Table 1 Allyltitanation of propanal from isoprenea
Table 2 Catalytic allyltitanation of several aldehydes from various
dienesa
Homoallylic alcohol
Entry
Diene
Aldehyde
Yield (%)
anti : synb
1
2
3
4
5
6
7
8
isoprene
isoprene
isoprene
isoprene
isoprene
isoprene
myrcene
butadiene
propanal
hexanal
citronellal
isobutyraldehyde
pivalaldehyde
benzaldehyde
propanal
90
88
80
48
24
90
86
35
60 : 40
64 : 36
63 : 37
Cp2Ti(g3-1,2-dimethylallyl)
formation conditions
Yield of
1b (%)
Entry
70 : 30
. 95 : 5
70 : 30
60 : 40
1
1. Cp2TiCl2, isoprene;
2. PrMgCl (10%)/0 uC; 3. PMHS
1. Cp2TiCl2, isoprene;
2. EtMgBr (10%)/0 uC; 3. PMHS
1. Cp2TiF2; 2. PMHS/60 uC/2 min;
3. isoprene
1. Cp2TiF2, isoprene;
2. PhSiH3 (10%); 3. PMHS
1. Cp2TiCl2, isoprene;
42
35
55
65
90
i
2
3
4
5
benzaldehyde
. 95 : 5
a
Conditions: Cp2TiCl2 (0.125 mmol), BuLi (0.25 mmol), isoprene
(10 mmol), PMHS (12 mmol). In all reactions the aldehyde
(2.5 mmol) was added at room temperature over 6 h. The ratio of
b
the two diastereoisomers (anti : syn) was determinated by 1H NMR
analysis.
2. BuLi (10%)/278 uC; 3. PMHS
a
Conditions: catalyst (0.125 mmol), isoprene (10 mmol), PMHS
(12 mmol). In all reactions the propanal (2.5 mmol) was added at
b
room temperature over 6 h. The ratio of the two diastereoisomers
(anti : syn) is equal to 60 : 40 in all the cases.
chain of the aliphatic primary aldehydes (entries 2,3). On the
contrary, increasing the steric hindrance at the a-carbon of the
aliphatic aldehyde led to increased anti diastereoselection but
decreased yields (entries 4,5). Since it has been shown that
the analog stoichiometric reactions gave satisfactory yields, we
interpret these results as indicating that the size of the alkyl group
of the aldehyde affects the key Ti–O cleavage step of the catalytic
process rather than the C–C bond formation itself. By analogy,
Buchwald has previously shown that the rate of hydrosilylation of
bulky ketones is slow.6a This catalytic allyltitanation procedure
tolerates aromatic aldehydes since the reaction of isoprene and
benzaldehyde gave good yield and diastereoselectivity (entry 6).
We also demonstrated the feasibility of the reaction with a more
complicated diene such as myrcene (entry 7). Finally we have
investigated one reaction with butadiene and benzaldehyde (entry
8). In order to use the butadiene as liquid, we conducted the
reaction at 215 uC. At this temperature, the reaction proceeded
but was considerably slowed down. The diastereoselectivity in
favor of the anti product is excellent as reported by Collins in the
stoichiometric case.2a
by Buchwald from Cp2TiF2 and PMHS could be used.6c Thus
titanocene difluoride was briefly heated with PMHS at 60 uC
(2 min) resulting in a colour change from yellow to dark blue. The
flask was then cooled at 0 uC, and isoprene was added dropwise
via syringe. The reaction mixture was stirred at room temperature
until the colour passed from blue to dark purple (15 min). Then
propanal was added to the in situ generated p-allyltitanium
complex over 6 h to afford the homoallylic alcohols with 55% yield
(entry 3). The use of phenylsilane as activating agent and of PMHS
as stoichiometric reductant allowed us to increase the yield to 65%
(entry 4). Although better yields were realised, more than 10% of
the undesirable product (2) was still formed. According to the
studies relating to the reaction of titanocenes with hydrosilanes it
can be assumed that several titanium hydrides and silyl
titanium hydrides are present in the reaction mixture.6c,9 So we
felt that among these species one catalyses the competitive
formation of 2. The fact that 2 is obtained whatever the catalytic
conditions used (Cp2TiCl2–RMgX–PMHS or Cp2TiF2–PMHS)
suggests that both systems have this species in common. Therefore
we continued our efforts towards developing an improved
protocol. Titanium hydride can also be formed by the reaction
of titanocene dichloride with n-BuLi in the presence of PMHS.10
The following reaction protocol employing lithium salt was then
attempted. Two equiv. of n-BuLi were added to a THF solution of
Cp2TiCl2 in the presence of isoprene at 278 uC. After 15 min,
PMHS (5 equiv. per equiv. of aldehyde) was added and the
reaction mixture was warmed to room temperature during which
time the colour of the reaction mixture turned to dark purple. The
propanal was then added at room temperature over 6 h and no
colour change was observed. Usual treatment afforded the
homoallylic alcohol (1) with 90% yield. It is important to point
out that this new protocol eliminates the formation of the side
product (2).
In summary, we have presented the first catalytic allyltitanation
using PMHS as stoichiometric reductant. Further efforts to extend
the scope of this reaction to other substrates such as ketones,
imines or acetals are underway.
Laurianne Bareille, Pierre Le Gendre* and Claude Mo¨ıse*
Laboratoire de Synthe`se et d’Electrosynthe`se Organome´tallique
LSEO-UMR 5188, Faculte´ des Sciences Gabriel, Dijon, France.
E-mail: pierre.le-gendre@u-bourgogne.fr;
claude.moise@u-bourgogne.fr; Fax: 33 3 8039 6098; Tel: 33 3 8039 6082
Notes and references
1 (a) F. Sato, S. Ijima and M. Sato, Tetrahedron Lett., 1981, 22, 243; (b)
B. Klei, J. H. Teuben and H. de Liefde Meijer, J. Chem. Soc., Chem.
Commun., 1981, 342.
2 (a) S. Collins, W. P. Dean and D. G. Ward, Organometallics, 1988, 7,
2289; (b) J. Szymoniak, H. Lefranc, J. Besanc¸on and C. Mo¨ıse,
Synthesis, 1995, 815; (c) H. Urabe, K. Yoshikawa and F. Sato,
Tetrahedron Lett., 1995, 31, 5595; (d) J. Szymoniak, D. Felix,
J. Besanc¸on and C. Mo¨ıse, Tetrahedron, 1996, 52, 4377; (e)
N. Thery, J. Szymoniak and C. Mo¨ıse, Eur. J. Org. Chem., 2000,
1483.
After optimizing our conditions using isoprene and propanal we
explored the scope of this methodology (Table 2). Several trends
are evident. The yield and the anti : syn ratio of the homoallylic
alcohol are almost unaffected by increasing the length of the alkyl
776 | Chem. Commun., 2005, 775–777
This journal is ß The Royal Society of Chemistry 2005