Homogeneous gold-catalyzed hydrosilylation of aldehydes

Article abstract of DOI:10.1016/j.jorganchem.2006.10.025

The catalytic hydrosilylation of aldehydes in the presence of PBu₃ modified Au(I)-complexes was investigated. In situ IR and NMR experiments have revealed that both, the ligand PBu₃ and the substrate aldehyde play an important role in stabilizing the gold catalyst and/or forming the catalytically active species. In their absence the reducing power of silane destabilizes the gold (I) catalyst giving rise to gold clusters or particles. Several side reactions involving water and oxygen were also investigated. A plausible reaction pathway as an alternative to the well-accepted mechanism for the transition-metal homogeneously catalyzed hydrosilylation of aldehydes has been proposed to accommodate the experimental observations.

Full text of DOI:10.1016/j.jorganchem.2006.10.025

Journal of Organometallic Chemistry 692 (2007) 1799–1805
www.elsevier.com/locate/jorganchem
Homogeneous gold-catalyzed hydrosilylation of aldehydes
a
b, ,1
*
b
a
a,
*
Di a´ na Lantos , Mar ´ı a Contel
, Sergio Sanz , Andrea Bodor , Istv a´ n T. Horv a´ th
a
Institute of Chemistry, E o¨ tv o¨ s University, H-1117 Budapest, P a´ zm a´ ny P e´ ter s e´ t a´ ny 1/A, Hungary
Departamento de Qu ´ı mica Inorg a´ nica – Instituto de Ciencia de Materiales de Arag o´ n, Universidad de Zaragoza – C.S.I.C., Zaragoza E-50009, Spain
b
Received 23 August 2006; received in revised form 12 October 2006; accepted 12 October 2006
Available online 19 October 2006
This paper is dedicated to Professor Gyula P a´ lyi on the occasion of his 70th birthday.
Abstract
The catalytic hydrosilylation of aldehydes in the presence of PBu
3
modified Au(I)-complexes was investigated. In situ IR and NMR
experiments have revealed that both, the ligand PBu and the substrate aldehyde play an important role in stabilizing the gold catalyst
3
and/or forming the catalytically active species. In their absence the reducing power of silane destabilizes the gold (I) catalyst giving rise to
gold clusters or particles. Several side reactions involving water and oxygen were also investigated. A plausible reaction pathway as an
alternative to the well-accepted mechanism for the transition-metal homogeneously catalyzed hydrosilylation of aldehydes has been pro-
posed to accommodate the experimental observations.
ꢀ
2006 Elsevier B.V. All rights reserved.
Keywords: Hydrosilylation; Homogeneous catalysis; Gold; Reaction mechanisms
1
. Introduction
could be the coordination properties of the substrates,
products, and side products as well as the ligands used.
The application of gold compounds as catalysts has
been gaining momentum as gold is considered as an envi-
ronmentally friendly metal and thus can be used in place
of toxic heavy metals [1]. In addition, gold and/or its com-
plexes could catalyze a great diversity of catalytic reactions
efficiently, some times with unexpectedly high selectivities
to products. Since the catalytically active species could be
mono- and/or polynuclear gold species and/or nanoparti-
cles in homogeneous catalytic systems, the molecular level
understanding of how to control the nuclearity or the par-
ticle size is crucial in the design of more stable, efficient and
even recoverable catalysts. Thus, molecular architecture
depends on the nature of the reaction itself and the addi-
tives used. In the case of homogeneous gold catalysis these
ð1Þ
Hydrosilylation of aldehydes and ketones is an impor-
tant reaction for the preparation of various intermediates
in organic synthesis and is generally catalyzed by rhodium,
ruthenium, or platinum complexes [2]. On the basis of a
PBu -modified gold catalyst for the hydrosilylation of alde-
3
hydes (Eq. 1) [3] we have recently reported several fluorous
gold catalysts to provide facile catalyst recycling [4]. Dur-
ing the development of the fluorous catalysts we have noted
several unusual features including the activity and stability
of the fluorous gold complexes and nanoparticles, both
depending on the ligand and aldehyde concentrations.
For example, it is rather surprising that the initial rate of
*
Corresponding authors.
E-mail address: istvan.t.horvath@hit-team.net (I.T. Horv a´ th).
Present address: Chemistry Department, Brooklyn College and The
1
the reaction of PhCHO (1a) with HSiMe Ph (2a) was
Graduate Center, City University of New York (CUNY), 2900 Bedford
Avenue, Brooklyn, NY 11210, United States.
2
hardly effected and the yield of PhCH OSiMe Ph (3a)
2
2
0
022-328X/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.10.025

1800
D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
plexes are catalytically inactive even in the presence of
excess PPh , the addition of PBu results in the formation
3
3
of the active species of unknown structure(s) [3]. First, we
have investigated the reaction of [AuCl(tht)] with various
amounts of PBu . It has been reported that while the reac-
3
tion [Et N][AuBr ] with 1 or 2 equiv. of PBu leads to the
4
2
3
formation of [AuBr(PBu )] or [Au(PBu ) ]Br, respectively,
3
3 2
higher PBu concentrations result in a rapidly exchanging
3
[
Au(PBu ) ]Br species even at ꢀ120ꢁC [5]. We have
3
3
observed a similar behavior at room temperature during
the titration of [AuCl(tht)] with various amounts of PBu
3
3
1
by P NMR (Fig. 1). The formation of [AuCl(PBu3)]
and [Au(PBu ) ]Cl could be clearly observed by the
3
2
appearance of the peaks at 23.5 ppm and 33.5 ppm, respec-
tively. Only one peak is observable above P/Au = 2, which
is shifted to higher fields and becomes broader at higher
Scheme 1.
was increased from 50% to almost full conversion by
ratios indicating the possible formation of [Au(PBu ) ]Cl
3 n
increasing the concentration of PBu3 (4) from 10 to
(n > 2) and the rapid exchange between these species
(Scheme 2).
2
0 mol% with respect to 3 mol% of the gold catalyst [3].
These results are difficult to explain by the well-accepted
mechanism of hydrosilylation of aldehydes, which includes
the oxidative addition of 2 to the metal center, the coordi-
nation and insertion of the carbonyl group of 1 into the
metal-silicon bond, and the reductive elimination of 3
Next we have investigated the performance of the PBu -
3
modified [AuCl(tht)] catalyst in the hydrosilylation of benz-
aldehyde (1a), propanal (1b), and nonanal (1c) using
Me PhSiH (2a) and Et SiH (2b) in CH CN, CH Cl or neat
2 3 3 2 2
reaction mixture (Table 1). It is important to emphasize that
in the absence of [AuCl(tht)] these hydrosilylation reactions
do not take place. The conversions of the aldehydes with 2a
were higher than that of with 2b. While the nature of the
two solvents employed has little effect, the reaction proceeds
somewhat faster in their absence. The products were identi-
fied by GC–MS and NMR (Table 2).
(
Scheme 1) [2]. We report here our detailed study on the
effects of key reaction parameters on rate and selectivity
including some of the side reactions.
2
. Results and discussion
The gold precursors used in the hydrosilylation of alde-
The dependence of the reaction rate on key reaction
parameters was investigated by in situ NMR measurements
at room temperature (see Fig. S1–S4 in Supplementary
hydes were either [AuCl(PPh )], as reported [3], or
3
[
AuCl(tht)] (tht = tetrahydrotiophene). While these com-
Au:P=1:0.5
60
50
40
30
20
10
Au:P=1:1
0
6
6
0
0
50
50
40
40
30
30
20
20
10
0
0
Au:P=1:1.5
10
Au:P=1:2
6
6
0
0
50
50
40
40
30
30
20
20
10
0
0
Au:P=1:2.5
10
Au:P=1:3
6
6
0
0
50
50
40
40
30
30
20
20
10
0
0
Au:P=1:3.5
10
Au:P=1:4
60
50
40
30
20
10
0
3
1
Fig. 1. P NMR spectra of the reaction mixture of 0.08 mmol of [AuCl(tht)] with PBu
3 3
(at molar ratios of Au to PBu : 1:0.5, 1:1, 1:1.5, 1:2, 1:2.5, 1:3,
1
:3.5 and 1:4) in 350 lL of dry and degassed CDCl under Ar at room temperature.
3

D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
1801
phosphine equilibria shown in Scheme 2. A similar effect
could be observed by changing the concentration of PBu3
(4), 1a, and 2a. A detailed kinetic investigation is underway
to establish the kinetic model for this system.
It should be noted, that at the Au/PBu = 1:1 or 1:2
3
ratios the reaction mixture became heterogeneous, evi-
denced by the formation of a black precipitate, and no reac-
tion could be observed. We have therefore increased the
PBu concentration, in accordance to previous work [3]
3
and our results. The hydrosilylation of 2 mmol of 1a with
4
mmol of 2a in the presence of 3 mol% [AuCl(PPh3)]
and 20 mol% 4 in 2 mL CH CN at 70 ꢁC was monitored
3
by in situ IR spectroscopy. To our surprise the colorless
reaction mixture turned deep purple when the IR bands
of 1a disappeared (e.g. at 100% conversion of 1a) and the
formation of a black precipitate was noticed. These results
have suggested that the presence of excess of PhCHO (1a)
Scheme 2.
material). By lowering the gold concentration from 3 mol%
to 1 mol%, the reaction rate decreases significantly, how-
ever not proportionally to the gold concentration. This
is probably due to the changing positions of the gold–
over HSiMe Ph (2a) is an important factor, in addition to
2
the excess of PBu (4) over gold [3]. The reaction was also
3
performed without added solvent. When 3.22 mmol 2a was
Table 1
a
Hydrosilylation catalyzed with AuCl(tht) and PBu
3
b
c
Aldehyde
Silane
Solvent
Product
Yield (%)
Benzaldehyde (1a)
Benzaldehyde (1a)
Benzaldehyde (1a)
Benzaldehyde (1a)
Propanal (1b)
Propanal (1b)
Propanal (1b)
Propanal (1b)
Nonanal (1c)
Nonanal (1c)
Nonanal (1c)
Nonanal (1c)
Benzaldehyde (1a)
Nonanal (1c)
a
Me
Me
2
PhSiH (2a)
PhSiH (2a)
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
–
3
2
3
2
3
2
3
2
3
2
3
2
CN
Cl
CN
Cl
CN
Cl
CN
Cl
CN
Cl
CN
Cl
(Benzyloxy)dimethylphenylsilane (3a)
(Benzyloxy)dimethylphenylsilane (3a)
(Benzyloxy)triethylsilane (3b)
(Benzyloxy)triethylsilane (3b)
Dimethylphenylpropoxysilane (3c)
Dimethylphenylpropoxysilane (3c)
Triethylpropoxysilane (3d)
81
82
78
55
69
56
30
5
82
95
16
3
2
2
Et
Et
3
SiH (2b)
SiH (2b)
3
2
Me
Me
2
PhSiH(2a)
PhSiH(2a)
2
2
Et
Et
3
SiH (2b)
SiH (2b)
3
2
Triethylpropoxysilane (3d)
Me
Me
2
PhSiH (2a)
PhSiH (2a)
Dimethylphenylsiloxynonane (3e)
Dimethylphenylsiloxynonane (3e)
1-Triethylsiloxynonane (3f)
1-Triethylsiloxynonane (3f)
(Benzyloxy)dimethylphenylsilane (3a)
Dimethylphenylsiloxynonane (3f)
2
2
Et
Et
3
SiH (2b)
SiH (2b)
3
2
Me
Me
2
PhSiH (2a)
PhSiH (2a)
100
100
2
–
3
3
mol% [AuCl(tht)] and 20 mmol% PBu (4) in a solution of 1 mmol aldehyde and 0.9 mmol silane in 0.5 mL solvent at 70 ꢁC for 3 h.
b
c
Characterized by NMR and GC–MS.
Conversion by H NMR.
1
Table 2
The characterization of the hydrosilylation products
Product
MS data (m/z)
NMR
1
(
Benzyloxy)dimethylphenylsilane
3a)
Benzyloxy)triethylsilane (3b)
227, 197, 164, 149, 135, 91
3
H (CDCl ) 7.2–7.5 (m, 3H), 7.26 (s, 5H), 4.66 (s, 2H), 0.4 (s, 6H)
(
(
193, 163, 135, 91
+
13
1
Dimethylphenylpropoxysilane (3c)
194 (M ), 179, 137, 135, 121, 105, 91,
C{ H} (CDCl
3
) 140.8, 137.5, 133.4, 129.5, 128.0, 127.8, 126.8, 126.2,
7
7, 61
26.2, 13.9, 10.4, ꢀ1.9
1
Triethylpropoxysilane (3d)
145, 117, 103, 87, 75, 59, 47, 45, 29, 27
H (CDCl
3
) 0.47–0.77 (m, 6H), 0.83–1.117 (m, 8H), 1.33–1.57 (c, 4H),
3
.93 (t, 2 H)
1
Dimethylphenylsiloxynonane (3e)
263, 200, 163, 137,121, 91, 69
H (CDCl
3
) 7.67 (m, 2H), 7.43 (m, 3H), 3.73 (t, 2H), 1.43 (s, 14H), 1.07
(
1
q, 5H), 0.49 (s, 6H)
3
1
3
C{ H} (CDCl ) 138.0, 133.3, 129.3, 127.7, 62.7, 32.8, 32.1, 29.5, 26.0,
2
2.8, 14.1 (2C), 13.9, ꢀ1.8
1
1
-Triethylsiloxynonane (3f)
,1,2,2-Tetramethyl-1,2-
diphenyldisiloxane
229, 135, 103, 89, 87, 83, 75, 69, 55
286 (M ), 271, 193, 135, 89
+
1
3
H (CDCl ) d 7.46 (m, 4H), 7.3 (m, 6H) and 0.33 (s, 12 H)
1,3-Hexaethyldisiloxane
217, 189, 161, 133, 105, 80, 66, 59

1802
D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
added to a solution of 3 mol% AuCl(PPh ) and 20 mol% 4
lowed by the subsequent gold-catalyzed conversion of 7a
to 6a. However, we could not detect 8 at all in the reaction
mixture.
3
in 20 mmol 1a, the hydrosilylation was completed at 90 ꢁC
within 25 min (Fig. 2 and see the reaction profile in Fig. S5
in Supplementary material) and the solution remained col-
orless. The addition of 2a can be continued (e.g. 6.44 mmol
and 3.86 mmol of 2a) without the appearance of the purple
color and the formation of a black precipitate until 1a
remains in slight excess.
Next we have studied the possible side reactions caused
by oxygen. Of course, the formation of OPBu is a poten-
3
tial side reaction, which could be avoided by the careful
exclusion of air during all operations. On the contrary,
the oxidation of benzaldehyde (1a) to benzoic acid (9b)
occurs readily opening a door for an unexpected side reac-
tion (Scheme 3), well know for much stronger acids such as
HCl (9a) [6].
These experiments suggest that benzaldehyde (1a) and
PBu (4) themselves or together play an important role in
3
stabilizing the gold catalyst and/or forming the catalyti-
cally active species. It is also evident that the reducing
The reaction of 3 mmol 1a, 0.3 mmol 4, and 0.6 mmol 9b
was monitored by NMR and the formation of 12b could be
power of HSiMe Ph (2a) is high enough to destabilize the
2
3
1
gold (I) catalyst giving rise to gold clusters or particles.
observed. The
P NMR exhibits a broad peak at
33.5 ppm, which becomes a singlet by lowering the temper-
ð2Þ
ature to ꢀ20 ꢁC. This chemical shift is similar to those dis-
played by ionic species containing positively charged
During the investigation of the hydrosilylation of alde-
hydes we have identified several side reactions caused by
the presence of small amounts of water and/or oxygen.
Surprisingly, HSiMe Ph (2a) could react with water in
2
the presence of the PBu -modified [AuCl(tht)] catalyst to
3
give H and (Me PhSi) O (6a) (Eq. 2), confirmed by GC–
2
2
2
MS. We have also established that 2a does not react with
water at room temperature or at 70 ꢁC. The hydrolysis of
product PhCH OSiMe Ph (3a) could have resulted in the
2
2
formation of Me PhSiOH (7a) and PhCH OH (8), fol-
2
2
Scheme 3.
o
1
4 C
6
PPh
3
PBu
3
o
-
-
8 C
*
o
20 C
3
0
20
10
0
-10
-20
-30
(ppm)
3
1
13
3
(4), C-labelled
Fig. 2. In situ IR spectra of the hydrosilylation of neat (20 mmol)
benzaldehyde (1a) in the presence of 0.6 mmol AuCl(PPh ) and 4 mmol
PBu (4) at 90 ꢁC. Me PhSiH (2a) was added in three portions (3.22, 6.44,
and 3.86 mmol) to the reaction mixture.
Fig. 3. P NMR spectra of the in situ reaction of PBu
3
benzaldehyde, and PhCOOH (mol ratio of 1: 10: 0.83) under Ar (with a
capillary filled with PPh
impurity.
3
2
3 3
in CDCl
as internal reference). * Unidentified

D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
1803
Table 3
3
1
1
P{ H} NMR data of the 1-hydroxy phosphonium salts
31
1
Compounds
P{ H} (ppm)
3
3
3
2
4.4
Bu
3
P
OOCPh
C
C
OH
Ph
H
1
2b
5.0
7.0
8.3
Bu
3
P
OOCCH
3
OH
P
Ph
H
1
2c
(HOC
3 6 3
H )
8 17
OOCC H
C
OH
C
8 17
H
H
1
3
1
3
1
13
Fig. 4. C{ H} NMR spectra of the in situ reaction of PBu
labelled benzaldehyde, and PhCOOH (mol ratio: 1:10:0.83) under Ar (with
a capillary filled with PPh in CDCl as internal reference). Zoomed for the
5–88 ppm region to show the CH carbon directly bonded to the
3
(4), C-
Me
3
P
OOCPh
OH
C
3
3
Ph
H
4
1
14
phosphorous atom (d = 67.2 ppm, JPC = 55 Hz).
phosphorous atoms [7]. The exact structure of 12b was
able reaction between 4 and freshly distilled 1a at room
temperature or at ꢀ20 ꢁC. Similar phosphonium salts have
previously been prepared by the addition of strong acids
[6]. Other alkyl phosphines can also react with aldehydes
in the presence of weak acids (Table 3).
Concerning the mechanism involved, our data suggest
that both PBu (4) and the aldehydes play a crucial role in
3
stabilizing the catalytically active gold species. On the con-
trary, the excess of hydrosilanes is detrimental due rapid
reduction of the gold catalyst. Although the formation of
1
3
delineated by using 99.9% C-labelled benzaldehyde. The
3
1
variable temperature
4.4 ppm, with a coupling constant of J = 55 Hz, indi-
P NMR shows a doublet at
1
3
PC
cating the presence of a P–C bond (Fig. 3). This observa-
tion was confirmed by a doublet at 67.2 ppm in the
1
3
1
C{ H} NMR spectrum (Fig. 4). Besides the coupling con-
1
1
stant J = 55 Hz, the coupling constant JCH = 125 Hz is
PC
also observable as expected for a methyne bonded to the
1
phosphorous atom. Finally, the H NMR spectrum shows
a singlet at 12.3 ppm for the acidic proton of the hydroxyl
group. It is important to emphasize, that there is no detect-
a tri-coordinated AuHP
[AuCl(xantphos)]-catalyzed dehydrogenative silylation of
intermediate was proposed for
2
Scheme 4.

1804
D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
alcohols [8], the formation of a mononuclear species with
freshly distilled from sodium benzophenone ketyl (Aldrich)
PBu seems unlikely. The increasing amount of 4 with
under N or Ar. [AuCl(tht)] was prepared by a method pre-
3
2
respect to gold should increase the level of substitution [5]
and thus the rate of the oxidative addition of the hydrosi-
lane should decrease (Scheme 1) resulting in a proportion-
ally slower catalytic reaction. Since the opposite effect was
observed, an alternative mechanism must be operational.
viously reported [10]. The aldehydes (Reanal and Aldrich)
were freshly distilled and stored under Ar. The silanes were
purchased from Aldrich and were used without further
purification. At carbonyl position 99.9% C-enriched
benzaldehyde was purchased from Euriso-top and was dis-
tilled before use and stored under Ar at ꢀ18 ꢁC.
1
3
One possibility is PBu (4) concentration dependent equilib-
3
rium positions between the [Au(PBu ) ]Cl (n = 2, 3, and 4)
3
n
species that may exhibit significantly different reactivity
towards the aldehydes depending on the electronic density
on the gold center (Scheme 4). While [Au(PBu ) ]Cl does
3
.1. Titration of [AuCl(tht)] with PBu (4)
3
3
4
An NMR tube was charged with 0.08 mmol (0.0260 g)
not have an open coordination site to activate the aldehyde,
of [AuCl(tht)] and 350 lL of dry and degassed CDCl3
under Ar and 10 lL PBu (4) was added eight times at
room temperature to the gold solution (molar ratios of
both [Au(PBu ) ]Cl (n = 2 and 3) could perform the activa-
3
n
3
tion of the aldehyde to form oxygen-bonded A2 and A3
adducts, respectively, with significantly different rates.
While gold(I) has little affinity for oxygen donor ligands,
tertiary phosphines have shown to stabilize Au(I)–O bonds.
2
:1, 1:1, 2:3, 1:2, 2:5, 1:3, 2:7 and 1:4).
3.2. General procedure for the hydrosilylation of aldehydes
AuCl(tht)] (0.03 mmol, 0.0098 g) was placed in a Kon-
+
The cation [AuL] (L = phosphine) is isolobal with a pro-
ton and shows a great affinity for bonding to various Lewis
bases [9]. The possibility for side-on coordination of the
aldehyde (C2) cannot be ruled out, but it seems plausible
for [Au(PBu ) ]Cl only. These intermediates could readily
[
tex tube in 0.5 mL of solvent and 0.2 mmol (50 lL) of
PBu3 (4), 1 mmol aldehyde (1a, 100 lL; 1b, 72 lL; 1c,
155 lL) and 0.9 mmol silane (2a, 140 lL; 2b, 140 lL) were
added to the reaction mixture, which was stirred at 70 ꢁC
for 3 h. The reaction vessel was then opened and the reac-
tion mixture was analyzed by NMR or GC–MS.
3
2
d+
dꢀ
undergo a concerted addition of the Si –H -bond to the
d+
dꢀ
C
–O -bond resulting in the formation of a coordinated
alkoxysilanes B2 and B3, which readily eliminate alkoxysi-
lane to regenerate the gold catalyst. The accelerating effect
of the PBu (4) could be also the result of converting A2
3
to A3 or B2 to B3.
3.3. Reaction of weak acids with aldehydes and phosphines
While the operation of a novel mechanism for gold
could lead to new applications in organic chemistry, the
stabilizing role of one of the substrates, e.g. the aldehydes,
is unusual in homogeneous transition-metal catalysis and
indeed surprising.
1
3.3.1. Reaction with C-labelled benzaldehyde
3
1
3
An NMR tube was charged with 3 mmol (300 lL) C-
benzaldehyde (1a), 0.3 mmol (74 lL) PBu (4), 0.25 mmol
3
(0.030 g) benzoic acid under Argon. The NMR spectra
were recorded at 300, 285, 265 and 253 K.
3
. Experimental
3.3.2. Reaction of aldehydes and phosphines in the presence
of weak acids
All manipulations and NMR experiments involving air-
or water-sensitive reagents were performed under an inert
atmosphere of dry argon with the use of Schlenk tech-
niques, and all solvents were dried and degassed before
3.3.2.1 An NMR tube was charged with 1.5 mmol
(150 lL) benzaldehyde (1a), 1.0 mmol (250 lL) PBu (4),
3
1
13
31
use. H, C and P NMR spectra were obtained on Bruker
ARX-250, Bruker Avance-400, and Bruker DRX-500
instruments in solvents like CDCl or CD Cl or neat reac-
and 1.35 mmol acetic acid (77 lL) under Argon and the
NMR spectra were recorded at RT.
3.3.2.2 An NMR tube was charged with 1.9 mmol
(325 lL) nonanal (1c) and 1.9 mmol (0.395 g) tris-hydroxy-
propylphosphine, and 0.04 mmol (0.006 g) nonanoic acid
under Argon and the NMR spectra were recorded at RT.
3.3.2.3 An NMR tube was charged with 3.45 mmol
3
2
2
tants by using a PPh insert (0.1 M PPh in CDCl ). Chem-
3
3
3
1
ical shifts are quoted relative to Si(CH ) (external, H and
C), H PO (internal standard, P). The in situ IR spectra
3
4
1
3
31
3
4
were recorded on a Mettler-Toledo’s ReactIRꢂ 1000
instrument using a SiComp probe. The spectra was
recorded every 2 min. GC–MS spectra were obtained on
an Agilent 6890GC instrument (with an 5973 selective mass
detector) using a HP-5MS column (30 m · 0.25 mm ·
(345 lL) benzaldehyde (1a), 0.23 mmol (24 lL) PMe (15)
3
and 0.51 mmol (0.063 g) benzoic acid (mol ratio 1:15:2.2)
in CD Cl under Ar and the NMR spectra were recorded
2
2
at RT.
0
.25lm).
Commercially available CDCl (Aldrich) and CD Cl
2
Acknowledgements
3
2
(
Aldrich) were dried with CaH (Aldrich) under N or
2 2
Ar. CH Cl (Aldrich) and CH CN (Reanal) were distilled
This work was partially supported by the Hungarian
National Scientific Research Fund (OTKA-T032850), the
2
2
3
from CaH (Aldrich) under N or Ar. THF (Reanal) was
2
2

D. Lantos et al. / Journal of Organometallic Chemistry 692 (2007) 1799–1805
(e) A.S.K. Hashmi, Gold Bull. 37 (2004) 1;
1805
Universidad de Zaragoza-Ibercaja (IBE2004B-CIE-04), the
(
f) C. Nieto-Oberhuber, S. L o´ pez, A.M. Echavarren, J. Am. Chem.
Soc. 127 (2005) 6178;
g) A.S.K. Hashmi, Angew. Chem., Int. Ed. 44 (2005) 6990;
h) J. Zhang, C.-G. Yang, C. He, J. Am. Chem. Soc. 128 (2006) 1798;
MEC–Universidad de Zaragoza, Research Contract ‘‘Ra-
m o´ n y Cajal’’, and CSIC-MTA (Spanish-Hungarian Col-
laborative Project 2004HU0003). S.S. would like to thank
the MEC-CSIC for a FPI PhD grant. The donation of
the ReactIR 1000 by Mettler-Toledo AutoChem Inc. is
greatly appreciated.
(
(
(i) A. Comas-Vives, C. Gonz a´ lez-Arellano, A. Corma, M. Iglesias,
nchez, G. Ujaque, J. Am. Chem. Soc. 128 (2006) 4756.
F. Sa
´
[
2] B. Marciniec, J. Gulinski, H. Maciejewski, in: I.T. Horv a´ th, E.
Iglesia, M.T. Klein, J.A. Lercher, J.A. Russell, E.I. Stieffel (Eds.),
Encyclopedia of Catalysis, vol. 4., Wiley-Interscience, 2003, p. 107.
3] H. Ito, T. Yajima, J. Tateiwa, A. Hosomi, Chem. Commun. (2000) 981.
4] D. Lantos, M. Contel, A. Larrea, D. Szab o´ , I.T. Horv a´ th, QSAR
Comb. Sci. 25 (2006) 719.
Appendix A. Supplementary data
[
[
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.jorganchem.
[5] (a) C.B. Colburn, W.E. Hill, C.A. McAuliffe, R.V. Parish, J. Chem.
Soc., Chem. Commun. (1979) 218;
2
006.10.025.
(
b) M.C. Gimeno, A. Laguna, Chem. Rev. 97 (1997) 511.
6] (a) C.A. Fyfe, M. Zbozny, Can. J. Chem. 50 (1972) 1713;
b) F. Ram´ırez, J.F. Pilot, C.P. Smith, Tetrahedron 22 (1966) 567.
7] S.W. Lee, W.C. Trogler, J. Org. Chem. 55 (1990) 2644.
[
References
(
[
[
1] Reviews and some selected examples (a) R. Casado, M. Contel, M.
Laguna, P. Romero, S. Sanz, J. Am. Chem. Soc. 125 (2003) 11925;
[8] H. Ito, K. Takagi, T. Miyahara, M. Sawamura, Org. Lett. 7 (2005)
3001.
(
(
(
(
b) X. Yao, C.-J. Li, J. Am. Chem. Soc. 126 (2004) 6884;
c) D.E. De Vos, B.F. Sels, Angew. Chem., Int. Ed. 43 (2004) 2;
d) M.R. Luzung, J.P. Markham, F.D. Toste, J. Am. Chem. Soc. 126
2004) 10858;
[9] J.P. Fackler Jr., W.E. van Zyl, B.A. Prihoda, in: H. Schmidbaur
(Ed.), Gold, Progress in Chemistry, Biochemistry and Technology,
20, Wiley, 1999, p. 798.
[10] R. Uson, A. Laguna, M. Laguna, Inorg. Synth. 26 (1989) 85.