Sc(OTf)3, an efficient catalyst for addition of allyltrimethysilane to aldehydes: Chemoselective addition to aldehydes in presence of ketone and in situ acylation (3-component coupling)

Article abstract of DOI:10.1055/s-1998-2222

Scandium triflate (2-10 mol%) have been found to be a highly efficient catalyst for the addition of allyltrimethylsilane to both activated aromatic and aliphatic aldehydes. However, deactivated (electron rich) aromatic aldehydes gave double allylated products instead. Using a stoichiometric amount of Ac₂O in the allylation reaction led directly to homoallylic acetates [in situ acylation catalysed by Sc(OTf)₃] and for the first time gave high yields of adducts with moderately electron rich aromatic aldehydes.

Full text of DOI:10.1055/s-1998-2222

FEATURE ARTICLE
1
822
Sc(OTf) , an Efficient Catalyst for Addition of Allyltrimethylsilane to Aldehydes;
3
Chemoselective Addition to Aldehydes in Presence of Ketones and in situ Acylation
(3 Component Coupling)
Varinder K. Aggarwal,* Graham P. Vennall
Department of Chemistry, University of Sheffield, Brook Hill, Sheffield S3 7HF, UK
Fax +44(114)2738673; E-mail: V.Aggarwal@sheffield.ac.uk
Received 29 May 1998; revised 9 July 1998
Abstract: Scandium triflate (2–10 mol%) has been found to be a
highly efficient catalyst for the addition of allyltrimethylsilane to
both activated aromatic and aliphatic aldehydes. However, deacti-
vated (electron rich) aromatic aldehydes gave double allylated
efficient catalyst for the desired reaction. The optimised
conditions were then applied to a range of aldehydes (Ta-
ble 1).
products instead. Using a stoichiometric amount of Ac O in the
2
allylation reaction led directly to homoallylic acetates [in situ acy-
Table 1. Scandium Triflate Catalysed Allylation of Aldehydes using
Allyltrimethylsilane in Nitromethane
lation catalysed by Sc(OTf) ] and for the first time gave high
3
yields of adducts with moderately electron rich aromatic alde-
hydes.
Key words: Lewis acid, Sc(OTf) , allylsilane, chemoselective
3
The addition of allylsilanes to aldehydes catalysed by
Lewis acids is an important synthetic transformation.1–5
However, generally, tight binding of the product homoal-
lylic alcohol to the Lewis acid catalyst, together with the
poor efficiency of silicon transfer to oxygen results in
poor turnover, thereby necessitating the use of stoichio-
Entry Aldehyde
Sc(OTf)3 Time Product Isolated
mol%
h
Yield %
1
2
3
4
5
6
7
8
9
0
Benzaldehyde
Benzaldehyde
0
2
2
2
2
48
14
14
14
14
14
14
14
14
18
18
–
0
79
73
98
84
32
43
0
1a
1b
1c
1d
1e
1f
–
1g
1h
1i
2-Naphthaldehyde
4-Chlorobenzaldehyde
4-Nitrobenzaldehyde
4-Methylbenzaldehyde 10
3-Methoxybenzaldehyde 10
4-Methoxybenzaldehyde 10
Phenylacetaldehyde
Valeraldehyde
Cyclohexane-
6
metric (or greater) amounts of Lewis acid. In the arena of
asymmetric addition of allylsilanes to aldehydes chiral
7
8
a
boron-based and titanium-based Lewis acids have been
reported which (unusually) give good yields even when
using only sub-stoichiometric amounts. Other notable cat-
alysts for the allylation reaction using allyltrimethylsilane
a
a
10
10
10
53
56
62
1
that have been used in sub-stoichiometric amounts are 11
+
– 9
4
10
Brønsted acids [TfOH B(OTf) , HN(SO F) ] and su-
carboxaldehyde
Pivalaldehyde
2
2
2
1
1
12
10
18
1j
71
per acids [Me SiB(OTf) ].
3
4
a
Unlike traditional Lewis acids (e.g. TiCl , AlCl ,
A range of other products were also obtained in these cases (see be-
low).
4
3
BF •OEt ), lanthanide and group III metal salts are known
3
2
to form strong but labile bonds with oxygen-donor
1
2
ligands. This feature has often allowed sub-stoichiomet-
ric amounts of the Lewis acid to be used to promote a va- All of the reactions had gone to completion overnight to
riety of reactions. Indeed, such Lewis acids are effective give mixtures of free and silylated alcohols. After treat-
catalysts in many fundamental processes, including ace- ment with tetrabutylammonium fluoride, and chromato-
1
3,14
15
16
talisation,
acylations,
Diels–Alder reactions,
graphic purification, the corresponding alcohols were ob-
Michael additions1 and Friedel–Crafts reactions. They tained in good yields.
7
18
have also been used to catalyse the addition of allylstan-
It was pleasing to find that both unactivated and activated
aromatic aldehydes worked well, giving high yields of the
products 1a–d (entries 2–5). Some of the most difficult
allylation reactions are with aliphatic aldehydes due to
their intrinsic lower reactivity. We were pleased to find
that allyltrimethylsilane coupled efficiently with aliphatic
aldehydes to give the corresponding adducts 1g–j (entries
1
9
nanes (both tetraallylstannane and the less reactive
2
0
21
allyltributylstannane ) and tetraallylgermane to carbo-
nyl compounds. However, no reports on the addition of
the much less reactive allylsilanes to aldehydes had been
reported using catalytic quantities of transition metals. In
this paper we describe our studies in carbonyl allylation.22
The successful application of lanthanide and group III 9–12). However, electron rich aromatic aldehydes were
metal triflates to the reactions described above led us to less successful in the allylation reaction. The weakly de-
screen a range of such catalysts in the reaction between activated aromatic aldehyde, 4-methylbenzaldehyde (en-
allyltrimethylsilane and benzaldehyde in a range of sol- try 6), required a higher catalyst loading (10 mol%) for
vents. After testing many solvent and catalyst combina- reaction to occur and gave a mixture of both the desired
tions, it was found that Sc(OTf) in nitromethane was an product 1e and over-reacted product 2a (Scheme 1).
3

December 1998
SYNTHESIS
1823
Additionally, ketones were shown to be completely unre-
active under these conditions and it was possible to
chemoselectively add allyltrimethylsilane to benzalde-
hyde in the presence of acetophenone (Scheme 4). None
of the allylated ketone was obtained. Careful chromato-
graphy gave recovered ketone and allylated aldehyde 1a
only. The chemoselectivity observed with allylsilane mir-
rors that observed with tetraallylgermanium21 using the
same conditions.
Scheme 1
The highly deactivated aromatic aldehyde, (4-methoxy-
benzaldehyde, entry 8) also required a high (10 mol%)
catalyst loading for reaction to occur but gave none of the
desired monoallylated product. Instead, complete conver-
sion to the double addition product 2b was evident
(Scheme 2).
Scheme 4
There is clearly a limitation in the reaction of allylsilanes
with aldehydes (poor reaction with electron rich aromatic
aldehydes) that is not observed with allylstannanes. This
is a result of the low nucleophilicity of allylsilanes and the
low electrophilicity of electron rich aromatic aldehydes
which therefore requires strong Lewis acids for activation.
Strong Lewis acids promote carbocation formation from
the product benzylic alcohols and carbocation formation
becomes increasingly facile as the aromatic aldehyde in-
creases in electron density resulting in double allylation.
Indeed, none of the catalytic systems reported to date (in-
cluding our own) have been applied to electron rich (de-
activated) aldehydes.
Scheme 2
The reaction of 3-methoxybenzaldehyde (entry 7) suggests
that this is an electronic effect because the desired product
1f was obtained in a moderate yield (43%) together with
another over-addition product 3 (12%) (Scheme 3).
Attempts to limit the extent of over addition by moderat-
Scheme 3
ing the Lewis acidity of Sc(OTf) using additional chelat-
3
Biographical Sketches
Varinder Aggarwal was born in Kalianpur in the northern state of Punjab in India in
961 and in 1963 his family emigrated to England. He received his BA in 1983 and PhD
1
in 1986 from Cambridge University under the guidance of Stuart Warren. He then carried
out post-doctoral research with Professor Gilbert Stork at Columbia University as a
Harkness Fellow and returned to a lectureship at Bath University in 1988. In 1991 he
moved to Sheffield University where, in 1997, he was promoted to Professor in Organic
Chemistry.
Graham Vennall was born in 1973 and graduated from Exeter University in 1994. He
received his PhD degree in 1998, at Sheffield University, under the supervision of Pro-
fessor Varinder Aggarwal, working on metal triflates as Lewis acids in organic synthesis.
Currently, he is back at Exeter University, working as Leverhulme Trust Research Fel-
low with Dr. Iain Coldham on chiral organolithium compounds.

1
824
Feature Article
SYNTHESIS
ing ligands were unsuccessful. We therefore attempted to Table 2. Acetic Anhydride Modified Scandium Triflate Catalysed
Allylation of Aldehydes using Allyltrimethylsilane
protect the alcohol in situ (Scheme 5) (it is presumably a
mixture of silyl ether and alcohol) to inhibit binding to the
Lewis acid and subsequent carbocation formation or for-
mation of acetals/hemiacetals, either of which would re-
sult in double allylation.
Entry Aldehyde
Sc(OTf)3 Product Isolateda
mol%
Yield %
1
2
3
4
5
6
7
Benzaldehyde
4-Nitrobenzaldehyde
4-Chlorobenzaldehyde
4-Methylbenzaldehyde
3-Methoxybenzaldehyde
4-Methoxybenzaldehyde
2
4a
–
74
0
51
68
47
2–20
5
2
2
2
4b
4c
4d
4e
–
Scheme 5
b
Perhaps an unlikely choice of protecting group was ace-
tate since a better leaving group is generated upon protec-
tion. However, if over-reaction was proceeding via acetal
formation
this strategy might work.
14
0
Cyclohexanecarboxaldehyde 2–20
a
Remainder of material was acylal, RCH(OAc)2.
The remainder of material in this case was the double allylated pro-
duct 2b.
1
3,14
and not via benzylic cation formation then
b
Yamamoto had previously shown15 scandium triflate to
be an extremely efficient catalyst for the acylation of a ferent but related roles: activation of the carbonyl group of
range of alcohols in MeCN (optimum solvent) and it was the aldehyde and activation of the anhydride for acylation.
decided to apply this methodology to our system. Treat-
ment of 1-phenylbut-3-en-1-ol 1a with acetic anhydride
In conclusion, scandium triflate in nitromethane has been
shown to be a highly active catalytic system for the addi-
(2.0 equiv.) in nitromethane under our previously deter-
tion of the weakly nucleophilic allyltrimethylsilane to a
range of both aromatic and aliphatic aldehydes. Products
were obtained in high yields for all aliphatic and most ar-
omatic aldehydes. However, electron rich aromatic alde-
hydes gave a range of over-reacted products.
mined conditions with Sc(OTf) (2 mol%) rapidly gave
3
the acylated product 4a in excellent yield (98%) (Scheme
6). This demonstrated that homoallylic alcohol 1a could
be acylated under conditions similar to those employed in
the earlier allylation process.
A complementary, in situ acylation, procedure was devel-
oped which worked more efficiently with electron-rich al-
dehydes. This modified procedure was applied to a range
of aromatic aldehydes and worked well with electron neu-
tral and moderately electron deficient/electron rich alde-
hydes. Strongly activated aldehydes (and aliphatic alde-
hydes) gave acylals and strongly deactivated aldehydes
gave double allylated products after long reaction times.
At short reaction times low yield of the monoallylated
product was obtained together with starting material.
Thus, for the first time, addition of allylsilane to electron
rich aromatic aldehydes has been achieved.
Scheme 6
The reaction of benzaldehyde with allyltrimethylsilane
was repeated in the presence of acetic anhydride and the
acylated product 4a was obtained in good yield (74%),
showing that an in situ acylation was viable. Under these
conditions it was found possible to extend this modified
procedure to a range of aromatic aldehydes (Table 2).
Compounds 1a, 1c–d, 1e, 1h–j, 4a, 4c–e25 had identical data
9
9
23
9
24
Thus, for the first time, a clean and high-yielding reaction
was obtained with 4-methylbenzaldehyde (entry 4) and
even 4-methoxybenzaldehyde (entry 6) gave allylated
product, albeit in low yield. 4-Nitrobenzaldehyde and cy-
clohexanecarboxaldehyde (entries 2 and 7) did not give the
desired products, due to the irreversible formation of acylals,
to that reported in the literature.
Scandium Triflate Catalysed Allylation of Aldehydes using Allyl-
trimethylsilane (Table 1); Typical Procedure:
To a stirred solution of scandium triflate (9.8 mg, 0.02 mmol) in ni-
tromethane (10 mL) at r.t. was added benzaldehyde (102 µL, 1.0
mmol) and allyltrimethylsilane (240 µL, 1.5 mmol). The mixture was
stirred at r.t. until complete consumption of the aldehyde was
achieved (as monitored by TLC). The mixture was then poured into
RCH(OAc) . 4-Methoxybenzaldehyde (entry 6) gave only
2
a low yield of product (14%) with the remainder isolated as
double-addition product 2b. Quenching the reaction after a
short reaction time (1 hour) gave the product 4e in the same
yield without any of the associated over-addition product
brine (10 mL), extracted using CH Cl2 (3 × 10 mL) and dried
2
(
Na SO ). After removal of the solvent under reduced pressure, the
2 4
residue was treated with tetrabutylammonium fluoride (1.0 mL, 1.1
mmol, 1.1 M in THF) and the deprotection was monitored by TLC.
2b, the remainder being unreacted aldehyde.
1-(2'-Napthyl)but-3-en-1-ol (1b)
This reaction is a novel three component coupling reac-
tion and allows for rapid construction of diverse products.
Chromatography (10% EtOAc/Petroleum ether) gave the product 1b
as a colourless oil (144 mg, 0.73 mmol, 73%), R 0.21 (10% EtOAc/
f
In addition, the catalyst Sc(OTf) is performing two dif-
Petroleum ether).
3

December 1998
SYNTHESIS
1825
IR (thin film): νmax 3385 (OH) and 1639 (C=C) cm–1.
MS (EI): m/z = 162 (M , 18%), 121 (57), 103 (52), 92 (100), 91 (58)
+
+
+
1
H NMR (250 MHz; CDCl ): δ = 2.18 (1 H, br s, OH), 2.57–2.65 (2
(Found M , 162.1039. C H O requires M , 162.1045).
3
11 14
H, m, CH ), 4.81 (1H, t, J = 7.2, CHOH), 5.17 (1 H, d, J = 10.6,
2
CH H =), 5.19 (1 H, d, J = 17.5, CH H =), 5.75–5.93 (1 H, m,
4-(4'-Methylphenyl)hepta-1,6-diene (2a):
A
B
A B
CH=) 7.45–7.52 (3 H, m, ArH), 7.79–7.87 (4 H, m, ArH).
Chromatography (10% EtOAc/Petroleum ether) gave 1e and 4-(4'-
methylphenyl)hepta-1,6-diene (2a) as a colourless oil (48 mg, 0.26
13
C NMR (63 MHz; CDCl ): δ = 43.7 (t), 73.4 (d), 118.4 (t), 124.0
3
(
1
d), 124.5 (d), 125.8 (d), 126.1 (d), 127.7 (d), 127.9 (d), 128.2 (d),
mmol, 26%), R 0.77 (2% EtOAc/Petroleum ether).
f
–
1
32.9 (s), 133.3 (s), 134.4 (d), 141.4 (s)
IR (CHCl ): ν
= 1640 (C=C) cm .
3
max
+
+
1
MS: m/z (EI) = 198 (M , 13%), 157 (100) and 129 (97) (Found M ,
1
H NMR (250 MHz; CDCl ): δ = 2.32 (3 H, s, CH ), 2.25–2.44 (4 H,
3 3
+
98.1048. C H O requires M , 198.1045).
m, 2 × CH ), 2.68 (1 H, tt, J = 8.1 and 6.1, ArCH), 4.93 (2 H, d, J =
2
1
4
14
1
1.1, 2 × CH H =), 4.97 (2 H, d, J = 17.0, 2 × CH H =), 5.58–5.75
A B A B
1
-(3'-Methoxyphenyl)but-3-en-1-ol (1f) and Di[1-(3'-methoxyphe-
(2 H, m, 2 × CH=), 7.04 (2 H, d, J = 8.2, ArH) and 7.10 (2 H, d, J =
nyl)but-3-en-1-yl] Ether (3)
Chromatography (10% EtOAc/Petroleum ether) gave the following
compounds:
8.2, ArH).
13
C NMR (63 MHz; CDCl ): δ = 21.1 (q), 40.4 (t), 45.2 (d), 116.0 (t),
3
127.6 (d), 128.9 (d), 135.5 (s), 136.9 (d), 141.5 (s)
MS (CI): m/z = 204 (MNH , 68%), 186 (M , 27), 162 (38), 145
+
4
+
+
+
1
-(3'-Methoxyphenyl)but-3-en-1-ol (1f) as a colourless oil (78 mg,
(100) (Found (MNH ) , 204.1749. C H N requires (MNH ) ,
4 14 22 4
0
.43 mmol, 43%), R 0.16 (10% EtOAc/Petroleum ether).
f
204.1752).
–
1
IR (CHCl ): ν
= 3606 (OH), 1602 (C=C) cm .
3
max
1
H NMR (250 MHz; CDCl ): δ = 2.45–2.54 (2 H, m, CH ), 3.82 (3
3
2
4-(4'-Methoxyphenyl)hepta-1,6-diene (2b):
Chromatography (2% EtOAc/Petroleum ether) gave the product 2b as
a colourless oil (114 mg, 0.56 mmol, 56%), R 0.47 (10% EtOAc/
Petroleum ether)
H, s, CH O), 4.70 (1 H, dd, J = 7.3, 5.5, CHOH), 5.14 (1 H, d, J =
3
1
0.0, CH CH =), 5.16 (1 H, d, J = 19.2, CH H =), 5.72–5.89 (1 H,
A B A B
f
m, CH=), 6.81 (1 H, ddd, J = 7.9, 2.4, 1.0, ArH), 6.90–6.95 (2 H, m,
ArH), 7.25 (1 H, t, J = 7.9, ArH).
–
1
IR (CHCl ): ν
= 1639 (C=C), 1611, 1513 (Ar) cm .
3
max
13
C NMR (63 MHz; CDCl ): δ = 43.8 (t), 55.2 (q), 73.1 (d), 111.2
1
3
H NMR (250 MHz; CDCl ): δ = 2.25–2.47 (4 H, m, 2 × CH ), 2.68
3
2
(
1
d), 112.9 (d), 118.1 (d), 118.4 (t), 129.4 (d), 134.4 (d), 145.6 (s) and
59.6 (s).
MS (EI): m/z = 178 (M , 9%), 137 (75), 109 (100), 94 (33), 77 (36)
(1 H, tt, J = 7.5, 5.9, ArCH), 3.79 (3 H, s, CH ), 4.93 (2 H, d, J = 10.0,
3
2
× CH H =), 4.97 (2 H, d, J = 17.5, 2 × CH H =), 5.58–5.75 (2 H,
A B A B
+
m, 2 × CH=), 6.84 (2 H, d, J = 8.8, ArH), 7.07 (2 H, d, J = 8.8, ArH).
+
+
(Found M , 178.1001. C H O requires M , 178.0994).
13
1
1
14
2
C NMR (63 MHz; CDCl ): δ = 40.5 (t), 44.7 (d), 55.2 (q), 113.6 (d),
3
1
16.0 (t), 128.6 (d), 136.7 (s), 136.9 (d), 157.8 (s).
Di[1-(3'-methoxyphenyl)but-3-en-1-yl] Ether (3), an inseparable
mixture of isomers (3:2) as a colourless oil (20 mg, 0.06 mmol, 12%),
+
+
MS (EI): m/z = 202 (M , 11%) and 161 (100) (Found M , 202.1358.
C H O requires M , 202.1361).
+
1
4 18
R 0.41 (10% EtOAc/Petroleum ether).
f
–
1
IR (CHCl ) ν
= 1600 (C=C) cm .
3
max
Acetic Anhydride Modified Allylation of Aldehydes using Allyl-
trimethylsilane (Table 2); Typical Procedure:
To a stirred solution of scandium triflate (9.8 mg, 0.02 mmol) and
benzaldehyde (102 µL, 1.0 mmol) in nitromethane (1.0 mL) at r.t. was
added Ac O (189 µL, 2.0 mmol) and allyltrimethylsilane (191 µL, 1.2
mmol). After stirring at r.t. overnight, the mixture was poured into
Major diastereomer:
1
H NMR (250 MHz; CDCl ): δ = 2.33 (2 H, dt, J = 13.8, 6.0,
3
2
× CH H C=), 2.54 (2 H, dt, J = 13.8, 7.3, 2 × CH H ), 3.80 (3 H,
A B A B
2
s, 2 × CH O), 4.10 (2 H, dd, J = 7.3, 6.0, 2 × CHO), 4.95 (2 H, d, J =
3
1
1.3, 2 × CH H =), 4.96 (2 H, d, J = 15.1, 2 × CH H ), 5.62–5.79 (2
A B A B
brine (4 mL), extracted with CH Cl (3 × 3 mL) and dried (MgSO ).
2
2
4
H, m, 2 × CH=), 6.74–6.85 (6 H, m, ArH), 7.20–7.27 (2 H, m, ArH).
13
C NMR (63 MHz; CDCl ): δ = 42.9 (t), 55.2 (q), 78.2 (d), 112.2 (d),
3
Acetic Acid 1-(4'-Chlorophenyl)but-3-enyl Ester (4b):
Chromatography (2% EtOAc/Petroleum ether) gave the product 4b as
a colourless oil (115 mg, 0.51 mmol, 51%), R 0.42 (10% EtOAc/
1
1
13.2 (d), 116.7 (t), 119.6 (d), 129.2 (d), 135.0 (d), 143.8 (s) and
59.7 (s).
f
1
Petroleum ether).
Minor diastereomer: H NMR (250 MHz; CDCl ): δ = 2.51 (4 H, dd,
3
= 1735 (C=O), 1635 (C=C) cm–1.
IR (CHCl ): ν
J = 6.2, 3.4, 2 × CH ), 3.65 (3 H, s, 2 × CH O), 4.41 (2 H, t, J = 6.2,
× CHO), 5.04 (2 H, d, J = 18.1, 2 × CH H =), 5.07 (2 H, d, J = 9.4,
× CH H =), 5.65–5.89 (2 H, m, 2 × CH=), 6.70–6.82 (6 H, m,
ArH), 7.17 (2 H, t, J = 6.9, ArH).
C NMR (63 MHz; CDCl ): δ = 41.6 (t), 55.1 (q), 79.5 (d), 112.0 (d),
12.9 (d), 117.0 (t), 119.2 (d), 129.0 (d), 134.7 (d), 144.1 (s) and
3
max
2
3
1
H NMR (250 MHz; CDCl ): δ = 2.05 (3 H, s, CH CO), 2.45–2.69
3
3
2
2
A B
(
2 H, m, CH ), 5.04 (1 H, d, J = 10.6, CH H =), 5.06 (1 H, d, J = 16.9,
2 A B
A
B
CH H =), 5.58–5.72 (1 H, m, CH=), 5.75 (1 H, dd, J = 7.8, 6.9,
A
B
13
CHO), 7.24 (2 H, d, J = 8.9, ArH), 7.34 (2 H, d, J = 8.9, ArH).
3
13
C NMR (63 MHz; CDCl ): δ = 21.0 (q), 40.5 (t), 74.5 (d), 118.3 (t),
3
1
1
127.9 (d), 128.5 (d), 132.8 (d), 133.6 (s), 138.5 (s), 170.0 (s).
59.4 (s).
+
37
+ 35
+
+
MS (EI): m/z = 226 [M ( Cl), 4%], 224 [M ( Cl), 14], 183 (73),
MS (EI): m/z = 338 (M , 24%), 161 (100) (Found M , 338.1892.
+
35
+
141 (100), 129 (19) and 77 (18) (Found M , 224.0602. C H O Cl
C H O requires M , 338.1882).
12 13 2
2
2
26
3
+
requires M , 224.0604).
1
-Phenylpent-4-en-2-ol (1g):
Chromatography (2% EtOAc/Petroleum ether) gave the product 1 g
We wish to thank Quest International and Sheffield University for
support of this work.
as a colourless oil (87 mg, 0.53 mmol, 53%), R 0.19 (2% EtOAc/
f
Petroleum ether).
–
1
IR (CHCl ) ν
= 3585 (OH), 1641 (C=C) cm .
3
max
1
H NMR (250 MHz; CDCl ): δ = 1.68 (1 H, d, J = 3.8, OH), 2.16–
3
2
.41 (2 H, m, CH ), 2.73 (1 H, dd, J = 13.1, 7.5, PhCH H ), 2.83 (1
2 A B
(1) Nishigaichi, Y.; Takuwa, A.; Naruta, Y.; Maruyama, K. Tetra-
hedron 1993, 49, 7395.
(2) Sakurai, H. Pure Appl. Chem. 1982, 54, 1.
H, dd, J = 13.1, 5.6, PhCH H ), 3.84–3.93 (1 H, m, CHOH), 5.15 (1
H, d, J = 11.3, CH H =), 5.17 (1 H, d, J = 15.6, CH H =), 5.82–5.96
A
B
A
B
A
B
(
1 H, m, CH=), 7.20–7.36 (5 H, m, ArH).
(3) Marshall, J. A. Chem. Rev. 1996, 96, 31.
1
3
C NMR (63 MHz; CDCl ): δ = 41.1 (t), 43.2 (t), 71.6 (d), 118.1 (t),
3
(4) Fleming, I. In Comprehensive Organic Synthesis; Vol. 2; Fle-
1
26.4 (d), 128.5 (d), 129.4 (d), 134.7 (d), 138.4 (s)
ming, I. and Trost, B. M., Ed.; Pergamon: Oxford, 1991; pp 563.

1
826
Feature Article
SYNTHESIS
(
5) Fleming, I.; Dunogues, J.; Smithers, R. Org. React. (N.Y.) 1989,
(16) Kobayashi, S.; Hachiya, I.; Ishitani, H.; Mitsuharu, A. Tetrahe-
dron Lett. 1993, 34, 4535.
3
7, 57.
(
(
6) Hollis, T. K.; Bosnich, B. J. Am. Chem. Soc. 1995, 117, 4570.
7) Ishihara, K.; Mouri, M.; Gao, Q. Z.; Maruyama, T.; Furuta, K.;
Yamamoto, H. J. Am. Chem. Soc. 1993, 115, 11490.
(17) Kobayashi, S.; Hachiya, I.; Takahori, T.; Araki, M.; Ishitani, H.
Tetrahedron Lett. 1992, 33, 6815.
(18) Kawada, A.; Mitamura, S.; Kobayashi, S. Synlett 1994, 545.
(19) Hachiya, I.; Kobayashi, S. J. Org. Chem. 1993, 58, 6958.
(20) Aspinall, H. C.; Browning, A. F.; Greeves, N.; Ravenscroft, P.
Tetrahedron Lett. 1994, 35, 4639.
(21) Akiyama, T.; Iwai, J. Tetrahedron Lett. 1997, 38, 853.
(22) For a preliminary account see: Aggarwal, V. K.; Vennall, G. P.
Tetrahedron Lett. 1996, 37, 3745.
(8) Gauthier, D. R.; Carreira, E. M. Angew. Chem., Int. Ed. Engl.
1
996, 35, 2363.
9) Davis, A. P.; Jaspars, M. J. Chem. Soc., Perkin Trans. I. 1992,
111.
(
2
(
(
10) Kaur, G.; Manju, K.; Trehan, S. Chem. Commun. 1996, 581.
11) Davis, A. P.; Jaspars, M. Angew. Chem., Int. Ed. Engl. 1992, 31,
4
70.
(23) Mukaiyama, T.; Minowa, N. Bull. Chem. Soc. Jpn. 1987, 60,
3697.
(
12) Kobayashi, S. Synlett 1994, 689.
(
13) Ishihara, K.; Karumi, Y.; Kubota, M.; Yamamoto, H. Synlett
(24) Falck, J. R.; Bhatt, R. K.; Ye, J. J. Am. Chem. Soc. 1995, 117,
5973.
(25) Hixson, S. S.; Franke, L. A.; Gere, J. A.; Xing, Y. J. Am. Chem.
Soc. 1988, 110, 3601.
1
996, 839.
14) Fukuzawa, S.; Tsuchimoto, T.; Hokata, T.; Hiyama, T. Synlett
995, 1077.
(
(
1
15) Ishihara, K.; Kubota, M.; Kurihara, H.; Yamamoto, H. J. Am.
Chem. Soc. 1995, 117, 4413.