MCPBA epoxidation of alkenes: Reinvestigation of correlation between rate and ionization potential

Article abstract of DOI:10.1021/ja981531e

A kinetic equation provides relative rate constants from product analysis under competition conditions. This has been applied to the epoxidation of a large number of alkenes with m-chloroperbenzoic acid in methylene chloride. The rate constants are well c

Full text of DOI:10.1021/ja981531e

J. Am. Chem. Soc. 1998, 120, 9513-9516
9513
MCPBA Epoxidation of Alkenes: Reinvestigation of Correlation
between Rate and Ionization Potential
Cheal Kim,* Teddy G. Traylor,† and Charles L. Perrin*
Contribution from the Department of Chemistry, UniVersity of California at San Diego,
9500 Gilman DriVe, La Jolla, California 92093-0358, and Department of Fine Chemistry,
Seoul National Polytechnic UniVersity, 172 Kongnung 2-Dong, Nowon-Gu, Seoul 139-743, Korea
ReceiVed May 4, 1998
Abstract: A kinetic equation provides relative rate constants from product analysis under competition conditions.
This has been applied to the epoxidation of a large number of alkenes with m-chloroperbenzoic acid in methylene
chloride. The rate constants are well correlated with ionization potentials, but there are separate linear
correlations for aliphatic and aromatic alkenes. However, the extent of electron transfer to the peracid is
minimal. These results can be interpreted in terms of transition-state imbalance (“nonperfect synchronization”),
frontier-orbital theory, and a transition state (1) that has little charge development at carbon.
Introduction
Peracids such as m-chloroperbenzoic acid (MCPBA) react
directly with alkenes to afford the corresponding epoxide (eq
1). A large amount of data show that the reaction rate is
increased by electron-donating groups on the alkene and by
electron-withdrawing groups on the peracids but that it is
insensitive to the steric environment.1 Since many epoxidation
reactions have been observed in biological systems2 and practical
syntheses, including catalytic processes,3 it is important to
understand the epoxidation mechanism.
Although several mechanisms for the epoxidation of alkenes
by peracids have been proposed, it is still not fully understood.
No mechanism encompasses all the results, including the specific
influence of solvent, the formation of rearranged byproducts,
and the induced decomposition of peracids.4 Bartlett’s “but-
terfly” model (1) for the transition state has been widely
accepted as accounting for the main features.1,5 This mechanism
involves a concerted attack on the alkene by the intramolecularly
hydrogen-bonded peracid.6 This mechanism is consistent with
the observations that the reaction is stereospecific and occurs
easily even in nonpolar solvents, indicating that free ions are
not involved.7
Hanzlik and Shearer observed unequal secondary kinetic
deuterium isotope effects at the two alkene carbons of p-
phenylstyrene.8 They proposed an asymmetric transition state
with partial charge on one of the carbons (2). This structure,
too, would be consistent with the previously reported kinetics,
stereospecificity, and Hammett correlations. In support, several
ab initio molecular orbital studies have suggested that asym-
metric transition states are energetically more favorable than
symmetric ones9 but not by much.10 In contrast, recent high-
level DFT calculations on epoxidation of ethylene indicate
synchronous or nearly synchronous formation of the two C-O
bonds.11 Moreover, the calculated kinetic isotope effects agree
with the experimental ones, whereas those calculated for an
asymmetric transition state do not. However, with substituted
ethylenes, such as butadiene, the transition state becomes
asymmetric.
† Deceased, June 14, 1993.
(1) Swern, D. Organic Peroxides; Wiley Interscience: New York, 1971;
Vol 2.
(2) (a) Ortiz de Montellano, P. R., Ed.; Cytochrome P-450: Structure,
Mechanism and Biochemistry; Plenum Press: New York, 1986. (b) Wallar,
B. J.; Lipscomb, J. D. Chem. ReV. 1996, 96, 2625.
Another measure of the symmetry of the transition state is
the relationship between the second-order rate constant k2 and
(3) (a) Meunier, B. Chem. ReV. 1992, 92, 1411. (b) Sisemore, M. F.;
Selke, M.; Burstyn, J. N.; Valentine, J. S. Inorg. Chem. 1997, 36, 979. (c)
Finn, M. G.; Sharpless, K. B. J. Am. Chem. Soc. 1991, 113, 113. (d) Al-
Ajlouni, A. M.; Espenson, J. H. J. Am. Chem. Soc. 1995, 117, 9243. (e)
Groves, J. T.; Lee, J.; Marla, S. S. J. Am. Chem. Soc. 1997, 119, 6269. (f)
Kim, C.; Chen, K.; Kim, J.; Que, L., Jr. J. Am. Chem. Soc. 1997, 119,
5964.
(4) Curci, R.; DiPrete, R. A.; Edwards, J. O.; Modena, G. A. J. Org.
Chem. 1970, 35, 740.
(5) Dryuk, V. G. Tetrahedron 1976, 32, 2855.
(8) Hanzlik, R. P.; Shearer, G. O. J. Am. Chem. Soc. 1975, 97, 5231.
(9) Plesnicˇar, B.; Tasevski, M.; Azˇman, A. J. Am. Chem. Soc. 1978,
100, 743. Yamabe, S.; Kondou, C.; Minato, T. J. Org. Chem. 1996, 61,
616.
(10) Bach, R. D.; Canepa, C.; Winter, J. E.; Blanchette, P. E. J. Org.
Chem. 1997, 62, 5191. Bach, R. D.; Glukhovtsev, M. N.; Gonzalez, C.;
Marquez, M.; Estevez, C. M.; Baboul, A. G.; Schlegel, H. B. J. Phys. Chem.
A 1997, 101, 6092. Bach, R. D.; Winter, J. E.; McDouall, J. J. W. J. Am.
Chem. Soc. 1995, 117, 8586. Bach, R. D.; Owensby, A. L.; Gonzalez, C.;
Schlegel, H. B.; McDouall, J. J. W. J. Am. Chem. Soc. 1991, 113, 2338.
(11) Singleton, D. A.; Merrigan, S. R.; Liu, J.; Houk, K. N. J. Am. Chem.
Soc. 1997, 119, 3385. Houk, K. N.; Liu, J.; DeMello, N. C.; Condroski, K.
R. J. Am. Chem. Soc. 1997, 119, 10147.
(6) Bartlett, P. D. Rec. Chem. Prog. 1957, 18, 111.
(7) Swern, D. J. Am. Chem. Soc. 1948, 70, 1235. March, J. AdVanced
Organic Chemistry, 4th ed.; John Wiley & Sons: New York, 1992; p 826.
S0002-7863(98)01531-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/09/1998

9514 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Kim et al.
the ionization potential of the alkene. Shellhamer and Aue
observed a linear relationship between log k2 for epoxidation
of some aliphatic alkenes (ethylene, propene, methylenecyclo-
propene, cyclopropene, cyclobutene, trans-2-butene, 2-methyl-
propene, cyclohexene, methylenecyclobutane, norbornene, cy-
clopentene, methylenecyclopentane, trans,trans-2,4-hexadiene,
and 2-methyl-2-butene) by MCPBA and the ionization potentials
of those alkenes.12 This is consistent with a frontier-orbital
interaction, between the LUMO of the peracid and the HOMO
of the alkene, as dominant in modulating reactivity. Since the
HOMO is delocalized over the alkene, the good correlation
suggests that little localization of charge is involved in the
transition state.
Yet recently, Shea and Kim have measured rates of MCPBA
epoxidation of a series of cyclic alkenes, including bridgehead
alkenes and trans-cycloalkenes.13 For these strained alkenes
there is a contribution to the rate acceleration from the relief of
strain energy on passing from alkene toward epoxide. However,
even with correction for the relief of strain, the overall
correlation of log k2 with ionization potential is poor. This is
so because conjugated alkenes, which have higher HOMO
energies than simple alkenes, are nevertheless less reactive. They
concluded that the HOMO-LUMO interaction, which is a
second-order perturbation, is not the dominant factor. More
important is the first-order Coulombic interaction associated with
the stabilization of partial positive charge in the transition state
by an additional alkyl group or by conjugation.
Because of our interest in hemin-catalyzed epoxidations,14
we wanted to use substituent effects in the most widely studied
MCPBA epoxidations as a standard for comparison. Therefore
we have reinvestigated MCPBA epoxidation for a series of
alkenes across a wide range of ionization potentials. Although
many rate constants have been published, it is not easy to make
comparisons among them because they were determined under
different conditions, especially of temperature or solvent.
Herein we report the development of a kinetic method to
determine relative rate constants for the reactions of a number
of alkenes with MCPBA. This method is applicable over a
range of ca. 103-fold in reactivity. Unexpectedly we find good
correlations of log k2 vs the ionization potential of the alkene,
in contrast to the results of Shea and Kim.
Figure 1. Time dependence of 460-nm absorbance during competitive
epoxidations by 2.20 × 10-3 M MCPBA of 7.30 × 10-6 M â-carotene
and 1-octene (0, 2.63 × 10-2, 4.39 × 10-2, 8.78 × 10-2 M from bottom
to top).
× 10-6 M â-carotene in CH2Cl2 was made up quickly at room
temperature. The disappearance of â-carotene was followed by a
decrease in absorbance at 460 nm (ꢀ ) 1.5 × 105). The pseudo-first-
order rate constant was divided by the concentration of ΜCPΒΑ to
afford a second-order rate constant of 1.5 ( 0.2 M-1 s-1
.
The relative rates of epoxidation of alkenes were determined by
competition with â-carotene as a reference alkene, whose disappearance
could be monitored at 460 nm. In all cases, at least three different
concentrations of the substrate alkene were used to determine the
relative rates. In such experiments, sufficient 0.1 M MCPBA solution
was added to a solution of 7.3 × 10-6 M â-carotene and at least 2.2 ×
10-2 M of a second alkene, all in methylene chloride, to bring the
oxidant concentration to 2.20 × 10-3 M in a total volume of 0.5 mL.
A higher concentration of less reactive alkenes was used for more
effective competition with the carotene. The rate of decrease of the
â-carotene absorbance at 460 nm was then followed. Plots of
absorbance vs time of a solution of â-carotene and MCPBA in the
presence of increasing amounts of 1-octene are shown in Figure 1. It
is clear that 1-octene reduces the extent of â-carotene epoxidation.
Data Analysis. The relative rates of reaction of an alkene and a
reference diene can be obtained from either the differential or the
integrated forms of the equations describing the plots in Figure 1. The
integrated form was used, adapted as follows from the competitive
epoxidation by an iodosylbenzene as catalyzed by a hemin:17
Experimental Section
Instrumentation. UV-visible absorption spectra and kinetic
measurements were obtained on a Kontron 810 spectrophotometer
interfaced to a Celerity computer.
If S and R are substrate and reference alkenes, reacting with oxidant
O with second-order rate constants kS and kR, respectively, and if the
absorbance of R (â-carotene) is monitored, then
Materials. m-Chloroperbenzoic acid was purified as previously
described.15 1-Octene (C&B Manufacturing Co., 97%) and norbornene
(Aldrich) were distilled before use. Benzonorbornadiene, 6b,8a-
dihydrocyclobut[a]acenaphthylene, vinylcyclopropane, and a mixture
of syn- and anti-7-vinyl-bicyclo[4.1.0]-heptanes were all obtained from
previous studies.16 Other alkenes were of the highest grade available
(Aldrich) and were used as received. The solvent was methylene
chloride (Fisher Spectrograde).
d[S]
-
) kS[S][O]
(2)
dt
and correspondingly for d[R]/dt, so that
k
d[S]
[S]
S d[R]
)
(3)
kR
R
Kinetic Measurements. All of the kinetic studies were carried out
at room temperature. A solution of 2.20 × 10-3 M MCPBA and 7.3
Integrating from 0 to t gives
(12) Shellhamer, D. F., Ph.D. Dissertation, University California, Santa
Barbara, 1974.
(13) Shea, K. J.; Kim, J.-S. J. Am. Chem. Soc. 1992, 114, 3044.
(14) Traylor, T. G.; Tsuchiya, S.; Byun, Y.-S.; Kim, C. J. Am. Chem.
Soc. 1993, 115, 2775. Traylor, T. G.; Kim, C.; Richards, J. L.; Xu, F.;
Perrin, C. L. J. Am. Chem. Soc. 1995, 117, 3468.
(15) Traylor, T. G.; Lee, W. A.; Stynes, D. V. J. Am. Chem. Soc. 1986,
106, 755. Hill, K. W., Ph.D. Dissertation, University California, Santa
Barbara, 1986.
k
[S]
[R]
ln
)
S ln
(4)
(5)
S0 kR R0
From the stoichiometry
O0 - [O] ) R0 - [R] + S0 - [S]
(16) Traylor, T. G. Pure Appl. Chem. 1991, 63, 265.
(17) Traylor, T. G.; Xu, F. J. Am. Chem. Soc. 1988, 110, 1953.
or

MCPBA Epoxidation of Alkenes
J. Am. Chem. Soc., Vol. 120, No. 37, 1998 9515
Table 1. Comparison of Rates or Selectivities in MCPBA
Epoxidation with Literature Values
(O0 - [O]) - (R0 - [R])
[S]
S0
) 1 -
(6)
S0
k2, M-1 s-1 or krel
this work
literature
ref
cyclohexene
trans-stilbene
norbornene/cyclohexene
1-octene/styrene
vinylcyclohexane/styrene
methylenecyclohexane/styrene
0.0375
0.0056
2.8
1.5
1.2
0.0225
0.0036
2.8
1
19
16
20
20
20
We neglect the complication due to diepoxidation of â-carotene, since
this contribution is expected to be small, owing to the reduction in
conjugation and to the electron-withdrawing character of an epoxy
group.18 If S0 . O0 or R0, the logarithm on the left in eq 4 can be
approximated, leading to
0.71a
1.1
21
33
k
(O0 - [O]) - (R0 - [R])
a 1-Decene, assumed to be equivalent to 1-octene.
[R]
[S]
S0
S ln
) ln
≈ -
(7)
(8)
(9)
kR R0
S0
Table 2. Relative Rates for Epoxidation of Alkenes and Literature
Ionization Potentials
and to
alkene
log k2rel IP, eV ref
â-carotene
0.0
6.73
21
kS
(O0 - [O]) - (R0 - [R])
benzonorbornadiene
1,4-diphenylbutadiene
trans-stilbene
-1.21
-2.09a
-2.41
-2.36
-0.98a
-0.66a
-2.57
-3.05
-1.51
-1.17
-1.82a
-1.83
-1.24
-1.60
-1.65
-1.24
-2.57
-2.47
-2.39
-3.00
≈ -
kR
S0 ln(R0/R∞)
7.55
8.00
8.20
8.25
8.25
8.47
8.57
8.75
8.81
8.82
8.90
8.97
9.00
9.02
9.04
9.41
9.50
9.52
9.85
22
23
23
24
25
25
25
16
26
25
16
26
25
26
27
16
16
25
16
cis-stilbene
At infinite time, [O] approaches 0, so
kS O0 - (R0 - R∞)
trans-2-cis-4-hexadiene
1,3-cyclohexadiene
styrene
2,6-dimethylstyrene
7-vinylbicyclo[4.1.0]heptane
norbornene
1,4-cyclohexadiene
vinylcyclopropane
methylenecyclohexane
cyclohexene
allyltrimethylsilane
cis-2-pentene
)
kR
S0 ln(R0/R∞)
Therefore the rate ratio kS/kR can be obtained from the stoichiometry
of the samples and from the initial and final concentrations of the
reference alkene R, measured by absorption spectrophotometry. Thus,
eq 9 can give the bimolecular rate constant for each of the substrate
alkenes, relative to that for â-carotene. The concentration of the alkene
was adjusted so as to aim for R∞ ≈ R0/2, which maximizes accuracy.
Table 1 shows that the results obtained by this method are in good
accord with those previously published, which validates our analytical
method. The largest discrepancy is for 1-octene, but our rate agrees
well with that of the similar vinylcyclohexane.
6b,8a-dihydrocyclobut[a]acenaphthylene
vinylcyclohexane
1-octene
vinyltrimethylsilane
a Corrected for statistics.
Results
The relative rates for epoxidation by MCPBA of a number
of alkenes, along with published ionization potentials, are listed
in Table 2. These rate constants krel are all relative to the rate
constant for epoxidation of â-carotene. Rate constants for dienes
have been corrected for statistics.
A plot of log krel vs gas-phase ionization potential IP is shown
in Figure 2. It is readily seen that the alkenes fall into two
groups, aromatic and simple aliphatic alkenes. The polyene
â-carotene does not fit on either line.
Discussion
Relative rates of epoxidation can be determined from the
extent to which they reduce the extent of epoxidation of
â-carotene. This method permits measurement under conditions
guaranteed to be identical, including temperature and medium.
(18) Mordi, R. C.; Walton, J. C.; Burton, G. W.; Hughes, L.; Ingold, K.
U.; Lindsay, D. A.; Moffatt, D. J. Tetrahedron, 1993, 49, 911.
(19) Schwartz, N. N.; Blumbergs, J. H. J. Org. Chem. 1964, 29, 1976.
(20) Collman, J. P.; Hampton, P. D.; Brauman, J. I. J. Am. Chem. Soc.
1990, 112, 2986.
(21) Mairanovsky, V. G.; Engovatov, A. A.; Ioffe, N. T.; Samokhvalov,
G. I. J. Electroanal. Chem. Interfacial Electrochem. 1975, 66, 123.
(22) Yip, K. L.; Lipari, N. O.; Duke, C. B. J. Chem. Phys. 1976, 64,
4020.
(23) Maier, J. P.; Turner, D. W. J. Chem. Soc., Faraday Trans.2 1973,
69, 196.
(24) Bieri, G.; Burger, F.; Heilbronner, E.; Maier, J. P. HelV. Chim. Acta
1977, 60, 2213.
(25) Kiser, R. W. Introduction to Mass Spectrometry and Its Application;
Prentice-Hall: Englewood Cliffs, NJ, 1965; p 308.
(26) Demeo, D. A.; El-Sayed, M. A. J. Chem. Phys. 1970, 52, 2622.
(27) Weidner, U.; Schweig, A. J. Organomet. Chem. 1972, 39, 261.
(28) Masclet, P.; Grosjean, D.; Mauvier, G.; Dubois, J. J. Electron
Spectrosc. Relat. Phenom. 1973, 2, 225.
Figure 2. Plot of log krel for MCPBA epoxidation of alkenes vs
ionization potential. Lines are separate least-squares fits for aliphatic
(O) and aromatic (0) alkenes, relative to â-carotene (b).
The slopes of log k vs IP for aromatic and aliphatic alkenes
in Figure 2 are -0.76 ( 0.24 and -1.36 ( 0.17 eV-1
,
respectively. The magnitude of the slope for the aromatics
would be even lower, with less error, if 2,6-dimethylstyrene
were excluded. This alkene might be retarded by steric
hindrance, although significant retardation has been seen only
in extreme cases.29
(29) Rebek, J., Jr.; Marshall, L.; McManis, J.; Wolak, R. J. Org. Chem.
1986, 51, 1649.

9516 J. Am. Chem. Soc., Vol. 120, No. 37, 1998
Kim et al.
Each of these correlations indicates that the epoxidation rate
increases as the alkene becomes more electron-rich. This is
certainly reasonable for a reaction with a reagent containing
electronegative oxygens. The uniformly low slopes of the two
correlation lines in Figure 2 suggest that the transition states of
simple aliphatic and aromatic alkenes in MCPBA epoxidation
are rather similar. Moreover, these correlations imply that the
epoxidation of alkenes with peracids is related to electron
density, such that the partially charged transition-state structure
(2) cannot be rejected. Nor can the frontier-orbital interaction
be neglected.
effect on ionization potential. In contrast, stabilization by alkyl
substitution, which does not operate via delocalization (at least
not to as great an extent), is not subject to a lag. Thus, aliphatic
alkenes are epoxidized faster than aromatic ones of comparable
ionization potential, and the rates of the former class are more
sensitive to ionization potential, as measured by the slope of
Figure 2.
Alternatively, this difference can be interpreted in terms of
frontier molecular orbital theory.33 One measure of the energy
of interaction between the alkene and an electrophilic reactant
is given by eq 10,
Since a lowering of activation energy by 1 eV, or 23 kcal/
mol, corresponds to a rate acceleration of 1.6 × 1017-fold at 20
°C, the observed slopes of log k vs IP represent only 4% and
8% electron transfer at the transition state, for aromatic and
aliphatic alkenes, respectively. Even though the reagent has
electronegative oxygens, there is little transfer of electron density
from carbon to those oxygens. Instead, the slopes are so low
that it might appear that electron transfer is hardly involved.
Indeed, Shea and Kim’s observation that the correlation is poor
led them to conclude that frontier-orbital interactions are not
dominant.13 In particular, they noted that 2,4-hexadiene and
1,4-diphenylbutadiene, which differ significantly in ionization
potential, nevertheless have nearly the same rate of epoxidation.
However, Figure 2 indicates that these lie on two different
correlations.
To resolve this discrepancy, it is necessary to understand why
there are two separate correlations in Figure 2. The divergence
is such that aromatic alkenes undergo epoxidation more slowly
than aliphatic ones of comparable ionization potential. The need
to consider these two classes separately has long been recog-
nized.30 In contrast, there is no such divergence in oxidation
by the oxene (Hm+dO) formed from an iron(III) porphyrin,
and the good correlation with ionization potential was taken as
evidence for an electron-transfer mechanism.17
2
E
cC /∆ꢀ
(10)
where cC is the coefficient of an alkene carbon in the highest
occupied MO (for utmost simplicity) and ∆ꢀ is the energy gap
between that MO and the lowest unoccupied MO of the
electrophile. It is this latter that represents the dependence on
ionization potential. In simple Hu¨ckel theory, the HOMO
coefficient at CR is 0.707 in ethylene, larger than the 0.595 for
styrene. Therefore, even at constant ∆ꢀ, the interaction with
an aliphatic alkene is stronger, and also more sensitive to the
ionization potential.
This analysis suggests that polyenes, which also stabilize the
transition state by delocalization, should lie on a line separate
from aromatic and aliphatic alkenes. Certainly â-carotene does.
However, the two examples in Figure 2 lie close enough to the
line for aliphatic alkenes that it is not possible to verify this
further.
These data do not permit us to distinguish directly between
symmetric (1) and asymmetric (2) transition states. However,
a substantial positive charge at one of the carbons, as in 2, is
not consistent with the very low slopes in Figure 2. The absence
of charge development suggests a transition state resembling
1, consistent with results of calculations on ethylene itself.10,11
For substituted ethylenes, there is no requirement for sym-
metry, and a decidedly asymmetric transition state was calcu-
lated for epoxidation of butadiene.11 Yet it must be recognized
that the positive charge in 2 is so low that the asymmetry cannot
be accompanied by any substantial delocalization into an
aromatic substituent, even though that delocalization ought to
enhance the asymmetry. Moreover, the slope for aromatic
alkenes is even lower than for aliphatic ones, owing to the lag
in delocalization that we propose. This lag is further evidence
against appreciable charge asymmetry in the transition state.
We propose that the difference between aromatic and aliphatic
alkenes is a consequence of transition-state imbalance (“non-
perfect synchronization”).31 Even though electron density is
transferred from alkene to peracid, the delocalization of positive
charge to the phenyl lags behind. To the extent that epoxidation
rates correlate with ionization potentials, the radical cation 3,
Conclusions
A kinetic equation has been developed to accomplish relative
rate measurements under common competition conditions. This
has permitted the measurement of rate constants for the reactions
of m-chloroperbenzoic acid with a large number of alkenes. The
rate constants vary over 3 orders of magnitude.
The rate constants are well correlated with ionization
potentials, suggesting that the frontier-orbital interaction is
important. However, the extent of electron transfer to the
peracid is minimal. Separate correlations are obtained for
aliphatic and aromatic alkenes. These results can be interpreted
in terms of transition-state imbalance (nonperfect synchroniza-
tion), frontier-orbital theory, and a transition state (1) that has
little charge development at either carbon.
with a full positive charge, is a model for the product of the
reaction. Some of that charge, +qPh0, is delocalized into the
0
phenyl, and some, +qC , remains at the double bond. In the
transition state (4), only a fraction +q of the full charge is
developed on the alkene. Some of that positive charge, +qPh,
is delocalized into the phenyl, and some, +qC () q - qPh)
remains at the double bond. According to Kresge’s model,32
qPh is proportional to q2, rather than simply to q. This leads to
qPh/q < qPh0, meaning that the extent to which charge is
delocalized into the phenyl ring is lower in 4 than in 3.
Consequently, the phenyl does not provide as much stabilization
to the transition state as might be expected on the basis of its
Acknowledgment. We are grateful to the National Institutes
of Health (T.G.T., Grant HL 13581) and the National Science
Foundation (T.G.T., Grant CHE 87-21364) for financial support.
(30) Swern, D. J. Am. Chem. Soc. 1947, 69, 1692.
(31) Bernasconi, C. F. Acc. Chem. Res. 1992, 25, 9.
(32) Kresge, A. J. Can. J. Chem. 1974, 52, 1897.
(33) Fleming, I. Frontier Orbitals and Organic Chemical Reactions; John
Wiley & Sons: Chichester, 1978.
JA981531E