ee (that is, 86% (R)-enantiomer). Compound 5 was converted
to 3c by treatment with KOH in dioxane at 60 °C, which
precipitated out with acidification. To determine if the
hydrolysis condition could racemize the chiral center of 3c,
we converted 3c back to 5 by reaction with HCl (g) and
MeOH/Et2O (1:3). HPLC analysis of 5 showed a slight
increase in ee (∼3.5%). This demonstrates that no racem-
ization took place during the hydrolysis of 5 to 3c.
On the basis of the interaction of the enantiomeric sensors
(R)- and (S)-1 with 3c of various enantiomeric compositions,
a linear relationship between the fluorescence intensity
difference ∆I [∆I ) (IS/IS0) - (IR/IR0); IS, fluorescence
intensity of (S)-1 in the presence of the acid; IR, fluorescence
intensity of (R)-1 in the presence of the acid; IS0 and IR0,
fluorescence intensity of (R)-/(S)-1 without the acid] and the
ee of 3c is established (Figure 3). Thus, using Figure 3, one
The optically active (R)- and (S)-3c obtained above were
allowed to interact with the fluorescent sensors (S)- and (R)-
1. We found that in a ternary solvent of THF/hexane/benzene
(1:5.5:18.5), the sensors (2.0 × 10-5 M) showed enantio-
selective fluorescent responses to the chiral acids (2.0 × 10-4
M) (Figure 1). THF was used to dissolve 3c, and the
Figure 3. Relationship between ∆I and the enantiomeric composi-
tion of mandelic acid.
Figure 1. Fluorescence spectra of (R)-1 (2.0 × 10-5 M) with (R)-
and (S)-3c (2.0 × 10-4 M) (λexc ) 332 nm, slit ) 3.5; 6.5 nm).
should be able to determine the enantiomeric composition
of a given sample of 3c from ∆I.
We studied the conversion of 2c to 3c by the TMSCN
addition in the presence of the chiral ligands (S)-/(R)-69 and
(+)-/(-)-DIPT in combination with Ti(OiPr)4 under various
conditions as summarized in Table 2. The product 3c was
combination of benzene and hexane was necessary for the
enantioselective fluorescent response. Although the enantio-
selectivity of the sensor for 3c is significantly lower than
that for the unsubstituted mandelic acid, it is still useful for
ee determination (vide infra).6
Samples of 3c (2.0 × 10-4 M) with various enantiomeric
compositions were prepared, and their interaction with (R)-1
(2.0 × 10-5 M) in the ternary solvent system was studied.
A monotonic relation between the fluorescence intensity and
the enantiomeric composition of the acid was obtained
(Figure 2). We also studied the interaction of the samples
with (S)-1 and observed a mirror image relationship for the
fluorescence responses. This confirmed the observed chiral
recognition of 3c by the fluorescent sensor.
Table 2. Asymmetric TMSCN Addition to 2c and Fluorescent
Ee Determination of Product 3c
ligand
(mol %)
Ti(OiPr)4
(mol %) (IS/IS0 - IR/IR0
∆I )
ee
(%)
entry solvent
)
1
2
3
4
5
6
7
8
9
CH2Cl2 (+)-DIPT (40)
CH2Cl2 (-)-DIPT (40)
CH2Cl2 (S)-6 (10)
CH2Cl2 (S)-6 (10)
CH2Cl2 (S)-6 (10)
40
40
8
10
12
10
10
10
10
2
1.92
-1.81
2.06
1.33
0.38
0.99
0.11
0.95
-1.26
-1.31
79
-77
84
54
14
39
3
38
-54
-56
THF
ether
(S)-6 (10)
(S)-6 (10)
toluene (S)-6 (10)
CH2Cl2 (R)-6 (10)
CH2Cl2 (R)-6 (10)
10
isolated in the range of 60-70% yield simply by centrifuga-
tion and filtration because of the insolubility of 3c in the
absence of THF. The 10 samples of 3c obtained from the
above catalyst screening were treated with the fluorescent
(7) Hayashi, M.; Matsuda, T.; Oguni, N. J. Chem. Soc., Perkin Trans. 1
1992, 22, 3135-3140. For the modification, see Supporting Information.
(8) Tanaka, K.; Mori, A.; Inoue, S. J. Org. Chem. 1990, 55, 181-185.
(9) Li, Z.-B.; Rajaram, A. R.; Decharin, N.; Qin, Y.-C.; Pu, L.
Tetrahedron Lett. 2005, 46, 2223-2226.
Figure 2. Fluorescence responses of (R)-1 (2.0 × 10-5 M) with
3c (2.0 × 10-4 M) at various R compositions (λexc ) 332 nm, slit
) 3.5; 6.5 nm).
Org. Lett., Vol. 7, No. 16, 2005
3443