Figure 3
presence of 4 among the electrolysis products is probably
due to the decomposition of the enolate ion via a ketene
pathway.12
Figure 2
When the ammonium salt was replaced by LiClO4, no
carboxylated product could be isolated from the electrolyzed
solution. Only 6% in 2a + 2b was obtained working with
divided cells in the same experimental conditions as those
of entry 9 (Table 1).
acts as a proton donor; in fact, when this solvent was used,
the yields in reduced product 3 were very high to the
detriment of those of the carboxylated products.
The role of Et3N in this reaction is not so clear: Evans11
reported that Et3N forms a kind of aggregated complex with
the enolate that derives from the reduction of the C-Br bond,
so it should enhance the diastereoselectivity of the carboxy-
lation reaction; on the other hand, the yields in 2a + 2b are
considerably lowered by the presence of the amine. The
In all cases, the more abundant diastereoisomer13 is 2a,
as demonstrated by the X-ray analysis (Figure 2). This is
probably due to the preferential “non-Evans” conformation14
(that is, the nonchelated one) of the enolate ion during the
carboxylation (Figure 3). In fact, in our experimental
conditions, the concentration of the metallic ion, which
derives from the consumption of the sacrificial anode and
could chelate both oxygen atoms, is probably too low with
respect to the concentration of the supporting electrolyte
cation R4N+, which can coordinate but not chelate both
carbonyl oxygen atoms. It is also possible that in a solvent
such as THF, the metallic cation is not free but ion-paired
with the bromide released during the electrolysis.
When the two diastereoisomeric esters (2a and 2b) were
allowed to epimerize separately (by reaction with Et3N in
CD3CN), the equilibrium ratio of 2a:2b was 2:3 in both cases,
so we can deduce that in our experimental conditions the
reaction is partially controlled by kinetic factors.
(9) General Procedure. A solution of 3-(2-bromopropionyl)-4R-phenyl-
2-oxazolidinone16 1 in 30 mL of THF-0.2 mol/L Bu4NBF4 was electrolyzed
(undivided cells, Pt or Pb cathode, Al or Mg anode, at -20 °C) under
galvanostatic conditions (I ) 4 mA/cm2) in the presence of carbon dioxide
(p ) 1 atm). After the consumption of 2 Faradays per mol of 1, the current
flow was stopped, the solvent evaporated under reduced pressure, and the
residue poured into water. This aqueous phase was extracted with diethyl
ether (3 × 30 mL), and this organic solution was worked up as usual, giving
the starting material 1, (R)-(-)-4-phenyl-3-propionyl-2-oxazolidinone 3,17
and (R)-(-)-4-phenyl-2-oxazolidinone 4, if any. The aqueous solution was
then acidified (pH ≈ 3) with dilute HCl and extracted again with ether.
This second ethereal phase was cooled at 0 °C and treated with ethereal
CH2N2.18 (CAUTION! Diazomethane is toxic and prone to cause develop-
ment of specific sensitivity; in addition, it is potentially explosive). The
usual workup gave the mixture of 2a and 2b, whose ratio was calculated
by 1H NMR. The two pure isomers were obtained after column chroma-
tography (8:2 n-hexanes-ethyl acetate as the eluent). 3-(2-Methoxycar-
bonylpropionyl)-4S-phenyl-2-oxazolidinone (Less Polar Isomer) 2a. 1H
NMR δ (CDCl3): 7.40-7.25 (m, 5H, ar), 5.46 (dd, 1H, J ) 8.9 Hz, J )
3.7 Hz, OCHH), 4.70 (t, 1H, J ) 8.9 Hz, N.-CH), 4.53 (q, 1H, J ) 7.2
Hz, CHCO2CH3), 4.29 (dd, 1H, J ) 8.9 Hz, J ) 3.7 Hz, OCHH), 3.71 (s,
3H, CO2CH3), 1.40 (d, 3H, J ) 7.2 Hz, CH3CHCO2CH3). 13C NMR δ
(CDCl3): 170.81, 169.08, 153.69, 138.80, 129.25, 128.85, 125.94, 70.19,
57.79, 52.53, 45.66, 13.19. GC-MS m/z: 277 (M+., 3%), 218 (M+ - CO2-
Any attempt to carboxylate by electrochemical means
(using electrogenerated bases) the starting material not
containing the bromine atom (i.e., N-propionyl-4R-phenyl-
oxazolidin-2-one) failed, so the use of a halogenated deriva-
tive seems necessary.15
(12) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737-1739.
(13) This kind of molecule exhibits a quite low kinetic acidity; they are,
in fact, quite stable to silica gel chromatography and can be isolated in
pure enantiomeric form. See: Evans, D. A.; Ennis, M. D.; Le, T. J. Am.
Chem. Soc. 1984, 106, 1154-1156.
(14) Yan, T.-H.; Tan, C.-W.; Lee, H.-C.; Lo, H.-C.; Huang, T.-Y. J.
Am. Chem. Soc. 1993, 115, 2613-2621.
(15) Any attempt to obtain 2a + 2b by metalation and carboxylation
(using Mg or Zn under classical reaction conditions, followed by the addition
of CO2) failed.
(16) Ito, Y.; Sasaki, A.; Tamoto, K.; Sunagawa, M.; Terashima, S.
Tetrahedron 1991, 47, 2801-2820.
(17) Ager, D. J.; Allen, D. R.; Schaad, D. R. Synthesis 1996, 1283-
1285.
CH3, 3%), 162 (73%), 104 (85%), 77 (79%), 59 (100%). [R]20 -96.9 (c
0.96, AcOEt). 3-(2-Methoxycarbonylpropionyl)-4R-phenyl-2-oxazolidi-
none (More Polar Isomer) 2b. H NMR δ (CDCl3): 7.40-7.31 (m, 5H,
D
1
ar), 5.44 (dd, 1H, J ) 9.0 Hz, J ) 3.9 Hz, OCHH), 4.71 (t, 1H, J ) 9.0
Hz, N.-CH), 4.49 (q, 1H, J ) 7.2 Hz, CHCO2CH3), 4.25 (dd, 1H, J ) 9.0
Hz, J ) 3.9 Hz, OCHH), 3.67 (s, 3H, CO2CH3), 1.39 (d, 3H, J ) 7.2 Hz,
CH3CHCO2CH3). 13C NMR δ (CDCl3): 170.67, 168.81, 153.68, 138.39,
129.08, 128.70, 125.82, 70.30, 57.88, 52.36, 45.55, 12.95. GC-MS m/z:
277 (M+., 3%), 218 (M+ - CO2CH3, 3%), 162 (73%), 104 (85%), 77 (79%),
59 (100%). [R]2D0 -44.6 (c 0.87, AcOEt).
(10) CAUTION! Although in more than 100 experiments no particular
safety problem has been encountered, the use of perchlorates in organic
solvent must be considered as potentially dangerous (explosive). Take
adequate precautions.
(11) Evans, D. A.; Rieger, D. L.; Bilodeau, M. T.; Urp`ı, F. J. Am. Chem.
Soc. 1991, 113, 1047-1049.
(18) de Boer, T. J.; Backer, H. J. Organic Syntheses; Wiley: New York,
1963; Collect. Vol. IV, pp 250-253.
Org. Lett., Vol. 4, No. 16, 2002
2619