516 J. Am. Chem. Soc., Vol. 121, No. 3, 1999
Holm
atoms on going from the ground state to the transition state.
Going from methyl iodide to the SN2 transition state does not
change the bonds between carbon and hydrogen significantly,
and KIEs are as mentioned above small. Bonds between
R-hydrogen and carbon in a free radical, however, are much
less “stiff” and more pliable than the bonds in methyl iodide.
This is known from the available IR spectroscopic data.7 A crude
estimate of the equilibrium
for the reaction with sodium naphthalenide. For these reactions
the high KIE indicates a reaction mechanism in which electron
transfer with formation of a methyl radical is the rate-
determining step. The KIE is only a fraction of the EIE, but
the variation observed indicates that the degree to which the
ET transition state has developed radical hybridization varies
for the individual reaction. C-I bond-breaking which is assumed
to be concerted with ET10 is probably not complete in the TS.
For two reactions in which the mechanism obviously must
include electron transfer to methyl iodide, very low KIEs are
found: (i) the reaction with magnesium to form the Grignard
reagent and (ii) the reduction with sodium in ammonia to form
methane.
Concerning (i), it was shown by the Whitesides group11 that
in the reaction of alkyl bromides and alkyl iodides with
magnesium the rate of the reaction did not vary with the nature
of the alkyl group. The reaction is mass transport or diffusion
controlled, and for the same reason the KIE for the reaction
must be unity as is actually found. The 13C KIE for the reaction
was likewise and for the same reason found to be unity.12
•
•
CD3 + CH3I h CD3I + CH3
(1)
[CD3I][CH3]/[CD3][CH3I] ) K1
which is the equilibrium isotope effect (EIE) for complete
homolysis of methyl iodide, may be obtained8 from the formula
K ) exp[0.7193(νR1 - νR2 - νP1 + νP2)/T]
where the ν values are the normal-mode frequencies for reactants
and products using only the frequencies which directly involve
hydrogen motion. Using literature values for C-H vibrations
and calculated values for C-D vibrations, it is found that the
right side in (1) is favored by the zero point energy correspond-
ing to 664 cal or K1 ) 3.07 at 25 °C.
The rate at which the equilibrium (1) is established is not
known. It might be possible that the results obtained are
influenced by reaction of the methyl radicals with methyl iodides
before hydrogens are abstracted from the donor. The values
found for kH/kD might in such cases tend to be too large. Since
the observed KIEs, however, in no case approach the calculated
EIE, it may be assumed that hydrogen abstraction and radical
recombination are faster than attack on methyl iodide.
The reactions listed in Table 1 may be grouped according to
the predicted reaction mechanisms.
The reactions with NaBH4 and with LiAlH4 (LAH) have low
KIEs, confirming that these reagents react via an SN2 transition
state by direct hydride transfer. This is most obvious for the
borohydride, for which the KIE is slightly inverse, corresponding
to a slightly smaller steric bulk of the CD3 group than of the
CH3 group.9 LAH likewise has a small, normal KIE and reacts
as a nucleophile.
It has been shown by Ashby et al. that LAH reacts with
1-halobornanes5b and with sterically hindered, neopentyl-like
alkanes5a,c by ET mechanisms. With halobornane the SN2
mechanism is impossible, and with branched alkyls the ET
mechanism is very much favored. Using the primary octyl
iodide, the Ashby group found4d an electron transfer mechanism
in the later phases of the reaction, while the initial phase was
strictly SN2. AlH3 reacted largely by ET. The determinations
were based on isotopic D/H exchange in the reaction product.
The normal H/D KIE found in the present competition experi-
ments is within the limits for KIEs for nucleophilic substitution.2
When the reaction was carried out in perdeuteriodiethyl ether,
only a trace amount of isotope exchange was, however, found
in the methane. ET is therefore of no importance in this reaction.
High KIEs on the order of kH/kD ) 1.40-1.70 (12-20% per
D) are found for ET to methyl iodide from a platinum cathode,
from benzophenone ketyl, and from sodium metal, for the
Kharasch reaction, for the reaction with hydrogen iodide, and
Concerning (ii), a high value for kH/kD in the reaction with
HI and a low value in the reaction with sodium in liquid
ammonia were interpreted as being characteristic for inner
sphere ET with a high KIE and for outer sphere ET with a low
KIE, respectively.4
In light of the new results it seemed to be a possibility that
low KIEs for this reaction are found because the ET for the
reaction is diffusion controlled. In the present study it has been
found that the reaction in flow stream experiments is complete
at the end of the mixing period, ∼10-3 s, meaning that the half-
life is less than 200 µs. The reaction is extremely fast and is
mixing controlled if not diffusion controlled. Competition
kinetics tend to be unreliable with very fast reactions.13 Whether
the distinction between outer sphere and inner sphere ET to
methyl iodide is relevant is therefore dubious. The rather
significant 13C KIE found for the reaction is dubious as well.4
Whether formation of methylsodium (anionic methyl) is a
reaction step in the reduction of methyl iodide with sodium in
ammonia should be considered. Monomeric methyllithium is
presumably a good model for methylsodium, and the position
of equilibrium (2) indicates the size of a possible KIE for
•
•
CH3 + CH3Li h CD3Li + CH3
(2)
electron transfer from methylsodium to methyl iodide. K2 is
found by combining (1) with (3) since K2 ) K1/ K3. Since K3
CD3Li + CH3I h CD3I + CH3Li
(3)
has been found14 to be 2.4, K2 is as low as 1.28. A rate-
determining step consisting of ET to methyl radical to form
methyl anion would therefore have a low KIE. A scheme for
(10) (a) Lexa, D.; Saveant, J. M.; Su, K. B.; Wang, D. L. J. Am. Chem.
Soc. 1988, 110, 7617. (b) Saveant, J. M. AdV. Phys. Org. Chem. 1990, 26,
1.
(11) Barber, J. J.; Whitesides, G. M. J. Am. Chem. Soc. 1980, 10, 02,
239
(12) Vogler, E. A.; Stein, R. L.; Hayes, J. M. J. Am. Chem. Soc. 1978,
100, 3163.
(13) (a) Francis, A. W. J. Am. Chem. Soc. 1926, 48, 655. (b) Tolgyesi,
W. S. Can. J. Chem. 1965, 43, 3, 343. (c) Felkin, H.; Frajerman, C.
Tetrahedron Lett. 1970, 1045.
(7) Pacansky, J.; Koch, W.; Miller, M. D. J. Am. Chem. Soc. 1991, 113,
317.
(8) Shiner, V. J., Jr.; Neumann, T. E. Z. Naturforsh. 1989, 44a, 337.
(9) Brown, H. C.; Azzaro, M. E.; Koelling, J. G.; McDonald, G. J. J.
Am. Chem. Soc. 1966, 88, 2520.
(14) Holm, T. J. Organomet. Chem. 1996, 506, 37.