Unique catalysis of gold nanoparticles in the chemoselective hydrogenolysis with H2: Cooperative effect between small gold nanoparticles and a basic support

Article abstract of DOI:10.1039/c2cc32850j

Gold nanoparticles on hydrotalcite act as a heterogeneous catalyst for the chemoselective hydrogenolysis of various allylic carbonates to the corresponding terminal alkenes using H₂ as a clean reductant. The combination of gold nanoparticles and basic supports elicited significantly unique and selective catalysis in the hydrogenolysis.

Full text of DOI:10.1039/c2cc32850j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 6723–6725
www.rsc.org/chemcomm
COMMUNICATION
Unique catalysis of gold nanoparticles in the chemoselective
hydrogenolysis with H2: cooperative effect between small gold
nanoparticles and a basic supportw
Akifumi Noujima,a Takato Mitsudome,a Tomoo Mizugaki,a Koichiro Jitsukawaa and Kiyotomi Kaneda*ab
Received 21st April 2012, Accepted 9th May 2012
DOI: 10.1039/c2cc32850j
Gold nanoparticles on hydrotalcite act as a heterogeneous
catalyst for the chemoselective hydrogenolysis of various allylic
carbonates to the corresponding terminal alkenes using H2 as a
clean reductant. The combination of gold nanoparticles and
basic supports elicited significantly unique and selective catalysis
in the hydrogenolysis.
heterosubstituents have been transformed into the corresponding
internal alkenes through hydrogenolysis using Pd/C with H2,9a,10
RhCl(PPh3)3 with N-propyl-1,4-dihydronicotinamide,11 and
Pd(PPh3)4 with silanes,12 potassium boron hydride,13 or metal
hydrides.14 On the other hand, catalytic systems for the
production of the corresponding terminal alkenes from allylic
heterosubstituents are rare; only Pd–phosphine complexes
with formates or SmI2 as a reductant have been reported.15
These Pd complex systems, however, suffer from the requirement
of additional phosphines and amines, the use of stoichiometric
amounts of hazardous reductants, and a tedious workup for the
separation and reuse of the catalysts. Therefore, the development
of a green catalytic system for the hydrogenolysis of allylic
heterosubstituents to terminal alkenes using heterogeneous
catalysts with H2 as a clean reductant is a challenging issue.
Herein, we report that hydrotalcite-supported Au NPs (Au/HT)
efficiently catalyze the hydrogenolysis of allylic carbonates to the
corresponding terminal alkenes with H2 as an ideally green
reductant (Scheme 1). The chemoselectivity is significantly superior
to those of other metal NPs. To the best of our knowledge, this
report is the first example of the selective hydrogenolysis of allylic
carbonates to terminal alkenes with H2.
For heterogeneous catalysts, the size of the metal nanoparticles
(NPs) and the interactions between the active metal species and
the support greatly influence the catalytic activity.1 In this
context, considerable effort has been devoted to controlling
the size of metal nanoparticles, screening supports, and altering
the composition of the supports to achieve high catalytic
performance. Recently, small Au NPs on appropriate supports
have been revealed to show significantly unique catalytic abilities
compared to other supported metal NPs, not only in oxidation
reactions, but also in reduction reactions.2,3 For example,
atomically precise Au25(SR)18 nanoparticles on Fe2O3 and
TiO2 catalyze the selective hydrogenation of a,b-unsaturated
carbonyl compounds to allylic alcohols through the coordination of
carbonyl oxygen to the vacant site of Au nanoparticles,3b and
Au/TiO2 can selectively hydrogenate substituted nitroaromatics to
the corresponding anilines where the favored adsorption of the
nitro group on the TiO2 surface leads to the high chemoselectivity.2a
We have also found that Au NPs supported on hydrotalcite
(Au/HT) and on hydroxyapatite (Au/HAP) showed high catalytic
activities towards the chemoselective reductions of epoxides4–6 and
amides7 to the corresponding alkenes and amines, respectively,
and the activities and chemoselectivities of the Au NPs were much
higher than those of other metal NPs.
The hydrogenolysis of cinnamyl methyl carbonate (1) using
various supported Au NPs of similar particle sizes16 was
carried out in toluene under an atmospheric pressure of H2
(Table 1). Au/HT exhibited the highest activity, affording the
corresponding terminal alkene 3-phenylpropene (2) and the
internal alkene b-methyl styrene (3) in 98% selectivity (2 : 3 =
10 : 1), as well as small amounts of the alkane by-products
3-phenylpropane (PP) and methyl 3-phenylpropyl carbonate
(PC) (entry 1). The basic MgO support also showed high
selectivity for 2, but the yield of 2 was much lower than that
obtained with Au/HT (entry 8). Weakly basic supports such as
CeO2 and Al2O3 resulted in moderate yields of 2 with lower
selectivities (entries 6 and 7). The non-basic supports TiO2 and
SiO2 showed extremely low activities in this transformation
(entries 9 and 10). No reaction occurred when unsupported
Hydrogenolysis of allylic heterosubstituents to alkenes is
important in organic chemistry because it allows the preparation
of various alkenes and the use of an allyl group as a protective
group for alcohols, carboxylic acids and amines.8,9 To date, allylic
a Department of Materials Engineering Science,
Graduate School of Engineering Science, Osaka University,
1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan.
E-mail: kaneda@cheng.es.osaka-u.ac.jp; Fax: +81 6-6850-6260;
Tel: +81 6-6850-6260
b Research Center for Solar Energy Chemistry, Osaka University,
1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan
w Electronic supplementary information (ESI) available: Experimental
details, EXAFS and TEM of Au/HT. See DOI: 10.1039/c2cc32850j
Scheme 1
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 6723–6725 6723

View Article Online
a
Table 1 Hydrogenolysis of 1 using various catalysts with H2
Ag/HT and Cu/HT, barely worked as catalysts (entries 16–18).
The use of Pd/C18 under the present conditions resulted in the
formation of undesired alkanes (entry 13). These phenomena
reveal that Au NPs have unique and high catalytic activity for
the hydrogenolysis of allylic carbonates to alkenes with H2.
The size effect of the Au NPs was then investigated. The
hydrogenolysis of 1 with H2 using a series of Au/HTs with
different particle sizes was carried out (entries 1, 19–22).19
Interestingly, the selectivity, as well as the yield of 2, increased
with a decrease in the size of the Au NPs. Larger Au NPs
showed a lower selectivity for 2 due to hydrogenation of the
CQC bonds of both 1 and 2. These results suggest, therefore,
that immobilization of small Au NPs on HT is essential for the
selective hydrogenolysis of 1 to the terminal alkene 2. The high
chemoselectivity of Au/HT with small Au NPs in the hydro-
genolysis of allylic carbonates to terminal alkenes was observed
with other substrates (Table 2). Both aliphatic and alicyclic allylic
carbonates, such as geranyl methyl carbonate, methyl perillyl
carbonate and methyl mirtenyl carbonate, were selectively trans-
formed into the corresponding terminal alkenes (entries 5–7).
To investigate the reaction mechanism for the Au/HT-
catalyzed hydrogenolysis, several control experiments were
carried out. Both the linear allylic carbonate trans-cinnamyl
methyl carbonate (L) and the branched allylic carbonate
1-phenyl-prop-2-enyl methyl carbonate (B) were treated with
Au/HT under a H2 atmosphere to afford the terminal alkene
3-phenylpropene with similar selectivity (Table 2, entries 1 and 2).
The use of D2 instead of H2 in the Au/HT-catalyzed hydrogenolysis
of both L and B provided 3-(D)-3-phenylpropene as a major
product (Scheme 2). No hydrogenation, isomerization or H–D
exchange reaction occurred when treating (H)-3-phenylpropene
with D2 in the presence of Au/HT. These results suggest that
Au/HT-catalyzed hydrogenolysis may mainly proceed via a p-allyl
Sel.b (%)
Conv.b
Entry
Catalyst
d (nm)
(%)
2 and 3 2 : 3
PP PC
1
Au/HT
Au/HT
Au/HT
Au/HT
Au/HT
Au/CeO2
Au/Al2O3
Au/MgO
Au/TiO2
Au/SiO2
Au NPsg
Pd/HT
2.7
2.7
2.7
2.7
2.7
3.8
3.6
3.1
3.7
14
>99
>99
>99
99
98
78
98
>99
>99
>99
>99
62
46
97
49
—
—
0
10 : 1 o1 o1
2c
10 : 1
10 : 1
10 : 1
10 : 1
8 : 1
7 : 1
9 : 1
8 : 1
0
0
0
0
3
12
1
8
—
—
10
16
6
15
—
—
—
2
0
0
0
3c,d
4c,e
5c,f
6
7
8
0
35
42
2
86
32
9
23
o1
43
—
—
90
84
94
85
—
—
—
4
10
11
12h
13h
14h
15
16
17
18
19
20
21
22
3.5
0
>99
>99
>99
17
Pd/Ci
0
0
0
0
Pt/HT
Rh/HT
Ru/HT
Ag/HT
Cu/HT
Au/HT
Au/HT
Au/HT
Au/HT
2
1
0
o1
—
93
82
61
43
5.8
7.9
11
87
78
9 : 1
8 : 1
7 : 1
7 : 1
4
8
15
13
31
42
60
31
20
a
Reaction conditions: catalyst (M: 2.3 mol%), toluene (5 mL), 1
b
(0.2 mmol). Determined by GC using an internal standard techni-
c
d
e
f
g
que. 60 1C, 16 h. 60 1C, 48 h. Reuse 1. Reuse 2. Prepared
according to ref. 17. 1 h. Obtained from Sigma Aldrich.
h
i
Au NPs were used and aggregation of the Au particles was
observed (entry 11).17 The above results clearly demonstrate
that basic supports are necessary for the selective formation of
2 from 1. Notably, lowering the reaction temperature to 60 1C
increased the selectivity for the alkenes to over 99%, while
suppressing the hydrogenation of the CQC bond of both
1 and 2 (entry 2). Moreover, the CQC bonds of the alkene
products were completely intact, even after prolonged heating
following the complete consumption of 1 (entry 3). To confirm
whether the Au/HT-catalyzed hydrogenolysis occurs on the
solid surface of Au/HT, Au/HT was removed by filtration
from the reaction mixture at 40% conversion of 1. Continuous
treatment of the resulting filtrate under similar reaction conditions
did not afford any products (see Fig. S3, ESIw). In addition,
inductively coupled plasma (ICP) analysis showed the absence of
Au species in the filtrate (detection limit of 0.1 ppm), revealing that
no leaching occurred during the reaction. These results confirm
that the hydrogenolysis proceeds on the Au NPs immobilized on
the HT. The recovered Au/HT catalyst was reusable for a
subsequent reaction without any pretreatment; Au/HT exhibited
a high catalytic activity and selectivity in the reuse experiments
without loss of its efficiency (entries 4 and 5).
a
Table 2 Au/HT-catalyzed hydrogenolysis of allylic carbonates with H2
Temp. Time Conv.b Sel.b Terminal :
Entry Substrate
(1C)
(h)
16
16
(%)
(%) internal
>99 10 : 1
>99 9 : 1
1
2
60
>99
>99
60
3
4
100
100
12
12
93
91
99 6 : 1
99 6 : 1
5
6
100
90
12
20
90
90
94 6 : 1
95 4 : 1
7
100
16
92
99 6 : 1
8
9
100
80
12
24
83
94
>99
—
Next, various metal NPs supported on HT were examined in
the hydrogenolysis of 1 under similar reaction conditions
(entries 12, 14–18). When Pd/HT, Pt/HT and Rh/HT were used
in place of Au/HT, the undesired hydrogenation of 1 occurred
to give alkanes, and no alkene products were obtained
(entries 12, 14 and 18). Other metal NPs on HT, such as Ru/HT,
>99 10 : 1
a
Reaction conditions: Au/HT (Au: 2.3 mol%), toluene (5 mL),
b
substrate (0.2 mmol). Determined by GC using an internal standard
technique.
c
6724 Chem. Commun., 2012, 48, 6723–6725
This journal is The Royal Society of Chemistry 2012

View Article Online
This work was supported by a Grant-in-Aid for Young
Scientists (A) (23686116) from the Japan Society for the
Promotion of Science (JSPS). We also thank Dr T. Uruga,
Dr H. Tanida, and Dr K. Nitta (SPring-8) for XAFS measurements
(2011A1295). A. N. thanks the JSPS Research Fellowships for
Young Scientists.
Scheme 2
Notes and references
1 (a) Y. Zhang, X. Cui, F. Shi and Y. Deng, Chem. Rev., 2012,
112, 2467; (b) A. Villa, M. Schiavoni and L. Prati, Catal. Sci.
Technol., 2012, 2, 673; (c) T. Tsukuda, H. Tsunoyama and
H. Sakurai, Chem.–Asian J., 2011, 6, 736.
´
2 (a) M. Boronat, P. Concepcion, A. Corma, S. Gonzalez, F. Illas
and P. Serna, J. Am. Chem. Soc., 2007, 129, 16230; (b) A. Corma
and P. Serna, Science, 2006, 313, 332.
3 (a) M.-M. Wang, L. He, Y.-M. Liu, Y. Cao, H.-Y. He and
K.-N. Fan, Green Chem., 2011, 13, 602; (b) Y. Zhu, H. Qian,
B. A. Drake and R. Jin, Angew. Chem., Int. Ed., 2010, 49, 1295;
(c) P. Claus, H. Hofmeister and C. Mohr, Gold Bull., 2004, 37, 181.
4 T. Mitsudome, A. Noujima, Y. Mikami, T. Mizugaki, K. Jitsukawa
and K. Kaneda, Angew. Chem., Int. Ed., 2010, 49, 5545.
5 T. Mitsudome, A. Noujima, Y. Mikami, T. Mizugaki,
K. Jitsukawa and K. Kaneda, Chem.–Eur. J., 2010, 16, 11818.
6 A. Noujima, T. Mitsudome, T. Mizugaki, K. Jitsukawa and
K. Kaneda, Angew. Chem., Int. Ed., 2011, 50, 2986.
7 Y. Mikami, A. Noujima, T. Mitsudome, T. Mizugaki,
K. Jitsukawa and K. Kaneda, Chem.–Eur. J., 2010, 17, 1768.
8 J. Tsuji and T. Mandai, Synthesis, 1996, 1.
9 (a) N. K. Garg, D. D. Caspi and B. M. Stoltz, J. Am. Chem. Soc.,
2004, 126, 9552; (b) R. J. Cvetovich, D. H. Kelly, L. M. DiMichele,
R. F. Shuman and E. J. J. Grabowski, J. Org. Chem., 1994,
59, 7704.
10 D. D. Caspi, N. K. Garg and B. M. Stoltz, Org. Lett., 2005,
7, 2513.
11 K. Nakamura, A. Ohno and S. Oka, Tetrahedron Lett., 1983,
24, 3335.
12 E. Keinan and N. Greenspoon, J. Org. Chem., 1983, 48, 3545.
13 R. O. Hutchins, K. Learn and R. P. Fulton, Tetrahedron Lett.,
1980, 21, 27.
14 (a) O. Dangles, F. Guibe and G. Balavoine, J. Org. Chem., 1987,
52, 4984; (b) E. Keinan and N. Greenspoon, Tetrahedron Lett.,
1982, 23, 241; (c) H. Matsushita and E. Negishi, J. Org. Chem.,
1982, 47, 4161; (d) R. O. Hutchins and K. Learn, J. Org. Chem.,
1982, 47, 4382; (e) F. Guibe and Y. S. M’Leux, Tetrahedron Lett.,
1981, 22, 3591.
Scheme 3
intermediate generated from an allylic carbonate. In our previous
studies on the deoxygenation of epoxides, it was found that the
polar hydrogen species Au-hydride and H+ were formed through
cooperative effects between the Au NPs and the basic sites of HT
when using 2-propanol or CO/H2O, as well as H2 as the
reductant.4,5 When 2-propanol or CO/H2O was employed instead
of H2, hydrogenolysis of 1 also occurred to provide 2 and 3 with
similar selectivity, which suggests the generation of Au-hydride
and H+ as active species in these reactions (Scheme 3).
Bearing in mind that the Au/HT-catalyzed hydrogenolysis
of allylic carbonates may involve the generation of Au-hydride
and H+ species and the formation of p-allyl intermediates,
the following reaction path is proposed. First, heterolytic
dissociation of H2 occurs at the interface between the
Au NPs and the basic sites of HT to give Au-hydride and
H+ on the HT surface.6 The H+ species generated on the
HT reacts with the allylic carbonate to form a p-allyl
Au intermediate, followed by attack of the hydride to afford
the corresponding alkene (see Fig. S4, ESIw). Hydride transfer
to the more substituted C3-side of the p-allyl intermediate in
the present Au/HT-H2 system is not similar to the previously
reported hydrogenolysis using Pd(PPh3)4 with metal hydride
affording the internal alkenes via the attack of Pd-hydride on
the less substituted C1-side.12–14,20 This phenomenon is also
different from the g-attack of H+ on Z1-allyl gold species
giving the corresponding alkenes.21
15 (a) A. Yoshida, T. Hanamoto, J. Inanaga and K. Mikami, Tetrahedron
Lett., 1998, 39, 1777; (b) T. Hayashi, H. Iwamura, M. Naito,
Y. Matsumoto and Y. Uozumi, J. Am. Chem. Soc., 1994,
116, 775; (c) T. Mandai, T. Matsumoto, M. Kawada and
J. Tsuji, J. Org. Chem., 1992, 57, 6090; (d) M. Oshima,
I. Shimizu, A. Yamamoto and F. Ozawa, Organometallics, 1991,
10, 1221; (e) J. Tsuji, I. Shimizu and I. Minami, Chem. Lett., 1984,
1017; (f) J. Tsuji, T. Yamakawa, M. Kaito and T. Mandai,
Tetrahedron Lett., 1978, 18, 2075; (g) H. Hey and H. J. Arpe,
Angew. Chem., Int. Ed. Engl., 1973, 12, 928.
In conclusion, Au/HT catalyzes the highly chemoselective
hydrogenolysis of allylic carbonates to the corresponding
terminal alkenes with H2 without CQC bond hydrogenation
of the substrates or products. The Au NPs exhibited significantly
different selectivity from those of other metal NPs. The polar
hydrogen species Au-hydride and H+ are formed through the
heterolytic cleavage of H2 at the interface between the Au NPs
and the HT surface, which may react with the allylic carbonates
to selectively afford terminal alkenes via the formation of p-allyl
intermediates. The effect of small Au NPs on the highly selective
hydrogenation of the C–O bond can be explained by the
increasing interfacial area between the Au NPs and the HT
which enables the exclusive formation of Au-hydride and H+.
The present study highlights the importance of the combination
of small gold nanoparticles and a basic support for achieving
high chemoselectivity.
16 T. Mitsudome, A. Noujima, T. Mizugaki, K. Jitsukawa and
K. Kaneda, Green Chem., 2009, 11, 793.
17 N. R. Jana, L. Gearheart and C. J. Murphy, Langmuir, 2001,
17, 6782.
18 It has been reported that commercially available Pd/C was effective
in the hydrogenolysis of cyclic allylic carbonate (ref. 10) as well as
various protective groups. See: (a) L. Jobron and O. Hindsgaul,
J. Am. Chem. Soc., 1999, 121, 5835; (b) E. A. Papageorgiou,
M. J. Gaunt, J. Yu and J. B. Spencer, Org. Lett., 2000, 2, 1049.
19 Au/HTs with different particle sizes were prepared by changing the
Au content. See ref. 6 for the preparation of Au/HTs.
20 Hydride transfer to the more substituted C3-side of the p-allyl
intermediate is observed in the Pd complex with formates. See
ref. 8 and 15.
21 (a) S. Komiya and S. Ozaki, Chem. Lett., 1988, 1431; (b) T. Sone,
S. Ozaki, N. C. Kasuga, A. Fukuoka and S. Komiya, Bull. Chem.
Soc. Jpn., 1995, 68, 1523.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 6723–6725 6725