Selective aerobic oxidation of benzylic alcohols catalyzed by Carbon-Based catalysts: A nonmetallic oxidation system

Article abstract of DOI:10.1002/anie.200904362

"Chemical Equation Presented" Swapping places: The use of carbon to replace transition metals in the selective aerobic oxidation of benzylic alcohols has been demonstrated to be feasible. The HNO₃ additive is activated by the carbon catalyst NSC, which serves as the primary oxidant for alcohol oxidations and is regenerated by O₂. NSC=nanoshell carbon.

Full text of DOI:10.1002/anie.200904362

Communications
DOI: 10.1002/anie.200904362
Aerobic Oxidation
Selective Aerobic Oxidation of Benzylic Alcohols Catalyzed by
Carbon-Based Catalysts: A Nonmetallic Oxidation System**
Yongbo Kuang, Nazrul M. Islam, Yuta Nabae, Teruaki Hayakawa, and Masa-aki Kakimoto*
Selective oxidation is of great importance in organic syn-
thesis,[1] and green, efficient, selective oxidation techniques
are highly sought after in the chemical industry.[2] In
particular, selective oxidation of alcohols is a fundamental
transformation in the synthesis of fine chemicals.[3] Recently,
aerobic oxidation processes have received increasing atten-
tion, as oxygen or air are used as the terminal oxidant to
replace the stoichiometric metal oxides such as chromates and
manganese oxides. The use of oxygen has great benefits from
both economic and green chemistry viewpoints, because
oxygen is relatively cheap and produces water as the only by-
product. Accordingly, varied heterogeneous and homogenous
catalysts derived from precious metals have been developed
for this purpose, for example, Pt, Ru, Pd, Au, and Ag.[4]
However, because of their rarity and high price, precious
metals are impractical for industrial use. Despite this, few
efforts have been made to seek out nonprecious metal
catalysts. One example is the 2,2,6,6-tetramethylpiperidine-
1-oxyl (TEMPO)/Br2/NaNO2 aerobic oxidation system
reported by Liu et al.,[5] in which TEMPO was used as a
homogeneous catalyst, and the in situ generated NO played a
crucial role in the activation of O2. Our interests are in the
exploration of the potential of carbon-based materials to be
applied as metal-free catalysts for selective aerobic oxidation
of alcohols. Several studies of carbon catalyst materials have
already revealed the ability to replace metal-based catalysts in
several important transformations, such as the Friedel–Crafts
reaction[6] and the oxidative dehydrogenation of aromatic
hydrocarbons[7] and alkanes.[8] Interestingly, carbon catalysts
with an ordered nanoshell structure or doped with heter-
oatoms have been reported to possess surprisingly high
oxygen reduction reaction (ORR) activities.[9] However, to
date there is no report on the selective aerobic oxidation of
alcohols using the same catalyst. Herein, we present a novel
protocol: nitric acid assisted carbon-catalyzed oxidation
System (NACOS), by which benzyl alcohol and its substituted
derivatives can be smoothly oxidized by oxygen under mild
reaction conditions, affording high conversions and good
selectivities into the corresponding aldehydes.
The nanoshell carbon (NSC) used in this investigation was
prepared by pyrolization of a blend of phenol resin and iron
phthalocyanine, similar to methods reported by Ozaki
et al.,[10] and subsequent acid washes using concentrated
HCl. The iron metal on the carbon surface can be thoroughly
removed by acid washes,[10] and the metal-free surface was
characterized by energy dispersive X-ray spectroscopy (EDS)
(Figure 1a) and X-ray photoelectron spectroscopy (XPS;
Table 4). The specific surface area of NSC was 330 m2 gÀ1,
determined by N2 adsorption. Initial experiments with benzyl
alcohol showed that the NSC alone could not activate oxygen
and no alcohol conversion was observed. It was postulated
that the catalyst might take effect in the presence of an
additive, so several acid and base additives were tested.
Among all the tested additives, concentrated nitric acid
[*] Y. Kuang, Dr. N. M. Islam, Dr. Y. Nabae, Dr. T. Hayakawa,
Prof. M.-a. Kakimoto
Department of Organic and Polymeric Materials
Tokyo Institute of Technology
2-12-1 O-okayama, Meguro-ku, Tokyo, 152-8552 (Japan)
Fax: (+81)3-5734-2875
E-mail: mkakimot@o.cc.titech.ac.jp
Homepage: http://www.op.titech.ac.jp/lab/kakimoto/index.html
[**] We thank New Energy and Industrial Technology Development
Organization (NEDO) for generous support; Nisshinbo Holdings
Inc. Business Development Division for supply of NSC. We are
grateful to Dr. Stephen Lyth for helpful discussions.
Figure 1. a) Comparison of EDS spectra between acid washed and
unwashed nanoshell carbon. The Cu characteristic emission results
from the sample holder grid, not from the carbon catalysts. b) and
c) TEM images of NSC.
436
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 436 –440

Angewandte
Chemie
Table 2: Oxidation of varied primary and secondary benzylic alcohols.
proved to be the most effective. Subsequent studies suggest
that HNO3 plays a critical role in the activation of O2.
With concentrated nitric acid as an additive, the NSC was
initially investigated for the oxidation of benzyl alcohol at
908C, with O2 (1 atm) as the oxidant and 1,4-dioxane as the
solvent. The reaction proceeded smoothly under these con-
ditions, with a conversion of 96%. Notably, the selectivity for
aldehyde formation was 92%, with benzoic acid being the
main by-product (Table 1, entry 1). From the viewpoint of
Entry[a] Substrate
Product
t [h] Conv. [%][b] Sel. [%][b]
1
3
4
2
5
4
5
5
97
84
83
98
82
78
99
98
2
92
3
94
4
95
5[c]
6[c]
7
92[d]
98[e]
90
Table 1: Aerobic oxidation of benzyl alcohol to benzaldehyde with
carbon-based catalysts.
Entry
Catalyst
Solvent
t [h]
Conv. [%][b]
Sel. [%][b,f]
1
NSC
NSC
NSC
–
–
AC
wAC
XC72
VGCF
1,4-dioxane
–
–
1,4-dioxane
1,4-dioxane
1,4-dioxane
1,4-dioxane
1,4-dioxane
1,4-dioxane
5
6
10
5
5
5
5
5
5
96
65
97
<3
3
88
90
91
3
92
82
38
98
96
70
68
58
99
2[c]
3[d]
4
8
4
98
100
5[e]
6
7
8
9
[a] General conditions: substrate (0.5 mmol), NSC (10 mg), 1,4-dioxane
(2 mL), 67% HNO3 (1 mmol), 908C. [b] Conversion and selectivity were
determined by gas chromatography mass spectrometry methods using an
internal standard (naphthalene). [c] Substrate was 1 mmol and temper-
ature was 1008C. [d] Under general conditions, conversion and selectivity
were 76% and 91%, respectively. [e] Under general conditions, conversion
and selectivity were 58% and 100% respectively.
[a] General conditions: catalyst (10 mg), alcohol (0.5 mmol), 1,4-dioxane
(2 mL), 67%HNO3 (1 mmol), 908C. [b] Determined by gas chromatog-
raphy mass spectrometry methods with an internal standard (naphtha-
lene). [c] Alcohol (20 mmol), 67%HNO3 (2 mmol), solvent free. [d] Alco-
hol (20 mmol), 40%HNO3 (4 mmol), solvent free. [e] NaNO2
(0.1 mmol) added. [f] Benzoic acid was the main by-product.
the reaction media[12a,b] or with NaNO2 as the catalyst.[12c]
Therefore it could be postulated that nitric acid ultimately
serves as the oxidant in NACOS. However in the absence of
NSC, it was shown that nitric acid alone was not active and
was unable to catalyze the transformation (Table 1, entry 4).
With the addition of 10 mol% NaNO2 (relative to HNO3),
still no improved conversion was obtained (Table 1, entry 5),
which indicates clearly a different mechanism with the
NaNO2 catalyzed system.[12c] To demonstrate that O2 rather
than HNO3, is the ultimate oxidant, we tracked the HNO3
remaining after the reaction. As shown in Table 3, nitric acid
green chemistry,[11] it is desirable to avoid the use of organic
solvents, therefore the oxidation of benzyl alcohol was
evaluated under solvent-free conditions. For a shorter reac-
tion time (6 h), medium conversion was observed (65%) and
the selectivity remained high (82%) (Table 1, entry 2). For a
longer reaction time (10 h), a good conversion was obtained
(97%), however the selectivity for aldehyde dropped below
40% (Table 1, entry 3), with benzoic acid being the major
product.
Following the success of benzyl alcohol oxidation, we
applied NACOS to a series of benzylic alcohols having
substitutions ranging from electron-donating groups to elec-
tron-withdrawing groups (Table 2). For all these derivatives,
good conversions and selectivities could be easily obtained.
Generally, the substrates with electron-donating groups
showed higher activity. For example, 4-methoxylbenzyl alco-
hol and 4-phenylbenzyl alcohol had conversions of around
95% at 908C after three and two hours respectively (Table 2,
entries 1 and 3). In contrast, 4-nitrobenzyl alcohol resulted in
conversion of less than 60% after a five hour reaction at 908C.
However, excellent results could be achieved simply by
raising the temperature to 1008C (Table 2, entry 6). Secon-
dary benzylic alcohols were also tested, giving good yields as
expected (Table 2, entries 7 and 8).
Table 3: Comparison of recoveries of HNO3 in benzyl alcohol oxidations
conducted under oxygen and nitrogen atmosphere.
[a]
Entry alcohol:HNO3 Atmosphere HNO3
Conv. [%] Sel. [%]
Recovery [%]
1
2
3
1:2
1:2
10:1
O2
N2
O2
96
13
90
96
95
81
92
42
51
was almost all recovered after the reaction in which the ratio
of substrate to acid was 1:2 (Table 3, entry 1). Even when the
substrate and acid ratio was 10:1, over 90% of the acid was
recovered (Table 3, entry 3). These results clearly indicate
that HNO3 does not serve as the ultimate oxidant, but O2
does. To additionally clarify the role of HNO3, a well-designed
reaction conducted under a N2 atmosphere (closed system)
provided some interesting results (Table 3, entry 2). In the
absence of oxygen, around 90% of the HNO3 was consumed,
suggesting that in the presence of the catalyst, HNO3 itself is
To study the mechanism of NACOS, first efforts focused
on elucidating the role of the additive. Nitric acid is one of the
most commonly produced compounds in the chemical
industry and has been studied intensively as an oxidizing
agent for alcohol oxidation. These reactions were carried out
either with concentrated acid (sulphuric acid or nitric acid) as
Angew. Chem. Int. Ed. 2010, 49, 436 –440
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
437

Communications
able to oxidize alcohol, and serves as a stoichiometric oxidant
hydrocarbons[7] and alkanes.[8] Therefore it is likely that the
NSC surface is oxidized by HNO3 under our reaction
conditions and the surface oxygen-containing group plays
an important role in this type of reaction. To examine this
assumption, the used catalyst (uNSC) was collected after the
reaction and analyzed by XPS (Table 4). The surface oxygen
content was shown to be increased from 6% to 9.5%.
However, IR spectrum of uNSC does not show a significant
increase at 1730 cmÀ1 (Figure 3b), which is associated with
under these conditions. Moreover, brown gas was observed
above the mixture during the reaction, indicating the
formation and accumulation of NO2. As the reaction pro-
ceeded, the brown color faded gradually, probably as a result
of the reaction between NO2 and benzaldehyde, which
produces benzoic acid as by-product. This reaction explains
the poor selectivity (42%).
Since the activation of HNO3 does not involve O2, NO2
should also be generated in the presence of O2. However,
brown gas was not observed under O2 bubbling. So NO2 did
not accumulate in the gas phase, but quickly reacted.
Considering the high recovery of HNO3, NO2 must largely
react with O2 towards the regeneration of HNO3. Thus under
an O2 atmosphere, there is a HNO3 consumption/regenera-
tion cycle in which HNO3 plays a role as primary oxidant for
alcohol oxidation, and O2 acts as secondary oxidant for HNO3
recovery. Therefore, HNO3 serves as a catalytic mediator, and
to the best of our knowledge, this is the first example of the
utilization of an in situ HNO3 recycling system in organic
synthesis.
Table 4: [a] The reported ratio is a molar ratio.Table 4
XPS results of nanoshell carbon (NSC) and nanoshell carbon collected
after reaction (uNSC).
Sample
C [%]
N [%]
O [%]
Fe [%]
NSC
uNSC
93
89
0.7
0.9
6.0
9.5
n.d.
n.d.
n.d.=not determined.
Alcohol conversion was not observed when the HNO3
additive was replaced by HBr or NaNO3, all other reaction
parameters remaining the same. This result shows that the
hydrogen cation and the nitrate ion alone cannot catalyze the
reaction. Furthermore, reactions carried out for five hours in
a 1,4-dioxane/water (1:1 to 1:4) mixed solvent showed a
significant drop in conversion with increasing water content
(Figure 2). Recall that HNO3 has a large dissociation
Figure 3. a) Comparison of benzyl alcohol oxidations using NSC,
uNSC, and oNSC under the identical reaction conditions: 1,4-dioxane
(2 mL), benzyl alcohol (0.5 mmol), catalyst (10 mg), HNO3 (1 mmol),
908C, 1 hour. b) IR spectra of NSC, uNSC, and oNSC.
[13b]
=
the C O stretching in lactones and carboxylic anhydrides.
In contrast, oNSC, prepared by treating NSC with 5m HNO3
solution under refluxing for five hours, shows a very strong
peak at 1750 cmÀ1. Therefore, the increased surface oxygen
detected by XPS is probably a result of a sum of a small
degree of surface oxidation and increased physical adsorption
of oxygen containing species. To evaluate the effect of surface
oxidation, three parallel reactions were conducted simulta-
neously under the same conditions with NSC, uNSC, and
oNSC respectively, using a multireactor. However, the con-
versions were at the same level after one hour of reaction,
indicating same catalytic performances of three catalysts. So
increased surface oxidation does not lead to higher reaction
rate.
Figure 2. Comparison of benzyl alcohol oxidations conducted with
1,4-dioxane/water mixture at various ratios. Solid squares: conversion,
open squares: selectivity to aldehyde. Reaction conditions: 1,4-diox-
ane/water mixture (2 mL), benzyl alcohol (0.5 mmol), NSC (10 mg),
67% HNO3 (1 mmol), 908C, atmospheric pressure, 5 h.
constant, as the ratio of water increases in the solvent, H+
and NO3À ions become more abundant. Thus we suggest that
the H+ and NO3À ions have no catalytic activity and only the
pseudo-acidic form of HNO3 is active for this type of reaction.
In contrast, nitric acid has been widely employed in the
surface modification of carbon materials.[13] And recent
studies have revealed the surface carbonyl groups as the
active sites for the oxidative dehydrogenation of aromatic
On the basis the evidence presented herein, we propose
an overall mechanism for NACOS (Scheme 1). HNO3 is
firstly adsorbed onto the NSC surface and activated. The
activation process probably starts from the weakening of the
À
N O single bond of HNO3. Upon heating, a nearby alcohol
molecule is attacked by the activated HNO3 to yield the
438
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 436 –440

Angewandte
Chemie
Scheme 1. Overall mechanism of NACOS.
À
aldehyde product accompanied with the rupture of N O bond
of HNO3 and the release of NO2. Then NO2 reacts with
surrounding O2 and water to regenerate HNO3. Therefore
considering the whole catalytic cycle, the net effect is that the
alcohol is oxidized by O2.
As shown in entries 6–9 in Table 1, several other candi-
dates for the oxidation of benzyl alcohol were tested,
including activated carbon (AC), carbon black (XC72), and
carbon nanofiber (VGCF). AC and XC72 showed good
conversions of around 90% after five hours, but the selectivity
was worse than that with NSC. Since the NSC used in our
study was washed by concentrated HCl, we treated AC with
the same acid-washing process to see whether it has any effect
on the selectivity. As shown in entry 7 of Table 1, it seems to
not help in the improvement of the selectivity. The high
conversions of AC and XC72 hint at the generality of this
mechanism, and the broad applicability of this method. In
contrast, VGCF had a much lower conversion of 3%. The low
catalytic activity of VGCF is probably because of the low
surface area (13 m2 gÀ1) compared to AC and XC72 (1100 and
250 m2 gÀ1, respectively). In the case of NSC, the good
conversion can be explained by the high surface area
(330 m2 gÀ1), however the reason for the excellent oxidation
selectivity may require an additional mechanism. It has been
reported that the carbonization of polymeric materials in the
presence of iron can result in the formation of hollow shell-
like crystalline carbon structures, leading to an improvement
in oxygen reduction activity.[10] Such crystalline structures
were also observed in this work in transmission electron
microscope (TEM) images (Figure 1b,c). The Raman spectra
obtained for these four carbon structures provides additional
support (Figure 4). As expected, AC and XC72 show very
broad bands at 1600 cmÀ1 (G band), whereas VGCF gives a
very sharp peak and a shift from 1600 cmÀ1 to 1580 cmÀ1, due
to the formation of organized structure similar to the multi-
walled carbon nanotube. The spectrum for NSC has a sharper
peak and a slight shift from 1600 cmÀ1 to 1590 cmÀ1 compared
with AC and XC72, which has been reported to be signs of the
formation of ordered carbon structures.[10] These findings
indicate that NSC not only possesses large surface area
(330 m2 gÀ1) but also has a degree of crystallinity. Because the
active sites of the NSC are likely to be on the carbon atoms
along exposed zigzag edges of graphitic layers,[14] the crystal-
line feature of NSC may be partially responsible for the
improved selectivity compared with AC and XC72, which are
both amorphous carbon structures with a low degree of
crystallinity.
Figure 4. X-ray photoelectron spectra of the different carbon surfaces.
The peaks represent ordered carbon structures.
was elucidated. This newly discovered method of O2 activa-
tion may have more potential applications in line with the
notion of green chemistry. NACOS works at mild temper-
atures and at atmospheric pressure, using low cost and
environmentally benign reactants, offering great ease of
operation and high potential for scalability. These factors
make NACOS an extremely attractive process for industrial
application.
Additional work is now in progress to study the detailed
reaction mechanism and to expand the scope of this process.
Experimental Section
Energy dispersive X-ray spectroscopy (EDS) and transmission
electron spectroscopy (TEM) was performed using a JEOL JEM-
2010F transmission electron microscope (accelerating voltage
200 kV). X-ray photoelectron spectroscopy (XPS) analysis was
performed with a PHI Quantum 2000 spectrometer. IR measure-
ments were conducted with JEOL JIR-SPX200 spectrometer.
RAMAN spectra were recorded with JASCO NRS-2100 spectrom-
eter. The samples were analyzed with a SHIMAZDU GCMS-QP2010
Plus, using a Restek Rxi-1 ms column. An internal standard method
(naphthalene) was used for quantitative analysis.
Preparation of nanoshell carbon (NSC): 1.00 g of iron phthalo-
cyanine was blended with 3.725 g of phenol resin in 300 mL of acetone
in a glass flask. The materials were well dispersed by sonication and
then acetone was removed by evaporation. After drying the residue
under reduced pressure, the blend was heated at 8008C under N2 flow
for 1 h. The obtained material was pulverized by ball milling to obtain
small particles. The pulverized black powder was washed in concen-
trated hydrochloric acid (37%) for 12 h. The washed NSC was filtered
and thoroughly washed by deionized water, then dried in a vacuum
oven for 24 h at 808C.
Determination of HNO3 recovery: The resultant mixture was
collected by filtration using a 45 mm syringe filter. In a beaker, 1 mL of
the filtrate was diluted to 40 mL with deionized water. The solution
was stirred for several minutes and then filtrated to remove insoluble
substances. A pH meter was used to determine the pH value of the
solution. Based on the pH value, the recovery of HNO3 after the
reaction was calculated.
This study revealed the feasibility of replacing costly
precious metals in selective aerobic oxidation with cheap and
abundant carbon materials. An efficient nitric acid assisted
carbon-catalyzed oxidation system (NACOS) was estab-
lished. The activation and recovery mechanism of HNO3
Angew. Chem. Int. Ed. 2010, 49, 436 –440
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
439

Communications
General oxidation method: The oxidation reactions were con-
[5] R. H. Liu, X. M. Liang, C. Y. Dong, X. Q. Hu, J. Am. Chem. Soc.
2004, 126, 4112 – 4113.
[6] F. Goettmann, A. Fischer, M. Antonietti, A. Thomas, Angew.
chem. 2006, 118, 4579 – 4583; Angew. Chem. Int. Ed. 2006, 45,
4467 – 4471.
[7] N. Keller, N. I. Maksimova, V. V. Roddatis, M. Schur, G. Mestl,
Y. V. Butenko, V. L. Kuznetsov, R. Schlogl, Angew. Chem. 2002,
114, 1962 – 1966; Angew. Chem. Int. Ed. 2002, 41, 1885 – 1888.
[8] J. Zhang, X. Liu, R. Blume, A. H. Zhang, R. Schlogl, D. S. Su,
Science 2008, 322, 73 – 77.
ducted in a 25 mL three-necked flask. The flask was charged with
benzyl alcohol (54.1 mg, 0.5 mmol), NSC (10 mg), naphthalene
(12.8 mg, 0.1 mmol) as an internal standard, and 1,4-dioxane
(2 mL). After the mixture was sonicated for 20 s, 67.6 mL (1 mmol)
of 67% nitric acid was added dropwise into the reaction mixture.
Oxygen gas was bubbled through a glass capillary. The reaction
mixture was heated to 908C for 5 h. Samples were taken at
appropriate intervals through a silicon septum using a hypodermic
needle.
[9] a) J. Ozaki, T. Anahara, N. Kimura, A. Oya, Carbon 2006, 44,
3358 – 3361; b) J. Ozaki, N. Kimura, T. Anahara, A. Oya, Carbon
2007, 45, 1847 – 1853; c) P. H. Matter, E. Wang, M. Arias, E. J.
Biddinger, U. S. Ozkan, J. Mol. Catal. A 2007, 264, 73 – 81; d) S.
Maldonado, K. J. Stevenson, J. Phys. Chem. B 2005, 109, 4707 –
4716.
[10] J. I. Ozaki, K. Nozawa, K. Yamada, Y. Uchiyama, Y. Yoshimoto,
A. Furuichi, T. Yokoyama, A. Oya, L. J. Brown, J. D. Cashion, J.
Appl. Electrochem. 2006, 36, 239 – 247.
Received: August 5, 2009
Revised: November 9, 2009
Published online: December 8, 2009
Keywords: aerobic oxidation · alcohols · aldehydes · carbon ·
.
metal-free catalysis
[11] P. T. Anastas, M. M. Kirchhoff, Acc. Chem. Res. 2002, 35, 686 –
694.
[1] R. A. Sheldon, J. K. Kochi, Metal-Catalyzed Oxidations of
Organic Compounds, Academic Press, New York, 1981.
[2] Vision 2020 Catalysis Report, www.ccrhq.org/vision2020/cataly-
sis.
[3] G. Tojo, M. Fernandez, Oxidations of Alcohols to Aldehydes and
Ketones, Springer, New York, 2006.
[4] For reviews, see: a) T. Mallat, A. Baiker, Chem. Rev. 2004, 104,
3037 – 3058; b) B. Z. Zhan, A. Thompson, Tetrahedron 2004, 60,
2917 – 2935; c) M. J. Schultz, M. S. Sigman, Tetrahedron 2006, 62,
8227 – 8241; d) T. Matsumoto, M. Ueno, N. Wang, S. Kobayashi,
Chem. Asian J. 2008, 3, 196 – 214.
[12] a) D. S. Ross, C. Gu, G. P. Hum, R. Malhotra, Int. J. Chem. Kinet.
1986, 18, 1277 – 1287; b) P. Strazzolini, A. Runcio, Eur. J. Org.
Chem. 2003, 526 – 536; c) S. R. Joshi, K. L. Kataria, S. B. Sawant,
J. B. Joshi, Ind. Eng. Chem. Res. 2005, 44, 325 – 333.
[13] a) H. P. Boehm, Carbon 2002, 40, 145 – 149; b) J. L. Figueiredo,
M. F. R. Pereira, M. M. A. Freitas, J. J. M. Orfao, Carbon 1999,
37, 1379 – 1389; c) P. V. Lakshminarayanan, H. Toghiani, C. U.
Pittman, Carbon 2004, 42, 2433 – 2442.
[14] T. Ikeda, M. Boero, S. F. Huang, K. Terakura, M. Oshima, J.
Ozaki, J. Phys. Chem. C 2008, 112, 14706 – 14709.
440
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 436 –440