Preparation and characterization of a new open-tubular capillary column for enantioseparation by capillary electrochromatography

Article abstract of DOI:10.1002/chir.23053

In order to use the enantioseparation capability of cationic cyclodextrin and to combine the advantages of capillary electrochromatography (CEC) with open-tubular (OT) column, in this study, a new OT-CEC, coated with cationic cyclodextrin (1-allylimidazolium-β-cyclodextrin [AI-β-CD]) as chiral stationary phase (CSP), was prepared and applied for enantioseparation. Synthesized AI-β-CD was characterized by infrared (IR) spectrometry and mass spectrometry (MS). The preparation conditions for the AI-β-CD-coated column were optimized with the orthogonal experiment design L₉(3⁴). The column prepared was characterized by scanning electron microscopy (SEM) and elemental analysis (EA). The results showed that the thickness of stationary phase in the inner surface of the AI-β-CD-coated columns was about 0.2 to 0.5?μm. The AI-β-CD content in stationary phase based on the EA was approximately 2.77?mmol·m^?2. The AI-β-CD-coated columns could separate all 14 chiral compounds (histidine, lysine, arginine, glutamate, aspartic acid, cysteine, serine, valine, isoleucine, phenylalanine, salbutamol, atenolol, ibuprofen, and napropamide) successfully in the study and exhibit excellent reproducibility and stability. We propose that the column, coated with AI-β-CD, has a great potential for enantioseparation in OT-CEC.

Full text of DOI:10.1002/chir.23053

Received: 13 August 2018
DOI: 10.1002/chir.23053
Revised: 5 December 2018
Accepted: 30 December 2018
R E G U L A R A R T I C L E
Preparation and characterization of a new open‐tubular
capillary column for enantioseparation by capillary
electrochromatography
Yingjie Li | Yimin Tang
| Shili Qin | Xue Li | Qiang Dai | Lidi Gao
College of Chemistry and Chemical
Engineering, Qiqihar University, Qiqihar,
PR China
Abstract
In order to use the enantioseparation capability of cationic cyclodextrin and to
combine the advantages of capillary electrochromatography (CEC) with open‐
tubular (OT) column, in this study, a new OT‐CEC, coated with cationic cyclo-
dextrin (1‐allylimidazolium‐β‐cyclodextrin [AI‐β‐CD]) as chiral stationary
phase (CSP), was prepared and applied for enantioseparation. Synthesized
AI‐β‐CD was characterized by infrared (IR) spectrometry and mass spectrome-
try (MS). The preparation conditions for the AI‐β‐CD‐coated column were opti-
mized with the orthogonal experiment design L9(34). The column prepared was
characterized by scanning electron microscopy (SEM) and elemental analysis
(EA). The results showed that the thickness of stationary phase in the inner
surface of the AI‐β‐CD‐coated columns was about 0.2 to 0.5 μm. The AI‐β‐
CD content in stationary phase based on the EA was approximately
2.77 mmol·m−2. The AI‐β‐CD‐coated columns could separate all 14 chiral
compounds (histidine, lysine, arginine, glutamate, aspartic acid, cysteine,
serine, valine, isoleucine, phenylalanine, salbutamol, atenolol, ibuprofen, and
napropamide) successfully in the study and exhibit excellent reproducibility
and stability. We propose that the column, coated with AI‐β‐CD, has a great
potential for enantioseparation in OT‐CEC.
Correspondence
Lidi Gao, University of Qiqihar, Building
6, Room D14, Qiqihar, Heilongjiang
161006, PR China.
Email: gaolidi@163.com
Funding information
48th overseas returnees project from the
Ministry of National Education; basic
special engineering general program of
Heilongjiang Provincial Education
Department, Grant/Award Number:
135209215; overseas scholar scientific
research project of Heilongjiang Provincial
Education Department, Grant/Award
Number: 1254HQ012
KEYWORDS
cationic cyclodextrin, chiral separation, open‐tubular capillary electrochromatography, thiol‐ene click
reaction
1 | INTRODUCTION
One enantiomer may produce the expected pharmacolog-
ical response whereas the opposite one may only show an
Chiral compounds have the same molecular formula and
similar chemical and physical properties, but they are not
the same completely because they have an opposite
arrangement of molecules structure like a mirror image,
so that they are very difficult to be distinguished.
Currently, more than two thirds of the drugs are chiral.
These enantiomers have different biological, physiologi-
cal, and pharmacological behaviors on human body.
antagonistic or a toxic response.1-3 These important dif-
ferences in pharmacological activities of enantiomers
cause a rigorous policy of the regulatory authorities,
which prescribed strict and specific guidelines for the
commercialization of chiral drugs. The guidelines empha-
size that approval will not be granted for a drug contain-
ing more than one isomer unless the pharmacokinetic
and pharmacodynamic properties of each isomer are
Chirality. 2019;1–10.
wileyonlinelibrary.com/journal/chir
© 2019 Wiley Periodicals, Inc.
1

2
LI ET AL.
described.4,5 In the agricultural field, the ineffective con-
figuration of chiral pesticides not only does not have the
effect of weeding but also may cause environmental pol-
lution and endanger human health. However, most of
the pesticides currently on the market are still in the form
of racemic forms.6,7 Therefore, it is necessary for
enantioseparation in the field of pharmaceutical and agri-
cultural science. Various methods have been developed
for enantioseparation, including high‐performance liquid
chromatography (HPLC), gas chromatography (GC),
supercritical fluid chromatography (SFC), and capillary
electrophoresis (CE). In addition, a hybrid technique,
such as capillary electrochromatography (CEC), has also
been used for enantioseparations.8-12
et al prepared eight single‐isomer ammonium‐β‐CD
derivatives with different side chains, and they were also
applied successfully for enantioseparation in CE.21 On
the other hand, Yao et al prepared a novel cationic CD
CSP by thiol‐ene click chemistry and used to separate
the dansyl amino acids, carboxylic aryl compounds, and
flavonoids by HPLC.22 Zhou et al prepared also a cationic
CD CSP and successfully used it to separate several race-
mic enantiomers by HPLC.23 However, most researches
were focused on CE or HPLC; a few researches were
reported to apply cationic CD to OT‐CEC for
enantioseparation.
Encouraged by the enantioseparation capability of
cationic CD and to combine the advantages of CEC with
OT column, in this study, a novel type of chiral OT‐CEC
column was prepared with cationic CD, AI‐β‐CD, as the
stationary phase and then used it to separate chiral
analytes, such as amino acids, drugs, and pesticide.
Meanwhile, the effects on enantioseparation by the inter-
actions between the analytes and CSPs were explored.
Finally, the repeatability and stability regarding the
AI‐β‐CD‐coated OT column were studied.
CEC is a very powerful tool for enantioseparations
because it combines the high sensitivity of HPLC with
the high‐separation efficiency of CE. In recent years,
enantioseparations using CEC attract more and more
attention because of the short analysis time and low sam-
ple and solvent consumption.13-16 In CEC, the column is
the most critical component. Depending on fabricating
techniques, three types of columns have been employed
including packed columns, monolithic columns, and OT
columns. Open‐tubular (OT)‐CEC has some advantages
in simple preparation process, infrequent bubble forma-
tion, low back pressure, and short analysis time, com-
pared with packed and monolithic CEC.17 However,
because of the limited amount of stationary‐phase coating
and/or its weak bonding strength to the inner wall of col-
umns, the OT‐CEC also suffers some problems, such as
relatively low sample capacity, short lifespan, and narrow
pH range, which result in the low‐separation capability
and the limited practical application. Early in 1992,
Mayer and Schurig separated enantiomers by OT‐CEC
using capillaries coated with a polydimethylsiloxane
(PDMS) bound permethylated Chirasil‐Dex (CD).18 Then,
Fang et al prepared a novel β‐CD‐GNPs‐coated OT‐CEC;
three tested enantiomers of four pairs were baseline
separated by this chiral column. Meanwhile, the
columns exhibited good repeatability and stability for
enantioseparation.14 To prepare a high‐performance and
stable OT column, it is the crucial to find a suitable mate-
rial as the stationary phase and immobilize it on the wall
of capillary in OT‐CEC for entioseparations.19 Cyclodex-
trin (CD) and their derivatives (CDs) have been widely
used as chiral stationary phases (CSPs), and the
researches grow tremendously.15 Cationic CDs become
an important branch of functionalized CDs and have
exhibited great potential in enantioseparations. For exam-
ple, Yu et al prepared a new cationic β‐CD derivative,
mono‐6‐deoxy‐6‐piperdine‐ β‐CD (PIP‐β‐CD), and suc-
cessfully applied for the enantioseparation of meptazinol
and its three intermediate enantiomers by CE.20 Wang
2 | MATERIALS AND METHODS
2.1 | Reagents and materials
All reagents used were of analytical grade unless other-
wise stated. β‐CD, anhydrous n,n‐dimethylformamide
(DMF), and dimethyl sulfoxide (DMSO) were ordered
from Kermel Chemical Reagent Co (Tianjin, China).
γ‐Mercaptopropyltrimethoxy‐silane
(γ‐MPTS),
azobisisobutyronitrile (AIBN), 1‐allylimidazole, and 4‐
toluenesulfonylchloride (TsCl) were purchased from
Acros Chemical Co (NJ, USA). All the amino acids race-
mates were obtained from Sinopharm Chemical Reagent
Co (Shanghai, China). Ibuprofen, atenolol, and
salbutamol were purchased from Pharmaceutical and
Biological Products Inspection (Beijing, China).
Napropamide was obtained from Aladdin Industrial Co
(Shanghai, China). All solvents used in CEC were of
chromatographic grade. Fused silica capillaries (50‐μm i.
d. and 375‐μm o.d.) were purchased from Yongnian Opti-
cal Fiber Factory (Hebei, China).
2.2 | Instrumentation
All the CEC experiments were performed on an Agilent
7100CE system (Agilent Technologies, Waldbronn,
Germany) equipped with diode‐array detector. A syringe
pump P200 (Elite, China) was employed to inject the mix-
tures into the fused silica capillary. Mass spectra were

LI ET AL.
3
recorded on 6230TOFMS mass spectrometer (CA, USA).
Scanning electron microscopy (SEM) images were
obtained by an S‐4300 (Hitachi, Japan) SEM. The
Fourier‐transform infrared (FTIR) spectra were measured
with SpectrumOne FTIR spectrometer (PE Co, USA). EA
was performed on PE2400 organic element analyzer
(PE Co, USA).
2.5 | CEC process
The AI‐β‐CD‐coated column was installed on the CE instru-
ment; detecting wavelength of analytes was set at 200 nm
except for napropamide (220 nm). Before started, the AI‐β‐
CD‐coated column was rinsed with ultrapure water for
5 minutes and running buffer for another 5 minutes. The
sample was injected into the AI‐β‐CD‐coated column by
50 mbar × 3 seconds and separated immediately by apply-
ing a running voltage of 15 kV. The CEC was performed
with 20‐mM acetate buffer (pH 3.0‐10.0) at room tempera-
ture. All the solutions including the test samples were
filtered through a 0.22‐μm pore‐size PES membrane.
2.3 | Synthesis of AI‐β‐CD
The preparation pathway of AI‐β‐CD was reported
previously, and it was diagramed in Figure 1.22,24
3 | RESULTS AND DISCUSSION
3.1 | Characterization of AI‐β‐CD
2.4 | Preparation of the AI‐β‐CD‐coated
OT‐CEC columns
FTIR analysis were carried out to verify the functioning
groups on AI‐β‐CD (Figure 2). The β‐CD had the charac-
teristic absorptions at 3400 cm−1(νO―H), 2927 cm−1
(ν―CH2―CH2―), 1640 cm−1(νC―H), and 1028 cm−1(νC―O).
The 6‐OTs‐β‐CD had the characteristic absorptions at
1365 and 843 cm−1, which represented the S¼O and
C―H of benzene rings (Figure 2B), respectively. The
AI‐β‐CD had the characteristic absorptions at 1566 and
1660 cm−1, which represented the C¼C in imidazole rings
and C¼C in allyl group, as shown in Figure 2C. Addition-
ally, mass spectrometry (MS) was applied to confirm fur-
ther AI‐β‐CD structure (Figure 3). In mass spectra of
AI‐β‐CD, m/z = 1225.49708 was assigned to the (M+) peak
in AI‐β‐CD. These results showed that allylimidazolyl has
been successfully introduced into AI‐β‐CD.
Briefly, the bare capillary was pretreated.25 Then, the cap-
illary was rinsed with 80% (v/v) ethanol solution of γ‐
MPTS for 30 minutes, both ends sealed and treated in
water bath for 24 hours at 40°C. In the process, γ‐MPTS
was bonded on the inner wall of bare capillary by
dehydration‐condensation reaction. After that, the capil-
lary was washed with methanol to remove the residual
reagents and dried by a nitrogen stream. At this point,
the inner wall of the capillary was modified by a layer
of γ‐MPTS with free ―SH groups. Next, AI‐β‐CD, AIBN,
and DMSO were mixed and sonicated for 20 minutes;
the mixture was poured into the modified capillary. With
both ends sealed, the capillary was heated in water bath
to couple the AI‐β‐CD with the layer of γ‐MPTS by
thiol‐ene click chemistry reaction.26 Finally, the AI‐β‐
CD‐coated columns were washed with methanol thor-
oughly to remove the residual of reactants, then flushed
with the running buffer for 10 minutes and filled with
ultrapure water and stored until use. The AI‐β‐CD‐coated
column was characterized by SEM and EA. The whole
process for the preparation was diagramed in Figure 1.
3.2 | Optimization of preparation
conditions of AI‐β‐CD‐coated columns
In order to optimize the preparation conditions of
AI‐β‐CD‐coated columns, reaction temperature, reaction
FIGURE 1 Synthesis pathway of the
AI‐β‐CD‐coated capillaries

4
LI ET AL.
(C)
(B)
(A)
FIGURE 2 Fourier‐transform infrared
(FTIR) spectra (A) β‐CD, (B) 6‐OTs‐β‐CD,
and (C) AI‐β‐CD
TABLE 1 The experimental design based on the Taguchi's L9(34)
orthogonal array and the results for the optimization on preparation
conditions of AI‐β‐CD‐coated column
Reaction
Temperature Reaction CAI‐β‐CD
Factor
Experiment
No.
(°C)
A
Time (h) (g/mL) Index
B
C
Rs
1
2
3
4
5
6
7
8
9
60
24
36
48
24
36
48
24
36
48
0.20
0.25
0.30
0.25
0.30
0.20
0.30
0.20
0.25
2.83
3.46
4.70
1.87
0.82
1.11
1.42
1.22
1.78
1.03
1.50
0.98
1.13
60
60
70
70
70
FIGURE 3 Mass spectra for AI‐β‐CD
80
80
80
time, and AI‐β‐CD concentration were investigated. The
orthogonal experiment design L9(34) was applied for the
optimization. The main significant factors and levels can
be identified by the analysis of variance as shown in
Table 1, where the resolution (Rs) of histidine enantio-
mers was used to weigh the performance of AI‐β‐CD‐
coated columns under different conditions.
According to the analyses, the performance of
AI‐β‐CD‐coated columns was influenced by the three
factors (Table 1), and the influence intensity was in the
order of AI‐β‐CD concentration > reaction tempera-
ture > reaction time. On the basis of the analysis results,
AI‐β‐CD solution with 0.3 g/mL, which is close to
saturation, was optimal one (T3C > T2C > T1C). Grafting
of the AI‐β‐CD to capillary columns by thiol‐ene reaction
can be realized at mild reaction temperature that is
benefit to long‐term stability.25,27 The maximum
T1j
T2j
T3j
Rj
3.35
4.03
3.61
0.68
3.54
3.87
3.58
0.33
DL‐histidine was used as a model analyte. The other experimental condi-
tions: capillary column: 40‐cm total length (26.5‐cm effective length),
50‐μm i.d.; 20‐mM acetate buffer solution; pH: 5.0; separation voltage:
15 kV; injection: 50 mbar × 3 seconds; detection wavelength: 200 nm. T
was the sum of indexes of the same level, R was the difference between the
maximum index and the minimum index from the same factor and level,
and j represented the different factors.
of 24 hours was not long enough to immobilize sufficient -
AI‐β‐CD for the enantioseparation; meanwhile, the reac-
tion time of 48 hours did not make the
resolution better. Therefore, 36 hours was chosen for
the preparation of the AI‐β‐CD‐coated columns
(T2B > T3B > T1B).
separation
efficiency
was
achieved
at
70°C
(T2A > T3A > T1A). In addition, the reaction time can also
affect the immobilization of AI‐β‐CD. The reaction time

LI ET AL.
5
3.3 | Characterization of the
AI‐β‐CD‐coated columns
In this work, AI‐β‐CD was readily immobilized on the
inner wall of a capillary column by thiol‐ene click chem-
istry reaction and served as an OT‐CEC stationary phase.
The morphological structure and thickness of the inner
surface were characterized by SEM. As shown in
Figure 4A, the inner surface of a bare capillary was very
smooth; after coating, it became much rough (Figure 4
B,C). It was clear that some aggregates were tightly
coated onto the inner wall of the AI‐β‐CD‐coated col-
umns and the thickness was about 0.2 to 0.5 μm, which
indicated that a solid phase was made successfully on
the inner wall of the capillary column.
Simultaneously, elemental analysis was conducted
and confirmed that the elements of C, H, and N exist
on the surface of the bare capillary column and the
AI‐β‐CD‐coated column, respectively. It could be found
that the difference between the nitrogen content of the
bare capillary column and the AI‐β‐CD‐coated column
FIGURE 5 Effect of eluent pH on the electroosmotic mobility.
The experimental conditions: sample: thiourea; 20‐mM acetate
buffer solution (pH 3.0‐10.0); separation voltage: 15 kV; injection:
50 mbar × 3 seconds; detection wavelength: 200 nm
blocking by silanol. The EOF reached 1.22 × 10−4 cm2 V
−1
s−1 for the coated column at pH 8.0, compared with
3.23 × 10−4 cm2 V−1 s−1 for the bare one. Additionally,
unlike that of the bare capillary column, EOF of the
AI‐β‐CD‐coated column was relatively stable in the stud-
ied pH range (Figure 5). The relative standard deviation
(RSD) of the EOF on the AI‐β‐CD‐coated column was
1.42% at pH 8.0, which was lower than that on the bare
one (RSD = 3.29%, n = 3). This result further indicates
that the coated surface is stable and reproducible, which
may facilitate the applications of AI‐β‐CD on CEC.
was significant; they were 0.70%
0.05% versus
0.85% 0.06%, respectively. The difference between the
carbon content of the bare capillary column and the
AI‐β‐CD‐coated column was also significant; they were
8.32%
0.08% to 11.47%
0.06%, respectively.
Meanwhile, the hydrogen content increased from
0.25%
0.66%
0.05% of the bare capillary column to
0.04% of the AI‐β‐CD‐coated column. The ele-
mental analysis results demonstrated that the coating
procedure was successful. The content of the AI‐β‐CD
precursor was calculated from elemental analysis, and it
3.4 | Enantioseparation performance of
AI‐β‐CD‐coated columns
was approximately 2.77 mmol·m−2
.
The electroosmotic flow (EOF) is the key factor in OT‐
CEC, and it is highly affected by buffer pH. In the present
study, we investigated the EOF values on both the bare
capillary column and the AI‐β‐CD‐coated column, with
varied pH of 3.0 to 10.0. As shown in Figure 5, the EOF
increased gradually regardless of the bare capillary or
the AI‐β‐CD‐coated column. The AI‐β‐CD‐coated column
had lower EOF than the bare capillary column had due to
The enantiomers, including 10 amino acids (histidine,
lysine, arginine, glutamate, aspartic acid, cysteine, serine,
valine, isoleucine, phenylalanine), three drugs
(salbutamol, atenolol, and ibuprofen), and one pesticide
(napropamide), were separated on the AI‐β‐CD‐coated
column by CEC. As shown in Table 2 and Figure 6, most
enantiomers could be separated successfully by the
(A)
(B)
(C)
FIGURE 4 Scanning electron microscopy (SEM) images for the inner surface of (A) bare capillary column (×5.0 k), (B) AI‐β‐CD‐coated
column (×5.0 k), and (C) AI‐β‐CD‐coated column (×20 k)

6
LI ET AL.
TABLE 2 The optimum pH for the separation of each analyte
Type
Analytes
Structure
PKa (PKa1/PKa2)
pI
Optimum pH
Rs
α
Alkaline amino acids
Histidine
1.82/9.17
7.59
5
1.78
1.87
Lysine
2.18/8.95
9.74
5
1.04
1.22
Arginine
2.17/9.04
2.19/9.67
10.76
3.22
8
9
2.47
0.71
2.29
1.21
Acidic amino acids
Glutamate
Aspartic acid
Cysteine
2.09/9.82
1.71/8.33
2.97
5.02
8
8
0.74
4.14
1.26
1.66
Serine
Valine
2.21/9.15
2.32/9.62
5.68
5.91
8
8
2.25
1.68
1.35
1.26
Isoleucine
2.36/9.68
6.02
8
0.89
1.13
Phenylalanine
Salbutamol
1.83/9.13
10.12
5.48
8
8
1.56
2.64
1.32
2.43
Chiral compounds
‐
Atenolol
‐
‐
‐
‐
6
3
3.47
1.51
3.14
1.22
Napropamide
Ibuprofen
4.55
‐
7
3.99
2.10
Assume that all the separation curves in this experiment satisfy the Gaussian curve. Rs: the resolution, when the baseline separation is achieved; the Rs can be
calculated by the formula, Rs = 2 (t2 − t1)/(W1 + W2); if the baseline separation has not been achieved, Rs can be calculated by the formula, Rs = 1.18 (t2 − t1)/
(Wh1 + Wh2), where W is the peak width (minute), Wh is the half width (minute), t is the the migration time (minute), and α is the the selectivity factor, α = t2/t1.
coated column, and the Rs and the selectivity factors (α)
varied from 0.71 to 4.14 and from 1.13 to 3.14, respec-
tively, for different enantiomers. All of the racemates
were completely separated with α > 1.05 under the stated
conditions.28 Meanwhile, the enantiomers of histidine,
arginine, cysteine, serine, valine, phenylalanine,
salbutamol, atenolol, napropamide, and ibuprofen
reached very good baseline separation with Rs less than
1.5. In this work, the influence of buffer pH on the
enantioseparation was evaluated from pH 3.0 to 10.0 for
alkaline amino acids (Figure 7A). When the pH is in
the range of 3.0 to 10.0, the AI‐β‐CD‐coated column
shows a more stable cathodic EOF. Typically, when histi-
dine is used for enantioseparation, since histidine is an
alkaline amino acid, if the pH is less than pI and greater
than pKa1, histidine will have a positive charge; its

LI ET AL.
7
FIGURE 6 Chiral separation chromatograms of analytes. 1. Histidine, 2. lysine, 3. arginine, 4. glutamate, 5. aspartic acid, 6. cysteine, 7.
serine, 8. valine, 9. isoleucine, 10. phenylalanine, 11. salbutamol, 12. atenolol, 13. napropamide, and 14. ibuprofen. The experimental
conditions: capillary column: 40‐cm total length (26.5‐cm effective length), 50‐μm i.d.; 20‐mM acetate buffer solution; separation voltage:
15 kV; injection: 50 mbar × 3 seconds; detection wavelength: 200 nm except for napropamide (220 nm); optimum pH of 14 analytes is shown
in Table 2
moving direction is consistent with the direction of EOF.
Herein, as the pH ranges from 3.0 to 6.0, the migration
time of histidine enantiomers shortens. However, the
resolution increases when pH ranges from 3.0 to 5.0 and
then decreases if pH increases up to 6.0. This might be
due to the shortening of the interaction time and the

8
LI ET AL.
TABLE 3 Effect of pH
Sample
pH
tL (min)
tD (min)
Rs
α
(A)
Histidine
3.0
4.0
5.0
6.0
3.56
2.53
2.08
1.22
4.56
3.91
3.89
2.24
0.76
1.42
1.78
1.30
1.28
1.55
1.87
1.84
The experimental conditions: capillary column: 40‐cm total length (26.5‐cm
effective length), 50‐μm i.d.; sample: histidine; 20‐mM acetate buffer solu-
tion; pH: 3.0 to 6.0; separation voltage: 15 kV; injection: 50 mbar × 3 seconds;
detection wavelength: 200 nm.
and the chiral separation was achieved successfully. If the
amino acids are acidic and the buffer pH is greater than
pI and less than pKa2, the amino acid will have the neg-
ative charge, which will result in repulsion between the
negatively charged amino acid and the cathode. When
the optimum pH for isolating these amino acids was
larger than their pI, such as cysteine, serine, valine, iso-
leucine, and phenylalanine, these amino acids can inter-
act with the cationic AI‐β‐CD by electrostatic
interaction and finally result in the separation for the
enantiomers (Figure 7B). However, for glutamate and
aspartic acids, they have a long retention time and low
resolution. As reported by Yao et al, the long retention
may be because of the stronger electrostatic interaction
(B)
22
generated by the second ―COO− moiety. Besides, the
competition of the two ―COO− moieties with the cat-
ionic sites on CSPs may weaken the “three‐point” interac-
tion; that is, three side groups of the chiral center have to
(C)
be held in place by specific interactions with its environ-
29
ment, either attractive or repulsive,
hence reduce the
enantioselectivity. Additionally, salbutamol, atenolol,
napropamide, and ibuprofen were separated successfully
based on their alkalinity or acidity; the optimum pH
was 8.0, 6.0, 3.0, and 7.0, respectively (Figure 7C). It
needs to emphasize that the hydrophobic interaction
between hydrophobic groups (such as benzene rings)
and the hydrophobic cavity of β‐CD also plays an impor-
tant role in the separation. The hydrophobic, hydrogen
bonding, π‐π, and electrostatic interactions between the
analytes and the CSPs all contribute to chiral recognition
process. These results show that the AI‐β‐CD‐coated col-
umn has excellent performance on enantioseparation. In
addition, the separation method used in this study has
the advantages of short time and low consumption.
FIGURE 7 Effect of pH on the resolution of the analytes. (A) 1.
Histidine, 2. lysine, and 3. arginine, (B) 4. glutamate, 5. aspartic
acid, 6. cysteine, 7. serine, 8. valine, 9. isoleucine, and 10.
phenylalanine, (C) 11. salbutamol, 12. atenolol, 13. napropamide,
and 14. ibuprofen
weakening of the interaction force between the analytes
and the CSPs. Thus, pH 5.0 was chosen for the chiral sep-
aration of histidine enantiomers based on the resolution
and migration time (Table 3). Since alkaline amino acids
have a group of ―NH2 or ―NH and easily bind to H+ to
form ―NH3+ and ―NH2+ when pH is less than pI value,
they produce a hydrogen bond with hydroxyl group on
the surfaces of β‐CD, so that the intermolecular forces
between the AI‐β‐CD and the enantiomers are enhanced,
3.5 | Reproducibility and stability
The reproducibility and stability about the AI‐β‐CD‐
coated columns were evaluated in terms of the RSDs on
the retention time and resolution using histidine enantio-
mers as the testing sample, and the results were exhibited

LI ET AL.
9
TABLE 4 The reproducibility of the column for the
enantioseparation
relatively stable EOF at varying pH values and better chi-
ral separation performance for 14 chiral compounds (his-
tidine, lysine, arginine, glutamate, aspartic acid, cysteine,
serine, valine, isoleucine, phenylalanine, salbutamol,
atenolol, ibuprofen, and napropamide). Moreover, the
AI‐β‐CD‐coated column exhibits an excellent reproduc-
ibility and stability. These results indicate that the OT‐
CEC column coated with AI‐β‐CD has great potential
for enantioseparation.
Type and Numbers
(n) of Experiments
Rs
(% RSD)
tL
tD
(% RSD)
(% RSD)
Run to run (n = 6)
Day to day (n = 6)
1.72
2.35
1.20
2.23
3.04
0.39
1.79
2.43
Column to column (n = 6) 3.13
Abbreviation: RSD, relative standard deviation. The experimental conditions:
capillary column: 40‐cm total length (26.5‐cm effective length), 50‐μm i.d.;
sample: histidine; 20‐mM acetate buffer solution; pH: 5.0; separation voltage:
15 kV; injection: 50 mbar × 3 seconds; detection wavelength: 200 nm.
ACKNOWLEDGMENTS
We gratefully appreciate the financial support from the
48th overseas returnees project from the Ministry of
National Education, the overseas scholar scientific
research project of Heilongjiang Provincial Education
Department (no. 1254HQ012), and basic special engineer-
ing general program of Heilongjiang Provincial Educa-
tion Department (no. 135209215), China.
ORCID
Yimin Tang https://orcid.org/0000-0003-1271-9592
REFERENCES
1. Declerck S, Heyden YV, Mangelings D. Enantioseparations of
pharmaceuticals with capillary electrochromatography:
a
FIGURE 8 The chromatogram of histidine enantiomers with
different runs. The experimental conditions were described as in
Table 1
review. J Pharm Biomed Anal. 2016;130:81‐99.
2. Fang L, Yu J, Jiang Z, Guo X. Preparation of a β‐cyclodextrin‐
based open‐tubular capillary electrochromatography column
and application for enantioseparations of ten basic drugs. Plos
One. 2016;11(1):e0146292.
in Table 4. All experiments were repeated for three times.
As shown in Table 4, the RSDs for run to run, day to day,
and column to column were less than 4%. Moreover, the
separation efficiency of the AI‐β‐CD‐coated columns did
not decrease obviously over 120 runs (Figure 8). These
results demonstrate that the AI‐β‐CD‐coated columns
prepared have excellent reliability and reproducibility.
Although the AI‐β‐CD‐coated column was flushed by
large amount of buffers varying pH from 3.0 to 10.0, the
retention time and chiral selectivity are reproducible
and indicate that the stability of the cationic CSP is great.
This may be attributed to the strong thiolether linkages
and stable imidazolium moiety, which are able to with-
stand various operating conditions.
3. Li L, Cheng B, Zhou R, Cao Z, Zeng C, Li L. Preparation and
evaluation of a novel n‐benzyl‐phenethylamino‐β‐cyclodextrin‐
bonded chiral stationary phase for HPLC. Talanta. 2017;174:
179‐191.
4. Ghanem A, Adly FG, Sokerik Y, Antwi NY, Shenashen MA,
El‐Safty SA. Trimethyl‐β‐cyclodextrin‐encapsulated monolithic
capillary columns: preparation, characterization and chiral
nano‐LC application. Talanta. 2016;Online(6):417‐431.
5. Adly FG, Antwi NY, Ghanem A. Cyclodextrin‐functionalized
monolithic capillary columns: preparation and chiral applica-
tions. Chirality. 2016;28(2):97‐109.
6. De Albuquerque NCP, Carrão DB, Habenschus MD, de Oliveira
ARM. Metabolism studies of chiral pesticides: a critical review. J
Pharmaceut Biomed. 2017;147:89.
7. Zhao P, Zhao J, Lei S, Guo X, Zhao L. Simultaneous enantio-
meric analysis of eight pesticides in soils and river sediments
by chiral liquid chromatography‐tandem mass spectrometry.
Chemosphere. 2018;204:210‐219.
4 | CONCLUSIONS
The preparation and charaterization of the AI‐β‐CD‐
coated OT‐CEC column for enantioseparation were
described. The AI‐β‐CD‐coated column possesses
8. Dong S, Sun Y, Zhang X, Li H, Luo G, Zhao L. Nanocellulose
crystals derivative‐silica hybrid sol open tubular capillary col-
umn for enantioseparation. Carbohydr Polym. 2017;165:359‐367.

10
LI ET AL.
9. Miao C, Bai R, Xu S, Hong T, Ji Y. Carboxylated single‐walled
carbon nanotubefunctionalized chiral polymer monoliths for
20. Yu J, Zhao Y, Song J, Guo X. Enantioseparation of meptazinol
and its three intermediate enantiomers by capillary electropho-
resis using a new cationic β‐cyclodextrin derivative in single
and dual cyclodextrin systems. Biomed Chromatogr. 2014;28(6):
868‐874.
affinity capillary electrochromatography.
J Chromatogr a.
2017;1487:227‐234.
10. Huang XY, Di DL. Chiral separation by counter‐current chro-
matography. Trac Trend Anal Chem. 2015;67:128‐133.
21. Wang Y, Zhou J, Liu Y, Tang J, Tang W. Evaluation of the chiral
separation ability of single‐isomer cationic β‐cyclodextrins in
capillary electrophoresis. Electrophoresis. 2015;35(19):2744‐2751.
11. De KK, Vander HY, Mangelings D. Generic chiral method
development in supercritical fluid chromatography and ultra‐
performance supercritical fluid chromatography. J Chromatogr
a. 2014;1363:311‐322.
22. Yao X, Tan T, Wang Y. Thiol‐ene click chemistry derived cat-
ionic cyclodextrin chiral stationary phase and its enhanced
separation performance in liquid chromatography.
Chromatogr A. 2014;1326:80‐88.
J
12. Mangelings D, Heyden YV. Enantioselective capillary
electrochromatography: recent developments and new trends.
Electrophoresis. 2011;32(19):2583‐2260.
23. Zhou J, Yang B, Tang J, Tang W. Cationic cyclodextrin clicked
chiral stationary phase for versatile enantioseparations in high‐
13. Kapnissi‐Christodoulou CP, Nicolaou AG, Stavrou IJ.
performance liquid chromatography.
J
Chromatogr A.
Enantioseparations
in
open‐tubular
capillary
2016;1467:169‐177.
electrochromatography: recent advances and applications. J
Chromatogr A. 2016;1467:145‐154.
24. Guo J, Zhang Q, Peng Y, et al. A facile and efficient one‐step
strategy for the preparation of β‐cyclodextrin monoliths. J Sep
Sci. 2013;36(15):2441‐2449.
14. Fang L, Wang P, Wen X, et al. Layer‐by‐layer self‐assembly of
gold nanoparticles/thiols β‐cyclodextrin coating as the station-
ary phase for enhanced chiral differentiation in open tubular
capillary electrochromatography. Talanta. 2017;167:158‐165.
25. Li Y, Song C, Zhang L, Zhang W, Fu H. Fabrication and evalu-
ation of chiral monolithic column modified by beta‐cyclodextrin
derivatives. Talanta. 2010;80(3):1378‐1384.
15. Guo J, Lin Y, Xiao Y, Crommen J, Jiang Z. Recent developments
in cyclodextrin functionalized monolithic columns for the
enantioseparation of chiral drugs. J Pharm Biomed Anal.
2016;130:110‐125.
26. Xu Y, Xiong X, Cai L, Tang Z, Ye Z. Thiol‐ene click chemistry.
Prog Chem. 2012;24:385‐394.
27. Tianjin University. Clicking preparation and application of allyl
cationic cyclodextrin chiral resolution material: China,
CN105435760[P]. 2016‐03‐30.
16. Al‐Hussin A, Boysen RI, Saito K, Hearn MTW. Preparation and
electrochromatographic characterization of new chiral β‐
cyclodextrin poly(acrylamidopropyl) porous layer open tubular
capillary columns. J Chromatogr a. 2014;1358:199‐207.
28. Wang X, Zhou A, Gao R. A review on chiral capillary
electrochromatography technique. Mod Instrum. 2003;02:10‐14.
17. Fu Q, Zhang K, Gao D, et al. Escherichia coli, adhesive coating
29. Davankov VA. The nature of chiral recognition: is it a three‐
point interaction? Chirality. 1997;9(2):99‐102.
as
a chiral stationary phase for open tubular capillary
electrochromatography enantioseparation. Anal Chim Acta.
2017;969:63‐71.
How to cite this article: Li Y, Tang Y, Qin S, Li
X, Dai Q, Gao L. Preparation and characterization
of a new open‐tubular capillary column for
enantioseparation by capillary
18. Mayer
S,
Schurig
V.
Enantiomer
separation
by
electrochromatography on capillaries coated with Chirasil‐Dex.
Dedicated to Professor Ernst Bayer on the occasion of his 65th
birthday. J High Resol Chromatogr. 1992;15(2):129‐131.
19. Li M. Application of graphene as the stationary phase for open‐
electrochromatography. Chirality. 2019;1–10.
https://doi.org/10.1002/chir.23053
tubular capillary electrochromatography.
J Chromatogr A.
2013;1277:93‐97.