Chiral cobalt-catalyzed enantioselective aerobic oxidation of α-hydroxy esters

Article abstract of DOI:10.1039/c0cc01917h

A chiral cobalt-catalyzed enantioselective aerobic oxidative kinetic resolution of (±)-α-hydroxy esters, using molecular oxygen as a sole oxidant, is reported and a maximum of selectivity factor (s) 31.9 was achieved with >99% enantiomeric excess for unreacted α-hydroxy esters.

Full text of DOI:10.1039/c0cc01917h

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Chiral cobalt-catalyzed enantioselective aerobic oxidation of
a-hydroxy esterswz
Santosh Kumar Alamsetti and Govindasamy Sekar*
Received 16th June 2010, Accepted 4th August 2010
DOI: 10.1039/c0cc01917h
A chiral cobalt-catalyzed enantioselective aerobic oxidative
kinetic resolution of (ꢀ)-a-hydroxy esters, using molecular
oxygen as a sole oxidant, is reported and a maximum of
selectivity factor (s) 31.9 was achieved with 499% enantiomeric
excess for unreacted a-hydroxy esters.
Scheme 1 Chiral cobalt catalyzed OKR of (ꢀ)-a-hydroxy esters.
for the first time, we report our initial findings regarding chiral
cobalt catalyzed oxidative kinetic resolution of racemic
a-hydroxy esters using molecular oxygen as the ultimate
oxidant (Scheme 1).
Enantiopure secondary alcohols are very important intermediates
in the field of asymmetric synthesis, as well as pharmaceutical,
agrochemical, and fine chemical industries.1 Although a plethora
of methods to synthesize enantiopure alcohols exist,2,3 the non-
enzymatic kinetic resolution route is particularly attractive and
versatile.4,5 Among the various kinetic resolution methods,
oxidative kinetic resolution (OKR) offers extraordinary practical
advantages, such as the use of molecular oxygen as the sole
stoichiometric oxidant and the formation of water as the ultimate
by-product. The pioneering work by Rychnovsky et al. using
chiral nitroxyl catalyst,6 followed by independent reports of
Sigman et al.7 and Stoltz et al.8 using chiral palladium complexes,
have contributed significantly to the field of oxidative kinetic
resolution.
We started our investigation with racemic methyl mandelate
(ꢀ)-1 as a model substrate for aerobic oxidative kinetic
resolution. When (ꢀ)-methyl mandelate was reacted with
5 mol% of TEMPO (2,2,6,6-tetramethylpiperidin-1-oxyl) and
5 mol% of enantiopure (S)-BINAM L1–Co(OAc)2 complex in
the presence of molecular oxygen at 80 1C, the oxidation failed
to take place (Table 1, entry 1). Similarly, when BINAM was
replaced with (ꢁ)-sparteine (L2), (R)-BINOL (L3), and (L)-
proline (L4), the oxidation reaction did not take place. When
Table 1 Ligand and Co salt screening for OKR of (ꢀ)-1a
Recently,
a chiral vanadium catalyst synthesized from
vanadium(V) tri-iso-propoxyoxide, in combination with
a
tridentate ligand derived from 3,5-di-tert-butylsalicylaldehyde
and (S)-tert-leucinol reported by Toste et al.9 for the resolutions
of racemic a-hydroxy esters, opened a new channel for the
synthesis of highly important enantiomerically-enriched
a-hydroxy esters.10 Later on, Chien et al. used a similar kind
of N-salicylidene-L-a-amino acid-based vanadyl(V) complex as
a catalyst for the OKR of racemic a-hydroxy esters.11 Very
recently, Jones and co-workers developed a polymer and
silica-supported tridentate Schiff base vanadium complex as
a catalyst for the asymmetric oxidation of racemic a-hydroxy
esters.12
Entry Ligand Co salt
Time C (%)b %ee of alcoholc sd
1
L1
L2
L3
L4
L5
L6
L5
L5
L5
L5
—
Co(OAc)2 4 d
Co(OAc)2 4 d
Co(OAc)2 4 d
Co(OAc)2 4 d
—
—
—
—
—
—
—
—
52
26
14
—
—
—
—
—
—
—
—
9.0
5.3
1.6
—
—
—
—
2
3
4
5
Co(OAc)2 40 h 43
6
7
Co(OAc)2 7 d
Co(NO3)2 3 d
30
46
—
—
—
—
8
9
10
11e
CoCl2
CoSO4
Co3O4
—
5 d
5 d
5 d
7 d
As a part of our ongoing research towards the synthesis of
enantiopure secondary alcohols by chiral copper (galactose
oxidase model), cobalt and iron complex catalyzed alcohol
oxidations,13 we are in the quest of finding an efficient,
economic, and easily available chiral metal catalyst for the
synthesis of enantiopure a-hydroxy esters through OKR under
mild reaction conditions where molecular oxygen will be used
as the sole oxidant (aerobic oxidation). In this communication,
Department of Chemistry, Indian Institute of Technology Madras,
Chennai, 600 036, India. E-mail: gsekar@iitm.ac.in;
Fax: +91 44 2257 4202; Tel: +91 44 2257 4229
a
Conditions: 5 mol% Co salt, 5 mol% ligand, 4 ml CH3CN and O2
b
(using oxygen balloon). Conversion was determined by 1H NMR
w Electronic supplementary information (ESI) available: Experimental
procedures and characterization data, full spectroscopic data for all
c
analysis of crude reaction mixture. %ee was determined by
1
compounds, H NMR and 13C NMR spectra for all the compounds,
d
HPLC using a Diacel chiralPAK AS-H column. s = krel(fast/slow)
=
all the 1H NMR spectra for conversion calculations and HPLC spectra
for %ee determination. See DOI: 10.1039/c0cc01917h
z This paper dedicated to Prof. S. Chandrasekaran.
e
ln[(1 ꢁ C)(1 ꢁ ee)]/ln[(1 ꢁ C)(1 + ee)]. Reaction only with TEMPO
without Co complex.
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun.

View Article Online
the (S)-BINAM based salen ligand L5–Co(OAc)2 complex was
used as the catalyst, the aerobic oxidative kinetic resolution
of (ꢀ)-methyl mandelate took place and it provided 43%
conversion (C)14 with selectivity (s)15 = 9 (entry 5). This
OKR provided 40% of methyl 2-oxo-2-phenylacetate 2 and
55% of recovered (R)-methyl mandelate with 52% ee. In this
oxidative kinetic resolution, the (S)-enantiomer of racemic
methyl mandelate was oxidized to the corresponding a-keto
ester and the slow-reacting (R)-enantiomer was recovered in
an optically active form.16 Upon replacing ligand L5 with L6,
the selectivity was reduced to 5.3 and the reaction took 7 days
to reach 30% conversion (entry 6).
Table 3 Effect of different ratio of chiral cobalt complex in OKR of
(ꢀ)-methyl mandelate
L5 Co(OAc)2
Entry (mol%) (mol%)
Conversion, %ee of
alcohol
Time C (%)
s
1
2
3
4
5
6
7
2.5
2.5
5
5
5
2.5
5
5
10
20
10
20
4 d
5 d
4 d
5 d
5 d
15 h
12 h
50.2
9
66
30
25
56.5
55
60
5
7.1
3.2
6.7
4.8
4.9
9.2
8.4
86
25
20
78
73
10
20
To increase the efficiency of the OKR of (ꢀ)-methyl
mandelate, the reaction was screened with several cobalt salts
with ligand L5 and the Co(OAc)2 turned out to be the best cobalt
salt (entry 5). When the OKR was carried out without Co
complex, the oxidation reaction did not provide even trace
amount of the oxidized product (entry 11). Then the reaction
was screened with several solvents to increase the efficiency of the
OKR. The results are summarized in Table 2. Although CH3CN,
benzene and xylene catalyzed the oxidation reaction, only
toluene turned out to be the solvent of choice by providing the
highest selectivity in shorter reaction time (Table 2, entry 1).
Next, we examined the effect of ratio of ligand L5 and
Co(OAc)2 in OKR of (ꢀ)-methyl mandelate and the results are
summarized in Table 3. Among the several combinations, 1 : 1
ratio of ligand L5 and cobalt acetate provided better selectivity
than 1 : 2 and 1 : 4 ratios (entries 1, 3, 6, and 7 vs. 2, 4 and 5).
Among 1 : 1 ratio catalyst, 10 mol% of ligand L5 and
10 mol% of Co(OAc)2 gave a maximum of s = 9.2 (entry 6). It
is important to mention that the reactivity and the selectivity
of the OKR were highly temperature dependent. When the
reaction was carried at 60 1C, it took 50 h for 53.6% conversion
and provided 7.1 selectivity (Table 4, entry 1). Increasing the
reaction temperature to 70 1C led to slightly increased reactivity
and selectivity (entry 2). At 80 1C, the reactivity increased and
it took just 15 h for 57% conversion, but the selectivity was
decreased to 9.2 (entry 3). Surprisingly, at 90 1C the reaction
took only 11 h for 49.3% conversion, with maximum observed
selectivity of 18.5 (entry 4). In following with the general
trend, at 100 1C, the reactivity increased, but the selectivity
was reduced to 5.9 (entry 5). To know the role of TEMPO, the
OKR was carried out without TEMPO at 90 1C, and the
selectivity was decreased to 7.3 (entry 6).
Table 4 Effect of temperature in the OKR of (ꢀ)-1
Entry
Temp./1C
Time/h
C (%)a
%ee of alcoholb
s
1
2
3
4
5
6
60
70
80
90
100
90
50
50
15
11
6
53.6
55.6
57.0
49.3
65.1
65.0
66
85
79
76
81
87
7.1c
13.8c
9.2c
18.5d
5.9c
17
7.3e
a
1
Conversion was determined by H NMR analysis of crude reaction
b
mixture. %ee was determined by HPLC using a Diacel chiralPAK
c
AS-H column. Selectivity (s) values represent an average of two
d
experiments. Selectivity (s) values represent an average of three
e
experiments. Reaction without TEMPO.
Using the above mentioned optimized reaction conditions, we
initiated our investigation into the scope of the L5–Co(OAc)2
complex catalyzed aerobic oxidative kinetic resolution of several
(ꢀ)-a-hydroxy esters, and the results are summarized in Table 5.
All kinds of aliphatic alcohol part of a-hydroxy esters such as
simple methanol, ethanol, long chain containing unbranched,
branched aliphatic alcohols, sterically hindered isopropyl alcohol
and highly sterically hindered tert-butyl alcohol were tolerated. A
maximum selectivity (s) = 31.9 at C = 60.9% was achieved for
the oxidative kinetic resolution when allylic alcohol is the alcohol
part of a-hydroxy ester with 99.9% ee for the recovered alcohol
(entry 11). When the phenyl ring of a-hydroxy ester contains both
the electron-releasing groups such as methoxy and methyl groups
(entries 12 and 13) and electron-withdrawing groups such as iodo
and bromo groups (entries 14 and 15), the OKR is well tolerated.
However, in the case of electron-releasing groups the reaction
took comparatively more time. All the (ꢀ)-a-hydroxy esters gave
the aerobic oxidation under mild conditions (60 or 90 1C, without
any base) with good to excellent selectivity (s = 6.4–31.9). In all
the oxidations, the (S)-enantiomers of (ꢀ)-a-hydroxy esters are
oxidized to corresponding a-keto esters and the slow reacting
(R)-enantiomers of a-hydroxy esters are recovered in highly
enantiomerically enriched form (78–99.9% ee).17
Table 2 Effect of solvents in the OKR of (ꢀ)-methyl mandelate
Entry
Solvent
Time
C (%)
%ee of alcohol
s
1
2
3
4
Toluene
CH3CN
Benzene
Xylene
DMF
15 h
5 d
30 h
20 h
5 d
56.5
14
48
48
—
—
—
78
6
51
5
—
—
—
9.2
2.3
5.7
1.2
—
5
In conclusion, we have described an efficient, economic, and
successful aerobic oxidative kinetic resolution of racemic
a-hydroxy esters to obtain enantiomerically-enriched a-hydroxy
esters, catalyzed by chiral cobalt complex, using molecular
6
THF
CHCl3
5 d
5 d
—
—
7a
a
Reaction was carried out at 60 1C.
c
Chem. Commun.
This journal is The Royal Society of Chemistry 2010

View Article Online
Table 5 OKR of (ꢀ)-a-hydroxy esters catalyzed by chiral cobalt
a-hydroxy esters in high yield and excellent enantioselectivity
(s up to 31.9). To the best of our knowledge, this is the first
report in the literature for chiral cobalt catalyzed oxidative
kinetic resolution of (ꢀ)-a-hydroxy esters. Efforts are
currently underway to provide detailed mechanistic insight
into the catalytic cycle and to expand the scope and synthetic
utility of the aerobic oxidative kinetic resolution.
a
complex using stoichiometric O2
Entry a-Hydrogen esters
Time/h C (%)b Yieldc %eed
s
We thank DST (Project No.: SR/S1/OC-06/2008), New
Delhi, for the financial support. SKA thanks CSIR, New
Delhi, for senior research fellowship.
1
11
16
8
49.7
66.5
68
47 (93) 78
31 (95) 97
31 (96) 91
40 (95) 84
30 (95) 90
38 (95) 84
31 (94) 94
40 (95) 82
28 (93) 98
24 (93) 98
19.9
2
11.1
7.2
10.9
7.3
9
Notes and references
1 (a) Comprehensive Organic Synthesis, ed. B. M. Trost, I. Fleming,
Pergamon Press, 1991; (b) F. A. Luzzio, Org. React., 1998, 53, 1;
(c) T. T. Tidwell, Org. React., 1990, 39, 297; (d) M. Hudlicky, in
Oxidations in Organic Chemistry, American Chemical Society,
Washington, DC, ACS Monograph Series, 1990.
2 (a) S. Itsuno, Org. React., 1998, 52, 395; (b) R. Noyori, O. Takeshi,
C. A. Sandoval and K. Muniz, Asymmetric Synth., 2007, 32.
3 (a) U. T. Bornscheuer and J. T. Kazlauskas, Survey of Enantio-
selective Lipase-Catalyzed Reactions, in Hydrolases in Organic
Synthesis, Wiley-VCH, Weinheim, 1999, pp. 65; (b) C.-H. Wong
and G. M. Whitesides, Enzymes in Synthetic Organic Chemistry,
Pergamon, New York, 1994.
3
4
9
57.5
67
5
8
6
9
60
4 (a) P. Muller, in Advances in Catalytic Processes, JAI Press Inc,
¨
Greenwich, CT, 1997, vol. 2, pp. 113–151; (b) D. E. J. E. Robinson
and S. D. Bull, Tetrahedron: Asymmetry, 2003, 14, 1407.
5 For general discussions, see: (a) H. B. Kagan and J. C. Fiaud, in
Topics in Stereochemistry, ed. E. L. Eliel, Wiley & Sons, New York,
1988, vol. 18, pp. 249–330; (b) A. C. Spivey, A. Maddaford and
A. Redgrave, Org. Prep. Proced. Int., 2000, 32, 333; (c) E. Vedeja
and M. June, Angew. Chem., Int. Ed., 2005, 44, 3974.
6 S. D. Rychnovsky, T. L. McLernon and H. Rajapakse, J. Org.
Chem., 1996, 61, 1194.
7 D. R. Jensen, J. S. Pugsley and M. S. Sigman, J. Am. Chem. Soc.,
2001, 123, 7475.
7
12
7
66
9.4
9.6
9.8
8.3
8
58
9
14
30
16
70
8 E. M. Ferreira and B. M. Stoltz, J. Am. Chem. Soc., 2001, 123, 7725.
9 T. A. Radosevich, C. Musich and D. F. Toste, J. Am. Chem. Soc.,
2005, 127, 1090.
10
11
73
10 Optically pure a-hydroxy esters, a-hydroxy carboxylic acid and
mandelic acid derivatives are important precursors toward enantio-
selective synthesis, drug development and biologically active anti-
bacterial compounds, see: (a) G. M. Coppola and H. F. Schuster,
a-Hydroxy Acids in Enantioselective Synthesis, VCH, Weinheim,
1997; (b) N. Momiyama and H. Yamamoto, J. Am. Chem. Soc.,
2005, 127, 1080; (c) A. Khalaj, H. Shadnia and M. Sharifzadeh,
Pharm. Pharmacol. Commun., 1998, 4, 373.
11 (a) S. W. Shiue, W. S. Mei, Q. K. Jun, S. M. Yogesh and
T. C. Chien, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 3522;
(b) T. C. Chien, B. Sampada, S. W. Shiue, D. P. Vijay, H. L. Ya,
Y. L. Cheng and Z. L. Way, J. Org. Chem., 2007, 72, 8175.
12 R. A. Shiels, K. Venkatasubbaiah and C. W. Jones, Adv. Synth.
Catal., 2008, 350, 2823.
60.9
36 (91) 99.9 31.9
12
13
14
56
6
61.9
67.7
74.5
66
35 (94) 80
29 (92) 90
6.8
7.0e
30
40
23 (94) 99.64 10.4e
13 (a) S. K. Alamsetti, S. Mannam, P. Muthupandi and G. Sekar,
Chem.–Eur. J., 2009, 15, 1086; (b) S. K. Alamsetti, P. Muthupandi
and G. Sekar, Chem.–Eur. J., 2009, 15, 5424; (c) P. Muthupandi,
S. K. Alamsetti and G. Sekar, Chem. Commun., 2009, 3288.
15
33 (95) 85
6.4e
1
14 Conversion was measured by H NMR analysis of crude reaction
mixture. See ESIw.
a
10 mol% Co(OAc)2, 10 mol% L5, O2, in 4 ml PhCH3 at 90 1C unless
otherwise mentioned. Conversion was determined by 1H NMR
c
analysis of crude reaction mixture. Isolated yield of enantiomerically
15 s = krel(fast/slow) = ln[(1 ꢁ C)(1 ꢁ ee)]/ln[(1 ꢁ C)(1 + ee)], where
b
C = conversion. See ESIw.
16 The absoluteconfiguration of a-hydroxy esters is determined from
sign of specific rotations in comparison with the literature values.
See ESIw.
17 In Table 5, all (ꢀ)-a-hydroxy esters gave (R)-a-hydroxy esters
as recovered starting materials except entries 8, 9, 11 and 14
(the absolute configuration of these compounds are not known in
the literature). Based on the stereochemical outcome of all other
reactions, we assume (ꢀ)-a-hydroxy esters of entries 8, 9, 11 and 14
were to be the same (R)-a-hydroxy esters as recovered starting
materials.
enriched a-hydroxy esters after silica gel column chromatography.
Numbers in parentheses are the total combined yields of the recovered
enantiomerically-enriched a-hydroxy ester and oxidized product
d
(a-keto ester). The %ee was determined by HPLC using a Diacel
e
chiralPAK AS-H column. Reaction was carried out at 60 1C.
oxygen as stoichiometric oxidant. The mild reaction conditions
of the catalytic system provided access to a wide range of
c
This journal is The Royal Society of Chemistry 2010
Chem. Commun.