Improved synthesis of both enantiomers of trans-cyclohex-4-ene-1,2- dicarboxylic acid

Article abstract of DOI:10.1016/S0957-4166(99)00345-6

A facile synthesis of both enantiomers of trans-cyclohex-4-ene-1,2- dicarboxylic acid on a multigram scale, that starts from inexpensive, commercially available compounds, is described.

Full text of DOI:10.1016/S0957-4166(99)00345-6

Pergamon
Tetrahedron: Asymmetry 10 (1999) 3403–3407
Improved synthesis of both enantiomers of
trans-cyclohex-4-ene-1,2-dicarboxylic acid
Anna Bernardi,a,∗ Daniela Arosio,a Donatella Dellavecchia a and Fabrizio Micheli b
aDipartimento di Chimica Organica eIndustriale, via Venezian 21, 20133 Milano, Italy
bGlaxoWellcome SpA, Medicines Research Centre, via Fleming 4, Verona, Italy
Received 23 July 1999; accepted 10 August 1999
Abstract
A facile synthesis of both enantiomers of trans-cyclohex-4-ene-1,2-dicarboxylic acid on a multigram scale, that
starts from inexpensive, commercially available compounds, is described. © 1999 Elsevier Science Ltd. All rights
reserved.
1. Introduction
trans-Cyclohex-4-ene-1,2-dicarboxylic acids 4 and ent-4 are versatile chiral building blocks which
have found numerous applications in the synthesis of complex molecular architectures, including natural
products.1–5 Recently we have introduced a novel application of this synthon, by demonstrating that the
cis diol obtained by osmium dihydroxylation of 4 can be effectively used as a 3,4-disubstituted galactose
mimic in the synthesis of pseudo-oligo saccharides.6 We now report a facile enantioselective preparation
of both enantiomers of 4, which can be carried out on a multigram scale with high-yields, starting from
inexpensive reagents and without intermediate purifications.
Besides classical resolution procedures,2 various syntheses have been reported for the title compound,
and are summarized below.
(i) Syntheses based on Diels–Alder reactions of chiral fumarates. The reaction was pioneered by
Walborsky1,2 and improved by H. Yamamoto3 using dimenthyl fumarate. These reactions are
reported to occur with very high yields and diastereoisomeric excess4 and have been used for
the synthesis of cyclohex-4-ene-1,2-dihydroxymethylene in enantiomerically pure form. However,
they are not useful for the stereoselective synthesis of the diacid, since the dimenthyl ester
extensively epimerizes and racemizes upon basic hydrolysis or under transesterification conditions.5
Alternatively, the diacid can be synthesized following Helmchen’s procedure,7 starting from the
∗
Corresponding author. E-mail: anna.bernardi@unimi.it
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00345-6

3404
A. Bernardi et al. / Tetrahedron: Asymmetry 10 (1999) 3403–3407
diethyl-lactylfumarate and using Ti catalysts. The resulting diester can be hydrolyzed with no loss of
configurational integrity using LiOH in MeOH/H2O.8 However, butadiene is not the best substrate
for Helmchen’s reaction and, on multigram scale, only diastereomeric ratios of up to 6:1 were
obtained.8
(ii) Syntheses based on Diels–Alder reactions using chiral catalysts. Although many catalysts have
been discovered for the enantioselective Diels–Alder reaction, few of them have been reported to
be effective for fumarate cycloadditions9–12 and, to the best of our knowledge, only one has been
described to promote the cycloaddition of butadiene to diethyl fumarate with 60% e.e.9
(iii) Synthesis based on enzymatic desymmetrization of the cis diester. Desymmetrization of the meso
dimethyl cis-cyclohex-4-ene-1,2-dicarboxylate using pig liver esterase (PLE) proceeds with high
yields to give the (+)-(1R,2S) enantiomer of the monoacid in >94% e.e.13 Treatment of this com-
pound with tBuOK results in the regioselective epimerization of C2 and affords the trans-(1R,2R)
monoacid as a 4:1 mixture with the starting cis-stereoisomer.14 Separation of these compounds is
exceedingly difficult, and can only be achieved by gradient column chromatography followed by
fractional distillation.14 The (1S,2S) enantiomer cannot be synthesized directly from this route.†
2. Results and discussion
We now present an improved synthesis of both enantiomers of the title compound, that takes advantage
of the chemical desymmetrization of anhydrides recently introduced by Bolm (Scheme 1).15 Treatment
of the commercially available tetrahydrophthalic anhydride 1 with MeOH at −50°C in the presence of
1.1 mol/equiv. of quinine or its quasienantiomer quinidine results in the highly stereoselective formation
of the (1S,2R) monoacid 2 (e.e. 90%) or the (1R,2S) enantiomer ent-2 (e.e. 93%), respectively. After
extraction of the base with HCl, the crude acids could be used without further purification. (The chiral
base could be quantitatively recovered by precipitation with 3 M NaOH.) The reaction could be carried
out on multigram scale with almost quantitative yield and consistently high stereoselectivity. The e.e.s
were determined by determining the specific rotations of the monoacids 2 and ent-2 and confirmed by
1H NMR of the corresponding ethylmandelate esters in C6D616 (Scheme 2).
Scheme 1. Bolm’s desymmetrization of tetrahydrophthalic anhydride
†
This compound can be prepared starting from the (+)-(1R,2S) enantiomer of the monoacid by orthogonal protection of the
acid (e.g. as a tBu ester), followed by hydrolysis of the methylester at C2, and C1 epimerization with tBuOK: unpublished
results from this laboratory.

A. Bernardi et al. / Tetrahedron: Asymmetry 10 (1999) 3403–3407
3405
Scheme 2. Synthesis of the ethyl mandelates
Starting from 2, the (1S,2S)-trans diacid 4 was obtained by the sequence depicted in Scheme 3.
Thus, tBuOK epimerization gave the 4:1 trans:cis mixture described by Seebach.14‡ Separation was
not attempted at this stage, but the crude was hydrolyzed (LiOH, MeOH) to give the desired diacid
4, contaminated by 20% of the cis isomer. The latter was selectively transformed in the starting
tetrahydrophthalic anhydride 1 by treating the crude hydrolysis product with 0.2 mol/equiv. of Ac2O
in refluxing benzene for 2 h. Under these conditions the trans diacid 4 was left unchanged and could be
isolated in 80% yield from the reaction crude by crystallization from benzene. A second crop of almost
pure trans diacid may be recovered by dissolving the mother liquor in dry dichloroethane and reacting the
anhydride with 1 mol/equiv. of polyethyleneglycol of MW 4600 for 18 h at reflux. This allows recovery
of the remaining diacid 4 after Et2O precipitation of the acylated polyethyleneglycol. Trapping of the
anhydride could also be achieved by using 1.1 equiv. of Wang resin in refluxing dichloroethane.
Scheme 3. Stereoselective synthesis of (1S,2S)-trans-cyclohex-4-ene-1,2-dicarboxylic acid 4
The enantiomeric excess of the diacid was checked at this stage both by determining the specific
rotation and by synthesizing the bis-mandelate ester (Scheme 2),16 and it was found to be 86%.
The whole sequence does not require any intermediate purification, and can be run on a multigram
scale in a matter of days, with >60% total yield starting from the tetrahydrophthalic anhydride.
‡
The reaction must be carried out in THF at room temperature for 1 h. If the reaction conditions are forced (longer time, higher
concentration or temperature), some transesterification also takes place, yielding variable amounts of a mono t-butylester, which
can be hydrolyzed to the diacid by treatment of the crude with trifluoroacetic acid in quantitative yield.

3406
A. Bernardi et al. / Tetrahedron: Asymmetry 10 (1999) 3403–3407
The same sequence can be run starting from the (1R,2S) enantiomer ent-2 of 93% e.e. to give the
(1R,2R)-trans-cyclohex-4-ene-1,2-dicarboxylic acid ent-4 with 88% e.e.
3. Experimental
3.1. General
1H NMR spectra were recorded at 200 MHz and 13C NMR spectra were recorded at 50 MHz. Chemical
shifts are reported in ppm relative to tetramethylsilane; coupling constants are reported in hertz. All
solvents were dried before use: THF using Na/benzophenone; benzene and toluene using Na; CCl4 using
4 Å molecular sieves; CH2Cl2 using CaH2.
3.2. Synthesis of (1S,2R)-cyclohex-4-ene-1,2-carboxylic acid monomethylester 2
To a suspension of tetrahydrophthalic anhydride 1 (5 g, 32.9 mmol, 1 equiv.) and quinine (11.8 g, 36.4
mmol, 1.1 equiv.) in 1:1 toluene:CCl4 (150 ml) MeOH (4 ml, 98.7 mmol, 3 eq) was added dropwise,
under N2 at −50°C. The solution was stirred at this temperature for 24 h, then the solvent was evaporated,
the residue dissolved in AcOEt, and the organic phase extracted with 6N HCl. The organic solvent was
dried with Na2SO4 and evaporated to yield 5.9 g of the cis monomethylester 2 (98%), which was used
without further purification. [α]D −13.7 (c 1.52, EtOH) {lit.14 for the (1R,2S) enantiomer [α]D +14.9
1
(c 1.48, EtOH)}. H NMR (200 MHz, CDCl3): 2.3–2.7 (m, 4H), 3.1 (m, 2H, H1 and H2), 3.7 (s, 3H,
COOMe), 5.7 (m, 2H, H4 and H5). 13C NMR (50 MHz, CDCl3): 25.4, 25.6, 39.3, 39.5, 51.8, 124.9, 125.0,
173.6, 179.6. The e.e. of 2 was determined to be 89% by synthesizing the mandelate ester (Scheme 2)
according to the reported procedure.16 1H NMR (C6D6, 200 MHz): 0.8 (t, 3H, J=7 Hz); 2.0–2.3 (m, 1H);
2.65–2.90 (m, 4H); 3.0–3.1 (m, 1H); 3.32 (s, 3H); 3.70–3.95 (m, 2H); 5.5 (m, 1H); 6.08 (s, 1H).
3.3. Synthesis of (1R,2S)-cyclohex-4-ene-1,2-carboxylic acid monomethyl ester ent-2
The same procedure as above was used; but quinidine (11.8 g, 36.4 mmol, 1.1 equiv.) was used instead
of quinine.
The crude (1R,2S) monomethyl ester had [α]D +13.7 (c 1.59, EtOH) {lit.14[α]D +14.9 (c 1.48, EtOH)}.
The e.e. of ent-2 was determined to be 93% by synthesizing the mandelate ester (Scheme 2) according to
the reported procedure.16 1H NMR (C6D6, 200 MHz): 0.8 (t, 3H, J=7 Hz); 2.0–2.3 (m, 1H); 2.5–2.9 (m,
4H); 3.0–3.1 (m, 1H); 3.30 (s, 3H); 3.70–3.95 (m, 2H); 5.5 (m, 1H); 6.12 (s, 1H).
3.4. cis–trans Equilibration of the monomethyl ester. Synthesis of 3
To a stirred solution of tBuOK (5.24 g, 46.7 mmol, 1.5 equiv.) in dry THF (27 ml), a solution of 2 (5.73
g, 31.1 mmol, 1 equiv.) in THF (67 ml) was added under N2 at 0°C. The solution was stirred at room
temperature for 1 h, then concentrated under vacuum to about 1/3 of the original volume and 6N HCl
was added to pH=1. The aqueous phase was extracted with Et2O, the organic phases dried with Na2SO4
1
and the solvent evaporated, to yield 5.67 g of crude in a 4:1 3:2 mixture, as evaluated by H NMR. 1H
NMR (200 MHz, CDCl3): 2.1–2.7 (m), 2.9 (m, 2H, H1 and H2 of the trans isomer 3), 3.1 (m, 2H, H1 and
H2 of the cis isomer 2), 3.7 (s), 5.7 (m).

A. Bernardi et al. / Tetrahedron: Asymmetry 10 (1999) 3403–3407
3407
3.5. Methylester hydrolysis. Synthesis of 4
To a solution of a 4:1 mixture of the trans and cis monomethyl esters 3 and 2 (1 g, 5.4 mmol, 1 equiv.)
in 3:1 MeOH:H2O (8 ml) LiOH·H2O (863 mg, 20.6 mmol, 3.8 equiv.) was added. The solution was
stirred at room temperature for 2 h, then concentrated under vacuum before adding 6N HCl to pH=1.
The aqueous phase was extracted with AcOEt, the organic solvent dried with Na2SO4 and evaporated to
1
yield 870 mg of 4:1 trans:cis diacids (95%). H NMR (200 MHz, CDCl3): 2.1–2.7 (m); 2.0 (m, 1H of
the trans diacid); 3.0–3.2 (2m, 1H of the cis diacid); 5.7 (m).
3.6. Isolation of 4
To a solution of a 4:1 mixture of the trans and cis diacids (933 mg total, containing 186.6 mg of cis
diacid, 1.09 mmol, 1 equiv.) in dry benzene (10 ml) Ac2O (120 µl, 1.09 mmol, 1 equiv.) was added under
1
N2. The solution was warmed to reflux for ca. 2 h (formation of the cis anhydride is monitored by H
NMR), then the solvent was evaporated and the residue crystallized from benzene (2 ml) to yield 595
1
mg of pure 4 (80%). [α]D +123 (c 2.78, EtOH) {lit.2 [α]D +160 (c 2.7, EtOH.)}; Mp 145–147°C; H
NMR (200 MHz, CDCl3): 2.1–2.7 (m, 4H), 2.85 (m, 2H, H1 and H2), 5.75 (m, 2H, H4 and H5), 9.6 (bs,
2H, COOH). 13C NMR (50.3 MHz, CDCl3): 28.5, 42.1, 125.5, 181.5. IR (CHCl3): 3300–2900, 1714.
Analysis calcd. for C8H10O4: C 56.47, H 5.92; found: C 56.25, H 6.09.
The enantiomeric excess of 4 was determined to be 85% by synthesizing the bis-mandelate ester
(Scheme 2) according to the reported procedure.16 1H NMR (C6D6, 200 MHz): 2.18–2.7 (m, 4H),
3.0–3.25 (m, 8H), 5.4 (m, 2H, H4 and H5), 6.08 (s, 1H, benzylic proton of the (1R,2R) diester), 6.11
(s, 1H, benzylic proton of the (1S,2S) diester), 7.0–7.5 (m, 5H).
Acknowledgements
A fellowship to D.D. from GlaxoWellcome is gratefully acknowledged.
References
1. Walborsky, H. M.; Barash, L.; Davis, T. C. J. Org. Chem. 1961, 26, 4778–4779.
2. Walborsky, H. M.; Barash, L.; Davis, T. C. Tetrahedron 1963, 19, 2333–2351.
3. Furuta, K.; Iwanaga, K.; Yamamoto, H. Tetrahedron Lett. 1986, 27, 4507–4510.
4. Solladie, G.; Lohse, O. J. Org. Chem. 1993, 58, 4555–4563.
5. Corey, E. J.; Su, W. Tetrahedron Lett. 1988, 29, 3423–3426.
6. Bernardi, A.; Checchia, A.; Brocca, P.; Sonnino, S.; Zuccotto, F. J. Am. Chem. Soc. 1999, 121, 2032–2036.
7. Hartmann, H.; Hady, A. F. A.; Sartor, K.; Weetman, J.; Helmchen, G. Angew. Chem., Int. Ed. Engl. 1987, 26, 1143–1145.
8. Bernardi, A.; Boschin, G.; Checchia, A.; Lattanzio, M.; Manzoni, L.; Potenza, D.; Scolastico, C. Eur. J. Org. Chem. 1999,
1311–1317.
9. Devine, P. N.; Oh, T. J. Org. Chem. 1992, 57, 396–399.
10. Evans, D. A.; Lectka, T.; Miller, S. J.; Tetrahedron Lett. 1993, 34, 7027–7030.
11. Evans, D. A.; Miller, S. J.; Lectka, T. J. Am. Chem. Soc. 1993, 115, 6460–6461.
12. Hawkins, J.; Loren, S. J. Am. Chem. Soc. 1991, 113, 7794–7795.
13. Schneider, M.; Engel, N.; Hoenicke, P.; Heinemann G.; Goerisch, H. Angew. Chem., Int. Ed. Engl. 1984, 23, 67–68.
14. Ito, Y. N.; Ariza, X.; Beck, A. K.; Bohac, A.; Ganter, C.; Gawley, R. E.; Kuehnle, F. N.; Tuleja, J.; Wang, Y.-M.; Seebach,
D. Helv. Chim. Acta 1994, 77, 2071–2110.
15. Bolm, C.; Gerlach, A.; Dinter, C. L. Synlett 1999, 195–196.
16. Parker, D. J. Chem. Soc., Perkin Trans. 2 1983, 83–88.