SnCl2-mediated carbonyl allylation in fully aqueous media

Article abstract of DOI:10.1016/j.tet.2004.05.046

Systematic studies were performed on SnCl₂-mediated carbonyl allylation reaction between aldehydes and allyl halides in fully aqueous media. Totally three valuable reaction systems were discovered, which were SnCl ₂/CuCl₂, SnCl₂/TiCl₃, and SnCl ₂/PdCl₂. They all provided good to excellent yields in the allylation of aliphatic and aromatic aldehydes under very mild and convenient conditions. SnCl₂, by itself, was also found to be effective for the allylation reaction when allyl bromide was employed. However, the SnCl ₂-only reaction could only tolerate very small amount of water as the solvent. The SnCl₂/CuCl₂, SnCl₂/TiCl ₃, and SnCl₂/PdCl₂-mediated reactions exhibited good regioselectivity favoring the γ-adduct when cinnamyl halides were employed as the allylation reagent. The same reactions with cinnamyl halides also showed good diastereoselectivity favoring the anti-product. Mechanistic studies using proton NMR techniques suggested that the additive (i.e., CuCl ₂, TiCl₃, PdCl₂) could accelerate the formation of allyltin intermediate, but this step was shown not to be the most important for the allylation. Thus we proposed that the Lewis acid catalysis effect exerted by the additive was the main reason for the observed reactivity enhancement.

Full text of DOI:10.1016/j.tet.2004.05.046

Tetrahedron 60 (2004) 6129–6136
SnCl -mediated carbonyl allylation in fully aqueous media
2
*
Xiang-Hui Tan, Yong-Quan Hou, Chao Huang, Lei Liu and Qing-Xiang Guo
*
Department of Chemistry, University of Science and Technology of China, Hefei 230026, China
Received 26 February 2004; revised 9 April 2004; accepted 19 May 2004
Available online 5 June 2004
Abstract—Systematic studies were performed on SnCl
aqueous media. Totally three valuable reaction systems were discovered, which were SnCl /CuCl , SnCl /TiCl , and SnCl /PdCl . They all
2
-mediated carbonyl allylation reaction between aldehydes and allyl halides in fully
2
2
2
3
2
2
provided good to excellent yields in the allylation of aliphatic and aromatic aldehydes under very mild and convenient conditions. SnCl
itself, was also found to be effective for the allylation reaction when allyl bromide was employed. However, the SnCl -only reaction could
only tolerate very small amount of water as the solvent. The SnCl /CuCl , SnCl /TiCl , and SnCl /PdCl -mediated reactions exhibited good
2
, by
2
2
2
2
3
2
2
regioselectivity favoring the g-adduct when cinnamyl halides were employed as the allylation reagent. The same reactions with cinnamyl
halides also showed good diastereoselectivity favoring the anti-product. Mechanistic studies using proton NMR techniques suggested that
the additive (i.e., CuCl , TiCl , PdCl ) could accelerate the formation of allyltin intermediate, but this step was shown not to be the most
2 3 2
important for the allylation. Thus we proposed that the Lewis acid catalysis effect exerted by the additive was the main reason for the
observed reactivity enhancement.
q 2004 Elsevier Ltd. All rights reserved.
1
. Introduction
So far, numerous metals have been reported to be effective
in mediating the aqueous Barbier reaction. Examples
5
6
7
8
Carbonyl allylation is a highly important synthetic trans-
formation in organic and pharmaceutical chemistry, because
the homoallylic alcohol product is a versatile subunit in
synthesis that can easily be converted to a number of other
include aluminum, magnesium, manganese, indium,
9
antimony, bismuth, lead, gallium, zinc, and tin.
10
11
12
13
14
Although good yields can often be obtained in these
reactions, the use of zero-valent metals unavoidably causes
some operational problems. For instance, it is often difficult
to stir the reaction mixture when a large amount of metal is
used. Metal oxide or hydroxide precipitation on the surface
of metal may slow or stop the reaction, so that organic
co-solvent or ultrasonic radiation has to be utilized in some
of the above reactions. Furthermore, some zero-valent
metals are too reactive and significant byproducts (e.g.,
pinacol reaction product, carbonyl reduction product, and
Wurtz reaction product) can be produced in the reaction.
1
useful functions. Until now, many methods have been
developed to accomplish carbonyl allylation, which usually
involve the synthesis of a certain allylic organometallic
reagent and the addition of the organometallic reagent to the
carbonyl compound in anhydrous organic solvents. How-
ever, because of the high reactivity and moisture-sensitivity
of most allylic organometallic reagents, this synthetic
approach is neither operationally simple nor safe to scale up.
An alternative and attractive approach to achieve carbonyl
2
allylation is to use the Barbier reaction, which refers to the
We thought that water-soluble reductive metal salts such as
SnCl , if applicable to the aqueous Barbier reaction, might
metal-mediated coupling between a carbonyl compound and
an organic halide in a one-pot fashion. Operational
simplicity is an obvious advantage of the Barbier reaction,
because no reaction intermediate needs to be isolated.
Moreover, it was found recently that the Barbier reaction
could be conducted in partially or even fully aqueous
2
solve some of the above problems associated with the zero-
valent metals. Thus we recently investigated the possible
use of SnCl in aqueous Barbier reaction. We found that a
2
combination of SnCl and Cu could efficiently mediate the
2
1
5
aqueous carbonyl allylation reactions. We also found that
TiCl3 could dramatically catalyze the SnCl -mediated
3
media. This ‘surprising’ discovery attracts considerable
attention because of the increasing public interest in Green
2
1
6
carbonyl allylation reactions in fully aqueous media. It
is worth mentioning that Roy et al. recently found that SnO/
Cu O and SnCl /CuCl could also mediate the carbonyl
4
Chemistry.
2
2
2
1
7
allylation reactions. However, in their reactions the use of
organic co-solvent such as THF was necessary. Further-
2
Keywords: Allylation; SnCl ; Aqueous medium; Green chemistry; Barbier
1
8
19
more, Samoshin and Yuan reported very recently that
SnCl /KI and SnCl /untrasonication could mediate the
carbonyl allylation reactions in water.
reaction.
*
Corresponding authors. Tel.: þ1-6465157344; fax: þ1-2129321289;
2
2
e-mail addresses: leiliu@chem.columbia.edu; qxguo@ustc.edu.cn
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.05.046

6130
X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
All the above findings suggest that SnCl -mediated allyl-
2
Table 1. Carbonyl allylation between benzaldehyde and allyl bromide
a
2 n
mediated by SnCl /MCl
ation in aqueous media might be developed into a valuable
reaction for organic synthesis. However, a lot of details
about this reaction still remain poorly understood at present.
For instance, the mechanism of the reaction is not fully clear
a
b
Entry
MCl
n
Yield (%)
1
2
3
4
5
6
7
8
9
MgCl
LaCl
2
Trace
6
99
22
27
32
40
27
99
and the role of the additive (e.g., Cu, TiCl ) in the reaction is
3
3
TiCl
CrCl
3
quite ambiguous. Secondly, the scope of the reaction has not
been adequately studied and the reaction condition may
require more optimization. Thirdly, most studies reported so
far only put emphasis on un-substituted allyl halides, while
the regio- and diastereo-chemistry associated with sub-
stituted allyl halides has received little attention.
3
MnCl
FeCl
CoCl2
NiCl
PdCl
CuCl
ZnCl
CdCl
InCl
PbCl
BiCl
2
2
2
2
1
1
1
0
1
2
2
95
2
Trace
Trace
Trace
Trace
Trace
In the present paper we wish to report some new findings
concerning the aqueous SnCl -mediated carbonyl allylation
2
13
3
2
1
1
4
5
2
reactions. We screened many Lewis acid additives for the
reaction and we studied the role of the additive in the
reaction. For the Lewis acid additives that were found to be
effective, we optimized the reaction conditions and
investigated the scope of the reaction. We also studied the
allylation reactions involving cinnamyl halides, where some
interesting regio- and diastereo-selectivities were observed.
Furthermore, we performed some NMR studies on the
reaction intermediates, from which we obtained some novel
insights into the reaction mechanism.
3
a
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of allyl bromide,
10 mmol of SnCl , 5 mmol of MCl (for PdCl this amount is reduced to
.5 mmol), 25 mL of water.
Yields were calculated using H NMR (300 MHz) after the reaction was
stirred for 24 h.
2
n
2
0
b
1
interesting water-compatible Lewis acid. This application of
Ti(III) has received very little attention before.
Using the SnCl /CuCl , SnCl /TiCl , and SnCl /PdCl
2
2
2
2
3
2
methods we examined the scope of the aqueous allylation
reactions (Table 2). It is found that both aliphatic and
aromatic aldehydes can be efficiently allylated using any
one of these three methods. Using allyl bromide as the
allylation reagent the reaction yields are mostly over 90%.
The yields are slightly lower (.80%) when allyl chloride is
used in the reaction.
2
. SnCl /MCl -mediated carbonyl allylation
2 n
In the previous studies Cu, CuCl , Cu O, and TiCl
3
2
2
additives were found to be able to catalyze the SnCl -
1
2
5 – 19
mediated carbonyl allylation reactions in water.
Herein, we examined a variety of water-soluble metal
chlorides as the additive for the aqueous SnCl -mediated
2
allylation reaction (Scheme 1). It is worth mentioning that
we did not examine the water-insoluble Lewis acids,
because the water-insoluble Lewis acids (usually solids)
might cause serious operational trouble for large scale
reactions.
It is worth noting that 2-nitrobenzyldehyde (entry 10 in
Table 2) cannot be effectively allylated using SnCl
because the NO group is considerably reduced in the
reaction. Similar NO reduction problem was noted before
in many zero-valent metal mediated carbonyl allylation
/TiCl ,
2
3
2
2
5
–14
reactions.
do not have this problem as their allylation yields for
-nitrobenzaldehyde are about 80–90%.
Nevertheless, SnCl /CuCl and SnCl /PdCl
2 2 2 2
2
In comparison with aldehydes, neither aliphatic nor
aromatic ketones can be efficiently allylated using any one
of the three methods (entries 11 and 12 in Table 2). In
Scheme 1.
Our results (Table 1) show that TiCl , CuCl , and PdCl can
2
particular, when the SnCl /PdCl method is employed and
2 2
3
2
effectively catalyze the SnCl -mediated allylation reaction
2
allyl chloride is used as the allylation reagent, only trace
amount of allylation product can be detected for both the
aliphatic and aromatic ketones. Therefore, one can use the
SnCl /PdCl method to selectively allylate aldehyde groups
2 2
in the presence of ketone groups.
(
yields .95%). LaCl , CrCl , MnCl , FeCl , CoCl , and
3 3 2 2 2
NiCl show some but fairly low catalytic effects (yields¼
2
6–40%). Other Lewis acids, for example, MgCl , ZnCl ,
CdCl , InCl , PbCl , and BiCl exhibit almost no effects as
2
2
2
3
2
3
very little product can be detected.
Interestingly, the trend of the catalytic activities of different
Lewis acids in our allylation reaction is not fully consistent
with that observed by Kobayashi et al. in their aqueous aldol
reactions. In their cases, Ln(III), Fe(II), Cu(II), Zn(II),
Cd(II), and Pb(II) were found to afford the desired aldol
adduct in high yields in aqueous solvents, whereas Pd(II)
was found to be much less active. In our aqueous allylation
3. SnCl -mediated carbonyl allylation
2
Yuan et al. reported very recently that SnCl can mediate
2
2
0
aqueous carbonyl allylation reactions under ultrasonic
1
9
condition without using any other additive. This report
immediately called upon our attention as we were interested
in the role of the additive in the allylation reaction. Thus we
carefully studied the aqueous allylation reactions using
reaction, however, PdCl is found to be the most effective.
2
Moreover, we found that Ti(III) could be used as an
SnCl alone.
2

X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
6131
Table 2. Carbonyl allylation mediated by SnCl
2
/CuCl
2
, SnCl
2
/TiCl , SnCl
3
2
/PdCl
2
, and SnCl
2
a
b
Entry
Substrates
X
Yield (%)
c
c
c
d
SnCl /CuCl
2
2
SnCl /TiCl
2
3
SnCl /PdCl
2
2
SnCl
2
only
b
Br
Cl
92
85
95
88
92
90
85
21
1
2
Br
Cl
95
90
98
94
99
99
95
25
Br
Cl
96
98
99
99
99
99
99
51
3
4
Br
Cl
93
90
99
91
99
99
99
36
Br
91
99
99
75
5
Cl
Br
87
96
99
99
99
99
25
87
6
7
Cl
Br
88
94
93
90
99
99
22
92
Cl
Br
Cl
Br
Cl
Br
83
92
79
90
84
88
85
93
90
96
96
51
90
94
92
80
95
88
41
87
30
65
15
49
8
9
1
1
0
1
Cl
Br
Cl
Br
79
43
41
23
Trace
25
80
Trace
Trace
Trace
Trace
Trace
Trace
17
17
9
12
Cl
11
Trace
Trace
Trace
a
b
c
X indicates whether the allylation reagent is allyl bromide (X¼Br) or allyl chloride (X¼Cl).
All the reactions were run for 24 h at room temperature.
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of allyl bromide, 10 mmol of SnCl
5 mL of water.
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of allyl bromide, 10 mmol of SnCl
2
, 5 mmol of MCl
n 2
(for PdCl this amount is reduced to 0.5 mmol),
2
d
2
, 2–5 mL of water.
We found that SnCl indeed could mediate the carbonyl
2
1 g of SnCl one can only add less than 3 g of water as
2
allylation reactions in water without using any additive
(
solvent. Indeed, in Yuan’s work only 10 mL of water was
used as solvent for a reaction involving 3.8 g of SnCl
(20 mmol).
Table 2). However, only allyl bromide can be used in these
2
1
9
reactions as allyl chloride provides very poor yields.
Moreover, we observed a very interesting water effect on
SnCl -mediated allylation reactions (Fig. 1).
Interestingly, the dramatic water effect on the SnCl2-
mediated allylation reaction is not exhibited by the SnCl2/
CuCl , SnCl /TiCl , or SnCl /PdCl -mediated reactions
2
As shown in Figure 1, the yield of the aqueous SnCl2-
mediated carbonyl allylation reaction is highly sensitive to
the amount of water used in the reaction. When less than
2
2
3
2
2
(Fig. 1). For these three reactions, increasing the amount
of water from 30 equiv. to over 200 equiv. only reduces the
allylation yield from 99% to about 94%. Therefore, CuCl2,
TiCl , and PdCl truly have some positive effects on the
3
0 equiv. of water (compared to SnCl ) is used in the
2
reaction, the allylation yield is over 98%. When 55 equiv. of
water is used, the yield drops to 75%. When over 100 equiv.
of water is used, the yield is lower than 10%. Therefore,
3
2
SnCl -mediated allylation reactions. The positive effects of
2
CuCl , TiCl , and PdCl are also revealed by the fact that
2
3
2
the key for the SnCl -mediated allylation reaction without
2
any additive is to use very little amount of water. That is, for
allyl chloride can be successfully used in these allylation
reactions.

6132
X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
CuCl , SnCl /TiCl , or SnCl /PdCl -mediated allylation
2
2
3
2
2
reactions, in which the g-adduct is obtained as the favored
isomer. For most cases, the ratio between the g- and
a-adduct is over 95:5. This type of regioselectivity has been
well known in literature.
On the other hand, when SnCl is used for the allylation
2
reaction without any additive, we observed a significant
amount of a-adduct so that the ratio between the g- and
a-adduct decreases to about 60:40. Similar observations
about the a-adducts have been reported recently by Loh
2
1
et al. They proposed that the metal salts formed from the
metal-mediated allylation can catalyze the g-adduct to
undergo a bond cleavage to generate the parent aldehyde in
situ followed by a concerted rearrangement, perhaps a
retroene reaction followed by a 2-oxonia[3,3]-sigmatropic
rearrangement to furnish the a-adduct.
Figure 1. Yields of benzaldehyde allylation mediated by SnCl
2
and
MCl /SnCl in different amount of water.
x
2
It is possible that SnCl serves as the Lewis acid catalyst in
2
the allylation reaction mediated by SnCl alone. However,
2
the fact that only a strictly limited amount of water can be
used indicates that SnCl is not as effective as CuCl , TiCl ,
or PdCl . Furthermore, there is one serious operational
Loh et al. found that the a-adduct became important only
2
2
2
3
1
when very little amount of water was used as solvent. This
indicates that the proposed rearrangement may only occur in
2
problem associated with the SnCl -mediated allylation
2
reaction without using any additive. That is, so little amount
of water can be used in the process that the reaction mixture
may become too dense to deal with. In comparison, the
SnCl /CuCl , SnCl /TiCl , and SnCl /PdCl reactions are
a highly concentrated solution of metal salts. In our SnCl -
2
mediated allylation without any additive, we had to use very
little water as solvent. Thus the a-adduct observed in the
present work may also be explained by Loh’s mechanism
(Table 3).
2
2
2
3
2
2
much easier to stir.
The diastereoselectivity of the allylation reaction (i.e., anti-
vs. syn-product) can be studied using both proton NMR and
4
. Regio- and diastereo-selectivity
22
GC–MS techniques. It is found in the present study that
the SnCl -mediated allylation reaction always exhibits the
2
In order to investigate the regio- and diastereo-selectivities
of the allylation reactions, we used cinnamyl and crotyl
halides as the allylation reagent. The allylation reactions
involving cinnamyl halides can provide both a-adduct and
g-adduct. The g-adduct can be either anti- or syn-product
anti-selectivity with or without CuCl , TiCl , or PdCl
2
2
3
catalyst for both cinnamyl halides and crotyl bromide
Table 4). The same anti-selectivity was also observed
(
before by Masuyama et al. and Roy et al.
22
17
(
Scheme 2).
5
. Mechanism
A possible (and crude) mechanism for the SnCl -mediated
2
allylation reaction is shown in Scheme 3. What remains
unclear is the role of CuCl , TiCl , or PdCl in the reaction.
2
3
2
Do they catalyze the first step (umpolung of the allyl
species) or the second step (allylation)?
A number of proton NMR experiments were conducted to
investigate the first step of the allylation reaction. In the first
series of experiments, we mixed SnCl (2 mmol) and allyl
Scheme 2.
2
It is found that SnCl , by itself, cannot effectively mediate
2
bromide (1.5 mmol) in 3 mL of D O. After the reaction was
2
run for 6 h, we observed a new peak in H NMR
1
the allylation reaction in water when cinnamyl chloride is
used as reactant. Thus CuCl , TiCl , or PdCl catalyst has to
2
3
2
corresponding to an allyltin species (d¼2.5 ppm, doublet,
1
19
be added to induce the reaction. The yields under these
conditions are usually 80–90%. On the other hand, when
cinnamyl bromide is used as reactant, SnCl is capable of
J ( Sn–H)¼155 Hz.) (Fig. 2A). Comparing this new peak
with the peak for allyl bromide (d¼4.1 ppm), we deter-
mined that the yield of the allyltin species was 57%.
2
mediating the aqueous allylation reaction by itself, although
in this case a strictly limited amount of water can be used in
the reaction. The yields under this condition are around 70–
Then we added PdCl (0.1 mmol) into the above mixture,
2
and continued the reaction for 18 h. The NMR spectrum of
the new mixture (Fig. 2B) is almost identical to that shown
in A. This experiment demonstrates that the allyltin species
can form without the help of PdCl . Addition of PdCl does
8
0%. Adding CuCl , TiCl , or PdCl catalyst increases the
2 3 2
yields to 80–95%. It is worth noting that by utilizing CuCl2,
TiCl , or PdCl as catalyst, we do not need to restrict the
amount of water to be used.
3
2
2
2
not change the nature and yield of the allyltin species, either.
Fairly good regioselectivities are observed for the SnCl2/
Since the allylation reaction mediated by SnCl without any
2

X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
6133
Table 3. The reactions of cinnamyl halides and aldehydes mediated by SnCl
2
, CuCl
2
2 3 2 2 2
/SnCl , TiCl /SnCl and PdCl /SnCl
a
b
b
Entry
Condition
Substrates
X
Yield (%)
g:a
anti:syn
c
2
c
1
2
3
SnCl
SnCl
SnCl
2
2
2
/CuCl
88
89
85
99:1
98:2
98:2
90:10
83:17
85:15
/TiCl
/PdCl
3
Cl
Cl
Cl
c
2
4
5
6
SnCl
SnCl
SnCl
2
2
2
/CuCl
2
80
83
91
94:6
95:5
89:11
67:33
90:10
75:25
/TiCl
/PdCl
3
2
7
8
SnCl
SnCl
SnCl
2
2
2
/CuCl
2
75
—
93
99:1
—
93:7
83:17
—
89:11
d
/TiCl
/PdCl
3
9
2
1
1
1
0
1
2
SnCl
SnCl
SnCl
2
2
2
/CuCl
2
90
94
94
99:1
88:12
97:3
88:12
75:25
90:10
/TiCl
/PdCl
3
Cl
Cl
2
13
14
15
SnCl
SnCl
SnCl
2
2
2
/CuCl
2
81
89
88
93:7
95:5
98:2
75:25
80:20
62:38
/TiCl
/PdCl
3
2
d
1
1
1
1
6
7
8
9
SnCl
SnCl
SnCl
SnCl
2
2
2
2
only
/CuCl
87
86
93
80
85:15
99:1
90:10
99:1
91:9
2
89:11
87:13
83:17
Br
Br
Br
/TiCl
/PdCl
3
2
20
21
22
23
SnCl
SnCl
SnCl
SnCl
2
only
71
84
87
92
58:42
98:2
96:4
97:3
62:38
84:16
83:17
75:25
2
2
2
/CuCl
2
/TiCl
/PdCl
3
2
24
25
26
27
SnCl
SnCl
SnCl
SnCl
2
2
2
2
only
/CuCl
70
77
—
85
55:45
90:10
—
91:9
96:4
—
2
/TiCl
/PdCl
3
2
94:6
85:15
2
2
3
3
8
9
0
1
SnCl
SnCl
SnCl
SnCl
2
only
/CuCl
81
92
95
92
56:44
87:13
96:4
85:15
77:23
82:18
86:14
2
2
/TiCl
/PdCl
3
Br
Br
2
2
2
99:1
32
33
34
35
SnCl
SnCl
SnCl
SnCl
2
only
79
87
84
82
75:25
95:5
96:4
94:6
68:32
78:22
77:23
81:19
2
2
2
/CuCl
2
/TiCl
/PdCl
3
2
a
X indicates whether the allylation reagent is cinnamyl bromide (X¼Br) or cinnamyl chloride (X¼Cl). All the reactions were run for 24 h at room
temperature.
Determined using GC–MS and 300 MHz H NMR.
b
c
1
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of cinnamyl bromide, 10 mmol of SnCl
.5 mmol), 25 mL of water.
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of cinnamyl chloride or bromide, 10 mmol of SnCl
2
, 10 mmol of MCl
n
(for PdCl
2
this amount is reduced to
0
d
2
, 2 mL of water.
additive has a strong restriction on the amount of water that
can be used, we also studied the water effect of the allyltin
intermediate formation. We added 10 mmol of SnCl and
Table 4. The reactions of crotyl bromide and aldehydes mediated by SnCl
2
,
a
CuCl
2
/SnCl , TiCl
2
3
/SnCl
2
2 2
and PdCl /SnCl
b
Yield (%) g:a anti:syn
b
2
Entry
Condition
Substrates
X
7.5 mmol of allyl bromide in 3 mL of D O. The proton
2
c
NMR of this reaction mixture (Fig. 2C) is very similar to
that in Figure 2A or B. Therefore, the yield of the allyltin
species is not dependent on the amount of water as the
solvent.
1
2
3
4
SnCl
SnCl
SnCl
SnCl
2
2
2
2
only
97
98
99
99
95:5 91:9
99:1 94:6
99:1 93:7
99:1 96:4
d
2
/CuCl
d
Br
/TiCl
/PdCl
3
d
2
5
6
7
8
SnCl
SnCl
SnCl
SnCl
2
2
2
2
only
/CuCl
/TiCl
/PdCl
68
81
82
88
78:22 62:38
82:18 66:34
80:20 71:29
84:16 69:31
The next question is whether or not the additive increases
the rate of allyltin formation. Therefore, we monitored the
appearance of the NMR peak at 2.5 ppm in the first 6 h of
2
Br
3
2
a
All the reactions were run for 24 h at room temperature.
1
Determined using GC–MS and 300 MHz H NMR.
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of crotyl
bromide, 10 mmol of SnCl , 10 mmol of MCl (for PdCl this amount
is reduced to 0.5 mmol), 25 mL of water.
b
c
2
n
2
d
Reaction condition: 5 mmol of benzaldehyde, 10 mmol of crotyl
bromide, 10 mmol of SnCl , 2 mL of water.
2
Scheme 3.

6134
X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
Figure 3. Yields of the allyltin species with or without additive in the first
6
2
h of the reaction. Conditions: SnCl (2 mmol), allyl bromide (1.5 mmol),
D
2
O (3 mL), CuCl
2
2
(when applicable, 1 mmol), PdCl (when applicable,
0
.1 mmol).
1
Figure 2. H NMR spectra in D
2 2
O (3 mL) for: (A) SnCl (2 mmol)/allyl
bromide (1.5 mmol), standing for 6 h; (B) the same mixture as in A with
0.1 mmol of PdCl
2
added after 6 h, standing for extra 18 h; (C) SnCl
2
(10 mmol)/allyl bromide (7.5 mmol), standing for 24 h.
Scheme 4.
the reaction (Fig. 3). It was found that PdCl and CuCl
2
2
indeed could accelerate the formation of the allytin species.
Nevertheless, the yield of the allyltin species is always ca.
6. Conclusion
6
0% with or without the additive. The maximum yield of the
allyltin species can also be achieved in 6 h with or without
the additive.
In the present paper we reported our systematic studies on
SnCl -mediated carbonyl allylation reaction between alde-
hydes and allyl halides in fully aqueous media. Totally three
valuable reaction systems were found, which were SnCl
CuCl , SnCl /TiCl , and SnCl /PdCl . They all provided
2
The above experiments show some effects of the additive on
the formation of the allyltin intermediate. However, none of
the above experimental findings can be used to explain the
following observation. That is, without appropriate additive
carbonyl allylation reaction does not occur even after 24 h
/
2
2
2
3
2
2
good to excellent yields in the allylation of aliphatic and
aromatic aldehydes under very mild and convenient
conditions. SnCl
for the allylation reaction when allyl bromide was
employed. However, the SnCl -only reaction could only
, by itself, was also found to be effective
2
(Table 1), although it is expected that the allyltin
intermediate is adequately produced in 6 h.
2
tolerate very small amount of water as the solvent.
Thus we have to propose that more important catalytic
effects (presumably Lewis acid catalysis) should be exerted
by the additives in the allylation step (Scheme 4). There are
two possible structures for the transition state. The first is an
The reactions exhibited good regioselectivity favoring the
g-adduct when cinnamyl halides were employed as the
allylation reagent. The same reactions with cinnamyl
halides also showed good diastereoselectivity favoring the
anti-product. Mechanistic studies using NMR techniques
2
3
acyclic one as proposed by Koreeda and Tanaka. They
reported that carbonyl allylation by (E)-cinnamyltributyltin
or (E)-cinnamyltriphenyltin catalyzed by BF ·Et O in
suggested that the additive (i.e., CuCl , TiCl , PdCl ) could
2 3 2
3
2
3
2
CH Cl also exhibited anti-diastereoselectivity. The
2
accelerate the formation of allyltin intermediate, but this
step was shown not to be the most important. Thus we
proposed that the Lewis acid catalysis effect exerted by the
additive was the main reason for the observed reactivity
enhancement.
2
second one, which is more popular, involves a cyclic six-
1
membered ring structure.
Currently we are not certain about which transition state
structure is correct. More detailed experiments need to
designed and conducted. Meanwhile theoretical studies are
required to compare the energies of all the possible
transition states. These studies are undergoing at the
moment in our lab and we will report the results in due
course.
7. Experimental
1
All the reactions were carried out in air. H NMR spectra
were recorded on a Bruker DPX-300 (300 or 400 MHz)

X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
6135
instrument using TMS as internal standard and CDCl or
3
D O as solvent. IR spectra were recorded on an FT/IR/410
2
116.2, 115.2, 113.7, 113.5, 113.4, 75.5, 74.5, 59.4, 58.6,
55.4, 55.3. HRMS (EI) m/z calcd for C H O 254.3280.
17 18 2
JASCO instrument. GC–MS was recorded on a TRACE
GC–MS instrument.
Found 254.3252. GC (column temp.¼250 8C) t ¼10.86
R
(anti), 11.06 (syn).
The yields reported in Tables 1 and 3 were determined using
7.1.3. 1-(4-Nitro-phenyl)-2-phenyl-but-3-en-1-ol. IR
21
1
1
the H NMR method. We obtained the H NMR spectrum of
the crude product from extraction. We integrated the peaks
for the aldehyde C–H proton (d<10) of the starting material
and for the allylic C–H proton (d<4.8) of the homoallylic
alcohol product. Since the aldehyde and homoallylic
alcohol in these cases are not volatile and they can be
fully extracted into the organic layer, we consider this
method to be accurate and convenient for yield
determination.
(neat, cm ) 3422, 3080, 3028, 1637, 1601, 1518, 1493,
1
1452, 1346, 759, 700. H NMR (400 MHz, CDCl , ppm) d
3
8.05 (d, J¼7.90 Hz, 2H), 7.30–7.18 (m, 5H), 7.04 (d,
J¼7.40 Hz, 2H), 6.25 (m, 1H), 5.34 (m, 2H), 4.94 (d,
1
3
J¼7.76 Hz, 1H), 3.50 (t, J¼8.54 Hz, 1H), 1.60 (br, 1H).
C
NMR (300 MHz, CDCl , ppm) (anti) 149.4, 139.7, 137.0,
3
128.8, 128.3, 127.6, 127.3, 123.2, 119.5, 76.5, 59.5. HRMS
(EI) m/z calcd for C H NO 269.2994. Found 269.2911.
1
6
15
3
GC (column temp.¼250 8C) t ¼13.12 (anti), 13.32 (syn).
R
1
The yields reported in Table 2 were determined as the
isolation yields. The a:g ratio in Table 3 were determined
using H NMR peak areas for the allylic C–H protons of the
homoallylic alcohol products (d<2.5 for a and d<3.4 for
g). The anti:syn ratio in Table 3 were determined using the
GC–MS method.
7.1.4. 3-Phenyl-undec-1-en-4-ol: (anti). IR (neat, cm2
)
1
3380, 3060, 1600. H NMR (300 MHz, CDCl , ppm) d
3
1
7.41–7.19 (m, 5H), 6.12 (m, 1H), 5.22 (q, J¼7.0 Hz, 2H),
3.80 (m, 1H), 3.25 (t, J¼8.13 Hz, 1H), 1.78(br, 1H), 1.48–
1
3
1.21 (m, 12H), 0.88 (t, J¼6.6 Hz 3H). C NMR (300 MHz,
CDCl , ppm) (anti) d 142.0, 138.5, 128.5, 128.4, 126.6,
3
126.1, 117.2, 74.0, 57.1, 34.6, 31.7, 29.6, 29.5, 25.7, 22.6,
14.0. HRMS (EI) m/z calcd for C H O 246.3918. Found
7
mediated carbonyl allylation
.1. Typical procedure of SnCl and MCl /SnCl -
2
x
2
17 26
246.3930. GC (column temp.¼250 8C) t ¼8.99 (anti), 9.09
R
(
syn).
To a mixture of carbonyl compound (10 mmol) and allyl
halide (15 mmol) in water, SnCl (20 mmol) and certain
7.1.5. 6-Methyl-3-phenyl-hept-1-en-4-ol: (anti). IR (neat,
1
2
2
catalyst were added (for the detailed amount of water and
catalyst please see Tables 1–3). The mixture was vigorously
stirred at room temperature for approximately 3 h. The
mixture was extracted with ether (3£30 mL). The combined
organic layers were washed by water (2£20 mL). Then the
cm ) 3421, 3078, 3025, 2985, 1640, 1600, 1495, 1450,
1
1000, 744, 699. H NMR (300 MHz, CDCl , ppm) d 7.41–
3
7.19 (m, 5H), 6.05 (m, 1H), 5.18 (q, J¼7.0 Hz, 2H), 3.87 (m,
1H), 3.20 (t, J¼8.03 Hz, 1H), 1.80 (br, 1H), 1.32 (m, 2H),
1
3
1.11 (m, 1H), 0.87 (d, J¼5.23 Hz, 6H).
C NMR
(300 MHz, CDCl , ppm) (anti) 141.9, 138.4, 128.8, 128.3,
organic layer was dried over anhydrous MgSO and was
4
3
filtered and evaporated. The residue, for most cases,
afforded the corresponding homoallylic alcohols of suffi-
cient purity as judged by TLC and 300 MHz H NMR
without the need for further purification. If necessary,
purification was performed by flash column chromato-
graphy (silica gel: 60–120 mesh; eluent: ethyl acetate–
petroleum, 1/5 v/v). The products in Table 2 are known
compounds. Some products in Table 3 are new compounds
and listed below.
126.7, 117.7, 72.1, 57.9, 43.9, 24.5, 23.8, 21.6. HRMS (EI)
m/z calcd for C H O 204.3114. Found 204.3158. GC
1
4 20
1
(column temp.¼250 8C) t ¼5.39 (anti), 5.54 (syn).
R
Acknowledgements
This work is supported by Chinese Academy of Science
(KJCXZ-SW-04) and National Natural Science Foundation
of China (No. 20332020).
7
.1.1. 1,2-Diphenyl-but-3-en-1-ol. IR (neat, cm21) 3422,
1
3078, 3029, 2930, 1643, 1600, 1496, 1450, 744, 699. H
NMR (300 MHz, CDCl , ppm) d 7.35–7.05 (m, 10H), 6.25
3
(
m, 1H), 5.30 (q, J¼8.73 Hz, 2H), 4.85 (m, 1H), 3.55 (t,
References and notes
1
3
J¼8.21 Hz, 1H), 2.30 (br, 1H). C NMR (300 MHz,
CDCl , ppm) d 141.8, 140.6, 140.3, 137.9, 137.7, 128.8,
3
1. (a) Yamamoto, Y.; Asao, N. Chem. Rev. 1993, 93, 2207.
(b) Hoppe, D. Houben-Weyl; Methods of Organic Chemistry;
Helmchen, G., Hoffmann, R. W., Mulzer, J., Schaumann, E.,
Eds.; Thieme: New York, 1996; Vol. E21, pp 1357–1602.
(c) Rottlander, M.; Boymond, L.; Berillon, L.; Lepretre, A.;
Varchi, G.; Avolio, S.; Laaziri, H.; Queguiner, G.; Ricci, A.;
Cahiez, G.; Knochel, P. Chem. Eur. J. 2000, 6, 76.
(d) Denmark, S. E.; Fu, J. Chem. Commun. 2003, 167.
(e) Zhang, Z.-J.; Wan, B.-S.; Chen, H.-L. Chin. J. Org. Chem.
2003, 23, 636. (f) Kennedy, J. W. J.; Hall, D. G. Angew.
Chem., Int. Ed. 2003, 42, 4732.
1
1
28.7, 128.4, 127.9, 127.7, 127.4, 127.1, 126.7, 126.6,
18.5, 117.2, 77.5, 77.2, 59.2, 58.5. HRMS (EI) m/z calcd
for C H O 224.1201. Found 224.1252. GC (column
1
temp.¼250 8C) t ¼8.51 (anti), 10.60 (syn).
6 16
R
7
(
.1.2. 1-(4-Methoxy-phenyl)-2-phenyl-but-3-en-1-ol. IR
1
neat, cm2 ) 3441, 3029, 2906, 1635, 1611, 1585, 1513,
1
1
7
5
453, 1248, 701, 676. H NMR (400 MHz, CDCl , ppm) d
3
.26–7.00 (m, 7H), 6.73 (d, J¼11.49 Hz, 2H), 6.25 (m, 1H),
.26 (m, 2H), 4.80 (d, J¼12.0 Hz, 1H), 3.74 (s, 3H), 3.52
1
3
(
m, 1H), 2.25 (br, 1H). C NMR (300 MHz, CDCl , ppm)
3
2. Recent examples: (a) Lee, A. S.-Y.; Chu, S.-F.; Chang, Y.-T.;
Wang, S.-H. Tetrahedron Lett. 2004, 45, 1551. (b) Cleghorn,
L. A. T.; Cooper, I. R.; Grigg, R.; MacLachlan, W. S.;
1
1
58.8, 140.4, 138.3, 137.9, 134.2, 134.1, 132.1, 128.9,
28.7, 128.0, 127.9, 127.2, 126.6, 126.5, 118.3, 117.3,

6136
X.-H. Tan et al. / Tetrahedron 60 (2004) 6129–6136
Sridharan, V. Tetrahedron Lett. 2003, 44, 7969. (c) Legros, J.;
Chem. 2000, 20, 269. (g) Hilt, G.; Smolko, K. I.; Waloch, C.
Tetrahedron Lett. 2002, 43, 1437. (h) Tan, K.-T.; Chng, S.-S.;
Cheng, H.-S.; Loh, T.-P. J. Am. Chem. Soc. 2003, 125, 2958.
(i) Loh, T.-P.; Yin, Z.; Song, H.-Y.; Tan, K.-L. Tetrahedron
Lett. 2003, 44, 911. (j) Paquette, L. A. Synthesis 2003, 765.
(k) Jang, T.-S.; Keum, G.; Kang, S. B.; Chung, B. Y.; Kim, Y.
Synthesis 2003, 775. (l) Kumar, S.; Kaur, P. Tetrahedron Lett.
2004, 45, 3413.
Meyer, F.; Coliboeuf, M.; Crousse, B.; Bonnet-Delpon, D.;
Begue, J.-P. J. Org. Chem. 2003, 68, 6444. (d) Zhang, J.;
Blazecka, P. G.; Berven, H.; Belmont, D. Tetrahedron Lett.
2003, 44, 5579. (e) Wada, S.; Hayashi, N.; Suzuki, H. Org.
Biomol. Chem. 2003, 1, 2160. (f) Williams, D. R.; Kiryanov,
A. A.; Emde, U.; Clark, M. P.; Berliner, M. A.; Reeves, J. T.
Angew. Chem., Int. Ed. 2003, 42, 1258. (g) Keh, C. C. K.; Wei,
C.; Li, C.-J. J. Am. Chem. Soc. 2003, 125, 4062. (h) Lannou,
M.-I.; Helion, F.; Namy, J.-L. Tetrahedron Lett. 2003, 43,
9. (a) Wang, W.; Shi, L.; Huang, Y. Tetrahedron 1990, 46, 3315.
(b) Ren, P.-D.; Jin, Q.-H.; Yao, Z.-P. Synth. Commun. 1997,
27, 2761. (c) Li, L.-H.; Chan, T. H. Tetrahedron Lett. 2000, 41,
5009. (d) Li, L.-H.; Chan, T. H. Can. J. Chem. 2001, 79, 1536.
10. (a) Minato, M.; Tsuji, J. Chem. Lett. 1988, 2049. (b) Wada, M.;
Ohki, H.; Akiba, K. Y. Bull. Chem. Soc. Jpn 1990, 63, 2751.
(c) Ren, P.-D.; Pan, S.-F.; Dong, T.-W.; Wu, S.-H. Chin.
J. Chem. 1996, 14, 462. (d) Laskar, D.; Gohain, M.; Prajapati,
D.; Sandhu, J. S. New J. Chem. 2002, 26, 193.
8
007. (i) Erdik, E.; Daskapan, T. Tetrahedron Lett. 2002, 43,
237. (j) Bian, Y.-J.; Li, J.-T.; Li, T.-S. Chin. J. Org. Chem.
002, 22, 227. (k) Auge, J.; Lubin-Germain, N.; Seghrouchni,
6
2
L. Tetrahedron Lett. 2002, 43, 5255. (l) Yadav, J. S.; Reddy,
B. V. S.; Reddy, P. M.; Srinivas, C. Tetrahedron Lett. 2002,
43, 5185. (m) Sugimoto, O.; Yamada, S.; Tanji, K.
Tetrahedron Lett. 2002, 43, 3355.
3
. (a) Li, C. J.; Chem, . Rev. 1993, 93, 2023. (b) Lubineau, A.;
Auge, J.; Queneau, Y. Synthesis 1994, 741. (c) Li, C. J.
Tetrahedron 1996, 52, 5643. (d) Zhang, Y.; Wang, M.; Wang,
D.; Huang, Z. Prog. Chem. 1999, 11, 394. (e) Liao, L.-A.; Li,
Z.-M. Chin. J. Org. Chem. 2000, 20, 306. (f) Li, C. J. Green
Chem. 2002, 4, 1. (g) Pae, A. Cho. Y.S., Curr. Org. Chem.
11. Zhou, J.-Y.; Sun, G.-F.; Zhang, M.-F.; Jia, Y.; Wu, S.-H. Chin.
J. Chem. 1997, 15, 361.
12. (a) Wang, Z.; Yuan, S.; Li, C.-J. Tetrahedron Lett. 2002, 43,
5097. (b) Wang, Z.-Y.; Yuan, S.-Z.; Zha, Z.-G.; Zhang, Z.-D.
Chin. J. Chem. 2003, 21, 1231.
13. (a) Petrier, C.; Luche, J. L. J. Org. Chem. 1985, 50, 910.
(b) Deng, D.; Lu, Z. Chin. Chem. Lett. 1994, 5, 173. (c) Deng,
D.-L.; Lu, Z.-H.; Wu, K. Chin. J. Org. Chem. 1996, 16, 462.
(d) Marton, D.; Stivanello, D.; Tagliavini, G. J. Org. Chem.
1996, 61, 2731. (e) Yi, X.; Haberman, J. X.; Li, C. J. Synth.
Commun. 1998, 28, 2999. (f) Zha, Z.-G.; Xie, Z.; Zhou, C.-L.;
Wang, Z.-Y.; Wang, Y.-S. Chin. J. Chem. 2002, 20, 1477.
(g) Chung, W.; Higashiya, S.; Oba, Y.; Welch, J. T.
Tetrahedron 2003, 59, 10031. (h) Zha, Z.; Xie, Z.; Zhou, C.;
Chang, M.; Wang, Z. New J. Chem. 2003, 27, 1297.
14. (a) Mukaiyama, T.; Harada, T. Chem. Lett. 1981, 1527. (b) Wu,
S.; Zhu, T. Acta Chim. Sin. 1987, 45, 1135. (c) Wu, S. H.;
Huang, B. Z.; Zhu, T. M.; Yiao, D. Z.; Chu, Y. L. Acta Chim.
Sin. 1990, 48, 372. (d) Zhou, J.-Y.; Chen, Z.-G.; Jia, Y.; Wu,
S.-H. Acta Chim. Sin. 1998, 56, 93. (e) Chan, T. H.; Yang, Y.;
Li, C. J. J. Org. Chem. 1999, 64, 4452. (f) Zhou, C.-L.; Zha,
Z.-G.; Wang, Z.-Y.; Wu, J.-H.; Zhang, J.-H. Chin. J. Chem.
2002, 20, 718.
2002, 6, 715. (h) Tan, X.-H.; Zhao, H.; Hou, Y.-Q.; Liu, L.;
Guo, Q.-X. Chin. J. Org. Chem. 2004, in press.
4
. (a) Anastas, P. T.; Warner, J. C. Green Chemistry: Theory and
Practice; Oxford University Press: New York, 1997. (b) Li,
C.-J.; Chan, T.-H. Organic Reactions in Aqueous Media;
Wiley: New York, 1997. (c) Lu, X. Prog. Chem. 1998, 10, 125.
(
d) Zhu, Z.-Q.; Zeng, J.-Q. Chin. J. Org. Chem. 2001, 21,
095. (e) Lindstroem, U. M. Chem. Rev. 2002, 102, 2751.
f) Zang, H.-J.; Li, Z.-M.; Wang, B.-L. Chin. J. Org. Chem.
1
(
2003, 23, 1058. (g) Hill, C. L. Angew. Chem., Int. Ed. 2004,
43, 402.
5
. (a) Nokami, J.; Otera, J.; Sudo, T.; Okawara, R. Organo-
metallics 1983, 2, 191. (b) Wada, M.; Ohki, H.; Akiba, K.
J. Chem. Soc. Chem. Commun. 1987, 708. (c) Araki, S.; Jin, S.
Idou. Y., Butsugan, Y. Bull. Chem. Soc. Jpn. 1992, 65, 1736.
(
d) Khan, R.; Prasada, R.; Turga, S. J. Chem. Res. (S) 1998,
02. (e) Tanaka, H.; Nakahata, S.; Watanabe, H.; Zhao, J.;
Kuroboshi, M.; Torii, S. Inorg. Chim. Acta 1999, 296, 204.
f) Estevam, I. S., Bieber, L. W. Tetrahedron Lett. 2003, 44,
67.
2
15. Tan, X.-H.; Shen, B.; Liu, L.; Guo, Q.-X. Tetrahedron Lett.
2002, 43, 9373.
(
6
16. Tan, X.-H.; Shen, B.; Deng, W.; Zhao, H.; Liu, L.; Guo, Q.-X.
Org. Lett. 2003, 5, 1833.
6
. (a) Wada, M.; Fukuma, T.; Morioka, M.; Takahashi, T.;
Miyoshi, N. Tetrahedron Lett. 1997, 38, 8045. (b) Li, C.-J.;
Zhang, W.-C. J. Am. Chem. Soc. 1998, 120, 9102. (c) Zhang,
W.-C.; Li, C.-J. J. Org. Chem. 1999, 64, 3230. (d) Lucas, P.;
Gajewski, J.; Chan, T. H. Can. J. Anal. Sci. Spectr. 2003, 48, 1.
17. (a) Sinha, P.; Roy, S. Organometallics 2004, 23, 67. (b) Kundu,
A.; Prabhakar, S.; Vairamani, M.; Roy, S. Organometallics
1997, 16, 4796.
18. Samoshin, V.; Gremyachinskiy, D. E.; Smith, L. L.; Bliznets,
I. V.; Gross, P. H. Tetrahedron Lett. 2002, 43, 6329.
19. Wang, J.; Yuan, G.; Dong, C.-Q. Chem. Lett. 2004, 33, 286.
20. (a) Kobayashi, S. Synlett 1994, 689. (b) Kobayashi, S.;
Manabe, K. Acc. Chem. Res. 2002, 35, 209. (c) Fu, N.-Y.;
Pang, M.-L.; Yuan, Y.-F.; Wang, J.-T. Chin. J. Org. Chem.
2003, 23, 1085.
(
e) Deng, W.; Tan, X.-H.; Liu, L.; Guo, Q.-X. Chin. J. Chem.
004, in press.
2
7
. (a) Li, C.-J.; Meng, Y.; Yi, X.-H.; Ma, J.; Chan, T.-H. J. Org.
Chem. 1998, 63, 7498. (b) Li, C.-J.; Meng, Y.; Yi, X.-H.; Ma,
J.; Chan, T.-H. J. Org. Chem. 1997, 62, 8632.
8
. (a) Chan, T. H.; Li, C. J.; Lee, M. C.; Wei, Z. Y. Can. J. Chem.
1994, 72, 1181. (b) Cintas, P. Synlett 1995, 1087. (c) Paquette,
L. A.; Mitzel, T. M. J. Am. Chem. Soc. 1996, 118, 1931.
21. Tan, K.-T.; Chng, S.-S.; Cheng, H.-S.; Loh, T.-P. J. Am. Chem.
Soc. 2003, 125, 2958.
(d) Chan, Y. H.; Yang, Y. J. Am. Chem. Soc. 1999, 121, 3228.
(e) Loh, T. P.; Zhou, J. R.; Yin, Z. Org. Lett. 1999, 1, 1855.
(f) Yuan, Y.-F.; Cao, Z.; Hu, A.-G.; Wang, J.-T. Chin. J. Org.
22. Takahara, P.; Masuyama, Y.; Kurusu, Y. J. Am. Chem. Soc.
1992, 114, 2577.
23. Koreeda, M.; Tanaka, Y. Chem. Lett. 1982, 1299.