σ-π chelation-controlled stereoselective hydrosilylation of ketones

Full text of DOI:10.1021/ja0042074

J. Am. Chem. Soc. 2001, 123, 6931-6932
6931
Scheme 1
σ-π Chelation-Controlled Stereoselective
Hydrosilylation of Ketones
Naoki Asao, Takeshi Ohishi, Kenichiro Sato, and
Yoshinori Yamamoto*
Department of Chemistry
Graduate School of Science
Tohoku UniVersity, Sendai 980-8578, Japan
ReceiVed December 8, 2000
Table 1. σ-π Chelation-Controlled Hydrosilylation of 4a
Since Cram’s pioneering work on chelation control in Grignard-
type addition to chiral alkoxy carbonyl substrates,1 a number of
studies on related subjects have appeared.2 Among them, the
Lewis acid-mediated chelation control is one of the most
fundamental and practically important concepts in modern organic
chemistry.3 The concept of chelation control has been applicable
to carbonyl compounds bearing heteroatom-containing function-
alities such as an alkoxy group in appropriate proximity (σ-σ
chelation). To the best of our knowledge, there is no example of
the chelation-controlled stereoselective reaction of carbonyl
compounds through σ-π chelation. Recently, we reported that
the chelation controlled regio- and chemoselective reaction which
proceeds via the coordination of π-electrons of triple bonds to
Lewis acids.4 Now, we wish to report the first example for the
stereoselective reactions which are controlled by the σ-π
chelation (Scheme 1).
substrate 4
yield of
ratio
entry
R1
R2
R3SiH
4a Et3SiH
4a Ph2MeSiH
4b Et3SiH
5 and 6 (%)b syn-5:anti-6
1
2
3
4
5
6
7
8
9
Ph
Ph
Ph
Ph
Ph
Et
c-C6H11
o-MePh
tBu
H
H
Me
Ph
90
99
quant
quant
quant
quant
93
7.0:1
6.8:1
5.0:1
3.0:1
7.7:1
4.4:1
5.0:1
15:1
4c Et3SiH
TMS 4d Et3SiH
H
H
H
H
4e Et3SiH
4f Et3SiH
4g Et3SiH
4h Et3SiH
94
quant
>30:1
a Reaction was performed with R3SiH (1 equiv) and B(C6F5)3 (2 mol
%) in toluene at 0 °C within 1 h. b Isolated yield.
Scheme 2
We examined the stereoselective hydrosilylation of various
ketones using R3SiH-B(C6F5)3 as a reducing agent.5 The reaction
of 2-methyl-1-phenyl-pentan-1-one 1 with Et3SiH in the presence
of catalytic amounts of B(C6F5)3 proceeded smoothly to give a
mixture of the hydrosilylated products 2 and 3 in 98% yield (eq
1). Slightly predominant formation of the anti-product 3 over syn-
product 2 was observed; the ratio of 2:3 was 1:1.5. We next
examined the hydrosilylation of 2-methyl-1-phenyl-pent-4-yn-1-
one 4a (R1 ) Ph, R2 ) H) under the same reaction conditions as
above. Interestingly, the syn-product 5a was afforded as the major
product (5a:6a ) 7:1) (eq 2). This result prompted us to examine
the hydrosilylation of 4a and related ketones 4b-4h to clarify
the generality of this unusual diastereoselectivity. The results are
summarized in Table 1.
TMS groups at the terminal position of alkyne, respectively, also
gave syn-selectivities (entries 3-5). Not only aromatic ketones
but also aliphatic ketones 4e, 4f, and 4h produced syn-products
selectively (entries 6, 7, and 9). Interestingly, stereoselectivities
t
increased from 4.4:1 (R1 ) Et) to >30:1 (R1 ) Bu) as the
substituents at R1 position became bulkier. These results clearly
indicate that the syn diastereoselectivity is widely observed in
the B(C6F5)3-catalyzed reduction of 4 with hydrosilanes.
The stereostructures of 5a and 6a were unambiguously
determined by converting 5a and 6a to 9a and 10, respectively,
as shown in Scheme 2. The treatment of a mixture of 5a and 6a
(4.9:1) with TBAF, followed by the protection of the resulting
alcohols by MPMCl under basic condition gave 7 in 88% yield.
The alkynyl part of 7 was converted to a carboxylic acid by
hydroboration-oxidative workup, which was subsequently es-
terified to give 8 in 47% yield. Deprotection of the MPM group
of 8 by CAN gave a mixture of the lactones 9a and 10 in a ratio
of 4.6:1 in 87% yield. The 1H NMR spectrum of 9a was identical
to that of the known compound.6 The stereostructure of 5h, which
was obtained from the aliphatic ketone 4h, was also determined
by converting 5h to cis-6-tert-butyl-5-methyl-tetrahydro-pyran-
2-one (9b) via similar routes. The stereostructures of 5b-g and
6b-g were assigned by their 1H NMR spectra on the analogy of
those of 5a, 6a, and 5h.
The predominant formation of the syn-product was also
observed in the reaction of 4a with other silanes such as Ph2-
MeSiH (entry 2). The reactions of 4b-d, bearing Me, Ph, and
(1) (a) Cram, D. J.; Kopecky, K. R. J. Am. Chem. Soc. 1959, 81, 2748-
2755. (b) Leitereg, T. J.; Cram, D. J. J. Am. Chem. Soc. 1968, 90, 4019-4026.
(2) (a) Mengel, A.; Reiser, O. Chem. ReV. 1999, 99, 1191-1223. (b)
Modern Carbonyl Chemistry; Otera, J., Ed.; Wiley-VCH: Weinheim, 2000.
(3) (a) Lewis Acids in Organic Synthesis; Yamamoto, H., Ed.; Wiley-
VCH: Weinheim, 2000; Vols. 1-2. (b) Lewis Acid Reagents; Yamamoto,
H., Ed.; Oxford University Press: New York, 1999. (c) Mahrwald, R. Chem.
ReV. 1999, 99, 1095-1120. (d) Santelli, M.; Pons, J. M. Lewis Acids and
SelectiVity in Organic Synthesis; CRC Press: Boca Raton, FL,1996. (e)
Shambayati, S.; Schreiber, S. L. In ComprehensiVe Organic Synthesis; Trost,
B. M., Fleming I., Eds.; Pergamon Press: Oxford, 1991; Vol. 1, pp 283-
324. (f) Yamaguchi, M. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Fleming I., Eds.; Pergamon Press: Oxford, 1991; Vol. 1, pp 325-353.
(4) Asao, N.; Asano, T.; Ohishi, T.; Yamamoto, Y. J. Am. Chem. Soc.
2000, 122, 4817-4818.
(5) Piers and co-workers found that B(C6F5)3-catalyzed hydrosilylation of
carbonyl functions, such as aldehydes, ketones and esters, proceeded very
smoothly to give the corresponding reduced compounds in high yields. (a)
Parks, D. J.; Piers, W. E. J. Am. Chem. Soc. 1996, 118, 9440-9441. (b) Parks,
D. J.; Blackwell, J. M.; Piers, W. E. J. Org. Chem. 2000, 65, 3090-3098.
(6) Oshima, M.; Yamazaki, H.; Shimizu, I.; Nisar, M.; Tsuji, J. J. Am.
Chem. Soc. 1989, 111, 6280-6287.
10.1021/ja0042074 CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/21/2001

6932 J. Am. Chem. Soc., Vol. 123, No. 28, 2001
Communications to the Editor
stronger.8 The proposed chelation model also can explain the
reason the syn-selectivity was obtained very predominantly or
exclusively in the reaction of 4g and 4h having bulky R1 groups
Figure 1.
The difference of the diastereoselectivities between eqs 1 and
2 clearly shows that the acetylenic bond of 4 exerts a crucial role
upon the observed syn-selectivity. Piers et al. proposed the
interesting silane activation mechanism in the B(C6F5)3-catalyzed
hydrosilylation of aldehydes and ketones; the ordinary mechanism,
in which the carbonyl oxygen of the electrophiles coordinates to
B(C6F5)3 and thus carbonyl substrates are activated, is not
operative in the B(C6F5)3-catalyzed reduction.5 Their extensive
mechanistic studies clarify that B(C6F5)3 activates the silane via
hydride abstraction to form the incipient silylium species which
enhances the electrophilicity of the carbonyl group, facilitating
the reduction by [HB(C6F5)3]- or R3SiH (Figure 1).
Most probably, a silylium species is generated here also, and
the σ-π coordination of this species is operative in the reaction
of 4. The anti diastereoselectivity in the reaction of 1 can be
accounted for by the ordinary Felkin-Anh model. The propyl
group at the R-position is regarded as the largest group and the
Me as the medium size (model 11). Accordingly, anti-3 is
produced with slight preference, and the observed low stereose-
lectivity is due to the small steric difference between propyl and
methyl group at the R-position. On the contrary, in the reaction
of 4, the reduction would proceed through the σ-π chelation of
R3Si+ (model 12): the hydride attack takes place from the less
hindered side to produce the syn-isomer 5.
(entries 8-9). There is a possibility that hydride may attack from
the bottom side of carbonyl group in the conformer 15, which
produces the anti-isomer 6. On the other hand, the axially oriented
methyl group prevents the hydride attack from the bottom side
in the conformer 16. The conformer 16 is more favored with the
bulkier R1 group because of the increased steric repulsion between
R1 and Me in 15.9
The 1,2-asymmetric induction via the σ-π chelation control
could be extended to the 1,3-system. The hydrosilylation of
3-methyl-1-phenyl-5-trimethylsilyl-4-pentyn-1-one 17 with Ph2-
MeSiH in the presence of 2 mol % of B(C6F5)3 gave the anti-
product 18 stereoselectively (18:19 ) 6.5:1) (eq 3). In contrast,
no selectivity was observed in the reaction of the saturated
analogue 20 (eq 4). The stereostructure of 18 was unambiguously
determined by converting 18 to 2310 via a similar route to that
shown in Scheme 2. The anti-stereoselectivity in the reaction of
17 can be accounted for by the σ-π chelation model 24, which
involves hydride attack on the less hindered face of a conforma-
tionally locked, internally chelated intermediate.2a
If the ordinary Felkin-Anh model is involved also in the case of
4, the anti-diastereomer 6 should be produced predominantly,
since a propargyl group is sterically larger than a Me group.7
Indeed, the anti-selectivity was observed with slight predominance
when the reduction of 4a was carried out using DIBAL-H, in
which the ratio of 13:14 was 49:51.
Now it is clear that the σ-π chelation is operative not only in
the 1,2- but also in the 1,3-asymmetric induction of certain
acetylenic ketones. The syn-diastereoisomers obtained either
exclusively or predominantly in the reaction of 4 or the anti-
isomer in the reaction of 17 can be converted, upon reduction of
the triple bond, to the anti-Felkin-Anh products which are not
easily available through the ordinary reducing methods. We are
now in a position to apply the σ-π coordination concept along
with the well-known σ-σ chelation to control stereoselectivities.
The stereoselectivities decreased as the substituents R2 of 4
became bulky (entries 1, 3, and 4). Presumably, a bulky R2 group
would make it difficult to form strong σ-π chelation in 12. Higher
selectivity obtained in the case of 4d may be explained by the
well-known â-silyl effect, which would make the chelation
(7) The B(C6F5)3-catalyzed hydrosilylation of 25 with Et3SiH afforded the
syn-isomer 26 as a sole product in 99% yield (eq 5). Both the σ-π chelation
and Felkin-Anh model leads to the syn-isomer, since isopropyl group at the
R-position of 25 is sterically larger than propargyl group. On the other hand,
the syn-selectivity was decreased (syn-28:anti-29 ) 98:2) in the hydrosilylation
of 27 bearing a saturated propyl group instead of a propargyl group at the
R-position, under the same reaction condition (eq 6). These results clearly
imply the σ-π chelation can be used not only for reversing the Felkin-Anh
selectivity but also for increasing it by choosing the substituent at the R-position
of carbonyl compounds.
Supporting Information Available: Spectroscopic and analytical data
for 2, 3, 5, 6, 9, 10, 18, 19, 21, 22, 23, 26, and 28, and the representative
procedure for the synthesis of 5h (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
JA0042074
(8) For reviews, see: (a) Lambert, J. B. Tetrahedron 1988, 46, 2677-
2689. (b) Apeloig, Y. In The Chemistry of Organic Silicon Compounds; Patai,
S., Rappoport, Z., Eds.; Wiley-Interscience: New York, 1989; pp 57-225.
(c) Fleming, I. Chemtracts: Org. Chem. 1993, 6, 113-116.
(9) Sarko, C. R.; Guch, I. C.; DiMare, M. J. Org. Chem. 1994, 59, 705-
706.
(10) Barbero, A.; Blakemore, D. C.; Fleming, I.; Wesley, R. N. J. Chem.
Soc., Perkin Trans. 1 1997, 1329-1352.