7490 J. Am. Chem. Soc., Vol. 123, No. 31, 2001
Gonza´lez-Nu´n˜ez et al.
jugative stabilization of diastereomeric transition states by long-
range interactions. For the first time this effect has been
evaluated based on a combination of the outstanding selectivity
shown by TFDO when reacting with saturated hydrocarbons
along with our methodology3e for single C-H bond oxygenation
of methylene groups. Moreover, the inductive effect of the
substituent modulates the reactivity with the distance while the
hyperconjugation determines the equatorial/axial selectivity.
Experimental Section
Solvents and reagents were purified by standard procedures.13
Methyl(trifluoromethyl)dioxirane (1) in dichloromethane solution was
prepared as described elsewhere.14 Trimethylsilylcyclohexane6a (2a) and
trifluoromethylcyclohexane6d (2d) were prepared by hydrogenation of
trimethylsilylbenzene and trifluoromethylbenzene following reported
procedures.6a,d Commercial tert-butylcyclohexane and methylcyclohex-
ane were purified by distillation. Alcohols 4b(C4)ax, 4b(C4)eq, 4c(C2)ax,
4c(C2)eq, 4c(C3)ax, 4c(C3)eq, 4c(C4)ax, 4c(C4)eq, 4d(C3)ax, 4d(C3)eq,
4d(C4)ax, and 4d(C4)eq were purchased as cis/trans mixtures or in pure
form. Trimethylsilyl-substituted cyclohexanols 4a(C3)eq, 4a(C3)ax,
4a(C4)eq, and 4a(C4)ax were prepared by means of a known procedure
for the catalytic hydrogenation of m- and p-trimethylsilyl-substituted
phenols.15a 3-tert-Butylcyclohexanol was prepared by means of catalytic
hydrogenation of 3-tert-butylphenol.15b Esters of trifluoroacetic acid
were prepared by treating the alcohol with an excess of trifluoroacetic
anhydride following a reported procedure.3e,f Unequivocal determination
of GC retention times for the equatorial and axial isomers was
Figure 6. Plot of ln eq/ax vs σI for monooxygenation at positions C3
and C4 of monosubstituted cyclohexanes 2.
It must be noted that since the progressive increase of the
substituent withdrawing effect simultaneously weakens Cieplak’s
interaction and enhances Anh’s interaction (Figure 5), there is
a σI value from which Anh’s hyperconjugation becomes
noticeable and progressively increases while Cieplak’s interac-
tion gradually vanishes. We should therefore expect that on
going from electron donor to electron withdrawing substituents,
the equatorial selectivity gradually diminishes up to a point
where the onset of Anh’s hyperconjugation starts to reverse the
trend. This balanced operation of both Cieplak’s and Anh’s
hyperconjugation as functions of the σI constant of the sub-
stituent has been observed experimentally in our preliminary
study of the diastereoselectivity in the oxygenation of the
2-substituted adamantane model.4
1
performed as reported below. Only the significant H NMR data for
the synthesized compounds are reported.
cis-3-Trimethylsilylcyclohexanol trifluoroacetate [3a(C3)eq]: H
1
NMR (DCCl3, 250 MHz) δ (ppm) 4.90 (dt, J1) 10.9 Hz, J2
4.5 Hz).
)
Applying this model to our data on the equatorial/axial
selectivity in the TFDO oxygenation of monosubstituted cy-
clohexanes, the results found for trifluoromethylcyclohexane
(2d) would be a consequence of an enhanced equatorial
selectivity due to Anh’s stabilization of the equatorial transition
state. Unfortunately, the necessity of dealing with conforma-
tionally homogeneous monosubstituted-cyclohexane rings re-
stricts the number of substituents in this study, thus precluding
the determination of the point of minimum equatorial/axial
selectivity derived from the balanced operation of both hyper-
conjugation models.4 Moreover, although the selectivity of the
reaction seems to be determined by the extended hyperconju-
gation, the flexibility of the monosubstituted cyclohexane model
could introduce distinct entropic factors for each diastereomeric
transition state, thus deviating the equatorial/axial ratio from
the linear trend predicted by the operation of hyperconjugative
interactions. It is worth noting that the selectivity induced by
the methyl and tert-butyl groups suggests that the hyperconju-
gative effect due to C-H and C-C bonds is roughly the same
within the experimental error.
trans-3-Trimethylsilylcyclohexanol trifluoroacetate [3a(C3)ax]: 1H
NMR (DCCl3, 250 MHz) δ (ppm) 5.25 (t, unresolved).
cis-4-Trimethylsilylcyclohexanol trifluoroacetate [3a(C4)ax]: H
NMR (DCCl3, 250 MHz) δ (ppm) 5.31 (t, unresolved).
trans-4-Trimethylsilylcyclohexanol trifluoroacetate [3a(C4)eq]: 1H
1
NMR (DCCl3, 250 MHz) δ (ppm) 4.88 (dt, J1) 11.3 Hz, J2
4.5 Hz).
)
cis-3-tert-Butylcyclohexanol trifluoroacetate [3b(C3)eq]: 1H NMR
(DCCl3, 250 MHz) δ (ppm) 4.90 (dt, J1) 10.9 Hz, J2 ) 4.3 Hz).
trans-3-tert-Butylcyclohexanol trifluoroacetate [3b(C3)ax]: 1H
NMR (DCCl3, 250 MHz) δ (ppm) 5.37 (t, unresolved).
Monooxygenation of Monosubstituted Cyclohexanes 2 with
TFDO (1). An equimolar amount of TFDO (1) in dichloromethane
was added to 1 mL of a 0.1 M solution of 2 in 1 M trifluoroacetic
anhydride in dichloromethane cooled to -40 °C. The reaction was
stirred at -40 °C until iodometric titration16 of the mixture showed
total consumption of dioxirane (2-6 h). The reaction was then warmed
to -10 °C and treated with potassium carbonate for 1 h. The crude
reaction mixture was filtered and analyzed by GC. The products were
identified by comparison with identical samples prepared as reported
above. The results are the average of at least three independent runs.
In summary, the oxygenation of methylene C-H bonds in
conformationally homogeneous monosubstituted cyclohexanes
seems to proceed along an in-plane trajectory attack of the
dioxirane while the diastereoselectivity of the reaction can be
qualitatively interpreted on the basis of the distinct hypercon-
Unequivocal Determination of GC Retention Times of Equatorial
and Axial Isomers of Compounds 3. The ratio of cis/trans isomers
1
of alcohols 4 and their trifluoroacetic esters 3 was determined by H
NMR analysis and then analyzed by means of GC. When the isomer
ratio did not allow the unequivocal determination of retention times
(i.e. cis:trans ratio approximately 1:1), the mixture of alcohols was fully
(12) One of the referees has pointed out that the deviation from linearity
seems greater for oxidations at C4. As the interactions between the remote
equatorial substituent and the equatorial transition state in the oxygenation
would seem to be much like those in solvolysis or DAST fluorination
reactions the proposition that nonlinearity is due to positive deviation for
X ) Me3Si cannot at present be dismissed. (a) Adcock, W.; Coope, J.;
Shiner, V. J., Jr.; Trout, N. A. J. Org. Chem. 1990, 55, 1411. (b) Lambert,
J. B.; Salvador, L. A.; So, J.-H. Organometallics 1993, 12, 697. (c) Adcock,
W.; Coton, J.; Trout, N. A. J. Org. Chem. 1994, 59, 1867. (d) Lambert, J.
B.; Ciro, S. M. J. Org. Chem. 1996, 61, 1940. (e) Cieplak, A. S. Chem.
ReV. 1999, 99, 1265.
(13) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: New York 1988.
(14) (a) Crandall, J. K. In Encyclopedia of Reagents for Organic
Synthesis; Paquette, L. A., Ed.; John Wiley & Sons: New York, 1995;
Vol. 3, pp 3622-3624. (b) Adam, W.; Curci, R.; Gonza´lez-Nu´n˜ez, M. E.;
Mello, R. J. Am. Chem. Soc. 1991, 113, 7654.
(15) (a) Fessenden, R. J.; Seeler, K.; Dagani, M. J. Org. Chem. 1966,
31, 2483. (b) Speier, J. L. J. Am. Chem. Soc. 1952, 74, 1003.
(16) Adam, W.; Asensio, G.; Curci, R.; Gonza´lez-Nu´n˜ez, M. E.; Mello,
R. J. Am. Chem. Soc. 1992, 114, 8345.