Indirect chiral ligand exchange chromatography for enantioseparation: a modification of conventional techniques

Abstract

Indirect chiral ligand exchange chromatography (ICLEC) was facilitated by substituting chiral ligands from the bidentate Cu(II) complex via racemic analytes in reaction vial. In-vial ligand exchange does not require the potentially harmful Cu(II)-containing mobile phase for HPLC separation, as it facilitates better pH control for improved complexation of enantiomers with the chiral Cu(II)–ligand complex. Resulting Cu(II)–analyte–ligand complexes have higher stability constants and show better chromatographic resolution than conventional (CLEC). The new technique also eliminates background noise, a problem with conventional CLEC due to higher concentrations of Cu-containing selector due to the Cu-free mobile phase. The new technique was found successful in determining enantiomeric purity of drugs in pharmaceutical formulations. The technique can be regarded as a novel surrogate for conventional CLEC and has the potential for biological and pharmaceutical applications.

---

1. Introduction

The chiral ligand exchange model suggests that during chiral ligand exchange chromatography (CLEC) the complex undergoes specific interactions with the enantiomer (i.e., with the ligand donating a lone pair of electrons). The metal ion thus simultaneously coordinates the chiral selector and the enantiomer to be separated. The principles of CLEC have been explained in the literature,^1–11 and the technique has been found to be successful for direct enantiomeric resolution of amino acids, beta-blockers and hydroxy acids by TLC¹² and HPLC.^2,13–21

The metal ion used in CLEC, especially the Cu(II), causes high levels of background noise and decreases sensitivity of the UV-visible detector.²² The optimal pH for complexation varies according to substance class; for example it is 4.3 for resolution of amino acids and hydroxy acids,² and >7 for amino alcohols.^22,23 At pH > 5, Cu precipitation is also reported,² which results in choking of the chromatographic system and column. Applications of other metal ions such as Zn(II) have been reported for chiral analysis using capillary electrophoresis,^24,25 but are found of limited use in chromatography compared to Cu(II)²² ions, which are used extensively in CLEC. The advantage of excellent complex-forming ability inherited in Cu(II) ions results in higher stability constants of diastereomeric complexes. However, this advantage of Cu(II) ions is subdued because the higher concentration of Cu in the mobile phase causes deterioration on chromatographic columns.

In its present form CLEC has been found to be less sensitive for chiral trace analysis especially in complex biological matrices such as plasma. The literature shows that the stability of diastereomeric complexes formed in CLEC is higher than the stability of the diastereomeric adducts formed by other chiral selectors,²⁶ and more stable ligand exchange complexes are formed in solution compared to the stationary phase.²²

Considering the vast potential of CLEC in enantioseparation and the fact that more stable ligand exchange complexes are formed in solution compared to the stationary phase,^22,27 and with the aim of overcoming the abovementioned limitations of CLEC, we report a modification in the technique regarding formation of the ligand exchange complex. The chiral ligand exchange complexes of racemic analytes were formed before chromatographic separation in the sample preparation vial itself. This is in view of literature reports on ligand-exchange TLC separation of β-blockers using various amino acids as the chiral selector when the racemic mixture was treated with the chiral selector before chromatogram development.²⁸

The modified technique can be safely called indirect chiral ligand exchange chromatography (ICLEC) because the chiral ligand exchange diastereomeric complexes were formed in off-column chromatography. Moreover, the need for a copper-containing mobile phase was eliminated as the chiral diastereomeric complexes were formed off-column unlike conventional LEC where comparatively large amounts of a copper-containing mobile phase are used. The ICLEC technique was applied for separating enantiomers of selected β-blockers (atenolol, propranolol, metoprolol, bisoprolol), hydroxyl acids (mandelic acid and lactic acid), dihydroxyphenyl-L-alanine (DOPA) and carnitine using Cu(II) ion as the ligand-binding metal ion and various L-amino acids (viz. L-proline (L-Pro), L-hydroxy proline (L-HPro), L-phenylalanine (L-Phe), L-leucine (L-Leu) and N,N-dimethyl-L-phenylalanine (N,N–(CH₃)₂–L-Phe) and chirally pure amino acid amides such as L-proline amide (L-Pro-NH₂), L-phenylalanine amide (L-Phe-NH₂) and L-leucine amide (L-Leu-NH₂)) as the chiral selector. Representative structures of the resolved compounds are shown in Fig. 1.

Fig. 1 Structure of compounds used in the present study.

As a real-world sample application, the technique was successfully applied to chiral analysis of the pharmaceutical dosage form of atenolol. The developed technique (ICLEC) was successfully optimized with respect to nature and concentration of organic modifier, pH of mobile phase, complex formation pH, and effect of various Cu(II) counter anions such as nitrate, sulphate, bromide, chloride and acetate.

To the best of our knowledge, this is the first report of its kind where diastereomeric ligand exchange complexes were formed before chromatography for separation of enantiomers by reversed-phase HPLC.

2. Experimental

2.1 Apparatus

The HPLC system consisting of a 10 mL pump head 1000, manager 5000 degasser, photodiode array detection system 2600, manual injection valve and Eurochrom operating software was from Knauer (Berlin, Germany). Other equipment used included pH meter, Cyber scan 510 (Singapore); Polarimeter P-3002 (Kruss, Hamburg, Germany); Milli-Q system, Millipore (Bedford, MA, USA); Perkin Elmer 1600 FT-IR spectrometer (Boardman, OH, USA); and Shimadzu UV-1601 spectrophotometer.

2.2 Materials

(R,S)-Atenolol, (R)-atenolol, (S)-atenolol, (R,S)-propranolol, (R)-propranolol and (S)-propranolol, (R,S)-metoprolol, (R,S)-bisoprolol, D and L-DOPA, DL and L-mandelic acid, DL-lactic acid, L-lactic acid, DL-carnitine, L-carnitine, L-proline, L-hydroxyproline, L-leucine, L-phenylalanine, L-prolinamide, L-leucinamide, L-phenylalaninamide and N,N-dimethyl-L-phenylalanine were obtained from Sigma-Aldrich (St. Louis, MO, USA). All other analytical grade chemicals and HPLC solvents were from E. Merck (Mumbai, India). Double distilled water purified (18.2 MΩ cm³) with Milli-Q system of Millipore (Bedford, MA, USA) was used throughout the studies.

Preparation of standard stock solutions and formation of Cu(II)–L-amino acid complex. Solution of Cu(II) (10 mM) having various counter ions as nitrate, sulphate, acetate, chloride, and of chiral ligands (20 mM), such as L-Pro, L-HPro, L-Leu, L-Phe, L-Pro-NH₂, L-Leu-NH₂, L-Phe-NH₂ and N,N–(CH₃)₂–L-Phe, was prepared in methanol triethylammonium phosphate (TEAP) buffer, 20 : 80, v/v. For formation of Cu(II)–L-chiral ligand complex, the two solutions were mixed in 1 : 2 ratio with final pH of 9 for β-blockers and 4–4.7 for DOPA, carnitine and hydroxy acids.

Preparation of diastereomeric ligand exchanged complexes. To the 1000 μL of chiral ligand complex solution (containing 20 μmol of L-proline) was added 800 μL of racemic atenolol solution (9 μmol). The combined solution was vortex mixed for 5 min at ambient temperature and was refrigerated until further analysis. The chiral ligand exchange diastereomeric complexes were prepared in the same way for all the racemic and chirally pure enantiomers. To the 100 μL of diastereomeric sample was added 200 μL of mobile phase, which was injected for the HPLC analysis.

3. HPLC

HPLC was performed on a Lichrospher C₁₈ (250 × 4.6 mm I.D., 5 μm) column from Merck (Darmstadt, Germany). The mobile phase used was methanol-triethylammonium phosphate (TEAP) buffer. The buffer of pH 5 was used for β-blockers while pH 4.3 was used for hydroxy acids, DOPA and carnitine; different ratios of methanol were used in the mobile phase. The mobile phase was filtered by membrane filter (0.45 μM) and degassed under reduced pressure prior to use. The separation was carried out at a flow rate of 1 mL min⁻¹ with a run time of 45 min with detection at 230 nm on photodiode array detector.

3.1 Validation procedures for analytical separation

Method validation was done using diastereomeric complexes of atenolol prepared with Cu(II)–L-phenylalanine in accordance to ICH guidelines. The calibration curves were obtained by plotting peak areas against concentrations, and linear regression equations were used to determine slopes and correlation coefficients. Recovery studies were carried out by analyzing solutions of various known concentrations and mean recovered values (six replicate runs) were represented as percentages of calculated values. Interday (3 days) and intraday stability studies were carried out to find precision and RSD. Limit of detection (LOD) and limit of quantification (LOQ) were evaluated.

4. Results and discussion

4.1 Formation of in situ diastereomeric chiral ligand exchange complexes

Using atenolol for formation of diastereomeric complex, we optimized the ratio of Cu(II) nitrate to L-phenylalanine. It was found that the best diastereomeric yield in terms of HPLC peak area was achieved when the ratio of chiral ligand exchange complex and atenolol was 2 : 1. Various concentrations of Cu(II) to L-phenylalanine were tried in ratios of 1 : 1, 1 : 2, 1 : 3 and 1 : 4, keeping the concentration of Cu(II) ions fixed at 10 mM concentration and varying the chiral selector concentration from 1 to 40 mM. The concentration of Cu(II) was also varied from 1 mM to 10 mM, thus providing various concentrations of 1 mM Cu(II)/2 mM L-phenylalanine, 2 mM Cu(II)/4 mM L-phenylalanine), 3 mM Cu(II)/6 mM L-phenylalanine, 4 mM Cu(II)/8 mM L-phenylalanine and 5 mM Cu(II)/10 mM L-phenylalanine to 10 mM Cu(II)/20 mM L-phenylalanine. The percent peak area obtained from HPLC indicated no appreciable increase in the amount of the complex formed with increases in molar concentrations of both the phenylalanine and Cu(II) beyond 1 : 2. Increased peak broadening and distortion were obtained at higher concentrations and simultaneous decrease in relative peak areas of diastereomers. This can be attributed to a reversible equilibrium reaction at the 1 : 3 molar ratio, while at 1 : 1 ratio the peak area of resolved diastereomers was reduced by almost half. Thus, the best resolution was obtained at the Cu(II) concentration of 10 mM and L-phenylalanine concentration of 20 mM. The other racemic molecules showed similar behavior with chiral selectors under study. Similarly, diastereomeric ligand exchange complexes were prepared using various copper salts as nitrate, sulphate, chloride, bromide and acetate. The diastereomeric chiral-ligand exchange complexes were found to be stable for about 5 days under refrigeration.

The reaction vials were placed in an incubator for a period of 5 to 60 min with an interval of 10 min, and the temperature for each time period was varied in the range of 10–45 °C. The temperature of 20 °C and reaction time of 5 min were successful for completed diastereomer formation. For the reaction carried out at 20 °C for 5 min, the diastereomeric peak area did not decrease appreciably; however, with an increase in time beyond 5 min for the reaction carried out at 40–45 °C, there was an appreciable decrease in the peak area. The latter observation indicates that a reversible ligand exchange reaction occurs at higher temperatures with time.⁷ A similar behavior was observed for all diastereomeric complexes.

The stepwise increase in pH from 4 to 9 showed a linear increase in the peak area; pH 4.3 was found to be successful for complete diastereomerization of DOPA and carnitine, while pH 4.7 was successful for the two hydroxy acids.

4.2 HPLC analysis

Chromatographic data for enantiomeric resolution of β-blockers, hydroxy acids, DOPA and carnitine using different chiral ligands (i.e., L-proline, L-hydroxyproline, L-phenylalanine, L-phenylalanine amide, L-leucine, L-leucine amide and N,N-dimethyl-L-phenylalanine) involved in the ligand exchanged complex formation are given in Table 1. All analytes were well resolved with good enantioselectivity and high resolution was obtained. The representative chromatograms for diastereomeric ligand exchanged complex of atenolol with different chiral selector complexes are shown in Fig. 2. The elution order of all analytes was confirmed by elution of chirally pure enantiomer, which indicated that in the case of β-blockers, the elution order was (S)- before (R)-enantiomer, while in other analytes it was D before L.

Table 1 Chromatographic data for resolution of various compounds under study^a

Analyte	Atenolol		Propranolol		Metoprolol		Bisoprolol		DOPA		Mandelic acid		Lactic acid		Carnitine
Chiral ligand	Rs	α	Rs	α	Rs	α	Rs	α	Rs	α	Rs	α	Rs	α	Rs	α
a Note: for beta blockers, the mobile phase used was CH3OH–TEAP (10 mM, pH 5) in ratio of 15 : 85, v/v. For hydroxyl acids, DOPA, and carnitine, the mobile phase used was CH3OH–TEAP (10 mM, pH 4.3) in ratio of 10 : 90, v/v.
L-Hydroxyproline	11.93	1.35	12.17	1.28	12.12	1.29	11.89	1.26	10.39	2.03	10.93	1.32	8.43	1.64	9.58	1.27
L-Proline	11.66	1.41	11.78	1.28	11.87	1.33	11.53	1.31	9.93	1.94	10.67	1.32	8.12	1.57	9.41	1.34
L-Proline amide	11.78	1.41	11.89	1.27	12.03	1.28	11.71	1.26	10.17	1.88	10.83	1.29	8.33	1.46	9.59	1.29
L-Phenylalanine	12.93	1.31	12.81	1.32	12.72	1.30	12.69	1.27	12.73	1.50	13.61	1.14	9.17	1.38	10.12	1.37
L-Phenylalanine amide	14.89	1.23	13.32	1.29	13.12	1.21	12.76	1.21	13.12	1.44	14.23	1.13	9.47	1.35	10.72	1.31
N,N-Dimethyl-L-phenylalanine	14.88	1.26	13.46	1.25	13.34	1.26	12.81	1.25	13.27	1.49	14.22	1.10	10.09	1.26	11.17	1.34
L-Leucine	11.51	1.29	11.43	1.07	11.32	1.20	10.82	1.14	12.01	1.48	12.63	1.28	7.78	1.36	8.57	1.34
L-Leucine amide	11.58	1.19	11.73	1.10	11.76	1.16	10.97	1.14	12.13	1.50	12.78	1.30	7.93	1.39	8.91	1.35

---

Fig. 2 Section of chromatograms showing resolution of atenolol with various chiral selectors. (1–8) represents the diastereomers of atenolol prepared with chiral ligand exchange Cu(II) complex of: (1) L-prolinamide; (2) N,N-dimethyl-L-phenylalanine; (3) L-phenylalanine; (4) L-phenylalaninamide; (5) L-proline; (6) L-hydroxy proline; (7) L-leucine; (8) L-leucinamide.

Effect of various chromatographic elements on resolution. Methanol was found to be best organic modifier for this study. When acetonitrile was used as the organic modifier, retention time decreased but at the cost of resolution and enantioselectivity. Further broader peaks with poor baseline resolution were obtained with acetonitrile as the mobile phase. Methanol concentration of 10% and 15% was thus found to be optimum for resolution of DOPA, carnitine, hydroxy acids and β-blockers.

pH is crucial in chromatographic studies; pH 5 was found to be optimum for resolution of all the β-blockers, and pH 4.3 was found to be the most effective for resolution of hydroxy acids, DOPA and carnitine. In the case of atenolol and other β-blockers, with an increase in pH up to 7 there was a small increase in resolution but at the cost of higher retention time and low enantioselectivity. In literature reports on CLEC, the diastereomers are formed online, which get the desired pH in the mobile phase for diastereomerization, while in the present case the diastereomeric complexes are formed in situ prior to chromatography.

The mobile phase flow rate was varied from 0.2 mL min⁻¹ to 1.5 mL min⁻¹. Increasing the mobile phase flow rate from 0.5 to 1 mL min⁻¹ was found to exert a linear decline in the retention factor, but the decrease in resolution was not sizeable. As the flow rate increased up to 1.5, the retention factor decreased and resolution diminished in linear proportion with no resolution observed at the flow rate of 1.5 mL min⁻¹. At a flow rate of less than 0.5 mL min⁻¹, very high retention factors with overlapping peaks were observed without any baseline resolution. Thus, the flow rate of 1 mL min⁻¹ was found to be successful for resolution of all the analytes.

Effect of various Cu(II) counterions on resolution. Table 2 shows the difference in resolution and enantioselectivity for diastereomeric complexes of the analytes under study that were formed upon use of Cu(II)–L-Phe-NH₂ complexes for ligand exchange in the presence of various Cu counter negative ions. The effect of various counterions such as nitrate, sulphate, bromide, chloride and acetate was investigated based on Natalini et al.,⁷ since Cu(II) has been used as the metal ion for complex formation, and therefore diastereomeric complexes after ligand exchange. Results were found to be in agreement with Natalini et al.⁷ showing the highest resolution when nitrate was the counterion.

Table 2 R_s values obtained in presence of different counter negative Cu(II) ions

Sample	Counter negative anions of Cu(II)
	Nitrate	Sulphate	Bromide	Chloride	Acetate
Atenolol	14.89	12.32	10.21	9.37	4.53
Propranolol	13.32	11.61	9.73	9.41	4.17
Metoprolol	13.12	11.18	8.87	7.98	3.12
Bisoprolol	12.76	11.11	9.01	8.15	3.73
DOPA	13.12	10.23	8.93	8.03	4.01
Mandelic acid	14.23	11.33	10.11	9.26	4.38
Lactic acid	9.47	6.17	5.13	4.47	1.96
Carnitine	10.72	6.33	4.88	4.23	1.77

---

Natalini et al.⁷ separated the enantiomers of amino acids using the Cu(II) complex of S-(R)-trityl cysteine; the highest resolution was obtained with Cu and nitrate as the counterion. They suggested that the increased resolution was due to lower activation energy of the transition state for the ligand-exchange process in the presence of nitrate counterions, resulting in improved column efficiency by Cu(II) nitrate over Cu(II) acetate-based solution. High activation energy in other cases causes slow mass transfer kinetics and poor resolution.

The explanation regarding the kinetics of mass transfer also holds true in the present case for LEC formation in situ (prior to chromatography) as revealed by the decrease in HPLC peak areas for separation of the diastereomer pair obtained from the same Cu(II)–L-Phe-NH₂ complex in the presence of four different counterions. At the same time resolution also decreases (Table 2). However, this factor alone does not account for the increased resolution obtained in the presence of nitrate ions. The interaction of diastereomeric ligand exchange complexes (that were formed prior to chromatography) with the stationary phase also controls the resolution rather than their formation kinetics as applicable in the conventional approach (where the ligand exchange complex formation takes place during the chromatographic process).

In the present case then, it could be suggested that the negative counterions may be influencing the hydrophobic interactions of the diastereomers with the reversed phase column during the process of chromatographic separation. Thus these counterions may theoretically be arranged by decreasing resolution as follows: NO₃⁻ > SO₄²⁻ > Br⁻ > Cl⁻ > CH3COO⁻.

4.3 Separation mechanism

The separation mechanism for indirect chiral ligand exchange chromatography (ICLEC) differs from conventional ligand exchange chromatography as it does not involve the formation of diastereomeric complex on the stationary phase and resembles that of indirect diastereomer separation formed after precolumn chiral derivatization where the interaction of already-formed diastereomers with the stationary phase is responsible for the resolution. The separation mechanism of the various compounds reported in this manuscript was depicted based on structure of diastereomers shown in Fig. 3.

Fig. 3 Graphical abstract of cis and trans ligand exchange complex.

Separation mechanism for β-blockers. In the case of β-blockers, the different chiral selectors used for CLE complex formation were L-Phe, L-Phe-NH₂, L-Leu, L-Leu-NH₂, L-Pro, L-Pro-NH₂, L-hydroxy proline and N,N-dimethyl-L-phenylalanine. The highest retention times and resolution were obtained for the diastereomeric complexes formed with the amino acid amides (used as chiral selector) compared to their amino acid counterparts used as chiral selectors. Among the amides (chiral selectors), the largest retention factor was obtained for the diastereomeric complex formed with L-leucine amide. This behavior could be correlated in terms of hydrophobicities of amino acids and corresponding amides in accordance with hydrophobicity scale for amino acids proposed by Bull and Breese.²⁹ Presence of the hydroxy group significantly reduced the hydrophobicity of L-hydroxyproline–Cu(II)–analyte complexes, and thus lowered the retention factors obtained on the reversed phase column.

The highest resolution was obtained with diastereomers formed with Cu(II) N,N-dimethyl-L-phenylalanine followed by those formed with L-phenylalanine amide. This can be attributed to the requirement that the chiral ligand (for better resolution in ligand exchange chromatographic enantioseparation) should possess a larger group to produce the space exclude function as well as certain lipophilia to be retained by the reversed phase column. The lowest resolution was shown by diastereomers formed with Cu(II) complexes of L-leucine and its amino acid amide (L-leucine amide), which have more hydrophobic side chains. The resolution order of diastereomers formed with Cu(II) ligand exchange complexes of L-proline, its amide (L-proline amide) and L-hydroxyproline falls in between the resolution shown by diastereomers formed with Cu(II) ligand exchange complexes of L-Phe, L-Leu and their corresponding amides. L-proline, its amide and the L-hydroxyproline formed a rigid network around the central metal ion leading to a higher stability constant of diastereomeric complexes formed and thus increased resolution. Compared to diastereomeric complexes of Cu(II)–L-phenylalanine and its corresponding amide, those formed with Cu(II)–L-proline and L-proline amide have greater steric constraints and hence lower resolution than the former. Among the proline series (L-proline, L-proline amide and L-hydroxy prolinamide), higher resolution was obtained for the diastereomers formed with L-hydroxyproline due to additional enantioselectivity provided by the presence of the hydroxy group.

The elution order in ligand exchange chromatography depends to a larger extent on stability and hydrophobicities of diastereomeric ligand exchange complexes. The complexes in which the hydrophobic groups (side chains) are on the same side (cis) of the copper plane leading to stronger hydrophobic interactions with the reversed phase stationary phase showed late elution. The diastereomeric complexes with hydrophobic side chains on the opposite side of the copper plane (trans) had weaker hydrophobic interactions with the apolar stationary phase and showed earlier elution of the diastereomeric complexes. The same explanation holds good for the elution order of the enantiomers of β-blockers in the present studies, which was found to be (S)- before (R)-enantiomer; the structure for diastereomers of (R,S)-atenolol prepared with Cu(II)–(L-Phe)₂ (as chiral selector) could thus be as shown in Fig. 3.

In Fig. 3 the complex formed by (S)-atenolol, the hydrophobic side chain of L-Phe and atenolol are on the opposite side of the copper plane and thus have a trans relationship as compared to the diastereomeric complex formed by (R)-atenolol where the hydrophobic side chains are on the same side of the copper plane (cis). Thus, because the cis complex is more hydrophobic, it interacts strongly with the apolar stationary phase and thus is eluted later than the less hydrophobic trans complex formed with (S)-atenolol. The same explanation seems to be true for all the β-blockers studied.

Separation mechanism of hydroxy acids, DOPA and carnitine. The separation mechanism suggested for β-blockers also holds true for enantiomers of hydroxy acids, DOPA and carnitine, in terms of the resolution and retention factor. The resolution and retention factor for ligand-exchange diastereomeric complexes of mandelic acid and DOPA were found to be higher compared to diastereomeric complexes of lactic acid and carnitine. This could be attributed to the presence of larger and more lipophilic side chains in mandelic acid and DOPA, which provide additional stereoselective interactions for the diastereomeric ligand exchange complex formed. On the basis of mandelic-acid diastereomer structure with Cu(II)–(L-Phe)₂ (can be generalized from Fig. 3), the elution order could be explained on the basis of hydrophobicities of diastereomeric complexes.³⁰ The diastereomeric complexes of D-mandelic acid and D-DOPA may have hydrophobic groups on the opposite side (trans) to the hydrophobic side chain of L-Phe in the Cu(II) plane (thus showing less hydrophobicity outside the complex) and less interaction with the apolar stationary phase compared to cis complexes formed by L-mandelic acid and L-DOPA. The L-L-diastereomers of L-mandelic acid and L-DOPA formed with Cu(II)–L-Phe eluted later than their corresponding D-L diastereomer due to more hydrophobic interaction with the stationary phase. A similar explanation holds true for lactic acid and carnitine enantiomers.

4.4 Method validation

Method validation was done in accordance with ICH guidelines using diastereomeric complexes of atenolol prepared with Cu(II)–L-Phe. The accuracy and precision studies of the developed method were carried out by replicate HPLC analysis (n = 6) of (RS)-atenolol with Cu(II)–L-Phe complexes at six different concentration levels (20, 25, 30, 35, 40, and 45 ng mL⁻¹), and the RSD were less than 2% in almost all cases. The relative standard deviation for (R)- and (S)-atenolol varied from 0.84 to 1.08% and 0.81 to 0.1.05% for intraday assay precision and 1.00 to 1.93% and 1.08 to 1.47% for interday assay precision. The recovery for (R)- and (S)- atenolol varied from 97.6 to 99.2% and 97.8 to 108.3% for the intraday assay and 95.9 to 98.2% and 95.14 to 111.5% for the interday assay, respectively.

The LOD and LOQ of the method were found to be 5 ng mL⁻¹ and 17 ng mL⁻¹ respectively.

4.5 Pharmaceutical sample applications

Determination of enantiomeric composition in pharmaceutical dosage forms. The developed ligand exchange HPLC method was successfully applied for determination of enantiomeric composition of pharmaceutical dosage forms with very good sensitivity. Ten Pertenol tablets, each containing 25 mg of (R,S)-atenolol, were ground to a fine powder and extracted with 100 mL methanol for 15 min using a sonicator at 25 °C. The solution was filtered through Whatman paper and the residue was further treated twice with methanol and filtered. The combined filtrate was concentrated in vacuum and left to cool until crystals appeared. The mother liquor was decanted and the crystals were recrystallized from MeOH–H₂O. The crystals were washed with diethyl ether and dried in air. The same procedure was repeated for extraction, isolation and purification of remaining β-blockers from their tablets. The recoveries were 96–98% of the quantities reported on the commercial labels. The purified compounds were used as standards and their stock solutions (10 mM) were prepared in methanol–TEAP buffer (pH 9). It was filtered through a 0.45 μm filter and used for subsequent analysis using diastereomeric ligand exchange complex formation with Cu(II)–L-phenylalanine. The drug formulations were found to be racemic according to HPLC peak areas of the diastereomers.

5. Conclusion

The developed ICLEC technique showed significant improvement over conventional CLEC in that it reduced chiral selector concentration and quantity, thereby overcoming the detection and sensitivity problems that plague CLEC due to high levels of background noise. A broader range of pH control for diastereomeric ligand exchange complex formation thus minimizes Cu(II) precipitation on the column, making the ICLEC technique suitable for resolution of compounds such as amino alcohols at a relatively lower mobile phase pH. ICLEC may thus be employed with greater ease and sensitivity for chiral bioanalysis and industrial applications with minimal use of a chiral selector.

Acknowledgements

Donation of Knauer HPLC equipment (to RB) by the Alexander von Humboldt-Stiftung, Bonn, Germany, is gratefully acknowledged.

Notes and references

1. V. A. Davankov and S. V. Rogozhin, J. Chromatogr., 1971, 60, 280–283.
2. M. G. Schmid, O. Lecnik and G. Gubitz, J. Chromatogr. A, 2000, 875, 307–314.
3. N. Sanaie and C. A. Haynes, J. Chromatogr. A, 2006, 1104, 164–172.
4. N. Sanaie and C. A. Haynes, J. Chromatogr. A, 2006, 1132, 39–50.
5. B. Natalini, N. Giacche, R. Sardella, F. Ianni, A. Macchiarulo and R. Pellicciari, J. Chromatogr. A, 2010, 1217, 7523–7527.
6. B. Natalini, A. Macchiarulo, R. Sardella, A. Massarotti and R. Pellicciari, J. Sep. Sci., 2008, 31, 2395–2403.
7. B. Natalini, R. Sardella, G. Carbone, A. Macchiarulo and R. Pellicciari, Anal. Chim. Acta, 2009, 638, 225–233.
8. B. Natalini, R. Sardella, N. Giacche, S. Palmiotto, E. Camaioni, M. Marinozzi, A. Macchiarulo and R. Pellicciari, Anal. Bioanal. Chem., 2010, 397, 1997–2011.
9. B. Natalini, R. Sardella, A. Macchiarulo and R. Pellicciari, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci., 2008, 875, 108–117.
10. B. Natalini, R. Sardella, A. Macchiarulo and R. Pellicciari, J. Sep. Sci., 2008, 31, 696–704.
11. R. Sardella, F. Ianni, B. Natalini, G. P. Blanch and M. L. R. del Castillo, Curr. Anal. Chem., 2012, 8, 319–327.
12. S. Yamazaki, T. Takeuchi and T. Tanimura, J. Chromatogr., 1991, 540, 169–175.
13. S. Yamazaki, K. Saito and T. Tanimura, J. Liq. Chromatogr., 1994, 17, 2559–2567.
14. G. Bazylak, J. Chromatogr. A, 1994, 665, 75–86.
15. H. Katoh, T. Ishida, S. Kuwata and H. Kiniwa, Chromatographia, 1989, 28, 481–486.
16. S. Kodama, A. Yamamoto, A. Matsunaga and K. Hayakawa, J. Chromatogr. A, 2001, 932, 139–143.
17. N. Nikolic, D. Veselinovic, J. Vucina, H. Lingeman and K. Karljikovic-Rajic, J. Pharm. Biomed. Anal., 2003, 32, 1159–1166.
18. J. Koidl, H. Hodl, M. G. Schmid, B. Neubauer, M. Konrad, S. Petschauer and G. Gubitz, J. Biochem. Biophys. Methods, 2008, 70, 1254–1260.
19. S. Y. Sun, Y. Z. Jia, N. Zeng and F. M. Li, Chromatographia, 2006, 63, 331–335.
20. S. Sotgia, A. Zinellu, E. Pisanu, G. A. Pinna, L. Deiana and C. Carru, J. Chromatogr. A, 2008, 1205, 90–93.
21. H. Hodl, A. Krainer, K. Holzmuller, J. Koidl, M. G. Schmid and G. Gubitz, Electrophoresis, 2007, 28, 2675–2682.
22. R. Marchelli, R. Corradini, T. Bertuzzi, G. Galaverna, A. Dossena, F. Gasparrini, B. Galli, C. Villani and D. Misiti, Chirality, 1996, 8, 452–461.
23. M. G. Schmid, M. Laffranchini, D. Dreveny and G. Gubitz, Electrophoresis, 1999, 20, 2458–2461.
24. H. Zhang, L. Qi, L. Mao and Y. Chen, J. Sep. Sci., 2012, 35, 1236–1248.
25. B. B. Sun, L. Qi, X. Y. Mu, J. Qiao and M. L. Wang, Talanta, 2013, 116, 1121–1125.
26. V. A. Davankov, J. Chromatogr. A, 1994, 666, 55–76.
27. B. Galli, F. Gasparrini, D. Misiti, C. Villani, R. Corradini, A. Dossena and R. Marchelli, J. Chromatogr. A, 1994, 666, 77–89.
28. R. Bhushan and S. Tanwar, J. Chromatogr. A, 2010, 1217, 1395–1398.
29. H. B. Bull and K. Breese, Arch. Biochem. Biophys., 1974, 161, 665–670.
30. P. M. P. Nazareth and O. A. C. Antunes, J. Braz. Chem. Soc., 2002, 13, 658–663.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra07061e

---