An efficient method for the synthesis of enantiopure cis-α,β-epoxy acids

Article abstract of DOI:10.1016/S0957-4166(99)00320-1

Enantiopure cis-α,β-epoxy acids were prepared via a modified Darzen's reaction employing the titanium-mediated bromination-aldolization of chiral acetate thioimide enolate.

Full text of DOI:10.1016/S0957-4166(99)00320-1

Pergamon
Tetrahedron: Asymmetry 10 (1999) 3249–3251
An efficient method for the synthesis of enantiopure
cis-α,β-epoxy acids
Ying-Chuan Wang, Chi-Lung Li, Hsi-Liang Tseng, Shang-Chi Chuang and Tu-Hsin Yan ∗
Department of Chemistry, National Chung-Hsing University, Taichung, Taiwan 400, Republic of China
Received 13 May 1999; accepted 4 August 1999
Abstract
Enantiopure cis-α,β-epoxy acids were prepared via a modified Darzen’s reaction employing the titanium-
mediated bromination–aldolization of chiral acetate thioimide enolate. © 1999 Published by Elsevier Science Ltd.
All rights reserved.
The great importance of chiral α,β-epoxy carboxylic acids as precursors to enantiopure α,β-epoxy lac-
tone and α-amino-β-hydroxy acids make their easy availability desirable.1 While Sharpless epoxidation,2
asymmetric nucleophilic epoxidation3 and modified Darzen’s epoxidation reactions4 are facile appro-
aches to chiral epoxides such as epoxy alcohols and α,β-epoxy esters, progress in the synthesis of
enantiopure epoxy acids and epoxy carboxylate salts has been sporadic.1,5 Although the asymmetric
epoxidation procedure reported by Sharpless has been demonstrated to be a useful approach to chiral
epoxy alcohols needed for the synthesis of enantiomerically pure α,β-epoxy acids, this method does not
generally extend to cis-epoxy acids and γ,δ-unsaturated α,β-epoxy acids.6 Herein we report an efficient
route to enantiomerically pure cis-α,β-epoxy acids and γ,δ-unsaturated α,β-epoxy carboxylate salts by
a modified Darzen’s reaction. Such a modification involves the use of one-pot bromination–aldolization
of N-acetyloxazolidinethione 17 and then converting the resulting bromohydrin aldol adducts to epoxy
acids. The reaction of thioimide 1 with TiCl4 (2.0 equiv.), diisopropylethylamine (2.2 equiv.) and bromine
(1.0 equiv.) in dichloromethane at −78°C for 0.5 h led to the bromoacetate titanium enolate. Addition
of representative aldehydes (1.5 equiv.) and reaction at −78°C for 2 h followed by extractive isolation
and silica gel purification gave bromohydrin aldols 2a–2d in excellent yields (90%) with complete
asymmetric induction (Scheme 1).8
Can the initial thioimide aldol adducts be directed toward epoxy acids? Exposure of a mixture of α-
bromo-β-hydroxy thioimide aldols 2a (1 mmol) and 10 equiv. of H2O in CH3CN (4 mL) to NEt3 (5
mmol) at 0°C for 4 h, and subsequently adding MeOH (1 mL)/saturated aqueous potassium carbonate (1
mL) and stirring at 25°C led to hydrolysis and epoxide formation to give α,β-epoxy acid 3a in 91% yield
∗
Corresponding author. E-mail: thyan@mail.nchu.edu.tw
0957-4166/99/$ - see front matter © 1999 Published by Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00320-1

3250
Y.-C. Wang et al. / Tetrahedron: Asymmetry 10 (1999) 3249–3251
Scheme 1.
(>99% ee)9 along with recovered oxazolidinethione.10 3a Showed a single set of absorptions in the 400
MHz NMR spectrum suggestive of stereospecific epoxide-formation. Hindered saturated aldol adduct
2b gave a similar result. Enantioselectivity was confirmed by chiral HPLC to be >99% by comparison to
racemic samples.10 The utility of this protocol was examined in the conversion of 2c to the corresponding
epoxy acids, which demands very mild conditions for workup due to the high propensity of phenyl
substituted epoxide functionality to rearrange in the presence of acid. Thus, under the workup conditions
described above, 3c was produced only in less than 10% yield. Replacing HCl by oxalic acid remarkably
avoided this problem and increased the yield to 84% (>99% ee).9,10 The appearance of only one set
of signals indicates no concurrent epimerization before cyclization. In the case of bromohydrin aldol
2d, all attempts to isolate the desired epoxy acid led to decomposition or bromohydrin acid 4 (96%
yield), depending on the conditions. However, adding 2.5 equiv. of sodium carbonate to a solution of 4 in
CH3CN:H2O (3:1, 4 mL) and stirring at 25°C for 10 h led to smooth epoxide formation to give sodium
epoxy carboxylate 4d, which was characterized by NMR spectroscopy.10
In conclusion, diastereopure syn-α-bromo-β-hydroxy thioimide aldols can be conveniently prepared
and serve as precursors to chiral cis-α,β-epoxy acids or sodium epoxy carboxylate.
References
1. (a) Cardani, S.; Barnardi, A.; Colombo, L.; Gennari, C.; Scolastico, C.; Venturini, I. Tetrahedron 1988, 44, 5563. (b)
Kurokawa, N.; Ohfune, Y. J. Am. Chem. Soc. 1986, 108, 6041.
2. Sharpless, K. B.; Katsuki, T. J. Am. Chem. Soc. 1980, 102, 5974.
3. (a) Reetz, M. T.; Lauterbach, E. H. Tetrahedron Lett. 1991, 32, 4477. (b) Marsman, B.; Wynberg, H. J. Org, Chem. 1979,
44, 2312.
4. (a) Corey, E. J.; Lee, D.-H.; Choi, S. Tetrahedron Lett. 1992, 33, 6735. (b) Evans, D. A.; Sjogren, E. B.; Weber, A. E.;
Conn, R. E. Tetrahedron Lett. 1987, 28, 39. (c) Abdel-Magid, A.; Lendon, N. P.; Drake, S. E.; Ivan, L. J. Am. Chem Soc.
1986, 108, 4595.
5. (a) Flisak, J. R.; Gombatz, K. J.; Holmes, M. M.; Jarmas, A. A.; Lantos, I.; Mendelson, W. L.; Novack, V. J.; Remich, J.
J.; Snyder, L. J. Org. Chem. 1993, 58, 6247. (b) Murakami, M.; Mukaiyama, T. Chem. Lett. 1982, 1271.
6. Chong, J. M.; Sharpless, K. B. J. Org. Chem. 1985, 50, 1560.

Y.-C. Wang et al. / Tetrahedron: Asymmetry 10 (1999) 3249–3251
3251
7. (a) Yan, T.-H.; Hung, A.-W.; Lee, H.-C.; Chang, C.-S.; Liu, W.-H. J. Org. Chem. 1995, 60, 3301. (b) Yan, T.-H.; Hung,
A.-W.; Lee, H.-C.; Chang, C.-S. J. Org. Chem. 1994, 59, 8187. (c) Yan, T.-H.; Tan, C.-W.; Lee, H.-C.; Lo, H.-C.; Huang,
T.-Y. J. Am. Chem. Soc. 1993, 115, 2613.
8. Assignment of the erythro configuration is based on the 1H NMR vicinal coupling using the well-established fact that Jthreo
(7–9 Hz)>Jerythro (3–6 Hz) [Ref. 7a and (a) House, H. O.; Crumrine, D. S.; Olmstead, H. D. J. Am. Chem. Soc. 1973, 95,
3310. (b) Kleschick, W. A.; Buse, C. T.; Heathcock, C. H. J. Am. Chem. Soc. 1977, 99, 247. (c) Heathcock, C. H.; Buse, C.
T.; Kleschick, W. A. J. Org. Chem. 1980, 45, 1066]. The syn-aldol adduct 2 was the only detected product in the 400 MHz
1H NMR analysis. The crude aldol adduct showed a single set of peaks in the 400 MHz 1H NMR spectrum suggestive of
complete asymmetric induction.
9. Enantiomeric excesses were determined by HPLC using a Chiralcel OD column (Daicel Chemical Industries),
hexane:iPrOH:CF3COOH (370:20:1), at 23°C with detection at 254 nm; elution times (1 mL/min flow) were 7.64 min
and 10.14 min.
10. New compounds have been characterized spectroscopically. Selected data for 3a: 1H NMR (400 MHz, CDCl3) δ 3.54 (d,
J=4.0 Hz, 1H, CHCHC_O), 3.22–3.18 (m, 1H, CHCHC_O), 1.69–1.37 (m, 4H, CH3CH2CH2), 0.92 (t, J=6.0 Hz, 1H,
CH3CH2CH2); 13C NMR (100 MHz, CDCl3) δ 173.38, 58.02, 52.51, 29.21, 19.53, 13.79; [α]2D5 −6.5 (c 2.4, CH2Cl2);
1
high-resolution MS m/e calcd for C6H11O3: 131.0717. Found: 131.0718. 3b: H NMR (400 MHz, CDCl3) δ 3.58 (d,
J=4.4 Hz, 1H, CHCHC_O), 2.92 (dd, 1H, J=9.2, 4.4 Hz, CHCHC_O), 1.72–1.56 (m, 1H, (CH3)2CHCH), 1.30 and
0.94 (2d, J=7.2 Hz, 1H, (CH3)2CH); 13C NMR (100 MHz, CDCl3) δ 173.12, 63.64, 53.01, 27.25, 20.18, 18.39; HPLC
(hexane:iPrOH:CF3COOH 370:20:1) 9.47 min (>99%); [α]2D5 −4.3 (c 0.5, CH2Cl2); high-resolution MS m/e calcd for
C6H11O3: 131.0717. Found: 131.0715. 3c: 1H NMR (400 MHz, CDCl3) δ 7.32–7.28 (m, 5H, C6H5), 4.27 (d, J=4.4 Hz,
1H, CHCHC_O), 3.80 (d, J=4.4 Hz, 1H, CHCHC_O); 13C NMR (100 MHz, CDCl3) δ 171.37, 132.02, 128.54, 128.04,
126.47, 57.95, 55.30; HPLC (hexane:iPrOH:CF3COOH 370:20:1) 9.76 min (>99%); [α]2D5 −22.3 (c 1.1, CH2Cl2); high-
resolution MS m/e calcd for C9H8O3: 164.0473. Found: 164.0474. 4d: 1H NMR (400 MHz, D2O) δ 6.25 (dq, J=15.2, 6.8
Hz, 1H, CH3HC_CH), 5.32 (dd, J=15.2, 7.2 Hz, 1H, CH3HC_CH), 3.72 (bs, 2H, CHCHC_O), 1.79 (d, J=6.8 Hz, 3H,
CH3CH_CH); 13C NMR (100 MHz, D2O) δ 174.67, 136.63, 123.35, 57.29, 57.15, 17.74.