5636 J . Org. Chem., Vol. 61, No. 16, 1996
Notes
leading to a conformation which is locked by an increase
in the barrier of rotation about the C-C bond. Formation
of the more stable trans bridged intermediate (Figure 1)
and subsequent approach by bromine, from the less
hindered face, would lead to the predominant (2R,3S)
diastereomer.
F igu r e 1. The preferred bridged configuration of the radical
intermediate in the bromination of 1a . Approach of bromine
from the less hindered lower face would give the diastereomer
4.
Con clu sion
A number of scalemic R-chloro and R-bromo carboxylic
acids can be brominated with absolute regioselectivity,
with reaction occurring at benzylic or tertiary positions.
The stereochemistry at the R-carbon is not compromised.
For 2-chloro-3-phenylpropanoic acid derivatives 1a and
9, bromination occurs with high diastereoselectivity to
give predominantly the 2R,3S diastereomers, 4 and 10,
respectively. However, selective hydroxylation of 4 was
not stereoselective, which suggests bridging of the chloro
substituent in the radical intermediate during bromina-
tion.
The high stereoselectivity observed in bromination of
1a , however, is in contrast to the nonstereoselective
bromination of 1c.1 In order to investigate the generality
of stereoselective bromination with R-chloro hydrocin-
namates, the 4-chlorophenyl derivative 9 was prepared
from 4-chlorophenylalanine and brominated under simi-
lar conditions. The reaction took longer to reach comple-
tion but afforded an 18:1 mixture of diastereomers. The
major diastereomer 10 was isolated by recrystallization
from hexane, and assigned the (2R*,3S*) configuration
based on the similarity in chemical shift and vicinal
coupling constant of the R- and â-protons to those of the
highly analogous compound 4.
The stereochemical outcome of the reactions of 1a and
9 can be envisaged to arise from either 1,2-asymmetric
induction10 or bridging by the neighbouring Cl.11 To
explore the possibility of the former, and also the selective
reaction of the 3-bromo over the 2-chloro substituent, the
chloro-bromo compound 4 was hydroxylated under condi-
tions which favor the formation of a benzylic cation. It
has been shown that in these resonance-stabilized sys-
tems reaction generally involves an open ion with no
participation from chlorine.12 Compound 4 in water/
acetone with AgNO3 gave a 1:1 mixture of the diastere-
omeric â-hydroxy compounds 11.13 This result is again
in contrast to the comparable N-phthaloyl phenylalanine
system where hydroxylation of the corresponding â-bro-
mo derivative afforded almost exclusively the 2S,3R
diastereomer.14 The contrasting stereoselectivities of the
radical bromination of 1a and the hydroxylation of 4
would suggest that the reaction intermediates of the two
reactions are not of similar configuration. The asym-
metric environment about the R-carbon does not, by itself,
induce stereoselectivity of reaction at the benzylic pos-
tion. Presumably, the radical intermediate adopts a
bridged configuration, where C-Cl bond deformation
occurs to bring the chlorine closer to, and eclipsing, the
unpaired electron orbital.15 Bonds to the other substit-
uents at the 2-position are appropriately rehybridized
Exp er im en ta l Section
Gen er a l. A Philips MLU 160-W (220-240 V) mercury lamp
was used for NBS brominations. R-Halo acid methyl esters 1a ,
1b, 2a , 2b, 3a , 3b, and 9 were prepared by diazotization7a-c of
the corresponding amino acids in either HCl or H2SO4/KBr, and
esterification of the R-halo acids overnight in methanol (excess),
which had been pretreated with thionyl chloride (1 mol equiv).
Gen er a l P r oced u r e for Br om in a tion . A mixture of the
appropriate R-halo acid methyl ester and NBS in CCl4 (250 mL)
was heated at reflux under nitrogen. The reaction was initiated
by irradiation with a 160 W mercury lamp. After the reaction
period the mixture was allowed to cool to rt, filtered, washed
with water, and evaporated under reduced pressure to give the
crude product.
Met h yl (2R,3S)-3-Br om o-2-ch lor o-3-p h en ylp r op a n oa t e
(4). Bromination of 1a (560 mg, 2.8 mmol) with NBS (640 mg,
3.6 mmol), as described above, for 2 h gave crude 4 and 5 (18:1
ratio determined NMR analysis) in 89% yield. Recrystallization
from hexane gave 4 in 53% overall yield: mp 97 °C; 1H NMR δ
3.90 (3H, s), 4.81 (1H, d, J ) 11.3 Hz), 5.24 (1H, d, J ) 11.3 Hz)
7.40 (5H, m); LRMS (EI) m/ z 276/278/280 ([M]+, BrCl). Anal.
Calcd for C10H10BrClO2: C, 43.25; H, 3.63; Br, 28.80; Cl, 12.78.
Found: C, 43.31; H, 3.40; Br, 29.04; Cl, 13.09. [R]D (c 0.2 in
CH3Cl) 90.4°.
Meth yl (2R*,3S*)-2,3-Dibr om o-3-p h en ylp r op a n oa te (6).
Bromination of 1b (500 mg, 2.1 mmol) with NBS (440 mg, 2.4
mmol), as described above, for 2 h gave crude 6. Recrystalliza-
tion of the mixture from hexane gave 6 in 55% yield: mp 112
°C; 1H NMR δ 3.90 (3H, s), 4.98 (1H, d, J ) 11.8 Hz), 5.34 (1H,
d, J ) 11.8 Hz), 7.39 (5H, m); LRMS (EI) m/ z 320/322/324 ([M]+,
Br2). Anal. Calcd for C10H10Br2O2: C, 37.28; H, 3.13; Br, 49.66.
Found: C, 37.56; H, 2.82; Br, 49.56.
Meth yl (R)-3-Br om o-2-ch lor o-3-m eth ylbu tan oate (7). Bro-
mination of 2a (360 mg, 2.0 mmol) with NBS (540 mg, 3.0 mmol),
as described above, for 18 h and vacuum distilation gave 7 as
an oil in 45% yield: 1H NMR δ 1.79 (3H, s), 1.88 (3H, s), 3.80
(3H, s), 4.50 (1H, s); LRMS (EI) m/ z 193/195 ([M - 35]+, Br).
Anal. Calcd for C6H10BrClO2: C, 31.40; H, 4.39; Br, 34.82, Cl
15.45. Found: C, 31.10; H, 4.27; Br, 35.23, Cl, 15.81. [R]D (c
0.2 in CH3Cl) 3.5°.
(10) (a) Porter, N. A., Giese, B.; Curran, D. P. Acc. Chem. Res. 1991,
24, 296. (b) Hart, D. J ., Krishnamuthy, R. J . Org. Chem. 1992, 57,
4457. (c) Smadja, W. Synlett 1994, 1.
(11) For an overview, see: Skell, P. S.; Shea, K. J . Bridged Free
Radicals. In Free Radicals; Kochi, J . K., Ed.; J ohn Wiley & Sons, Inc.:
New York, 1973; Vol. II, pp 809-852.
(12) (a) Cabaleiro, M. C., J ohnson, M. D. J . Chem. Soc. B 1967, 565.
(b) Fahey, R. C. J . Am. Chem. Soc. 1966, 88, 4681. (c) Fahey, R. C.,
Schubert, R. C. J . Am. Chem. Soc. 1965, 87, 5172. (d) de la Mare, P.
B. D., Koenigsberger, R. J . Chem. Soc. 1964, 5327. (e) Cristol, S. J .,
Stermitz, F. R., Ramsey, P. S. J . Am. Chem. Soc. 1956, 78, 4939.
(13) Compound 4 and AgNO3 (1.2 mol equiv) were stirred in a 1:1
mixture of acetone/water overnight. The mixture was filtered, and the
solvents were evaporated under reduced pressure. The residue was
dissolved in deuteriochloroform and analyzed by 1H NMR. The
diastereomeric mixture of the hydroxy derivative 11 was identified by
comparison of the 1H NMR spectrum with published spectral data:
de la Mare, P. B. D.; Wilson, M. A. J . Chem. Soc., Perkin Trans. 2
1973, 653.
Meth yl (S)-4-Br om o-2-ch lor o-4-m eth ylp en ta n oa te (8a ).
Bromination of 3a (1.0 g, 6.1 mmol) with NBS (1.6 g, 9.1 mmol),
as described above, for 11 h and chromatography on silica gel
(EtAc/hexane) gave 8a as an oil in 43% yield: 1H NMR δ 1.75
(3H, s), 1.85 (3H, s), 2.35 (1H, dd, J ) 15.4 Hz, J ) 5.2 Hz), 2.81
(1H, dd, J ) 15.4 Hz, J ) 7.0 Hz), 3.81 (3H, s), 4.60 (1H, dd, J
) 7.0 Hz, J ) 5.2 Hz); LRMS (CI, C4H10) m/ z 321/323/325 ([M]+,
BrCl). Anal. Calcd for C7H12BrClO2: C, 34.52, 4.97; Br, 32.81,
(15) (a) Lyons, A. R.; Symons, M. C. R. J . Am. Chem. Soc. 1971, 93,
7330. (b) Bowles, A. J .; Hudson, A., J ackson, R. A. Chem. Phys. Lett.
1970, 5, 552. (c) Edge, D. J ., Kochi, J . K. Tetrahedron Lett. 1972, 1341.
(d) Edge, D. J ., Kochi, J . K. J . Am. Chem. Soc. 1972, 94, 6485.
(14) Easton, C. J ., Hutton, C. A., Tan, E. W., Tiekink, E. R.
Tetrahedron Lett. 1990, 31, 7059.