Immobilization and chiral recognition of 3,5-dimethylphenylcarbamates of cellulose and amylose bearing 4-(trimethoxysilyl)phenylcarbamate groups

Article abstract of DOI:10.1002/chir.20722

A small amount of 4-(trimethoxysilyl)phenyl groups was randomly introduced onto the 3,5-dimethylphenylcarbamates of cellulose and amylose by a onepot method. The obtained derivatives were then effectively immobilized onto silica gel as chiral packing materials (CPMs) for high-performance liquid chromatography through intermolecular polycondensation of the trimethoxysilyl groups. The effects of the amount of 4-(trimethoxysilyl)phenyl groups on immobilization and enantioseparation were investigated. Also, the solvent durability of the immobilized-type CPMs was examined with the eluents containing chloroform and tetrahydrofuran. When these eluents were used, the chiral recognition abilities of the CPMs for most of the tested racemates were improved to some extent depending on the compounds.

Full text of DOI:10.1002/chir.20722

CHIRALITY 22:165–172 (2010)
Immobilization and Chiral Recognition of
3,5-Dimethylphenylcarbamates of Cellulose and Amylose
Bearing 4-(Trimethoxysilyl)phenylcarbamate Groups
SHOUWAN TANG,1,2 TOMOYUKI IKAI,1 MASASHI TSUJI,1 AND YOSHIO OKAMOTO1,3
*
1EcoTopia Science Institute, Nagoya University, Furo-Cho, Chikusa-Ku, Nagoya, Japan
2Department of Chemistry, School of Pharmaceutics and Chemical Engineering, Taizhou College, Linhai, Zhejiang,
People’s Republic of China
3College of Material Science and Chemical Engineering, Harbin Engineering University, Harbin, People’s Republic of China
ABSTRACT
A small amount of 4-(trimethoxysilyl)phenyl groups was randomly
introduced onto the 3,5-dimethylphenylcarbamates of cellulose and amylose by a one-
pot method. The obtained derivatives were then effectively immobilized onto silica gel
as chiral packing materials (CPMs) for high-performance liquid chromatography
through intermolecular polycondensation of the trimethoxysilyl groups. The effects of
the amount of 4-(trimethoxysilyl)phenyl groups on immobilization and enantiosepara-
tion were investigated. Also, the solvent durability of the immobilized-type CPMs was
examined with the eluents containing chloroform and tetrahydrofuran. When these elu-
ents were used, the chiral recognition abilities of the CPMs for most of the tested race-
mates were improved to some extent depending on the compounds. Chirality 22:165–
C
VV
172, 2010.
2009 Wiley-Liss, Inc.
KEY WORDS: chiral recognition; chiral separation; chiral stationary phase; chiral
packing materials; high-performance liquid chromatography; HPLC;
immobilization; polysaccharide derivative
INTRODUCTION
Till now, various interesting methods, such as the
bifunctional reagent method using diisocyanate,15,16
radical polymerization,17–20 photoirradiation,21 enzyme-cat-
alyzed polymerization,22 and so on, have been developed
for the immobilization of polysaccharide derivatives onto
silica gel. Moreover, Chiralpak IA, Chiralpak IB, and Chir-
alpak IC (Daicel, Tokyo, Japan), which are composed of
immobilized amylose 3,5-dimethylphenylcarbamate, cellu-
lose 3,5-dimethylphenylcarbamate, and cellulose 3,5-
dichlorophenylcarbamate, respectively, have been com-
mercialized as CPMs.23–25
Polysaccharide derivatives-based chiral packing materi-
als (CPMs), such as the phenylcarbamates, especially 3,5-
dimethylphenylcarbamates 1 and 2 in Figure 1, and ben-
zoates of cellulose and amylose, exhibit high chiral recog-
nition abilities toward a wide range of chiral compounds
and have been recognized as the most powerful CPMs for
direct separation of enantiomers by high-performance liq-
uid chromatography (HPLC).1–6 These polysaccharide-
based CPMs have been traditionally prepared by coating
the derivatives onto macroporous silica gel. Because of
their coated nature, solvents such as tetrahydrofuran
(THF) and chloroform, which can dissolve or swell the
polysaccharide derivatives, cannot be used as eluents.
Therefore, only a limited range of solvents, such as a hex-
ane–alcohol mixture for normal phase separation and a
water–acetonitrile mixture for reversed phase separation,
can be used as the eluents for these coated-type CPMs.
The strict limitation of eluent selection is sometimes a seri-
ous problem for the efficient analytical and preparative res-
olution of enantiomers.7,8 On the other hand, better resolu-
tion,9,10 reversed elution orders of enantiomers,11 and
higher sample solubility of some racemates8,12–14 can be
achieved by utilizing the prohibited eluents. To confer uni-
versal solvent compatibility upon these kinds of CPMs,
synthesis of immobilized-type CPMs is a direct approach.
This can significantly broaden the choice of solvents as
eluents or sample dissolving reagents.
Recently, we found that polysaccharide 3,5-dimethylphe-
nylcarbamates bearing a small amount of alkoxysilyl
groups could be effectively immobilized onto silica gel
through intermolecular polycondensation of these groups
under an acidic condition.26–28 For instance, 89% of the cel-
lulose derivative (3 in Fig. 1) and 86% of the amylose de-
rivative (4 in Fig. 1) could be immobilized onto silica gel
when only 2% and 1% of 3-(triethoxysilyl)propyl groups
Contract grant sponsor: Daicel Chemical Industries.
*Correspondence to: Prof. Yoshio Okamoto, EcoTopia Science Institute,
Nagoya University, Furo-Cho, Chikusa-Ku, Nagoya 464-8603, Japan; Col-
lege of Material Science and Chemical Engineering, Harbin Engineering
University, 145 Nantong St., Harbin 150001, People’s Republic of China.
E-mail: okamoto@apchem.nagoya-u.ac.jp
Received for publication 11 December 2008; Accepted 13 February 2009
DOI: 10.1002/chir.20722
Published online 19 May 2009 in Wiley InterScience
(www.interscience.wiley.com).
C
VV
2009 Wiley-Liss, Inc.

166
TANG ET AL.
phenyl trimethoxysilane from Gelest (PA, USA). Triphos-
gene and trimethylsilyl chloride were purchased from TCI
(Tokyo, Japan).
Synthesis of 4-(Trimethoxysilyl)phenyl isocyanate
4-(Trimethoxysilyl)phenyl isocyanate was synthesized
from 4-aminophenyl trimethoxysilane and triphosgene by
a conventional method, as follows. Triphosgene (2.67 g,
9.0 mmol) was dissolved in 50 ml of dry THF. A mixture
of 4-aminophenyl trimethoxysilane (3.84 g, 18 mmol) and
triethylamine (5.6 ml, 40 mmol) in dry THF (40 ml) was
slowly dropped into the aforementioned solution kept in
an ice bath. After the dropping, the temperature was
raised to 708C, and the mixture was allowed to react at
708C for 2.5 h. After the reaction, triethylhydrogenammo-
nium chloride was removed by filtration, and THF was
evaporated from the filtrate. After most of the THF was
removed, 4-(trimethoxysilyl)phenyl isocyanate was then
collected by distillation under reduced pressure. Yield 1.2
g (28% from 4-aminophenyl trimethoxysilane). IR (cm21),
2270 (N¼¼C¼¼O); 1H NMR d (ppm) (300 MHz, CDCl3),
3.60 (9H, s, Si(OCH3)3), 7.05–7.13 (2H, d, 2 3 Ar-H),
7.54–7.63 (2H, d, 2 3 Ar-H). Anal. calcd for C10H13NO4Si:
C, 50.19; H, 5.48; N, 5.85; Found: C, 50.07; H, 5.46; N,
6.08.
Fig. 1. Structures of cellulose (1, 3, and 5) and amylose (2, 4, and 6)
derivatives.
were introduced on the derivatives, respectively. The
obtained CPMs showed high chiral recognition abilities
and universal solvent compatibility.
One-Pot Synthesis of Polysaccharide Derivatives Bearing
4-(Trimethoxysilyl)phenyl Groups
According to the previous procedure,27 the derivatives
of cellulose 5 and amylose 6 bearing 4-(trimethoxysilyl)-
phenyl groups were synthesized by a one-pot method as
indicated in Figure 2. After cellulose was dissolved in a
mixture of DMA/LiCl/pyridine, 3,5-dimethylphenyl isocya-
nate (0.83 equiv relative to the hydroxyl groups of cellu-
lose) was added, and the mixture was allowed to react at
808C for 8 h. Various amounts of 4-(trimethoxysilyl)phenyl
isocyanate were then added and allowed to react at 808C
for 13 h. Finally, the residual hydroxyl groups of cellulose
In the present study, to investigate the influence of the
trialkoxysilyl groups on the performance of CPMs, the
polysaccharide derivatives 5 and 6 bearing a small
amount of 4-(trimethoxysilyl)phenyl groups in Figure 1
were synthesized by a one-pot method and immobilized
onto silica gel by intermolecular polycondensation of the
trimethoxysilyl groups. The relationships among the
amount of 4-(trimethoxysilyl)phenyl groups introduced,
immobilization, and enantioseparation of the obtained
CPMs were investigated using a hexane/2-propanol mix-
ture as eluent. Also, the chiral recognition abilities of the
obtained CPMs were evaluated with the eluents containing
chloroform and THF.
EXPERIMENTAL
Materials
Cellulose (Avicel, DP 5 200) was purchased from
Merck (Darmstadt, Germany). Amylose (DP 5 300), 3,5-
dimethylphenyl isocyanate, Chiralpak IA, Chiralpak IB,
Chiralcel OD, and Chiralpak AD-H (25 3 0.46 cm i.d.)
were kindly provided by Daicel Chemical Industries
(Tokyo, Japan). The macroporous silica gel (Daisogel
SP-1000) with a mean particle size of 7 lm and a mean
pore diameter of 100 nm was a gift from Daiso Chemical
(Osaka, Japan). The dehydrated solvents such as N,N-
dimethylacetamide (DMA), pyridine, and THF were
obtained from Kanto (Tokyo, Japan). Triethylamine was
obtained from Kanto (Tokyo, Japan), anhydrous lithium
Fig. 2. Scheme of the synthesis of cellulose (5) and amylose (6) deriv-
chloride from Nakalai Tesque (Kyoto, Japan), and 4-amino- atives bearing 4-(trimethoxysilyl)phenyl groups.
Chirality DOI 10.1002/chir

IMMOBILIZATION AND CHIRAL RECOGNITION OF DERIVATIVES
167
were treated with an excess of 3,5-dimethylphenyl isocya- quantitative reaction of the 4-(trimethoxysilyl)phenyl iso-
nate (0.83 equiv relative to the hydroxyl groups of cellu- cyanate with the hydroxyl groups of polysaccharide.
lose) at 808C for 8 h. The cellulose derivatives 5a–c were Through this procedure, the content of 4-(trimethoxysilyl)-
obtained as methanol-insoluble fractions. Also, the amy- phenyl groups introduced onto the polysaccharides can be
lose derivatives 6a–c were synthesized by a similar controlled by the amount of isocyanate added.
method.
Six polysaccharide derivatives (see Fig. 1) were synthe-
sized using the one-pot method. The H NMR spectra of
1
the obtained cellulose (5b) and amylose (6a) derivatives
in Figure 3 clearly indicate that the introduction of the 4-
(trimethoxysilyl)phenyl groups was successful, as shown
by the triplet signals (SiOCH3) at about 3.55–3.70 ppm.
The content of 4-(trimethoxysilyl)phenyl groups in 5a–c
and 6a–c was estimated to be 0.7, 1.2, 2.1% and 0.9, 1.3,
2.5%, respectively. The ratio of (3,5-dimethylphenylcarba-
mate)/(4-(trimethoxysilyl)phenylcarbamate) (R1/R3) was
determined from the ratio of (Ar-CH3)/(SiOCH3) in the 1H
NMR spectra. It should be pointed out that because the
peaks of glucose units of the cellulose derivatives overlap
those of SiOCH3 to some extent (Fig. 3a), the R3 contents
of these derivatives calculated by this method may not be
as accurate as those of the amylose derivatives. In fact,
when the cellulose and amylose derivatives were prepared
under the same conditions, the R3 contents of the cellulose
derivatives were always smaller than those of the amylose
derivatives. However, the R3 contents of amylose deriva-
tives as well as those of cellulose derivatives determined
by this method were found to be proportional to the
amount of 4-(trimethoxysilyl)phenyl isocyanate added.
The derivatives were coated and immobilized onto bare
silica gel to obtain the CPMs, and their chiral recognition
abilities were evaluated by HPLC using 10 racemates
shown in Figure 4.
Immobilization of Polysaccharide Derivatives
onto Silica Gel
The immobilization of polysaccharide derivatives having
4-(trimethoxysilyl)phenyl groups onto bare silica gel was
similarly performed as reported previously.26–28 In brief,
each polysaccharide derivative (0.35 g) in THF (8 ml) was
coated onto bare silica gel (1.40 g). The 5- or 6-coated
silica gel (0.65 g) was then suspended in a mixture of etha-
nol (6 ml), water (1.5 ml), and trimethylsilyl chloride (0.1
ml) and allowed to react at 1108C for 10 min. After being
washed extensively with THF and dried, the 5- and 6-im-
mobilized CPMs were obtained. The immobilization effi-
ciency was calculated from the organic contents of the
CPMs before and after immobilization estimated by ther-
mogravimetric (TG) analysis.
Apparatus and Chromatography
1
The 300 and 500 MHz H NMR spectra were measured
on a Varian Mercury and a Varian INOVA-500 spectrome-
ters, respectively. The TG analysis was performed using a
Seiko EXSTRA 6000 system (Seiko, Chiba, Japan). The
HPLC experiments were carried out on a Jasco chromato-
graph (Jasco, Tokyo, Japan) consisting of a multiwave-
length detector (Jasco MD-910), a circular dichrosim de-
tector (Jasco CD-1595 YS), an intelligent HPLC pump
(Jasco PU-1580), and an intelligent column thermostat
(Jasco CO-1560). The obtained CPMs were packed into
stainless-steel columns (25 3 0.20 cm i.d.) by a slurry
packing method. The dead time (t0) was determined using
1,3,5-tri-t-butylbenzene as the nonretained compound.29
The plate numbers of the packed columns were about
2000 for benzene using hexane/2-propanol (90/10, v/v) as
eluent at a flow rate of 0.1 ml/min. The columns were
maintained at 258C by a column thermostat during the
chromatographic experiments.
Effect of 4-(Trimethoxysilyl)phenyl Groups on Chiral
Recognition Abilities of Cellulose Derivatives
The 4-(trimethoxysilyl)phenyl groups introduced onto
cellulose derivative 1 for the intermolecular polycondensa-
tion may significantly influence the chiral recognition abil-
ity of 1. Therefore, 20 wt % of the derivatives 5a–c were
coated onto bare silica gel using a conventional method,
and the resolution results of the racemates on these
CPMs are shown in Table 1, together with those on the
commercially available Chiralcel OD coated with cellulose
derivative 1, having no 4-(trimethoxysilyl)phenyl group.
The elution orders for the enantiomers were the same on
the 5-coated CPMs, whereas the a values, which repre-
sent the chiral recognition ability, were slightly decreased
RESULTS AND DISSUSION
Synthesis of 3,5-Dimethylphenylcarbamates of Cellulose
and Amylose Bearing 4-(Trimethoxysilyl)phenyl Groups
Polysaccharide 3,5-dimethylphenylcarbamates contain- with the increased content of R3 in 5. Compared with Chir-
ing a small amount of 3-(triethoxysilyl)propyl groups have alcel OD, slightly smaller a values were observed on the
been prepared by a one-pot method.26,27 In a similar way, 5-coated CPMs, except for racemates 7, 11, and 16,
the derivatives of cellulose 5a–c and amylose 6a–c bear- whose values were larger. Because the R3 content of 5a
ing 4-(trimethoxysilyl)phenyl residues were synthesized was smaller than those of 5b and 5c, the a values on the
by adding successively 3,5-dimethylphenyl isocyanate, 5a-coated CPM were more similar to those on Chiralcel
4-(trimethoxysilyl)phenyl isocyanate, and 3,5-dimethyl- OD. One point should be mentioned here is that the parti-
phenyl isocyanate as indicated in Figure 2. The roles of cle sizes of the silica gels used for the commercially avail-
the first addition of 3,5-dimethylphenyl isocyanate are not able columns are 10 lm for Chiralcel OD and 5 lm for
only to convert the hydroxyl groups of polysaccharide into Chiralpak IA and Chiralpak IB, which are different from
phenylcarbamate residues but also to eliminate a small that on the CPMs prepared in our experiments with 7 lm
amount of water in the reaction system, which disturbs silica gels.
Chirality DOI 10.1002/chir

168
TANG ET AL.
Fig. 3. 1H NMR spectra of the derivatives of (a) cellulose 5b and (b) amylose 6a in pyridine-d5 at 808C. Py, pyridine.
Immobilization of Cellulose Derivatives and Their Chiral
Recognition Abilities
attained through the intermolecular polycondensation
among the trimethoxysilyl groups of 5 as well as the previ-
ous study,27,28 because the immobilization efficiency did
not decrease when 3-aminopropyl silica gel was used
instead of the bare silica gel. Because of the hydrolysis of
(CH3)3SiCl, HCl was produced in the reaction system and
Figure 5 shows a possible immobilization process. The
5-coated silica gel was dispersed into a mixture of EtOH/
H2O/(CH3)3SiCl and then 5 was immobilized onto the
silica gel by heating at 1108C for 10 min.26–28 After the
immobilization, the obtained CPMs were extensively
washed with THF to exclude the nonimmobilized deriva-
tives and dried. The immobilization efficiency was esti-
mated by TG analysis. In Table 2 are given the immobiliza-
tion efficiencies of the cellulose derivatives. The immobili-
zation efficiency increased with an increase in the R3
content of 5: 78% for 5a with 0.7% R3, 87% for 5b with 1.2%
R3, and 93% for 5c with 2.1% R3. It can be seen that the
immobilization method is very efficient; even if the R3 con-
tent was reduced to 0.7% (5a), 78% of the cellulose deriva-
tive could be immobilized. The immobilization was mainly
Chirality DOI 10.1002/chir
Fig. 4. Molecular structures of racemates.

IMMOBILIZATION AND CHIRAL RECOGNITION OF DERIVATIVES
169
TABLE 1. Separation results on the 5-coated CPMs and Chiralcel OD
5a (99.3/0.7)
5b (98.8/1.2)
5c (97.9/2.1)
Chiralcel ODb,c
Derivatives (R1/R3)a
Racemates
0
0
0
0
k1
k1
a
k1
a
k1
a
a
7
8
9
10
11
12
13
14
15
16
0.66 (1)
0.44 (–)
1.46 (1)
0.91 (–)
0.74 (–)
1.08 (–)
0.66 (–)
0.84 (–)
0.66 (1)
1.02 (–)
1.37
1.76
1.41
1.20
1.21
2.73
0.79 (1)
0.50 (–)
1.71 (1)
1.00 (–)
0.84 (–)
1.26 (–)
0.65 (–)
1.00 (–)
0.83 (1)
1.15 (–)
1.35
1.67
1.37
1.23
1.21
2.54
0.66 (1)
0.48 (–)
1.52 (1)
0.96 (–)
0.67 (–)
1.09 (–)
0.64 (–)
0.85 (–)
0.72 (1)
1.07 (–)
1.40
1.44
1.35
1.14
1.22
2.43
1.05 (1)
0.79 (–)
2.44 (1)
1.50 (–)
1.22 (–)
2.29 (–)
0.42 (–)
1.46 (–)
0.91 (1)
3.74 (–)
1.26
2.11
1.57
1.27
1.14
2.87
1.13
1.40
2.63
1.47
ꢀ1
ꢀ1
ꢀ1
1.30
2.00
3.34
1.23
1.70
3.26
1.22
1.41
2.78
Eluent, hexane/2-propanol (90/10, v/v); column size, 25 3 0.20 cm i.d.; flow rate, 0.1 ml/min; temperature, 258C. The signs in parentheses represent
the CD detection (254 nm) of the first-eluted enantiomer.
aThe ratio of R1/R3 was determined from the ratio of the (Ar-CH3)/(SiOCH3) in the 1H NMR spectrum.
bData from ref. 17.
cColumn size, 25 3 0.46 cm i.d.; flow-rate, 0.5 ml/min.
the acidic condition was favorable for the immobilization.27 larger values for racemates 7, 13, and 16. These results
In addition, the silanol groups on the silica gel surface are indicate that this method is valuable for efficient immobili-
expected to be simultaneously end-capped with trimethyl- zation of the cellulose derivatives onto silica gel without
silyl groups during the immobilization process.26–28
losing their chiral recognition abilities. Therefore, the
Figure 6 shows the chromatogram for the resolution of effect of R3 groups on the chiral recognition abilities of the
racemic Tro¨ger’s base 7 on the 5b-immobilized CPM with immobilized-type CPMs was negligible, when its content
hexane/2-propanol (90/10, v/v) as eluent. The dead time was small.
(t0) was estimated to be 9.35 min. The enantiomers were
The data for the commercially available Chiralpak IB
eluted 0at retention times t1 and t2. The retention factors consisting of the immobilized cellulose 3,5-dimethylphe-
0
(k1 , k2 ), which are estimated as (t1 2 t0)/t0 and (t2 2 t0)/ nylcarbamate as the chiral selector are included in Table 2
t0, were determined to be 0.73 and0 1.09, respectively, for comparison. Basically, for racemates 7, 11, 13, and
0
which led the separation factor a (5 k2 /k1 ) to be 1.49.
16, better resolution was attained on the 5-immobilized
In Table 2 are summarized the resolution results of 10 CPMs than on Chiralpak IB. For the rest of the racemates,
racemates on the 5-immobilized CPMs using hexane/2- the 5a-immobilized CPM showed a values comparable
propanol (90/10, v/v) as eluent. By comparing the resolu- with those of Chiralpak IB, whereas the 5b- and 5c-immo-
tion results on the 5-coated and 5-immobilized CPMs bilized CPMs showed slightly smaller values. In addition,
shown in Tables 1 and 2, we can easily recognize the the a values on CPM 3 immobilized using 3-(triethoxysi-
effect of the immobilization process on the chiral recogni- lyl)propyl residue are listed in Table 2 for comparison with
tion abilities of the cellulose derivative 5. Generally, the a the 5b-immobilized CPM. Both CPMs were prepared by
values on 5a, 5b, and 5c were similar before and after intermolecular polycondensation of different amounts of
immobilization, except for the slightly smaller a values for trialkoxysilyl residues, and the immobilization efficiencies
racemates 8, 12, and 15 after immobilization, and the of the cellulose derivatives were close, being 89% and 87%
Fig. 5. Scheme of immobilization of polysaccharide derivatives onto silica gel. [Color figure can be viewed in the online issue, which is available at
www.interscience.wiley.com.]
Chirality DOI 10.1002/chir

170
TANG ET AL.
TABLE 2. Immobilization efficiencies and separation results on the 5-immobilized CPMs, Chiralpak IB, and 3-immobilized
CPM
5a (R1/R3 5
99.3/0.7)
a78%b
5b (R1/R3 5
98.8/1.2)
a87%b
5c (R1/R3 5
97.9/2.1)
a93%b
Chiralpak
IBc,d
3e (R1/R2 5 98/2)
89%b
Cellulose
derivative
Racemate
0
0
0
0
k1
k1
a
k1
a
k1
a
a
a
7
8
9
10
11
12
13
14
15
16
0.59 (1)
1.39
1.64
1.36
1.17
1.21
2.56
0.73 (1)
0.42 (–)
1.51 (1)
0.90 (–)
0.75 (–)
1.13 (–)
0.57 (–)
0.90 (–)
0.86 (1)
1.16 (–)
1.49
1.34
1.32
1.15
1.26
2.19
1.15
1.20
1.32
3.78
0.79 (1)
0.52 (–)
1.76 (1)
1.02 (–)
0.81 (–)
1.24 (–)
0.70 (–)
1.03 (–)
0.98 (1)
1.28 (–)
1.48
1.28
1.29
1.16
1.26
2.17
1.14
1.17
1.28
3.71
0.96 (1)
0.55 (–)
2.00 (1)
1.12 (–)
1.00 (–)
1.54 (–)
3.15 (–)
1.13 (–)
0.86 (1)
1.48 (–)
1.22
1.77
1.33
1.22
1.14
2.42
1.53 (1)
1.51 (–)
1.35 (1)
1.15 (–)
1.23 (–)
2.35 (–)
0.35 (–)
1.24 (1)
0.72 (–)
0.63 (–)
0.88 (–)
0.71 (–)
0.70 (–)
0.57 (1)
0.74 (–)
ꢀ1
ꢀ1
ꢀ1
(–)
1.24
1.92
4.16
1.26
1.89
2.72
1.21 (–)
1.65 (1)
3.52 (–)
Eluent, hexane/2-propanol (90/10, v/v); column size, 25 3 0.20 cm i.d.; flow rate, 0.1 ml/min; temperature, 258C. The signs in parentheses represent
the CD detection (254 nm) of the first-eluted enantiomer.
aThe ratio of R1/R3 was determined from the ratio of the (Ar-CH3)/(SiOCH3) in the 1H NMR spectrum.
bImmobilization efficiency estimated from TG analysis.
cData from Ref. 17.
dColumn size, 25 3 0.46 cm i.d.; flow-rate, 0.5 ml/min.
eData from Ref. 27.
for the 3- and 5b-immobilized CPMs, respectively. The a mates were improved to some extent depending on the
values on the 3- and 5b-immobilized CPMs were similar racemates except for racemates 7, 15, and 16. A typical
except for the slightly smaller values for racemates 8, 12, example is the separation of racemate 12 with eluents
0
and 15 on the 5b-immobilized CPM, and slightly larger containing chloroform. Its k1 and a values were signifi-
values for racemates 13 and 16, indicating that the chiral cantly increased when the eluent was changed from hex-
0
recognition abilities of these two CPMs are rather similar. ane/2-propanol (90/10, v/v; k1 5 0.88, a 5 2.56) to hex-
0
These results suggest that the method reported in this ane/chloroform/2-propanol (90/10/1, by volume; k1
study is valuable for immobilizing cellulose derivatives 7.27, a 5 3.68). The increased k1 value may be attributed
5
0
onto silica gel without significantly reducing their chiral to the increased H-bonding interactions between the enan-
recognition abilities.
tiomers of 12 and the 5a-immobilized CPM resulted from
The effect of eluents on the chiral recognition ability of the decreased portion of the H-bonding solvent, 2-propa-
the 5a-immobilized CPM was also investigated. The
results are shown in Table 3. By utilizing the eluents con-
taining chloroform and THF, which cannot be used for
those coated-type CPMs, the a values for most of the race-
TABLE 3. Separation results on the 5a-immobilized CPM
with the eluents containing chloroform and THF
5a-Immobilized CPM
H/I
H/C/I
H/T/I
(90/10)
(90/10/1)
(90/10/1)
0
0
0
Racemates
k1
a
k1
a
k1
a
7
8
9
10
11
12
13
14
15
16
0.59 (1) 1.39 0.75 (1)
0.35 (–) 1.64 0.43 (–)
1.24 (1) 1.36 1.98 (1)
1.31 0.75 (1)
1.31
2.11 0.43 (1) ꢀ1
1.41 1.62 (1)
1.22 0.74 (–)
1.24 0.65 (–)
3.68 0.73 (–)
1.34
0.72 (–)
0.63 (–)
0.88 (–)
1.17 1.30 (–)
1.21 0.73 (–)
2.56 7.27 (–)
1.29
1.19
2.71
0.71 (–) ꢀ1
0.66 (–) ꢀ1
1.76 (–) ꢀ1
0.70 (–)
1.24
0.91(–)
1.29 0.98 (–)
7.52 (1)
1.27 1.02 (–)
1.27
1.73
1.12
a
0.57 (1) 1.92
–
–
0.74 (–)
4.16 0.95 (–)
Column size, 25 3 0.20 cm i.d.; temperature, 258C; flow rate, 0.1 ml/min.
The signs in parentheses represent the CD detection (254 nm) of the
first-eluted enantiomer. H, hexane; I, 2-propanol; C, chloroform; T,
tetrahydrofuran.
Fig. 6. Chromatographic resolution of 7 on the 5b-immobilized CPM.
Eluent, hexane/2-propanol (90/10, v/v); column size, 25 3 0.20 cm i.d.;
flow-rate, 0.1 ml/min; temperature, 258C. The signs in the parentheses
represent the CD detection (254 nm) of the first-eluted enantiomer.
aEnantiomers were not eluted in 4 h.
Chirality DOI 10.1002/chir

IMMOBILIZATION AND CHIRAL RECOGNITION OF DERIVATIVES
171
TABLE 4. Separation results on the 6-immobilized CPMs, Chiralpak IA, and 4-immobilized CPM
6a (R1/R3 5
99.1/0.9)
a58%b
6b (R1/R3 5 98.7/
6c (R1/R3 5
97.5/2.5)
a 95%b
1.3)
Chiralpak
IAc
4d (R1/R2 5 99/1)
a77%b
86%b
Amylose
derivative
Racemate
0
0
0
0
k1
a
k1
a
k1
a
k1
a
a
7
8
9
10
11
12
13
14
15
16
0.31 (1)
0.14 (1)
1.16 (–)
0.86 (–)
0.30 (–)
0.50
1.30
2.76
1.08
2.09
0.42 (1)
0.22 (1)
1.55 (–)
1.09 (–)
0.34 (–)
0.64
1.32
2.52
1.07
2.04
0.85 (1)
0.40 (1)
2.93 (1)
1.59 (–)
0.73 (–)
1.04
0.95 (1)
0.96 (1)
3.52 (1)
1.87 (–)
1.24
1.98
ꢀ1
1.80
ꢀ1
1.0
0.76 (1)
0.59 (1)
3.82 (–)
2.99 (–)
0.81 (–)
1.70 (1)
0.96 (1)
1.18 (1)
3.14 (1)
2.92 (–)
1.50
2.72
1.26
2.40
1.04
1.19
1.44 (1)
2.83 (1)
1.18 (–)
2.11 (–)
ꢀ1
ꢀ1
ꢀ1
1.0
1.0
(–)
1.0
1.0
1.0
1.0
0.34
0.43
ꢀ1
ꢀ1
0.38 (1)
1.37 (1)
0.90 (–)
ꢀ1
0.49 (1)
1.78 (1)
1.21 (–)
ꢀ1
1.16
1.18
1.84
1.09
1.65
2.21
1.08 (1)
3.49 (1)
2.27 (–)
2.79
2.14
1.88
2.10
Eluent, hexane/2-propanol (90/10, v/v); column size, 25 3 0.20 cm i.d.; flow rate, 0.1 ml/min; temperature, 258C. The signs in parentheses represent
the CD detection (254 nm) of the first-eluted enantiomer.
aThe ratio of R1/R3 was determined from the ratio of the (Ar-CH3)/(SiOCH3) in the 1H NMR spectrum.
bImmobilization efficiency estimated from TG analysis.
cColumn size, 25 3 0.46 cm i.d.; flow-rate, 0.5 ml/min; temperature, 258C.
dData from Ref. 27.
nol. Moreover, the reversal elution orders of enantiomers method shown in Figure 2 and immobilized onto silica gel
for racemate 8 were observed by CD detection with a mix- by a similar method applied to the cellulose derivatives.
ture of hexane/THF/2-propanol (90/10/1, by volume) as The results of the immobilization and enantioseparation
the eluent and confirmed by HPLC analysis of the pure on the obtained CPMs are shown in Table 4. Similar to the
enantiomers of 8 isolated using Chiralpak AD-H (25 3 cellulose derivatives, as the content of 4-(trimethoxysilyl)-
0.46 cm i.d.) with hexane/2-propanol (90/10, v/v) as phenyl groups in the amylose derivatives increased, the
reported previously.28 The conformational change in the immobilization efficiency increased and the chiral recogni-
cellulose derivative in these eluents may partially contrib- tion ability decreased. The reversal of elution orders of
ute to the above results.
enantiomers for racemate 9 were observed by CD detec-
tion on the 6c-immobilized CPM containing a higher con-
tent of R3. In this case, same eluent was used and thus
this is a true reversal of elution orders. The results on a
commercial column, Chiralpak IA, composed of immobi-
lized amylose 3,5-dimethylphenylcarbamate as the chiral
selector are included in Table 4 for comparison. The a val-
ues on the 6a-immobilized CPM were comparable with
those on Chiralpak IA, whereas those on the 6b- and 6c-
immobilized CPMs were somewhat lower. In Table 4 are
also included the separation data for the 4-immobilized
CPM. Compared with the 4-immobilized CPM (immobili-
zation efficiency, 86%), the 6b-immobilized CPM (immobi-
lization efficiency, 77%) showed slightly smaller a values.
The solvent compatibility of the 6a-immobilized CPM
was also investigated and the results are shown in Table 5.
With the eluents containing chloroform and THF, race-
mates 9, 11, 12, and 14 were better resolved. Addition-
ally, with hexane/THF/2-propanol (90/10/1, by volume)
as the eluent, the elution orders of the enantiomers for 9
were reversed by CD detection and confirmed by the
HPLC analysis of the pure enantiomers of 9.
Immobilization of Amylose Derivatives and Their Chiral
Recognition Abilities
The amylose derivatives 6a–c bearing 0.9, 1.3, and 2.5%
of R3, respectively, were prepared using the one-pot
TABLE 5. Separation results on the 6a-immobilized CPM
with the eluents containing chloroform and THF
6a-Immobilized CPM
H/I
(90/10)
H/C/I
(90/10/1)
H/T/I
(90/10/1)
0
0
0
Racemate
k1
a
k1
a
k1
a
7
8
9
10
11
12
13
14
15
16
0.31 (1)
0.14 (1)
1.16 (–)
0.86 (–)
1.30 0.93 (1)
2.76 0.20 (1)
1.08 2.10 (–) ꢀ1
2.09 1.62 (–)
0.59 (1)
1.0 3.04 (–)
1.22 0.53 (1)
2.64 0.16 (1)
1.41 (1)
1.60 0.85 (–)
1.41 0.25
1.23
2.00
1.20
1.75
1.0
0.30 (–) ꢀ1
0.50
0.34
1.45 0.44
1.0
1.0 0.33 (1) ꢀ1
0.73 (1) ꢀ1
0.38 (1) ꢀ1
0.54 (1)
1.39 0.44 (1)
–
1.16
–
1.85
a
a
1.37 (1)
2.79
2.14 3.60 (–)
–
–
0.90 (–)
2.10 1.18 (–)
CONCLUSION
Column size, 25 3 0.20 cm i.d.; temperature, 258C; flow rate, 0.1 ml/min.
The signs in parentheses represent the CD detection (254 nm) of the
first-eluted enantiomer. H, hexane; I, 2-propanol; C, chloroform; T,
tetrahydrofuran.
The 3,5-dimethylphenylcarbamates of cellulose and am-
ylose containing a small amount of 4-(trimethoxysilyl)-
phenyl groups were synthesized by a one-pot method and
immobilized onto silica gel via intermolecular polyconden-
Chirality DOI 10.1002/chir
aEnantiomers were not eluted in 4 h.

172
TANG ET AL.
14. Cirilli R, Simonelli A, Ferretti R, Bolasco A, Chimenti P, Secci D, Mac-
cioni E, La Torre F. Analytical and semipreparative high performance
liquid chromatography enantioseparation of new substituted 1-thiocar-
bamoyl-3,5-diaryl-4,5-dihydro-(1H)-pyrazoles on polysaccharide-based
chiral stationary phases in normal-phase, polar organic and reversed-
phase conditions. J Chromatogr A 2006;1101:198–203.
sation of trimethoxysilyl residues under an acidic condi-
tion. When a suitable amount of 4-(trimethoxysilyl)phenyl
groups was introduced, the obtained CPMs exhibited high
chiral recognition ability, which are comparable with those
of the commercially available immobilized-type CPMs.
The solvents unusable with the conventional coated-type
CPMs, such as chloroform and THF, could be used for the
immobilized-type CPMs, and the chiral recognition abil-
ities of these CPMs were improved for most of the race-
mates. As in the previous studies,26–28 this immobilization
method seems to be valuable.
15. Okamoto Y, Aburatani R, Miura S, Hatada K. Chiral stationary phases
for HPLC: cellulose tris(3,5-dimethylphenylcarbamate) and tris(3,5-
dichlorophenylcarbamate) chemically bonded to silica gel. J Liquid
Chromatogr 1987;10:1613–1628.
16. Yashima E, Fukaya H, Okamoto Y. 3,5-Dimethylphenylcarbamates of
cellulose and amylase regioselectively bonded to silica gel as chiral
stationary phases for high-performance liquid chromatography.
J Chromatogr A 1994;667:11–19.
17. Chen XM, Yamamoto C, Okamoto Y. Influence of vinyl monomers
and temperature on immobilization of cellulose 3,5-dimethylphenylcar-
bamate onto silica gel as chiral stationary phases for high-performance
liquid chromatography. J Chromatogr A 2006;1104:62–68.
ACKNOWLEDGMENT
The authors thank Dr. X.M. Chen for his experimental
supports.
18. Franco P, Senso A, Oliveros L, Minguillo´n C. Covalently bonded poly-
saccharide derivatives as chiral stationary phases in high-performance
liquid chromatography. J Chromatogr A 2001;906:155–170.
LITERATURE CITED
19. Kubota T, Yamamoto C, Okamoto Y. Preparation of chiral stationary
phase for HPLC based on immobilization of cellulose 3,5-dimethylphe-
nylcarbamate derivatives on silica gel. Chirality 2003;15:77–82.
1. Okamoto Y, Kawashima M, Hatada K. Chromatographic resolution.
XI. Controlled chiral recognition of cellulose triphenylcarbamate deriv-
atives supported on silica gel. J Chromatogr 1986;363:173–186.
20. Chen XM, Qin F, Liu YQ, Huang XD, Zou HF. Synthesis of chiral
stationary phases with radical polymerization reaction of cellulose
phenylcarbamate derivatives and vinylized silica gel. J Chromatogr A
2004;1034:109–116.
2. Okamoto Y, Kaida Y. Resolution by high-performance liquid chroma-
tography using polysaccharide carbamates and benzoates as chiral
stationary phases. J Chromatogr A 1994;666:403–419.
3. Yashima E, Okamoto Y. Chiral discrimination on polysaccharides
derivatives. Bull Chem Soc Jpn 1995;68:3289–3307.
21. Francotte E, Huynh D. Immobilized halogenophenylcarbamate deriva-
tives of cellulose as novel stationary phases for enantioselective drug
analysis. J Pharm Biomed Anal 2002;27:421–429.
4. Yashima E, Okamoto Y. Polysaccharide derivatives for chromato-
graphic separation of enantiomers. Angew Chem Int Ed Engl 1998;37:
1020–1043.
22. Enomoto N, Furukawa S, Ogasawara Y, Akano H, Kawamura Y,
Yashima E, Okamoto Y. Preparation of silica gel-bonded amylose
through enzyme-catalyzed polymerization and chiral recognition abil-
ity of its phenylcarbamate derivatives in HPLC. Anal Chem 1996;68:
2798–2804.
5. Tachibana K, Ohnishi A. Reversed-phase liquid chromatographic sep-
aration of enantiomers on polysaccharide type chiral stationary
phases. J Chromatogr A 2001;906:127–154.
6. Yamamoto C, Okamoto Y. Optically active polymers for chiral separa-
tion. Bull Chem Soc Jpn 2004;77:227–257.
23. Zhang T, Nguyen D, Franco P. Enantiomer resolution screening strat-
egy using multiple immobilised polysaccharide-based chiral stationary
phases. J Chromatogr A 2008;1191:214–222.
7. Yashima E, Yamamoto C, Okamoto Y. Polysaccharide-based chiral LC
columns. Synlett 1998;344–360.
24. Zhang T, Franco P. Analytical and preparative potential of immobilized
polysaccharide-derived chiral stationary phases. In: Subramanian G,
editor. Chiral separation techniques: a practical approach (3rd edi-
tion). Weinheim: Wiley-VCH; 2007. p 99–134.
8. Francotte E. Enantioselective chromatography as a powerful alterna-
tive for the preparation of drug enantiomers. J Chromatogr A 2001;
906:379–397.
9. Zhang T, Kientzy C, Franco P, Ohnishi A, Kagamihara Y, Kurosawa
H. Solvent versatility of immobilized 3,5-dimethylphenylcarbamate of
25. Zhang T, Nguyen D, Franco P, Isobe Y, Michishita T, Murakami T.
Cellulose tris(3,5-dichlorophenylcarbamate) immobilised on silica: a
novel chiral stationary phase for resolution of enantiomers. J Pharm
Biomed Anal 2008;46:882–891.
amylose in enantiomeric separations by HPLC.
2005;1075:65–75.
J Chromatogr A
10. Zhang T, Nguyen D, Franco P, Murakami T, Ohnishi A, Kurosawa H.
Cellulose 3,5-dimethylphenylcarbamate immobilized on silica: a new
chiral stationary phase for the analysis of enantiomers. Anal Chim
Acta 2006;557:221–228.
26. Ikai T, Yamamoto C, Kamigaito M, Okamoto Y. Efficient immobiliza-
tion of cellulose phenylcarbamate bearing alkoxysilyl group onto silica
gel by intermolecular polycondensation and its chiral recognition.
Chem Lett 2006;35:1250–1251.
11. Ghanem A. True and false reversal of the elution order of barbiturates
on bonded cellulose-based chiral stationary phase. J Chromatogr A
2006;1132:329–332.
27. Ikai T, Yamamoto C, Kamigaito M, Okamoto Y. Immobilization of
polysaccharide derivatives onto silica gel: facile synthesis of chiral
packing materials by means of intermolecular polycondensation of
triethoxysilyl groups. J Chromatogr A 2007;1159:151–158.
12. Francotte E. Contribution of preparative chromatographic resolution
to the investigation of chiral phenomena. J Chromatogr A 1994;666:
565–601.
28. Tang SW, Okamoto Y. Immobilization of cellulose phenylcarbamate
onto silica gel via intermolecular polycondensation of triethoxysilyl
groups introduced with (3-glycidoxypropyl)triethoxysilane. J Sep Sci
2008;31:3133–3138.
13. Zhang T, Schaeffer M, Franco P. Optimization of the chiral separation
of a Ca-sensitizing drug on an immobilized polysaccharide-based chi-
ral stationary phases: case study with
J Chromatogr A 2005;1083:96–101.
a
preparative perspective.
29. Koller H, Rimbo¨ck KH, Mannschreck A. High-pressure liquid chroma-
tography on triacetylcellulose. J Chromatogr 1983;282:89–94.
Chirality DOI 10.1002/chir