Process research of (R)-cyclohexyl lactic acid and related building blocks: A comparative study

Article abstract of DOI:10.1021/op030202q

(S)-Cyclohexyl lactic acid is a component of the selective E-selectin inhibitor 2 ((S)-cHexLact-2-0-(3-Galβ(1→3)ddGlc-(4→1)αFuc). We describe the evaluation of various synthetic routes to this building block: (A) diazotation of phenylalanine followed by phenyl ring hydrogenation; (B) phenyl ring hydrogenation of phenyl alanine followed by diazotation; (C) acidic hydrolysis of the cyanohydrin derived from phenylacetaldehyde, enantiomeric resolution of the resulting, racemic phenyl lactic acid via diasteromeric salt formation and phenyl ring hydrogenation; (D) enantioselective dihydroxylation of a cinnamate ester, followed by hydrogenation of the benzylic hydroxy group and the aromatic nucleus; (E) enantioselective biocatalytic reduction of phenylpyruvic acid, followed by phenyl ring hydrogenation. The development of (2R)-2-0-(4-nitrophenyl)sulfonyl-cyclohexyl lactic acid p-bromobenzylester 21 as a buidling block with improved crystallinity and stability is also described.

Full text of DOI:10.1021/op030202q

Organic Process Research & Development 2003, 7, 559−570
Process Research of (R)-Cyclohexyl Lactic Acid and Related Building Blocks:
A Comparative Study‡
Thomas Storz* and Peter Dittmar
Process Research and DeVelopment, Chemical and Analytical DeVelopment; NoVartis Pharma AG,
CH-4002 Basel, Switzerland
Pierre Franc¸ois Fauquex and Philippe Marschal
Biotechnology DeVelopment, NoVartis Pharma AG, CH-4002 Basel, Switzerland
Willy Urs Lottenbach and Heinz Steiner
Hydrogenation and High-Pressure SerVice, SolVias AG, CH-4002 Basel, Switzerland
Abstract:
leukocyte surface via interaction with the common tetrasac-
charide epitope sialyl Lewisx (sLex, 1).5 Analogues of sialyl
Lewisx have therefore been proposed as potential therapeutics
for a host of inflammatory disorders (e.g., ischemia/reper-
fusion injury following organ transplantation).6 The sLex
analogue 2 (Figure 1) was discovered by Thoma et al. at
Novartis Pharmaceuticals Corporation.7 Compared to sLex,
2 showed a 30-fold improved affinity in a cell-free E-selectin-
ligand binding assay.8
As a simplified analogue of SLex, 2 features the (S)-
cyclohexyl lactic acid moiety as a substitute9 for the
structurally more complex N-acetylneuraminic acid residue.
The former is introduced into the trisaccharide precursor of
2 via nucleophilic displacement reaction of a 2-O-sulfonyl
substituted (R)-cyclohexyl lactic acid building block. Enan-
tiomerically pure (R)-cyclohexyl lactic acid10 is not com-
mercially available. Hence, an efficient synthetic access to
this hydroxy acid needed to be elaborated.
(S)-Cyclohexyl lactic acid is a component of the selective
E-selectin inhibitor 2 ((S)-cHexLact-2-O-(3-Galâ(1f3)ddGlc-
(4f1)rFuc). We describe the evaluation of various synthetic
routes to this building block: (A) diazotation of phenylalanine
followed by phenyl ring hydrogenation; (B) phenyl ring
hydrogenation of phenyl alanine followed by diazotation; (C)
acidic hydrolysis of the cyanohydrin derived from phenylac-
etaldehyde, enantiomeric resolution of the resulting, racemic
phenyl lactic acid via diasteromeric salt formation and phenyl
ring hydrogenation; (D) enantioselective dihydroxylation of a
cinnamate ester, followed by hydrogenation of the benzylic
hydroxy group and the aromatic nucleus; (E) enantioselective
biocatalytic reduction of phenylpyruvic acid, followed by phenyl
ring hydrogenation. The development of (2R)-2-O-(4-nitrophe-
nyl)sulfonyl-cyclohexyl lactic acid p-bromobenzylester 21 as a
buidling block with improved crystallinity and stability is also
described.
Results and Discussion
(A) The Medicinal Chemistry Synthesis. The synthesis
(Scheme 1)7 of the (R)-cyclohexyl lactic acid building
block 6 started from commercially available (R)-3-phenyl
lactic acid10b (3). Hydrogenation of the aromatic nucleus,
followed by benzyl ester formation with CsCO3/BnBr and
O-sulfonylation with triflic anhydride, gave the building
block 6 in three steps and 60% total yield after two
chromatographies. The triflate 6 proved to be an unstable
oil at room temperature and resisted all attempts to crystal-
lization.
Introduction
Excessive leukocyte influx from blood vessels into the
surrounding tissues has been linked to acute or chronic
reactions as observed in reperfusion injuries, psoriasis, stroke,
rheumatoid arthritis or respiratory diseases.1-3 Selectin-
dependent leukocyte adhesion is the first step in the cascade
of events that leads to the extravasation of leukocytes.4 In
particular, E-Selectin on the surface of the endothelial blood
vessel walls recognizes complex glycoprotein ligands on the
‡ Dedicated to Professor Andrea Vasella on the occasion of his 60th birthday.
* To whom correspondence should be addressed. Current address: Amgen
Inc., One Amgen Center Drive, Thousand Oaks, CA 91320-1799, U.S.A.
E-mail: tstorz@amgen.com.
(5) Kansas, G. S. Blood 1996, 88, 3259.
(6) For a review, see: Simanek, E. E.; McGarvey, G. J.; Jablonowsky, J. A.;
Wong, C.-H. Chem. ReV. 1998, 98, 833.
(7) Thoma, G.; Kinzy, W.; Bruns, C.; Patton, J. T.; Magnani, J. L.; Ba¨nteli, R.
J. Med. Chem. 1999, 42, 4909.
(8) Ba¨nteli, R.; Herold, P.; Bruns, C.; Patton, J. T.; Magnani, J. L.; Thoma, G.
HelV. Chim. Acta 2000, 83, 2893.
(9) Kolb, H. C.; Ernst, B. Chem. Eur. J. 1997, 1571.
(10) (a) v. Braun, J.; Nelles, J. Chem. Ber. 1933, 66, 1464 (racemate). (b) Bajusz,
S.; Baraba´s, E.; Fauszt, I.; Fehe´r, A.; Horva´th, Gy.; Juha´sz, A.; Szabo´, A.
G.; Sze´ll, E. Bioorg. Med. Chem. 1995, 3, 1079 (R-enantiomer).
(1) Mousa, S. A. Drugs Fut. 1996, 21, 283.
(2) Mousa, S. A.; Cheresh, D. A. Drug DiscoVery Today 1997, 2, 187.
(3) Cines, D. B.; Pollak, E. S.; Buck, C. A.; Loscalzo, J.; Zimmermann, G. A.;
McEver, R. P.; Pober, J. S.; Wick, T. M.; Konkle, B. A.; Schwartz, B. S.;
Barnathan, E. S.; McCrae, K. R.; Hug, B. A.; Schmidt, A.-M.; Stern, D.
M. Blood 1998, 91, 3527.
(4) Spertini, O.; Luscinskas, F. W.; Gimbrone, M. A.; Tedder, T. F. J. Exp.
Med. 1992, 175, 1789.
10.1021/op030202q CCC: $25.00 © 2003 American Chemical Society
Published on Web 06/27/2003
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
559

Figure 1.
Table 1. Aromatic ring hydrogenation of phenyl lactic acid: catalyst screening
(concentration, mmol)
batch size (%)
catalyst,
solvent (%),
reaction time, h
entry
1
H2 pressure
temperature (°C)
yield (%)/purity (%)
12.0
(5)
5% Rh/C
(10 wt %/wt)
15 bar
EtOH (94)
50
2
92.9 (GLC)
conversion: 99.8
2
3
4
5
300.9
(5)
5% Rh/C
(10 wt %/wt)
15 bar
10% Ru/C
(15 wt %/wt)
15 bar
5% Rh/C
(10 wt %/wt)
15 bar
10% Ru/C
(10 wt %/wt)
30 bar
EtOH (94)
50
5
99.6, crude
12.0
(5)
AcOH
50-70-100
59 h
89.4% (GLC)
conversion: 99.9%
12.0
(14.3)
AcOH/H2O
(8:6) 50-80
108
70.6 (GLC)
conversion: 99.9
12.0
(13.5)
1 M NaOH
60
16
7.0 (GLC)
conversion: 13.2
Scheme 1. Medicinal chemistry synthesis
nucleus of the resulting phenyl lactic acid over a more
economic catalyst: Although literature reports11-14 had
claimed that diazotation of (R)-phenyl alanine to (R)-phenyl
lactic acid 3 proceeds under complete retention of config-
uration (only optical rotation values were cited as proof of
enantiomeric purity), we invariably observed a loss of
enantiomeric purity as unequivocally determined by HPLC
analysis on a chiral column (2-5%, see Experimental
Section). The hydrogenation of the aromatic nucleus in 3
was investigated under various conditions (Table 1). In
general, none, or only insignificant loss (<0.1%) of enan-
tiomeric purity was observed (see typical result in Experi-
mental Section).
(B) Process R&D. Surprisingly, there is only one
literature report for the production of (R)-cyclohexyl lactic
acid (4): aromatic ring hydrogenation of D-(+)-3-phenyl
lactic acid (3) over a platinum oxide catalyst.10b Due to the
high price and limited availability of 3, and the unfavorable
physicochemical properties combined with the instability of
the triflate 6, we aimed to find a more economic access to
4 and, ultimately, also to replace the triflate 6 by a crystalline,
more stable activated building block.
In our first strategy, (R)-phenyl alanine (7) was chosen
as starting material. Two variants were investigated:
(A) diazotation of phenylalanine under retention of
configuration,11 followed by hydrogenation of the aromatic
The best overall results of approach A are summarized
in Scheme 2.
(B) Hydrogenation of the aromatic nucleus of phenyl-
alanine,15 followed by diazotation of the resulting cyclohexyl
alanine 8 under retention of configuration: The hydrogena-
tion of phenylalanine showed less variability than the
corresponding hydrogenation of phenyl lactic acid (Table 2).
(12) Yamada, S.-I.; Koga, K.; Juang, T. M.; Achiwa, K. Chem. Lett. 1976, 927.
(13) Terashima, S.; Tseng, C. C.; Koga, K. Chem. Pharm. Bull. 1979, 27, 747.
(14) Urban, F. J.; Moore, B. S. J. Heterocycl. Chem. 1992, 29, 431.
(15) Bain, J. D.; Wacker, D. A.; Kuo, E. E.; Chamberlin, A. R. Tetrahedron
1991, 47, 2389.
(11) Cohen, S. G.; Weinstein, S. Y. J. Am. Chem. Soc. 1964, 86, 5326.
560
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development

Table 2. Aromatic ring hydrogenation of phenyl alanine: catalyst screening
concentration (mmol),
batch size (%)
catalyst,
solvent,
reaction time,
yield/purity
entry
1
H2 pressure
temperature (°C)
12.1
6.7
5% Rh/C
(10 wt %/wt)
4 bar
1 M HCl in H2O
50
3.25 h
96.0% (GLC)
conversion: 99.9%
2
3
4
5
6
7
12.1
5
5% Rh/C
(10 wt %/wt)
15 bar
10% Ru/C
(10 wt %/wt)
15 bar
10% Ru/C
(10 wt %/wt)
15 bar
5% Rh/C
(10 wt %/wt)
5 bar
5% Rh/C
(10 wt %/wt)
5 bar
10% Ru/C
2.6 wt %/wt)
30 bar
EtOH (94 5%)
43 h
50-100
94.1% (GLC)
conversion: 99.5%
12.1
5
AcOH
50
19 h
no conversion (TLC)
12.1
6.7
1 M NaOH
60
3 h
90.1% (GLC)
conversion: 99.7%
12.1
14.3
AcOH/H2O
(4:3)
50
AcOH/H2O
(4:3)
50
17 h
96.9% (GLC)
conversion: >99%
302.7
14.3
7 h
100% crude
12.1
13.3
1 M NaOH
60
7 h
91.5% (GLC)
conversion: >99.9%
Scheme 2. Diazotization strategies: approach A
ously difficult substrates and, in general, tend to give lower
ee16f,17,18 than their aromatic counterparts. A kinetic resolution
of racemic phenylacetaldehyde-cyanohydrin on an analytical
scale (no product isolated, no yields given) with a mutant
Pseudomonas strain to give enantiomerically enriched (ca.
75% ee) (S)-(-)-phenyl lactic acid has been reported.19 One
study reporting the synthesis of enantiomerically enriched
(88% ee) (R)-phenylacetaldehyde-cyanohydrin in 83% yield
on a millimolar scale via oxynitrilase-catalyzed transcyana-
tion with acetone cyanohydrin has appeared.20 However, due
to the high substrate dilution, high enzyme concentration (ca.
1000 units/mmol substrate) and the requirement for a special
enzyme purification of the commercial enzyme prior to use,
this method appeared not practical for preparative scale
synthesis.
Scheme 3. Diazotization strategies: approach B
On the other hand, we obtained racemic phenyl lactic acid
((-PLA) 11 without need for the isolation of the intermediate
1021 via cyanohydrin reaction of the bisulfite adduct22 of 9
However, the subsequent diazotation of 8 proceeded only
in very poor yields. Hence, approach B, as summarized in
Scheme 3, was quickly abandoned.
(C) Cyanohydrin reaction of phenylacetaldehyde, followed
by acidic hydrolysis to (R/S)-phenyl lactic acid, resolution
of the racemic acid via diastereomeric salt formation and
hydrogenolytic reduction of the phenyl ring as in approach
(A):
(16) For reviews, see: (a) Groger, H. AdV. Synth. Catal. 2001, 343, 547. (b)
Van Der Gen, A.; Brussee, J. NATO Sci. Ser. 1: Disarm. Technol. 2000,
33, 365. (c) Gregory, R. J. H. Chem. ReV. 1999, 99, 3649. (d) Effenberger,
F. Chimia 1999, 53, 3. (e) Johnson, D. V.; Griengl, H. Chim. Oggi 1997,
15, 9. (f) North, M. Synlett 1993, 807. (g) Kruse, C. G. In Chirality in
Industry; Collins, A. N., Sheldrake, G. N., Crosby, J., Eds.; Wiley: New
York, 1992; Chapter 14, p 279.
(17) Zandbergen, P.; Van Der Linden, J.; Brussee, J.; Van Der Gen, A. Synth.
Commun. 1991, 21, 1387.
(18) Ziegler, T.; Ho¨rsch, B.; Effenberger, F. Synthesis 1990, 575.
(19) Hashimoto, Y.; Kobayashi, E.; Endo, T.; Nishiyama, M.; Horinouchi, S.
Biosci. Biotechnol. Biochem. 1996, 60, 1279.
Although much progress has recently been achieved with
enantioselective cyanohydrin reactions,16 R-methylene-alde-
hydes, such as arylacetaldehydes, have proved to be notori-
(20) Ognyanov, V. I.; Datcheva, V. K.; Kyler, K. S. J. Am. Chem. Soc. 1991,
113, 6992.
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
561

Scheme 4. Approach C: resolution of racemic PLA
was prone to acid-catalyzed transesterification, as the
intermediate 3-phenyl lactic acid was isolated as the methyl
ester after the reaction (Scheme 7).
Because of this finding and the unfavorable physical
properties of the benzyl ester 13B in combination with the
lower ee in the dihydroxylation reaction of benzyl cinnamate,
it was decided to focus on the crystalline methyl ester 13A.
Initially, we isolated the intermediate phenyl lactic ester 14
before starting a fresh hydrogenolysis experiment in a
separate vessel with the second catalyst. It was then
discovered that equivalent results were also obtained by
sequential addition of the two catalysts in one pot in the same
solvent (Scheme 8). This represents a significant progress
over previous methodology for this transformation.24
In the sequence from the dihydroxylation product 13A
to (R)-cyclohexyl lactic acid 4, the enantiomeric purity
changes only very slightly. The initial osmium impurity in
13A (200 ppm) dropped to a mere 4 ppm in the final product
4 without any special purification measures. Overall, ap-
proach D delivered a very practical and readily scaleable
process based on a cheap and abundant starting material
(methyl cinnamate).
followed by acid-catalyzed hydrolysis in 71% yield after
recrystallization of the crude product from toluene (Scheme
4). A variety of readily available optically pure bases were
then screened for their potential to form diastereomeric salts23
(Table 3).
Resolution with the most promising bases, (+)-dehydro-
abietylamine, (S)-phenylglycinol and (S)-phenylalaninol was
optimized (data not shown); typical results are shown in
Table 4.
Especially the resolution via the diastereomeric (S)-salt
separation with (+)dehydroabietylamine looked very promis-
ing, due to the very low cost of the chiral amine. Resolutions
with (S)-phenylglycinol and (S)-phenylalaninol also worked
very well, but required recycling of the (expensive) chiral
base.
Finally, we also investigated the enantioselective biocata-
lytic reduction of phenylpyruvic acid 29 as an alternative
access to (R)-3-phenyl lactic acid 3. The enantioselective
reduction of 2-oxo carboxylic acids to the corresponding (R)-
2-hydroxy carboxylic acids by resting cells of Proteus
mirabilis or Proteus Vulgaris was reported by H. Simon and
collaborators.27 The stereospecificity is extremely high (ee
> 98%)28 and the substrate specificity very broad. Proteus
possesses a formate dehydrogenase as well as a hydrogenase,
in addition to its specific 2-oxocarboxylate reductase activity.
Therefore, the basic principle of the reaction involves the
use of formate or hydrogen gas as reducing agents for the
2-oxocarboxylic acids and, of a catalytic amount of viologen
(V2+), for example benzyl viologen or methyl viologen, as
electron carrier:
(D) Enantioselective dihydroxylation of methyl cinnamate,
followed by selective hydrogenolytic removal of the benzylic
hydroxy group and hydrogenation of the aromatic nucleus.24
Cinnamate esters are very cheap starting materials (Me,
Et, Bn: $8-15/kg), and readily scaleable protocols for their
enantioselective dihydroxylation to R,â-dihydroxy-phenyl-
propionates in good yield and enantiomeric purity are well
established.25 We aimed at an efficient and selective one pot-
reduction process, employing only a noble metal catalyst and
hydrogen as the sole reducing agent (Scheme 5).
The methyl and benzyl esters of the corresponding (2R,-
3S)-3-phenyl-2,3-dihydroxypropionate26 were synthesized as
starting materials. The methyl ester is easily purified by
recrystallization and can be obtained in very high enantio-
meric purity, whereas the benzyl ester has to be purified by
chromatography and, in our hands, was routinely obtained
in inferior optical purity (Scheme 6).
HCOO- + 2 V++ f H+ + CO2 + 2 V•+
or: H2 + 2 V++ f 2 H+ + 2 V•+
The subsequent hydrogenolytic removal of the benzylic
hydroxy group proceeded in good yield and without signifi-
cant loss in optical purity, but the benzyl ester apparently
R-CO-COO- + 2 V•+ + 2 H+ f
(R)-R-CHOH-COO- + 2 V++
(21) (a) Erlenmeyer, E.; Lipp, A. Ann. Chem. 1883, 219, 187. (b) Ruggli, P.;
Hegedues, B. HelV. Chim. Acta 1942, 25, 1285.
(22) (a) Biquard, D. Ann. Chim. (Paris) 1933, 10, 97. (b) Meerpoel, L.; Hoornaert,
G. Synthesis 1990, 905. (c) Chesters, N. C. J. E.; O’Hagan, D.; Robins, R.
J. J. Chem. Soc., Perkin Trans. 1 1994, 1159.
(23) The only diastereomeric salt resolution for (D)-(+)-3-phenyl lactic acid we
are aware of (base: morphine!) was reported by A. McKenzie and H. Wren
(J. Chem. Soc. 1910, 97, 1358), who later also described the resolution of
racemic PLA via its diastereomeric (-)-menthol esters (J. Chem. Soc. 1921,
119, 801).
The simplest procedure for the reduction of 2-oxo carboxylic
acids is the stirring of the Proteus cells together with the
substrate, benzyl viologen and formate in phosphate buffer
pH 7, at 37 °C under an atmosphere of nitrogen, in a batch-
mode process.29 When oxidised, one molecule of formate
delivers two electrons but only one proton. To provide the
additional proton necessary for the substrate reduction and
to maintain a constante formate concentration, formic acid
is continuously added, using an automatic pH control
system.The optimal reaction time to obtain a 2-oxo carboxy-
(24) To the best of our knowledge, only two selective reductions of R,â-
dihydroxy-dihydro-cinnnamic esters to 3-aryl lactic acid esters have been
reported: (a) Nakajima, M.; Tomioka, K. and Koga, K. Tetrahedron 1993,
49, 10807 (Et3SiH/TFA); (b) Rho, H. S.; Ko, B.-S. Synth. Commun. 2001,
31, 283 (Mg-reduction of the corresponding thionocarbonate derivative).
A Mn-catalyzed oxidation of cinnamates in the presence of triphenyl silane
also leads to 3-aryl lactic acid esters: Tanaka, M.; Mukaiyama, C.;
Mitsuhashi, H.; Maruno, M.; Wakamatsu, T. J. Org. Chem. 1995, 60, 4339.
(25) For a review, see: Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B.
Chem. ReV. 1994, 94, 2483.
(27) Simon, H.; Bader, J.; Gu¨nther, H.; Neumann, S.; Thanos, J. Angew. Chem.,
Int. Ed. Engl. 1985, 24, 539-553.
(28) Gu¨nther, H.; Neumann, S.; Simon, H. J. Biotechnol. 1987, 5, 53-65.
(29) Schummer, A.; Yu, H.; Simon, H. Tetrahedron 1991, 47, 9019-9034.
(26) Wang, Z. M.; Kolb, H. C.; Sharpless, K. B. J. Org. Chem. 1994, 59, 5104.
562
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development

Table 3. Resolution of racemic PLA: base screening
result yield
(%)
R/S recryst.
entry scale (mmol)
base
(R)-1-phenyl-ethylamine
cinchonine
cinchonidine
quinine
quinidine
(-)-ephedrine
solvent
R/S crude
(yield, %)
1
2
15
6
6
6
6
6
6
6
6
6
6
6
6
6
6
EtOAc/iPrOH 4:1 (oil)
n.d.
n.d.
51:49
EtOAc/iPrOH
EtOAc/iPrOH
EtOAc/iPrOH
EtOAc/iPrOH
EtOAc
EtOAc
EtOAc
EtOAc
EtOAc
(oil)
crystals 61
3
4
5
6
7
8
9
n.d.
(oil)
(oil)
(oil)
crystals, >100
crystals, >100
crystals, >100
crystals, >100
(oil)
n.d.
n.d.
n.d.
56:44
52:48
50:50
50:50
n.d.
(+)-pseudoephredine
L-phenylglycinol
L-phenylalaninol
L-threo-2-amino-1-phenyl-1,3-propandiol
70:30 (67)
89:11 (73)
n.d.
10
11
12
13
14
15
D-threo-2-Amino-1-nitrophenyl-1,3-propandiol EtOAc
n.d.
(R)-2-amino-1-butanol
N-methyl-D-glucamine
(-)-spartein
EtOAc
iPrOH/MeOH
EtOAc
(oil)
(oil)
crystals, >100
n.d.
n.d.
33:67
(+)-dehydro-abietylamine
EtOAc
14:86 (87)
Table 4. Resolution of racemic PLA by diastereomeric salt formation: optimization
scale
entry (mmol)
yielda of
diast. salt (%)
yielda of recovered
acid (%) (R/S)
base
price of
base
solvent
recovery (%) base ($/kg)
A
B
C
24
24
24
L-phenyl-glycinol
L-phenyl-alaninol
(+)-dehydro-abietylamine 4-methyl-2-pentanone
sec-butanol
2-butanone
93.8 (R)
87.6 (R)
n.d. (S)b
93.5 (99:1)
79
99
100
995
725
20
52.2 (100: 0)
73.0 (80:20)
after recryst.:
49.0 (100: 0)
a Based on (R)-enantiomer. b (S)-Salt precipitates out, (R)-acid isolated from mother liquor.
Scheme 5. Approach D: asymmetric dihydroxylation/
Scheme 6. Approach D: starting materials
selective reduction
After optimization of the reaction conditions and of the
extraction procedure, we applied this process successfully
to the reduction of phenylpyruvic acid (approach E): the
product (R)-3-phenyl lactic acid was obtained with a conver-
sion of >99.8%, a total yield (bioconversion and recovery)
of 87% and an enantiomeric purity (R-form) of >99.5%
(Scheme 9).
lic acid conversion of about 100% is in the range of 24 h.
After removal of the cells by filtration or by centrifugation,
the (R)-2-hydroxy carboxylic acid produced can be recovered
by extraction in ethyl acetate and crystallization by addition
of methylcyclohexane or cyclohexane.30
(30) Fauquex, P.; Sedelmeier, G. EP 0371408, 1989.
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
563

Scheme 7. Approach D: hydrogenation of benzylic
Scheme 10. Ester screening for (R)-cyclohexyl lactic acid
hydroxy group
Table 6. Evaluation of esters of (R)-cyclohexyl lactic acid
physical
state
melting
yield (%)
yield (%)
(cryst.)
ester
R
point (°C) (chromat.)
5
16
17
18
19
20
benzyl
solid
oil
oil
solid
oil
solid
54-56
-
87
50
93
88
37
94
70
-
4-chlorobenzyl
2,5-dichlorobenzyl
4-bromobenzyl
4-nitrobenzyl
-
-
50-52
-
54-55
82
-
73
2-naphthyl-methyl
Scheme 8. Approach D: methyl ester variant
However, both A and E featured relatively expensive starting
materials. On the other hand, with options C and D, two
more sustainable processes were developed, delivering
material of acceptable optical purity from cheap starting
materials. Overall, process D was found to be the most
attractive option for technical scale manufacture of (R)-
cyclohexyl lactic acid.
We then initiated a screening process for crystalline esters
of (R)-cyclohexyl lactic acid. Esters were readily accessible
by base-catalyzed, selective O-alkylation with the corre-
sponding benzyl halide. The cesium carbonate/DMF-alkyl-
ation procedure used in the research synthesis could be
replaced by a more convenient (no aqueous workup required)
alkylation protocol (anhydrous potassium carbonate in
acetone, cat. TBAI) without appreciable loss in yield
(Scheme 10). Also, no epimerization was observed under
these reaction conditions, as proven by determination of
enantiomeric purity in the p-bromobenzylester 18 by HPLC
on a chiral stationary phase (see Experimental Section).
The best candidates, on basis of their crystalline properties
and ease of preparation, were the 4-bromobenzyl (18) and
the 2-naphthylmethyl (20) esters, which were readily acces-
sible on a preparative scale (no chromatography required)
(Table 6). Several O-sulfonyl derivatives of these two esterss
and also of the benzyl ester 5swere synthesized. Their
syntheses and properties are summarized in Table 7. The
nosylate of the 4-bromobenzyl ester (21) appeared especially
attractive. In preliminary experiments with model nucleo-
philes (data not shown), sulfonate 21 displayed electrophilic
reactivity comparable to the triflate derivative 6 used in the
research synthesis while featuring good crystallinity and
improved stability.
Scheme 9. Approach E: biocatalytic reduction
Table 5. Comparison of different syntheses for
(R)-3-cyclohexyl lactic acid
starting
material
(price)
number
of
steps
yield optical purity
(%) R/S
synthesis
(A)
(C)
D-phenylalanine
2
58-75 95:5-98:2
17-32 99:1-100:0
(∼165 $/kg)
phenylacetaldehyde 5
(∼30 $/kg)
(3 steps
“one-pot”)
Conclusions
(D)
(E)
methyl cinnamate
4
68-72 g99.3:0.7
80-85 99.5:0.5
(∼10 $/kg)
(2 steps
In summary, four robust processes amenable for large-
scale, chromatography-free preparation of (R)-3-cyclohexyl
lactic acid were developed from four different starting
materials. Asymmetric dihydroxylation of methyl cinnamate,
followed by a one-pot catalytic reduction of the benzylic
hydroxy group and the aromatic nucleus, and subsequent
saponification of the methyl ester proved to be the most
sustainable process with regard to price and stability of
“one pot”)
phenylpyruvic acid 2
(≈ 880 $/kg)
Table 5 compares the four processes: the shortest
syntheses investigated were processes A and E, with the latter
being much more reliable in terms of the enantiomeric purity.
564
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development

Table 7. Evaluation of different sulfonates
ester
R1
R2
melting point
yield (%) (chromat.)
yield (%) (cryst.)
21
22
23
24
25
26
27
28
4-brom-benzyl
4-brom-benzyl
4-brom-benzyl
2-naphthyl-methyl
2-naphthyl-methyl
benzyl
4-nitrophenyl-
methyl-
4-bromophenyl-
methyl-
4-nitrophenyl-
4-nitrophenyl-
4-bromophenyl-
4-methylphenyl
71-72 °C
-
81
48
79
58
80
63
60
60
-
52
-
-
-
-
-
oil
68-69 °C
oil
oil
oil
oil
oil
benzyl
benzyl
starting material, total yield and optical purity of the product
(Table 5). As an alternative to the unstable and oily triflate
6, an activated building block with improved stability and
crystallinity (p-bromobenzylester nosylate 21) was found.
under reduced pressure until crystallization started to set in
(approximately one-third of the volume), diluted with hexane
(300 mL) and allowed to crystallize at 4 °C overnight, which
gave a second crop (33 g, 9.4%) of crystals with comparable
purity. Total yield of 3: 211.8 g (60.2%) of a colorless
crystalline solid. TLC: Rf ) 0.45 (toluene/EtOAc/CH2Cl2/
HCOOH ) 24:40:40:4). MS (EI): m/z 166 (M+), 148 (M+
- H2O), 121, 103, 91, 77, 65.
Experimental Section
Starting materials, reagents and solvents were obtained
from commercial suppliers and were used without further
purification. All the melting points are uncorrected and
Enantiomeric purity was determined by HPLC on a chiral
column (Chiralpak AD, 25 mm × 4.6 mm, eluent: n-hexane/
i-PrOH/TFA 96:4:0.2, flow: 0.5 mL/min, λ ) 210 nm, 40
°C; [R]- and [S]-enantiomers from Fluka as analytical
standards): R/S ) 97.4:2.6. A smaller batch (25-g scale) gave
78% total yield, enantiomeric purity R/S ) 95.5:4.6.
(C) (By Resolution of Racemic PLA via Diastereomeric
Salt Separation). A suspension of phenylacetaldehyde (50.0
g, 416.1 mmol) in 228 mL of commercial (38%) aqueous
sodium bisulfite solution (832.3 mmol) was stirred in an
icebath at 0 °C for 30 min before a solution of potassium
cyanide (108.39 g, 1.665 mol) in 200 mL of water was slowly
added over 45 min. The reaction mixture was stirred at 5-10
°C for additional 60 min. Water (100 mL) was added, and
the mixture was extracted with TBME (2 × 250 mL).
Organic phases were dried over sodium sulfate and, after
addition of 1 drop of concentrated H2SO4, were evaporated
under reduced pressure. To the resulting 10, a yellow oil,
was added concentrated aqueous HCl (300 mL), and the
mixture was heated under reflux (95 °C) for 60 min. The
mixture was allowed to cool to room temperature overnight,
and the clear, yellow solution was extracted with TBME (8
× 300 mL) and ethyl acetate (300 mL). Organic phases were
dried over sodium sulfate and evaporated under reduced
pressure. The crude 11 (65.54 g, 95%), a brown oil, was
dissolved in toluene (180 mL) and heated to reflux. Upon
cooling, a colorless precipitate formed, which was filtered
and dried in a vacuum oven (50 °C). Yield 48.44 g (70.1%
based on phenylacetaldehyde) racemic phenyl lactic acid
(PLA) 11 as colorless crystals.
1
determined on a Buchi apparatus. H NMR spectra were
recorded at 400 MHz, and 13C NMR spectra were recorded
at 100 or 125 MHz on a Bruker DPX 400/500 instrument.
IR spectra were measured on a Bruker IFS660 spectrometer.
The enantiopurity of 3, 4, 5, 13 and 14 was determined on
a Hewlett-Packard Series 1100 HPLC system using a
Chiralcel OD or Chiralpak AD column.
(2R)-2-Hydroxy-3-phenylpropionic Acid (3) (A) (by
Diazotation of D-Phenylalanine, Improved Variant of the
Procedure of Cohen and Weinstein11). A 10-L Buchi
reactor was charged with a solution of D-phenylalanine (7)
(350 g, 2.12 mol, R/S g 99.5:0.5) in 3245 mL of 0.5 M
H2SO4. At 5 °C, a solution of sodium nitrite (219.3 g, 3.18
mol) was added over a 3-h period. The reaction temperature
was kept between 0 and 10 °C. After the addition was
complete, stirring was continued, and the reaction mixture
was allowed to reach room temperature overnight. The
resulting suspension was cooled to 5 °C, and sulfamic acid
(5.2 g, 53.5 mmol) was added portionwise under vigorous
stirring until iodine-cadmium reagent paper showed no more
nitrite. Upon addition of isopropyl acetate (3000 mL), a clear
biphasic mixture formed. The organic phase was separated
and washed with water (1000 mL) and 20% aqueous NaCl
(750 mL). Collected water phases were extracted with
isopropyl acetate (2 × 500 mL) and the combined organic
phases dried over MgSO4 and reduced to about 20% of their
volume under reduced pressure, just until crystallization
started to set in. Hexane was then added, and the suspension
was cooled to 0 °C and filtered. The filter cake was washed
with 250 mL of cold hexane/isopropyl acetate (1:1-mixture)
and 300 mL of cold hexane and dried under reduced pressure
at 40 °C (178.8 g, 50.8%). The filtrate was again evaporated
(C1): Resolution of Racemic PLA with L-(+)-Phenyl-
glycinol. 11 (4.0 g, 24.1 mmol) and L-(+)-phenylglycinol
(3.63 g, 26.5 mmol) were dissolved in 2-butanol (80 mL) at
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
565

80 °C. The clear solution was heated to 115 °C and slowly
cooled to room temperature. At 70 °C, TBME (80 mL) was
added. At 50 °C, a few seed crystals of recrystallized (R)-
salt 12A from a previous batch were added. The resulting,
thick suspension was stirred in an ice bath for several hours
and filtered. The filter cake was washed with ice cold TBME
(20 mL) and dried in a vacuum oven (35 °C). The crude
salt 12A was dissolved in methylene chloride (40 mL) at
room temperature. Aqueous 1 M sodium hydroxide (40 mL)
was added, and after stirring vigorously for 5 min, the
aqueous phase was separated and extracted with methylene
chloride (40 mL) before adjusting the pH to 1 by addition
of concentrated aqueous HCl (5 mL). The aqueous phase
was then extracted with ethyl acetate (4 × 50 mL). The
combined extracts were evaporated to yield crystalline,
colorless 3 (1.87 g, 93.5%). Enantiomeric purity (vide
supra): R/S ) 99.0:1.0.
(C2): Resolution of Racemic PLA with L-Phenylalani-
nol. 11 (4.56 g, 27.45 mmol) and L-phenylalaninol (4.61 g,
30.5 mmol) were suspended in 2-butanone (180 mL) and
heated to 70 °C. After filtration, the clear filtrate was slowly
cooled to room temperature. At 30 °C, a few seed crystals
of recrystallized (R)-salt 12B from a previous batch were
added. The resulting, thick suspension was stirred for another
1.5 h at rt and 1.5 h in an ice bath and filtered. The filter
cake was washed with ice-cold TBME (30 mL) and dried in
a vacuum oven (35 °C). The resulting salt 12B (3.82 g,
87.6%) was then suspended in methylene chloride (20 mL)
and 1 M sodium hydroxide (20 mL). After stirring vigorously
for 5 min, the aqueous phase was separated and extracted
with methylene chloride (20 mL). Both methylene chloride
phases were washed with aqueous 1 M sodium hydroxide
(10 mL). Water phases were combined, and the pH was
adjusted to pH 1 by addition of concentrated aqueous HCl
(4 mL). The resulting suspension was stirred in an ice bath
for 30 min and filtered. The filter cake was dissolved in
TBME and evaporated to yield crystalline, colorless 3 (1.19
g, 52.2%). Enantiomeric purity (vide supra): R/S ) 100:0.
(C3): Resolution of Racemic PLA with (+)-Dehydroa-
bietylamine. To a solution of 11 (4.0 g, 24.1 mmol) in
isobutyl methyl ketone (50 mL) was added a solution of (+)-
dehydroabietylamine (Merck, z.S.) (8.22 g, 26.5 mmol) in
the same solvent (100 mL). After stirring the initially clear
solution at room temperature for approximately 10-15 min,
a precipitate formed. The suspension was heated to 95 °C,
whereupon a clear yellow solution formed that was slowly
cooled to room temperature. At 85-90 °C, a few seed
crystals of recrystallized (S)-salt 12C from a previous batch
were added. At 70 °C, a thick suspension of (S)-salt 12C
formed which was stirred for 1.5 h and filtered. The filter
cake was washed with isobutyl methyl ketone (50 mL). The
filtrate was evaporated to dryness under reduced pressure,
the oily residue taken up in methylene chloride (30 mL),
and the clear solution extracted with 1 M sodium hydroxide
(3 × 20 mL). The combined water phases were extracted
with methylene chloride (20 mL), and the pH of the water
phase was adjusted to pH 1-2 by addition of concentrated
aqueous HCl (5 mL). The resulting suspension was extracted
with ethyl acetate (4 × 30 mL). EtOAc phases were
combined and evaporated under reduced pressure to yield
crude 3 (1.46 g, 73%). Enantiomeric purity (vide supra): R/S
) 80.4:19.6. A portion of the crude material (1.28 g) was
recrystallized from deionized water (7 mL) to give colorless,
crystalline 3 (0.86 g, 49.0%). Enantiomeric purity (vide
supra): R/S ) 100:0.
(E) (By Enantioselective Biocatalytic Reduction of
Phenyl Pyruvic Acid). Proteus mirabilis DSM 30115 was
grown on 5-L scale at 37 °C, under anaerobic conditions
(nitrogen flow via sparger) and with pH 6.5 regulation
(corrected with a 30% NaOH solution in case of decrease
below pH 6.5), in a medium of the following composition:
20 g/L soy peptone, 5 g/L yeast extract, 5 g/L K2HPO4, pH
7.2, inoculated with a preculture of 10% volume. During
the fermentation a 70% (w/v) glucose monohydrate solution
was fed at a constant flow rate of 7.2 mL/h. After 23 h of
fermentation, the culture was cooled to 15 °C, and the cells
were harvested by centrifugation (Cryofuge 8000, Heraeus,
35 min at 5000 rpm, at 4 °C). The wet cells obtained (about
17 g of wet packed cells/L of culture) were frozen and stored
during 2 years at -20 °C.
The reduction of phenyl pyruvic acid was conducted on
100-mL scale in a stirred glas reactor with jacket (Metrohm)
at 37 °C, under anaerobic conditions (nitrogen flow subsur-
face). A mixture was prepared with the following compo-
nents added (and dissolved) in the same sequence as indicated
below: 43 g of purified water, 0.256 g of KH2PO4, 0.384 g
of K2HPO4, 1.36 g of potasium formate, 7 g of phenyl
pyruvic acid (Fluka No. 78190, purum, ∼99%), 24 g of poly-
(ethylene glycol) 4000, ∼5.73 g of a 50% (w/w) KOH
solution (to adjust pH 7.0), 11.27 g of purified water, 40
mg of benzyl viologen previously dissolved in 1.5 mL of
purified water, 1.5 g of wet packed cells previously
suspended in 4 mL of purified water, a drop of antifoam
polypropylene glycol 2025, the whole giving a total weight
of 100 g (corresponding to a volume of about 90 mL). During
the reaction, the pH was controlled and continuously adjusted
to pH 7.0 by addition of a 50% (v/v) formic acid solution
with an automatic buret (Dosimat Metrohm 665 with a 20-
mL buret, Impulsomat Metrohm 614, pH meter Metrohm
632). Samples were taken at different times for determination
of the percentage of conversion by HPLC analysis: 20 µL
of a 400-fold diluted sample (with 0.1 M KH2PO4 pH 3.0)
was applied to a 5 µ Nucleosil C18 column (length: 16 cm,
diameter: 4.6 mm, mobile phase: 950 mL of 0.05 M KH2-
PO4 pH 3.0 + 500 mL of acetonitrile, flow rate: 1 mL/min,
detection at 216 nm). The end of the reaction was reached
after 20 h, with a measured phenyl pyruvic acid conversion
of >99.8% and a formic acid consumption of 3.16 mL.
The (R)-3-phenyl lactic acid was recovered as indicated
below. The reaction mixture (about 91 mL) was adjusted to
pH 2.0 with 3.69 g of H2SO4 96%, and 90 mL of ethyl acetate
was added at room temperature under stirring. The mixture
was transferred to a separatory funnel for partial phase
separation during 30 min and then filtered through a 1-cm
thick Hyflo layer (on a Bu¨chner filter of diameter 6.5 cm).
The filter was rinsed with 6 mL of ethyl acetate, and the
566
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development

filtrate was tranfered to a separatory funnel for complete
phase separation (during 30-60 min). The aqueous phase
was recovered and subjected to a second extraction step at
pH 2.0 (adjusted with H2SO4 96%) with 90 mL of ethyl
acetate, and after phase separation, the organic phase was
recovered and added to the organic phase of the first
extraction step. The organic pool (168 mL) was subjected
to two consecutive wash steps with 0.2 volume of a saturated
Na2SO4 solution (52 g of Na2SO4 per 100 g of purified water)
at pH 2.0 (adjusted with H2SO4 96%). The recovered, washed
organic pool was then concentrated to about 30 mL in a
rotavapor. Fresh ethyl acetate (30 mL) was added to the
concentrate, and the solution was reduced again to 30 mL.
The crystallization of the product was performed by addition
of 100 mL methylcyclohexane to the concentrate, and
complete removal of ethyl acetate by evaporation in the
rotavapor. The product suspension in methylcyclohexane was
then cooled to 10 °C and filtered on a paper filter. After
two rinse steps with 10 mL of ice cold methylcyclohexane,
the final product was recovered and dried at 65 °C, to give
white (colorless), crystalline (R)-3-phenyl lactic acid (3) (6.17
g, 87%). Enantiomeric purity (vide supra): R/S > 99.5:0.5.
General Procedure for the Hydrogenation Screening
Experiments (Tables 1 and 2). The starting material was
placed in stainless steel pressure vessel and dissolved in the
solvent indicated, and the moistened catalyst was added. The
reaction vessel was closed, flushed twice with nitrogen gas
and twice with hydrogen gas. The reactor was pressurized
to the hydrogen pressure indicated and stirred at the given
temperature for the indicated period of time. The consump-
tion of hydrogen ceased within this period of time. The
reactor was cooled to room temperature, the hydrogen
pressure released, and the reactor flushed twice with nitrogen
gas. The crude reaction mixture was filtered and evaporated
to dryness, and the crude product mixture was analyzed by
GLC (percent area) after derivatization with AcCl/iPrOH/
TFAA.
(2R)-2-Hydroxy-3-cyclohexylpropionic acid (4).10 (A)
From Phenyl Lactic Acid. A Buchi reactor was charged with
a solution of (2R)-2-hydroxy-3-phenyl lactic acid (3) (50.0
g, 0.3 mol, R/S >99.5:0.5, from enantioselective bioreduction
with Proteus mirabilis; see procedure E) in 94% ethanol
(1000 mL) and 5% Rh-C hydrogenation catalyst (5.0 g).
The mixture was hydrogenated at 50 °C under 15 bar H2
pressure under vigorous stirring for 3 h. The suspension was
filtered, the clear filtrate was evaporated, and the resulting
crystalline solid was dried in a vacuum-drying oven at 50
°C. Yield 50.2 g (96.9%). Enantiomeric purity (vide supra):
R/S ) 99.5:0.5. Mp 92-94 °C. 1H NMR (400 MHz, DMSO)
δ 0.75-1.0 (m, 2 H), 1.1-1.3 (m, 3 H), 1.35-1.52 (m, 3
H), 1.55-1.82 (m, 5 H), 3.0-3.8 (br s, 1 H, H-O-C(2)), 3.98
(m, 1 H, H-C(2)). 13C NMR (100 MHz, DMSO) δ 26.5, 26.8,
27.0 (C(3′), C(4′), C(5′) cHexyl), 32.7/34.4 (C(2′), C(6′)
cHexyl), 34.1 (C(1′) cHexyl), 42.4 (C(3), 68.4 (C(2), 177.3
(C(1)). MS(EI) m/z 173 (MH+), 154 (M+ - H2O), 136, 127,
109, 83, 67, 55.
mmol) in methanol (310 mL) was added 98 mL of 2 M
methanolic HCl and 5% Pd-C hydrogenation catalyst (10.5
g). Under vigorous stirring in a Buchi glass hydrogenation
reactor, the suspension was hydrogenated at 67 °C under 4
bar H2 pressure for 5 h. The temperature was lowered to 50
°C, and 5% Rh-C hydrogenation catalyst (3.5 g) was added.
Hydrogenation was continued at 50 °C for another 17 h
before the suspension was filtered, the clear filtrate was
evaporated, and the resulting oil (15)33 was dried in a
vacuum-drying oven at 35 °C/1 mbar for 1 h. The oil was
taken up in methanol (633 mL) and water (211 mL). Lithium
hydroxide monohydrate (10.7 g, 255.0 mmol) was added at
30 °C, and the turbid solution was stirred at 25 °C overnight.
Amberlyst IR-120 (H+-form, dry, 60.9 g ≈ 268.1 mEq H+)
was added, the ion-exchange resin was filtered off andwashed
with excess methanol and water, and the clear filtrate was
evaporated under reduced pressure to yield a crystalline
material which was dried in a vacuum drying oven (40 °C,
1 mbar). Yield: 28.95 g (94.2% over three steps) 4 as off-
white, crystalline solid. Enantiomeric purity (vide supra): R/S
) 99.3:0.7. Os-content: 4 ppm (by elemental analysis).
(2R,3S)-2,3-Dihydroxy-3-phenylpropionic Acid Methyl
Ester (13A). According to the procedure of Sharpless et al.,26
from 41.85 g of methyl cinnamate were obtained 37.72 g
(74.5%) 13A, R/S ) 98.8:1.2 after first recrystallization from
toluene; respectively 35.40 g (69.9%), R/S ) 99.7:0.3 after
second recrystallization from toluene. Os-content: 102 ppm
(by elemental analysis).
(2R,3S)-2,3-Dihydroxy-3-phenylpropionic Acid Benzyl
Ester (13B).31 Following the same procedure,26 from 15.9 g
of benzyl cinnamate was obtained 17.9 g of 13B, 99% yield,
R/S ) 90.7:9.3 after chromatograpy from toluene/ethyl
acetate 5:1. MS(EI) m/z 272 (M+), 197, 166, 148, 107, 91,
79, 65.
(2R)-2-Hydroxy-3-phenylpropionic Acid Methyl Ester
(14A).32 (A) From Hydrogenolysis/Transesterification of the
Benzyl Ester Diol. To a solution of 13B (R/S ) 90.7:9.3)
(8.7 g, 32.0 mmol) in methanol (100 mL) was added 8.8
mL of 1.1 M methanolic HCl and 5% Pd-C hydrogenation
catalyst (870 mg). Under vigorous stirring, the suspension
was hydrogenated at 25 °C under 2 bar H2 pressure for 16
h. The suspension was filtered, the clear filtrate was
evaporated, and the resulting crystalline solid was dried in a
vacuum-drying oven at 50 °C. Yield 4.805 g (83.3%).
Enantiomeric purity (vide supra): R/S ) 93.3:6.7.
(B) From Hydrogenolysis of the Methyl Ester Diol. To a
solution of 13A (R/S ) 98.6:1.4) (5.7 g, 29.1 mmol) in
methanol (50 mL) was added 16 mL of 1.1 M methanolic
HCl and 5% Pd-C hydrogenation catalyst (770 mg). Under
vigorous stirring, the suspension was hydrogenated at 25 °C
under 1.7-4 bar H2 pressure for 30 h. The suspension was
filtered, the clear filtrate was evaporated, and the resulting
crystalline solid was dried in a vacuum-drying oven at 50
(31) Mukaiyama, T.; Sugimura, H.; Ohno, T.; Kobayashi, S. Chem. Lett. 1989,
1401.
(32) (a) Schoofs, A.; Weidmann, R.; Collet, A.; Horeau, A. Bull. Soc. Chim. Fr.
1976, 2031. (b) Mor, G.; Ehret, A.; Cohen, S. G. Arch. Biochem. Biophys.
1976, 174, 158.
(B) From Methyl Dihydroxyphenylpropionate 13A (Vide
infra). To a solution of 13A (R/S ) 99.7:0.3) (35.0 g, 178.4
(33) Damon, D. B.; Hoover, D. J. J. Am. Chem. Soc. 1990, 112, 6439.
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
567

°C. Yield 5.22 g (99.0%). Enantiomeric purity (vide supra):
R/S ) 98.2:1.8.
(d, 1 H, J ) 6.0 Hz, HO-C(2)), 7.32-7.53 (AA′BB′, 4 H,
ClPh). 13C NMR (100 MHz, DMSO) δ 26.5, 26.7, 26.9, 32.8,
34.0, 34.2, 42.3, 65.5, 68.7, 110.0, 122.1, 131.1, 132.2, 136.5,
175.3. MS(CI): m/z 367 [M + C2H5-gas adduct], 343, 342,
341, 340 (M+), 339, 295, 293, 277, 211, 198, 171, 169, 127,
125, 109, 83. IR (KBr, cm-1) 3408s/br, 2923s, 2848s, 1752ss,
1728s, 1489m, 1223m, 1209m, 1186m, 1170m, 1142m,
1091m, 1070s, 1012m, 805s. Elemental analysis calcd for
C16H21BrO3 (341.24): C 56.32%, H 6.20%, Br 23.42%;
found: C 56.31%, H 6.07%, Br 23.30%.
The potential for epimerization during this reaction was
checked in two separate batches: Exp. A: 3.3 mmol of 4
(R/S ) 98.2:1.8) gave 88% of 18, R/S ) 99.5/0.5; Exp. B:
50 mmol of 4 (R/S ) 99.3:0.7) gave 82% of 18, R/S ) 99.8:
0.2).
(2R)-2-Hydroxy-3-cyclohexylpropionic acid p-nitroben-
zyl ester (19): 37% yield, yellow oil, (chromat. n-hexane/
EtOAc 4:1), 1H NMR (400 MHz, DMSO) δ 0.80-1.77 (m,
13 H, H2-C(3) + cHexyl), 4.20 (dt, J ) 6.0 Hz, 6.7 Hz, 1
H, H-C(2)), 5.28 (A2, 2 H, NO2Ph-CH2), 5.46 (d, 1 H, J )
6.0 Hz, HO-C(2)), 7.65, 8.27 (AA′BB′, 4 H, ClPh). 13C NMR
(100 MHz, DMSO) δ 25.8, 26.0, 26.2, 32.0, 33.3, 33.5, 41.5,
64.4, 68.0, 123.7, 128.7, 144.1, 147.3, 174.5. MS(EI): m/z
308, 307 (M+), 289, 262, 211, 193, 171, 153, 137, 127, 109,
107, 97, 83, 81, 67, 55. IR (Film, cm-1) 3485s/br, 2924ss,
2851s, 1741ss, 1608m, 1524ss, 1449m, 1348ss, 1275s, 1194s,
1142s, 1098s, 1013m, 853w, 738w. Elemental analysis calcd
for C16H21NO5 (307.35): C 62.53%, H 6.89%, N 4.56%;
found: C 62.79%, H 7.02%, N 4.49%.
General Synthesis Procedure for the Esters 5, and 16-
20. Anhydrous potassium carbonate (1.15 equiv) was added
to a solution of 4 (10-50 mmol) in dry acetone (80-400
mL). The white suspension was stirred at room temperature
for 3 h, whereupon tetrabutylammonium iodide (1.0 equiv)
and the corresponding benzyl halide (1.15 equiv) were added,
and the suspension was stirred for a further 12-18 h at 40°C,
until TLC had indicated complete consumption of the
hydroxy acid. After cooling to room temperature, the
suspension was filtered over a sand/Celite pad, and the filter
cake was washed with acetone. Evaporation under reduced
pressure gave the crude product as a colorless-to-yellowish
oil, which was further purified or crystallized by the solvent
system indicated below with the individual compounds.
Following this procedure we obtained:
(2R)-2-Hydroxy-3-cyclohexyl-propionic acid benzyl
ester (5):7 87% yield (chromat. from n-hexane/EtOAc 9:1),
colorless oil, crystallizes slowly in the refrigerator, mp 54-
1
56 °C, H NMR + 13C NMR identical to lit.7 MS(EI) m/z
262 (M+), 181, 166, 153, 127, 109, 91, 83, 67, 55. IR (KBr,
cm-1) 3314s/br, 2923s, 2849m, 1751ss, 1742s, 1538m,
1511s, 1463w, 1444w, 1212m, 1192s, 1143s, 1087s, 1078s,
738m, 695m.
(2R)-2-Hydroxy-3-cyclohexylpropionic acid p-chlo-
robenzyl ester (16): 50% yield, (chromat. n-hexane/EtOAc
1
9:1), colorless oil, H NMR (400 MHz, DMSO) δ 0.79-
1.73 (m, 13 H, H2-C(3) + cHexyl), 4.12 (dt, J ) 6.4 Hz,
6.7 Hz, 1 H, H-C(2)), 5.12 (AA′, 2 H, ClPh-CH2), 5.40 (d,
1 H, J ) 6.1 Hz, HO-C(2)), 7.39-7.47 (AA′BB′, 4 H, ClPh).
13C NMR (100 MHz, DMSO) δ 26.5, 26.7, 26.9, 32.7, 34.0,
34.1, 34.2, 42.3, 65.5, 68.7, 129.3, 129.8, 130.8, 133.6, 135.6,
136.1, 175.3. MS(EI) m/z 298, 296 (M+), 171, 153, 127, 125,
109, 89, 83, 81, 67, 55. IR (Film, cm-1) 3479s/br, 2924ss,
2851s, 1736ss, 1494s, 1449m, 1274s, 1198s, 1142s, 1094s,
1017m, 965w, 845w, 809m. Elemental analysis calcd for
C16H21ClO3 (296.79): C 64.75%, H 7.13%, Cl 11.95%;
found: C 64.84%, H 6.90%, Cl 11.88%.
(2R)-2-Hydroxy-3-cyclohexylpropionic acid 2-naphth-
ylmethyl ester (20): 94% yield, mp 54-55 °C, (chromat.
from n-hexane/EtOAc 9:1), colorless oil, crystallizes slowly
1
while standing at rt, mp 54-55 °C, H NMR (400 MHz,
DMSO) δ 0.77-1.74 (m, 13 H, H2-C(3) + cHexyl), 4.18
(dt, J ) 6.2 Hz, 6.8 Hz, 1 H, H-C(2)), 5.30 (AA′, 2 H,
Naphth-CH2), 5.42 (d, 1 H, J ) 6.2 Hz, HO-C(2)), 7.50-
7.57 (m, 3 H, Naphth-H), 7.90-7.95 (m, 4 H, Naphth-H).
13C NMR (100 MHz, DMSO) δ 26.5, 26.7, 26.9, 32.8, 34.0,
34.2, 42.4, 66.5, 68.8, 126.8, 127.2, 127.3, 127.7, 128.5,
128.6, 129.0, 133.5, 133.6, 134.6, 175.4. MS(CI): m/z 312
(M+), 281, 141, 109, 83. IR (KBr, cm-1) 3506s, 3483s,
2923s, 2849m, 1708ss, 1742s, 1465w, 1291m, 1258m,
1233w, 1132m, 980m, 823s, 750m. Elemental analysis calcd
for C20H24O3 (312.41): C 76.89%, H 7.74%; found: C
76.77%, H 7.75%.
General Synthesis Procedure for the Sulfonates 16-
28. Toluene-4-sulfonyl chloride, 4-bromobenzenesulfonyl
chloride or 4-nitrobenzenesulfonyl chloride (1.3 equiv) was
added to a solution of the corresponding hydroxyester (5,
18 or 20) (1-5 mmol) in dry toluene (5-25 mL) at room
temperature. N-Methylmorpholine (1.4 equiv) was added, and
the suspension was stirred for a further 2-4 h at room
temperature and 10-18 h at 50 °C, until TLC had indicated
complete consumption of the hydroxy ester. After cooling
to room temperature, the suspension was washed with 20%
NaHCO3 (2 × 10 mL) and water (2 × 15 mL). Combined
water phases were extracted with toluene (1 × 15 mL).
Drying of the toluene phases over sodium sulfate and
(2R)-2-Hydroxy-3-cyclohexylpropionic acid 2,5-dichlor-
benzyl ester (17): 93% yield, (chromat. n-hexane/EtOAc
1
9:1), colorless oil, H NMR (400 MHz, DMSO) δ 0.74-
1.71 (m, 13 H, H2-C(3) + cHexyl), 4.10 (br dt, 1 H, H-C(2)),
5.31 (AA′, 2 H, Cl2Ph-CH2), 5.43 (d, 1 H, J ) 6.0 Hz, HO-
C(2)), 7.42-7.58 (m, 3 H, Cl2Ph). 13C NMR (100 MHz,
DMSO) δ 26.5, 26.7, 26.9, 32.9, 34.0, 34.1, 34.15, 42.5,
61.7, 68.7, 129.6, 131.7, 132.5, 136.9, 175.2. MS(CI) m/z
331 (MH+), 285, 171, 159, 125, 79. IR (Film, cm-1) 3477m/
br, 2924ss, 2851s, 1738ss, 1583m, 1565m, 1438s, 1248m,
1200s, 1142m, 1096s, 989w, 965w, 779m, 769m. Elemental
analysis calcd for C16H20Cl2O3 (331.24): C 58.02%, H
6.09%, Cl 21.41%; found: C 57.92%, H 6.09%, Cl 21.30%.
(2R)-2-Hydroxy-3-cyclohexylpropionic acid p-bro-
mobenzyl ester (18): 88% yield (chromat. n-hexane/EtOAc
4:1), respectively 82% (recryst. from n-hexane), yellowish
1
crystals, mp 50-52 °C, H NMR (400 MHz, DMSO) δ
0.77-1.75 (m, 13 H, H2-C(3) + cHexyl), 4.12 (dt, J ) 6.0
Hz, 6.5 Hz, 1 H, H-C(2)), 5.11 (AA′, 2 H, BrPh-CH2), 5.38
568
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development

evaporation under reduced pressure gave the crude product
ususally as a yellow oil, which was further purified or
crystallized by the solvent system indicated below with the
individual compounds.
1175ss, 1142m, 1043m, 1001s, 965s, 922m, 858m, 818m,
797m.
(2R)-2-(4-Nitrophenyl)sulfonyloxy-3-cyclohexyl-propi-
onic acid 2-naphthylmethyl ester (25): 58% yield, (crude
product chromat. from n-hexane/EtOAc 9:1), yellowish oil,
1H NMR (400 MHz, DMSO) δ 0.70-1.52 (m, 11 H,
cHexyl), 1.64 (m, 1 H, H-C(3)), 1.75 (m, 1 H, H′-C(3)), 5.15
(dd, 1 H, J ) 4.1, 9.3 Hz, H-C(2)), 5.30 (Naphth-CH2),
7.42-7.56 (m, 3 H, Naphth), 7.87-7.93 (m, 4 H, Naphth),
8.21-8.38 (AA′BB′, NO2PhSO2). 13C NMR (125 MHz,
DMSO) δ 25.8, 26.0, 26.1, 26.2, 32.0, 33.2, 33.5, 60.2, 67.5,
68.2, 77.5, 81.1, 123.8, 125.2, 126.5, 126.9, 127.0, 127.4,
128.1, 128.2, 128.3, 128.6, 128.8, 130.0, 132.9, 133.0, 133.1,
133.2, 141.1, 147.7, 151.1, 154.8, 168.7, 170.6. MS(CI): m/z
526 [M+ + C2H5 gas adduct], 497 [M+], 156, 141 [Naphth-
(2R)-2-(4-Nitrophenyl)sulfonyloxy-3-cyclohexyl-propi-
onic acid p-brombenzyl ester (21): 60% yield, (crude
product recryst. from EtOH), yellowish crystals, mp 71-72
1
°C, H NMR (400 MHz, DMSO) δ 0.67-1.76 (m, 13 H,
H2-C(3) + cHexyl), 4.12 (dt, J ) 6.0 Hz, 6.5 Hz, 1 H,
H-C(2)), 5.09-5.15 (m, 3 H, H-C(2) + BrPh-CH2), 7.28-
7.56 (AA′BB′, 4 H, BrPh), 8.20-8.43 (AA′BB′, 4 H, NO2-
Ph). 13C NMR (100 MHz, DMSO) δ 26.1, 26.4, 26.5, 26.6,
32.1, 32.4, 33.6, 33.7, 33.9, 67.0, 77.8, 122.5, 124.2, 125.7,
127.8, 130.4, 131.2, 132.3, 135.4, 141.5, 151.6, 169.1. MS-
(EI): m/z 527 [MH+], 356, 322 [M+ - NO2PhSO3H], 230,
+
+
186, 171, 169 [BrPhCH2 ], 138, 94, 83. IR (KBr, cm-1)
CH2 ]. IR (Film, cm-1) 3100w, 2925s, 2852m, 1757s,
3120m, 2926s, 1752s, 1734s, 1535ss, 1372s, 1351s, 1190ss,
1171m, 1166m, 982s, 961s, 858m, 843s, 748m, 617s.
Elemental analysis calcd for C22H24BrNO7S (526.41): C
50.20%, H 4.60%, Br 15.18%, N 2.66%, S 6.09%; found:
C 50.17%, H 4.59%, Br 15.05%, N 2.58%, S 6.10%.
(2R)-2-Methansulfonyloxy-3-cyclohexyl-propionic acid
p-brombenzyl ester (22): 81% yield, (crude product chro-
mat. from n-Hexane/EtOAc 9:1), colorless oil, 1H NMR (400
MHz, DMSO) δ 0.85-1.79 (m, 13 H, H2-C(3) + cHexyl),
3.25 (s, 3 H, CH3SO2), 5.14-5.24 (m, 3 H, H-C(2) + BrPh-
CH2), 7.36-7.61 (AA′BB′, 4 H, BrPh). 13C NMR (125 MHz,
DMSO) δ 25.9, 26.0, 26.2, 32.1, 33.2, 33.3, 38.3, 66.4, 76.0,
122.1, 130.9, 131.9, 135.2, 169.5. MS(CI): m/z 421 [MH+],
420, 419, 418, 417, 339, 337, 323, 321, 249, 171, 138, 94.
IR (KBr, cm-1) 3030w, 2926ss, 2852s, 1756ss, 1490m,
1449m, 1361ss, 1277m, 1174ss, 1141m, 1071m, 1010s, 967s,
922m, 858m, 801s.
(2R)-2-(4-Bromphenyl)sulfonyloxy-3-cyclohexyl-propi-
onic acid p-brombenzyl ester (23): 52% yield (crude
product recryst. from EtOH), respectively 48% yield (crude
product chromatographed from n-hexane/EtOAc 9:1), yel-
lowish crystals, mp 68-69 °C, 1H NMR (400 MHz, DMSO)
δ 0.67-1.71 (m, 13 H, H2-C(3) + cHexyl), 4.94 (dd, J )
3.8 Hz, 9.6 Hz, 1 H, H-C(2)), 5.12 (s, 3 H, BrPh-CH2), 7.29-
7.61 (AA′BB′, 4 H, BrBn), 7.88 (s, 4 H, BrPhSO2). 13C NMR
(125 MHz, DMSO) δ 25.7, 26.0, 26.1, 31.6, 33.2, 66.5, 76.6,
122.0, 128.2, 129.3, 130.1, 130.8, 131.0, 131.9, 132.0, 133.3,
134.8, 135.0, 168.9. MS(ESI): dec, m/z 236, 234 [BrPhSO3H].
IR (Film, cm-1) 2924s, 1760ss, 1745sh/s, 1576s, 1490m,
1392s, 1377s, 1278m, 1190s, 1071s, 1012s, 919s, 861m,
823s, 797m, 611s.
1534ss, 1381m, 1350s, 1278m, 1188ss, 991m, 920w, 853m,
819w, 745m.
(2R)-2-(4-Nitrophenyl)sulfonyloxy-3-cyclohexyl-propi-
onic acid benzyl ester (26): 80% yield, (crude product
chromat. from n-hexane/EtOAc 9:1), yellowish oil, 1H NMR
(400 MHz, DMSO) δ 0.72-1.76 (m, 13 H, H2-C(3) +
cHexyl), 5.11-5.14 (m, 3 H, H-C(2) + Ph-CH2), 7.30-7.43
(m, 5 H, Ph), 8.21-8.42 (AA′BB′, NO2PhSO2). 13C NMR
(100 MHz, DMSO) δ 26.1, 26.4, 26.5, 32.1, 33.2, 33.6, 33.7,
39.4, 67.8, 77.9, 125.7, 129.1, 129.2, 129.3, 130.4, 135.9,
141.5, 147.7, 151.6, 169.1. MS(EI): m/z 447 [M+], 356 [M+
•
- PhCH2 ], 330, 312, 244, 186 [NO2PhSO2+.], 153, 138,
122, 109, 92, 67, 55. IR (Film, cm-1) 3100w, 2925s, 2852m,
1758s, 1534ss, 1404m, 1381s, 1351s, 1314m, 1278m, 1188ss,
1100m, 920m, 853m, 819w, 745s, 617m.
(2R)-2-(4-Bromphenyl)sulfonyloxy-3-cyclohexyl-propi-
onic acid benzyl ester (27): 63% yield, (crude product
chromat. from n-hexane/EtOAc 9:1), yellowish oil, 1H NMR
(500 MHz, DMSO) δ 0.80-1.68 (m, 13 H, H2-C(3) +
cHexyl), 4.91 (dd, 1 H, J ) 2.3, 9.7 Hz, H-C(2)), 5.11 (s, 2
H, Ph-CH2), 7.30-7.41 (AA′BB′, 4 H, BrPhSO2), 7.85-
7.87 (m, 5 H, Ph). 13C NMR (125 MHz, DMSO) δ 25.7,
26.0, 26.1, 31.6, 33.2, 39.0, 67.3, 76.6, 128.6, 128.8, 128.9,
129.3, 130.2, 133.3, 134.8, 135.6, 168.9. MS(CI): m/z 481
•
[M+], 390 [M+ - PhCH2 ], 244, 220, 181, 153, 138, 91. IR
(Film, cm-1) 3100w, 2924s, 2850m, 1759s, 1534ss, 1450m,
1392s, 1378s, 1278m, 1190ss, 1069m, 1012m, 1001m, 919m,
861m, 777w, 749m, 613m. Elemental analysis calcd for
C22H25BrO5S (10% toluene) (442.48): C 55.65%, H 5.31%,
S 6.52%; found: C 55.74%, H 5.30%, S 6.57%.
(2R)-2-(4-Tolyl)sulfonyloxy-3-cyclohexyl-propionic acid
benzyl ester (28): 60% yield, (crude product chromat. from
(2R)-2-Methansulfonyloxy-3-cyclohexyl-propionic acid
2-naphthylmethyl ester (24): 79% yield, (crude product
chromat. from n-hexane/EtOAc 9:1), slightly yellowish oil,
1H NMR (400 MHz, DMSO) δ 0.88-1.75 (m, 13 H, H2-
C(3) + cHexyl), 3.25 (s, 3 H, CH3SO2), 5.20 (m, 1 H,
H-C(2)), 5.40 (Naphth-CH2), 7.51-7.56 (m, 3 H, Naphth),
7.92-7.95 (m, 4 H, Naphth). 13C NMR (125 MHz, DMSO)
δ 25.8, 26.0, 26.2, 32.2, 33.2, 33.3, 38.3, 67.3, 76.2, 126.4,
126.9, 127.7, 128.1, 128.3, 128.6, 133.1, 133.3, 169.6. MS-
1
n-hexane/EtOAc 9:1), colorless oil, H NMR (500 MHz,
DMSO) δ 0.60-1.69 (m, 13 H, H2-C(3) + cHexyl), 2.42
(s, 3 H, CH3-PhSO2), 4.83 (dd, 1 H, J ) 3.6, 9.1 Hz, H-C(2)),
5.12 (s, 2 H, Ph-CH2), 7.32-7.43 (m, 5 H, Ph), 7.46-7.82
(AA′BB′, 4 H, MePhSO2). 13C NMR (100 MHz, DMSO) δ
26.1, 26.4, 26.5, 32.0, 33.5, 33.6, 39.6, 67.6, 76.4, 128.6,
129.0, 129.2, 129.3, 131.0, 133.0, 136.1, 146.3, 169.5. MS-
(CI): m/z 417 [MH+], 327, 325, 299, 281, 245, 227, 181,
138, 107, 91. IR (Film, cm-1) 3050w, 2925s, 2852m, 1761s,
1740sh, 1595m, 1498m, 1450m, 1374s, 1278s, 1190ss,
(CI): m/z 390 [MH+], 141 [Naphth-CH2 ]. IR (Film, cm-1)
+
3035w, 2925ss, 2852s, 1753ss, 1455m, 1361ss, 1277m,
Vol. 7, No. 4, 2003 / Organic Process Research & Development
•
569

1178ss, 1002s, 918m, 861m, 815m, 698m, 666s, 554m.
Elemental analysis calcd for C23H28O5S (416.54): C 66.32%,
H 6.78%, S 7.70%; found: C 66.38%, H 6.75%, S 7.79%.
HPLC. Mr. Dominique Grimler and Mr. Michael Haller are
thanked for their technical assistance with some scale-up
experiments, and Dr. C.-P. Mak for his encouragement.
Acknowledgment
We gratefully acknowledge the support of Dr. Christoph
Spoendlin for the determination of enantiomeric purities by
Received for review February 27, 2003.
OP030202Q
570
•
Vol. 7, No. 4, 2003 / Organic Process Research & Development