Spectral evidence for single electron transfer in nucleophilic aliphatic substitution of a carbanion by methyl iodide

Full text of DOI:10.1021/ja9620402

J. Am. Chem. Soc. 1997, 119, 2291-2292
2291
Communications to the Editor
Spectral Evidence for Single Electron Transfer in
Nucleophilic Aliphatic Substitution of a Carbanion
by Methyl Iodide
Laren M. Tolbert,* Joanne Bedlek, Michael Terapane, and
Janusz Kowalik
School of Chemistry and Biochemistry
Georgia Institute of Technology
Atlanta, Georgia 30332-0400
ReceiVed June 17, 1996
ReVised Manuscript ReceiVed January 8, 1997
Introduction
Despite the simmering debate within the community of
physical organic chemists over the scope and importance of
electron transfer in nucleophilic aliphatic substitution,1 evidence
continues to accumulate that electron transfer may play a role
in some mechanisms for this substitution,2 particularly when
the electrophiles are sterically hindered alkyl iodides. In a recent
paper, we described the ground state and excited state chemistry
of 9-phenylfluorenyl anion (9PF-) with neopentyl-type iodides.3
Reaction with Me3CCH2I only occurred upon irradiation (Φ )
1.0), whereas reaction with the sine qua non of alkyl iodides,
methyl iodide, occurred rapidly in the dark. Thus, we contem-
plated the mechanistic consequences of using a sterically
hindered nucleophile, 9-mesitylfluorenyl anion (9MsF-) with
methyl iodide. We now report the first evidence that this
reaction unequivocally produces radicals by direct and nearly
quantitative observation of radical intermediates.
Figure 1. Evolution of 9-mesitylfluorenyllithium absorption spectrum
at -40 °C: (A) after addition of butyllithium; (B) 15 min after addition
of MeI; (C) 20 min after addition of MeI; (D) 35 min after addition of
MeI; (E) after addition of PhNO2.
absorption spectrum of 9-mesitylfluorenyl radical,2 consisting
of a peak at 495 nm and a broader peak at 465 nm. When the
carbanion was oxidized with nitrobenzene, the same absorptions,
superimposed on benzenenitronate absorptions, could be pro-
duced in equivalent amounts (Figure 1).
The spectroscopic observations are consistent with an electron
transfer pathway in this alkylation reaction. Above -40 °C,
9MsF- undergoes dissociative electron transfer to methyl iodide
to yield methyl radical. Methyl radical can recombine with
mesitylfluorenyl radical 9MsF• in the solvent cage to yield
methylated product. Alternatively, radical pair separation leads
to irreversible methyl radical dimerization or hydrogen atom
abstraction, leaving the persistent mesitylfluorenyl radical.
From the unreactivity of 9PF- toward neopentyl iodides, its
strong reactivity toward sterically unhindered iodides, and the
sluggish reactivity of 9MsF- toward methyl iodide, a coherent
picture begins to emerge. All of these alkyl iodides undergo
dissociative electron transfer on the polarographic time scale,
making direct electrochemical comparisons difficult. However,
there is little electronic distinction between the sterically
hindered and unhindered alkyl iodides, and thus, these would
be expected to exhibit similar reduction potentials. Although
the reduction potential of alkyl iodides cannot be measured due
to their irreversibility, their reduction potential is below ca. -2
V vs SCE.2b Similarly, 9-arylfluorenyl anions show little
distinction between oxidation potentials7 and at ca. -0.5 V vs
SCE5 provide a roughly constant but insufficient driving force
for “outersphere” electron transfer. Hence, the mechanism must
involve approach of the donor carbanion to the alkyl iodide at
a near bonding distance which allows the electron transfer to
occur at lower potentials (i.e., an “innersphere” electron transfer).
What distinguishes the three cases mentioned here is the ease
of steric approach, bearing in mind that, regardless of mecha-
nism, the frontier orbital of the electrophile will be the σ* C-I
orbital associated with Walden inversion (Scheme 1). Although
SN2 mechanisms are known to exhibit significant steric con-
straints, of which Walden inversion is the most profound,
evidence has been presented that electron transfer processes may
have steric components as well.2e For methyl iodide and
9-phenylfluorenyl anion, backside approach is sterically feasible;
whereas, with the same nucleophile and neopentyl iodide, it is
not, and photoactivation is required. Conversely, for 9MsF-
9-Mesitylfluorenyl anion (9MsF-) was prepared in tetrahy-
drofuran by deprotonation of 9-mesitylfluorene with butyllithium
at -78 °C.4 Solutions of the anion were stable at room
temperature and exhibited the characteristic three-peak absor-
bance at 460, 496, and 532 nm associated with the poorly
allowed π-π* absorptions of fluorenyl anion.5 When MeI was
added at 0 °C, the orange solution turned yellow, and gas
chromatography/mass spectroscopic analysis indicated the for-
mation of the methylated product, 9-methyl-9-mesitylfluorene,
in ca. 60% yield. Methylation of 9-mesitylfluorene in Me2SO/
KOH produced the same product in >70% yield and allowed
isolation of the methylated product.6 The color changes
associated with the reaction in tetrahydrofuran could be
monitored in the absorption spectrophotometer at low temper-
atures. Upon warming the sample from -78 to -40 °C, the
characteristic absorption of the anion was replaced by the known
(1) (a) Newcomb, M.; Curran, D. Acc. Chem. Res. 1988, 21, 206. (b)
Newcomb, M.; Varick, T. R.; Choi, S.-Y. J. Org. Chem. 1992, 57, 373. (c)
Park, S.-U.; Chung, S. K.; Newcomb, M. J. Org. Chem. 1987, 52, 3275.
(2) (a) Haberfield J. Am. Chem. Soc. 1995, 117, 3314. (b) Ahbala, M.;
Hapiot, P.; Houmam, A.; Jouini, M.; Pinson, J.; Save´ant, J.-M. J. Am. Chem.
Soc. 1995, 117, 11488. (c) Daasbjerg, K.; Christensen, T. B. Acta Chem.
Scand. 1995, 49, 128. (d) Lund, H.; Daasbjerg, J.; Lund, T.; Pedersen, S.
U. Acc. Chem. Res. 1995, 28, 313. (e) Ashby, E. C. Acc. Chem. Res. 1988,
21, 414. (f) Ashby, E. C.; Park, B.; Patil, G. C. J. Org. Chem. 1993, 58,
424. (g) Ashby, E. C.; Pham, T. N.; Amrollah-Madjdabadi, A. J. Org. Chem.
1991, 56, 1596.
(3) Tolbert, L. M.; Sun, X.; Ashby, E. C. J. Am. Chem. Soc. 1995, 117,
2681.
(4) Chandross, E. A.; Sheley, C. F. J. Am. Chem. Soc. 1968, 90, 4395.
(5) Tolbert, L. M.; Fitzwater, S. M.; Netzel, T. L.; Raya, N.; Stapleton,
M. J. Phys. Chem. 1992, 96, 4492.
(6) Spectral data: 1H NMR (CDCl3) δ 7.77 (d, J ) 7.5, 2H), 7.33 (m,
2H), 7.22 (m, 4H), 6.93 (s, 1H), 6.58 (s, 1H), 2.90 (s, 3H), 2.23 (s, 3H),
1.88 (s, 3H), 1.01 (s, 3H); MS z/e (intensity) 298 (88), 203 (100), 268 (19),
253 (12), 252 (11), 179 (29), 178 (28). The gas chromatographic retention
times of the two products were also identical.
(7) Bordwell, F. G.; Bausch, M. J. J. Am. Chem. Soc. 1986, 108, 1979.
S0002-7863(96)02040-9 CCC: $14.00 © 1997 American Chemical Society

2292 J. Am. Chem. Soc., Vol. 119, No. 9, 1997
Communications to the Editor
Scheme 1
regioselectivity for alkylation of carbonyl radical anion for one-
vs two-electron processes.9 In any event, this distance depen-
dence for the electron transfer associated with nucleophilic
substitution provides further permissive evidence for the
intervention of SET in these substitutions and provides a further
complication for distinguishing between SET and SN2 pathways.
That is, an innersphere SET pathway will have similar stereo-
chemical constraints to the SN2 pathway. Admittedly, the
observation of SET is invariably the result of sterically inhibiting
the normal presumptive SN2 pathway and reducing its rate. We
note, however, that this is apparently the first example of methyl
iodide acting as an oxidant in this fashion.
reported here, backside approach is possible but not to within
an SN2 bonding distance. Rather, “innersphere” single electron
transfer (SET) occurs.2d Thus, the potential energy surface for
the electron transfer mechanism is probably shallower but at a
distance longer than that for the SN2 transition structure. Indeed,
the orbital interaction associated with electron transfer may not
be at C-9 of the fluorenyl anion but at the more unhindered
C-3 position.8 This regioselective electron transfer has prece-
dence in the calculations of Shaik, who predicts a difference in
Acknowledgment. We are indebted to the National Science
Foundation for support of this work through grant number CHE-
9111768 and through an REU fellowship for Joanne Bedlek.
JA9620402
(9) (a) Sastry, G. N.; Shaik, S. J. Phys. Chem. 1996, 100, 12241. (b)
Sastry, G. M.; Reddy, A. C.; Shaik, S. Angew. Chem., Int. Ed. Engl. 1995,
34, 1495. (c) Sastry, G. M.; Shaik, S. J. Am. Chem. Soc. 1995, 117, 3290.
(d) Pross, A. Acc. Chem. Res. 1985, 18, 212.
(8) Electron transfer induced alkylation of triphenylmethyl anions occur
at both para and R positions: Tolbert, L. M. J. Am. Chem. Soc. 1980, 102,
6808.