Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation

Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation

ARTICLE IN PRESS JID: CHROMA [m5G;September 11, 2019;19:1] Journal of Chromatography A xxx (xxxx) xxx Contents lists available at ScienceDirect J...

2MB Sizes 0 Downloads 11 Views

ARTICLE IN PRESS

JID: CHROMA

[m5G;September 11, 2019;19:1]

Journal of Chromatography A xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation Siru Ren a, Song Xue b, Xiaodong Sun c, Mengjie Rui a, Li Wang a, Qi Zhang a,∗ a b c

School of Pharmacy, Jiangsu University, 301 Xuefu Road, Zhenjiang, Jiangsu 212013, PR China Department of Pharmacy, Affiliated Hospital of Jiangsu University, Zhenjiang, 212013, PR China Medical School (In preparation), Shanghai University, Shanghai 200444, PR China

a r t i c l e

i n f o

Article history: Received 3 May 2019 Revised 2 September 2019 Accepted 3 September 2019 Available online xxx Keywords: Non-aqueous capillary electrophoresis Enantioseparation Synergistic system Chiral ionic liquids β -cyclodextrin Amino acids

a b s t r a c t In this work, tetraalkylammonium amino acid ionic liquids (TAA-AAILs) were first applied to non-aqueous capillary electrophoresis (NACE) to establish synergistic systems with a conventional chiral selector, native β -cyclodextrin (β -CD). Excellent enantioseparations of some dansyl-amino acid (Dns-AA) samples were achieved. A series of comparison experiments and a molecular docking study were performed to validate the synergistic effect of TAA-AAILs and β -CD in NACE. Several interesting results were observed compared with previously reported chiral ILs-related aqueous CE studies. In particular, the direct enantioselectivity of TAA-AAILs was observed for the first time by using it as sole chiral selector in NACE. This was an encouraging finding because it was the first direct and convincing evidence that AAILs were able to participate in the enantiorecognition process in the conventional chiral selectors-based synergistic systems. The new TAA-AAILs synergistic NACE system was further optimized in terms of alkyl chain length, TAA-AAILs concentration, β -CD concentration, electrolyte composition and applied voltage, etc. Best enantioseparations of Dns-AAs were obtained when 100 mM β -CD and 10 mM tetramethylammonium-l-arginine (TMA-l-Arg) were added in an NMF buffer containing 50 mM Tris and 35 mM CA (apparent pH 7.85) under UV detection. The applied voltage was set at 30 kV. The method was then successfully employed for the determination of enantiomeric impurities of a real AA sample. This work proved that the use of chiral ILs as additives in NACE is a promising approach for enantioseparation. © 2019 Elsevier B.V. All rights reserved.

1. Introduction The enantiomers of a chiral compound always exhibit virtually identical physical and chemical properties. However, when it comes to a living system (e.g. a human body), the enantiomers may be distinctly different in biological activities. Therefore, chiral separation has always been an important research field in separation science [1,2]. For decades, chromatographic techniques, especially HPLC, have been one of the most effective analytical methods for enantioseparation [3–5]. Although various chiral columns have shown powerful enantioselectivities, they are usually very expensive and have relatively short lifespans. Besides, the large chiral reagents consumption and relatively low column efficiency of HPLC methods have also limited their applications. Capillary electrophoresis (CE) has emerged as a powerful alternative to conventional chromato∗

Corresponding author. E-mail address: [email protected] (Q. Zhang).

graphic techniques for chiral separation. The most attractive advantages of CE are its high separation efficiency, small sample consumption as well as its diverse separation modes such as capillary zone electrophoresis (CZE), micellar electrokinetic chromatography (MEKC), non-aqueous capillary electrophoresis (NACE), etc [6–8]. To obtain a successful CE chiral separation, a chiral selector (or dual selector systems [9–12]) is usually necessary to selectively recognize the enantiomers [13–15]. Take β -cyclodextrin (β CD) derivatives for example: they can form complexes with enantiomers in the running buffer via non-covalent interactions such as inclusion, hydrogen bonding, electrostatic and steric interactions. Enantioseparation occurs because of the different thermodynamic stability of the complexes between the selector and both enantiomers. Compared with β -CD derivatives, the native β -CD has been paid much less attention due to the lack of functional groups such as carboxymethyl, sulfobutyl ether or amino, etc. Another major defect of native β -CD is its poor solubility in aqueous solutions (approximately 18 mg/mL, 20 °C), which greatly restricts its application in aqueous CE [16–18].

https://doi.org/10.1016/j.chroma.2019.460519 0021-9673/© 2019 Elsevier B.V. All rights reserved.

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA 2

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

The principle of NACE was first introduced in 1984 by Wahlbroehl and Jorgenson [19]. It is particularly suitable for the cases when analytes or electrolytes are insoluble or partially soluble in water. In NACE, the ion-solvation effects and molecular recognition process can lead to different selectivity and electrophoretic behavior. In addition, higher electric field strengths can be used in certain non-aqueous solvents to obtain higher separation efficiency compared with that in aqueous media [20,21]. The use of β -CD in NACE for enantioseparation has been reported in several literatures [20,22]. However, the separations of the enantiomers were not very satisfactory in terms of resolutions (Rs) and peak shapes. In fact, the focus of most chiral NACE studies is the choice of various non-aqueous solvents and suitable electrolytes. Further modifications and improvements (e.g. to explore more versatile chiral additives/modifiers) are essential to extend the enantioseparation capability of NACE. Ionic liquids (ILs) are a group of organic salts with melting points below 100 °C or more often close to room temperature [23– 31]. They are excellent additives for CE because of their unique physical and chemical properties such as relatively high conductivity, exceptional chemical and thermal stabilities, as well as the considerable solubility in both organic and inorganic solvents. Besides, it is feasible to design and synthesize various task-specific ILs by altering their anion-cation combinations [32,33]. Chiral ILs, which have a chiral cation or anion, or both, have received much attention in chiral science. They have proven to be qualified additives in chiral CE for improving enantioseparation. So far, however, all of these studies were carried out in aqueous CE, and the investigations of the separation mechanisms of these multicomponent enantiorecognition systems (chiral selectors, chiral ILs, enantiomers, etc.) were not sufficient [34-39]. In this work, tetraalkylammonium amino acid ionic liquids (TAA-AAILs) were first applied to NACE to cooperate with native β CD for enantioseparation. We aim to investigate the performance of chiral ILs as additives in NACE, and to validate the synergistic effect between TAA-AAILs and β -CD in NACE by a series of comparison experiments and mechanism studies.

Fig. 1. Structures of TAA-AAILs used in this work.

2.2. CE experiments All NACE experiments were performed on a CL1030 CE system (Beijing Huayang Liming Instrumental Co., Beijing, China) with a UV detector and an HW-20 0 0 Chemstation. A 40 cm (32 cm effective length) × 50 μm id uncoated fused-silica capillary (Hebei Yongnian Ruifeng Chromatography Ltd., Hebei, China) was used throughout the experiment. Sample injections were performed by hydrodynamic mode with sampling height at 10 cm for 5 s. All NACE experiments were carried out using a voltage of 30 kV. The wavelength for detection was 254 nm for all Dns-AAs. The CE system was operated in the conventional mode with the anode at the injector end of the capillary. The new capillary was flushed (2 bar) with 1.0 M NaOH (30 min), followed by 0.1 M NaOH (10 min), water (10 min), NMF (10 min) respectively. Between consecutive injections, the capillary was rinsed (2 bar) with NMF and running buffer for 3 min each. The background electrolyte (BGE) was freshly prepared by dissolving appropriate amount of β -CD and/or other additives in NMF containing a certain concentration of Tris and CA. The apparent pH (pH∗ ) was measured by a combination electrode (E-201-C) connected to a digital pH meter, type PHS-3C (INESA Scientific Instrument Co., Ltd. Shanghai, China). The running buffers were filtered with 0.45 μm pore membrane filters and degassed by sonication prior to use. 2.3. Dansylation of AAs

2. Experimental 2.1. Chemicals Alanine (Ala), Cysteine (Cys), Histidine (His), Phenylalanine (Phe), Tryptophan (Trp) and Tyrosine (Tyr) were purchased from Dalian Meilun Biotechnology Co., LTD (Dalian, China). Native β CD (purity > 99%) was purchased from Zibo Qianhui Biotechnology (Shandong, China). Tetramethylammonium-l-arginine (TMAl-Arg, purity > 99%), Tetraethylammonium-l-arginine (TEA-lArg, purity>99%) and Tetrabutylammonium-l-arginine (TBA-l-Arg, purity>99%) were purchased from Shanghai Cheng Jie Chemical Co., Ltd. (Shanghai, China). The structures of the TAA-AAILs are shown in Fig. 1. Lithium carbonate (purity > 99.99%), dansyl chloride (purity>98%) and N-Methylformamide (NMF, purity > 99% m/m) were purchased from Aladdin industrial Co., Ltd. (Shanghai, China). Nylon filters (0.45 μm), methanol, ethanol and acetonitrile (ACN), all of HPLC grade, were purchased from Sinopharm Chemical Reagent (Nanjing, China). Thiourea, tetramethylammonium chloride (TMA-Cl), tris (hydroxymethyl) aminomethane (Tris), citric acid (CA) and disodium tetraborate decahydrate (Na2 B4 O7 ·H2 O) were of analytical grade from Nanjing Chemical Reagent Co., Ltd. (Nanjing, China). Ultrapure water was used throughout the experiments.

All AAs were dansylated according to the previous literature [40]. Briefly, an aliquot of 200 μL AAs solution (1.0 mg/mL in 40 mM lithium carbonate buffer) was mixed with 300 μL labeling solution of dansyl chloride (1.5 mg/mL in acetonitrile). The mixed solution was allowed to react at room temperature for 35 min, and then was terminated by adding 50 μL 2% (v/v) ethylamine. The obtained Dns-AAs solutions were kept at 4 °C before use. 2.4. Molecular docking simulations The 3D molecular model of β -CD was obtained from RCSB Protein Data Bank (PDB ID: 3CGT). The molecular models of all enantiomers and chiral ILs were constructed using ChemOffice Pro 2017 software. The 3D structures were optimized with MM2 force field calculation. A PM3 quantum mechanical method was used for the geometry optimization of all enantiomers and chiral selectors. The molecular docking simulations were performed with AutoDock 4.2.6 [41]. The Lamarckian genetic algorithm (LGA) was employed to determine the binding conformations of a flexible ligand (enantiomers) to a target receptor (chiral selectors). The grid ˚ with a grid-point spacmaps of dimensions (50 A˚ × 50 A˚ × 50 A) ing of 0.375 A˚ were generated using Auto-Grid module to cover the receptor molecule. Each atom type (called probe) of the ligand was placed at the grid points and its interaction energy with all

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

ARTICLE IN PRESS

JID: CHROMA

[m5G;September 11, 2019;19:1]

1.37/1.020 3.96/1.032 2.21/1.024

L −2.20/−2.32 −2.05/−2.34 −1.39/−1.55 1.91/1.033 4.30/1.071 2.16/1.035 1.31/1.020 3.20/1.046 1.63/1.021

Rs/α

13.7 8.11 5.64 6.68 13.2 10.2 Single β -CDs in aqueous systema Single β -CDs in NACEb β -CDs/TMA-l-Arg synergistic NACEc β -CDs + TMA-Cl in NACEd β -CDs + l-Arg in NACEe β -CDs + TMA-Cl + l-Arg in NACEf

Conditions: a 10 mM β -CDs in 50 mM NaH2 PO4 -100 mM borate aqueous buffer (pH 9.00), applied voltage, 20 kV. b 100 mM β -CDs in NMF containing 50 mM Tris-30 mM CA (pH∗ 7.83); applied voltage, 30 kV. c 100 mM β -CDs, 10 mM TMA-l-Arg in NMF containing 50 mM Tris-35 mM CA (pH∗ 7.85); applied voltage, 30 kV. d 100 mM β -CDs, 10 mM TMA-Cl in NMF containing 50 mM Tris-35 mM CA (pH∗ 7.90); applied voltage, 30 kV. e 100 mM β -CDs, 10 mM l-Arg in NMF containing 50 mM Tris-30 mM CA (pH∗ 7.89); applied voltage, 30 kV. f 100 mM β -CDs, 10 mM TMA-Cl, 10 mM l-Arg in NMF containing 50 mM Tris-33 mM CA (pH∗ 7.86); applied voltage, 30 kV. g 10− 5 cm2 V− 1 s− 1 . Other conditions as in Section 2. (NS: Not separated; L: Rs < 0.5). ∗ See typical electropherograms of the deformed peaks in the Supporting Information (Fig. S1).

1.84/1.022 4.60/1.047 2.36/1.026

NS −2.42/−2.54 −2.49/−2.59 −1.90/−2.02 NS 1.59/1.026 −2.44/−2.57 4.57/1.072 −2.20/−2.35 1.89/1.022 −1.95/−2.07 Deformed peaks∗ Deformed peaks∗ NS −2.95/−3.04 −2.74/−2.94 −1.98/−2.08

μep1/ μep2 g μep1/ μep2 g

Dns-Tyr

Rs/α

μep1/ μep2 g

Dns-Phe

Rs/α Dns-Trp

μep1/ μep2 g μeof g

3.2.1. β -CD in aqueous CE and NACE For comparison, the enantioseparation performance of single β CD aqueous system and single β -CD NACE system were summarized in Table 1. As can be seen, none of the Dns-AAs were baseline separated in the single β -CD aqueous system even though we have used the optimized conditions reported by previous literature [42]. This may be ascribed to the weak complex formation of the enantiomers with CDs in aqueous media without the assistance of other additives/modifiers. Another reason might be that the solubility of β -CD in aqueous solution is very low (0-18 mg/mL, 20 °C)

Chiral systems

3.2. Performance of different separation systems

Table 1 Separation performance of different chiral systems

A few papers have reported the application of chiral AAILs in aqueous CE to cooperate with conventional chiral selectors (e.g. cyclodextrins, polysaccharides, antibiotics, etc.) for enantioseparation [35,36]. Synergistic effects between AAILs and chiral selectors were considered as the main contribution for the improved separations. Various indirect comparison experiments were performed in different studies to prove the existence of the synergistic effects. However, most of these observations were not convincing enough to support the claim because, in most cases, the enantiomers could not be separated using only the AAILs in the running buffer. In this work, a TAA-AAIL, TMA-l-Arg, was selected to establish chiral synergistic system with β -CD in NACE. At first, the use of TMA-l-Arg as sole chiral selector was investigated in a nonaqueous buffer (NMF solution). Interestingly, an obvious separation tendency of the Dns-AAs enantiomers was observed with the presence of 10 mM TMA-l-Arg in the running buffer (see Fig. 2). This was the first time that TAA-AAILs showed direct enantioseparation capability in CE, which was an encouraging finding even though the Rs is far from satisfactory. It was also the first direct evidence that AAILs were able to participate in the enantiorecognition process in the conventional chiral selectors-based synergistic systems, rather than just indirectly influence the enantioseparation by modifying the EOF or the ionic strength, etc. We further determined the migration order of the Dns-AA enantiomers in the sole chiral ILs system, and the result showed out that the first eluting peak was D-enantiomer.

Rs/α

3.1. TAA-AAIL as sole chiral selector in NACE

NS −3.05/−3.15 −2.74/−2.93 −1.90/−2.06

Dns-His

3. Results and discussion

μep1/ μep2 g

Rs/α

Dns-Cys

Rs/α

The Rs and selectivity factor (α ) of Dns-AA enantiomers were calculated by Rs = 2(t2 − t1 )/(w1 + w2 ) and α = t2 /t1 , where t1 and t2 are the migration times of the two enantiomers, and w1 and w2 are the widths of their peaks at the baseline. The plate number (N) was calculated according to N = 5.54 (t/w1/2 )2 , where t is the migration time and w1/2 is the peak width at half of the peak height. The electroosmotic flow (EOF) and apparent mobilities (μapp ) were calculated by μeo f = (L × l )/(V × t0 ), and μapp = (L × l )/(V × tm ), where L, l, V, t0 and tm are the total capillary length, effective capillary length, applied voltage, migration time of thiourea (a neutral marker) and migration time of the enantiomers, respectively. Effective mobilities (μeff ) were calculated by μe f f = μapp − μeo f .

μep1/ μep2 g

Dns-Ala

2.5. Calculations

NS −2.33/−2.45 −2.55/−2.68 −2.13/−2.22

the atoms of the receptor was computed and assigned to the corresponding grid point. 100 LGA runs, each with 200 individuals in the population, were performed. Results differing by less than 1 A˚ in a positional root mean square deviation (rmsd) were clustered together. In each group, the lowest binding energy conformation with the highest percentage frequency was selected as the group representative.

3

1.29/1.019 6.78/1.089 1.47/1.031

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA 4

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

Fig. 2. Enantioseparations of Dns-Phe with (A) and without (B) TMA-l-Arg as sole chiral selector in NACE. Conditions: (A) 50 mM Tris-35 mM CA in NMF buffer containing 10 mM TMA-l-Arg in NMF (pH∗ 7.88); (B) 50 mM Tris-30 mM CA in NMF buffer without TMA-l-Arg in NMF (pH∗ 7.85). Applied voltage, 30 kV; other conditions as in Section 2. ∗ : Enantioseparation of Dns-Phe (D:L = 2:1) in condition (A).

[43]. According to Wren and Rowe, the optimum concentrations of chiral selectors are determined by the association constants of the host-guest complexes formed by enantiomers and chiral selectors. The greater the difference between the stability of the complexes, the better the enantioselectivity of the separation [44]. For DnsAAs, the range of β -CD concentration that can be studied in aqueous solution is narrow. Therefore, efficient separations are difficult to obtain in aqueous systems. The separation of Dns-AA enantiomers in single β -CD NACE system was then performed. Considering the solubility of β -CD, NMF was chosen as the non-aqueous solvent for this work because it can dissolve more than 700 mM of β -CD. Alcohol solvents such as methanol or ethanol were not considered since β -CD is insoluble in these solvents. As shown in Table 1, the separation of DnsAAs enantiomers was well improved with 100 mM (optimized concentration) β -CD in NMF. However, the Rs (mostly in the range of 1.0–2.0) were still not satisfactory for analytical purposes. In addition, the separation efficiency was relatively low (mostly in the range of 50 0 0 0-10 0 0 0 0) due to the peak tailing. 3.2.2. β -CD/TAA-ILs synergistic system in NACE The performance of β -CD/TAA-ILs synergistic NACE system was investigated by the combined use of TMA-l-Arg and β -CD with NMF as the non-aqueous media. The results showed that the separations of Dns-AAs enantiomers were significantly improved in terms of resolutions and peak shapes (see Fig. 3). Since the TMAl-Arg was proven to have direct enantioseparation capability towards Dns-AAs, and yield the same migration order as in the sing β -CD system, we speculate that the synergistic effect between TMA-l-Arg and β -CD was an essential contribution for the increased enantioselectivity. The separation efficiency was also increased (mostly in the range of 150,0 0 0–250,0 0 0) mainly due to the improved peak shapes, especially the suppressed peak tailings. This phenomenon was very similar to the aqueous system as reported in previous papers [45]. The TMA+ ions were able to adsorb to the capillary inner wall, and thus decreases the adsorption of separated enantiomers. A series of other comparison experiments were conducted including the combined use of “β -CD + l-Arg”, “β -CD + TMA-Cl”

and “β -CD + TMA-Cl + l-Arg” in NACE. Their performance was also summarized in Table 1. It was observed that all Dns-AAs peaks were deformed (see typical electropherograms in Fig. S1 in Supporting Information) once free l-Arg was added in the NACE systems (e.g. “β -CD + l-Arg” and “β -CD + TMA-Cl + l-Arg” systems). This phenomenon was hard to explain because in aqueous system, the addition of free AAs as additives was even able to improve enantioseparation by cooperating with CDs [46]. Further studies are need to demonstrate the mechanism of this result. In the “β -CD + TMA-Cl” system, the separation of Dns-AAs enantiomers was slightly improved in terms of resolutions, mainly due to the decreased EOF and suppressed peak tailing (A typical electropherogram can be seen in Fig. S1 in Supporting Information). This result was similar with previous literature using TMA-Cl as additive to improve enantioseparations in aqueous systems [36]. However, the improvements were still far less than that in the β CD/TMA-l-Arg synergistic NACE system. All these results indicate that the decreased EOF is not the main contribution for the improved separations. In other words, the use of chiral ILs as additives in NACE has significant superiority over conventional modification methods. 3.2.3. Study of the enantiorecognition mechanism by molecular docking A molecular docking study was performed to compare and analyze the enantioseparation mechanism of sing β -CD and β CD/TAA-ILs synergistic system. The binding free energy (࢞G) data were utilized to demonstrate binding affinity of the Dns-AA enantiomers to the chiral selectors (࢞G = −RTlnK). In this part, the TMA-l-Arg is assumed to exist in two states, ionic associated and dissociated, considering that previous studies have shown that ILs may be partially dissociated in the aqueous or non-aqueous solutions due to their hydrophilic or hydrophobic nature [47,48]. Table 2 shows the molecular docking results for all Dns-AAs in different separation systems. As can be seen, the addition of TMA-l-Arg to the single β -CD system (whether dissociated or not) significantly improved the values of |࢞G| for all studied drugs, suggesting the enhanced binding affinity between enantiomers and chiral selectors. In particular, the differ-

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

5

Fig. 3. Enantioseparations of all tested Dns-AAs in the β -CD/TMA-l-Arg synergistic NACE system. Conditions: 100 mM β -CD, 10 mM TMA-l-Arg in NMF containing 50 mM Tris-35 mM CA (pH∗ 7.85); applied voltage, 30 kV; other conditions as in Section 2.

ences of ࢞G between d- and l- enantiomers (|࢞࢞G|) were also enlarged in the β -CD/TMA-l-Arg synergistic system, and the data were well consistent with the experimental results of α (enantioselectivity) as listed in Table 1. In comparison, the absolute values of ࢞G or ࢞࢞G in the associated TMA-l-Arg/β -CD system were generally higher than those in the dissociated TMAl-Arg system, indicating that the existing state of chiral ILs in synergistic system might be an important factor influencing the enantiorecognition.

Generally, there are multiple interactions participating in a chiral recognition including hydrogen bonding interaction, electrostatic interaction, host-guest inclusion, dipole-dipole, π -π , etc. Although not all the interactions could be observed directly, the molecular docking simulation is still able to provide visualized evidence for the improved enantioseparation. A typical simulation image is shown in Fig. 4, in which we can see that the inclusion of Dns-His enantiomers in β -CD was well improved with the presence of TMA-l-Arg (whether dissociated (b) or not (c)) in the sep-

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

ARTICLE IN PRESS

JID: CHROMA 6

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx Table 2 Summary of molecular docking results with AutoDock. Dns-His

Mode A Mode B Mode C

Dns-Ala

Dns-Phe

࢞GL a

|࢞࢞G|a , b

࢞GD a

࢞GL a

|࢞࢞G|a , b

࢞GD a

࢞GL a

|࢞࢞G|a , b

࢞GD a

࢞GL a

|࢞࢞G|a , b

−9.92 −11.05 −18.42

−10.34 −12.43 −20.55

0.42 1.38 2.13

−10.13 −12.18 −19.09

−10.46 −13.19 −21.81

0.33 1.01 2.72

−9.92 −10.59 −19.26

−10.55 −12.43 −21.22

0.63 1.84 1.96

−11.97 −13.02 −18.79

−12.43 −14.73 −21.31

0.46 1.71 2.52

Dns-Trp

Mode A Mode B Mode C

Dns-Lys

࢞GD a

Dns-Tyr

࢞GD a

࢞GL a

|࢞࢞G|a , b

࢞GD a

࢞GL a

|࢞࢞G|a , b

−12.81 −13.35 −20.01

−13.14 −14.53 −22.10

0.33 1.18 2.09

−9.54 −9.80 −19.00

−10.00 −11.55 −21.64

0.46 1.75 2.64

Modes A: single β -CD combined with drug enantiomers respectively. Mode B: dissociated TMA-l-Arg and β -CD combined with drug enantiomers. Mode C: associated TMA-l-Arg and β -CD combined with drug enantiomers. a kJ/mol. b Absolute value of difference of ࢞G between the d- and l- amino acids.

aration environment. Visualized hydrogen bonding interactions (labeled by green lines) between enantiomers and β -CD were dramatically enhanced compared with single β -CD environment. All of the above results suggested that the chiral ILs participated in the enantiorecognition process, and exhibited synergistic effect with β -CD during chiral separations. Nevertheless, the study of the enantiorecognition mechanism of a multicomponent separation system has always been a challenge in the analytical chemistry community. Most existing methods (including the molecular docking method) are not able to completely present the chiral environment and the enantiorecognition process. Further studies are warranted to elucidate the precise mechanism of the chiral ILs synergistic separation systems. 3.3. Optimization of separation conditions 3.3.1. Effect of alkyl chain length of TAA-ILs The effect of alkyl chain length of TAA-ILs on enantioseparation was studied by using TMA-l-Arg, TEA-l-Arg and TBA-l-Arg, respectively, to cooperate with β -CD. The results showed that the migration time of enantiomers was inversely proportional to the alkyl chain length of TAA-ILs. This may be due to the fact that the smaller cations (TMA+ ) were more easily to be adsorbed to the capillary inner wall, so as to decrease the EOF. A typical electropherogram of the separation of Dns-AAs enantiomers was shown in Fig. 5, in which Dns-Trp was resolved in all three β -CD/TAAILs synergistic NACE systems. However, TMA-l-Arg seemed to be a better choice because it yielded the best Rs and α , while the migration times of enantiomers were still acceptable. Thus, the β CD/TMA-l-Arg synergistic NACE system was selected for the further optimization. 3.3.2. Effect of TAA-ILs and β -CD concentration In this work, the effect of TAA-ILs concentration was investigated over the range of 0–20 mM with β -CD concentration fixed at 100 mM. As expected, the presence of TMA-l-Arg in the running buffer dramatically enhanced the separations of Dns-AAs enantiomers (Fig. 6A). With the concentration of TMA-l-Arg increasing from 5 to 10 mM, the migration times of enantiomers moderately increased from about 10 min to 25 min. However, the enantiomer peaks (including EOF peak) suddenly vanished (not detectable in 100 min) when the concentration of TMA-l-Arg reached 20 mM. This was another interesting result observed in NACE systems compared with aqueous chiral ILs synergistic system, in which the migration times normally increased stably with the concentration of chiral ILs in most cases. We speculated that the adsorption of

TMA+ on the capillary inner wall may be more intense in the NACE system. The saturated adsorption was relatively easy to achieve and therefore dramatically suppressed the EOF. As a result, the enantiomers were not able to migrate to the cathode for detections. The optimum β -CD concentration was studied in the range of 50–150 mM while keeping the TMA-l-Arg concentration constant at 10 mM. Fig. 6B shows the effect of β -CD concentration on Rs of all Dns-AAs enantiomers. The best Rs values were obtained at 100 mM β -CD. At higher concentrations, the Rs had no significant change or even decreased for most analytes. This observation was mainly attributed to the saturated enantiorecognition reactions between enantiomers and chiral selectors. Based on the above results, 10 mM of TMA-l-Arg and 100 mM of β -CD were eventually selected for the synergistic NACE systems (Typical electropherograms of the enantioseparations under different chiral selector concentrations were given in Fig. S2 and Fig. S3 in Supporting Information). 3.3.3. Effect of electrolyte concentration (pH∗ ) The electrolyte concentration is another critical factor in NACE because it determines the buffer capacity and has a significant influence on EOF and the ionization degree of analytes [20]. In this work, the Tris concentration was fixed at 50 mM to guarantee the buffer capacity. The effect of electrolyte on enantioseparation was studied by varying the CA concentrations in the range of 15– 55 mM. pH∗ was used as an indicator for illustration. The result (Fig. 7) showed that a weak basic environment was favorable for the β -CD/TMA-l-Arg NACE system. The best separations of DnsAAs in the synergistic NACE system were obtained with 35 mM CA (pH∗ = 7.85). When the pH∗ was lower than 6.3 or higher than 9.8, all separations declined to a low level. (Typical electropherograms of the enantioseparations under different pH∗ were given in Fig. S4 in Supporting Information) 3.3.4. Choice of applied voltage In CE, the applied voltage is not always a critical parameter. However, it affects the separation efficiency and the electrophoretic velocities of analytes. Usually a high applied voltage is preferred to increase separation efficiency and shorten analysis time, on condition that the joule heating (indicated by the current value) is not too strong. Compared with aqueous CE, NACE usually yields much lower current value with the same electrolyte composition. In this work,

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

7

Fig. 4. Molecular docking simulation of the enantiorecognition of Dns-His in, (a) single β -CD system; (b) dissocated TMA-l-Arg/β -CD synergistic system and (c) associated TMA-l-Arg/β -CD synergistic system. The hydrogen bonding is indicated by green dotted line. O red, N blue, H white, C grey. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article).

the maximum applied voltage of 30 kV was chosen because the current was only 11 μA under the optimized conditions. The separation efficiency was excellent (e.g. the plate numbers of Dns-Cys peaks were more than 20 0,0 0 0) and the baseline was very stable. This can be considered as another advantage of the TAA-ILs synergistic NACE system because in aqueous systems, a relatively low applied voltage may be necessary to avoid high current, even though this will result in longer analysis time and a loss of separation efficiency.

3.4. Method application The new NACE method was validated in terms of limit of detection (LOD), limit of quantitation (LOQ), linearity, accuracy and precision. β -CD/TMA-l-Arg NACE system and Trp were selected as the model system and analyte, respectively. The peak identification was achieved by adding extra d-Trp in the racemic Trp before derivatization. The first eluting peak was identified as Dns-d-Trp, and the second eluting peak was Dns-l-Trp. The LOD and LOQ of

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA 8

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

Fig. 5. Effect of alkyl chain length of TAA-AAILs on the enantioseparation of Dns-Trp. Conditions: 50 mM Tris/CA in NMF buffer (pH∗ 7.85 ± 0.05), containing (A) 100 mM β -CD; (B) 100 mM β -CD, 10 mM TBA-l-Arg; (C) 100 mM β -CD, 10 mM TEA-l-Arg; (D) 100 mM β -CD, 10 mM TMA-l-Arg; applied voltage, 30 kV; other conditions as in Section 2.

Fig. 6. Effect of TMA-l-Arg (A) and β -CD (B) concentrations on enantioseparation. Conditions: 50 mM Tris/CA in NMF buffer (pH∗ 7.85 ± 0.05), containing (A) 5–15 mM TMA-l-Arg, 100 mM β -CD and (B) 10 mM TMA-l-Arg, 40–120 mM β -CD; applied voltage, 30 kV; other conditions as in Section 2.

l-Trp were determined based on the signal to noise ratio, 3:1 and 10:1. As a result, the LOD and LOQ for l-Trp were 2.2 μg/mL and 7.3 μg/mL, respectively. The calibration curve was linear over the range of 7.3–500 μg/mL (r2 = 0.9986). The intra-day repeatability of the migration times (t1 ) was 1.69% (RSD, n = 6). The inter-day repeatability of the migration times (t1 ) was 3.48% (RSD, n = 6). The diagram of the standard curve, as well as the electropherograms corresponding to run-to-run and day-to-day were given in

Supporting Information (Fig. S5 and Fig. S6). To study the accuracy of the NACE method, recovery experiments were carried out with the standard addition procedures at three l-Trp concentrations (40 μg/ml, 50 μg/ml, 60 μg/ml) in triplicate. The average recoveries of the impurities (l-Trp) were obtained in the range of 96.1–102.8% (RSD = 2.68%, n = 9). The β -CD/TMA-l-Arg NACE system was then applied to determine the l-Trp impurity in a spiked d-Trp sample solution (1%

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

9

Fig. 7. Effect of electrolyte concentrations (pH∗ ) on enantioseparation. Conditions: 10 mM TMA-l-Arg, 100 mM β -CD in NMF containing 50 mM Tris and 15–55 mM CA. Applied voltage, 30 kV; other conditions as in Section 2. Note: 15 mM CA (pH∗ = 9.81); 25 mM CA (pH∗ = 8.99); 35 mM CA (pH∗ = 7.85); 45 mM CA (pH∗ = 7.02); 55 mM CA (pH∗ = 6.30).

l-Trp in d-Trp) and a commercial d-Trp sample (l-Trp: not detectable). The results were summarized in Table S1 in the Supporting Information. The RSDs of the detected impurity concentrations (n = 6) were less than 1.21%, which demonstrated that this proposed NACE method was able to precisely determine the enantiomeric purity of real samples. 4. Conclusion In this paper, several TAA-AAILs were first applied to NACE to establish synergistic system with β -CD. Excellent enantioseparations of some Dns-AAs samples were achieved. The synergistic effect between TAA-AAILs and β -CD was validated by a series of comparison experiments and a molecular docking study. The direct enantioselectivity of TAA-AAILs was also observed for the first time by using TMA-l-Arg as sole chiral selector in NACE. The results indicate that TAA-AAIL is a superior additive for chiral NACE. The present system can be expanded to a variety of other conventional chiral selectors, e.g. other cyclodextrins or antibiotics which have poor water solubilities. The design and application of novel chiral ILs for NACE are also worth more attention. Declaration of Competing Interest The authors have declared no conflict of interest. Acknowledgements This work was supported by the Project of National Natural Science Foundation of China (No.: 81703465, 21707053), the Natural Science Foundation of Jiangsu Province (No.: BK20170533, BK20181445) and the Senior Talent Cultivation Program of Jiangsu University (No.: 16JGD055). Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.chroma.2019.460519.

References [1] A.M. Stalcup, Chiral separations, Annu. Rev. Anal. Chem. 3 (2010) 341–363, doi:10.1146/annurev.anchem.111808.073635. [2] T.J. Ward, K.D. Ward, Chiral Separations: a review of current topics and trends, Anal. Chem. 84 (2012) 626–635, doi:10.1021/ac202892w. [3] A. Cavazzini, N. Marchetti, R. Guzzinati, M. Pierini, A. Ciogli, D. Kotoni, I. D’Acquarica, C. Villani, F. Gasparrini, Enantioseparation by ultra-highperformance liquid chromatography, TrAC, Trends Anal. Chem. 63 (2014) 95– 103, doi:10.1016/j.trac.2014.06.026. [4] D.C. Patel, M.F. Wahab, D.W. Armstrong, Z.S. Breitbach, Advances in highthroughput and high-efficiency chiral liquid chromatographic separations, J. Chromatogr. A 1467 (2016) 2–18, doi:10.1016/j.chroma.2016.07.040. [5] X. Huang, D. Pei, J. Liu, D. Di, A review on chiral separation by counter-current chromatography: development, applications and future outlook, J. Chromatogr. A 1531 (2018) 1–12, doi:10.1016/j.chroma.2017.10.073. [6] R.L.C. Voeten, I.K. Ventouri, R. Haselberg, G.W. Somsen, Capillary electrophoresis: trends and recent advances, Anal. Chem. 90 (2018) 1464–1481, doi:10. 1021/acs.analchem.8b0 0 015. [7] R. Fang, G. Chen, L. Yi, Y. Shao, L. Zhang, Q. Cai, J. Xiao, Determination of eight triazine herbicide residues in cereal and vegetable by micellar electrokinetic capillary chromatography with on-line sweeping, Food Chem 145 (2014) 41– 48, doi:10.1016/j.foodchem.2013.08.028. [8] Y. Hong, X. Guo, G. Chen, J. Zhou, X. Zou, X. Liao, T. Hou, Determination of five macrolide antibiotic residues in milk by micellar electrokinetic capillary chromatography with field amplified sample stacking, J. Food Saf 38 (2018) e12382, doi:10.1111/jfs.12382. [9] L. Müllerová, P. Dubský, B. Gaš, Twenty years of development of dual and multi-selector models in capillary electrophoresis: a review, Electrophoresis 35 (2014) 2688–2700, doi:10.1002/elps.201400149. [10] L. Müllerová, P. Dubský, B. Gaš, Separation efficiency of dual-selector systems in capillary electrophoresis, J. Chromatogr. A 1330 (2014) 82–88, doi:10.1016/j. chroma.2014.01.006. [11] L. Müllerová, P. Dubský, B. Gaš, Generalized model of electromigration with 1: 1 (analyte: selector) complexation stoichiometry: part II. application to dual systems and experimental verification, J. Chromatogr. A 1384 (2015) 147–154, doi:10.1016/j.chroma.2015.01.055. [12] S. Krait, M. Heuermann, G.K. Scriba, Development of a capillary electrophoresis method for the determination of the chiral purity of dextromethorphan by a dual selector system using quality by design methodology, J. Sep. Sci. 41 (2018) 1405–1413, doi:10.1002/jssc.201701166. [13] I.J. Stavrou, M.C. Mavroudi, C.P. Kapnissi-Christodoulou, Chiral selectors in CE: recent developments and applications (2012-mid 2014), Electrophoresis 36 (2015) 101–123, doi:10.10 02/elps.20140 0310. [14] J. Stavrou Ioannis, A. Agathokleous Eliana, P. Kapnissi-Christodoulou Constantina, Chiral selectors in CE: Recent development and applications (mid2014 to mid-2016), Electrophoresis 38 (2016) 786–819, doi:10.1002/elps. 201600322.

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519

JID: CHROMA 10

ARTICLE IN PRESS

[m5G;September 11, 2019;19:1]

S. Ren, S. Xue and X. Sun et al. / Journal of Chromatography A xxx (xxxx) xxx

[15] R.B. Yu, J.P. Quirino, Chiral selectors in capillary electrophoresis: trends during 2017–2018, Molecules 24 (2019) 1135, doi:10.3390/molecules24061135. ˇ ˇ [16] P. Rezanka, K. Navrátilová, M. Rezanka, V. Král, D. Sýkora, Application of cyclodextrins in chiral capillary electrophoresis, Electrophoresis 35 (2014) 2701– 2721, doi:10.10 02/elps.20140 0145. [17] J.M. Saz, M.L. Marina, Recent advances on the use of cyclodextrins in the chiral analysis of drugs by capillary electrophoresis, J. Chromatogr. A 1467 (2016) 79– 94, doi:10.1016/j.chroma.2016.08.029. [18] Q. Zhu, G.K. Scriba, Advances in the use of cyclodextrins as chiral selectors in capillary electrokinetic chromatography: fundamentals and applications, Chromatographia 79 (2016) 1403–1435, doi:10.1007/s10337- 016- 3167- 0. [19] Y. Walbroehl, J.W. Jorgenson, On-column UV absorption detector for open tubular capillary zone electrophoresis, J. Chromatogr. A 315 (1984) 135–143, doi:10.1016/S0021-9673(01)90731-9. [20] I. Ali, M.M. Sanagi, H.Y. Aboul-Enein, Advances in chiral separations by nonaqueous capillary electrophoresis in pharmaceutical and biomedical analysis, Electrophoresis 35 (2014) 926–936, doi:10.10 02/elps.20130 0222. [21] M.-L. Riekkola, M. Jussila, S.P. Porras, I.E. Valkó, Non-aqueous capillary electrophoresis, J. Chromatogr. A 892 (20 0 0) 155–170, doi:10.1016/S0 021-9673(0 0) 00108-4. [22] F. Wang, M.G. Khaledi, Chiral separations by nonaqueous capillary electrophoresis, Anal. Chem. 68 (1996) 3460–3467, doi:10.1021/ac960537o. [23] T. Welton, Room-temperature ionic liquids. Solvents for synthesis and catalysis, Chem. Rev. 99 (1999) 2071–2084, doi:10.1021/cr980032t. [24] M.J. Trujillo-Rodríguez, H. Nan, M. Varona, M.N. Emaus, I.D. Souza, J.L. Anderson, Advances of ionic liquids in analytical chemistry, Anal. Chem. 91 (2019) 505–531, doi:10.1021/acs.analchem.8b04710. [25] H. Yang, L. Chen, C. Zhou, X. Yu, A.E.A. Yagoub, H. Ma, Improving the extraction of l-phenylalanine by the use of ionic liquids as adjuvants in aqueous biphasic systems, Food Chem 245 (2018) 346–352, doi:10.1016/j.foodchem.2017.10.110. [26] W. He, H. Wang, Z. Jing, D. Cui, J. Zhu, Z. Li, H. Ma, Highly efficient synthesis of hydrophilic phytosterol derivatives catalyzed by ionic liquid, J. Am. Oil Chem. Soc. 95 (2018) 89–100, doi:10.1002/aocs.12024. [27] W. He, Q. Liu, H. Yu, X. Si, J. Zhang, Efficient synthesis of octacosanol linoleate catalyzed by ionic liquid and its structure characterization, J. Am. Oil Chem. Soc. 93 (2016) 509–517, doi:10.1007/s11746- 016- 2793- x. [28] W. Zhu, P. Wu, L. Yang, Y. Chang, Y. Chao, H. Li, Y. Jiang, W. Jiang, S. Xun, Pyridinium-based temperature-responsive magnetic ionic liquid for oxidative desulfurization of fuels, Chem. Eng. J. 229 (2013) 250–256, doi:10.1016/j.cej. 2013.05.115. [29] S. Xun, W. Zhu, D. Zheng, L. Zhang, H. Liu, S. Yin, M. Zhang, H. Li, Synthesis of metal-based ionic liquid supported catalyst and its application in catalytic oxidative desulfurization of fuels, Fuel 136 (2014) 358–365, doi:10.1016/j.fuel. 2014.07.029. [30] W. Jiang, W. Zhu, H. Li, X. Wang, S. Yin, Y. Chang, H. Li, Temperature-responsive ionic liquid extraction and separation of the aromatic sulfur compounds, Fuel 140 (2015) 590–596, doi:10.1016/j.fuel.2014.09.083. [31] X. Jiaojiao, Z. Bin, Z. Gangbin, W. Ping, L. Zhenjiang, Quick separation and enzymatic performance improvement of lipase by ionic liquid-modified Fe3 O4 carrier immobilization, Bioprocess. Biosyst. Eng. 41 (2018) 739–748, doi:10.1007/ s00449- 018- 1907- 2. [32] P. Sun, D.W. Armstrong, Ionic liquids in analytical chemistry, Anal. Chim. Acta 661 (2010) 1–16, doi:10.1016/j.aca.20 09.12.0 07.

[33] T.D. Ho, C. Zhang, L.W. Hantao, J.L. Anderson, Ionic liquids in analytical chemistry: fundamentals, advances, and perspectives, Anal. Chem. 86 (2014) 262– 285, doi:10.1021/ac4035554. [34] C.P. Kapnissi-Christodoulou, I.J. Stavrou, M.C. Mavroudi, Chiral ionic liquids in chromatographic and electrophoretic separations, J. Chromatogr. A 1363 (2014) 2–10, doi:10.1016/j.chroma.2014.05.059. [35] M. Greño, M.L. Marina, M. Castro-Puyana, Enantioseparation by capillary electrophoresis using ionic liquids as chiral selectors, Crit. Rev. Anal. Chem. (2018) 1–18, doi:10.1080/10408347.2018.1439365. [36] Q. Zhang, Ionic liquids in capillary electrophoresis for enantioseparation, TrAC, Trends Anal. Chem. 100 (2018) 145–154, doi:10.1016/j.trac.2018.01.001. [37] Q. Zhang, J. Zhang, S. Xue, M. Rui, B. Gao, A. Li, J. Bai, Z. yin, E.M. Anochie, Enhanced enantioselectivity of native α -cyclodextrins by the synergy of chiral ionic liquids in capillary electrophoresis, J. Sep. Sci. 41 (2018) 4525–4532, doi:10.10 02/jssc.20180 0792. [38] M. Greño, M.L. Marina, M. Castro-Puyana, Effect of the combined use of γ cyclodextrin and a chiral ionic liquid on the enantiomeric separation of homocysteine by capillary electrophoresis, J. Chromatogr. A 1568 (2018) 222–228, doi:10.1016/j.chroma.2018.07.023. [39] Q. Zhang, X. Qi, C. Feng, S. Tong, M. Rui, Three chiral ionic liquids as additives for enantioseparation in capillary electrophoresis and their comparison with conventional modifiers, J. Chromatogr. A 1462 (2016) 146–152, doi:10.1016/j. chroma.2016.07.066. [40] L. Qi, Y. Chen, M. Xie, Z. Guo, X. Wang, Separation of dansylated amino acid enantiomers by chiral ligand-exchange CE with a zinc(II) l-arginine complex as the selecting system, Electrophoresis 29 (2008) 4277–4283, doi:10.1002/elps. 20 080 0 064. [41] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell, A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility, J. Comput. Chem. 30 (2009) 2785–2791, doi:10.1002/jcc. 21256. [42] M. Yoshinaga, M. Tanaka, Effect of urea addition on chiral separation of dansylamino acids by capillary zone electrophoresis with cyclodextrins, J. Chromatogr. A 710 (1995) 331–337, doi:10.1016/0 021-9673(95)0 0501-3. [43] I.E. Valkó, H. Sirén, M.-L. Riekkola, Chiral separation of dansyl-amino acids in a nonaqueous medium by capillary electrophoresis, J. Chromatogr. A 737 (1996) 263–272, doi:10.1016/0 021-9673(96)0 0 0 03-9. [44] S.A. Wren, R.C. Rowe, Theoretical aspects of chiral separation in capillary electrophoresis: I. initial evaluation of a model, J. Chromatogr. A 603 (1992) 235– 241, doi:10.1016/0021- 9673(92)85366- 2. [45] Q. Zhang, Y. Du, S. Du, J. Zhang, Z. Feng, Y. Zhang, X. Li, Tetramethylammonium-lactobionate: a novel ionic liquid chiral selector based on saccharides in capillary electrophoresis, Electrophoresis 36 (2015) 1216–1223, doi:10.10 02/elps.20140 0358. [46] S. Chalavi, A.R. Fakhari, S. Nojavan, P. Mirzaei, Evaluation of the synergistic effect with amino acids for enantioseparation of basic drugs using capillary electrophoresis, Electrophoresis 39 (2018) 2202–2209, doi:10.10 02/elps.20180 0128. [47] H. Wang, J. Wang, S. Zhang, Y. Pei, K. Zhuo, Ionic association of the ionic liquids [C4mim][BF4], [C4mim][PF6], and [Cnmim]Br in molecular solvents, ChemPhysChem 10 (2009) 2516–2523, doi:10.10 02/cphc.20 090 0438. [48] P. Yee, J.K. Shah, E.J. Maginn, State of hydrophobic and hydrophilic ionic liquids in aqueous solutions: are the ions fully dissociated? J. Phys. Chem. B 117 (2013) 12556–12566, doi:10.1021/jp405341m.

Please cite this article as: S. Ren, S. Xue and X. Sun et al., Investigation of the synergistic effect of chiral ionic liquids as additives in non-aqueous capillary electrophoresis for enantioseparation, Journal of Chromatography A, https://doi.org/10.1016/j.chroma.2019.460519