C O M M U N I C A T I O N S
Table 2. Electrochemically Promoted Cyclobutanation of
Bis(enones) 1a, 1b, 1d, and 1ea
is formed, cycloaddition in these cases is highly stereo- and
periselective.
The pronounced cis-diastereoselectivity is presumed to derive
from relatively strong electrostatic interactions between the sodium
ion and both carbonyl oxygens in the transition state for cyclization.
Although the anion radical moiety is presumed to reside primarily
upon just one of the aroyl groups in the product anion radical, it
appears plausible that in the transition state the anion radical moiety
is substantially delocalized over both benzoyl groups. The circum-
stances under which the Diels-Alder product is not formed at all
are attractive synthetically but are also of particular interest
mechanistically. The most intriguing possibility is that, as a result
of enhanced thermodynamic driving force, the cyclobutanation may
have become concerted. The more likely possibility, however, is
that the second cyclization step may have become so rapid that the
bond rotations which may be necessary in order to generate the
conformations appropriate for Diels-Alder formation may be too
slow to compete with the covalent bond formation which leads to
the cyclobutane product.
The further possibility of accelerating the rate of the second
cyclization step, and therefore increasing the yield of pericyclic
products, by installing electron-withdrawing groups on the aromatic
ring was also investigated. As demonstrated by the reaction of
substrate 1e, a 4-chloro substituent is indeed observed to signifi-
cantly enhance the yield of cyclobutane product cis-2e, but in
contrast to the results obtained for 1c and 1d, a small amount of
Diels-Alder adduct 3e is also formed. Notably, products of
dechlorination are not observed. Finally, with increased chloro
substitution, as in the case of 3,4-dichlorobenzoyl-derived bis(enone)
1f, the yield of pericyclic products is relatively modest. In this case
a large amount of polymer formation is observed, suggesting the
anion radical moiety may now be too stabilized for efficient
intramolecular cycloaddition.
In all cases, these chemically induced cycloadditions appear to
be stoichiometric, rather than catalytic radical chain processes.
Optimum yields were obtained when about 70-120 mol % of the
chrysene anion radical was used relative to the substrate. Quite
possibly, tight ion-pairing of the sodium ions with the product anion
radicals in the somewhat nonpolar solvent has the effect of retarding
the rate of intermolecular electron transfer to substrate molecules,
which would be required for chain propagation. Notably, although
four products are formed in some cases, the cyclobutane products
cis-2a-cis-2f are predominantly the major reaction products (Table
1). Finally, use of samarium iodide failed to generate any cyclo-
butane products. Instead, the previously mentioned aldol-type
products were formed in a reaction requiring 2 mol of samarium
iodide.
For selected substrates, electrochemical promotion of the cy-
cloaddition was examined. For cycloadditions conducted under
electrochemical conditions, reduction of the distonic anion radical
intermediate does not contribute as a competitive reaction pathway.
However, both cis- and trans-cyclobutane isomers are formed, as
would perhaps be expected considering the relatively more polar
solvent involved and the steric characteristics of the tetrabutylam-
monium counterion, both of which would tend to diminish ion-
pairing interactions. The total yields of pericyclic products are seen
to be quite good, and since the cis-cyclobutanes are readily
converted to the more stable trans-cyclobutanes,4 the procedure
affords a high-yielding approach to the latter. As with the chemical
procedure, no pericyclic products are formed when the COY groups
are acetyl or ester. An especially attractive feature of the electro-
chemical procedure is that it is electrocatalytic, furnishing modest
turnovers of between 3 and 5 (Table 2).
cis-2
trans-2
3
pericyclic products
(total yield%)
substrate
(yield%)b
(yield%)b
(yield%)b
1a
1bc
1d
1e
17.1
38.8
14.4
11.1
58.6
20.9
28.9
52.0
12.6
28.4
8.2
88.3
88.1
51.3
74.7
11.6
a RepresentatiVe Procedure: To a flask charged with 1a (100 mg, 0.329
mmol, 100 mol %) under a positive flow of N2(g) was added 22 mL of a
0.1 M solution Bu4NBF4 in acetonitrile. This solution was added to the
working electrode (WE) compartment, and a further 6 mL of the electrolyte
solution was placed in the counter electrode (CE) compartment. Reticulated
carbon electrodes were used for the WE and CE, alongside a silver wire
“pseudo-standard” reference electrode (RE) encased in porous Vycor glass.
The solution was subjected to electrolysis at constant voltage, with stirring
under an atmosphere of N2(g) at room temperature, between -1.5 and -2.0
V vs RE. Upon complete consumption of starting material, the reaction
mixture was partitioned by between water and benzene. The organic layer
was separated, dried (Na2SO4), filtered, and evaporated to give the crude
reaction products. b Isolated yield after purification by silica gel chroma-
tography. c For 1b, MgClO4 was used as electrolyte and electrolysis was
performed at -2.5 to -3.0 V.
further variation of the single-electron reductant. The scope of these
transformations, including intermolecular cross-cyclobutanation and
Diels-Alder cycloaddition, and their reaction mechanisms are
currently under active investigation in these laboratories.
Acknowledgment. This research was supported, in part, by the
Robert A. Welch Foundation (F-149, N.L.B.) and the NIH (RO1
GM65149-01, M.J.K.).
Supporting Information Available: Spectral data for all new
compounds (1H NMR, 13C NMR, IR, HRMS) (PDF). Single-crystal
X-ray crystallographic data for compounds cis-2b, trans-2b, 3b, and
4b (CIF). This material is available free of charge via the Internet at
References
(1) Bauld, N. L. Tetrahedron 1989, 45, 5307.
(2) Bauld, N. L. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-
VCH: Weinheim, 2001; Vol. 2, p 133.
(3) For electrochemically induced anion radical cyclobutanation, see: (a)
Delaunay, J.; Mabon, G.; Orliac, A.; Simonet, J. Tetrahedron Lett. 1990,
31, 667. (b) Jannsen, R.; Motevalli, M.; Utley, J. H. P. J. Chem. Soc.,
Chem. Commun. 1998, 539. (c) Roh, Y.; Jang, H.-Y.; Lynch, V.; Bauld,
N. L.; Krische, M. J. Org. Lett. 2002, 4, 611.
(4) Anion radicals haVe been implicated as reactiVe intermediates in the Co-
catalyzed [2 + 2] cycloaddition of bis(enones): (a) Baik, T.-G.; Wang,
L.-C.; Luiz, A.-L.; Krische, M. J. J. Am. Chem. Soc. 2001, 123, 6716. (b)
Wang, L.-C.; Jang, H.-Y.; Lynch, V.; Krische, M. J. J. Am. Chem. Soc.
2002, 124, 9448.
(5) For other pericyclic reactions of anion radicals, see: (a) Bauld, N. L.;
Chang, C.-S.; Farr, F. R. J. Am. Chem. Soc. 1972, 94, 7164. (b) Bauld,
N. L.; Cessac, J.; Chang, C.-S. Farr, F. F.; Holloway, R. J. Am. Chem.
Soc. 1976, 98, 4561. (c) Fox, M. A.; Hurst, J. R. J. Am. Chem. Soc. 1984,
106, 7626. (d) Borhani, D. W.; Greene, F. D. J. Org. Chem. 1986, 51,
1563. (e) Witmer, M.; Altmann, E.; Young, H.; Sancar, A.; Begley, T. P.
J. Am. Chem. Soc. 1989, 111, 9264. (f) Pasto, D. J.; Yang, S.-H. J. Org.
Chem. 1989, 54, 3544.
(6) â,â-Coupling of bis(enones) with and without subsequent aldolization has
previously been observed under the conditions of electron transfer. For
selected examples, see: (a) Enholm, E. J.; Kinter, K. S. J. Am. Chem.
Soc. 1991, 113, 7784. (b) Hays, D. S.; Fu, G. C. J. Org. Chem. 1996, 61,
4.
(7) It appears unlikely that the aldol-type cyclization occurs at the stage of
distonic anion radical in view of the fact that aldol type products are not
formed in the electrochemical reactions, which also involve distonic anion
radical intermediates.
In summation, we report the first anion radical cycloadditions
mediated by chemical agents in solution. The present results clearly
suggest the feasibility of expanding this new reaction type through
JA030543J
9
J. AM. CHEM. SOC. VOL. 126, NO. 6, 2004 1635