separation of enantiomers

Transcription

1 1050 Electrophoresis 2006, 27, Feng Qin Chuanhui Xie Shun Feng Junjie Ou Liang Kong Mingliang Ye Hanfa Zou National Chromatographic Research & Analysis Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, P. R. China Received August 25, 2005 Revised October 14, 2005 Accepted October 15, 2005 Research Article Monolithic silica capillary column with coated cellulose tris(3,5-dimethylphenylcarbamate) for capillary electrochromatographic separation of enantiomers Monolithic silica capillary columns were prepared by a sol-gel process in fused-silica capillaries with an inner diameter of 50 mm and were modified by coating of cellulose tris(3,5-dimethylphenylcarbamate). Influences of the factors in the modification process on enantiomer separations were investigated. The prepared columns were used to perform enantiomer separations by CEC. Fifteen and two pairs of enantiomers were separated under aqueous and nonaqueous mobile phases, respectively, and most of them were baseline-separated with very high column efficiencies. The Van Deemter curve was found flat under high linear velocity of the mobile phase, which indicated favorable kinetic properties of the prepared columns. Baseline separation of a pair of enantiomers was achieved in 90 s with high-column efficiency by short-end separation under high voltage. Keywords: Capillary electrochromatography / Cellulose tris(3,5-dimethylphenylcarbamate) / Enantiomer separation / Monolithic silica capillary column DOI /elps Introduction HPLC based on chiral stationary phases (CSPs) has proven to be one of the most useful methods for the separation of enantiomers in the past decades [1]. Many CSPs have been prepared and until now more than 100 CSPs have been introduced on the market [2]. Miniaturization of chiral separations has attracted more and more interest as it not only allows the low cost of expensive chiralpacking materials and high-purity organic solvents, but also makes the analysis safety to the environment. Several techniques such as capillary LC (CLC), CE, and CEC have been used in the miniaturized chiral separations. Among these techniques, CEC appears to be promising as it combines the high selectivity of HPLC with the high efficiency of CE [3, 4]. Correspondence: Professor Hanfa Zou, National Chromatographic Research & Analysis Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian , P. R. China hanfazou@dicp.ac.cn Fax: Abbreviations: CDMPC, cellulose tris(3,5-dimethylphenylcarbamate); CLC, capillary LC; CSP, chiral stationary phase; EL, effective length; TL, total length In chiral separations by CEC, open-tubular capillaries and packed capillaries with CSPs are normally used. In the former case, the inside wall of the capillaries was modified with appropriate chiral selectors including CD derivative [5], proteins [6, 7], and polysaccharide derivatives [8], etc. The chiral resolution ability was not very high because of the low content of chiral selectors. In the latter case, the CSPs were packed into the capillaries and were stabled by retaining frits. CSPs that were used in HPLC have been successfully transferred to CEC, such as CD and its derivatives [9 11], proteins [12], macrocyclic antibiotics [13], chiral polyacrylamides [14], and Pirkle-type CSPs [15]. However, it suffered from the bubble formation and repeatability problems caused by the frit. More recently, monolithic-type CSPs have been developed for enantioselective CEC. In one approach, in situ copolymerization of monomeric chiral selectors and methacrylic (or other) type of comonomers were used to prepare the CSPs, such as molecular imprinted polymer [16], allyl carbamoylated b-cd [17], vinylized chiral crown ether [18], and functionalized quinidine [19, 20]. In another approach, on-column modifications of the preprepared monolithic rods with chiral selectors were used, such as ligand exchange CSPs [21, 22], and macrocyclic antibiotics [23, 24]. The chiral resolution ability of monolithic

2 Electrophoresis 2006, 27, CE and CEC 1051 columns was comparable to their packed-type counterparts. In addition, the disadvantages caused by the frits in packed column could be avoided. Polysaccharide derivative CSPs have proven to be one of the most widely used CSPs in HPLC because of their broad applicability to various kinds of enantiomers [2]. Recently, those kinds of CSPs were also transferred to CEC. Packed-column CEC with coated or bonded cellulose tris(3,5-dimethylphenylcarbamate) (CDMPC) [14, 25 32], cellulose tris(4-methylbenzoate) [33, 34], cellulose tris(3,5-dichlorophenylcarbamate) [35 38], and amylose (3,5-dimethylphenylcarbamate) [27, 34, 39] were reported by several research groups. Surprisingly, no attention has been paid to the polysaccharide derivative-based monolithic CSPs for CEC separation, although Chankvetadze et al. [40 42] reported promising application of such kinds of columns in HPLC and CLC. In our opinion, several advantages can be found if the polysaccharide derivative-based monolithic columns were used under CEC: the powerful chiral recognition ability offered by the polysaccharide derivatives; the favorable dynamic properties offered by monolithic columns and the high-column efficiency offered by CEC. In this study, CDMPC-modified monolithic silica capillary columns were prepared and the enantiomer separations on these columns by CEC were performed to validate the above assumption. 2 Materials and methods 2.1 Chemicals and reagents Microcrystalline cellulose, thiourea, and ACN were obtained from Merck (Darmstadt, Germany). 3,5-Dimethylphenyl isocyanate and diethylamine were purchased from Sigma-Aldrich (Steinheim, Germany). Tetramethoxysilane (TMOS) was obtained from Chemical Factory of Wuhan University (Wuhan, P. R. China), and poly- (ethylene glycol) (PEG, MW = ) was from Aldrich (Milwaukee, WI, USA). Pyridine and THF were purchased from Tianjin Second Chemical Plant (Tianjin, P. R. China). Methanol (HPLC grade) was obtained from Yuwang Chemical Company (Shandong, P. R. China). Water was purified by a Milli-Q system (Millipore, Milford, MA, USA). Racemic benzoin, indapamide, praziquantel, Tröger s base, trans-stilbene oxide, alprenolol, pindolol, and propranolol were purchased from Sigma (St. Louis, MO, USA). Racemic flavanone was obtained from Acros (NJ, USA). Racemic drug candidate A, tetrahydropalmatine, ranolazine, and hydroxyzine were obtained from National Institute for the Control of Pharmaceutical and Biological Products (Beijing, P. R. China). Racemic 4,4 -dimethoxy-5,6,5,6 -dimethylenedioxy-biphenyl-2,2 -dicarboxylate (DDBD) was a gift from Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences (Lanzhou, P. R. China). Racemic 3- butyl-phthalide was obtained from CSPC Pharmaceutical Technology (Shijiazhuang, P. R. China). Chemical structures of the tested enantiomers are listed in Fig Preparation of CDMPC-modified monolithic silica capillary columns Fused-silica capillaries (50 mm ID6365 mm OD), which were purchased from Yongnian Optic Fiber Plant (Hebei, China), were used to prepare monolithic silica columns. The native silica monolith was prepared in the capillary column by a sol-gel process, which was reported elsewhere [43]. In situ activation of the monolithic silica was performed by a rehydroxylation process in order to maximize the number of silanol groups on the silica surface. A 20% v/v HCl solution was continuously pumped through the monolithic capillary column under constant pressure (6.9 MPa) for 6 h at 607C. The column was flushed to neutral with water and then dried overnight with nitrogen purging at 1207C. CDMPC was prepared as described previously [31]. The obtained CDMPC was dissolved in acetone, filtered through a 0.46 mm membrane, reprecipitated in methanol and dried in vacuum at 407C. Then different amounts of the dried CDMPC (30, 60, and 90 mg) were dissolved into 1.0 ml of acetone, respectively, and the solutions were ultrasonicated before use. The in situ modification of the monolithic silica capillary column was performed according to Chankvetadze et al. [42]. Briefly, the capillary column containing monolithic silica (35 cm) was connected to an empty HPLC column ( mm ID), and the opposite end of the HPLC column was connected to an Elite P230 pump (Elite Company, Dalian, China). The capillary column was flushed with acetone under 6.9 MPa for 20 min, and then dried with nitrogen purging at 407C for 30 min. After that, the HPLC column was filled with the prepared CDMPC solution, and a constant pressure of 6.9 MPa was applied to drive the CDMPC solution through the capillary column. The pressure was kept for 60 min after the capillary column was completely filled with CDMPC solution. Then the capillary column was disconnected from the HPLC column and dried at atmosphere pressure under room temperature. After complete evaporation of acetone, which took about 4 7 days due to the different concentration of CDMPC solutions, the capillary column was dried in vacuum at 407C for 2 h. On-line detection window was made at appropriate position of the capillary column by removing of the outside coating. The column was installed in the cassette and cut into appropriate length for further use.

3 1052 F. Qin et al. Electrophoresis 2006, 27, Figure 1. Structure of the tested enantiomers. 2.3 CEC CEC enantiomer separations were performed on two sets of instruments. (i) A P/ACE System MDQ (Beckman, Fullerton, CA, USA) equipped with a diode-array UV detector was used to perform the separations under aqueous mobile phases. (ii) A Hewlett-Packard 3D CE system equipped with a UV detector (Waldbronn, Germany) was used to perform the separations under nonaqueous mobile phases. The temperature of the columns was set at 257C. Detection wavelength was set at 214 nm. Samples were dissolved into the mobile phase (0.1, 0.5 mg/ ml). The samples and mobile phases were filtered and ultrasonicated before use. Electrokinetic (for aqueous

4 Electrophoresis 2006, 27, CE and CEC 1053 mobile phases) or pressurized (for nonaqueous mobile phases) injections were adopted. During the CEC separations, no pressure was added to the inlet and outlet vials, and no bubble formation was observed in the enantiomer separations. Long-end separations were adopted in the experiment except for other illustrations. 2.4 Calculations In this study, t 1 and t 2 are the retention times of the firstly and lately eluted enantiomers. R s is the resolution factor and can be expressed as R s ¼ 1:18 ðt 2 t 1 Þ w 1 þ w 2 where w 1 and w 2 are the half band widths of the firstly and lately eluted enantiomers. N 1 and N 2 (plates/m) are the column efficiencies of the firstly and lately eluted enantiomers, and can be expressed as: N 1 ¼ 5:54 EL t 2 1 ; N 2 ¼ 5:54 w 1 EL t 2 2 w 2 where EL is the effective length (m) of the capillary column. 3 Results and discussion 3.1 Characterization of the monolithic columns The characterization of the prepared native silica monolith was presented elsewhere [43]. The monolithic silica formed in the capillary column has the morphology of a continuous skeleton and large through-pores, and the monolithic rod was bonded to the inner wall of the capillary column. These characters lead to favorable dynamic properties of the monolithic silica capillary columns, and no frits are needed to retain the stationary phase. The pressure drop of the monolithic column increased little after modified with CDMPC (60 mg/ml), which suggested that the modification process had little effect on the dynamic property of the column. In CEC separations, EOF is the driving force to transport the mobile phase through the capillary columns. The generation of EOF is believed to be contributed mainly by the ionic groups on the surface of the stationary phase. In this study, the monolithic silica column was modified with neutral CDMPC; thus, only silanol groups contributed to the EOF. This could be proved by the direction of EOF, which always kept from anode to cathode under all the mobile phases used. Figure 2 presents the dependence of EOF on the ph of the phosphate buffers. It was Figure 2. Dependence of EOF on the ph of the phosphate buffers. Thiourea was adopted as t 0 marker. Mobile phases: ACN/phosphate buffer (2 mm) with different ph (40/60, v/v). Capillary column: 50 mm ID monolithic silica column modified with 60 mg/ml CDMPC, total length (TL) 30.2 cm, EL 20 cm. Applied voltage: 10 kv. observed that the EOF increased with the increment of the ph. The EOF generated under acidic mobile phases (ph, 6) seemed too low to elute the uncharged compounds; thus, neutral or basic mobile phases are preferred in the enantiomer separations. The effects of voltages on the currents were investigated by using a CDMPC-modified column (60 mg/ml) under the mobile phase of ACN/aqueous buffer (2 mm NaH 2 PO 4 10 mm NHEt 2, ph 9.60) (40/60 v/v). It was obtained that the currents increased linearly from 1.2 to 15 ma when the applied voltage increased from 2.5 to 25 kv, and the correlation coefficient (r) was This indicated the Joule heating did not seem to be a cause of concern. In addition, the EOF velocity linearly increased with the increase of the applied voltage (r = ); thus, fast separation of enantiomers could be obtained by applying high voltage. 3.2 Influences of the factors in modification process on enantiomer separations CDMPC-modified monolithic silica capillary columns were prepared by Chankvetadze et al. [42], and such columns were used for enantiomer separations under CLC, but the influences of the modification process on enantiomer separations were not investigated. Based on repeatability and separation ability goal, such investigations are necessary. Viscous solution was obtained when different amounts of CDMPC were dissolved in acetone; thus, during the in situ modification, the diffusion of the solution took a long time

5 1054 F. Qin et al. Electrophoresis 2006, 27, when spreading through the monolith. In addition, concentration gradient of the CDMPC solution along the axis of the capillary column may be formed because of the adsorption of CDMPC on the surface of the silica. If the driving pressure was released immediately after the monolithic column was filled with CDMPC solution, inhomogeneous film may be formed on the surface of the silica, which may lead to several disadvantages, such as low column-to-column reproducibility, bad peak shape and low column efficiency. This phenomenon was observed in our experiments. Thus, the driving pressure was kept for more 60 min after observing the path of CDMPC solution through the capillary column, and this did work in the improvement of column-to-column reproducibility (see Section 3.5) and peak shape. In the capillary columns prepared in this study, the neutral CDMPC was used as chiral selector to resolve the enantiomers and the monolithic silica was used as support as well as to generate EOF. Thus, the amount of CDMPC loaded onto the monolithic silica affects not only the chromatographic properties, such as the retention, the separation factor, the resolution, and the peak efficiency of the enantiomer, but also the EOF in the capillary column. In this study, different concentrations of CDMPC were used in the modification process, which consequently resulted in different loading of them on the monolithic silica. Figure 3 presents the electrochromatograms for the separation of indapamide and praziquantel on the prepared columns (30 and 60 mg/ml). As can be seen, the void times on the two columns were 5.13 and 6.22 min, respectively, while on the column prepared with 90 mg/ml of CDMPC, the void time was larger than 60 min (data not shown, this is the reason why no electrochromatograms were obtained on this column). That is, increasing the amount of CDMPC loaded onto the monolithic silica would result in decreased EOF, which is similar to the particle-type of CSPs [31, 39]. It could also be seen that the retention, separation factor and resolution of the enantiomers increased with the increasing loading of CDMPC, while the column efficiency decreased along this tendency. Obviously, from the viewpoints of shorter analysis time and higher column efficiency, lower loading of CDMPC was preferred, but this could only be used for the limited enantiomers which could be well resolved by CDMPC. To separate as many enantiomers as possible in appropriate time, column prepared with 60 mg/ ml of CDMPC was chosen in this study. 3.3 Separation of enantiomers under aqueous and nonaqueous mobile phases Polysaccharide derivative CSPs were widely used under the mobile phase of hydrocarbon/alcohol (normal phase mode), and they were also useful in combination with Figure 3. Electrochromatograms for separation of indapamide (a) and praziquantel (b) on different columns. Capillary columns: 50 mm ID monolithic silica column modified with 30 mg/ml (I) or 60 mg/ml (II) of CDMPC, TL 30.2 cm, EL 20 cm. Mobile phase: ACN/phosphate buffer (2 mm, ph 6.80) (40/60, v/v). Applied voltage: 10 kv. aqueous organic (RP mode) and purely organic (polar organic mode) mobile phases. Enantioseparations using particle type of such CSPs by CEC under the three mobile phase modes have been reported [25 39]. Here we report the enantiomer separations using CDMPC-modified monolithic silica capillary columns by CEC. Fifteen pairs of enantiomers were tested under aqueous and nonaqueous mobile phases, and the electrochromatographic data are listed in Table 1. The nine pairs of neutral enantiomers were separated under the neutral mobile phases of ACN/phosphate buffer (2 mm, ph 6.80). Seven pairs of them were baseline separated and high column efficiencies were obtained. As

6 Electrophoresis 2006, 27, CE and CEC 1055 Table 1. Electrochromatographic data for separation of the tested enantiomers under aqueous and nonaqueous mobile phases Racemates Mobile phase Applied voltage (kv) t 1 (min) t 2 (min) N 1 (plates/m) N 2 (plates/m) R s Benzoin a Butyl-phthalide a DDBD a Indapamide a Praziquantel a Tröger s base a d Drug candidate A b Flavanone b trans-stilbene oxide b Alprenolol c Hydroxyzine c Pindolol c Propranolol c Ranolazine c d Tetrahydropalmatine c Mobile phases: (a) ACN/phosphate buffer (2 mm, ph 6.80) (40/60, v/v), (b) ACN/phosphate buffer (2 mm, ph 6.80) (60/40, v/v), (c) ACN/aqueous buffer (2 mm NaH 2 PO 4 10 mm NHEt 2, ph 9.60) (40/60, v/v), (d) methanol/ammonium acetate (5 mm). Capillary column: 50 mm ID monolithic silica column modified with 60 mg/ml of CDMPC. For mobile phases a c, TL of capillary column, 30.2 cm, EL, 20 cm. For mobile phases d, TL, 32.5 cm; EL, 24 cm. Injections: 3, 8s610 kv for analytes under mobile phases a c; 5 bar68 s for analytes under mobile phase d. can be seen, the column efficiencies were larger than , and the largest value was up to plates per meter. Typical electrochromatograms for the resolution of drug candidate A and benzoin are shown in Figs. 4a and b. These column efficiencies were much larger than what had been obtained in HPLC, CLC, and even CEC, although the mobile phases used were not the same. The high column efficiencies may originate from the plug-like profile of the EOF in the capillary columns and the enhanced mass-transfer kinetics provided by the monolithic silica. Such a gain in column efficiency can significantly improve the separation of the enantiomers. For example, 3-butyl-phthalide was hard to be resolved by conventional HPLC method because the retention times of the two enantiomers were too close, but its baseline separation was achieved by CEC in this study. The six pairs of basic enantiomers were also separated under basic mobile phases of ACN/aqueous buffer (2 mm NaH 2 PO 4 10 mm NHEt 2, ph 9.60) (40/60 v/v). As can be seen in Table 1, five pairs of basic enantiomers were baseline separated. High column efficiencies were also obtained (, to, ). Typical electrochromatograms for the resolution of alprenolol and pindolol are shown in Figs. 4c and d. In our experiment, we tried to separate several acidic enantiomers under acidic mobile phases, but this was not realized because under such mobile phases, which are necessary to keep the analytes neutral, the EOF was too low to elute the analytes. Among the 15 pairs of the enantiomers tested, only Tröger s base and ranolazine were separated under nonaqueous mobile phase of methanol/ammonium acetate (5 mm). Compared to aqueous mobile phases, the EOF generated under nonaqueous mobile phase was much lower; thus, higher voltage (20 kv) was adopted. The electrochromatograms are shown in Figs. 4e and f. As is seen, the column efficiencies of the two enantiomers under nonaqueous mobile phase were higher than those obtained under aqueous mobile phases, which may be due to the lower viscosity of the mobile phase in the former case. 3.4 Effect of the linear velocity of the mobile phase on the column efficiency The Van Deemter plot of the firstly eluted enantiomer of benzoin was measured by varying the applied voltages from 2.5 to 27.5 kv in steps of 2.5 kv (the linear velocity of

7 1056 F. Qin et al. Electrophoresis 2006, 27, Figure 4. Electrochromatograms for separation of enantiomers under aqueous and nonaqueous mobile phases. Solutes: (a) drug candidate A, (b) benzoin, (c) alprenolol, (d) pindolol, (e) Tröger s base, (f) ranolazine. Experimental conditions were the same as described in Table 1.

8 Electrophoresis 2006, 27, CE and CEC 1057 Figure 5. Effect of the linear velocity of mobile phase on the plate height of the firstly eluted enantiomer of benzoin. Mobile phase: ACN/phosphate buffer (2 mm, ph 6.80) (40/60, v/v). Capillary column: 50 mm ID monolithic silica column modified with 60 mg/ml of CDMPC, TL 30.2 cm, EL 20 cm. Applied voltages varied from 2.5 to 27.5 kv in steps of 2.5 kv. the mobile phase linearly increased from 0.12 to 1.39 mm/ s), and the obtained H-u plot is presented in Fig. 5. It can be seen that a flat Van Deemter curve was obtained when the linear velocity of the mobile phase was larger than 0.53 mm/s, which was one of the typical behaviors of monolithic-type stationary phases. This indicated that the C parameter in the Van Deemter equation had little contribution to the plate height. Thus, the favorable dynamic properties of the monolithic silica did contribute to the enhanced mass transfer. Fast enantiomer separations are becoming more promising in recent years because the modern synthesis, such as combinatorial chemistry, can generate numbers of enantiomers in short time. Increasing the mobile phase flow rate is a direct way to shorten the analysis time, but this often leads to increased back-pressure and peak broadening in conventional particle packed HPLC. For the CEC enantiomer separations reported in this study, back-pressure was not a problem that should be concerned, even more, the favorable dynamic properties of the monolithic silica made it possible to get acceptable column efficiency under high linear flow rate. Thus, fast enantiomer separations can be realized by shortening the EL and applying high voltage. Figure 6 shows the electrochromatograms for the fast enantiomer separations of benzoin and drug candidate A. As is seen, the two pairs of enantiomers were separated in a very short time (160 and 90 s, respectively) with good peak shape. Figure 6. Electrochromatograms for fast-speed separation of benzoin (a) and drug candidate A (b). Mobile phase: (a) ACN/phosphate buffer (2 mm, ph 6.80) (40/60, v/v); (b) ACN/phosphate buffer (2 mm, ph 6.80) (60/40, v/ v). Capillary column: 50 mm ID monolithic silica column modified with 60 mg/ml CDMPC, TL 30.2 cm, EL 10.2 cm. Applied voltage: (a) 25 kv; (b) 30 kv. 3.5 Column-to-column reproducibility and stability Another column was prepared with the same process (60 mg/ml) as described in materials and methods to investigate the column-to-column reproducibility. Several enantiomers were chosen as test samples, and the electrochromatographic data are listed in Table 2. As is seen, good resolution and high column efficiencies were obtained on the newly prepared column, and the electrochromatographic data are similar with those obtained in Table 1, which suggested that the column-to-column reproducibility was satisfactory.

9 1058 F. Qin et al. Electrophoresis 2006, 27, Table 2. Electrochromatographic data for column-to-column reproducibility Racemates t 1 (min) t 2 (min) N 1 (plates/m) N 2 (plates/m) R s Benzoin Butyl-phthalide DDBD Tröger s base Mobile phase: ACN/phosphate buffer (2 mm, ph 6.80) (40/60, v/v). Capillary column: 50 mm ID monolithic silica column modified with 60 mg/ml of CDMPC, TL 30.2 cm, EL 20 cm. Applied voltage: 10 kv. No significant decline of the resolution factors and efficiencies were observed after the columns being used under neutral, and even basic, mobile phases (aqueous and nonaqueous) for 40 injections, which indicated good stability of the prepared columns. 4 Concluding remarks Preparation of CDMPC-coated monolithic silica capillary columns was reported. The integration of CDMPC with monolithic silica capillary column and CEC provided advantages of powerful chiral recognition ability, enhanced mass transfer kinetics, and high-column efficiency in enantiomer separations. In total 15 pairs of enantiomers were successfully resolved by CEC under aqueous and nonaqueous mobile phases. The columnto-column reproducibility was satisfactory, and the prepared columns were stable under the adopted mobile phases. The financial supports from the Natural Science Foundation of China (No ) and High-Tech Program Grant from CAS (KGCX2-SW ) are gratefully acknowledged. 5 References [1] Mainer, N., Franco, P., Lindner, W., J. Chromatogr. A 2001, 906, [2] Okamoto, Y., Yashima, E., Angew. Chem. Int. Ed. 1998, 37, [3] Lämmerhofer, M., Svec, F., Fréchet, J., Lindner, W., Trends Anal. Chem. 2000, 19, [4] Mangelings, D., Maftouh, M., Heyden, Y., J. Sep. Sci. 2005, 28, [5] Wang, Y., Zeng, Z., Guan, N., Cheng, J., Electrophoresis 2001, 22, [6] Liu, Z., Zou, H., Ni, J., Zhang, Y., Anal. Chim. Acta 1999, 378, [7] Liu, Z., Zou, H., Ye, M., Ni, J., Zhang, Y., Electrophoresis 1999, 20, [8] Francotte, E., Jung, M., Chromatographia 1996, 42, [9] von Brocke, A., Wistuba, D., Gfrorer, P., Stahl, M., Schurig, V., Bayer, E., Electrophoresis 2002, 23, [10] Wistuba, D., Cabrera, K., Schurig, V., Electrophoresis 2001, 22, [11] Gong, Y., Lee, H., Anal. Chem. 2003, 75, [12] Lloyd, D., Li, S., Ryan, P., J. Chromatogr. A 1995, 694, [13] Fanali, S., Catarcini, P., Presutti, C., Quaglia, M., Righetti, P., Electrophoresis 2003, 24, [14] Krause, K., Girod, M., Chankvetadze, B., Blaschke, G., J. Chromatogr. A 1999, 837, [15] Vickers, P., Smith, N., J. Sep. Sci. 2002, 25, [16] Schweitz, L., Andersson, L., Nilsson, S., Anal. Chem. 1997, 69, [17] Koide, T., Ueno, K., J. Chromatogr. A 2000, 893, [18] Koide, T., Ueno, K., J. Chromatogr. A 2001, 909, [19] Lämmerhofer, M., Peters, E., Yu, C., Svec, F., Fréchet, J., Lindner, W., Anal. Chem. 2000, 72, [20] Lämmerhofer, M., Svec, F., Fréchet, J., Lindner, W., Anal. Chem. 2000, 72, [21] Chen, Z., Hobo, T., Anal. Chem. 2001, 73, [22] Chen, Z., Hobo, T., Electrophoresis 2001, 22, [23] Kornyšova, O., Owens, P., Maruška, A., Electrophoresis 2001, 22, [24] Kornyšova, O., Jarmalavičienė, R., Maruška, A., Electrophoresis 2004, 25, [25] Chen, X., Jin, W., Qin, F., Liu, Y., et al., Electrophoresis 2003, 24, [26] Chen, X., Qin, F., Liu, Y., Kong, L., Zou, H., Electrophoresis 2004, 25, [27] Girod, M., Chankvetadze, B., Blaschke, G., Electrophoresis 2001, 22, [28] Mayer, S., Briand, X., Francotte, E., J. Chromatogr. A 2000, 875, [29] Otsuka, K., Mikami, C., Terabe, S., J. Chromatogr. A 2000, 887, [30] Kawamura, K., Otsuka, K., Terabe, S., J. Chromatogr. A 2001, 924, [31] Chankvetadze, L., Kartozia, I., Yamamoto, C., Chankvetadze, B., et al., Electrophoresis 2002, 23, [32] Qin, F., Liu, Y., Chen, X., Kong, L., Zou, H., Electrophoresis 2005, 26, [33] Chen, X., Zou, H., Ye, M., Zhang, Z., Electrophoresis 2002, 23,

10 Electrophoresis 2006, 27, CE and CEC 1059 [34] Meyring, M., Chankvetadze, B., Blaschke, G., J. Chromatogr. A 2000, 876, [35] Girod, M., Chankvetadze, B., Okamoto, Y., Blaschke, G., J. Sep. Sci. 2001, 24, [36] Chankvetadze, B., Kartozia, I., Breitkreutz, J., Okamoto, Y., Blaschke, G., Electrophoresis 2001, 22, [37] Chankvetadze, B., Kartozia, I., Okamoto, Y., Blaschke, G., J. Sep. Sci. 2001, 24, [38] Chankvetadze, B., Kartozia, I., Breitkreutz, J., Girod, M., Knobloch, M., Okamoto, Y., Blaschke, G., J. Sep. Sci. 2001, 24, [39] Girod, M., Chankvetadze, B., Blaschke, G., J. Chromatogr. A 2000, 887, [40] Chankvetadze, B., Yamamoto, C., Okamoto, Y., Chem. Lett. 2003, 32, [41] Chankvetadze, B., Ikai, T., Yamamoto, C., Okamoto, Y., J. Chromatogr. A 2004, 1042, [42] Chankvetadze, B., Yamamoto, C., Tanaka, N., Nakanishi, K., Okamoto, Y., J. Sep. Sci. 2004, 27, [43] Xie, C., Hu, J., Xiao, H., Su, X., Dong, J., Tian, R., He, Z., Zou, H., Electrophoresis 2005, 26,