A,3-Allylic Strain as a Control Element in the Paterno-Buechi Reaction of Chiral Silyl Enol Ethers: Synthesis of Diastereomerically Pure Oxetanes Containing Four Contiguous Stereogenic Centers

Article abstract of DOI:10.1021/ja963827v

The facial diastereoselectivity in the Paterno-Buechi reaction of chiral enol ethers and benzaldehyde was studied. The substituents (R^S, R^L) at the stereogenic carbon atom (-C*HR^SR^L) attached to the β-position o

Full text of DOI:10.1021/ja963827v

J. Am. Chem. Soc. 1997, 119, 2437-2445
2437
1,3-Allylic Strain as a Control Element in the Paterno`-Bu¨chi
Reaction of Chiral Silyl Enol Ethers: Synthesis of
Diastereomerically Pure Oxetanes Containing Four Contiguous
Stereogenic Centers
Thorsten Bach,* Kai Jo1dicke, Kristian Kather, and Roland Fro1hlich†
Contribution from the Organisch-Chemisches Institut der Westfa¨lischen Wilhelms-UniVersita¨t
Mu¨nster, D-48149 Mu¨nster, Germany
ReceiVed NoVember 4, 1996X
Abstract: The facial diastereoselectivity in the Paterno`-Bu¨chi reaction of chiral silyl enol ethers and benzaldehyde
was studied. The substituents (RS, RL) at the stereogenic carbon atom (-C*HRSRL) attached to the â-position of the
silyl enol ether were varied in order to evaluate the influence of steric bulk and electronic effects. The combined
yields for the two diastereomeric 3-(silyloxy)oxetanes a and b range between 44% and 76%. In accordance with the
1,3-allylic strain model the facial diastereoselectivity (diastereomeric ratio (dr) ) a/b) was best with large (RL )
t-Bu, SiMe2Ph) and polar (RL ) OMe) substituents at the γ-position of the silyl enol ether (dr up to >95/5). Two
regioselective ring opening reactions were applied to the product oxetanes 29a, 50, and 51. They furnished
diastereomerically pure diols (52, 53) and triols (31) in excellent yields.
Introduction
Scheme 1
The Paterno`-Bu¨chi reaction has long been established as one
of the most important and most useful photochemical transfor-
mations.1 Since up to three new stereogenic centers are formed
in the course of the reaction, there have been many studies as
to both the relative and the absolute configurations of the product
oxetanes.2 The former aspect of relative configuration is related
to two basic topics, i.e., the simple diastereoselectivity of C-C
bond formation3 and the stereospecifity of the process with
regard to the alkene configuration.4 The latter aspect of absolute
configuration addresses the problem of facial diastereoselectivity
if a chiral moiety is attached to either one of the reaction partners
or of enantioselectivity if a catalytic approach is pursued.5 From
a synthetic point of view it is desirable to minimize the numbers
of stereoisomers in order to retain maximum efficiency. We
have recently shown that silyl enol ethers of the general type I
react well with aromatic aldehydes and yield the diastereomeri-
cally pure oxetanes II (Scheme 1).6
relatively long lived triplet species (3D), and the reaction
proceeds therefore in a stereoconvergent manner; i.e., it is
independent of the initial olefin configuration.6d In a second
selection step the simple diastereoselectivity between the newly
formed stereogenic centers at C-2 and C-3 is determined. The
impact of viable intersystem crossing (ISC) geometries8 can be
counterbalanced or overridden by the competition between
retrocleavage and ring closure in a short-lived singlet biradical
(1D). We have postulated that it is this competition which
dictates the simple diastereoselectivity for the silyl enol ether
photocycloaddition.6b Oxetanes of the general structure II
represent 1,2,3-trifunctional building blocks which can be used
readily in successive ring-opening reactions.9
If one of the two alkene substituents R or R1 is chiral, the
question of facial diastereoselectivity comes into play. In order
to achieve a complete face differentiation, the attachment of a
stereogenic center to the â-position of the silyl enol ether III
(R1 chiral) appeared to be particularly promising to us (Scheme
2). The obvious reason for this notion was the preferred
conformation III′ to be predicted on the grounds of 1,3-allylic
The relative configuration established in this reaction is
3
determined in the intermediate 1,4-biradical D.7 The bulky
alkyl groups R and R1 are oriented trans to each other in the
* To whom correspondence should be addressed. New address: Philipps-
Universita¨t Marburg, Fachbereich Chemie, D-35032 Marburg, Germany.
† To whom inquiries about the X-ray analysis should be addressed at
the Universita¨t Mu¨nster.
X Abstract published in AdVance ACS Abstracts, February 15, 1997.
(1) (a) Paterno`, E.; Chieffi, G. Gazz. Chim. Ital. 1909, 39-1, 341-361.
(b) Bu¨chi, G.; Inman, C. G.; Lipinsky, E. S. J. Am. Chem. Soc. 1954, 76,
4327-4331.
(2) For recent reviews, see: (a) Mattay, J.; Conrads, R.; Hoffmann, R.
In Methoden der Organischen Chemie (Houben-Weyl) 4te Aufl.; Helmchen,
G., Hoffmann, R. W., Mulzer, J., Schaumann, E., Eds.; Thieme: Stuttgart,
1995; Vol. E 21c, pp 3133-3178. (b) Porco, J. A.; Schreiber, S. L. In
ComprehensiVe Organic Synthesis; Trost, B., Ed.; Pergamon Press: Oxford,
1991; Vol. 5, pp 151-192. (c) Carless, H. A. J. In Synthetic Organic
Photochemistry; Horspool, W. M., Ed.; Plenum Press: New York, 1984;
pp 425-487. (d) Jones, G. In Organic Photochemistry; Padwa, A., Ed.;
Dekker: New York, 1981; Vol. 5, pp 1-123.
(3) Bach, T. Angew. Chem. 1996, 108, 976-977; Angew. Chem., Int.
Ed. Engl. 1996, 35, 884-886 and references cited therein.
(4) Khan, N.; Morris, T. H.; Smith, E. H.; Walsh, R. J. Chem. Soc.,
Perkin Trans. 1 1991, 865-870 and references cited therein.
(5) Review: Inoue, Y. Chem. ReV. 1992, 92, 741-770.
(6) (a) Bach, T. Tetrahedron Lett. 1991, 32, 7037-7038. (b) Bach, T.;
Jo¨dicke, K. Chem. Ber. 1993, 126, 2457-2466. (c) Bach, T. Tetrahedron
Lett. 1994, 35, 5845-5848. (d) Bach, T. Liebigs Ann. Chem. 1995, 855-
866.
(7) First detection of a preoxetane 1,4-biradical: Freilich, S. C.; Peters,
K. S. J. Am. Chem. Soc. 1981, 103, 6255-6257. Freilich, S. C.; Peters, K.
S. J. Am. Chem. Soc. 1985, 107, 3819-3822.
(8) For a comprehensive review on the various intersystem crossing
geometries related to oxetane formation, see: Griesbeck, A. G.; Mauder,
H., Stadtmu¨ller, S. Acc. Chem. Res. 1994, 27, 70-75.
(9) (a) Bach, T. Tetrahedron Lett. 1994, 35, 1855-1858. (b) Bach, T.
Liebigs Ann. Chem. 1995, 1045-1053. (c) Bach, T.; Kather, K. J. Org.
Chem. 1996, 61, 3900-3901. (d) Bach, T.; Lange, C. Tetrahedron Lett.
1996, 37, 4363-4364.
S0002-7863(96)03827-9 CCC: $14.00 © 1997 American Chemical Society

2438 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Bach et al.
Scheme 2
Scheme 3a
a Reagents and conditions: (a) -78 f +25 °C (THF); (b) Me2CuLi,
-78 f 0 °C, TMSCl, NEt3, 0 °C (THF); (c) 30 °C (PhH).
Table 1. Facial Diastereoselectivity in the Photocycloaddition of
Chiral, γ-Alkyl-Substituted Silyl Enol Ethers
alkene
RG
product yielda (%) drb (a/b) regioselectivityc
5
6
7
8
Et
i-Pr
Ph
9
10
11
12
61
57
76
58
61/39
70/30
71/29
95/5
87/13
82/18
80/20
75/25
strain.10 As depicted in Scheme 2 a conformation III′ in which
the hydrogen atom at the stereogenic center and the R-carbon
atom of the silyl enol ether are locked in a synperiplanar fashion
is strongly favored. An approach of the photoexcited carbonyl
compound should occur from the less shielded face. If the
further course of the reaction included the previously encoun-
tered selection steps (Vide supra), the stereogenic center in the
γ-position of the silyl enol ether III was to control the relative
configuration of all stereogenic centers within the oxetane V.
Contrary to the observations made by Scharf et al. in their work
on the auxiliary assisted Paterno`-Bu¨chi reaction of chiral phenyl
glyoxylates,11 the retrocleavage was not expected to significantly
influence the facial diastereoselectivity. For the remote stereo-
genic center in the 1,4-biradical IV, a conformational preference
with regard to the prostereogenic radical centers was not
foreseen. An impact on the retrocleavage in either one of the
two diastereoisomers IVa or IVb seemed unlikely. This
hypothesis which was borne out experimentally (Vide infra)
simplifies the discussion as only the C-O bond formation step
has to be considered to explain the facial diastereoselectivity.
The strategy of acyclic stereoselection has rarely been
exploited in Paterno`-Bu¨chi reactions.12 Most studies are
concerned with the facial diastereoselectivity in cyclic systems.13
Furthermore, concave auxiliaries such as 8-phenylmenthol have
been established as reliable control elements.11 Examples of
good diastereofacial control based on 1,3-allylic strain was
reported by Schreiber and Hoveyda in the context of intramo-
lecular Paterno`-Bu¨chi reactions.14 In this paper we present
our results on the successful embodiment of 1,3-allylic strain
in the intermolecular photocycloaddition of chiral silyl enol
ethers and aldehydes.15 In addition, two facile ring-opening
t-Bu
a Total yield of the two diastereoisomers a and b. b Diastereomeric
1
ratio determined by H NMR analysis of the crude product mixture.
c Ratio of regioisomers (3- vs 2-(silyloxy)oxetane) determined by GLC
analysis of the crude product mixture.
reactions are described which demonstrate the synthetic versatil-
ity of functionalized oxetanes.
Results and Discussion
Influence of Steric Bulk. The first series of silyl enol ethers
we synthesized and tested in the Paterno`-Bu¨chi reaction carried
a tert-butyl substituent in the R-position (R ) t-Bu) and a methyl
group at the stereogenic center in the γ-position (RS ) Me).
The additional γ-substituent RL was varied in size. In general,
the ketones 1-316 are readily transferred to the corresponding
silyl enol ethers 6-8 by kinetically controlled deprotonation
and subsequent quench with chlorotrimethylsilane (TMSCl) at
-78 °C (Scheme 3).17 As the ketones are generated by
conjugate cuprate addition to 2,2-dimethyl-4-hexen-3-one, the
enolate can also be directly trapped with TMSCl.18 In the case
of compound 8 the yield improved considerably by employing
the latter protocol. For the preparation of silyl enol ether 5 the
conjugate addition/enolate quench was applied to ketone 4.19
By irradiation of the olefinic substrates 5-8 in the presence of
benzaldehyde, we obtained preliminary information as to the
steric influence of the γ-substituent RL.
From the data recorded in Table 1 it can be seen that only
two stereoisomeric oxetanes a and b were isolated. The relative
configuration within the four-membered ring is identical in either
isomer, but the isomers differ in the stereochemical relation
between the exocyclic stereogenic center and the carbon atom
(10) Review: Hoffmann, R. W. Chem. ReV. 1989, 89, 1841-1860.
(11) (a) Buschmann, H.; Scharf, H.-D.; Hoffmann, N.; Plath, M. W.;
Runsink, J. J. Am. Chem. Soc. 1989, 111, 5367-5373. (b) Buschmann, H.;
Scharf, H.-D.; Hoffmann, N.; Esser, P. Angew. Chem. 1991, 103, 480-
518; Angew. Chem., Int. Ed. Engl. 1991, 30, 477-515 and references cited
therein.
(12) (a) Jarosz, S.; Zamojski, A. Tetrahedron 1982, 38, 1453-1456. (b)
Schreiber, S. L. Science 1985, 227, 857-863. (c) Zagar, C.; Scharf, H.-D.
Chem. Ber. 1991, 124, 967-969. (d) Bach, T.; Jo¨dicke, K.; Wibbeling, B.
Tetrahedron 1996, 52, 10861-10878.
1
C-4. Their ratio which can be determined by both H NMR
and GLC reflects the extent of facial diastereoselection exerted
by the stereogenic center in the silyl enol ether. The higher
the steric demand of the substituent RL the better is the
diastereomeric ratio (dr). On the basis of the preferred
conformation III′ depicted in Scheme 2, the results can be
(13) (a) Morton, D. R.; Morge, R. A. J. Org. Chem. 1978, 43, 2093-
2101. (b) Araki, Y.; Senna, K.; Matsuura, K.; Ishido, Y. Carbohydr. Res.
1978, 60, 389-395. (c) Vasudevan, S.; Brock, C. P.; Watt, D. S.; Morita,
H. J. Org. Chem. 1994, 59, 4677-4679.
(16) (a) 1: Azzena, U.; Luisi, P. L.; Suter, U. W. HelV. Chim. Acta 1981,
64, 2821-2829. (b) 2: Malmberg, H.; Nilson, M.; Ullenius, C. Acta. Chem.
Scand., Ser. B 1981, B35, 625-629. (c) 3: House, H. O:; Weeks, P. D. J.
Am. Chem. Soc. 1975, 97, 2770-2777.
(17) Heathcock, C. H. In Modern Synthetic Methods; Scheffold, R., Ed.;
Verlag Helvetica Chimica Acta: Basel, 1992; Vol. 6, pp 1-102.
(18) Stork, G.; Hudrlik, P. F. J. Am. Chem. Soc. 1968, 90, 4462-4464.
(19) Ketone 4: Oare, D. A.; Henderson, M. A.; Sanner, M. A.;
Heathcock, C. H. J. Org. Chem. 1990, 55, 132-157.
(14) Hoveyda, A. H. Ph.D. Thesis, Yale University, 1986, cited in ref
2b.
(15) Preliminary communication: Bach, T., Jo¨dicke, K.; Kather, K.;
Hecht, J. Angew. Chem. 1995, 107, 2455-2457; Angew. Chem., Int. Ed.
Engl. 1995, 34, 2271-2273.

Paterno`-Bu¨chi Reaction of Chiral Silyl Enol Ethers
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2439
Scheme 4a
a Reagents and conditions: (a) neat, 80 °C; (b) CuCl, -78 °C (Et2O);
(c) -78 f +25 °C (THF); (d) 30 °C (PhH).
Figure 3. Possible conformations which explain the facial differentia-
tion in the Paterno`-Bu¨chi reaction of the silyl enol ethers 5-8.
Figure 1. Structure of oxetane 11a in the crystal.
expected to be predominantly influenced by the activation
entropy difference ∆∆S *. In the case of oxetane 11 we were
able to show that there is indeed a linear dependence of ln(k/k′)
and 1/T in the (experimentally) limited temperature range from
-30 to +40 °C (Eyring plot in Figure 2) where k/k′ represents
the facial diastereoselection as determined by the diastereomeric
ratio (dr). The calculated values for ∆∆H q and ∆∆S q are 4.5
kJ mol-1 and 21.3 J mol-1 K-1, respectively. These figures
clearly indicate an entropically determined diastereoselection.
In the temperature range we studied there is no apparent change
in the mechanism. The competition beween retrocleavage and
ring closure in a 1,4-biradical which is known to be enthalpy
determined11a does not influence the facial diastereoselectivity
even at comparably high temperature. Preparatively, the
increase of the diastereomeric ratio from 58/42 at -30 °C to
70/30 at +40 °C is significant and provides a handle for
improving the selectivity of the photocycloaddition.
Figure 2. Eyring plot relating the facial diastereoselectivity observed
in the photocycloaddition of silyl enol ether 7 and benzaldehyde to
1/T.
readily understood. It is the difference in bulk between RL and
RS ) Me which favors the Si-attack. With the extremely bulky
tert-butyl group (RL ) t-Bu) a clear-cut preference (dr ) >95/
5) was noted. The total yields for both diastereomers a and b
range between 57% and 72%. The decrease in yield as
compared with related 3-(silyloxy)oxetanes6d is partially con-
nected to an increasing amount of regioisomer (2-(silyloxy)-
oxetane) formed by attack of the photoexcited aldehyde at the
R-position of the silyl enol ether. The regioselectivity as defined
in Table 1 was determined by GLC integration. For this purpose
the peak assignment was based on GLC-MS data recorded
under otherwise identical conditions. In the MS the main
fragmentation pathway is the retro-[2 + 2]-cycloaddition which
occurs in either direction, i.e., between O/C-2 and C-3/C-4 as
well as between O/C-4 and C-2/C-3. A distinction between
the regioisomers by this method is therefore facile and unam-
biguous.
The structure elucidation was carried out by single-crystal
X-ray analysis. Figure 1 represents the structure of the major
diastereoisomer 11a in the crystal. It is remarkable that even
in the oxetane the proton at the exocyclic stereogenic center
and the carbon atom C-3 which carries the silyloxy group adopt
an eclipsed orientation. From Figure 1 one can easily envisage
the approach of benzaldehyde to the less shielded Si-face of
the silyl enol ether 7 to yield the major product 11a. The
conformational preference is governed by 1,3-allylic strain as
still evident in the product.
Since we intended to prove that the oxetane formation
proceeds without racemization, the silyl enol ether 7 was also
prepared in enantiomerically pure form. For this purpose the
readily available carboxylic acid 1321 was converted into the
ketone (+)-2 via the corresponding acyl chloride 14 (Scheme
4). Silyl enol ether formation and subsequent Paterno`-Bu¨chi
reaction proceeded uneventfully and provided the oxetane (+)-
11a whose enantiomeric purity was determined to be >95%
enantiomeric excess (ee) by shift experiments with tris[3-
[(heptafluoropropyl)hydroxymethylene]-D-camphorato]europi-
um (Eu(hfc)3). With respect to the ee determination the tert-
butyl substituent at the C-3 position of the oxetane was
1
deliberately chosen. Its H NMR signal was most sensitive
toward the shift reagent.
Polar Groups in the γ-Position. On the basis of stereo-
electronic considerations and in agreement with the Curtin-
Hammett principle,22 one may also trace back the facial
diastereoselection in the cases described above to the conforma-
tion III′′ depicted in Figure 3.23 It differs only slightly from
the conformation III′ discussed earlier and should be readily
accessible. The large substituent RL is positioned orthogonal
to the double bond plane, increasing the HOMO energy of the
nucleophilic π-system by σπ-interaction. In the transition state
the antiperiplanar orientation of its σ-bond to the developing
σ*-orbital leads to a considerable stabilization.23b The attack
of the carbonyl triplet and the C-O bond formation occur from
The decisive facial diastereoselection occurs in the C-O-
bond-forming step, i.e. the attack of the photocexcited carbonyl
triplet at the â-carbon of the olefin (cf. Scheme 2). In the loose
early transition state of such a reaction the entropy levels should
be narrowly spaced.20 As a consequence the selectivity is
(21) Mukaiyama, T.; Iwasawa, N. Chem. Lett. 1981, 913-916.
(22) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
Wiley: New York, 1994; pp 651-654.
(23) (a) McGarvey, G. J.; Williams, J. M. J. Am. Chem. Soc. 1985, 107,
1435-1437. (b) Houk, K. N.; Paddon-Row, M. N.; Rondan, N. G.; Wu,
Y.-D.; Brown, F. K.; Spellmayer, D. C.; Metz, J. T.; Li, Y.; Lonchrich, R.
J. Science 1986, 231, 1108-1117 and references cited therein.
(20) Skell, P. S.; Cholod, M. S. J. Am. Chem. Soc. 1969, 91, 7131-
7137.

2440 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Bach et al.
Scheme 5
Scheme 6a
the face opposite RL. Whereas conformation III′ accounts well
for a face differentiation on steric grounds according to the 1,3-
allylic strain model, conformation III′′ takes the stereoelectronic
aspects into consideration.
Precedence for the addition of an electrophilic radical to a
prostereogenic double bond carrying a chiral substituent can
be found in the work of Curran et al. who employed enol ethers
derived from chiral alcohols.24 We are, however, not aware of
studies in which the â-substituent of a heteroatom-substituted
alkene is chiral. The reverse case, i.e., a nucleophilic radical
attack to an acceptor substituted alkene with a chiral â-sub-
stituent, is amply documented.25 For comparison, reports about
electrophilic reactions may be consulted, i.e., on the alkylation
of chiral enolates. The facial diastereoselectivities observed in
these reactions parallel our results for simple alkyl substituents
RL.26 It is interesting to note that polar substituents have a strong
impact on the facial diastereoselectivity. An example is shown
in Scheme 5.23a Enolate alkylation of ester 15 results in
preferential formation of diastereoisomer 16 (dr ) 90/10). The
success of the TBDMSOCH2 group was correlated with its
donor character which makes it act as RL in a conformation
similar to III′′.
a Reagents and conditions: (a) n-BuLi, -30 f 0 °C, Piv2O, -50
f +25 °C (THF). (b) -78 °C (Et2O); (c) Pd(OH)2, 25 °C (MeOH);
(d) -78 f +25 °C (THF); (e) 30 °C (PhH).
Scheme 7a
As the A values of Me and ROCH2 are roughly the same,27
their difference in size can be considered marginal. A silyl enol
ether which bears a chiral substituent -CHMeCH2OR in the
â-position was selected to distinguish between stereoelectronic
(conformation III′′) and sheer steric (conformation III′) effects.
For its synthesis, the conjugate addition of known R-alkoxy-
anions28 to 2,2-dimethyl-4-hexen-3-one proved to be not suited.
A different route was chosen. The TBDMS-protected propar-
gylic alcohol 1729 served as the starting material which was
acylated according to a known procedure.30 Conjugate addition
to the alkyne 18 delivered ketone 19 as a mixture of dia-
stereoisomers (dr ) 60/40) (Scheme 6). Hydrogenation of this
intermediate to its saturated analogue 20 and subsequent enolate
silylation furnished the desired silyl enol ether 21. Upon
irradiation in the presence of benzaldehyde a facile photo-
cycloaddition occurred. Disappointingly, the two oxetanes 22
were formed in a ratio of roughly 1:1 (regioselectivity 88/12).
In contrast to the above mentioned enolate alkylations, the
stereoelectronic donor capacity of the ROCH2 substituent does
not significantly influence the facial diastereoselectivity in
Paterno`-Bu¨chi reactions of chiral silyl enol ethers. Conse-
quently, our picture of a conformation III′ which is attacked
preferentially from the less shielded face in an early tran-
sition state describes best these and the other results so far
obtained.
a Reagents and conditions: (a) 1,8-bis(dimethylamino)naphthalene,
25 °C (CH2Cl2); (b) -78 f +25 °C (THF); (c) 30 °C (PhH).
Next we turned our attention to the possible installation of
an alkoxy group at the stereogenic center. Attempts to disilylate
the hydroxy ketone 2331 were not successful although there was
precedence for a similar reaction conducted with a â-hydroxy
ester.32 Elimination products were observed presumably be-
cause the intermediate â-silyloxy ketone enolate is silylated
slowly and the E1cB process is favored. The same problem
was anticipated with other hydroxy-protected derivatives of
compound 23. As it turned out, the methoxy (RL ) OMe)33
substituent resists an elimination successsfully and we obtained
the silyl enol ether 27 conveniently in two steps from 2,2-
dimethyl-5-hydroxy-3-heptanone (23) via ketone 25 (Scheme
7). The photocycloaddition of 27 to benzaldehyde proceeded
with good facial diastereoselectivity (dr ) 85/15), and the major
diastereoisomer 29a was isolated in 54% yield. Its relative
configuration was determined by X-ray crystallography after
desilylation (K2CO3 in MeOH) to the corresponding oxetanol.15
If the size of the alkyl substituent (RS) in the γ-position is
increased, the selectivity decreases. With ethyl compound 28
which was obtained from ketone 2434 in analogy to the silyl
enol ether 27 the diastereomeric ratio of oxetane 30 dropped
slightly to 80/20. The yield is significantly lower, a fact which
can be attributed to the deterioration of the regioselectivity and
to a higher tendency for hydrogen abstraction reactions in 28.
(24) Curran, D. P.; Geib, S. J.; Kuo, L. H: Tetrahedron Lett. 1994, 35,
6235-6238. We thank a reviewer for providing us with this reference.
(25) Reviews: (a) Smadja, W. Synlett 1994, 1-26. (b) Curran, D. P.;
Porter, N. A.; Giese, B. Stereochemistry of Radical Reactions; VCH:
Weinheim, 1996.
(26) (a) Fleming, I.; Lewis, J. J. J. Chem. Soc., Chem. Commun. 1985,
149-151. (b) Yamamoto, Y.; Maruyama, K. Ibid. 1984, 904-905.
(27) Hirsch, J. A. Top. Stereochem. 1967, 1, 199-222.
(28) Corey, E. J.; Eckrich, T. M. Tetrahedron Lett. 1983, 24, 3165-
3168.
(29) Bundy, G. L.; Lin, C. H.; Sih, J. C. Tetrahedron 1981, 37, 4419-
4430.
(30) Chopra, A. K.; Moss, G. P.; Weedon, B. C. L. J. Chem. Soc., Perkin
Trans. 1 1988, 1371-1382.
(31) Bienvenue, A. J. Am. Chem. Soc. 1973, 95, 7345-7353.
(32) Shirai, F.; Nakai, T. Tetrahedron Lett. 1988, 29, 6461-6464.
(33) Methylation: Diem, M. J.; Burow, O. F.; Fry, J. L. J. Org. Chem.
1977, 42, 1801-1802.
(34) Singh, B. M.; Aguliar, F. J.; Brunet, E. Tetrahedron 1994, 50, 8169-
8184.

Paterno`-Bu¨chi Reaction of Chiral Silyl Enol Ethers
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2441
Scheme 8
Scheme 9a
Figure 4. Electrostatic repulsion as possible reason for the observed
facial diastereoselection in the Paterno`-Bu¨chi reaction of γ-methoxy-
substituted silyl enol ethers.
To explain the face selectivity in the photocycloaddition of
silyl enol ether 27, we again rely on the initially mentioned
1,3-allylic strain model (Scheme 2). Electrophilic addition
reactions to chiral allylic ethers are well precedented,35 and
several rationalizations have been proposed which try to account
for the partially contradictory results.36 For cases in which a
(Z)-substituent on the double bond severely increases the 1,3-
allylic strain, it is generally accepted that a conformation such
as III′ is strongly favored by 12-15 kJ mol-1 10
There is no
.
a Reagents and conditions: (a) -78 f +25 °C (THF); (b) NMO,
molecular sieves 4 Å, 25 °C (CH2Cl2); (c) -78 f +25 °C (THF); (d)
30 °C (PhH).
reason to believe that this preference will dramatically change
by going from the ground state to the transition state. Still,
small variations in the dihedral angle ((30°) of the bond
connecting the stereogenic carbon atom and the double bond
are possible. In our example an approach of the electrophilic
triplet has taken place from the face shielded by the methyl
group (RS ) Me) and opposite the methoxy substituent (RL )
OMe). On stereoelectronic and electrostatic37 grounds such a
trajectory appears not particularly appealing at first sight. We
have earlier seen (Vide supra), however, that stereoelectronic
considerations do not apply to the photocycloaddition. It
remains to clarify why the attack occurs opposite the methoxy
group in conformation III′, i.e., why the electrophile avoids the
nucleophilic substituent. In our opinion the clue to this apparent
paradox lies in the electronic situation of the reactive nπ*-
excited carbonyl group which embraces both an electrophilic
carbonyl oxygen and a nucleophilic carbon atom. As depicted
in Figure 4 the C-O bond formation will occur perpendicularly
to the double bond plane or slightly twisted. During the
approach to the double bond the electron rich carbonyl carbon
atom will turn away from the olefinic π-system. It will therefore
get into close proximity to the γ-substituents and it will avoid
for electrostatic reasons the nucleophilic methoxy group. The
attack from the opposite face is favored. By increasing the steric
bulk of the alkyl substituent situated at this face (RS ) Et), the
selectivity decreases.
further. An example is shown in Scheme 8. We currently
investigate additional consecutive reactions and their use for
the synthesis of biologically relevant products (Vide infra).
For the synthesis of enantiomerically pure compounds (S)-
malic acid serves as a readily available starting material.
Aldehyde 32,39 for example, can be converted to the silyl enol
ether 35 in three steps (Scheme 9). The t-BuMgCl addition
proceeded well and gave alcohol 33. Oxidation of the alcohol
33 with NMO/TPAP (cat.)40 and subsequent enolate silylation
gave the desired alkene 35. The choice of the diol protective
group has not yet been optimized. Attempts to use a carboxylic
acid or its derivatives as electrophile to directly obtain ketone
34 remained unsuccessful. The photocycloaddition of silyl enol
ether 35 proceeded smoothly (regioselectivity >95/5) and
furnished diastereomerically pure oxetane 36 which was isolated
in 70% yield. The facial diastereoselectivity (dr ) 90/10) is
good, which again reveals that the steric influence of the CH2OR
substituent is comparable to the one of a methyl group (cf. 21)
at the stereogenic center.
Further Variations in the r- and γ-Positions of the Silyl
Enol Ether. The R-substituent R on the silyl enol ether has
been held constant in all previous experiments in order to allow
an undisturbed evaluation of the facial diastereoselection. A
variation of this group is possible, of course, and we have studied
two cases in more detail. A protected ketone and a carboxylic
acid moiety were introduced into the R-position. Starting from
the known ketone 37,41 an aldol reaction lead to alcohol 38
(Scheme 10). Methylation33 furnished the corresponding methyl
ether 39 which was subsequently converted to the silyl enol
ether 40 according to the standard protocol. The photocyclo-
addition proceeded smoothly (regioselectivity >95/5), and the
oxetane 41 was isolated in diastereomerically pure form after
chromatography. The diastereomeric ratio (dr ) 79/21) in the
crude product mixture was slightly lower than in the comparable
tert-butyl case (cf. oxetane 29).
Oxetanes such as 29a contain four contiguous stereogenic
centers which can be employed for the construction of interesting
cyclic and acyclic molecules. The relative configuration
established at C-3, C-4, and the exocyclic stereogenic center
for instance correlates to the stereochemical arrangement found
in branched sugars.38 Simple hydrogenolysis of the 2-aryl-
oxetanes yields open chain products which may be manipulated
(35) Only some selected references are provided. (a) Diels-Alder
reaction: McDougal, P. G.; Rico, J. G.; VanDerveer, D. J. Org. Chem.
1986, 51, 4494-4497. Kozikowski, A. P.; Jung, S. H.; Springer, J. P. J.
Chem. Soc., Chem. Commun. 1988, 167-169. See also ref 23. (b)
Dihydroxylation: Cha, J. K.; Christ, W. J.; Kishi, Y. Tetrahedron 1984,
40, 2247-2255. (c) 1,3-Dipolar cycloaddition: Houk, K. N., Moses, S.
R.; Wu, Y.-D.; Rondan, N. G., Ja¨ger, V., Schohe, R.; Fronczek, F. R. J.
Am. Chem. Soc. 1984, 106, 3880-3882. Kozikowski, A. P.; Ghosh, A. K.
J. Org. Chem. 1984, 49, 2762-2772. (d) Mukaiyama aldol reaction: See
ref 32.
For the installation of a protected carboxylic acid in the
R-position the synthesis commenced with the nitrile 42.42
(36) (a) Tripathy, R.; Franck, R. W.; Onan, K. D. J. Am. Chem. Soc.
1988, 110, 3257-3262. (b) Adam, W.; Gla¨ser, J.; Peters, K.; Prein, M.
Ibid. 1995, 117, 9189-9193
(39) Saito, S.; Hasegawa, T.; Inabe, M.; Nishida, R. Chem. Lett. 1984,
1389-1392.
(40) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639-666.
(37) Kahn, S. D.; Pau, C. F.; Chamberlin, A. R.; Hehre, W. J. J. Am.
Chem. Soc. 1987, 109, 650-663.
(41) Bo¨eseken, J.; Tellegen, F. Recl. TraV. Chim. Pays-Bas 1938, 57,
133-143.
(42) Stetter, H.; Hunds, A. Liebigs Ann. Chem. 1984, 1577-1590.
(38) Yoshimura, J. AdV. Carbohydr. Chem. Biochem. 1984, 42, 69-
134.

2442 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Bach et al.
Scheme 10a
Scheme 12a
a Reagents and conditions: (a) -78 °C (THF); (b) 1,8-bis(dimethyl-
amino)naphthalene, 25 °C (CH2Cl2); (c) -78 f +25 °C (THF); (d)
30 °C (PhH).
a Reagents and conditions: (a) (PhMe2Si)2CuLi, -23 f 0 °C,
TMSCl, NEt3, 0 f 25 °C (THF); (b) 30 °C (PhH); (c) 25 °C (THF).
Scheme 11a
addition of (PhMe2Si)2CuLi43 to the R,â-unsaturated ketones
4631 and 4744 and in situ quench of the intermediate enolates
(Scheme 12). As expected from the steric requirements of the
silyl group, the facial diastereoselectivity in the subsequent
Paterno`-Bu¨chi reaction was excellent. By this means the
diastereomerically pure oxetanes 50 (dr ) 95/5, regioselectivity
70/30) and 51 (dr ) 83/17, regioselectivity 80/20) can be readily
generated. Treatment with tetrabutylammonium fluoride (TBAF)
promoted a smooth ring opening to the desired diols 52 and 53
which are valuable intermediates for further functionalization.
The (E)-configuration of the double bond suggests an anti-
periplanar stereospecific E2 type elimination. The two-step
sequence photocycloaddition/fragmentation allows the formal
allyl transfer of a bulky allylic R-alkoxy anion to a carbonyl
group with good stereocontrol. It is complementary to known
allyl transfer reactions which enable stereoselective access to
less crowded 1,2-diols.45 Since the photocycloaddition proceeds
racemization-free, enantiomerically pure material should be
readily accessible from enantiomerically pure chiral silyl enol
ethers.46
a Reagents and conditions: (a) -25 f -15 °C (THF); (b) t-Bu2CuLi,
-78 f 0 °C, TMSCl, NEt3, 0 f 25 °C (THF); (c) 30 °C (PhH).
Addition of 1-propenylmagnesium bromide gave the R,â-
unsaturated ketone 43 which was transformed to the silyl enol
ether 44 by the one-pot sequence conjugate addition/silylation
(Scheme 11). The oxetane formation was initiated by irradiation
of the olefin 44 in the presence of benzaldehyde. Only a single
diastereoisomer 45 was detected (dr ) >95/5) which was
accompanied by regioisomeric impurities (regioselectivity 85/
15). After hydrolysis of the excess silyl enol ether, the pure
oxetane 45 was obtained in 52% yield.
Summary
Despite the success encountered in terms of facial diastereo-
selection, it must be noted that the regioselectivity highly
depends on the bulkiness of the R-substituent. Attempted
photocycloaddtion reactions with an i-Pr-substituted silyl enol
ether exhibited a strong deterioration in this type of selectivity
(regioselectivity 70/30). As can be seen from the above-
mentioned examples, there are nonetheless several protected
functional groups which can be introduced and used success-
fully. In an earlier study we have already shown that other
aromatic aldehydes may be employed in the photocycloaddition
to silyl enol ethers.6d Aliphatic aldehydes, however, appear
unsuited for this particular reaction. Attempts to react acet-
aldehyde and related carbonyl compounds in a fashion analogous
to that described for benzaldehyde remained unsuccessful and
gave only low yields of the corresponding oxetane (<5%). It
can be assumed that the singlet state is not quenched efficiently
because the concentration of the alkene is comparably low. In
the triplet manifold the well-documented reactions of aliphatic
aldehydes2d (Norrish type cleavage, hydrogen abstraction) take
over and inhibit oxetane formation.
In summary, our studies have revealed that chiral silyl enol
ethers of the general formula III (Scheme 2) are suitable alkene
components in stereoselective Paterno`-Bu¨chi reactions. The
yields of the major product oxetanes range between 44% and
76%. In order to ensure good facial diastereoselectivity (dr )
g85/15), the substituent RL in the γ-position should be bulky
(RL ) SiMe2Ph, t-Bu) or polar (RL ) OR). In a single step
three new stereogenic centers are established in a predictable
manner. The formation of 1 out of 16 possible isomeric
oxetanes is strongly favored. The reaction proceeds racemiza-
tion-free, and as a consequence, enantiomerically pure oxetanes
are equally well accessible. Ring-opening reactions which occur
selectively at the bond either between O and C-2 or between O
and C-4 have been developed and applied to selected oxetanes.
With these methods the stereochemical information generated
in the photochemical step can be transformed to acyclic target
molecules.
(43) (a) Ager, D. J.; Fleming, I.; Patel, S. K. J. Chem. Soc., Perkin Trans.
1 1981, 2520-2526. (b) Fleming, I.; Newton, T. W. Ibid. 1984, 1805-
With respect to a projected ring opening between O and C-4
in the photocycloaddition product, we prepared a couple of silyl
enol ethers which bear a silicon substituent in the γ-position.
The idea was to facilitate a fragmentation of the product oxetane
in order to generate diastereomerically pure 3-butene-1,2-diols.
The corresponding starting materials were prepared by conjugate
1809.
(44) 47 was prepared in full analogy to 46 from the corresponding
hydroxy ketone 38.
(45) Hoffmann, R. W.; Metternich, R. Liebigs Ann. Chem. 1985, 2390-
2402.
(46) Matsumoto, Y.; Hayashi, T., Ito, Y. Tetrahedron 1994, 50, 335-
346.

Paterno`-Bu¨chi Reaction of Chiral Silyl Enol Ethers
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2443
silyl enol ether 6 (855 mg). The crude product (dr ) 70/30;
regioselectivity 82/18) was purified by flash chromatography (CH/EA
) 99/1). Total yield: 195 mg (57%). Major isomer (2RS,3RS,4SR,1′RS)-
10a. Rf ) 0.32 (95/5). 1H NMR (CDCl3): -0.30 (s, 9 H), 0.77 (d, 3J
) 6.8 Hz, 3 H), 0.95 (d, 3J ) 7.2 Hz, 3 H), 0.97 (d, 3J ) 7.2 Hz, 3 H),
1.09 (s, 9 H), 1.68 (dsept, 3J ) 7.2 Hz, 3J ) 1.9 Hz, 1 H,), 2.14 (ddq,
Experimental Section
General Procedures. See ref 6d.
2,2,5-Trimethyl-3-[(trimethylsilyl)oxy]-3-heptene (5). Typical
Procedure A. To a stirred suspension of 2 mmol (179 mg) of CuCN
in 5 mL of THF was added dropwise at 0 °C a solution of 3 mmol of
MeLi (2 mL of a 1.5 M solution) in ether. After 30 min the mixture
was cooled to -78 °C and a solution of 1 mmol of R,â-unsaturated
ketone 419 (140 mg) in 5 mL of THF was added dropwise. The reaction
mixture was stirred for 30 min, warmed to 0 °C, and stirred for an
additional 30 min. At this temperature 1 mmol of triethylamine (101
mg, 139 µL) and 2 mmol of TMSCl (217 mg, 254 µL) were added
successively by syringe. After stirring for another 5 min, the mixture
was warmed to ambient temperature. It was quenched with an aqueous
saturated NH4Cl solution, and the product was extracted into ether.
The combined organic layers were washed with brine and dried over
MgSO4. After filtration the solvents were removed in Vacuo. The
residue proved to be sufficiently pure (GLC) for further transformations.
Yield: 206 mg (90%). Rf ) 0.72 (90/10). 1H NMR (CDCl3): 0.22
3
3
3
3J ) 10.9 Hz, J ) 6.8 Hz, J ) 1.9 Hz, 1 H), 4.45 (d, J ) 10.9 Hz,
1 H), 5.62 (s, 1 H), 7.25-7.40 (m, 5 H). Anal. Calcd for C21H36O2Si
(348.6): C, 72.36; H, 10.41. Found: C, 72.10; H, 10.16. For further
analytical data see the Supporting Information. Minor isomer
(2SR,3SR,4RS,1′RS)-10b. Rf ) 0.36 (95/5). 1H NMR (CDCl3): -0.30
3
3
(s, 9 H), 0.64 (d, J ) 6.9 Hz, 3 H), 0.83 (d, J ) 6.9 Hz, 3 H), 1.03
(d, 3J ) 7.2 Hz, 3 H), 1.08 (s, 9 H), 2.09 (ddq, 3J ) 10.7 Hz, 3J ) 6.9
Hz, J ) 2.9 Hz, 1 H), 2.39 (dsept, J ) 6.9 Hz, J ) 2.9 Hz, 1 H),
4.40 (d, J ) 10.7 Hz, 1 H), 5.63 (s, 1 H), 7.25-7.38 (m, 5 H).
3
3
3
3
3-(1,1-Dimethylethyl)-2-phenyl-4-(1-phenylethyl)-3-[(trimethyl-
silyl)oxy]oxetane (11). As described in typical procedure C the oxetane
formation was carried out on a 1 mmol-scale starting with silyl enol
ether 747 (967 mg). The crude product (dr ) 71/29; regioselectivity
80/20) was purified by flash chromatography (CH/EA ) 99/1). Total
yield: 289 mg (76%). Major isomer (2RS,3RS,4SR,1′RS)-11a. Rf )
0.25 (95/5). 1H NMR (CDCl3): -0.27 (s, 9 H,), 0.71 (s, 9 H), 1.45
(d, 3J ) 6.9 Hz, 3 H), 3.50 (dq, 3J ) 10.2 Hz, 3J ) 6.9 Hz, 1 H), 4.77
3
3
(s, 9 H), 0.85 (t, J ) 7.4 Hz, 3 H), 0.91 (d, J ) 6.7 Hz, 3 H), 1.04
3
(s, 9 H), 1.15-1.35 (m, 2 H), 2.20-2.35 (m, 1 H), 4.29 (d, J ) 9.3
Hz, 1 H). 13C NMR (CDCl3): 1.0 (q) 11.7 (q), 20.6 (q), 28.7 (q), 30.7
(t), 32,7 (d), 36.1 (s), 110.5 (d), 157.0 (s). Anal. Calcd for C13H28OSi
(228.4): C, 68.35; H, 12.35. Found: C, 68.20; H, 12.44.
3
(d, J ) 10.2 Hz, 1 H), 5.65 (s, 1 H), 7.13-7.48 (m, 5 H). Anal.
2,2,5,6-Tetramethyl-3-[(trimethylsilyl)oxy]-3-heptene (6). Typical
Procedure B. To a stirred solution of 10 mmol of LDA prepared from
10.5 mmol of diisopropylamine (1.06 g, 1.47 mL) and 10 mmol of
n-BuLi (4 mL of a 2.5 M solution in n-hexane) at 0 °C in 20 mL of
THF was slowly added at -78 °C ketone 116a (10 mmol, 1.70 g). After
20 min 11 mmol of TMSCl was added (1.20 g, 1.40 mL) within 10
min. After another 30 min, the mixture was warmed to room
temperature and the solvent was removed in Vacuo. The residue was
filtered, and the solid in the filter was washed thoroughly with pentane.
The solvent was removed, and the procedure (filtration, washing, solvent
removal) was repeated until a clear solution resulted. Purification by
flash chromatography (CH/EA ) 99/1) yielded 1.29 g (53%) of the
desired product. Rf ) 0.70 (95/5). 1H NMR (CDCl3): 0.22 (s, 9 H),
Calcd for C24H34O2Si (382.6): C, 75.34; H, 8.96. Found: C, 75.45;
H, 8.85. For further analytical data see the Supporting Information.
Enantiomerically pure oxetane (+)-11a was prepared from silyl enol
ether (+)-7 in the same fashion. [R]20 ) +76.7 (c ) 1.2, acetone).
D
Minor isomer (2SR,3SR,4RS,1′RS)-11b. Rf ) 0.17 (95/5). 1H NMR
3
(CDCl3): -0.21 (s, 9H), 1.14 (s, 9 H), 1.15 (d, J ) 6.7 Hz, 3 H),
3.37 (dq, 3J ) 10.6 Hz, 3J ) 6.7 Hz, 1 H), 4.74 (d, 3J ) 10.6 Hz, 1 H),
5.61 (s, 1 H), 7.13-7.48 (m, 5 H). Anal. Calcd for C24H34O2Si
(382.6): C, 75.34; H, 8.96. Found: C, 75.38; H, 8.95. For further
analytical data see the Supporting Information. Enantiomerically pure
oxetane (+)-11b was prepared from silyl enol ether (+)-7 in the same
fashion. [R]20
) +155.7 (c ) 1.0, acetone). The temperature
D
dependent measurements of the diastereomeric ratio 11a/11b were
3
3
3
0.84 (d, J ) 6.7 Hz, 3 H), 0.85 (d, J ) 6.7 Hz, 3 H), 0.88 (d, J )
carried out in an immersion apparatus as previously described.6b
6.7 Hz, 3 H), 1.04 (s, 9 H), 1.44 (virt oct, 3J ≈ 6.7 Hz, 1 H), 2.16 (virt
X-ray crystal structure analysis of 11a: formula C24H34O2Si, M )
382.60, 0.7 × 0.4 × 0.35 mm, a ) 15.977(3) Å, b ) 17.205(3)Å, c )
3
3
3
dquint, J ) 9.8 Hz, J ≈ 6.7 Hz, 1 H), 4.35 (d, J ) 9.8 Hz, 1H).
Anal. Calcd for C14H30OSi (242.5): C, 69.35; H, 12.47. Found: C,
69.46; H, 12.29. For further analytical data see the Supporting
Information.
8.575(2) Å, â ) 101.06(3)°, V ) 2313.4(8) Å3, Fcalc ) 1.099 g cm-3
,
µ ) 9.96 cm-1, no absorption correction, Z ) 4, monoclinic, space
group P21/c (no. 14), λ ) 1.541 78 Å, T ) 293 K, ω/2θ scans, 5205
reflections collected ((h, -k, +l), [(sin θ)/λ]max ) 0.62 Å-1, 4867
independent and 4021 observed reflections [I e 2σ(I)], 252 refined
parameters, R ) 0.046, wR2 ) 0.127, maximum residual electron
density 0.21 (-0.29) e Å-3, hydrogens calculated and riding.
3-(1,1-Dimethylethyl)-4-(1-methylpropyl)-2-phenyl-3-[(trimeth-
ylsilyl)oxy]oxetane (9) Typical Procedure C. A quartz tube was
charged with 3.5 mmol of silyl enol ether 5 (799 mg) dissolved in 6
mL of benzene. The sample was irradiated at 300 nm (RPR 3000 Å)
in a merry-go-round unit. Benzaldehyde (1 mmol, 106 mg, 102 µL)
was added slowly as a solution in 1 mL of benzene via syringe (within
2 h). Upon complete addition the reaction was monitored by TLC
and GLC. After the aldehyde was fully consumed the irradiation was
stopped. The solvent was evaporated in Vacuo, and the residue was
3-(1,1-Dimethylethyl)-2-phenyl-4-(1,2,2-trimethylpropyl)-3-[(tri-
methylsilyl)oxy]oxetane (12a). As described in typical procedure C
the oxetane formation was carried out on a 1 mmol scale starting with
silyl enol ether 847 (900 mg). The crude product (dr ) >95/5;
regioselectivity 75/25) was purified by flash chromatography (CH/EA
) 99/1). Yield: 248 mg (58%). Rf ) 0.25 (95/5). 1H NMR
(CDCl3): -0.22 (s, 9 H), 0.94 (s, 9 H), 1.10 (s, 9 H), 1.17 (d, 3J ) 7.2
1
analyzed by H NMR spectroscopy and by GLC-MS to determine
the diastereo- and regioselectivity (dr ) 60/40; regioselectivity 87/13).
Silyl enol ether was recovered from the crude product mixture by
distillation in a Kugelrohr apparatus. Further purification was carried
out by flash chromatography (CH/EA ) 99/1). The diastereoisomers
were not fully separable. Total yield: 222 mg (61%). Rf ) 0.44 (90/
10). Anal. Calcd for C20H34O2Si (334.6): C, 71.80; H 10.24.
Found: C, 71.66; H, 10.13. Major isomer (2RS,3RS,4SR,1′RS)-9a. 1H
NMR (CDCl3): -0.31 (s, 9 H), 0.76 (d, 3J ) 6.4 Hz, 3 H), 0.92 (t, 3J
) 7.1 Hz, 3 H), 1.08 (s, 9 H), 1.30-2.10 (m, 3 H), 4.21 (d, 3J ) 10.5
Hz, 1 H), 5.62 (s, 1 H), 7.20-7.50 (m, 5 H). 13C NMR (CDCl3): 2.3
(q), 10.7 (q), 13.1 (q), 24.4 (q), 24.5 (t), 28.8 (s), 36.2 (d), 86.1 (d),
87.5 (s), 89.3 (d), 127.6 (d), 128.0 (d), 128.2 (d), 139.9 (s). Minor
isomer (2SR,3SR,4RS,1′RS)-9b. 1H NMR (CDCl3): -0.30 (s, 9 H),
3
3
3
Hz, 3 H), 1.50 (dq, J ) 7.2 Hz, J ) 0.8 Hz, 1 H), 5.18 (d, J ) 0.8
Hz, 1 H), 5.79 (s, 1 H), 7.21-7.44 (m, 5 H). Anal. Calcd for C22H38O2-
Si (362.6): C, 72.87; H, 10.56. Found: C, 72.85; H, 10.28. For further
analytical data see the Supporting Information.
2,2-Dimethyl-5-methoxy-3-hexanone (25). Typical procedure D.
To a solution of 70 mmol of hydroxyketone 2331 (10.1 g) in 100 mL
of CH2Cl2 were added 100 mmol of trimethyloxonium tetrafluoroborate
(14.8 g) and 120 mmol of 1,8-bis(dimethylamino)naphthalene (25.7
g). The suspension was stirred at room temperature for 20 d and was
subsequently quenched with ice-cold water. The product was thor-
oughly extracted into ether (6 × 50 mL). The combined organic layers
were washed with an aqueous saturated NaHCO3 solution and with
brine and dried over MgSO4. After filtration the solvents were removed
in Vacuo. The residue was filtered through a plug of silica (ether as
eluent) and subsequently distilled in Vacuo. Yield: 6.42 g (58%). Bp:
100-110 °C (20 mbar). 1H NMR (CDCl3): 1.08 (s, 9 H), 1.13 (d, 3J
3
3
0.77 (d, J ) 6.6 Hz, 3 H), 0.94 (t, J ) 7.4 Hz, 3 H), 1.07 (s, 9 H),
1.30-2.10 (m, 3 H), 4.27 (d, 3J ) 10.5 Hz, 1 H), 5.63 (s), 7.20-7.50
(m, 5 H).
3-(1,1-Dimethylethyl)-4-(1,2-dimethylpropyl)-2-phenyl-3-[(tri-
methylsilyl)oxy]oxetane (10). As described in typical procedure C
the oxetane formation was carried out on a 1 mmol scale starting with
(47) For preparation see the Supporting Information.

2444 J. Am. Chem. Soc., Vol. 119, No. 10, 1997
Bach et al.
2
3
) 6.2 Hz, 3 H), 2.43 (dd, J ) 16.9 Hz, J ) 6.2 Hz, 1 H), 2.84 (dd,
2J ) 16.9 Hz, 3J ) 6.2 Hz, 1 H), 3.33 (s, 3 H), 3.83 (virt sex, 3J ≈ 6.2
Hz, 1 H). Anal. Calcd for C9H18O2 (158.2): C, 68.31; H, 11.47.
Found: C, 68.49; H, 11.46. For further analytical data see the
Supporting Information.
starting with silyl enol ether 3547 (205 mg). The crude product (dr )
90/10; regioselectivity >95/5) was purified by flash chromatography
(CH/EA ) 99/1). Yield: 100 mg (70%). [R]20 ) +32.2 (c ) 0.8,
D
CH2Cl2). Rf ) 0.36 (90/10). 1H NMR (CDCl3): -0.24 (s, 9 H), 1.10
2
3
(s, 9 H), 1.33 (s, 3 H), 1.40 (s, 3 H), 4.05 (dd, J ) 8.4 Hz, J ) 5.0
Hz, 1 H), 4.27 (dd, 2J ) 8.4 Hz, 3J ) 6.3 Hz, 1 H), 4.37 (ddd, 3J ) 8.8
2,2-Dimethyl-5-methoxy-3-[(trimethylsilyl)oxy]-3-hexene (27). As
described in typical procedure B the silyl enol ether formation was
carried out on a 35 mmol scale starting with ketone 25 (5.54 g). The
crude product proved to be sufficiently pure (GLC) for further
transformations. Yield: 7.70 g (95%). Bp: 150 °C (2 mbar). Rf )
0.53 (90/10). 1H NMR (CDCl3): 0.24 (s, 9 H), 1.06 (s, 9 H), 1.19 (d,
3
3
3
Hz, J ) 6.3 Hz, J ) 5.0 Hz, 1 H), 4.53 (d, J ) 8.8 Hz, 1 H), 5.64
(s, 1 H), 7.23-7.39 (m, 5 H). 13C NMR (CDCl3): 2.6 (q), 25.4 (q),
25.7 (q), 27.2 (q), 36.6 (s), 67.1 (t), 74.6 (d), 86.7 (d), 86.9 (d), 87.3
(s), 110.2 (s), 127.3 (d), 128.0 (d), 128.8 (d), 140.8 (s). Anal. Calcd
for C21H34O4Si (378.6): C, 66.63; H, 9.05. Found: C, 66.94; H, 9.18.
3-(2-Methyl-1,3-dioxolan-2-yl)-4-(1-methoxyethyl)-2-phenyl-3-
[(trimethylsilyl)oxy]oxetane (41). As described in typical procedure
C the oxetane formation was carried out on a 0.86 mmol scale starting
with silyl enol ether 4047 (785 mg). The crude product (dr ) 79/21;
regioselectivity >95/5) was purified by flash chromatography (CH/
3J ) 6.5 Hz, 3 H), 3.22 (s, 3 H), 4.07 (dq, J ) 9.0 Hz, J ) 6.5 Hz,
1 H), 4.50 (d, 3J ) 9.0 Hz, 1 H). Anal. Calcd for C12H26O2Si
(230.4): C, 62.55; H, 11.37. Found: C, 62.44; H, 11.26. For further
analytical data see the Supporting Information.
3
3
3-(1,1-Dimethylethyl)-4-(1-methoxyethyl)-2-phenyl-3-[(trimeth-
ylsilyl)oxy]oxetane (29). As described in typical procedure C the
oxetane formation was carried out on a 1 mmol-scale starting with silyl
enol ether 27 (806 mg). The crude product (dr ) 85/15; regioselectivity
91/9) was purified by flash chromatography (CH/EA ) 99/1). Total
yield: 209 mg (62%). Major isomer (2RS,3RS,4SR,1′SR)-29a. Mp:
96 °C. Rf ) 0.21 (95/5). 1H NMR (CDCl3): -0.26 (s, 9 H), 1.08 (s,
9 H), 1.31 (d, 3J ) 6.2 Hz, 3 H), 3.33 (s, 3 H), 3.80 (dq, 3J ) 8.9 Hz,
3J ) 6.2 Hz, 1 H), 4.37 (d, 3J ) 8.9 Hz, 1 H), 5.64 (s, 1 H), 7.26-7.36
(m, 5 H). 13C NMR (CDCl3): 2.5 (q), 14.0 (q), 25.4 (q), 36.0 (s),
55.5 (q), 74.5 (d) 85.9 (d), 86.5 (d), 88.1 (s), 127.6 (d), 127.7 (d), 128.1
(d), 139.4 (s). Anal. Calcd for C19H32O3Si (336.5): C, 67.81; H, 9.58.
Found: C, 67.85; H, 9.33. Minor isomer (2SR,3SR,4RS,1′SR)-29b. Rf
) 0.11 (90/10). 1H NMR (CDCl3): -0.30 (s, 9 H), 1.07 (d, 3J ) 6.2
Hz, 3 H), 1.07 (s, 9 H), 3.59 (s, 3 H), 3.81 (dq, 3J ) 9.3 Hz, 3J ) 6.2
EA
) 95/5). Total yield: 235 mg (75%). Major isomer
(2RS,3RS,4SR,1′SR)-41. Mp: 80 °C. Rf ) 0.20 (90/10). 1H NMR
(CDCl3): -0.29 (s, 9 H), 1.27 (d, 3J ) 6.4 Hz), 1.32 (s, 3 H), 3.36 (s,
3 H), 3.86 (dq, J ) 8.8 Hz, J ) 6.4 Hz, 1 H), 4.00-4.25 (m, 4H),
4.43 (d, J ) 8.8 Hz, 1 H), 5.68 (s, 1 H), 7.28-7.40 (m, 5 H). NOE
3
3
3
experiment (CDCl3): H (4.43), H(5.68) [0.51%]; H (1.32), H(4.43) [1.7%].
13C NMR (CDCl3): 1.8 (q), 14.3 (q), 19.6 (q), 56.1 (q), 65.1 (t), 74.3
(d), 84.6 (d), 85.2 (s), 86.3 (d), 109.7 (s), 126.1 (d), 126.5 (d), 127.0
(d), 136.4 (s). Anal. Calcd for C19H30O5Si (366.5): C, 62.26; H, 8.25.
Found: C, 62.56; H, 8.22. Minor isomer (2SR,3SR,4RS,1′SR)-41. Rf
) 0.05 (90/10). 1H NMR (CDCl3): -0.31 (s, 9 H), 1.11 (d, 3J ) 6.3
3
3
Hz), 1.27 (s, 3 H), 3.56 (s, 3 H), 3.86 (dq, J ) 9.0 Hz, J ) 6.3 Hz,
3
1 H), 4.01-4.19 (m, 4H), 4.58 (d, J ) 9.0 Hz, 1 H), 5.73 (s, 1 H),
7.27-7.45 (m, 5 H). NOE experiment (CDCl3): H (5.73), H(4.58)
[0.42%]; H (1.27), H(5.73) [0.56%], H(4.58) [1.15%]. For further analytical
data see the Supporting Information.
3
Hz, 1 H), 4.48 (d, J ) 9.3 Hz, 1 H), 5.71 (s, 1 H), 7.25-7.46 (m, 5
H). Anal. Calcd. for C19H32O3Si (336.5): C, 67.81; H, 9.58.
Found: C, 67.97; H, 9.43. For further analytical data see the Supporting
Information.
1-(2,4,10-Trioxaadamant-3-yl)-2-buten-1-one (43). To a stirred
solution of 1.0 mmol of nitrile 4242 (167 mg) in 10 mL of THF was
slowly added at -78 °C (within 25 min) 1 mmol of MeCHCHMgBr
(0.6 mL of a 1.7 M solution in THF). The solution was warmed to
-10 °C and stirred at this temperature for another 1.5 h. The mixture
was quenched with a cold aqueous HCl solution (1 N), and the product
was extracted into CH2Cl2. The combined organic layers were dried
over MgSO4. After filtration the solvents were removed in Vacuo. The
residue was purified by flash chromatography (CH/EA ) 60/40).
Yield: 149 mg (71%). Mp: 80 °C. Rf ) 0.26 (40/60). 1H NMR
(CDCl3): 1.79 (d, 2J ) 12.7 Hz, 3 H), 1.93 (dd, 3J ) 7.0 Hz, 4J ) 1.7
2,2-Dimethyl-5-methoxy-3-(phenylmethyl)-3,4-hexanediol (31).
Oxetane 29a (2.5 mmol, 840 mg) was dissolved in 20 mL of methanol,
and 0.5 mmol of the catalyst Pd/C [10% w/w] (265 mg) was added to
the solution. The hydrogenolysis was carried out in a conventional
hydrogenation apparatus at ambient temperature and atmospheric
pressure. The progression of the reaction was indicated by the volume
of consumed hydrogen and was further monitored by TLC. Upon
complete transformation (72 h) the mixture was filtered and the solvent
removed in Vacuo. The residue was purified by chromatography (FC,
CH/EA ) 95/5). Yield: 565 mg (85%). Mp: 112 °C. Rf ) 0.45
(75/25). 1H NMR (CDCl3): 0.99 (d, 3J ) 5.9 Hz, 3 H), 1.08 (s, 9 H),
1.67 (d, J ) 6.2 Hz, 1 H), 2.80 (s, 3 H), 2.94 (dq, J ) 9.0 Hz, J )
5.9 Hz, 1 H), 2.95 (d, 2J ) 13.8 Hz, 1 H), 3.03 (d, 2J ) 13.8 Hz, 1 H),
3.60 (dd, 3J ) 9.0 Hz, 3J ) 6.2 Hz, 1 H), 4.48 (s, 1 H), 7.17-7.47 (m,
5 H). 13C NMR (CDCl3): 15.6 (q), 26.2 (q), 36.5 (t), 39.5 (s), 54.7
(q), 75.2 (d), 78.4 (d), 79.8 (s), 125.9 (d), 127.6 (d), 131.5 (d), 139.7
(s). Anal. Calcd for C16H25O3 (266.4): C, 72.14; H, 9.46. Found:
C, 71.94, H 9.62.
2
3
Hz, 3 H), 2.75 (d, J ) 12.7 Hz, 3 H), 4.54 (s, 3 H), 6.61 (dq, J )
15.6 Hz, 4J ) 1.7 Hz, 1 H), 7.18 (dq, 3J ) 15.6 Hz, 3J ) 7.0 Hz, 1 H).
Anal. Calcd for C11H14O4 (210.2): C, 62.85; H, 6.71. Found: C,
62.58; H, 6.73. For further analytical data see the Supporting
Information.
3
3
3
1-(2,4,10-Trioxaadamant-3-yl)-3,4,4-trimethyl-1-[(trimethylsilyl)-
oxy]-1-pentene (44). As described in typical procedure A the silyl
enol ether formation was carried out on a 3 mmol scale starting with
ketone 43 (631 mg) and employing 9 mmol of t-BuLi (6 mL of a 1.5
M solution in pentane). The crude product was purified by flash
chromatography (CH/EA ) 75/25). Yield: 864 mg (85%). Mp: 57
°C. Rf ) 0.66 (40/60). 1H NMR (CDCl3): 0.21 (s, 9 H), 0.85 (s, 9
H), 0.90 (d, 3J ) 6.9 Hz, 3 H), 1.68 (d, 2J ) 12.5 Hz, 3 H), 2.25 (dq,
3J ) 10.2 Hz, 3J ) 6.9 Hz, 1 H), 2.64 (d, 2J ) 12.5 Hz, 3 H), 4.43 (s,
3 H), 5.12 (d, 3J ) 10.2 Hz, 1 H). Anal. Calcd for C18H32O4Si
(339.5): C, 63.68; H, 9.50. Found: C, 63.54; H, 9.57. For further
analytical data see the Supporting Information.
2-Phenyl-4-(1,2,2-trimethylpropyl)-3-(2,4,10-trioxaadamant-3-yl)-
3-[(trimethylsilyl)oxy]oxetane (45). As described in typical procedure
C the oxetane formation was carried out on a 0.65 mmol scale starting
with silyl enol ether 44 (766 mg). The crude product (dr ) >95/5;
regioselectivity 85/15) was dissolved in 80 mL of ether, and upon
addition of 50 mL of 2 N HCl, the mixture was stirred for 21 d. After
neutralization with an aqueous saturated NaHCO3 solution, the product
was extracted into ether. The combined organic layers were washed
with brine and dried over MgSO4. After filtration the solvents were
removed in Vacuo. The residue was purified by flash chromatography
(CH/EA ) 90/10). Yield: 151 mg (52%). Mp: 141 °C. Rf ) 0.43
(75/25). 1H NMR (CDCl3): 0.05 (s, 9 H), 1.20 (s, 9 H), 1.40 (d, 3J )
1,2-Dihydroxy-5,5-dimethyl-1,2-O-isopropylidene-4-hexanone (34).
To a solution of 5.9 mmol of alcohol 3347 (1.19 g) in 50 mL of CH2Cl2
were added 12 mmol of N-methylmorpholine N-oxide (1.40 g) and 4
g of powdered molecular sieves 4 Å. The suspension was stirred at
room temperature for 10 min. Then 0.6 mmol of tetrapropylammonium
perruthenate (211 mg) was added, and stirring was continued for 6 h.
The crude product mixture was filtered through a plug of silica (ethyl
acetate as eluent), and the solvents were removed in Vacuo. The residue
proved to be sufficiently pure (GLC) for further transformations.
Yield: 967 mg (82%). [R]20 ) +7.9 (c ) 0.3, CH2Cl2). Rf ) 0.18
D
(90/10). 1H NMR (CDCl3): 1.16 (s, 9 H), 1.30 (s, 3 H), 1.36 (s, 3 H),
2.67 (dd, 2J ) 17.4 Hz, 3J ) 7.3 Hz, 1 H), 3.12 (dd, J ) 17.4 Hz, 3J
2
) 5.7 Hz, 1 H), 3.51 (dd, 2J ) 8.1 Hz, 3J ) 6.9 Hz, 1 H), 4.15 (dd, 2J
3
3
3
) 8.1 Hz, J ) 5.9 Hz, 1 H), 4.42 (virt tt, J ≈ 7.3 Hz, J ≈ 5.8 Hz,
1 H). Anal. Calcd for C11H20O3 (200.3): C, 65.96; H, 10.07. Found:
C, 65.55; H, 10.67. For further analytical data see the Supporting
Information.
4-(2,2-Dimethyl-1,3-dioxolan-4-yl)-3-(2,2-dimethylethyl)-2-phen-
yl-3-[(trimethylsilyl)oxy]oxetane (36). As described in typical pro-
cedure C the oxetane formation was carried out on a 0.38 mmol scale

Paterno`-Bu¨chi Reaction of Chiral Silyl Enol Ethers
J. Am. Chem. Soc., Vol. 119, No. 10, 1997 2445
3
3
2
7.4 Hz, 3 H), 2.14 (dq, J ) 7.4 Hz, J ) 3.1 Hz, 1 H), 2.02 (d, J )
12.5 Hz, 3 H), 2.94 (d, 3 H, 2J ) 12.5 Hz), 4.75 (s, 3 H), 5.57 (d, 3J )
3.1 Hz, 1 H), 6.22 (s, 1 H), 7.48-7.71 (m, 5 H). 13C NMR (CDCl3):
2.2 (q), 10.4 (q), 27.8 (q), 28.7 (s), 33.1 (t), 42.4 (d), 68.0 (d), 82.7
(d), 84.1 (d), 85.4 (s), 109.0 (s), 127.1 (d), 127.3 (d), 128.0 (d), 138.7
(s). Anal. Calcd for C25H38O5Si (446.7): C, 67.23; H, 8.57. Found:
C, 67.32; H, 8.45.
(CH/EA ) 98/2). The diastereoisomers were not fully separable. Total
yield: 18 mg (61%). Anal. Calcd for C26H38O4Si2 (470.8): C, 66.34;
H, 8.14. Found: C, 66.40; H, 7.95. Major isomer (2RS,3SR,4RS,1′SR)-
51. Rf ) 0.23 (90/10). 1H NMR (CDCl3): δ -0.27 (s, 9 H), 0.34 (s,
3 H), 0.37 (s, 3 H), 1.13 (s, 3 H), 1.29 (d, 3J ) 7.5 Hz, 3 H), 1.48 (dq,
3J ) 7.5 Hz, 3J ) 4.3 Hz, 1 H), 3.82-4.04 (m, 4 H), 4.89 (d, 3J ) 4.3
Hz, 1 H), 5.68 (s, 1 H), 7.25-7.37 (m, 6 H), 7.43 (dd, 3J ) 7.6 Hz, 4J
) 1.4 Hz, 2 H), 7.53-7.60 (m, 2 H). For further analytical data see
the Supporting Information. Minor isomer (2SR,3RS,4SR,1′SR)-51 Rf
) 0.26 (90/10). 1H NMR (CDCl3): δ -0.31 (s, 9 H), 0.40 (s, 3 H),
0.44 (s, 3 H), 0.75 (d, 3J ) 7.4 Hz, 3 H), 1.23 (s, 3 H), 1.89 (dq, 3J )
11.4 Hz, J ) 7.4 Hz, 1 H), 4.02 - 4.15 (m, 4 H), 4.63 (d, J ) 11.4
Hz, 1 H), 5.61 (s, 1 H), 7.21-7.29 (m, 4 H), 7.31-7.37 (m, 4 H),
7.57-7.61 (m, 2 H). For further analytical data see the Supporting
Information.
2,2-Dimethyl-3-(hydroxyphenylmethyl)-4-hexen-3-ol (52). Typi-
cal procedure F. To a stirred solution of 0.2 mmol of oxetane 50 (88
mg) in 1 mL of THF was added dropwise via syringe a solution of
0.44 mmol of TBAF (0.44 mL of a 1 M solution) in THF. After stirring
for 3 h the mixture was diluted with water and saturated with solid
NaCl. The product was extracted into CH2Cl2. The combined organic
layers were dried over MgSO4. After filtration the solvents were
removed in Vacuo. The residue was purified by flash chromatography
(CH/EA ) 98/2). Yield: 40 mg (85%). Mp: 87-88 °C. Rf ) 0.13
2,2-Dimethyl-5-(dimethylphenylsilyl)-3-[(trimethylsilyl)oxy]-3-
hexene (48). Typical Procedure E. To a stirred suspension of 18.75
mmol (1.68 g) of CuCN in 15 mL of THF was added dropwise at 0 °C
a solution of 37.5 mmol of PhMe2SiLi (62.5 mL of a 0.6 M solution)
in THF. After 30 min the mixture was cooled to -23 °C and a solution
of 15 mmol of R,â-unsaturated ketone 4631 (1.90 g) in 75 mL of THF
was added dropwise (within 30 min). The reaction mixture was stirred
for 1 h and warmed slowly to 0 °C. At this temperature 90 mmol of
triethylamine (9.11 mg, 12.5 mL) and 90 mmol of TMSCl (9.78 g,
11.4 mL) were added successively by syringe. The mixture was
warmed to ambient temperature and stirred for 18 h. It was diluted
with pentane and quenched with an aqueous saturated NaHCO3 solution.
After 10 min of rapid stirring an aqueous layer was formed on the
bottom of the flask from which the organic layer was decanted and
filtered through a plug of silica. From the aqueous layer the product
was extracted into pentane. The extracts were also filtered through a
plug of silica. The organic layers were combined, and the solvents
were removed in Vacuo. The crude product was purified by flash
chromatography (CH/EA/NEt3 ) 1000/3/1). Yield: 3.22 g (64%). Rf
) 0.63 (95/5). 1H NMR (CDCl3): δ 0.21 (s, 9 H), 0.24 (s, 3 H), 0.25
(s, 3 H), 0.93 (d, 3J ) 7.4 Hz, 3 H), 1.02 (s, 9 H), 1.93 (dq, 3J ) 11.0
3
3
3
4
(95/5). 1H NMR (CDCl3): δ 0.92 (s, 9 H), 1.66 (dd, J ) 6.5 Hz, J
) 1.7 Hz, 3 H), 4.82 (s, 1 H), 5.34 (dq, 3J ) 15.7 Hz, 3J ) 6.5 Hz, 1
3
4
H), 5.93 (dq, J ) 15.7 Hz, J ) 1.7 Hz, 1 H), 7.19-7.31 (m, 5 H).
13C NMR (CDCl3): δ 17.7 (q), 26.4 (q), 37.6 (s), 77.4 (d), 79.8 (s),
125.8 (d), 127.7 (d), 128.0 (d), 128.5 (d), 131.2 (d), 141.4 (s). Anal.
Calcd for C15H22O2 (234.3): C, 76.88; H, 9.46. Found: C, 76.93; H,
9.45.
3
3
Hz, J ) 7.4 Hz, 1 H), 4.27 (d, J ) 11.0 Hz, 1 H), 7.30-7.60 (m, 5
H). Anal. Calcd for C19H34OSi2 (334.6): C, 68.19; H, 10.24.
Found: C, 68.17; H, 10.37. For further analytical data see the
Supporting Information.
2-(2-Methyl-1,3-dioxolan-2-yl)-1-phenyl-3-penten-1,2-diol (53).
As described in typical procedure F the diol formation was carried out
on a 250 µmol scale starting with oxetane 51 (118 mg). The crude
product was purified by flash chromatography (CH/EA ) 85/15).
Yield: 51 mg (77%). Mp: 89-90 °C. Rf ) 0.18 (70/30). 1H NMR
3-(Dimethylphenylsilyl)-1-(2-methyl-1,3-dioxolan-2-yl)-1-[(tri-
methylsilyl)oxy]-1-butene (49). As described in typical procedure E
the silyl enol ether formation was carried out on a 2.7 mmol scale
starting with ketone 4744 (427 mg). The crude product was purified
by flash chromatography (CH/EA/NEt3 ) 1000/20/1). Yield: 850 mg
(86%). Rf ) 0.35 (95/5). 1H NMR (acetone-d6): δ 0.19 (s, 9 H), 0.25
3
4
(CDCl3): δ 1.43 (s, 3 H), 1.68 (dd, J ) 6.4 Hz, J ) 1.7 Hz, 3 H),
1.79 (s, 1 H), 3.79 (s, b, 1 H), 3.98-4.14 (m, 4 H), 5.00 (s, 1 H), 5.40
(dq, 3J ) 15.7 Hz, 3J ) 6.4 Hz, 1 H), 5.95 (dq, 3J ) 15.7 Hz, 4J ) 1.7
Hz, 1 H), 7.23-7.35 (m, 5 H). Anal. Calcd for C15H20O4 (264.3): C,
68.16; H, 7.63. Found: C, 68.18; H, 7.85. For further analytical data
see the Supporting Information.
3
(s, 3 H), 0.26 (s, 3 H), 0.97 (d, J ) 7.4 Hz, 3 H), 1.39 (s, 3 H), 2.05
(dq, 3J ) 11.1 Hz, 3J ) 7.4 Hz, 1 H), 3.78-3.91 (m, 4 H), 4.83 (d, 3J
) 11.1 Hz, 1 H), 7.32-7.36 (m, 3 H), 7.49-7.55 (m, 2 H). Anal.
Calcd for C19H32O3Si2 (364.6): C, 62.59; H, 8.85. Found: C, 62.51;
H, 8.83. For further analytical data see the Supporting Information.
3-(2,2-Dimethylethyl)-4-[1-(dimethylphenylsilyl)ethyl]-2-phenyl-
3-[(trimethylsilyl)oxy]oxetane (50). As described in typical procedure
C the oxetane formation was carried out on a 1.5 mmol scale starting
with silyl enol ether 48 (2.26 g). The crude product (dr ) 95/5;
regioselectivity 70/30) was purified by flash chromatography (CH/EA
) 99/1). Yield: 291 mg (44%). Rf ) 0.43 (95/5). 1H NMR
(CDCl3): δ -0.25 (s, 9 H), 0.36 (s, 3 H), 0.38 (s, 3 H), 0.97 (s, 9 H),
Acknowledgment. This work was generously supported by
the Deutsche Forschungsgemeinschaft (Grant Ba 1372/1-2), by
the Fonds der Chemischen Industrie, and by the Gesellschaft
zur Fo¨rderung der Westfa¨lischen Wilhelms-UniVersita¨t. We
thank Dr. B. Riedl (Bayer AG, Wuppertal) for a gift of chemicals
and Prof. Dr. D. Hoppe for his continuing support. This paper
is dedicated to Professor Waldemar Adam on the occasion of
his 60th birthday.
3
3
3
1.20 (d, J ) 7.6 Hz, 3 H), 1.55 (dq, J ) 7.6 Hz, J ) 5.9 Hz, 1 H),
4.86 (d, J ) 5.9 Hz, 1 H,), 5.62 (s, 1 H), 7.21-7.59 (m, 10 H). 13C
3
Supporting Information Available: Details of the X-ray
crystal structure analysis of 11a, experimental procedures for
the preparation of compounds (+)-2, 7, 8, 18-22, 26, 28, 30,
33, 35, and 38-40, and additional spectroscopic information
(NMR assignments, IR, MS) for all new compounds (33 pages).
See any current masthead page for ordering and Internet access
instructions.
NMR (CDCl3): δ -4.6 (q), -3.0 (q), 2.9 (q), 10.3 (q), 21.6 (d), 25.9
(q), 36.9 (s), 84.9 (d), 85.8 (d), 90.0 (s), 127.7 (d), 127.8 (d), 128.0
(d), 128.4 (d), 128.9 (d), 134.0 (d), 138.6 (s), 139.6 (s). Anal. Calcd
for C26H40O2Si2 (440.8): C, 70.85; H, 9.15. Found: C, 71.02; H, 9.23.
4-[1-(Dimethylphenylsilyl)ethyl]-3-(2-methyl-1,3-dioxolan-2-yl)-
2-phenyl-3-[(trimethylsilyl)oxy]oxetane (51). As described in typical
procedure C the oxetane formation was carried out on a 63 µmol scale
starting with silyl enol ether 49 (70 mg). The crude product (dr )
83/17; regioselectivity 80/20) was purified by flash chromatography
JA963827V