1
522
K. He et al.
LETTER
H
tion. However, it is gratifying that ligand (+)-(S)-2 was
very stable and could be readily recovered and reused
without loss of its catalytic activity and asymmetric induc-
tion ability. It was found that a 4:1 molar ratio of ligand to
Ph
Me
OH
N
O
P
N
Ti(Oi-Pr) resulted in better enantioselectivity, whereas
4
Me
the use of one equivalent of ligand (+)-(S)-2 per Ti(Oi-
Ph
1
H
Pr) led to higher chemical yield (entries 1, 5 and entries
4
6
, 9). Buono reported that the introduction of i-PrOH as an
Figure 1
additive has a dramatic influence on the enantioselectivity
5
d
2
i
in asymmetric silylcyanation. However, only a little in-
crease in selectivity was observed in our research (entries
1
Dibenzylation of (+)-3 derived from D-camphor led to
N,N′-dibenzyl-1,2,2-trimethylcyclopentane-1,3-diamine
, 4 and entries 6, 11). The reaction temperature was also
(
4). The cyclization of the latter with O-1-naphthyl phos-
found to be an essential factor to the reaction. The reaction
at 20 °C generally led to better results than that carried out
at 0 °C. An increase in reaction temperature resulted in a
detrimental effect to the reaction due to the instability of
the adduct silyl ether. The nature of the substrate aromatic
aldehyde has a dramatic influence on the catalytic effect.
Generally, the enantioselectivity of aldehyde substituted
with electron donating group (methyl and methoxy) on the
benzene ring was better than that of electron withdrawing
group (chloro and nitro) substituted one. Moreover, the
enantioselectivity was also affected by the position of the
substituent on the benzene ring. When methoxy substitut-
ed benaldehyde was employed as the substrate, it was
found that the enantioselectivity falls in the order: o>m>p
phorodichloridate afforded O-naphthyl phosphorodiami-
date (5). A pair of diastereomers of 5 was obtained via
column chromatography. A subsequent P-O to P-C rear-
rangement upon treatment of (–)-5 with n-BuLi resulted in
the formation of cyclic (+)-o-hydroxynaphthylphos-
phonodiamide (2). The corresponding rearrangement
product of (+)-5 was not obtained under the same condi-
tions. The absolute configuration of the phosphorus atom
in (+)-5 was determined as S via X-diffraction analysis.7
Thus the absolute configuration of the phosphorus atom in
(
–)-5 should be R. At the same time, crystallographic
study shown that the absolute configuration of the phos-
8
phorus atom in (+)-2 was S. Therefore, the rearrange-
ment from (–)-5 to (+)-2 proceeded with total retention of
configuration at the phosphorus atom.
(
entries 6, 14 and 12). This finding indicates that not only
the electrical effect but also the position of the subtituent
or steric effect) had a decisive role on the enantioselectiv-
The catalytic effect of the titanium complex formed in situ
(
from (+)-(S)-2 and Ti(Oi-Pr) in asymmetric silylcyana-
4
ity of the reaction.
tion of aromatic aldehydes was investigated. The experi-
mental results were listed in Table 1.
In conclusion, a new chiral cyclic o-hydroxynaphthyl-
phosphonodiamide (+)-2 was synthesized and the abso-
lute configuration of phosphorus atom was determined by
the X-ray diffraction analysis. Excellent results (up to
Usually, the silylcyanation reaction was best conducted in
methylene chloride. We firstly examined the influence of
the amount of ligand used on the enantioselectivity of the
reaction. It was found that a variation of decrease in yield
and enantiomeric excesses value was observed depending
on the substrate employed with the reducing of the
amount of (+)-(S)-2 from 40 mol% to 20 mol%. As to the
substrate o-methoxybenzaldehyde whose ee value was
9
8.3% ee) were achieved in the asymmetric silylcyanation
of aromatic aldehydes using (+)-2 as the ligand catalyst in
10
the presence of Ti(Oi-Pr) . Investigations on further ex-
4
tending the range of substrates and application of this
compound for other asymmetric reaction are continuing in
our laboratory.
9
8% and 97%, respectively, only a very slight change in
yield and enantioslectivity was observed (entries 6 and 7).
While for some substrates, such as p-methoxybenzalde-
hyde, m-methoxybenzaldehyde and a-naphthyl aldehyde,
Acknowledgment
We are grateful to the National Natural Science Foundation of
this change led to an obvious decrease in enantioselectiv- China (No. 20272025) and the Ph. D. Programs of Ministry of
ity (entries 12 and 13, 14 and 15, 18 and 19). Further re- Education of China for generous financial support for our programs.
ducing the amount of (+)-(S)-2 to 10 mol% resulted in
remarkable decrease both in yield and enantioselectivity
References
(
entry 8). These results showed that the change from the
(
1) (a) North, M. Tetrahedron: Asymmetry 2003, 14, 147.
b) Schmidt, M.; Herve, S.; Klempier, N.; Griengl, H.
ligand 1 to (+)-(S)-2 led to a significant improvement in
the enantioselectivity. Under the same condition, the use
of 40 mol% of ligand 1 led to only 90% ee for the substrate
o-methoxybenzaldehyde. Although a slight high ligand
loading (20–40 mol%) was required for the silylcyana-
(
Tetrahedron 1996, 52, 7833. (c) Effenberger, F. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1555. (d) Mori, A.; Nitta,
H.; Kudo, M.; Inous, S. Tetrahedron Lett. 1991, 32, 4333.
Synlett 2004, No. 9, 1521–1524 © Thieme Stuttgart · New York