Development of a method for preparing a highly reactive and stable, recyclable and environmentally benign organopalladium catalyst supported on sulfur-terminated gallium arsenide(001): A three-component catalyst, {Pd}-S-GaAs(001), and its properties

Article abstract of DOI:10.1002/adsc.200606025

We have developed a method for preparing a recyclable and environmentally benign organopalladium catalyst for the Heck reaction supported on sulfur-terminated gallium arsenide(001). This three-component catalyst, {Pd}-S-GaAs(001), exhibited high stability and activity, furthermore, it tolerated reuse in 10 runs of the Heck reaction (average yield, 97%) under aerobic conditions. The sulfur layer was very important to stabilize this catalyst. Only trace amounts of Pd were leached from this catalyst to the reaction mixture, as measured by ICP-mass. The valence of immobilized Pd was zero by XPS spectrometry.

Full text of DOI:10.1002/adsc.200606025

FULL PAPERS
DOI: 10.1002/adsc.200606025
Development of a Method for Preparing a HighlyReactive and
Stable, Recyclable and Environmentally Benign Organopalladium
Catalyst Supported on Sulfur-Terminated Gallium Arsenide
ACHTREUNG
A Three-Component Catalyst, {Pd}-S-GaAs
ties
ACHTREUNG
Mitsuhiro Arisawa,a, d,* Masahiro Hamada,a Ikuko Takamiya,a
Masahiko Shimoda,b Shiro Tsukamoto,c Yasuhiko Arakawa,c
and Atsushi Nishidaa,*
a
Graduate School of Pharmaceutical Sciences, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba 263-8522, Japan
Fax : (+81)-43-290-2909; e-mail: arisawa@pharm.hokudai.ac.jp or nishida@p.chiba-u.ac.jp
National Institute for Materials Science, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan
Nanoelectronics Collaborative Research Center, University of Tokyo, 4-6-1 Komaba, Meguro, Tokyo 153-8505, Japan
Present address: Graduate School of Pharmaceutical Sciences, Hokkaido University, Kita12, Nishi 6, Kita-ku, Sapporo
b
c
d
060-0812, Japan
Received: January 24, 2006; Accepted: March 25, 2006
Abstract: We have developed a method for prepar- layer was very important to stabilize this catalyst.
ing a recyclable and environmentally benign organo- Only trace amounts of Pd were leached from this
palladium catalyst for the Heck reaction supported catalyst to the reaction mixture, as measured by ICP-
on sulfur-terminated gallium arsenide
A
three-component catalyst, {Pd}-S-GaAs(001), exhib- XPS spectrometry.
AHCTREUNG
ited high stability and activity, furthermore, it toler-
ated reuse in 10 runs of the Heck reaction (average Keywords: Heck reaction; heterogeneous catalysis;
yield, 97%) under aerobic conditions. The sulfur immobilization; palladium; semiconductors; sulfur
Introduction
be active, leach-free and easy to handle, and have re-
ported an entirely novel organopalladium catalyst
Organopalladium catalysts are very important in or- supported on a semiconductor surface, GaAs(001),
G
ganic chemistry and widely used in the synthesis of terminated by elemental sulfur evaporation using mo-
biologically active compounds.[1] The Heck reaction lecular beam epitaxy (MBE)[5] or ammonium sulfide
has received considerable attention because it offers a solution (Figure 1).[6] As far as we know, the ability of
reliable method for carbon-carbon bond formation.[2] a semi-conductor to support an organometallic cata-
Like many organic reactions using organometallic re- lyst has not been reported previously. We expected
agents and catalysts, the standard procedure under this new material to amplify the recyclability or to ac-
homogeneous conditions suffers from wasted noble tivate differently the aryl halides. In these reports, we
metals or catalytically active metals which are difficult have presented the following findings: 1) a {Pd}-S-
to recover and lost in aqueous work-up. With regard GaAs
ACHTREUNG
to green chemistry, heterogeneous organometallic cat- efficiently than Pd
alysts are favored and currently being extensively
studied.[3] However, most of these systems do not
give as high a level of activity as homogeneous cata-
lysts. Recently, it has reported that some heterogene-
ous catalysts showed a higher activity than homogene-
ous ones, and the mechanism was studied in detail.[4]
ACHTREUNG
We have been interested in the development of
more practical heterogeneous catalysts, which should
Figure 1. Preparation of three-phase catalyst, {Pd}-S-GaAs-
(001), catalyst A.
AHCTREUNG
Adv. Synth. Catal. 2006, 348, 1063 – 1070
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1063

Mitsuhiro Arisawa et al.
FULL PAPERS
homogeneous catalyst, and could be reused at least 10 times (Scheme 1). The chemical yield of the 10th run,
times, 2) inactivated {Pd}-S-GaAs(001) could be revi- however, was only 24% and the average chemical
talized by further treatment with Pd(PPh3)4, 3) sulfur yield of 10 runs was 56% (Table 1, entry 1). To identi-
termination in this catalyst was important for stability fy a more reactive catalyst, we surveyed the sources
of this catalyst.[5,6]
of Pd. Catalysts B – F were prepared by the same
In this paper, we present a method for preparing method as catalyst A except for the source of Pd, and
{Pd}-S-GaAs(001), which is drastically improved in then repeatedly subjected to the Heck reaction as
AHCTREUNG
AHCTREUNG
ACHTREUNG
both catalytic activity and stability. Some chemical shown in Scheme 1. The results are summarized in
and physical properties of the catalyst were elucidated Table 1. Catalyst B, prepared with PdCl2, showed
using X-ray photoelectron spectroscopy (XPS), and almost no activity (entry 2). Catalysts C and D, pre-
scanning electron microscopy (SEM). Furthermore, pared with Pd
N
N
experiments were carried out in order to attempt ly, were unstable and lost most of their catalytic activ-
identification of the real active Pd species.
ity before the 7th run (entries 3 and 4). Catalyst E,
prepared with Pd(dba)2, retained its activity until the
AHCTREUNG
10th run, although its activity was lower than that of
Results and Discussion
catalyst A (entry 5). In contrast, catalyst F, prepared
with Pd
(001), lyst A and the coupling product was obtained in 79%
(OAc)2 is
step procedure, (NH4)2Sx treatment (aqueous solu- the best source of Pd among those tested for prepar-
tion, 608C, 30 min.), and Pd adsorption [Pd(PPh3)4, ing {Pd}-S-GaAs(001).
7.2 mM in MeCN, 1008C, 12 h] followed by washing Next, we examined heated-washing conditions after
(OAc)2, had a much higher activity than cata-
As we have previously reported,[6] {Pd}-S-GaAs
ACHTREUNG
which is named catalyst A, was prepared by a three- yield in the 10th run (entry 6). Therefore, Pd
ACHTREUNG
A
ACHTREUNG
at room temperature with acetonitrile (MeCN). This Pd adsorption, since it is possible that excess Pd on
catalyst A could be used for the Heck reaction of io- surface, which is not bound to sulfur, could deactivate
dobenzene (1a) and methyl acrylate (2a) over 10 {Pd}-S-GaAs(001) by forming a less active or inactive
N
colloidal Pd species.[7] Hence, catalysts G and H were
prepared by a new three-step procedure, which con-
sisted of (NH4)2Sx treatment (608C, 30 min.), Pd ad-
sorption [Pd
N
N
heated-washing in refluxing MeCN for 12 h. The cata-
lysts were subjected to the same Heck reaction
(Scheme 1) over 10 times, respectively. As shown in
Table 2, the activity of catalyst G, which was prepared
Scheme 1. Heck reaction.
Table 1. Effect of Pd sources in Pd adsorption.
Entry
Cat.
Pd source
Yield of 3a [%][a]
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
ave.
1
2
3
4
5
6
A
B
C
D
E
F
Pd
PdCl2
Pd
Pd
Pd
Pd
N
93
1
51
43
74
100
87
1
31
8
66
99
71
0
14
1
31
91
72
-
15
0
46
98
57
-
10
1
12
97
54
-
12
-
9
88
35
-
1
-
5
37
-
-
-
1
25
-
-
-
1
24
-
-
-
4
56
-
-
-
25
88
81
72
71
79
[a]
Yields of isolated products.
Table 2. Effects of washing conditions after Pd adsorption.
Entry
Cat.
Pd source
Heated-washing
Yield of 3a [%][a]
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
ave.
1
2
3
4
A
G
F
Pd
Pd(PPh3)4
Pd(OAc)2
Pd(OAc)2
C
No
Yes
No
Yes
93
90
100
96
87
55
99
100
71
67
91
91
72
66
98
93
57
57
97
98
54
54
88
100
35
46
81
96
37
43
72
99
25
43
71
98
24
39
79
95
56
56
88
97
H
[a]
The yields were determined by NMR analysis.
1064
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1063 – 1070

A Three-Component Catalyst, {Pd}-S-GaAs(001), and its Properties
A
FULL PAPERS
Figure 2. Chemical yield and amount of leached Pd using
catalyst H.
Figure 3. Chemical yield and amount of leached Pd using
catalyst N. Unit of leached Pd is ng not mg.
with Pd(PPh3)4 followed by heated-washing, was the H, which has been washed in MeCN at 1008C(bath
A
same as that of catalyst A, which was not treated by temperature), had the best catalytic activity. Next, we
heated-washing. In contrast, catalyst H, which was continued to examine the effect of solvent in heated-
prepared with Pd(OAc)2, had a higher activity than washing at higher temperature (1358C, bath tempera-
A
catalyst F, that is, the catalytic activity of H did not ture). As the result, we found that toluene can be an-
decrease over 10 runs (95%, entry 4), and the average other candidate for heated-washing, but its catalytic
chemical yield over 10 runs was 97%.
activity was lower than that of catalyst H (entry 6). In
We measured the amount of leached Pd into the re- entry 7, we carried out heated-washing in xylene at
action solution after the Heck reaction using catalyst 1358C. The activity of catalyst N was as high as that
H by inductively coupled plasma mass spectrometry of catalyst H with good reproducibility and catalyst N
(ICP-mass). Figure 2 summarizes the chemical yield kept its high activity through 10 runs (entry 7). When
of the product and the amount of leached Pd in each the heated-washing was carried out in xylene at
run. While the chemical yield was almost quantitative 1508C, the activity of catalyst O decreased slightly
through 10 runs, the amount of leached Pd decreased (entry 8).
after 4 uses. The amount of Pd leached after the 5th
When catalyst N was employed, ICP-mass analysis
run is a few mg, less than 1.7 ppm. Considering that indicated that the amount of Pd leached into the reac-
the amount of immobilized Pd after the 10th run is tion mixture was extremely low in each run (0.04 –
125 mg, only trace amounts of Pd were leached after 0.26 ppm, Figure 3). These results suggested that
5th run.
heated-washing in a non-coordinating solvent can pre-
Unfortunately, after several experiments, we found pare highly reactive and stable catalysts with good re-
that the reproducibility of the catalytic activity for producibility and lower leaching, although further ex-
catalyst H was poor. Hence, we optimized the heated- periments are required to explain why these differen-
washing conditions. First, we carried out the heated- ces in catalytic activities come about, merely by
washing using some solvents at 1008C(Table 3, en-
changing solvent and temperature in heated-washing.
tries 1 – 4). Among the solvents we examined, catalyst
Table 3. Optimization of heated-washing conditions.
Entry
Cat.
Heated-washing
Solvent
Yield of 3a [%][a]
T [8C][b]
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
ave.
1
2
3
4
5
6
7
8
H
I
J
K
L
M
N
O
MeCN
toluene
DMF
DMSO
DMF
toluene
xylene
xylene
100
100
100
100
135
135
135
150
96
98
98
88
70
95
96
99
100
96
86
78
80
93
100
99
91
92
95
64
78
93
99
100
93
82
89
38
58
97
98
97
98
82
80
21
52
94
100
99
100
63
84
17
49
91
91
91
96
55
75
16
43
91
96
97
99
48
77
12
32
90
97
94
98
62
69
13
31
90
92
92
95
63
70
14
31
85
91
86
97
74
82
36
52
92
96
95
[a]
The yields in entry 1 were determined by NMR analysis. The yields in entries 2 to 8 were determined by HPLCanalysis.
[b] Bath temperature.
Adv. Synth. Catal. 2006, 348, 1063 – 1070
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1065

Mitsuhiro Arisawa et al.
FULL PAPERS
Table 4. Heck reaction of various aryl halides.
Entry Substrate [Ar-X]
No. Ar
Product
3a[b]
Yield of 3a [%][a]
X
1st
2nd 3rd 4th
5th
6th
7th
8th
9th 10th ave.
1
2
3
4
5
6
7
8
1b
1c
1d
1e
1f
1g
1h
1i
C6H5
C6H5
Br
0
0
98
0
0
90
0[c]
0[c]
94
0[d]
0[d]
81
0[e]
0[e]
72
-
-
63
56
-
-
52
59
-
-
49
47
-
-
43
46
-
-
0[f]
0[f]
68
68
99
76
54
56
OTf 3a[b]
2-MeC6H4
3-MeC6H4
4-MeC6H4
4-MeOC6H4
2-BrC6H4
3-BrC6H4
4-BrC6H4
4-NO2C6H4
4-MeC(O)C6H4
xylene
I
I
I
I
I
I
I
I
I
I
3b
3c
3d
3e
3f
3g
3h
3i
36
38
96
54
25
29
42
-
100 100
100 100
100 92
100 72
62
100 100 100 100 100 100 98
100 87
75
65
47
65
0
89
58
39
49
-
68
36
40
42
-
48
28
38
41
-
50
28
25
38
-
92
71
80
81
67
10
51
57
57
56
4
100 100
9
1j
1k
1l
98
82
91
99
81
34
59
58
10
11
12
26[f]
57[f]
94
3j[h]
5
44
38
94
-
-
-
92
-
-
4
100
100 99
100 83
91
83
[a]
1
Yields were determined by H NMR spectra; PhNO2 (0.25 mmol) was employed as an internal standard.
[b] Yields were determined by HPLCanalysis.
[c]
Reaction time was 24 h.
[d] Reaction was carried out in toluene at 1358C(bath temperature) for 12 h.
[e]
Reaction was carried out in toluene at 1358C(bath temperature) for 24 h.
[f] Average yield through 1 to 5 runs.
[g]
0.50 mmol of acetophenone was used as an internal standard.
Next, to probe the scope and limitations of catalyst
N, it was subjected to the Heck reaction of other sub-
strates. The results are summarized in Tables 4 and 5.
Bromobenzene (1b) and phenyl trifluoromethanesul-
fonate (1c), which are widely used as a substrate for
Heck reaction,[8] were inactive under our conditions
(Table 4, entries 1 and 2). In these cases, catalyst N,
which was used over 5 runs, still possessed catalytic
activity for the Heck reaction between 1a and 2a,
therefore, we assume that our catalyst could not have
undergone oxidative addition with 1b and 1c under
these reaction conditions. When 2-, 3-, and 4-iodoto-
luene were used as substrates, chemical yields of the
corresponding product were varied depending on the
position of the methyl group on the aromatic ring (en-
tries 3 – 5). When 4-iodoanisole (1g) was used as a
substrate, the reaction proceeded, although the yields
decreased gradually (entry 6). When bromo-1-iodo-
benzene (1h–j) was used, the reaction proceeded che-
moselectively to give the corresponding bromocinna-
mate, but the reactivity also gradually decreased (en-
tries 7–9). In entries 10 and 11, aryl iodides with elec-
Figure 4. XPS spectra of catalyst H.
tron-withdrawing groups at the para-position were
used to yield the corresponding products at the 1st
run, however, after the 2nd run, the chemical yields as a substrate and gave the corresponding product in
decreased drastically, and no reactivity remained after excellent yield over 10 runs and the average yield was
the 5th run. In entry 12, 1-iodonaphthalene, which 94%. The fact that non- or de-activated aryl iodides
was sometimes known for its low activity,[9] was used react better than activated ones (entries 3 – 6 versus
1066
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1063 – 1070

A Three-Component Catalyst, {Pd}-S-GaAs(001), and its Properties
A
FULL PAPERS
Figure 5. SEM observation of catalyst H.
Table 5. Heck reaction of methyl vinyl ketone.
Entry
Substrate [2]
Product
Yield of 3a [%][a]
R
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
10th
ave.
1
2
2a
2b
OMe
Me
3a
3k
96
76
100
80
99
79
98
78
100
77
91
76
96
72
97
75
92
74
91
71
96
76
[a]
1
Yields were determined by H NMR spectra; PhNO2 (0.25 mmol) was employed as an internal standard.
8 – 9) might suggest that the use of the support modi- Catalysis Properties
fies slightly the reaction mechanism, or displaces the
limiting step of the reaction from the oxidative addi- Based on above discoveries, we investigated the char-
tion of the aryl halide to Pd(0) to the reductive elimi- acteristics of catalyst H in detail. First, the status of
nation to give the expected compound, although we Pd on the surface both before and after these reac-
do not have enough data to discuss this matter in tions, was examined by XPS measurement[11] of cata-
detail at this stage.
lyst H using Mg Ka radiation (K.E. = 1253 eV) in a
When methyl vinyl ketone (2b) was used instead of separate ultrahigh vacuum chamber. The energy shift
2a, the chemical yield of the 1st run was 76%,[10] but due to sample charging was corrected as the C1s
almost the same chemical yields were maintained peak is located at 285.0 eV. The XPS spectrum of cat-
through 10 runs (Table 5).
alyst H before the Heck reaction exhibited clear
From these results, we found that our catalyst could peaks of Pd 3d core-level photoemission, indicating
catalyze only aryl iodides, and the catalytic activities that the organopalladium was definitely immobilized
were also influenced by the substituent on aryl io- as expected (spectrum I in Figure 4). The peak width
dides.
is slightly wider than that of metal Pd (not shown),
Adv. Synth. Catal. 2006, 348, 1063 – 1070
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1067

Mitsuhiro Arisawa et al.
FULL PAPERS
dobenzene (7) according to the Daviesꢁ protocol,[14]
and carried out the Heck reactions of 1a and 7 with
2a in the presence of {Pd}-S-GaAs, catalyst N
(Scheme 2). After the reaction mixture had been re-
fluxed for 12 h, the catalyst was removed. Then the
reaction mixture was filtered. From the filtrate, 2% of
coupling product 3a was obtained, while the resin was
treated with an excess amount of trifluoroacetic acid
(TFA) at room temperature for 2 h to give 8 in 30%
yield. As a control, we ran the Heck reaction of only
7 using {Pd}-S-GaAs. In this case, the coupling prod-
uct 8 was not obtained. Furthermore, using the homo-
geneous catalyst, Pd(OAc)2, the Heck reaction pro-
A
ceeded to give 8 in 62% yield.
Next, we examined hot and cold filtration tests to
see whether leached Pd species have catalytic activity
for the Heck reaction.[15] A mixture of 1a, 2a, Et3N,
and {Pd}-S-GaAs in MeCN was refluxed for 6 h with-
out stirring, and then the catalyst was removed from
the ꢂhotꢁ mixture. The yield of 3a was 13% in this
step. The ꢂhotꢁ solution was continuously refluxed
without {Pd}-S-GaAs. After 24 h (total time 30 h), the
yield of 3a was 35%. In addition, a mixture of 1a, 2a,
Et3N, and {Pd}-S-GaAs in MeCN was refluxed for 6 h
without stirring. After the mixture was cooled to
Scheme 2. Three-phase test.
suggesting a complex chemical environment and/or room temperature, the catalyst was removed from the
partial oxidation of Pd. The binding energy of the Pd ꢂcoldꢁ mixture. The yield of 3a was 13% in this step.
3d5/2 peak, 335.9 eV, is close to that of metal Pd, The ꢂcoldꢁ clear solution was refluxed for 24 h (total
335.0 eV, indicating that the valence of Pd is zero,[12] time 30 h) without {Pd}-S-GaAs to give 3a in 28%
while that of the Pd source, Pd(OAc)2, is 2. Even yield. In both cases, the amount of product increased,
A
after repeated cycles of the Heck reaction, catalyst H even when the catalyst was removed from the reac-
showed almost the same Pd 3d peaks (spectrum II in tion mixture. However, these chemical yields were
Figure 4), suggesting that the oxidation state of the much lower than that of the reaction in which {Pd}-S-
immobilized Pd did not changed in repeated the GaAs was employed from the beginning to the end
Heck reactions.
(Figure 6).
Second, to obtain further information about the
Considering that there is no relationship between
surface both before and after 10 runs of the Heck re- chemical yield and amount of leached Pd (Figures 2
action, catalyst H was directly surveyed by SEM and and 3) for the Heck reaction, these results indicate
compared to the GaAs
Figure 5, catalyst H before the Heck reaction had a cies was leached from {Pd}-S-GaAs into the reaction
characteristic structure on GaAs(001) (Figure 5b) and mixture. 2) The leached Pd species has some activity
A
AHCTREUNG
after 10 runs of the Heck reaction, the surface struc- in the Heck reaction. 3) The catalytic activity of
ture was obviously changed (Figures 5c and d), al- leached Pd is lower than that of {Pd}-S-GaAs. 4) For
though the catalytic activity showed no significant dif- a highly efficient Heck reaction, the presence of {Pd}-
ference between the 1 st run and the 10th run of the S-GaAs is critical, although we do not know whether
Heck reaction.
{Pd}-S-GaAs works as the actual active species or as
the pre-catalyst.
Mechanistic Study
Conclusions
For the development of a novel heterogeneous cata-
lyst, to identify the real catalyst is one of the impor- We have successfully developed a method for prepar-
tant problems.[4] First, we examined a three-phase ing {Pd}-S-GaAs
(001), the catalytic activity of which
G
test.[13] This test can clarify which soluble or insoluble for the Heck reaction does not decrease after 10 uses
species is the real active catalyst by running a metal- and which shows only trace amounts of Pd leached
catalyzed reaction at the presence of a substrate im- into the reaction mixture after the 5th run. In
mobilized on a polymer. We prepared immobilized io- addition, it was clear that the immobilized Pd on S-
1068
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1063 – 1070

A Three-Component Catalyst, {Pd}-S-GaAs(001), and its Properties
A
FULL PAPERS
Figure 6. Hot and cold filtration tests.
in a flask and dried at room temperature for 10 min, then
the sample was heated at 250 – 3008Cusing a heat-gun for
20 min under reduced pressure (ca. 6 mmHg). The resulting
sulfur-terminated GaAs plate [S-GaAs] was placed in a so-
lution of Pd source (see Table 1) in MeCN (3 mL) and stir-
red at 1008C(bath temperature) for 12 h under an argon at-
mosphere. The plate was rinsed with MeCN (3 mL ”100)
until the washings showed no activity in the Heck reaction,
as shown in Scheme 1 and dried under vacuum to give {Pd}-
S-GaAs.
GaAs(001) is in the oxidation state Pd(0). This cata-
A
lyst is easy to recover and transfer under air using a
pair of tweezers. We hope that this catalyst might lead
to new opportunities in environmentally benign im-
mobilized catalysts and microreactors.
Experimental Section
Catalysts G – O (heated-washing conditions): After sulfur-
terminating on GaAs by the same procedure as for catalysts
1H NMR spectra were obtained on JEOL JNM-GSX 400a
and ecp400 spectrometers with tetramethylsilane as the in-
ternal standard. HPLCwas carried out using a Cica Mighty-
sil RP-18 and the HPLCspectra were detected by a Hitachi
variable wavelength UV monitor 638–41 (254 nm) and re-
corded by a Hitachi D-2500. ICP-mass spectra were ob-
tained on Hewlett Packard 4500 series and 7500 series in-
struments. XPS was carried out using a VG ESCALAB Mk-
II. All air- and moisture-sensitive reactions were carried out
under an argon atmosphere. Heck reactions were carried
out with Eyela ChemiStationTM PPS-1510. MeCN (Kokusan
Chemicals) and DMF (Kanto Chemicals) were distilled with
CaH2 (Wako Chemicals) before use.
A – F, the S-GaAs plate were placed in Pd
A
Pd(OAc)2 (0.022 mmol) in degassed MeCN (3 mL) and stir-
AHCTREUNG
red at 1008C(bath temperature) for 12 h under an argon at-
mosphere. The plate was rinsed with MeCN (3 mL ” 100)
until the washings showed no activity in the Heck reaction
shown in Scheme 1, and dried under vacuum. The plate was
placed in the solvent (3 mL) shown in Tables 2 and 3, and
kept at an appropriate temperature (see Tables 2 and 3) for
12 h under an argon atmosphere. The plate was rinsed with
MeCN (3 mL ” 50) until the washings showed no activity in
the Heck reaction, as shown in Scheme 1 and dried under
vacuum to prepare {Pd}-S-GaAs. ICP-mass (Hewlett Pack-
ard 4500 or 7500 series) revealed that 104 mg to 209 mg of Pd
were adsorbed on the surface.
Preparation of {Pd}-S-GaAs(001)
U
Heck Reaction
Catalysts A – F (non-heated-washing conditions): A 13”11”
0.6 mm3 sample of GaAs
(001) [(Sumitomo Electric Indus-
AHCTREUNG
A mixture of iodobenzene (1a; 56.0 mL, 0.50 mmol), methyl
acrylate (2a; 56.2 mL, 0.63 mmol), and Et3N (87.1 mL,
0.63 mmol) in MeCN (3 mL) was heated in the presence of
{Pd}-S-GaAs at 1008C(bath temperature) for 12 h under an
argon atmosphere without stirring. After the mixture had
been cooled to room temperature, the {Pd}-S-GaAs plate
was removed from the reaction solution and rinsed several
tries, Ltd. for catalysts A – I and K, (Dowa Mining Co.,
Ltd.) for catalysts J and L – O, there are no differences be-
tween suppliers in catalytic activity in our system] was
placed in (NH4)2Sx (ammonium sulfide solution, yellow,
sulfur content: 5 – 7%, 3 mL, Kishida Chemical Co., Ltd.)
at 608Cfor 30 min and then rinsed in succession with H O
2
(3 mL ” 10) and MeCN (3 mL ” 5). The sample was placed
Adv. Synth. Catal. 2006, 348, 1063 – 1070
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.asc.wiley-vch.de
1069

Mitsuhiro Arisawa et al.
FULL PAPERS
times with MeCN. Yields of methyl trans-cinnamate (3a)
were determined by isolated weight, 1H NMR analysis
(0.25 mmol of nitrobenzene was used as an internal stan-
dard), or HPLCanalysis (0.05 mmol of nitrobenzene was
used as an internal standard). In addition, the amount of Pd
species in the reaction mixture was examined by ICP-mass
after treatment with concentrated nitric acid for a few
weeks. The recovered {Pd}-S-GaAs plate was again subject-
ed to the above reaction conditions as a 2nd run. This proce-
dure was repeated for a total of 10 runs.
Akiyama, M. Ueno, T. Kitamori, S. Kobayashi, Science
2005, 304, 1305–1308.
[4] a) Y. Ji, S. Jain, R. J. Davis, J. Phys. Chem. B 2005, 109,
17232–17238; b) F. Zhao, M. Shirai, Y. Lkushima, M.
Arai, J. Mol. Catal. A: Chem. 2002, 180, 211–219.
[5] M. Arisawa, S. Tsukamoto, M. Shimoda, M. Pristovsek,
A. Nishida, Jpn. J. Appl. Phys. 2002, 41, L1197–L1199.
[6] I. Takamiya, S. Tsukamoto, M. Shimoda, N. Miyashita,
M. Arisawa, Y. Arakawa, A. Nishida, Chem. Lett. 2004.
33, 1208–1209.
[7] a) W. A. Herrmann, C. Brossmer, K. ꢃfele, M. Beller,
H. Fisher, J. Mol. Catal. A: Chem. 1995, 103, 133–146;
b) M. Beller, H. Fisher, K. Kꢄhlein, C. P. Reisinger,
W. A. Herrman, J. Organomet. Chem. 1996, 520, 257–
259; c) L. Djakovitch, K. Koehler, J. Am. Chem. Soc.
2001, 123, 5990–5999.
Acknowledgements
This research was supported by an IT program and a Grant-
in-Aid for Exploratory Research and the Encouragement of
Young Scientists (A) from the Ministry of Education, Cul-
ture, Sports, Science and Technology, Japan. M. A. is also
grateful for financial support from Iketani Foundation,
Yazaki Memorial Foundation for Science and Technology
and a Mitsui Chemical Ltd. Award in Synthetic Organic
Chemistry. We also thank Professors Dr. K. T. Suzuki and
Dr. Y. Ogra, Chiba University, for experiments and discus-
sions regarding ICP-mass used in this research. We thank
Keyence Corporation for conducting the SEM experiments.
[8] M. Gruber, S. Chouzier, K. Koehler, L. Djakovitch,
Appl. Catal. A: Gen. 2004, 265, 161–169.
[9] Z. Xiong, N. Wang, M. Dai, A. Li, J. Chen, Z. Yang,
Org. Lett. 2004, 6, 3337–3340.
[10] Reactivity of methyl vinyl ketone is usually lower than
that of methyl acrylate. A 76% of chemical yield is
useful in the literature; a) L. Botella, C. Nꢅjera, J. Org.
Chem. 2005, 70, 4360–4369; b) L. Botella, C. Nꢅjera,
Tetrahedron Lett. 2004, 45, 1833–1836; c) D. Yang, Y.-
C. Chen, N.-Y. Zhu, Org. Lett. 2004, 6, 1577–1580;
d) S. Cacchi, G. Fabrizi, A. Goggiamani, Arkivoc 2003,
58–66; e) H. Hagiwara, Y. Shimizu, T. Hoshi, T.
Suzuki, M. Ando, K. Ohkubo, C. Yokoyama, Tetrahe-
dron Lett. 2001, 42, 4349–4351; f) J. H. Clark, D. J.
Macquarrie, E. B. Mubofu, Green Chem. 2000, 2, 53–
56; g) S. Cacchi, G. Fabrizi, F. Gasparrini, C. Villani,
Synlett 1999, 345–347.
References
[1] a) E. Negishi, Handbook of Organopalladium Chemis-
try for Organic Synthesis, Wiley Interscience, New
York, 2002; b) J. Tsuji, Palladium Reagents and Cata-
lysts, Wiley, New York, 2004.
[11] M. Shimoda, S. Tsukamoto, N. Koguchi, Surf. Sci. 1998,
402, 669–672.
[2] a) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000,
100, 3009–3066; b) A. Biggis, M. Zecca, M. Basato, J.
Mol. Catal. A 2001, 173, 249–274.
[3] a) Q.-H. Fan, Y.-M. Li, A. S. C. Chan, Chem. Rev. 2002,
102, 3385–3466; b) N. E. Leadbeater, M. Marco.
Chem. Rev. 2002, 102, 3217–3274; c) C. E. Song, S.-G.
Lee, Chem. Rev. 2002, 102, 3495–3524; d) J. Dupont,
R. F. de Souza, P. A. Z. Auarez, Chem. Rev. 2002, 102,
3667–3692; e) J. Kobayashi, Y. Mori, K. Okamoto, R.
[12] The palladium(II) 3d5/2 peak from PdCl2 is observed at
338.3 eV.
[13] a) J. Rebek, F. Gavina, J. Am. Chem. Soc. 1974, 96,
7112–7114; b) J. Rebek, D. Brown, S. Zimmerman, J.
Am. Chem. Soc. 1975, 97, 454–455.
[14] I. A. Davies, L. Matty, D. L. Heghes, P. J. Reider, J.
Am. Chem. Soc. 2001, 123, 10139–10140.
[15] C. Baleizao, A. Corma, H. Garcia, A. Leyva, J. Org.
Chem. 2004, 69, 439–446.
1070
www.asc.wiley-vch.de
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2006, 348, 1063 – 1070