ꢀ
L. Chacꢀon-Garcıa et al. / Tetrahedron Letters 45 (2004) 2141–2145
2144
entry 12) led to diol (S)-16a, which possesses the same
configuration as diol obtained from adduct 13a (Table 1,
entry 9). Therefore, and as was stated for S,O-acetals 6
and 7, this result indicates that the stereochemistry at C-
11 in S,O-acetals 17 and 18 is (S) and (R), respectively. It
is noteworthy that S,O-acetals 8 (R3 ¼ Ph) and 19
(R3 ¼ Me) afford the a-hydroxycarbonyl derivatives
possessing the opposite absolute configuration than
those obtained with oxathianes 2a and 3b, or oxathianes
2b and 3c, respectively. This fact represents a synthetic
complement when acyloxathianes are used as chiral
auxiliaries, since to obtain a given enantiomeric pair it
was necessary to interchange the R groupof the acyl
moiety followed by addition of the appropriate Grig-
nard reagent.5
the above S,O-acetals are opening a new avenue to
substantially improve the synthesis of the corresponding
acyloxathianes. Spectroscopic data for compounds 8
and 10a are showed.19;20
Acknowledgements
This work was supported by CONACyT (grant 35013E)
and CGPI-IPN (grant 20030702). L.C.-G., M.E.V.-D.
and S.L.-R. thank CONACyT (92069, 125225 and
165282, respectively) and CGPI/IPN (PIFI) post-
graduate fellowships. S.P.-E. acknowledges CGPI/IPN
(PIFI) for undergraduate fellowship.
Although the chiral auxiliary is recovered in good yield,
either as 4, 4a or 15, the main shortcoming of this
methodology is that in the next cycle of induction, when
reusing the chiral auxiliary 4, the separation of S,O-
acetals 6 and 7 (when using 4), or 8 and 9 (if 4a is used)
should be faced. Therefore, at that stage it should be
decided if equilibration of the crude mixture of 6 and 7
will be carried out or not, depending on the desired
absolute configuration of the target compound.
References and notes
ꢀ
ꢀ
ꢀ
ꢀ
1. Martınez-Ramos, F.; Vargas-Dıaz, M. E.; Chacon-Garcıa,
L.; Tamariz, J.; Joseph-Nathan, P.; Zepeda, L. G.
Tetrahedron: Asymmetry 2001, 12, 3095.
ꢀ
ꢀ
ꢀ
ꢀ
2. Vargas-Dıaz, M. E.; Chacon-Garcıa, L.; Velazquez, P.;
Tamariz, J.; Joseph-Nathan, P.; Zepeda, L. G. Tetrahe-
dron: Asymmetry 2003, 14, 3225.
3. For some a-phenyl-a-hydroxycarbonyl derivatives, see: (a)
Yasohara, Y.; Miyamoto, K.; Kizaki, N.; Hasegawa, J.;
Ohashi, T. Tetrahedron Lett. 2001, 42, 3331; (b) Nakajima,
Y.; Takeyama, T.; Furusato, T.; Ooya, H.; Nakayama,
M.; Sasabe, S. JP Patent 06220049 A2 19940809, 1994;
Chem. Abstr. 1995, 122, 133170; (c) Hoechst-Schering, DE
Patent, 4319887 A1 19941222, 1994; Chem. Abstr. 1995;
122, 132777.
On the other hand, taking advantage of the isolation of
S,O-acetals 6 and 7, and assuming their intermediacy in
the formation of oxathiane 2a, they were converted to
the last compound in 78% isolated yield by refluxing in
CHCl3 in the presence of p-TsOH during 1 h (Scheme 1).
This result is a notable yield improvement concerning
the synthesis of benzoyloxathiane 2a.
ꢀ
4. Solladie-Cavallo, A.; Balaz, M.; Salisova, M.; Welter, R.
J. Org. Chem. 2003, 68, 6619.
5. Lynch, J. E.; Eliel, E. L. J. Am. Chem. Soc. 1984, 106,
2943.
From the last reaction were isolated dimeric com-
pounds17 in roughly 10% yield, which are structurally
related to dimer 22 isolated from the treatment of
hydroxythiol 4 with paraformaldehyde (Scheme 1).4;18
As mentioned above, such behavior could be explained
in terms of angular strain. In other words, the lower
conformational freedom of hydroxythiol 4, that is, as
compared to hydroxythiol prepared from pulegone,5
partially precludes the formation of an additional ring
(in this case the oxathiane ring), affording dimeric
compounds with less content of angular strain. This is
the reason why dimer 15 is likely formed in lieu of the
sultine,12 which was not previously isolated.1;2
6. Starting from diastereomerically pure S,O-acetal, certain
degree of epimerization at C-11 is observed when CH2Cl2
is not anhydrous. OTBS protection in DMF as solvent
should also be avoided due to epimerization of pure S,O-
acetal.
7. S,O-Acetals 6 and 7 are easier separated by flash chroma-
tography than compounds 8 and 9 because their larger Rf
value differences; however, the later can be efficiently
separated by applying the following typical procedure: In
a chromatography column (id ¼ 3.5 cm) packed with 90 cm
of flash silica gel, 584 mg of the crude mixture of S,O-
acetals 8 and 9 were eluted with a mixture of EtOAc–
hexane (1:49) at 15 psi, giving 344 mg (59%) of 8, 192 mg
(33%) of 9 and 46 mg (8%) of a mixture of both
compounds.
8. This was corroborated by converting diastereomerically
pure 6–8 and comparing its Rf values with the crude
mixture of 8 and 9.
9. Grignard reagents should be freshly prepared; otherwise
epimerization of pure 8 or 9 at C-11 occurs before addition
takes place, affording mixtures of adducts 10 and 11, or 12
and 13, respectively, in unpredictable ratios.
In search to improve the synthesis of acetyloxathiane 2b,
the above protocol was applied to the crude mixture of
S,O-acetals 17 and 18. However, in this case the yield of
the desired oxathiane 2b resulted similar to that
described without previous isolation of S,O-acetals 17
and 18.1
In conclusion, the synthesis of S,O-acetals 6, 7, 17, and
18 provides clear evidence that our experimental pro-
tocol can be extended to the preparation of a wide
variety of analogous S,O-acetals. In addition, was
demonstrated that such compounds represent an
attractive synthetic alternative to prepare the some times
desirable enantiomeric pair of molecules in high optical
purity. Finally, was demonstrated that the isolation of
10. This by-product is formed by hydride transfer from the b-
position of the Grignard reagent: Hamelin, A. Bull. Soc.
Chim. Fr. 1961, 1211.
11. (a) Cram, D. J.; Kopecky, K. R. J. Am. Chem. Soc. 1959,
81, 2748; For evidence of Cram-chelated transition state,
see: (b) Frye, S. V.; Eliel, E. L.; Cloux, R. J. Am. Chem.
Soc. 1987, 109, 1862; (c) Frye, S. V.; Eliel, E. L. J. Am.