4
43
t-butyl-CH3
+
SiOCH2CH3
O
SiCH CH
N
2
2
O
Ph
pyridine-H
Ph Ph CH
OH
C
O
Ph CPh3 CH Ph
OH
H O
2
N
Ph
t-butyl-CH3
3
4
5
6
7
&
SOCH CH
3
i
2
SiCH2
X
50
O
H
C
CONHPh
OH
Ph-H
2.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2
(ppm)
Co(acac)3
9
SiOCH
CONHPh
Ph
2
CF3
O
glucose-H
N-H
8
10
11
11
10
9
8
7
6
5
4
3
2
1
0
(
ppm)
Figure 4. Structures of racemates 311.
Figure 3. 1H NMR (500 MHz) spectrum of the amylose derivative 2b in
pyridine-d5 at 80 °C.
O
Ph
the catalyst) were added to the solution at 40 °C, and the reaction
was continued with stirring for 21 days until the esterification of
the 2-hydroxy group was complete. The reaction mixture was
then dispersed into a large excess of 2-propanol, and the product
t1
7
t2
(
A-1) was recovered as an insoluble fraction in 98% yield. The
t0
obtained mono ester A-1 was dissolved in a mixture of N,N-
dimethylacetamide, lithium chloride, and pyridine. After being
completely dissolved, 3,5-dichlorophenyl isocyanate was added
and the reaction was continued for 6 h at 80 °C. A calculated
amount of 3-(triethoxysilyl)propyl isocyanate was then added
and allowed to react for 16 h at 80 °C. The ratio of the 3,5-
dichlorophenylcarbamate to 3-(triethoxysilyl)propylcarbamate
residues was controlled by the amount of 3-(triethoxysilyl)pro-
pyl isocyanate. Finally, the underivatized hydroxy groups were
allowed to react with an excess of 3,5-dichlorophenyl isocyanate
for 7 h at 80 °C. Two amylose derivatives 2a and 2b bearing
different amounts of the 3-(triethoxysilyl)propylcarbamate res-
idues at 3- and 6-positions were isolated as a methanol-insoluble
fraction. The structures of the derivatives were confirmed by
0
20
40
Elution time / min
Figure 5. Chromatogram for the resolution of 7 on 2b-IM.
((t2 ¹ t0)/t0), were estimated to be 2.20 and 4.58, respectively,
which resulted in the separation factor ¡ (k ¤/k ¤) to be 2.08.
2
1
The chromatographic resolution of 9 racemates 311 on 2a-
1
2
IM and 2b-IM are summarized in Table 1. For comparison,
resolution on commercial polysaccharide-based chiral columns,
Chiralpak IA, IB, and IC, which are immobilized CPMs
consisting of amylose tris(3,5-dimethylphenylcarbamate), cellu-
lose tris(3,5-dimethylphenylcarbamate) and cellulose tris(3,5-
dichlorophenylcarbamate) as the chiral selectors, are also
included. For some racemates, the derivatives 2a and 2b bearing
a 4-tert-butylbenzoyl group at the 2-position showed equivalent
or higher chiral recognition compared to Chiralpak IA, IB, and
IC. Racemates 5 and 7, which cannot be resolved efficiently on
the three commercial immobilized columns, were better resolved
on both 2a and 2b. Racemate 9, which is rather difficult to
resolve on Chiralpak IA and IB, was sufficiently resolved on
both 2a and 2b, and these two CPMs both exhibited a much
higher recognition ability for racemate 7 compared to other
commercially available Daicel columns (Table 2). For prepara-
tion of the immobilized CPM with high chiral recognition as
1
1
H NMR and elemental analysis. The H NMR spectrum of 2b is
shown in Figure 3, from which the ratio of (3,5-dichlorophen-
ylcarbamate)/[3-(triethoxysilyl)propylcarbamate] could be esti-
mated from (aromatic proton)/(SiCH2) intensity ratio.
For immobilization, two amylose derivatives 2a and 2b
(0.35 g) were first dissolved in THF (8 mL) and coated onto
silica gel (Daiso gel SP-1000, 7 ¯m, 1.40 g) in the previous
7
manner. The 2a- and 2b-coated silica gels (0.65 g) were then
added into a mixture of ethanol, water, and trimethylsilyl
chloride (6/1.5/0.1, v/v/v). After stirring for 10 min at 110 °C,
the immobilized CPMs (2a-IM and 2b-IM) were extensively
washed with THF and acetone and dried at 50 °C in vacuo. The
immobilization efficiency was estimated by thermogravimetric
(
TG) analysis.
The 2a-IM and 2b-IM were packed in a stainless-steel tube
25 © 0.20 cm i.d.) as a slurry. The plate numbers of the packed
3
(
well as coated CPM, it is preferable for the R residue content to
columns were 13001800 for benzene using a hexane/2-
propanol (90/10, v/v) mixture as the eluent at the flow rate of
be as low as possible. However, as shown in Table 2, the
3
immobilization efficiency decreased as the R content decreased.
¹
1
3
0
.1 mL min . 1,3,5-Tri-tert-butylbenzene was used as a non-
Therefore, the CPM obtained from 2b bearing 2% R at 3- and
11
3
retained compound to estimate the dead time (t0).
6-positions seems to be better than that from 2a bearing 1% R .
The chiral recognition abilities of the new amylose
derivatives were evaluated with 9 racamates 311 (Figure 4).
Figure 5 shows the chromatogram of the resolution of racemic
If the immobilization efficiency is low, bare silica surfaces must
reduce the chiral recognition through nonchiral interaction.
Under the same chromatographic conditions using a
standard eluent consisting of hexane/2-propanol mixtures, 2b-
IM showed chiral recognition comparable to 1-coated CPM
(Table 3). The higher order structure of 2b after the immobi-
lization seems to be similar to that of 1-coated on the silica gel.
2-phenylcyclohexanone (7) on the amylose derivative 2b. The
enantiomers were eluted at the retention time t1 and t2 with a
baseline separation. The dead time (t ) was determined to be
0
7.78 min. The retention factors, k1¤ [(t1 ¹ t0)/t0] and k2¤
Chem. Lett. 2010, 39, 442444
© 2010 The Chemical Society of Japan
Shen, Jun
