6854
P. Cankarˇ et al. / Tetrahedron Letters 46 (2005) 6851–6854
secondary carbamate 19, electrochemical reduction at
À2.65 V at the mercury cathode afforded the cyclic car-
bamate 20 without any formation of cinnamyl alcohol
or 1-phenylpropene (Fig. 4).
(CFI), Ontario Innovation Trust (OIT), TDC Research
Foundation Fellowship to S.C.B., and the Canada Re-
search Chair Foundation.
References and notes
3. Conclusions
1. Solis-Oba, A.; Hudlicky, T.; Koroniak, L.; Frey, D.
Tetrahedron Lett. 2001, 42, 1241.
2. Oshima, M.; Shimizu, I.; Yamamoto, A.; Ozawa, F.
Organometallics 1991, 10, 1221.
3. Oshima, M.; Sakamoto, T.; Maruyama, Y.; Ozawa, F.;
Shimizu, I.; Yamamoto, A. Bull. Chem. Soc. Jpn. 2000, 73,
453.
4. Lautens, M.; Paquin, J. Org. Lett. 2003, 5, 3391.
5. Yoshida, A.; Hanamoto, T.; Inanaga, J.; Mikami, K.
Tetrahedron Lett. 1998, 39, 1777.
6. Maruyama, Y.; Sezaki, T.; Tekawa, M.; Sakamoto, T.;
Shimizu, I.; Yamamoto, A. J. Organomet. Chem. 1994, 473,
257.
7. Hansen, J.; Freeman, S.; Hudlicky, T. Tetrahedron Lett.
2003, 44, 1575.
8. Experimental procedure: Cyclic voltammetry experiments
were performed with 15–30 mg of substrate in 30 mL of
CH3CN containing 0.15 M Et4NBr (or 0.1 M n-Bu4NPF6
for values reported in Table 1). Glassy carbon was used as
the working electrode. Platinum and AgNO3/Ag (0.1 M
CH3CN) were used as counter and references electrodes,
respectively. The working and counter electrodes were
polished on alumina before use. iR compensations were
applied for all experiments for potential measurements.
Sweep rates were typically 100 mV/s. The reported poten-
tials are for peak maxima. The electrolyses were performed
at controlled potential in a cell with two compartments. The
cathode was a mercury pool and the anode was platinum.
The reactions were monitored by observing the drop in the
current and by TLC. The solution was decanted into ether
and filtered to remove the electrolyte. The filtrate was
concentrated and the products isolated by column
chromatography.
The electrochemical reduction of carbonates versus car-
bamates proceeds with complete selectivity with the car-
bonate moiety reduced in preference to carbamate. In
the absence of a proton source, tertiary carbamates are
not reduced and secondary carbamates cyclize via one
of the possible pathways proposed in Figure 5. In carba-
mates, the tendency to form the ketyl versus styryl
radical anion may be similar (as in esters). We offer an
initial speculation regarding two options that may be
possible: one involving the proton transfer as in path-
way A (Fig. 5), the other proceeding through the ketyl
cyclization as in pathway B. Both pathways would lead
to the same product, 22, after a loss of an electron from
ketyl 21.
In the presence of a proton source, tertiary carbamates
are fully reduced to amines and cinnamyl alcohol.
Secondary carbamates are prone to cyclization follow-
ing the first electron transfer and are not further
reduced. The selectivity of removal of the cinnamyl
group favors the less basic atom (oxygen), in agreement
with previously reported observations for ethers versus
amines.7 The mechanism of the observed electrocycliza-
tion will be investigated in detail, and the results as well
as potential applications in synthesis will be reported in
due course.
Acknowledgements
9. The conversions of substrates to products were complete
(vide TLC analysis). Isolation and purification of products
at the scale at which the initial experiments were performed
(0.2–0.7 mM) were complicated by the large amount of
electrolyte used. Optimized larger scale procedures will
obviate these problems.
This work was funded by National Science and Engi-
neering Research Council (NSERC) of Canada,
NSERC-PDF to S.C.B., Petroleum Research Fund
administered by the American Chemical Society (PRF-
38075-AC), Canadian Foundation for Innovation