P. Chen, W. Yang / Tetrahedron Letters 55 (2014) 2290–2294
2293
Table 2
For discussion of data shown in Table 1, we take 1a (entry 1,
E = 59.0) as the parent compound for comparison with the others.
We can see the phenomena as follows:
Configuration assignment of 2a, 2h, 2o, 2q, 3b, 3d, and 3p by comparison with the
reported [ data
a
]
D
Compound
Measured values
Reported values
Ref.
[
a
]
ee (%)
90.1
[
a
]
ee (%)
D
D
R
3
R
R
COOMe
OH
2a
2h
2o
2q
3b
3d
3p
+100.04
(c 0.75, CHCl3)
+115.7
+124.6
(c 0.82, CHCl3)
+101.6
(c 1.09, CHCl3)
-7.5
(c 1.3, CHCl3)
+124.6
71 (S)
16
78
79
86
80
81
82
2
n
96.1
93.8
92.4
89.8
18.2
94.3
100 (S)
>99 (S)
71 (S)
90 (R)
62 (R)
99 (R)
1
(c 0.98, CHCl3)
À6.21
(c 0.5, CHCl3)
+110.4
(1) Compounds 1b, 1f–1h, 1j, and 1l (entries 2, 6–8, 10 and 12,
E = 70.1 to 192.5), which bear a substituent of moderate size
on meta- or para-position of the phenyl group (n = 0, R3 = H,
R1 or R2 with moderate size), gave higher E values, while
compounds 1m (entry 13, E = 15.0) and 1n (entry 14, no
reaction), which bear a substituent of larger size on para-
position (n = 0, R2 = R3 = H, R1 with large size), gave signifi-
cantly lower E values. This fact implies that size of substitu-
ent (i.e. the size of R group) has an influence on the
interaction of the substrate with the enzyme.
(2) When the size of substituents on the phenyl group is small
enough, such as F and Br (whose atomic radius is 42 and
94 pm, respectively, while that of H is 53 pm),77 steric hin-
drance would not be a key factor for the interaction of the
substrates and the enzyme, while electron-withdrawing
effect would make some positive contribution to the E val-
ues (entries 6–8, and 10, E = 70.1 to 192.5, while in entry
1, E = 59.0). However, no reaction occurred when the man-
delate substrates have a fluorine substituent on ortho-posi-
(c 0.97, CHCl3)
À93.1
(c 0.82, CHCl3)
À82.6
(c 0.74, CHCl3)
À57.2
(c 1.00, CHCl3)
À45
(c 0.59, CHCl3)
À23.6
(c 4.0, CHCl3)
À28.7
(c 1.07, CHCl3)
(c 1.61, CHCl3)
OH
OH
M
M
L
L
unfavored enentiomert
favored enantiomer
L= large size group; in this paper, L=aryl group
M= middle size groip; in this paper, M= alkynyl group
Figure 2. A schematic presentation of Kazlauskas’ rule.
tion (n = 0, R3 = F) (entries
attributed to that six-membered ring intramolecular
hydrogen-bond structure was formed between F and the
-hydroxyl group of the methyl mandelate (Fig. 1), which
9 and 11). This might be
a
a
inhibited the acylation of the hydroxyl group.
(3) Substitution on ortho-position of the phenyl group (n = 0,
R1 = R2 = H, R3 with large size) (entry 3, compound 1c,
E = 1.5) causes stereo-hindrance to the hydroxyl group, lead-
ing to dramatical decrease of E values.
Configuration of the other products (2c, 2e, 2f, 2g, 2j, 2l, 2m,
2n), where the aryl groups stand for the large size group and the
ester groups stand for the middle size group, was proposed to have
the same designation as the known ones based on Kazlauskas’ rule
(Fig. 2).83,84
In summary, we have achieved the KR of seventeen racemic
methyl mandelic esters via stereoselective acylation using lipase
PS-30 as the catalyst. Majority of the substrates gave values of E
close to or larger than 50, the best one achieves 197.5.
(4) In compounds 1p and 1q (n = 1 and 2, R1 = R2 = R3 = H, R1)
(entries 16 and 17, E = 57.8 and 197.5, respectively), the phe-
nyl group is relatively far away from the hydroxyl group,
leading to the decrease of stereo-hindrance around the
hydroxyl group, which is in favor of the interaction of the
substrate with the enzyme thus giving the best resolution.
(5) Ethyl mandelate (entry 18, compound 1r) gave lower E value
(28.0) than that of methyl mandelate (entry 1, E = 59.0). No
reaction occurred when the mandelic esters have alkoxyl
groups larger than ethoxy, therefore they could not be used
as the substrate of this enzymatic kinetic resolution reaction.
Acknowledgments
This work is financially supported by a grant of The National
Natural Science Foundation of China (NSFC, Grant No. 21272305),
The Natural Science Foundation of Shanghai Municipality (Grant
No. 12ZR1400200) and The Fundamental Research Funds for the
Central Universities.
Configuration of some acylated products (2a, 2h, 2o, and 2q)
and the unreacted alcohols (3b, 3d, and 3p) is further confirmed
by comparison of the observed optical rotations with those re-
ported in the literatures (Table 2). Some other products have been
reported in the references (2b–f, 2q in Ref. 85; 2l, 2p in Ref. 13,
Supplementary data
respectively), but their [
sponding references.
a
]
D
data could not be found in the corre-
Supplementary data associated with this article can be found, in
H
O
References and notes
F
OMe
O
Figure 1. A proposed intramolecular hydrogen-bond between F and a-hydroxyl
group of methyl 2-fluoro-mandelate.