undergo regioselective (7 and 10:1 rr) methylene insertion to give
cyclopentanones 8a and 8b in >75% yield (entries 8 and 9). We
have also demonstrated tolerance for ortho substitution within the
monoarylated series. As shown in entries 10-12, o-chloro-,
o-bromo-, and o-methoxyphenyl cyclobutanones transform to the
corresponding 5-C ring ketones with g5:1 regioselectivity and
63-72% yield. A representative dialkyl cyclobutanone (11) was
tested as well. Selectivity falls to 3:1 in the absence of the bulky
arene substituent, but in this case the R-quaternary cyclopentanone
(12) is still recovered with good efficiency (70% yield).
homologation of 1a proceeds cleanly to an 8:1 mixture of
diastereomeric ꢀ-keto silanes with a 12:1 predominance of
secondary vs quaternary carbon migration (Scheme 3). The
Scheme 3
.
Diversifying Reactions Made Possible by hfac
Ligation
Closer analysis of the reaction pathway for homologation of
ketone 1a was revealing. As shown in Scheme 2, real-time
Scheme 2
.
Carbosilane Intermediate Detected by reactIR
Analysis
major product (13, 76%) is stable to silica gel and readily
separable from the minor components of the reaction.15 Its 2,5-
trans configuration is assigned by analogy to the more enabling
transformation depicted at the bottom of Scheme 3. Thus,
Sc(hfac)3-catalyzed reaction with PhMe2SiCHN216 smoothly
converts 1a to organosilane 15 with comparable diastereo- and
regioselectivity. Comparative NOE analysis on both diastere-
omers established the relative configuration in 15 as that shown.
After stereoselective LAH reduction to a secondary alcohol,
repeated attempts to oxidize the C-Si bond gave product
mixtures. Silyl protecting groups did not hold up to the acidic
conditions needed, but reaction on the derived acetate was
favorable. Cyclopentanediol 16 can be accessed in 54% yield
over four steps that include Fleming oxidation17 of the acetoxy
silane and acetate ester removal. Subsequent acetalization with
p-bromobenzaldehyde gives a cis-configured dioxolane, con-
firming the stereochemistry of the reduction. Interestingly, upon
revisiting the issue of Sc(OTf)3 catalysis of Brook rearrangement
in 13, we found protodesilylation (to 2a) as the exclusive result
(Scheme 3). At this time, it is not clear why enol silane 14, if
generated catalytically from 13, is stable only under the
preparative conditions for 1-C ring expansion. A possible role
for the donor-acceptor nucleophile in mediating silicon transfer
is considered next in a discussion of reaction mechanism.
monitoring of the event by reactIR spectroscopy confirms
a short-lived (1 h) intermediacy of ꢀ-keto silane 13, whose
carbonyl stretch (1717 cm-1) is easily discernible from that of
the starting 1a (1775 cm-1). Since authentic 13 is a stable
species (vide infra), Sc(OTf)3 appears to play a dual role in the
transformation, catalyzing (1) rapid (<30 min) insertion of the
TMSCH moiety into the less hindered R C-C bond of 1a and
(2) a more gradual 1,3-Brook rearrangement13 of 13 to 14. At
this point, we sought to gain access to carbosilane 13 by other
means so that C f O Si transfer could be studied under more
controlled conditions. While synthetic methods exist14a for such
useful intermediates,14b it was intriguing to consider that a less
Lewis acidic Sc trication might halt reaction at ꢀ-keto silane
13, especially since the kinetics of carbon insertion appeared
faster than Brook isomerization.
(11) Other recent literature on the utility of this reagent: (a) Dias, E. L.;
Brookhart, M.; White, P. S. J. Am. Chem. Soc. 2001, 123, 2442–2443. (b)
Greenman, K. L.; Carter, D. S.; Van Vranken, D. L. Tetrahedron 2001,
57, 5219–5225. (c) Aggarwal, V. K.; Sheldon, C. G.; Macdonald, G. J.;
Martin, W. P. J. Am. Chem. Soc. 2002, 124, 10300–10301. (d) Lebel, H.;
Paquet, V. J. Am. Chem. Soc. 2004, 126, 320–328. (e) Ni, Y.; Montgomery,
J. J. Am. Chem. Soc. 2004, 126, 11162–11163. (f) Ku¨hnel, E.; Laffan,
D. D. P.; Lloyd-Jones, G. C.; del Campo, T. M.; Shepperson, I. R.; Slaughter,
J. L. Angew. Chem., Int. Ed. 2007, 46, 7075–7078. (g) Reference 7.
(12) For commentary on this trend, see: Seyferth, D.; Dow, A. W.; Menzel,
H.; Flood, T. C. J. Am. Chem. Soc. 1968, 90, 1080–1082.
(13) (a) Brook, A. G. Acc. Chem. Res. 1974, 7, 77–84. (b) Moser, W. H.
Tetrahedron 2001, 57, 2065–2084. (c) Brook, A. G.; Limburg, W. W.;
MacRae, D. M.; Fieldhouse, S. A. J. Am. Chem. Soc. 1967, 89, 704–706.
(14) (a) Corey, E. J.; Ru¨cker, C. Tetrahedron Lett. 1984, 25, 4345–
4348. (b) Obayashi, M.; Utimoto, K.; Nozaki, H. Bull. Chem. Soc. Jpn.
1979, 52, 1760–1764.
(15) The regioisomeric R-silyl, ꢀ-quaternary ketone was recovered as a
single diastereomer (6% yield) along with 10% of the R-silyl epimer
of 13.
(16) Prepared by a procedure for TMSD from (chloromethyl)dimethylphe-
nylsilane: Shiori, T.; Aoyama, T.; Mori, S. Org. Synth. 1990, 68,
1-4.
(17) Fleming, I.; Sanderson, P. E. J. Tetrahedron Lett. 1987, 28, 4229.
In a screen with other commercial Sc(III) salts, Sc(acac)3 and
Sc(tmhd)3 were completely ineffective (<2% conv), but
Sc(hfac)3 gave the desired outcome. With 10 mol % catalyst,
3600
Org. Lett., Vol. 12, No. 16, 2010