2.3
Ar2
Ar1
+
1 Ar1 = 2-pyridyl (5 equiv.)
(Eo = –2.31 V vs. SCE)
4 Ar2 = 2-quinolyl
(Eo = –1.92 V vs. SCE)
1.8
1.3
at –1.4 V vs. Ag/AgBr
ca. –1.76 V vs. SCE
electrolysis
Ar1
Ar2
Ar1
+
+
0.8
Ar2
Ar2
Ar1
5e (14%)
5c (ca. 50%)
5a (<5%)
0.3
Scheme 2
0
20
40
60
80
100
120
–0.2
aryl)cyclobutanes are formed by oxidation of the intermediate
radical anion (tentatively represented as 8); in principle this
could be the result of disproportionation (route B) or of a chain
process as in the vinyl-sulfone case3 (route C). The results from
the reaction profile (Fig. 1) indicate cyclobutane formation via
the chain process with no overall consumption of charge.
Co-electrolysis gives further insight (Scheme 2 and Table 1).
Co-electrolysis of 3 in the presence of a five-fold excess of 1, at
the potential of the more easily reduced substrate 3, gave the
cross-coupled cyclobutane 5d and the homo-coupled cyclo-
butane and linear dimer 6b, respectively. Cyclobutane 5a was
found in only trace amounts. Similar behaviour was observed
for the co-electrolysis of 4 and 1. Cross-coupled cyclobutanes
are therefore formed by reaction between the radical-anion
formed at the lower potential and the other, unreduced,
component. Formation of both radical anions, e.g. by homo-
geneous electron transfer, would give all three possible
cyclobutanes in comparable amounts.
Thus there is compelling evidence for the formation of the
cyclobutanes by allowed cycloaddition between vinylpyridines
or vinylquinolines and the radical anions derived from them. A
detailed mechanistic and kinetic examination is underway,
aimed at distinguishing conclusively between the possibilities
for the follow-up reactions as outlined in Scheme 1.
We acknowledge support from the EPSRC and from the EU
Human Capital & Mobility Institutional Grant # ERBCH-
BGCT940590.
Q / coulombs
Fig. 1 Reaction profile for controlled potential electrolysis of 1 (see Table
1); (5) (CB)/(EHD), (:) (VP)/[Total], (/) (CB)/[Total], (-) (EHD)/
[Total], (+) (VPH2)/[Total]. [Total] = (VD) + (CB) (EHD) + (VPH2).
cm22) for concentrations in the range 60 mm–0.3 m. Relevant
reduction potentials10 are given in Table 2.
Vinylpyridines 1 and 3 give chemically irreversible reduction
on cyclic voltammetry at low scan rates but reversibility is
apparent for the reductions of 2 and 4 at modest scan rates ( < 10
V s21). Thus, apart from direct further reduction of the
cyclobutanes it is possible that electron transfer from the
persistent first-formed radical anions will take place to give
redox-catalyzed cleavage of the cyclobutanes.
A reaction profile was constructed (Fig. 1) for cathodic
constant potential reduction of 1 by using GLC analysis to
follow relative concentrations of reactant and products as a
function of charge passed. The results show clearly that 1 (VP)
was consumed using 0.70 F. Furthermore cyclobutane 5a (CB)
and linear hydrodimer 6a (EHD) were formed in almost
constant proportion (2:1) throughout the electrolysis. A third
product was that of 2 F cathodic hydrogenation (2-ethylpyr-
idine). The profile and coulometry are consistent with consump-
tion of the first-formed radical anion in parallel reactions;
dimerisation leading to linear EHD, cycloaddition leading in a
catalytic chain process to the cyclobutane, and cathodic
hydrogenation leading to 2-ethylpyridine. The final molar
proportions of products were 6a (0.29), 5a (0.51) and
2-ethylpyridine (0.20). Consequently the proportion of charge
consumed, given that EHD is a 1 F process and hydrogenation
a 2 F process, must be linear hydrodimer (0.29 F) and
2-ethylpyridine (0.4 F), totalling 0.69 F. 2-Vinylpyridine 1 was
consumed in 0.70 F, which indicates that the cyclobutane is
formed without overall charge consumption. This experiment is
reproducible and repeated experiments gave similar results.
The possibilities are detailed in Scheme 1, which illustrates
formation of the linear hydrodimer by the usual radical anion–
radical anion route (A), by disproportionation of the first
product of cycloaddition, the radical anion route (B), and by
subsequent 2e reduction of the cyclobutanes. The 1,2-di(hetero-
Notes and References
† E-mail: j.utley@qmw.ac.uk
1 F. Zhou and A. J. Bard, J. Am. Chem. Soc., 1994, 116, 393.
2 I. Fussing, M. Gu¨llu¨, O. Hammerich, A. Hussain, M. F. Nielsen and
J. H. P. Utley, J. Chem. Soc., Perkin Trans. 2, 1996, 649.
3 N. L. Bauld, Advances in Electron Transfer Chemistry, 1992, vol. 2,
pp. 1–61 and references cited therein.
4 J. Delaunay, G. Mabon, A. Orliac and J. Simonet, Tetrahedron Lett.,
1990, 31, 667. See also, J. Delaunay, A. Orliac and J. Simonet,
J. Electrochem. Soc., 1995, 142, 3613.
5 J. D. Anderson, M. M. Baizer and E. J. Prill, J. Org. Chem., 1965, 30,
1645.
6 Crystal data for 5b: colourless crystal, monoclinic, P21n, a
=
26.095(10), b = 7.138(2), c = 6.134(3) Å, b = 92.59(3)°, V =
1141.4(8) Å3, Z = 4, R = 0.0706, GOF = 0.677. CCDC 182/719.
7 All cyclobutanes were characterised by high resolution mass spectrome-
try, 1H NMR (250 and 600 MHz) and 13C NMR (62.5 MHz)
spectroscopy; signal assignments were made using 600 MHz TOCSY,
NOESY and HMQC spectroscopy. Satisfactory elemental analyses
were obtained from all crystalline compounds, while the purity of oils
was checked by GC analysis.
Ar
Ar
Ar
Ar
e
–
Ar
•
.-
Ar
8
(A)
(C) chain process
dimerisation
(B)
disproportionation
Ar
Ar
8 T. Nakano, A. Martin, C. Rivas and C. Pe´rez, J. Heterocycl. Chem.,
1977, 14, 921.
Ar
–
2–
Ar
•
2e
(stepwise?)
Ar
Ar
9 S. Takamuku, B. R. Dih-Nghoe and W. Schnabel, Z. Naturforsch, Teil
A, 1978, 33, 1281; S. Takamuku, H. Kighawa, S. Suematsu, S. Tokai, K.
Tsumoni and H. Sakurai, J. Phys. Chem., 1982, 86, 1861.
10 Formal reduction potentials (E°) were measured in collaboration with O.
Hammerich and M. F. Nielsen (University of Copenhagen).
2 H+
Ar
Scheme 1
Received in Cambridge, UK, 16th October 1997; 7/07460C
540
Chem. Commun., 1998