Copper nanoparticles supported on doped graphenes as catalyst for the dehydrogenative coupling of silanes and alcohols

Article abstract of DOI:10.1002/anie.201405669

Copper nanoparticles (NPs) supported on a series of undoped and doped graphene materials (Gs) have been obtained by pyrolysis of alginate or chitosan biopolymers, modified or not with boric acid, containing Cu²⁺ ions at 900 °C under inert atmosphere. The resulting Cu-G materials containing about 17 wt% Cu NPs (from 10 to 200 nm) exhibit high catalytic activity for the dehydrogenative coupling of silanes with alcohols. The optimal material consisting on Cu-(B)G is more efficient than Cu NPs on other carbon supports.

Full text of DOI:10.1002/anie.201405669

Angewandte
Chemie
DOI: 10.1002/anie.201405669
Graphene in Catalysis
Copper Nanoparticles Supported on Doped Graphenes as Catalyst for
the Dehydrogenative Coupling of Silanes and Alcohols**
Juan F. Blandez, Ana Primo,* Abdullah M. Asiri, Mercedes ꢀlvaro, and Hermenegildo Garcꢁa*
Abstract: Copper nanoparticles (NPs) supported on a series of
undoped and doped graphene materials (Gs) have been
obtained by pyrolysis of alginate or chitosan biopolymers,
modified or not with boric acid, containing Cu2+ ions at 9008C
under inert atmosphere. The resulting Cu-G materials con-
taining about 17 wt% Cu NPs (from 10 to 200 nm) exhibit
high catalytic activity for the dehydrogenative coupling of
silanes with alcohols. The optimal material consisting on Cu-
(B)G is more efficient than Cu NPs on other carbon supports.
stituting the physical limit for 2D materials. G as support
offers unique properties. Specifically, G has a very large
specific surface area of about 2500 m2 gÀ1 totally accessible
and the extended p orbitals on the G can interact strongly
with the empty d orbitals of transition metal atoms at the
interface. In addition, G can also adsorb substrates and
reagents, bringing them near the active metal NPs. It has been
observed that the activity of the metal NPs on G can be higher
than on other supports because of the combination of the
above effects.[20,21]
A
lcoxysilanes are high added-value commodity reagents for
A third aspect of the present work is the innovative
procedure reported here for the simultaneous preparation of
Cu NPs on G by carbonization of natural biopolymers. The
procedure also allows doping or co-doping of the G sheet with
heteroatoms. The interaction of metal NPs with heteroatoms
on doped G has been proposed from theoretical calculations
as a tool to control the activity of G.[22] The resulting graphitic
material, after exfoliation, exhibits high activity for the
dehydrogenative coupling of silanes with alcohols
(Scheme 1).
surface coating and modification[1–3] and for preparation of
hybrid organic–inorganic materials.[4–6] One possibility for the
synthesis of these reagents is the use of supported gold
nanoparticles (NPs) as heterogeneous catalysts for the
dehydrogenative coupling of silanes with alcohols
(Scheme 1).[7–9]
The preparation method for the series of Cu-G materials
is summarized in Scheme 2. It starts with an aqueous solution
Scheme 1. Catalytic dehydrogenative o-silylation of alcohols by hydro-
silanes.
One line of research of continued interest aims at
replacing gold and precious metals by abundant and afford-
able base metals.[10–14] In this context, it is important to
determine if the reaction shown in Scheme 1 can also be
catalyzed by copper.
In addition to determine the catalytic activity of Cu,
another point of this study is the use of graphene (G) as
support. Carbonaceous materials are widely used as supports
of metal NPs.[15,16] G is becoming increasingly used in catalysis
as support of metal NPs.[17–19] G is a one-carbon-atom-thick
layer of sp2 carbon atoms in hexagonal arrangement, con-
Scheme 2. Preparation procedure for the present Cu-G materials. The
polysaccharide can be alginate or chitosan. The dopant salt can be
borate. 1) Gelification of biopolymers by Cu2+, 2) pyrolysis at 9008C in
Ar 3) graphitization and 4) ultrasonication.
of biopolymer (alginate or chitosan) modified or not by boric
acid that is precipitated with a Cu2+ salt. The resulting Cu2+-
modified biopolymer hydrogel is dried and submitted to
pyrolysis at 9008C under inert atmosphere to render a Cu-
containing graphitic carbon residue. During the pyrolysis
under reductive conditions, Cu2+ undergoes spontaneous
reduction to Cu0 forming Cu NPs, simultaneously to graphi-
tization of the biopolymer. Pyrolysis of alginate and other
natural polysaccharides form graphitic carbon residues that
can be easily exfoliated into G sheets.[23–25] In the case of
chitosan, a polysaccharide of glucosamine, the resulting
graphitic residue contains 7 wt% of N, according to combus-
tion analysis. If the biopolymer contains boric acid, then, the
resulting graphitic residue becomes doped with boron.[23]
Sonication of the Cu-containing graphitic carbon residue
[*] J. F. Blandez, Dr. A. Primo, Prof. M. ꢀlvaro, Prof. H. Garcꢁa
Instituto de Tecnologꢁa Quꢁmica and
Chemistry Department (UPV-CSIC)
Avda/de los Naranjos s/n 46022 Valencia (Spain)
E-mail: hgarcia@qim.upv.es
Prof. A. M. Asiri, Prof. H. Garcꢁa
Center of Excellence for Advanced Materials Research
King Abdulaziz University, Jeddah (Saudi Arabia)
[**] Financial support by the Spanish Ministry of economy and
competitiveness (Severo Ochoa and CTQ2012-32315) and Gener-
alitat Valenciana Prometeo 2012-013) is gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201405669.
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12581

.
Angewandte
Communications
leads to suspended Cu-G and Cu-doped G. The list of
materials prepared and their main analytical data are
presented in Table S1 (see the Supporting Information).
The obtained Cu-graphitic carbon atoms were character-
ized by chemical analysis, XRD, and electron microscopy. The
Cu content on the materials was determined by ICP-AAS
(see Table S1).
Because of the high Cu percentage and its crystallinity
arising from the high pyrolysis temperature, the samples Cu-
graphitic carbon exhibit an intense X-ray diffractogram
(Figure S9), corresponding to the (1.0.0), (2.0.0), and (2.2.0)
Cu facets, without observation of any peak of graphitic
carbon.
Figure 1. Dark field scanning TEM image (left; scale bar=200 nm)
and TEM image of Cu-G (right; scale bar=100 nm).
Sonication in n-butanol of the Cu-graphitic carbon atoms
leads to persistent inks where graphitic platelets were
suspended. After removal the residual non/dispersed solid
residue (typically about 20% of the initial weight), the
suspended material was characterized by microscopy. TEM
shows a 2D morphology of micrometric length with the
presence of wrinkles because of the flexibility of the G layers.
The presence of Cu-NPs was also observed (Figure 1).
Statistical analysis of the size distribution indicates that
Cu NPs show a broad range of particle sizes from 10 to
200 nm (Figures S1–S4). The average particle diameter of
Cu NPs for Cu-G was 24 Æ 7 nm. It was found that the
G samples containing nitrogen as
dopant elements is reflected in Raman spectroscopy by
increasing D band intensity.
X-ray photoelectron spectroscopy (XPS) determines the
composition of the Cu NPs supported on G and the distribu-
tion of each element in different oxidation states. Depending
on the sample, the C1s peak can be deconvoluted in three or
four component, the major one appearing at 284.5 eV
characteristic of sp2 G carbon atoms. Other components of
C1s appear at higher binding energy and correspond to
C atoms bonded to the dopant element and oxygen. C atoms
dopant element exhibit larger average
size Cu NPs. This range of Cu particle
dimension is remarkable considering
that the samples were obtained by
pyrolysis at 9008C. It is very likely that
the biopolymers and the carbon resi-
dues derived from them, control the
growth of Cu NPs during the pyrolysis
as it has been proposed for metal
oxides.[26–29] Cu NPs appear well-dis-
persed on G having a round morphol-
ogy. In addition the presence of nano-
metric holes on the sheet can also be
observed (see Figure 1). These holes
could be formed by CO2 evolution
during pyrolysis and could play a role
on the stabilization of metal NPs and
on the catalytic activity of Gs.[30]
The single layer morphology for
some platelets was determined by
AFM. Figure 2 shows a view of some
of these platelets for Cu-(B)G, with
the count of the average height of
Cu NPs from 10 to 25 nm and the
vertical height corresponding to
a single layer (B)G.
Raman spectroscopy of Cu NPs
supported on graphitic carbon resi-
Figure 2. AFM images of Cu-(B)G (a and b). Panel (a) presents a top AFM view of Cu-(B)G with
a 2.5ꢂ2.5 mm2 area where the height of the Cu NPs was measured. Panel (b) presents higher
magnification of the (B)G sheet (lower part of the image) at the border and the dashed line
indicates the points through which the vertical height of the sheet was measured. Panel (c) shows
the height distribution (x axis) of Cu NPs inside the square delimited in panel (a). Panel (d) shows
dues shows the expected 2D, G, and
D peaks at 2700, 1590, and 1350 cmÀ1
,
respectively. The intensity ratio of the
G and D peaks varied from 1.3 to 1.6.
It should be noted that the presence of the height variations along the white line in panel (b).
12582 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586

Angewandte
Chemie
sizes in Figures S1–S4). Under optimal con-
ditions measured for an experiment using
ten times higher amount of 1a and 2a
a minimum turnover number (TON) of
2000 with a turnover frequency (TOF) of
4.16 hÀ1 was obtained for Cu-(B)G. The
initial reaction rate of Cu-G was higher
than that of Cu-(B)G, but the conversion of
1a in the presence of Cu-G levels off earlier
than for Cu-(B)G. One explanation is faster
deactivation of Cu-G compared to Cu-(B)G.
Figure 3. XPS peaks recorded for the Cu-(B)G sample and the best deconvolution into
individual components.
associated to dopant elements or at defects, are in a propor-
Table 1: Dehydrogenative coupling of dimethylphenylsilane (1a) and n-
butanol (2a) in the presence of Cu-containing catalysts.[a]
tion that agrees with the percentage of dopant element.
Table S1 summarizes the percentage for each component in
the corresponding XPS peaks of each sample.
As expected, depending on the sample, the presence in
XPS of the peak corresponding to the dopant element, either
boron, nitrogen or both was also detected. Figure 3 shows the
XPS peaks corresponding to the three elements present in
Cu-(B)G (see also Figures S12 and S13 for additional XPS).
The intensity of the XPS peak for each element and the
corresponding response factor allowed determination of the
elemental composition of the sample. In addition, the 2p3/2
peak corresponding to Cu was also observed for all the
samples with very similar binding energy. Deconvolution of
the Cu2p3/2 peak allows the determination of the atomic
proportion of CuII with respect to the combined amount of
Cu0 and CuI. Overall, the above characterization provides
firm evidence of the success on the formation of Cu NPs
simultaneously with the synthesis of doped Gs. Thus, chemical
analysis and XPS detect the presence of the expected
elements, while the morphology and crystallinity of G were
determined by microscopy. AFM shows that sonication
produces exfoliation of the graphitic carbon atoms to single-
or few-layer platelets.
Entry Catalyst
Conditions [8C] Conversion [%][b] Selectivity [%][b]
1
2
3
4
5
6
7
8
9
Cu-G
Cu-(B)G
Cu-G
Cu-(B)G
Cu-(B,N)G 80
Cu-G
Cu-(B)G
Cu-(B,N)G 100
Cu-(N)G 100
sonication, RT 28
sonication, RT 33
32[c]
21[c]
93
80
80
74
47
92
47
87
100
100
80
95
100[c]
51
100[d]
43[c]
80
65
[a] Reaction conditions: 1a (5 mmol), 2a (10 mmol), catalyst
(0.5 mol%), Ar atmosphere, 24 h. [b] Conversion and selectivity were
determined by GC using internal standard. [c] Incomplete mass balance
because of adsorption is responsible for the low selectivity. [d] 48 h.
The purpose of this work is to study the catalytic activity
of Cu NPs supported on Gs. Besides showing that Cu NPs are
active for the dehydrogenative coupling of silane and
alcohols, another aspect is to show how doping alters the
activity of G.
In the initial stage of our work we selected as model
reaction the coupling of dimethylphenylsilane (1a) with n-
butanol (2a). A preliminary screening to establish the
efficiency of the series of Cu-containing catalysts was carried
out at various temperatures. The only compound observed
was butoxydimethylphenylsilane (3a). Particularly notable is
the absence of the disiloxane. The results obtained are
presented in Table 1, while the time–conversion plots for
representative experiments are shown in Figure 4. In those
experiments where the selectivity was low (see footnote [c] in
Table 1), the reason was silane adsorption on G, making low
the mass balances (see Figure S12).
The presence of dopant elements exhibits a remarkable
influence on the catalytic activity, the most efficient catalyst
being Cu-(B)G followed by Cu-G. In view of the previous
particle size characterization, the most likely explanation for
this performance is the larger particle size of Cu NPs for those
G samples having N as dopant element (compare particle
Figure 4. Time–conversion plot for the reaction of 1a and 2a in the
presence: Cu-G (a), Cu-(B)G (b), Cu-(B,N)G (c), and Cu-(N)G (d).
To address this issue we performed twin experiments in which
the reaction of 1a and 2a were carried out in the presence of
Cu-G and after 1500 minutes an additional amount of Cu-G
was added in one of them (see Figure S13). While the reaction
in which no fresh Cu-G was added did not progress further,
the twin experiment adding fresh Cu-G catalyst re-start
increasing the conversion, reaching 92% at 3000 minutes.
This experiment clearly evidences faster deactivation of Cu-G
during the reaction than that of Cu-(B)G. This deactivation is
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12583

.
Angewandte
Communications
likely to be due to product adsorption, in view of the
reusability tests. In the case of Cu-(B)G a kinetic study of
product deactivation showed that the time–conversion plots
were not influenced by the presence of 3a in the range from 0
to 0.4m (Figure S13).
observed. Only adsorption is responsible for some apparent
decrease in selectivity. Thus, observation of disiloxane
formation in entries 2 to 6 in Table 2 indicates the superiority
of G as support for this reaction.
Reusability of Cu-(B)G and catalyst stability was tested
by performing a series of three consecutive runs using the
same sample recovered by filtration, washed with hexane, and
redispersed by sonication. It was observed that conversion of
1a decreased from 87 to 74 and 51% from the first to the
second and third use, respectively (see Figure S20). This
indicates a gradual deactivation of Cu-(B)G upon use. TEM
images of the three times used Cu-(B)G do not show
agglomeration of Cu NPs (see Figure S.21 in SI) although it
is a common cause of deactivation. In fact, the average Cu NP
size of the used catalyst is even smaller. In contrast, surface
area determination in the liquid phase using methylene blue
as probe shows a significant decrease in the surface area from
1527 to 1014 m2 gÀ1. Analysis of Cu in n-butanol after the
reaction indicates Cu leaching to some extent. Thus, at
present, the most likely causes for the gradual catalyst
deactivation seem to be the partial stacking of (B)G layers
and leaching of Cu from (B)G to the solution.
The activation energy (Ea) of the dehydrogenative
coupling of 1a and 2a was determined by carrying out this
reaction in the range of temperatures from 40 until 1008C,
determining the initial reaction rate from the slope of the time
conversion plot at zero reaction time. Ea was obtained from
the best fit of the logarithm of the reaction rate versus the
inverse of the absolute temperature, giving 27.7 Æ 0.3 kJmolÀ1
(see Figure S14). This value for Ea is relatively low indicating
that Cu NPs are efficient catalysts. Because of the low Ea
value, the possibility that the reaction could be limited by
diffusion was considered and dismissed, since a series of
experiments in which the stirring rate was set at 600 and
1200 rpm did not show any difference in the time–conversion
plot (see Figure S18).
To compare the activity of Cu NPs on G, we screened the
activity of a series of Cu-containing catalysts on various
supports including different carbon materials (diamond NPs
and high-surface-area-active carbon atoms) as well as on
organic biopolymer (chitosan) and metal oxide (MgO) (see
the Supporting Information for relevant data of these
catalysts).
In view of the high activity of Cu-(B)G as catalyst for the
coupling of 1a and butanol (see Table 1, entry 4), we
proceeded to study the scope of this catalyst for other silanes
and alcohols. The results obtained are presented in Table 3.
The activity data for the reaction of 1a and 2a catalyzed
by various Cu-containing catalysts is summarized in Table 2.
Table 3: Activity data for the dehydrogenative coupling of silanes and
alcohols catalyzed by Cu-(B)G.[a]
Table 2: Results of the dehydrogenative coupling of silane 1a with n-
butanol 2a catalyzed by Cu NPs supported on different solids.[a]
Entry Silane
Alcohol t [h] T [8C] Conver. [%][b] Selec. [%][b]
1
2
3
4
5
6
7
8
9
Me2PhSiH BuOH
Me2PhSiH MeOH 24
Me2PhSiH EtOH
Ph2SiH2
Ph2SiH2
Ph2SiH2
PhSiH3
PhSiH3
PhSiH3
48
100
70
>99
69
>99
90
Entry
Catalyst
Conversion [%][b]
Selectivity [%][b]
24
70
54
94
1
2
3
4
5
6
Cu-G
80
89
44
20
94
95
26
6
29
9
BuOH
MeOH
EtOH
BuOH
MeOH
EtOH
0.5 100
>99
>99
>99
>99
>99
90
100
100
100
100
100
93
Cu-Diamond NPs
Cu-Silylated Diamond
Cu/MgO
Cu/chitosan
Cu/active carbon
first use
1
1
70
70
0.5 100
1
5
70
70
99
30
44
<5
second use
[a] Reaction conditions: 1 (5 mmol), 2 (entries 1–3: 10 mmol, entries 4–
6: 20 mmol; entries 7–9: 30 mmol), catalyst (0.5 mol%), Ar atmos-
phere; [b] The selectivity and the conversion were obtained by GC with
internal standard.
[a] Reaction conditions: 1a (5 mmol), 2a (10 mmol), catalyst
(0.5 mol%), Ar atmosphere, 1008C, 24 h. [b] Conversion and selectivity
determined by GC with internal standard.
While the activity of supported Cu NPs as catalysts for the
dehydrogenative coupling is general, the selectivity of the
process is low or moderate, because of the competitive
formation of disiloxane as the major product. In the case of
Cu/active carbon that shows a moderate selectivity, the
activity fades in the second use because of Cu leaching. As
a summary of Table 2, the combination of Cu NPs and G as
support renders the best performing material, showing the
advantage of G as support.
To provide a valid comparison of the selectivity of various
catalysts, selectivity values should be compared at similar
conversion. In the present case, it was, however, observed that
for Cu NPs supported on G selectivity was almost constant
with conversion (see Figure S19) and no other products are
As expected, using Cu-(B)G high to moderate conversions
were achieved with excellent selectivity towards the target
alcoxysilane.
As it can be seen in Table 3, the dehydrogenative coupling
can also be performed using methanol or ethanol as alcohols.
In these cases, however, because of their lower boiling point
compared to n-butanol, the reaction was carried out at lower
temperature and, for this reason, the conversion of silane 1a
was significantly lower than that achieved at the same time for
n-butanol at higher temperature. Comparing methanol and
ethanol, the first was more reactive probably because of steric
effects.
Besides dimethylphenylsilane, other silanes were also
tested. Diphenylsilane reacts much faster than dimethylphe-
12584 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586

Angewandte
Chemie
nylsilane affording in a very short period of time the dialcoxy
product (Table 3, entry 4). This silane also reacts with
methanol and ethanol to the corresponding diethoxy and
dimethoxy products, although longer reaction times are
needed because of the lower reaction temperature (Table 3,
performance than other supports that can be derived from the
large surface area, high adsorption capacity and strong metal–
G interaction.
entries 5 and 6). In these cases, the evolution of the reaction Experimental Section
Cu NPs supported on Gs were prepared in a single step starting from
mixture shows the presence of low concentrations of the
corresponding monoalcoxysilane that appears as primary but
instable product. Phenylsilane also reacts promptly with
nBuOH, MeOH, and EtOH to the corresponding trialcoxy
phenylsilane (Table 3 entries 7, 8, and 9). Also in these cases,
the formation of intermediates of the corresponding mono-
and dialcoxy phenylsilanes was observed. Apparently, the
commercial alginate or chitosan that were precipitated with Cu-
(NO3)2 and submitted subsequently to pyrolysis and exfoliation as
indicated in the supporting information. The preparation of the
catalysts included in Table 2 is also described in the Supporting
Information. An example to illustrate the procedure is included here.
Synthesis of Cu/G: Alginic acid sodium salt from brown algae
(0.5 g, Sigma) was dissolved in a 50 mL of Cu(NO3)2 aqueous solution
(1 wt% with respect to the alginic acid sodium salt). Before pyrolysis,
the solution was concentrated by heating in an oven at 1008C
overnight. The pyrolysis was performed in an electrical horizontal
tubular oven in an argon stream (50 mLminÀ1) using the following
oven program: 2008C during 2 h for annealing and then heating at
108CminÀ1 up to 9008C for 6 h. This graphitic carbon residue is
sonicated at 700 W during 1 h for obtaining the dispersed Cu/G in the
reaction mixture.
General catalytic procedure for the dehydrogenative silylation
with alcohols: The catalyst (0.5 mol%) was charged in a 5 mL
reinforced glass reactor equipped with a magnetic bar. Then, alcohol
was added (2–4 mmol) and the reactor was sonicated under dry N2
atmosphere to deoxygenate. Finally, the silane was introduced in the
reactor with a syringe. The reaction was stirred at 1008C or the
required temperature from 30 minutes to 48 h, depending on the
substrates, in an oil bath. The reaction mixture was allowed to cool to
room temperature and the catalyst removed by filtration. For GC
analysis, a known amount of dodecane as internal standard was
added. Products were purified by flash chromatography on silica gel
with anhydrous ethyl acetate/hexane mixture and were characterized
by the spectroscopic techniques (see the Supporting Information for
spectroscopic data).
À
reactivity of silanes increases as the number of Si H bonds
increases.
Based on the results obtained and the nature of the
catalyst, a reasonable mechanistic proposal is that silane
interacts with Cu NPs to form Cu-Si and Cu-H species, while
the alcohol interacts with Lewis acid sites on G leading to
chemisorption with formation of alcoxide and protons.
Preliminary FTIR measurements of n-butanol adsorption on
Cu-(B)G show the presence of a band at about 1050 cmÀ1
upon contacting n-butanol at 808C that disappears subse-
quently upon further heating. This absorption band could be
attributed to the alcoxide interacting with the (B)G sheet. The
coupling between the silyl group bonded to Cu and the
alcoxide on the Lewis acid on G will render the final product
(Scheme 3). The primary role of the dopant element on the
Received: May 27, 2014
Revised: July 24, 2014
Published online: September 5, 2014
Scheme 3. Pictorial illustration of Cu NPs supported on doped G
À
sheets. Activation of Si H bonds of silane would take place on
Keywords: alcoxysilanes · dehydrogenative coupling ·
graphene · heterogeneous catalysis · metal nanoparticles
.
Cu NPs, while the G sheet would cooperate by adsorbing the alcohol.
This adsorption could take place preferentially on the dopant elements
(indicated as X).
[1] P. G. Belelli, M. L. Ferreira, D. E. Damiani, J. Mol. Catal. A
2000, 159, 315.
catalytic activity seems to be the control of the Cu particle size
distribution as previously proposed when commenting the
TEM images, but also should participate in adsorption of
reagents and products (see Figure S14) and probably also in
other steps in the reaction mechanism. The combined
information of in situ spectroscopic techniques and computa-
tional calculations are necessary to gain deeper insight into
the influence of dopants elements in the catalyst activity.
Thus, in the present study we have shown that Cu NPs
supported on G materials are efficient catalysts for the
dehydrogenative coupling of silanes and alcohols. Pyrolysis of
a conveniently modified biopolymer containing Cu2+ leads to
the simultaneous formation of doped graphitic carbon
residues and Cu NPs. The solid can be easily exfoliated by
sonication. Doping of G exhibits a notable influence on the
catalytic activity. Comparison with other types of carbon
atoms containing Cu indicates that G exhibits a superior
[2] I. Dꢀaz, J. Perez-Pariente, Chem. Mater. 2002, 14, 4641.
[3] M. Matheron, T. Gacoin, J. P. Boilot, A. Bourgeois, A. Brunet-
Bruneau, J. Rivory, A. Jimenez, J. Biteau, Stud. Surf. Sci. Catal.
2005, 156, 327.
[4] A. Gꢁmez-Avilꢂs, P. Aranda, F. M. Fernandes, C. Belver, E.
Ruiz-Hitzky, J. Nanosci. Nanotechnol. 2013, 13, 2897.
[5] N. N. Herrera, J. M. Letoffe, J. L. Putaux, L. David, E. Bourgeat-
Lami, Langmuir 2004, 20, 1564.
[6] M. Jaber, F. O. M. Gaslain, J. Miehe-Brendle, Clays Clay Miner.
2009, 57, 35.
[7] H. Garcꢀa, M. Stratakis, Chem. Rev. 2012, 112, 4469.
[8] T. Mitsudome, Y. Yamamoto, A. Noujima, T. Mizugaki, K.
Jitsukawa, K. Kaneda, Chem. Eur. J. 2013, 19, 14398.
[9] T. Taguchi, K. Isozaki, K. Miki, Adv. Mater. 2012, 24, 6462.
[10] S. Rendler, O. Plefka, B. Karatas, G. F. Auer, R. C. Muck-
Lichtenfeld, S. Grimme, M. Oestreich, Chem. Eur. J. 2008, 14,
11512.
[11] K. T. Kira, H. A. Hamajima, T. Baba, S. Takai, M. Isobe,
Tetrahedron 2002, 58, 6485.
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12585

.
Angewandte
Communications
[12] J. W. Park, C. H. Jun, Org. Lett. 2007, 9, 4073.
[13] H. S. Hilal, A. Rabah, I. S. Khatib, A. F. Schreiner, J. Mol. Catal.
1990, 61, 1.
[22] J. Ding, M. Wang, X. Zhang, C. Ran, RSC Adv. 2013, 3, 14073.
[23] A. Dhakshinamoorthy, A. Primo, P. Concepcion, M. Alvaro, H.
Garcia, Chem. Eur. J. 2013, 19, 7547.
[14] H. Ito, K. Takagi, T. Miyahara, M. Sawamura, Org. Lett. 2005, 7,
3001.
[15] J. A. Zazo, J. A. Casas, A. F. Mohedano, J. J. Rodrꢀguez, Appl.
Catal. B 2006, 65, 261.
[24] M. Latorre-Sꢃnchez, A. Primo, H. Garcꢀa, Angew. Chem. Int.
Ed. 2013, 52, 11813; Angew. Chem. 2013, 125, 12029.
[25] I. Lazareva, Y. Koval, M. Alam, S. Strçmsdçrfer, P. Mꢄller, Appl.
Phys. Lett. 2007, 90, 262108.
[16] S. Ikeda, S. Ishino, T. Harada, N. Okamoto, T. Sakata, H. Mori, S.
Kuwabata, T. Torimoto, M. Matsumura, Angew. Chem. Int. Ed.
2006, 45, 7063; Angew. Chem. 2006, 118, 7221.
[17] P. V. Kamat, J. Phys. Chem. Lett. 2010, 1, 520.
[18] C. Xu, X. Wang, J. Zhu, J. Phys. Chem. C 2008, 112, 19841.
[19] L. Shang, T. Bian, B. Zhang, D. Zhang, L. Z. Wu, C. H. Tung, Y.
Yin, T. Zhang, Angew. Chem. Int. Ed. 2014, 53, 250; Angew.
Chem. 2014, 126, 254.
[26] M. Buaki-Sogo, M. Serra, A. Primo, M. Alvaro, H. Garcꢀa,
ChemCatChem 2013, 5, 513.
[27] A. El Kadib, A. Primo, K. Molvinger, M. Bousmina, D. Brunel,
Chem. Eur. J. 2011, 17, 7940.
[28] A. El Kadib, K. Molvinger, T. Cacciaguerra, M. Bousmina, D.
Brunel, Microporous Mesoporous Mater. 2011, 142, 301.
[29] C. Lavorato, A. Primo, R. Molinari, H. Garcia, ACS Catal. 2014,
4, 497.
[20] T. Sun, Z. Zhang, J. Xiao, C. Chen, F. Xiao, S. Wang, Y. Liu,
Nature 2013, 493, 195.
[30] P. Shi, R. Su, F. Wan, M. Zhu, D. Li, S. Xu, Appl. Catal. B 2012,
123 – 124, 265.
[21] E. Yoo, T. Okata, T. Akita, M. Kohyama, J. Nakamura, I.
Honma, Nano Lett. 2009, 9, 2255.
12586 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 12581 –12586