hydroxylated fluorides, not only as potential catalytic materi-
als, but also as potential supports for active catalytic functions.
In conclusion, we have described the synthesis of a
bifunctional ionic gold/magnesium fluoride catalyst using an
easy and facile incipient wetness impregnation method.
Moreover, this procedure did not require additional treat-
ment with highly toxic compounds such as KCN or high-
temperature activation steps, as usually needed for the
generation of isolated cationic gold sites. The obtained
material is able to catalyze both steps of the synthesis of
À1
(
d=0.855 gmL , 1.0 mL, 5.6 mmol, 860 mg) was dissolved in 5 mL
of toluene. The catalyst (100 mg) was added to this mixture. After the
reactor was closed, it was pressurized with H (15 bar), and heated up
2
to 808C, and the reaction mixture was stirred for 22 h. After the
reaction the catalyst was filtered off and the reaction mixture was
1
13
analyzed by GC chromatography and H and C NMR spectroscopy.
The diastereoselectivity (ds) is the selectivity for (Æ )-isopulegol with
respect to the other isopulegol isomers.
The separated catalyst was kept in air at room temperature until
dried. The dried powder was transferred into the pressurized reactor
with the reaction product from the first charge and kept under the
same reaction conditions for another 22 h. After the second reaction
the product was separated, analyzed, and used in a third charge with
the same catalyst.
(
Æ)-menthol—cyclization of citronellal to (Æ )-isopulegol and
hydrogenation of (Æ )-isopulegol—resulting for the first time,
in a diastereoselective one-pot heterogeneous synthesis of
(
Æ)-menthol. Our results may have important implications
Received: April 8, 2010
Published online: September 20, 2010
for the synthesis of other gold/porous material catalysts and
their application as bifunctionalized catalysts. In addition, the
presented Au/MgF catalyst may also be applied to other
reactions requiring both acid and hydrogenation activity.
2
Keywords: diastereoselective synthesis · gold catalysts ·
hetereogeneous catalysis · magnesium fluoride · menthol
.
Experimental Section
[1] G. C. Bond, D. T. Thompson, Catal. Rev. Sci. Eng. 1999, 41, 319 –
388.
[2] F. Z. Su, Y. M. Liu, L. C. Wang, Y. Cao, H. Y. He, K. N. Fan,
Angew. Chem. 2008, 120, 340 – 343; Angew. Chem. Int. Ed. 2008,
47, 334 – 337.
[3] M. Valden, X. Lai, D. W. Goodman, Science 1998, 281, 1647 –
1650.
[4] A. Grirrane, A. Corma, H. Garcia, Science 2008, 322, 1661 –
1664.
The nanoscopic fluoridew were synthesized as reported else-
[
11,29,30]
where
using aqueous 71 wt% HF solution. XRD, MAS NMR
spectroscopy, TEM, thermal analysis, and elemental analysis were
applied to study the structure, composition, and thermal behavior of
the bulk material. XPS measurements, FTIR with probe molecules,
and the determination of N and Ar adsorption–desorption isotherms
2
[
11]
were carried out to investigate the surface properties.
Ir-based catalysts were prepared by incipient wetness impregna-
tion using [Ir(acac) ] as the metal precursor, and H-b-zeolite
[5] A. Wolf, F. Schꢁth, Appl. Catal. A 2002, 226, 1 – 13.
[6] M. D. Hughes, Y.-J. Xu, P. Jenkins, P. McMorn, P. Landon, D. I.
Enache, A. F. Carley, G. A. Attard, G. J. Hutchings, F. King,
E. H. Stitt, P. Johnston, K. Griffin, C. J. Kiely, Nature 2005, 437,
1132 – 1135.
3
2
À1
(
Sꢁdchemie, SiO /Al O = 21.6, BET-surface area = 739 m g ) as
2 2 3
the support. Further details regarding the catalystsꢂ preparation can
be found in the Supporting Information.
The gold-based MgF samples were prepared by incipient wetness
2
impregnation. The amount of solid HAuCl needed for 4.0 wt% Au in
[7] S. Biella, L. Prati, M. Rossi, J. Catal. 2002, 206, 242 – 247.
[8] X. Zhang, H. Shi, B.-Q. Xu, Angew. Chem. 2005, 117, 7294 –
7297; Angew. Chem. Int. Ed. 2005, 44, 7132 – 7135.
[9] S. Carrettin, J. Guzman, A. Corma, Angew. Chem. 2005, 117,
2282 – 2285; Angew. Chem. Int. Ed. 2005, 44, 2242 – 2245.
[10] C. Raptis, H. Garcia, M. Stratakis, Angew. Chem. 2009, 121,
3179 – 3182; Angew. Chem. Int. Ed. 2009, 48, 3133 – 3136.
[11] S. Wuttke, S. M. Coman, G. Scholz, H. Kirmse, A. Vimont, M.
Daturi, S. L. M. Schroeder, E. Kemnitz, Chem. Eur. J. 2008, 14,
11488 – 11499.
4
the final material was dissolved in a volume of pure water
corresponding to the pore volume of the support. The impregnation
solution was added dropwise to the support during intensive stirring.
After the addition of the solution was complete, the support was
slightly damp.
The resulting catalyst precursor was divided in two parts and
dried for 4 h at either 1008C or 1508C under vacuum (Au-100 and Au-
1
50). The same procedure was applied using H-b-zeolite as the
support and the amount of HAuCl needed for a finding loading of
4
0
.5, 1.0, 3.0, 4.0, and 5.0 wt% Au.
[12] S. M. Coman, P. Patil, S. Wuttke, E. Kemnitz, Chem. Commun.
2009, 460 – 462.
The catalysts were characterized by numerous conventional
techniques such as ICP-AES, XRD, XPS, EXAFS, TG-DTA, and
UV/Vis spectroscopy. X-ray diffraction was recorded by using a
Siemens D5000 X-ray diffractometer with nickel-filtered CuKa
radiation (l = 1.5418 ꢀ) at a scanning rate of 0.1 min in the 2q
range of 10–808. The local Au environment was investigated by
[13] S. M. Coman, V. I. Parvulescu, S. Wuttke, E. Kemnitz, Chem-
CatChem 2010, 2, 92 – 97.
[14] Y. Nakatani, K. Kawashima, Synthesis 1978, 147 – 148.
[15] C. Milone, A. Parri, A. Pistone, G. Neri, S. Galvagno, Appl.
Catal. A 2002, 233, 151 – 157.
À1
EXAFS spectroscopy at the Au L edge. The absorption spectra of the
[16] K. Arata, C. Matsuura, Chem. Lett. 1989, 1797 – 1798.
3
Angew. Chem. Int. Ed. 2010, 49, 8134 –8138
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8137

Communications
[
17] F. Nea t¸ u, S. Coman, V. I. Pꢃrvulescu, G. Poncelet, D. de Vos,
[25] K. H. Schulte-Elte, G. Ohloff, Helv. Chim. Acta 1967, 50, 153 –
165.
P. A. Jacobs, Top. Catal. 2009, 52, 1292 – 1300.
[
[
18] A. Corma, M. Renz, Chem. Commun. 2004, 550 – 551.
19] Z. Yongzhong, N. Yuntong, S. Jaenicke, G. K. Chuah, J. Catal.
[26] F. Nea t¸ u, Z. Li, R. Richards, P. Y. Toullec, J.-P. GenÞt, K.
Dumbuya, J. M. Gottfried, H.-P. Steinrꢁck, V. I. Pꢃrvulescu, V.
Michelet, Chem. Eur. J. 2008, 14, 9412 – 9418.
2005, 229, 404 – 413.
[
20] S. Wuttke, S. M. Coman, J. Krꢄhnert, F. C. Jentoft, E. Kemnitz,
Catal. Today 2010, 152, 2 – 10.
[27] G. K. Chuah, S. H. Liu, S. Jaenicke, L. J. Harrison, J. Catal. 2001,
200, 352 – 359.
[
21] D. E. OꢂReilly, E. M. Peterson, C. E. Scheie, J. M. Williams, J.
Chem. Phys. 1971, 55, 5629 – 5635.
[28] A. Comas-Vives, C. Gonzalez-Arellano, A. Corma, M. Iglesias,
F. Sanchez, G. Ujaque, J. Am. Chem. Soc. 2006, 128, 4756 – 4765.
[29] S. Rꢁdiger, U. Groß, E. Kemnitz, J. Fluorine Chem. 2007, 128,
353 – 368.
[
22] R. B. Greegor, F. W. Lytle, J. Catal. 1980, 63, 476 – 486.
23] H. H. Kung, M. C. Kung, C. K. Costello, J. Catal. 2003, 216, 425 –
[
4
32.
[30] S. M. Coman, S. Wuttke, A. Vimont, M. Daturi, E. Kemnitz,
Adv. Synth. Catal. 2008, 350, 2517 – 2524.
[
24] B. Hammer, J. K. Nørskov, Nature 1995, 376, 238 – 240.
8138
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 8134 –8138