The comparison in enantioseparation ability of the chiral stationary phases with single and mixed selector-The selectors derived from two D -tartrates

Article abstract of DOI:10.1002/chir.20904

(2S,3S)-2,3-Bis(3,5-dimethylphenylcarbonyloxy)-3-(benzyloxycarbonyl) -propanoic acid and (2S,3S)-2,3-bis(1-naphthalenecarbonyloxy)-3- (benzyloxycarbonyl)-propanoic acid were synthesized from D-tartaric acid. These two compounds were chlorinated to afford

Full text of DOI:10.1002/chir.20904

CHIRALITY 23:228–236 (2011)
The Comparison in Enantioseparation Ability of the Chiral Stationary
Phases with Single and Mixed Selector—The Selectors Derived from
Two D-Tartrates
JUN CHEN,1 MU-ZI LI,1 YAN-HUA XIAO,1 WEI CHEN,1 SHI-RONG LI,2 AND ZHENG-WU BAI
1
*
1Key Laboratory of Green Chemical Process of Ministry of Education, Hubei Key Laboratory of Novel Chemical Reactor and Green Chemical
Technology, Wuhan Institute of Technology, Wuhan, China
2Hubei Key Laboratory of Biologic Resources Protection and Utilization, Hubei Institute for Nationalities, Enshi, Hubei, China
ABSTRACT
(2S,3S)-2,3-Bis(3,5-dimethylphenylcarbonyloxy)-3-(benzyloxycarbonyl)-propanoic
acid and (2S,3S)-2,3-bis(1-naphthalenecarbonyloxy)-3-(benzyloxycarbonyl)-propanoic acid were
synthesized from D-tartaric acid. These two compounds were chlorinated to afford two chiral
selectors for chiral stationary phases (CSPs). The selectors were separately immobilized on ami-
nated silica gel to give two single selector CSPs; and were simultaneously immobilized to obtain
a mixed selector CSP. Comparing to the single selector CSPs, the mixed selector CSP bears the
enhanced enantioseparation ability, suggesting that the two selectors in the mixed selector CSP
are consistent for chiral recognition in most mobile phase conditions. Chirality 23:228–236,
C
VV
2011.
2010 Wiley-Liss, Inc.
KEY WORDS: single selector; mixed selector; chiral stationary phase; enantioseparation; high-
performance liquid chromatography
INTRODUCTION
(1R,2R)-1,2-diphenylethylenediamine.18 However, its enantio-
separation ability is lower than that of single selector CSPs.
The reason probably is these two selectors may reversely
contribute the chiral recognition and as a result, the enantio-
separation ability of the CSP is impaired. To investigate the
fact causing the impair in enantioseparation ability of the
mixed selector CSP, in this work, another mixed selector
CSP was synthesized, where the two selectors were both pre-
pared from D-tartaric acid and they are close in their struc-
tures. The enantioseparation ability of the relevant single se-
lector and mixed selector CSPs was evaluated.
Chiral high-performance liquid chromatography (HPLC)
has been proved to be a very effective technique to separate
racemates into their enantiomers thus being widely used for
the enantioseparation of drugs, especially for the develop-
ment of new drugs.1–4 Chiral stationary phases (CSPs) are
essential absorbent materials for chiral HPLC columns.5 In
this technique, enantioseparation results from the difference
of the interactions between a pair of enantiomers and CSPs.6
Many authors discussed the mechanism of chiral recognition
during enantioseparation.7–10 Generally, it is believed that
temporary diastereoisomers are formed when chiral selec-
tors interact with chiral analytes through H bonding, p-p
interaction, dipole-dipole interaction, or van der Waals
force.6,10 In addition, the stereo-hindrance in the interaction
between chiral selectors and chiral analytes is another ele-
ment for chiral discrimination.11 Under the guidance of these
principles, various types of CSPs were developed. To
improve the enantioseparation ability, some biselector CSPs
were reported.12–16 However, these biselector CSPs do not
always show enhanced enantioseparation ability. In the
reported works, two chiral compounds were connected with
a cross-linker of multiple reactive groups to form a biselector,
which was then immobilized on a support to afford a biselec-
tor CSP. Kraak and coworkers17 first prepared a mixed selec-
tor CSP by immobilizing two derivatives of phenylglycine on
aminated silica gel through acid–base reaction. The enantio-
separation ability of this CSP was tested only in limited mo-
bile phases with few chiral analytes, because the selector will
disassociate from the CSP when the acidity of mobile phases
changes dramatically. The CSP cannot be subjected to the
enantioseparation of the chiral analytes whose acidity is
stronger than that of carboxylic acid or whose basicity is
stronger than that of amine. Otherwise, the disassociation
also takes place. To establish a method to prepare mixed se-
lector CSPs by covalent immobilization, in our previous
work, a mixed selector CSP was synthesized. The two chiral
selectors were prepared from L-dibenzoyl tartaric acid and
MATERIALS AND METHODS
Materials and Chemicals
D-Tartaric acid was purchased from Chengdu Likai Chiral Tech.
(China). 3,5-Dimethylbenzoic acid and 1-naphthalenecarbonyl chloride
were, respectively, available from Shanghai Zhuorui and Changzhou
Wujin Chemical (China). 3-Aminopropyltriethoxysilane (APTES) was
obtained from Novel Organic Silicon Materials of Wuhan University
(China) and redistilled before use. Silica gel (Lichrosorb Si 100) was
obtained from Merck (Germany) with a particle size of 5 lm, a pore size
2
of 100 A, and a surface area of 300 m g21. Triethylamine (TEA) was
˚
dried over phosphorous pentoxide and redistilled. Pyridine was dried
with NaOH and CaH2 in sequence and redistilled. All other chemicals
used for the CSPs synthesis were of analytical grade and used as
received.
Instruments and Measurements
IR spectra were recorded on a Nicolet FTIR instrument (USA) with
KBr pellets. Elemental analysis (EA) measurement was conducted on an
Contract grant sponsor: National Natural Science Foundation of China
(NSFC); Contract grant number: 20675061
Contract grant sponsor: Hubei Provincial Department of Education of China;
Contract grant number: Z 20081501
*Correspondence to: Zheng-Wu Bai, Key Laboratory of Green Chemical Pro-
cess of Ministry of Education, Wuhan Institute of Technology, Wuhan
430073, China. E-mail: zhengwu_bai@yahoo.com
Received for publication 1 February 2010; Accepted 22 June 2010
DOI: 10.1002/chir.20904
Published online 29 September 2010 in Wiley Online Library
(wileyonlinelibrary.com).
C
VV
2010 Wiley-Liss, Inc.

CSPs OF SINGLE AND MIXED SELECTOR
229
Fig. 1. The synthetic scheme of CSPs 1–3.
Elementar VarioEL III CHNOS apparatus (Germany). 1H NMR spectra
were performed on a Varian INOVA 500 spectrometer (USA). The stain-
less steel HPLC empty columns (250 mm 3 4.6 mm) were purchased
from Hypersil (US). The CSPs were packed into the empty columns with
an Alltech Model 1666 slurry packer (USA). The enantioseparation was
run on a Waters chromatograph (USA) equipped with a Waters 996 pho-
todiode array detector, a Waters 600E Quat Pump, a Waters Millenium
32 system controller, a Waters 717 plus autosampler.
258C) d: 2.3 (12H, CH3), 5.8 (2H, CꢀꢀH), 7.3–7.6 (m, 6H, ArꢀꢀH); 13C
NMR (DMSO, 258C) d: 72 (CH), 125–135 (m, ArꢀꢀC), 160 (ꢀꢀCO2ꢀꢀ),
166, 168 (ꢀꢀCO2H).
Compound 2 was prepared in a similar manner to prepare Compound
1, with 1-naphthalenecarbonyl chloride (10.59 g, 55.6 mmol) and tartaric
acid (2.79 g, 18.6 mmol). The reaction mixture was stirred overnight at
1008C. The crude product was collected by heat-filtration and washed
thoroughly with chloroform. Compound 2 (6.48 g) was obtained as a
white powder after drying in vacuo. Yield: 76%; m.p: 179–1818C; [a]D20
:
142.88 (C 1.0, methanol); IR (KBr, cm21): 3651–3141 (ꢀꢀCO2H), 3091,
3046 (aromatic CꢀꢀH), 2917, 2842 (CꢀꢀH), 1764 (ꢀꢀCO2ꢀꢀ), 1727
(ꢀꢀCO2H), 1453, 1378 (CꢀꢀH); EA (C26H18O8ꢁ5H2O, %): Calcd C 56.93, H
5.15; Found C 57.11, H 5.23; 1H NMR (DMSO, 258C) d: 6.2 (2H, CH),
7.6–8.9 (m, 14H, ArꢀꢀH); 13C NMR (DMSO, 258C) d: 72 (CꢀꢀH), 125–
135 (m, ArꢀꢀC), 160 (ꢀꢀCO2ꢀꢀ), 166ꢀꢀ168 (ꢀꢀCO2H).
Preparation of Chiral Stationary Phases
SOCl2 (18 ml, 0.25 mol) was added dropwise to 3,5-dimethylbenzoic
acid (90.8 g, 60.5 mmol) with stir. The resulting solution was then stirred
at 608C for 3 h to ensure the completion of the reaction. The excess
SOCl2 was removed by distillation in vacuo to afford yellow oil, in which
D-tartaric acid (3.03 g, 20.2 mmol) was added. The mixture was stirred at
1108C for 6 h. After cooling to ambient temperature, the mixture solidi-
fied. The solid was suspended in 100 ml water, and the suspension was
stirred at 1008C for 1 h. The solid was filtered and washed with benzene
thoroughly. Compound 1 (6.36 g) was white powder after drying in
vacuo. Yield: 76%; m.p: 206–2088C;[a]2D0: 144.88 (C 1.0, methanol); IR
(KBr, cm21): 3672–3079 (ꢀꢀCO2H, aromatic CꢀꢀH), 2944, 2913 (CꢀꢀH),
1764 (ꢀꢀCO2ꢀꢀ), 1714 (ꢀꢀCO2H), 1453, 1378 (CꢀꢀH); EA (C22H22O8ꢁH2O,
%): Calcd C 61.11, H 5.59; Found C 61.31, H 5.57; 1H NMR (DMSO,
Compound 1 (14.49 g, 35.0 mmol) and Na2CO3 (3.14 g, 29.6 mmol)
were mixed in water (100 ml). To this solution, toluene (80 ml), benzyl
chloride (3.69 g, 29.8 mmol), and catalytic amount of PEG-400 were
added. After stirring at 858C for 18 h, the reaction mixture was cooled
and separated into two phases. The toluene layer was neutralized with
diluted hydrochloric acid under vigorous stir and was then washed with
water till aqueous phase became neutral in pH. The organic phase was
dried over anhydrous MgSO4. Yellow oil was given after toluene was
removed in vacuo. (2S,3S)-2,3-Bis(3,5-dimethylphenylcarbonyloxy)-3-
Chirality DOI 10.1002/chir

230
CHEN ET AL.
Fig. 2. The structures of chiral analytes resolved on CSPs 1–3 (1–38).
(benzyloxycarbonyl)-propanoic acid (Compound 3) (5.71 g) was recrys-
tallized from cyclohexane as a white solid. Yield: 38%; m.p: 128–1308C;
[a]2D0: 118.48 (C 1.0, methanol), IR (KBr, cm21) 3676–3108 (ꢀꢀCO2H, ar-
omatic CꢀꢀH), 2958, 2912 (CꢀꢀH), 1752, 1742 (ꢀꢀCO2ꢀꢀ), 1727
(ꢀꢀCO2H), 1457, 1374 (CꢀꢀH), EA (C29H28O8, %): Calcd C 69.04, H 5.59;
Found C 69.46, H 5.42; 1H NMR (DMSO, 258C) d: 2.4 (12H, CH3), 5.2
(m, 2H, CH2), 5.9 (1H, CH), 6.0 (1H, CH), 7.1–7.6 (m, 11H, ArꢀꢀH); 13C
NMR (DMSO, 258C) d: 21 (CH3), 52 (CH2), 68 (CH), 72 (CH), 127–139
(ArꢀꢀC), 160 (ꢀꢀCO2ꢀꢀ), 165–167 (ꢀꢀCO2H).
(2S,3S)-2,3-Bis(1-naphthalenecarbonyloxy)-3-(benzyloxycarbonyl)-pro-
panoic acid (Compound 4) was prepared in the same manner to prepare
Compound 3, with Compound 2 (10.19 g, 22.2 mmol), Na2CO3 (2.37 g,
22.4 mmol) and benzyl chloride (2.80 g, 22.2 mmol). It was purified with
column chromatography packed with silica gel and eluted with a mixture
of cyclohexane/ethyl acetate/acetic acid (2/1/0.02, v/v/v). Yield: 43%;
m.p: 46–488C; [a]2D0: 12.88 (C 1.0, methanol); IR (KBr, cm21): 3680–
3150 (ꢀꢀCO2H), 3089, 3057 (aromatic CꢀꢀH), 2955, 2848 (CꢀꢀH), 1763
(ꢀꢀCO2ꢀꢀ), 1726 (ꢀꢀCO2H), 1464, 1383 (CꢀꢀH), EA (C33H24O8, %): Calcd
Chirality DOI 10.1002/chir

CSPs OF SINGLE AND MIXED SELECTOR
231
Fig. 3. The structures of chiral analytes resolved on CSPs 1–3 (39–60).
C 72.26, H 4.41; Found C 72.38, H 4.86; 1H NMR (CHCl3, 258C) d: 4.9–
Column Packing and Enantioseparation
5.0 (m, 2H, CH2), 5.9 (2H, CH), 6.7–8.8 (m, 19H, ArꢀꢀH); 13C NMR
(CHCl3, 258C) d: 68–72 (m, CH2), 76–77 (CH), 124–134 (m, ArꢀꢀC), 165–
166 (m, ꢀꢀCO2ꢀꢀ), 171 (ꢀꢀCO2ꢀꢀ), 172 (ꢀꢀCO2H).
The columns were packed using a pressurized slurry technique. CSPs
1–3 (ꢂ3.3 g) were added in chloroform (30 ml), subsequently followed
by sonication for 8 min to form slurries. These slurries were packed into
empty HPLC columns with methanol as displacer solvent, under the
pressure no more than 50 Mpa. The packed columns were flushed with
isopropanol and isopropanol/n-hexane in turn. The column efficiency
was measured by biphenyl, with isopropanol/n-hexane (10/90, v/v) as
the mobile phase. The sample solutions (1 mg ml21) were prepared by
dissolving the chiral compounds in acetonitrile and were filtered through
0.2-lm membrane. The injection volume was 15 ll. All mobile phases
were filtered and degassed before use. The flow rates were set at 1 ml
min21. Column temperature was set at 258C.
The retention factors (k1 and k2) were calculated according to the for-
mulas of (t1 2 t0)/t0 and (t2 2 t0)/t0, where t1 and t2 are, respectively,
the retention time of the first and the second-eluted enantiomers, and t0
is determined by measuring the retention time of NaNO2 solution. The
separation factor (a) was calculated from the formula of k2/k1. The reso-
lution (Rs) was calculated from the formula of 2(t2 2 t1)/(w1 1 w2),
where w1 and w2 are the bandwidth of the first and the second-eluted
enantiomers, respectively.
3-Aminopropyl silica gel was prepared with dried silica gel and
APTES.19 EA (%): C 5.58, H 1.50.
Compound 3 (2.56 g, 5.1 mmol) was placed in a 50 ml three-necked
round bottom flask, in which SOCl2 (10 ml) was added dropwise with
stir. The resulting solution was heated to 758C and stirred for 8 h. The
excess SOCl2 was thoroughly removed in vacuo to give Selector 1 as vis-
cous oil, which was mixed with 3-aminopropyl silica gel (3.65 g), pyri-
dine (15 ml), and TEA (3 ml). The suspension was gently stirred over-
night at 808C. The solid was collected by filtration and washed with ace-
tone in a Soxhlet apparatus. CSP 1 (4.03 g) was a pale yellow solid after
the removal of the solvent. IR (KBr, cm21): 3461, 1544 (ꢀꢀNHꢀꢀ), 2980,
2938 (ꢀꢀCHꢀꢀ), 1603, 1646 (ꢀꢀNHꢀꢀCOꢀꢀ), 1111 (SiꢀꢀO); EA (%): C
12.60, H 2.05; Suspension-state 1H NMR (D2O, 258C) d: 0.8 (SiꢀꢀCH2),
1.30 (SiCH2CH2CH2), 2.4 (CH3, NH2), 3.3 (SiCH2CH2CH2), 5.4 (CH,
CH2), 5.8 (CH), 7.2–8.9 (m, ArꢀꢀH).
CSP 2 (3.88 g) as a brown solid was prepared in the same manner to
prepare CSP 1 using Compound 4 (3.87 g, 7.1 mmol), SOCl2 (10 ml),
and 3-aminopropyl silica gel (3.50 g). IR (KBr, cm21): 3461, 1545
(ꢀꢀNHꢀꢀ), 1646 (ꢀꢀNHꢀꢀCOꢀꢀ), 1111 (SiꢀꢀO); EA (%): C 12.63, H 1.94;
Suspension-state 1H NMR (D2O, 258C) d: 0.8 (SiꢀꢀCH2), 1.3
(SiCH2CH2CH2), 2.3 (NH2), 2.8–3.3 (m, SiCH2CH2CH2), 5.3–5.6 (m, CH,
CH2), 5.9 (CH), 7.3–8.9 (m, ArꢀꢀH).
RESULTS AND DISCUSSION
Preparation and Characterization of CSPs 1–3
Figure 1 shows the synthetic scheme of CSPs 1–3.
Because the molecular size of Selector 2 is larger than that
of Selector 1, in the preparation of CSP 3, the fed amount of
Compound 4 is a little more than that of Compound 3 in
molar number to ensure both selectors are sufficiently immo-
bilized on aminated silica gel. There are many residual
amino groups on the surface of aminated silica gel. In the
FTIR spectra of CSPs 1–3, the absorbance approximately at
CSP 3 (3.86 g) as a brown solid was prepared in the same manner to
prepare CSP 1 using a mixture of Compound 3 (1.83 g, 3.6 mmol) and
Compound 4 (2.40 g, 4.0 mmol), SOCl2 (10 ml), and 3-aminopropyl silica
gel (3.44 g). IR (KBr, cm21): 3457, 1548 (ꢀꢀNHꢀꢀ), 2944 (ꢀꢀCHꢀꢀ), 1731
(ꢀꢀCO2ꢀꢀ), 1641 (ꢀꢀNHꢀꢀCOꢀꢀ), 1111 (SiꢀꢀO); EA (%): C 14.03, H 2.05;
Suspension-state 1H NMR (D2O, 258C) d: 0.8 (SiꢀꢀCH2), 1.3
(SiCH2CH2CH2), 2.1 (NH2), 2.9–3.3 (m, SiCH2CH2CH2), 5.0–5.6 (m, CH,
CH2), 6.8–7.8, 8.6–9.1 (m, ArꢀꢀH).
Chirality DOI 10.1002/chir

TABLE 1. The chromatographic data of chiral compounds resolved by CSPs 1–3
CSP 1
CSP 2
CSP 3
S/N
k1
a
Rs
Eluent
k1
a
Rs
Eluent
k1
a
Rs
Eluent
1
2
3
4
5
6
7
8
1.96
2.14
1.15
1.20
0.42
0.43
A(70/30)a
B(40/60)b
0.44
0.35
0.24
0.31
0.64
1.18
1.27
1.18
1.46
1.24
0.51
0.39
0.28
0.92
0.30
C(20/80)c
D(80/20)d
E(70/30)f
D(90/10)h
C(10/90)c
1.39
0.27
0.50
0.39
0.30
0.20
0.74
0.80
0.35
0.78
0.70
0.32
0.63
0.20
1.31
3.14
3.64
1.32
2.52
1.17
1.87
2.01
1.21
1.16
1.83
1.45
0.45
2.94
3.06
1.35
4.38
1.00
0.77
0.60
0.38
0.58
0.36
0.90
0.85
0.86
0.65
0.61
B(90/10)c
A(10/90)e
E(50/50)e
D(90/10)g
E(60/40)e
B(90/10)e
B(90/10)g
B(90/10)g
F(30/35/35)c
B(90/10)i
E(50/50)i
C(20/80)c
C(20/80)c
E(90/10)i
1.95
1.92
2.07
1.22
1.22
1.22
0.34
0.35
0.44
A(10/90)g
A(10/90)g
E(90/10)h
9
1.90
1.88
3.10
1.20
1.30
1.08
0.50
0.77
0.43
A(10/90)b
C(10/90)c
B(90/10)g
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.24
0.65
1.90
1.11
0.33
0.52
B(30/70)f
C(80/20)i
0.55
1.76
0.73
E(50/50)j
0.27
0.20
1.64
1.38
0.65
0.59
E(70/30)i
E(60/40)i
1.92
2.25
1.37
1.11
0.37
0.28
E(40/60)i
A(30/70)i
0.75
0.33
0.91
0.28
1.18
1.52
1.24
1.48
0.19
0.53
0.50
0.25
C(10/90)i
C(10/90)f
B(60/40)f
D(90/10)f
0.25
0.28
0.25
0.52
0.27
0.31
0.27
0.25
1.81
0.23
0.28
0.17
1.37
1.29
1.39
1.19
1.19
1.30
1.17
1.25
1.29
1.74
1.21
1.49
0.42
0.32
0.59
0.38
0.31
0.42
0.19
0.27
0.25
1.08
0.42
0.82
A(30/70)i
D(90/10)h
C(30/70)b
E(70/30)f
E(90/10)e
E(60/40)i
C(10/90)f
A(30/70)h
A(50/50)f
A(50/50)f
E(90/10)g
D(80/20)i
2.00
1.91
2.19
1.11
1.15
1.08
0.30
0.46
0.19
A(30/70)h
D(90/10)f
D(90/10)g
0.23
0.25
0.21
1.98
1.27
1.32
1.10
0.17
0.25
C(10/90)f
A(50/50)f
C(10/90)f
1.91
1.10
0.27
A(30/70)h
0.24
1.73
0.51
A(50/50)f
1.92
1.84
1.93
1.90
1.08
1.09
1.14
1.30
1.15
1.16
0.23
0.38
0.65
0.35
0.17
A(50/50)c
A(30/70)h
C(20/80)c
A(30/70)g
E(70/30)a
0.23
0.27
0.26
1.34
1.19
1.23
0.51
0.31
0.41
E(90/10)h
E(90/10)g
E(90/10)g
0.42
0.21
0.99
0.38
1.18
1.67
1.72
1.10
0.56
0.48
0.80
0.15
F(30/35/35)f
B(90/10)f
D(70/30)i
E(50/50)e
1.99
1.11
0.30
D(90/10)f
0.13
0.22
0.26
1.39
1.49
1.74
0.14
0.23
0.49
A(70/30)h
A(70/30)f
A(30/70)f
0.38
0.28
0.44
1.58
1.43
1.89
1.06
0.45
1.91
E(50/50)b
B(30/70)f
E(40/60)i
1.70
2.33
1.75
1.10
1.17
1.18
0.36
0.48
0.52
A(70/30)g
C(10/90)i
A(70/30)a
0.16
0.24
0.18
0.29
1.77
1.31
1.92
1.16
0.85
0.33
1.01
0.21
E(90/10)g
E(70/30)f
E(90/10)g
E(90/10)d
0.76
0.99
2.15
1.35
1.06
0.35
B(90/10)i
D(70/30)i
1.90
1.89
2.11
3.45
2.18
1.10
1.13
1.19
1.08
1.18
0.27
0.32
0.47
0.33
0.38
A(70/30)a
A(70/30)a
C(10/90)c
E(70/30)a
B(40/60)b
1.32
1.10
0.18
E(40/60)i
0.29
1.06
1.14
1.15
0.26
0.68
E(90/10)d
D(70/30)i
1.77
1.97
2.44
1.90
2.09
1.13
1.21
1.15
1.12
1.19
0.46
0.61
0.50
0.38
0.51
A(70/30)i
C(30/70)b
C(10/90)c
A(50/50)c
C(30/70)b
0.26
1.21
1.30
1.07
0.66
0.42
A(50/50)c
C(80/20)b
0.29
0.24
0.10
0.27
1.16
1.26
4.12
1.26
0.31
0.38
0.75
0.40
E(90/10)g
E(90/10)i
B(70/30)i
E(90/10)g
0.28
1.20
0.22
C(10/90)i
1.89
2.03
1.15
1.10
0.43
0.19
A(10/90)b
E(90/10)h
0.22
0.27
0.24
1.40
1.28
1.79
0.42
0.44
0.85
A(70/30)g
E(90/10)g
A(30/70)g
0.86
0.29
0.25
2.76
0.26
0.24
1.70
1.23
1.29
2.14
2.69
3.24
2.67
0.27
0.28
1.47
0.65
1.26
D(80/20)f
C(10/90)g
A(30/70)f
2.13
1.17
0.58
B(70/30)h
A(10/90)i
0.06
3.53
0.30
A(70/30)g
C(10/90)g
F(30/35/35)f
Eluent: A: Methanol/ethanol; B: n-Hexane/isopropanol; C: n-Hexane/ethanol; D: Methanol/water; E: Acetonitrile/water; F: n-Hexane/ethanol/methanol. aThe
wavelength of UV detection 230 nm.
bThe wavelength of UV detection 245 nm.
cThe wavelength of UV detection 254 nm.
dThe wavelength of UV detection 200 nm.
eThe wavelength of UV detection 215 nm.
fThe wavelength of UV detection 210 nm.
gThe wavelength of UV detection 220 nm.
hThe wavelength of UV detection 225 nm.
iThe wavelength of UV detection 235 nm.
jThe wavelength of UV detection 270 nm.

CSPs OF SINGLE AND MIXED SELECTOR
233
TABLE 2. The representative chromatographic resolution of eight chiral compounds on the CSPs under the same
separation conditions
CSP 1
CSP 2
CSP 3
S/N
2
k1
a
Rs
k
k1
a
Rs
k
k1
a
Rs
k
Eluent
2.10
No separation
No separation
1.21
0.38
254
No separation
0.69
0.62
0.29
0.60
0.69
0.35
0.78
0.32
0.50
0.44
0.44
0.26
0.46
0.24
0.27
0.41
0.33
1.11
1.16
2.44
1.24
1.14
1.21
1.16
8.20
2.12
1.17
2.80
1.24
3.14
1.34
1.38
1.39
1.15
0.51
0.71
1.56
0.31
0.45
0.36
0.90
1.99
0.78
0.28
2.74
0.35
1.11
0.64
0.66
1.25
0.12
254
254
225
235
210
254
235
210
254
245
235
215
235
220
215
220
275
A(30/70)
B(70/15/15)
C(10/90)
D(80/20)
A(40/60)
B(30/35/35)
A(90/10)
E(70/30)
F(50/50)
F(60/40)
F(60/40)
E(70/30)
F(50/50)
E(50/50)
F(90/10)
C(80/20)
E(50/50)
0.75
0.22
0.35
0.77
No separation
No separation
0.27
0.70
0.20
No separation
0.45
0.51
No separation
0.27 1.26
No separation
0.23 1.35
1.08
0.39
0.73
0.39
0.90
254
210
200
210
3
1.85
1.27
1.49
6
No separation
1.98
3.10
No separation
0.55 1.76
No separation
No separation
No separation
10
11
13
1.19
1.08
0.45
0.43
254
220
2.66
1.39
1.38
1.22
0.62
0.59
210
220
235
0.73
270
15
38
41
1.58
1.20
0.48
0.23
254
210
44
53
55
1.95
No separation
2.37 1.25
No separation
1.15
0.36
1.21
240
210
0.40
0.30
220
220
Eluent: A: n-Hexane/isopropanol; B: n-Hexane/ethanol/methanol; C: n-Hexane/ethanol; D: Methanol/water; E: Ethanol/methanol; F: Acetonitrile/water. UV
detection: k (nm).
1650 cm21 designating to the amine is much stronger than
that approximately at 1720 cm21 designating to the esters in
the selectors, which are much less than amino groups in
amount. The absorbance peaks of esters are partially over-
lapped by those of amino groups, thus not separately appear-
ing in the spectra. On the basis of the increment of the con-
tents of carbon, the capacities of the two selectors on CSPs 1
atom introduced into CSP 3 during the immobilization of the
Selector 1, where one hydrogen atom left amianted silica gel
yielding hydrogen chloride when Selector 1 was immobi-
lized. The same case is for (n2H 2 1) during the immobiliza-
tion of Selector 2. DCC and DCH are available from the EA of
CSP 3 and the aminated silica gel. By resolving the equation
group consisting of eqs. 3 and 4, the capacities of Selectors
1 and 2 on CSP 3 were estimated as 0.43 lmol m22 and
0.30 lmol m22, respectively.
and 2 are estimated as 0.67 lmol m22 and 0.60 lmol m22
,
respectively.20 The capacities of Selectors 1 and 2 on CSP 3
are calculated as follows: Suppose x and y present the capaci-
ties (lmol m22) of Selectors 1 and 2 on CSP 3; nC1 and n2C
are carbon atom numbers contained in each molecule of
Selectors 1 and 2; n1H and n2H present hydrogen atom num-
bers contained in each molecule of Selectors 1 and 2. After
the immobilization of the two selectors, the content of carbon
and hydrogen on aminated silica gel increases and the incre-
ments are defined as DCC (%) and DCH (%). The carbon con-
tent increment that comes from Selectors 1 and 2 are
defined as DC1C and DC2C; the hydrogen content increment
are defined as DC1H and DCH2 . The relationship between the
selector capacities and the carbon content increment is
expressed by eqs. 1 and 2:
General Enantioseparation Evaluation of CSPs 1–3
The efficiency of the columns packed with CSPs 1–3 is
determined as 21,900, 20,500, and 15,100 plates per meter.
TABLE 3. The influence of the alcohols content on the
enantioseparation of the chiral analytes resolved by CSP 1 in
normal phase mode and polar organic mode
S/N k1
a
Rs UV detection (nm)
Eluent
4
2.06 1.24 0.52
2.02 1.24 0.44
2.02 1.22 0.41
2.10 1.36 0.78
2.10 1.27 0.66
2.00 1.16 0.38
2.13 1.36 0.91
2.07 1.29 0.62
2.05 1.19 0.37
1.89 1.17 0.39
1.93 1.11 0.35
1.92 1.12 0.34
1.58 1.68 1.24
1.91 1.33 0.86
1.86 1.37 0.85
1.83 1.33 0.75
1.88 1.26 0.58
1.97 1.21 0.61
1.99 1.17 0.42
1.97 1.17 0.42
225
220
235
220
220
220
245
254
254
254
230
254
245
254
220
215
220
245
245
245
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Ethanol/n-hexane: 60/40
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Methanol/ethanol: 10/90
Methanol/ethanol: 50/50
Methanol/ethanol: 70/30
Ethanol/n-hexane: 60/40
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
5
10 3 DCC1 ¼ 12 3 n1C 3 x::::::
10 3 DCC2 ¼ 12 3 n2C 3 y::::::
ð1Þ
ð2Þ
9
Equation 1 adds with eq. 2 to give eq. 3:
12n1Cx þ 12n2Cy ¼ 10ðDCC1 þ DCC2 Þ ¼ 10DCC::::::
ð3Þ
21
Similarly, based on the hydrogen content increment, eq. 4
is obtained:
31
32
13ðn1H ꢀ1Þx þ13ðnH2 ꢀ1Þy ¼ 10ðDCH1 þDCH2 Þ ¼ 10DCH ð4Þ
49
In eqs. 1 and 2, 12 is the relative atom mass of carbon. In
eq. 4, (n1H 2 1) refers to the actual numbers of hydrogen
Chirality DOI 10.1002/chir

234
CHEN ET AL.
TABLE 4. The influence of the alcohols content on the enantioseparation of the chiral analytes resolved by CSP 2 in normal phase
mode and polar organic mode
S/N
3
k1
a
Rs
UV detection (nm)
Eluent
0.22
0.22
0.22
0.23
0.22
0.24
0.23
0.30
0.11
0.11
0.45
0.13
0.27
0.12
0.20
0.18
0.23
0.21
0.30
0.25
0.08
1.95
1.85
1.44
1.33
2.86
1.71
1.61
3.39
2.21
1.90
1.58
1.71
3.40
1.93
1.53
1.50
1.78
1.54
2.50
1.99
2.79
0.91
0.73
0.40
0.31
1.78
0.98
0.86
1.42
0.47
0.25
0.48
0.20
1.40
0.35
0.61
0.42
1.00
0.42
1.03
0.76
0.32
210
210
210
220
210
220
220
220
254
225
254
225
235
225
220
220
220
210
210
235
210
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Methanol/ethanol: 30/70
Methanol/ethanol: 50/50
Ethanol/n-hexane: 70/30
Ethanol/methanol: 90/10
Ethanol/methanol: 70/30
Methanol/ethanol/n-hexane: 35/35/30
Methanol/ethanol/n-hexane: 45/45/10
Methanol/ethanol: 50/50
Methanol/ethanol: 30/70
Methanol/ethanol: 50/50
Methanol/ethanol/n-hexane: 35/35/30
Methanol/ethanol/n-hexane: 45/45/10
Methanol/ethanol/n-hexane: 25/25/50
Methanol/ethanol/n-hexane: 35/35/30
Ethanol/n-hexane: 50/50
23
38
41
55
59
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Ethanol/methanol: 50/50
Under normal and reversed phase modes and polar organic
mode, and in the linear separation conditions, CSPs 1–3
were undergone enantioseparation evaluation toward struc-
turally various chiral compounds (Figs. 2 and 3). The enan-
tioseparation was confirmed by the same UV absorbance fea-
ture of a pair of two enantiomers. The chromatographic
results are tabulated in Tables 1 and 2. CSPs 1 and 2,
respectively, separated 37 and 35 chiral compounds, whereas
47 chiral compounds were separated by CSP 3. Because the
selectors are close in their capacities and structures, CSPs 1
and 2 have the equivalent enantioseparation ability. In the
view of the numbers of separated chiral compounds, CSP 3
has the best enantioseparation ability. Under most tested
conditions, the resolutions on CSP 3 are better than those
on CSPs 1 and 2.
The difference in enantioseparation ability between CSP 3
and the two single selector CSPs is related to the structural
similarity of the two selectors in CSP 3. Based on the theory
of forming diastereoisomers during chiral recognition,6,7 two
enantiomers of a racemate interact with a selector through
stereoselective complexation to give two diastereoisomers.
The stability difference of the two diastereoisomers leads to
enantioseparation. The enantiomer forms the less stable dia-
stereoisomer is eluted out early, and another one is eluted
out late. There are two selectors in CSP 3, with which a pair
of enantiomers of a chiral analyte interact to produce two
group diastereoisomers. If the diastereoisomers results from
R (or S)-form isomer and the two selectors are both more
stable than the ones results from S (or R)-form isomer and
the two selectors, the effects of Selectors 1 and 2 on enantio-
separation enhance each other. In the event that these dia-
stereoisomers are not both more stable than the ones formed
between the S (or R)-form isomer and the two selectors, the
effects of Selectors 1 and 2 on enantioseparation are just
slightly enhanced, or not enhanced, or even impaired.
Because Selectors 1 and 2 were both prepared from tartaric
acid, except the derivatization with different substituents
(Fig. 1), they have the same steric configurations, which can
match that of an enantiomer forming stable diastereoisom-
ers. Relatively, the steric configuration of another enantiomer
cannot well fit those of Selectors 1 and 2 forming less stable
diastereoisomers, or not forming diastereoisomers. Further,
the substituents, i.e., 3,5-dimethylbenzoyl and 1-naphthalene-
carbonyl, are complementary in their structures or electronic
effects for the diastereoisomers formation by the complexa-
tion between the two selectors with an enantiomer. Accord-
ingly, the two selectors in CSP 3 consistently contribute
enantioseparation resulting in the better enantioseparation
ability, although the capacity of individual selectors in CSP 3
is lower in comparison with CSPs 1 and 2.
In the previous work, a mixed selector CSP was prepared
from L-dibenzoyl tartaric acid and (1R,2R)-1,2-diphenylethyle-
nediamine, where the two selectors are not so close in their
structures as those in this work. An enantiomer cannot
simultaneously well fit the structures of the two selectors.
Thus, the mixed selector CSP did not show enhanced enan-
tioseparation ability.
Influence of Alcohols Content of Mobile Phases on
Enantioseparation
Mobile phases not only elute chiral analytes out but also
impact enantioseparation. Alcohols are usually used as the
compositions for chiral HPLC. The influence of alcohols
upon enantioseparation is caused by modulating the polarity
of mobile phases or by involving in the interaction between
chiral analytes and selectors. Under normal phase mode and
polar organic mode, the effect of the content of methanol,
ethanol, and isopropanol on the enantioseparation was inves-
tigated (Tables 3–5). With the content increment of alcohols
or higher-polarity alcohols, the overall polarity of mobile
phase increases, however, the resolution decreases. The rea-
son for the observed trend is related to the following fact: the
functional groups in Selectors 1 and 2 are ester, while only
one amide bond is employed to link the selectors and the
Chirality DOI 10.1002/chir

CSPs OF SINGLE AND MIXED SELECTOR
235
TABLE 5. The influence of the alcohols content on the enantioseparation of the chiral analytes resolved by CSP 3 in normal phase
mode and polar organic mode
S/N
6
k1
a
Rs
UV detection (nm)
Eluent
0.60
0.42
0.40
0.39
0.34
0.33
0.23
0.22
0.51
0.21
0.26
0.25
0.24
0.29
0.28
0.28
0.75
0.25
0.24
0.26
0.26
0.22
0.32
0.28
0.26
0.26
0.31
0.30
0.30
0.23
0.26
0.25
0.28
0.29
1.13
1.18
1.17
1.14
1.16
1.16
1.32
1.27
1.16
1.26
1.22
1.20
1.19
1.49
1.34
1.24
1.11
1.28
1.21
1.29
1.24
1.58
1.30
1.46
1.40
1.30
1.73
1.68
1.52
1.61
3.14
3.48
1.96
2.08
0.50
0.42
0.30
0.15
0.26
0.22
0.41
0.27
0.69
0.32
0.20
0.23
0.25
0.49
0.38
0.25
0.65
0.35
0.26
0.43
0.35
0.91
0.60
0.72
0.56
0.39
1.02
0.91
0.58
0.76
0.62
1.01
0.50
0.49
210
220
215
215
210
205
215
225
210
220
235
220
205
220
225
215
220
220
205
235
215
210
235
210
220
235
220
220
210
210
215
215
220
210
Ethanol/n-hexane: 40/60
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
Ethanol/n-hexane: 90/10
Methanol/ethanol: 50/50
Methanol/ethanol: 70/30
Methanol/ethanol: 10/90
Methanol/ethanol: 30/70
Methanol/ethanol/n-hexane: 25/25/50
Methanol/ethanol/n-hexane: 45/45/10
Methanol/ethanol: 10/90
Methanol/ethanol: 30/70
Methanol/ethanol: 50/50
Ethanol/n-hexane: 60/40
Ethanol/n-hexane: 70/30
Ethanol/n-hexane: 80/20
11
23
Methanol/ethanol/n-hexane: 15/15/70
Methanol/ethanol/n-hexane: 35/35/30
Methanol/ethanol/n-hexane: 45/45/10
Methanol/ethanol: 10/90
41
55
Methanol/ethanol: 30/70
Methanol/ethanol/n-hexane: 25/25/50
Methanol/ethanol/n-hexane: 45/45/10
Ethanol/n-hexane: 40/60
Ethanol/n-hexane: 50/50
Ethanol/n-hexane: 60/40
Isopropanol/n-hexane: 40/60
Isopropanol/n-hexane: 50/50
Isopropanol/n-hexane: 60/40
Ethanol/n-hexane: 40/60
Ethanol/n-hexane: 80/20
Methanol/ethanol: 30/70
60
Methanol/ethanol: 50/50
Methanol/ethanol: 70/30
support. In addition to the p-p interaction between the aro-
matic rings in the selectors and in the chiral analytes, the
complexation between the selectors and the analytes is
mainly owed to the dipole-dipole interaction and H bonding
between the esters in the selectors and the functional groups
in analytes, such as carbonyl, carboxyl, etc. These interac-
tions will be weakened if the content of the alcohols or
higher-polarity alcohol contained in mobile phase increases,
because there is a competition for the analytes to interact
with alcohols and with the selectors.
tigations that possibly offers new considerations for the de-
velopment of chiral packing materials.
LITERATURE CITED
1. Welch CJ. Microscale chiral HPLC in support of pharmaceutical process
research. Chirality 2009;21:114–118.
2. Ryoo JJ, Heo KS, Choi E-S, Park JH, Lee W. Enantioseparation of tiropra-
mide by HPLC. Chirality 2004;16:S51–S54.
3. Wang X, Ching CB. Chiral separation of b-blocker drug nadolol by
HPLC—a kinetic and equilibrium study. J Sep Sci 2007;30:333–343.
4. Leonard WR Jr, Henderson DW, Miller RA, Spencer GA, Sudah OS,
Biba M, Welch CJ. Strategic use of preparative chiral chromatography
for the synthesis of a preclinical pharmaceutical candidate. Chirality 2007;
19:693–700.
CONCLUSIONS
Contrary to the previous work, in which the two selectors
in a mixed selector CSP were prepared from different chiral
compounds, the selectors in the present work were both
derived from the identical chiral compound. The correspond-
ing mixed selector CSP shows enhanced enantioseparation
ability. It remains unknown whether all mixed selector CSPs
whose selectors are very close in their structures show
improved enantioseparation ability comparing with the corre-
sponding single selector CSPs. Therefore, for mixed selector
CSPs, the relationship between the selector structures and
the enantioseparation ability is worth wider and deeper inves-
5. Cancelliere G, D’Acquarica I, Gasparrini F, Misiti D, Villani C. Synthesis
and application of novel, highly efficient HPLC chiral stationary phases: a
chiral dimension in drug research analysis. Pharm Sci Technol Today
1999;2:484–492.
6. Pirkle WH, Pochapsky TC. Considerations of chiral recognition relevant
to the liquid chromatographic separation of enantiomers. Chem Rev
1989;89:347–362.
7. Berthod A. Chiral recognition mechanisms. Anal Chem 2006;78:2093–
2099.
8. Berthod A. Chiral recognition mechanisms with macrocyclic glycopep-
tide selectors. Chirality 2009;21:167–175.
Chirality DOI 10.1002/chir

236
CHEN ET AL.
9. Kafri R, Lancet D. Probability rule for chiral recognition. Chirality 2004;
16:369–378.
15. Kontrec D, Vinkovic V, Sunjic V. New chiral stationary phases based on
(R)-1-naphthylethylamine bound to 2,4,5,6-tetrachloro-1,3-dicyanoben-
zene. Chirality 1999;11:722–730.
10. Zhang G, Sun Q, Hou Y, Hong Z, Zhang J, Zhao L, Zhang H, Chai Y.
New mathematic model for predicting chiral separation using molecular
docking: mechanism of chiral recognition of triadimenol analogues. J Sep
Sci 2009;32:2401–2407.
16. Iuliano A, Attolino E, Salvadori P. Biselector enantioselective stationary
phases for HPLC: dependence of the chiral discrimination properties on
stereochemistry and chemical nature of each unit of the chiral auxiliary.
Tetrahedron Asymmetry 2002;13:1805–1815.
11. Siret L, Tambute´ A, Be´gos A, Rouden J, Caude M. Steric hindrance influ-
ence on the enantiorecognition ability of tyrosine-derived chiral station-
ary phases. Chirality 1991;3:427–435.
17. Kip J, Haperen PV, Kraak JC. R-N-(Pentafluorobenzoyl)phenylglycine as
a chiral stationary phase for the separation of enantiomers by high-per-
formance liquid chromatography. J Chromatogr 1986;356:423-427.
12. Oxelbark J, Gidlund P. Investigation of a tartaric acid-based linear poly-
amide and dimer as chiral selectors in liquid chromatography. Chirality
2005;17:79–84.
18. Wei W-J, Deng H-W, Chen W, Bai Z-W, Li S-R. Preparation and enantiosepa-
ration of mixed selector chiral stationary phase derived from benzoylated tar-
taric acid and 1,2-diphenylethylenediamine. Chirality 2010;22:604–611.
13. Iuliano A, Attolino E, Salvadori P. (S)-Leucine and [(S)-1-(1-naphthyl)e-
thyl]amine as chiral building blocks for a bifunctional system—synthesis
of a new chiral stationary phase and evaluation of its biselector properties
19. Ihara T, Sugimoto Y, Asada M, Nakagama T, Hobo T. Influence of the
method of preparation of chiral stationary phases on enantiomer separations
in high-performance liquid chromatography. J Chromatogr A 1995;694:49–56.
in the HPLC resolution of racemic compounds. Eur
2001;2001:3523–3529.
J Org Chem
20. Yin C-Q, He B-J, Li S-R, Liu Y-Q, Bai Z-W. Immobilization of (1S,2R)-(1)-
2-amino-1,2-diphenylethanol derivates on aminated silica gel with differ-
ent linkages as chiral stationary phases and their enantioseparation evalu-
ation by HPLC. Chirality 2009;21:442–448.
14. Iuliano A, Lecci C, Salvadori P. The S-triazine moiety as a scaffold for connect-
ing different chiral auxiliaries: synthesis of new biselector CSPs for enantiose-
lective chromatography. Tetrahedron Asymmetry 2003;14:1345–1353.
Chirality DOI 10.1002/chir