A new nucleophilic catalyst for kinetic resolution of racemic sec-alcohols

Article abstract of DOI:10.1246/cl.2002.1114

A new tertiary amine-based nucleophilic catalyst, derived from a simple combination of commercially available compounds, affords good to excellent kinetic resolution of racemic sec-alcohols.

Full text of DOI:10.1246/cl.2002.1114

1114
Chemistry Letters 2002
A New Nucleophilic Catalyst for Kinetic Resolution of Racemic sec-Alcohols
Kyu-Sung Jeong,ꢀ Soong-Hyun Kim, Hyun-Jin Park, Kyoung-Jin Chang, and Kwan Soo Kimꢀ
Department of Chemistry, Yonsei University, Seoul 120-749, Korea
(Received August 7, 2002; CL-020667)
A new tertiary amine-based nucleophilic catalyst, derived
from a simple combination of commercially available com-
pounds, affords good to excellent kinetic resolution of racemic
sec-alcohols.
In recent years, remarkable advances have been made in the
development of chiral nucleophilic catalysts for the kinetic
resolution of racemic alcohols via acylations. Fu,1 Fuji,2
Oriyama,3 Spivey,4 and others5 designed and synthesized tertiary
amine-based nucleophilic catalysts, with which the kinetic
resolution of racemic alcohols has been successfully achieved.
On the other hand, Vedejs6 reported that chiral phosphines could
also serve as efficient nucleophilic catalysts for the enantioselec-
tive acylation of racemic alcohols. However, there are still some
limitations for practical applications of these catalysts, and
continued efforts are required for the development of a
synthetically more accessible catalyst that can efficiently resolve
racemic compounds. Herein we report a new, easily prepared
tertiary amine-based nucleophilic catalyst 1 that gives good to
excellent resolution of racemic alkylarylcarbinols.
The catalyst 1 consists of three basic components: 4-(N; N-
dimethylamino)pyridine (DMAP) as a catalytic component,
binaphthyl as a chiral segment, and Kemp’s triacid7 as a linker.
The chiral binaphthyl unit could be tethered to either the 2-(ortho)
or 3-position (meta) of the DMAP. We here chose the meta
position because the ortho substitution may deteriorate the
catalytic activity due to the resulting steric congestion around the
nitrogen where acyl-transfer actually occurs. The divergent
relationship between the meta position and the ring nitrogen of the
DMAP requires a specific linker molecule that can effectively
bring the chiral environment into close proximity with the
reaction site. Molecular modeling suggests that Kemp’s triacid
would be an ideal molecule for this purpose. The energy-
minimized structure of the catalyst 1, obtained by a Monte Carlo
search, is shown in Figure 1. It should be noted that by rotating the
C (meta in DMAP)-N (imide) bond in 1, the dimethylamino group
of the DMAP is placed away from the neighboring binaphthyl unit
to avoid otherwise severe steric repulsion between two groups.
Consequently, the ring nitrogen atom is completely surrounded
by the dissymmetric binaphthyl moiety.
Figure 1. Energy-minimized structure of 1 obtained by a Monte
Carlo search using MacroModel 5.5 (MM3 force field).
Synthesis of the catalyst 1 is straightforward and outlined in
Scheme 1. 3-Amino-DMAP 2 and anhydride acid 3 were heated at
reflux in pyridine to give DMAP-imide acid 4 (96%). After
activation of 4 with oxalyl chloride, coupling with (S)-(ꢁ)-1,10-
binaphthyl-2,20-diamine (47%), followed by acetylation (67%)
with acetic anhydride, gave the nucleophilic catalyst 1.8
We first examined the kinetic resolution of 1-phenylethanol
(0.3 M) in various conditions with the catalyst 1 (1 mol%) and
acetic anhydride (0.1–0.2 equiv).9 The enantioselectivity
strongly depends on solvent; in dichloromethane, benzene,
toluene, tetrahydrofuran, and dimethoxyethane, selectivity
factors10 (s = rate constant of fast-reacting enantiomer/rate
constant of slow-reacting enantiomer) are 1.4–2.1, while in tert-
amyl alcohol s ¼ 6:2 (0 ꢂC, 7% conversion).11 In addition, owing
to a small extent of the background reaction that occurs without
the catalyst 1, the s values slightly depend on other factors such as
the amounts of acetic anhydride and catalyst, temperature, and
added bases. General trends are as follows. First, s values
increased with increasing mol% of catalyst 1, but decreased with
increasing amounts of acetic anhydride. Second, higher s values
were observed at 0 ꢂC than at room temperature. Third, the
reaction proceeded slightly slower in the absence of bases
(triethylamine, diisopropylethylamine), but resulted in a small
increase of the selectivity factor (s).12
Scheme 1. Synthesis of the catalyst 1.
Copyright Ó 2002 The Chemical Society of Japan

Chemistry Letters 2002
1115
On the basis of these observations, we conducted the kinetic
resolution of various racemic sec-alcohols (0.3 M) in tert-amyl
alcohol with acetic anhydride (1 equiv) and the catalyst 1
(1 mol%) for 2–18 h at 0 ꢂC, and the results are summarized in
Table 1. As seen in entries 1–4, the stereoselectivity factor (s)
increases as the steric bulk of the alkyl groups
(Me < Et < i-Pr < t-Bu) increases. Especially, the stereoselec-
tivity factor is remarkably increasing up to 21 in the case of trans-
2-phenylcyclohexanol (entry 8).
References andNotes
1
G. C. Fu, Acc. Chem. Res., 33, 412 (2000) and references
therein.
2
a) T. Kawabata, M. Nagato, K. Takasu, and K. Fuji, J. Am.
Chem. Soc., 119, 3169 (1997). b) T. Kawabata, K. Yamamoto,
Y. Momose, H. Yoshida, Y. Nagaoka, and K. Fuji, Chem.
Commun., 2001, 2700.
3
4
5
a) T. Oriyama, K. Imai, T. Hosoya, and T. Sano, Tetrahedron
Lett., 39, 397 (1998). b) T. Oriyama, K. Imai, T. Sano, and T.
Hosoya, Tetrahedron Lett., 39, 3529 (1998).
a) A. C. Spivey, T. Fekner, S. E. Spey, and H. Adams, J. Org.
Chem., 64, 9430(1999). b) A. C. Spivey, T. Fekner, and S. E.
Spey, J. Org. Chem., 65, 3154 (2000).
a) E. Vedejs and X. Chen, J. Am. Chem. Soc., 119, 2584
(1997). b) E. R. Jarvo, G. T. Copeland, N. Papaioannou, P. J.
Bonitatebus, Jr., and S. J. Miller, J. Am. Chem. Soc., 121,
11638 (1999).
Table 1. Kinetic resolution of racemic sec-alcohols with
catalyst 1 (1 mol%)
6
7
a) E. Vedejs, O. Daugulis, and S. T. Diver, J. Org. Chem., 61,
430(1996). b) E. Vedejs and O. Daugulis, J. Am. Chem. Soc.,
121, 5813 (1999).
a) D. S. Kemp and K. S. Petrakis, J. Org. Chem., 46, 5140
(1981). b) D. P. Curran, K.-S. Jeong, T. A. Heffner, and J.
Rebek, Jr., J. Am. Chem. Soc., 111, 9238 (1989). c) K.-S.
Jeong, K. Parris, P. Ballester, and J. Rebek, Jr., Angew.
Chem., Int. Ed. Engl., 29, 555 (1990).
8
Incorporation of any chiral amine or alcohol to 4 may
potentially provide a DMAP-based chiral nucleophilic
catalyst, but thus far the catalyst 1 was found to be the most
efficient for the kinetic resolution of racemic alcohols.
Physical and spectroscopic properties of 1: mp: 252–
24
254 ꢂC; ½ꢀꢃ
¼ ꢁ92:9ꢂ (c ¼ 0:1, CHCl3); 1H NMR
D
(CDCl3, 500 MHz) ꢁ 8.56–8.53 (m, 2H), 8.15–8.14 (m, 1H),
8.14 (s, 1H), 8.09 (d, 1H), 8.04 (d, 1H), 7.94–7.91 (m, 2H),
7.44–7.40(m, 2H), 7.26 (m, 2H), 7.15 (s, 1H, NH), 7.15–7.13
(m, 1H), 6.97 (d, 1H), 6.83 (s, 1H, NH), 6.62 (d, 1H), 2.72–
2.69 (m, 1H), 2.69 (s, 6H), 2.00(d, 1H), 1.49 (s, 3H), 1.40(d,
1H), 1.33 (d, 1H), 1.26 (s, 3H), 1.09 (s, 3H), 0.99 (d, 2H), 0.57
(s, 3H); 13C NMR (CDCl3, 125.7 MHz) ꢁ 175.9, 175.6, 173.0,
169.5, 154.2, 151.5, 149.6, 136.5, 134.5, 132.4, 132.2, 131.6,
131.3, 130.8, 129.7, 128.7, 128.5, 127.5, 127.4, 125.6, 125.5,
125.2, 124.8, 122.8, 121.4, 120.7, 120.6, 120.4, 111.7, 45.2,
44.1, 44.0, 43.2, 42.0, 40.9, 40.8, 31.5, 26.4, 26.3, 24.3; IR
ꢂmax (KBr)/cmꢁ1 3411, 3372, 1734, 1691; Anal. Calcd for
C41H41N5O4: C, 73.74; H, 6.19; N, 10.49. Found: C, 73.72; H,
6.23; N, 10.53%.
In summary, considering that the catalyst 1 is prepared by a
simple combination of commercially available compounds, and
also effects kinetic resolution with a high enantioselection, this
can be regarded as a significant first step toward the development
of a practical nucleophilic catalyst. Currently, work is underway
on the modification of 1, aiming at increasing further selectivity
factors and revealing the origin of the enantioselectivity.
9
The conversion rate of the reaction between 1-phenylethanol
and acetic anhydride with the catalyst 1 was nearly equal to or
slightly slower than that with the DMAP itself under similar
reaction conditions. This is possibly because the reaction
center (ring nitrogen) of 1 is sterically more congested than
the DMAP.
10H. B. Kagan and J. C. Fiaud, Top. Stereochem., 18, 249
(1988).
This research was financially supported by the Center for
Molecular Design and Synthesis (CMDS) at KAIST. We are
grateful to Prof. J. S. Moore for allowing the use of his
MacroModel program and Prof. M. Gin for proofreading of this
manuscript.
11 For a similar trend of solvent effects on the kinetic resolution,
see: J. C. Ruble, J. Tweddell, and G. C. Fu, J. Org. Chem., 63,
2794 (1998).
12 For the same observation, see: B. Tao, J. C. Ruble, D. A. Hoic,
and G. C. Fu, J. Am. Chem. Soc., 121, 5901 (1999).