Louisiana State University LSU Digital Commons

LSU Doctoral Dissertations Graduate School

2004 Analytical separations using packed and open- tubular capillary electrochromatography Constantina P. Kapnissi Louisiana State University and Agricultural and Mechanical College, [email protected]

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_dissertations Part of the Chemistry Commons

Recommended Citation Kapnissi, Constantina P., "Analytical separations using packed and open-tubular capillary electrochromatography" (2004). LSU Doctoral Dissertations. 595. https://digitalcommons.lsu.edu/gradschool_dissertations/595

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Doctoral Dissertations by an authorized graduate school editor of LSU Digital Commons. For more information, please [email protected].

ANALYTICAL SEPARATIONS USING PACKED AND OPEN- TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY

A Dissertation

Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy

in

The Department of Chemistry

by Constantina P. Kapnissi B.S., University of Cyprus, 1999 August 2004

Copyright 2004 Constantina Panayioti Kapnissi All rights reserved

ii

DEDICATION

I would like to dedicate this work to my husband Andreas Christodoulou, my parents

Panayiotis and Eleni Kapnissi, and my sisters Erasmia, Panayiota and Stella Kapnissi. I want to thank all of you for helping me, in your own way, to finally make one of my dreams come true. Thank you for encouraging me to continue and achieve my goals.

Thank you for your endless love, support, and motivation. Andreas, thank you for your continuous patience and for being there for me whenever I needed you during this difficult time. I will never forget what you did for me.

iii

ACKNOWLEDGMENTS

I would like to acknowledge the following people for their contributions in this work:

Dr. Isiah M. Warner for his guidance and excellent advice. Thank you for encouraging me to finish what I started. I want to also thank you for the financial and moral support.

You are an excellent mentor, and you prove that every day with your constant encouragement and support. Thank you again for everything.

Dr. Rezik A. Agbaria for his helpful scientific discussions during my general exam and my project with laser scanning confocal microscopy.

Dr. Charles William Henry III for training me how to use the capillary electrophoresis instrument and the packing apparatus. Dr. Henry and Dr. Shahab Shamsi helped me understand the most important concepts of my research.

Dr. Joseph B. Schlenoff and his group at Florida State University in Tallahassee,

Florida for training me on the polyelectrolyte multilayer coating procedure.

Dr. Lei Geng and Mark Lowry at University of Iowa in Iowa City, Iowa for helping me conduct some experiments on the laser scanning confocal microscope, and for their helpful discussions about this part of my research.

Dr. Akbay Cevdet and Bertha C. Valle for providing the polymeric surfactants I used in my research work.

Warner research group and Warner family for their encouragement and their total support.

Bertha C. Valle for her friendship, her advice, and the long useful discussions we had about research.

iv

Dr. Nikki Gill, Dr. Xiaofeng Zhu, and Dr. Abdul Numan for editing a few of my reports.

My family and friends in Cyprus for believing in me, and supporting all my choices.

v

TABLE OF CONTENTS

DEDICATION ...... iii

ACKNOWLEDGMENTS ...... iv

LIST OF TABLES ...... ix

LIST OF FIGURES ...... x

LIST OF ABBREVIATIONS ...... xiv

ABSTRACT ...... xviii

CHAPTER 1. INTRODUCTION ...... 1 1.1 Capillary Electrophoresis ...... 1 1.2 Micellar Electrokinetic Chromatography ...... 9 1.3 Capillary Electrochromatography ...... 15 1.3.1 Packed-Capillary Electrochromatography ...... 17 1.3.2 Open-Tubular Capillary Electrochromatography ...... 18 1.3.2.1 Adsorption ...... 20 1.3.2.2 Covalent Bonding and/or Crosslinking ...... 29 1.3.2.3 Porous Layers ...... 31 1.3.2.4 Chemical Bonding After Etching ...... 32 1.3.2.5 Sol-Gel Technique ...... 35 1.4 Chirality ...... 38 1.5 Chiral Selectors in Open-Tubular Capillary Electrochromatography ...... 40 1.5.1 Derivatized Cyclodextrins ...... 40 1.5.2 Cellulose ...... 42 1.5.3 Proteins ...... 43 1.5.4 Molecular Imprinted Polymers ...... 44 1.5.5 Polymeric Surfactants ...... 46 1.6 Laser Scanning Confocal Microscopy ...... 47 1.7 Scope of Dissertation ...... 51 1.8 References ...... 54

CHAPTER 2. SEPARATION OF BENZODIAZEPINES BY USE OF CAPILLARY ELECTROCHROMATOGRAPHY ...... 65 2.1 Introduction ...... 65 2.2 Experimental ...... 68 2.2.1 Reagents and Chemicals ...... 68 2.2.2 Instrumentation and Conditions ...... 68 2.2.3 Sample and Buffer Preparation ...... 69 2.2.4 Preparation and Conditioning of Packed Capillary Columns ...... 69 2.3 Results and Discussion ...... 70

vi

2.3.1 Effect of the Nature of Organic Modifier ...... 70 2.3.2 Effect of Mobile-Phase Composition ...... 73 2.3.3 Effect of Applied Voltage ...... 73 2.3.4 Effect of Tris Concentration ...... 75 2.3.5 Effect of Column Temperature ...... 76 2.3.6 CEC Separation of Drugs from a Urine Sample ...... 78 2.4 Conclusion ...... 81 2.5 References ...... 81

CHAPTER 3. ANALYTICAL SEPARATIONS USING MOLECULAR MICELLES IN OPEN-TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY ...... 85 3.1 Introduction ...... 85 3.2 Experimental ...... 87 3.2.1 Apparatus and Conditions ...... 87 3.2.2 Reagents and Chemicals ...... 87 3.2.3 Sample and Buffer Preparation ...... 88 3.2.4 Synthesis of Monomeric and Polymeric Surfactant ...... 88 3.2.5 Procedure for Polyelectrolyte Multilayer (PEM) Coating ...... 88 3.3 Results and Discussion ...... 89 3.3.1 Endurance of PEM Coating ...... 89 3.3.2 Stability of PEM Coating ...... 91 3.3.3 Reproducibilities ...... 92 3.3.4 Voltage Study ...... 92 3.3.5 Temperature Study ...... 92 3.3.6 Comparison Between Monomeric and Polymeric Surfactants ...... 94 3.4 Conclusion ...... 95 3.5 References ...... 98

CHAPTER 4. CHIRAL SEPARATIONS USING POLYMERIC SURFACTANTS AND POLYELECTROLYTE MULTILAYERS IN OPEN-TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY ...... 101 4.1 Introduction ...... 101 4.2 Experimental ...... 103 4.2.1 Apparatus and Conditions ...... 103 4.2.2 Reagents and Chemicals ...... 104 4.2.3 Sample and Background Electrolyte Preparation ...... 104 4.2.4 Synthesis of Polymeric Surfactant ...... 105 4.2.5 Procedure for Polyelectrolyte Multilayer Coating ...... 106 4.3 Results and Discussion ...... 106 4.3.1 Effect of Additives in Polymer Deposition Solutions ...... 107 4.3.2 Effect of NaCl Concentration ...... 107 4.3.3 Effect of Column Temperature ...... 108 4.3.4 Effect of Bilayer Number ...... 108

vii

4.3.5 Reproducibilities ...... 113 4.3.6 Chiral Separation of Analytes ...... 113 4.4 Conclusion ...... 118 4.5 References ...... 118

CHAPTER 5. INVESTIGATION OF THE STABILITY OF POLYELECTROLYTE MULTILAYER COATINGS IN OPEN-TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY USING LASER SCANNING CONFOCAL MICROSCOPY ...... 121 5.1 Introduction ...... 121 5.2 Experimental ...... 123 5.2.1 Apparatus and Conditions ...... 123 5.2.2 Reagents and Chemicals ...... 124 5.2.3 Sample and Background Electrolyte Preparation ...... 124 5.2.4 Synthesis of Polymeric Surfactant ...... 125 5.2.5 Procedure for Polyelectrolyte Multilayer Coating ...... 125 5.3 Results and Discussion ...... 126 5.3.1 Open-Tubular Capillary Electrochromatography ...... 126 5.3.1.1 Coating Stability ...... 126 5.3.1.2 Capillary Recovery and Coating Regeneration ...... 130 5.3.2 Laser Scanning Confocal Microscopy ...... 133 5.4 Conclusion ...... 137 5.5 References ...... 138

CHAPTER 6. CONCLUSIONS AND FUTURE STUDIES ...... 141 6.1 References ...... 147

VITA ...... 148

viii

LIST OF TABLES

Table Page

1.1 Classification of surfactant molecules used in MEKC ...... 11

3.1 Reproducibilities of PEM capillary coating. Conditions: same as Figure 3.3 ...... 94

4.1 Run-to-run reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration was varied ...... 114

4.2 Capillary-to-capillary reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration was varied...... 116

4.3 Reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration, 0.1 M...... 116

5.1 Stability of PEM capillary coating. Conditions: same as Figure 5.1 ...... 130

ix

LIST OF FIGURES

Figure Page

1.1 Schematic diagram of a CE system ...... 2

1.2 Origin of EOF – Electrical Double Layer ...... 5

1.3 (a) EOF flow profile, (b) Pressure driven flow profile and their corresponding solute zone profiles ...... 6

1.4 Solute migration ...... 7

1.5 Structure of a surfactant molecule ...... 10

1.6 Equilibria between a surfactant molecule, a surface monolayer, and a micelle .....12

1.7 Elution time window for neutral analytes in MEKC ...... 15

1.8 Origin of EOF in a packed capillary ...... 19

1.9 (a) Intrinsic, (b) Extrinsic charge compensation ...... 26

1.10 Scheme of the PEM-coated capillary ...... 27

1.11 Model for the aperture-based near-field optical microscope ...... 49

1.12 Schematic diagram of a confocal microscope ...... 52

2.1 Structures of the seven benzodiazepine analytes ...... 71

2.2 Effect of the nature of organic modifier on the CEC separation of benzodiazepines. Conditions: C18 stationary phase; 40 cm packed x 100 µm i.d.; electrolyte, 10 mM Tris (pH 8)-ACN (40:60), 10 mM Tris (pH 8)-MeOH (30:70), 10 mM Tris (pH 8)-THF (55:45); applied voltage, 30 kV; electrokinetic injection, 15 kV for 5 s; temperature, 25 °C; UV detection, 220 nm ...... 74

2.3 Effect of mobile-phase composition on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the mobile phase composition (ACN-Tris) was varied ...... 75

x

2.4 Effect of applied voltage on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the applied voltage was varied; electrolyte, 10 mM Tris (pH 8)-ACN (60:40) ...... 77

2.5 Effect of Tris concentration on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the Tris buffer concentration was varied; electrolyte, Tris (pH 8)-ACN (60:40); applied voltage, 20 kV ...... 78

2.6 Effect of column temperature on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the temperature was varied; electrolyte, 30 mM Tris (pH 8)-ACN (60:40); applied voltage, 20 kV ...... 79

2.7 CEC separation of drugs from a urine sample. (a) Without spiking. Conditions: C18 stationary phase; 40 cm packed x 100 µm i.d.; electrolyte, 30 mM Tris (pH8)-ACN (60:40); applied voltage, 20 kV; electrokinetic injection, 15 kV for 5 s; temperature, 15 °C; UV detection, 220 nm. (b) With spiking. Separation conditions are the same as above, except electrokinetic injection (urine sample), 20 kV for 10 s; electrokinetic injection (2 mg/ml standard oxazepam), 15 kV for 5 s ...... 80

3.1 Structures of the (a) monomeric SUG and (b) polymeric SUG ...... 90

3.2 Scheme of the PEM-coated capillary ...... 91

3.3 Stability studies of PEM coating. Conditions: 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SUG) with 0.2 M NaCl; pressure injection, 0.1 psi for 1 s; electrolyte, 50 mM Na2HPO4 (pH 9.2); applied voltage, 20 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm ...... 93

3.4 Effect of applied voltage on the OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except applied voltage was varied ...... 96

3.5 Effect of temperature on the OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except the temperature was varied ...... 97

xi

3.6 Comparison between monomeric and polymeric surfactants for OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except applied voltage, 30 kV. (a) 0.5% (w/v) poly (L-SUG) (b) 0.5% (w/v) mono (L-SUG) ...... 98

4.1 Structure of poly (L-SULV) ...... 105

4.2 Chiral separation of BNP. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV); pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); applied voltage, 30 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm. (a) 0.01 M NaCl; (b) 0.01 M 1E-3MI-HFP; (c) 0.01 M 1B-3MI-TFB ...... 109

4.3 Effect of NaCl concentration on the chiral separation of BNP. Conditions: same as Figure 4.2, except NaCl concentration was varied ...... 110

4.4 Effect of column temperature on the chiral separation of BNP. Conditions: same as Figure 4.2, except temperature was varied ...... 111

4.5 Effect of bilayer number on the chiral separation of BNP. Conditions: same as Figure 4.2, except bilayer number was varied ...... 112

4.6 Illustration of a run-to-run reproducibility for the chiral separation of BNP. Conditions: same as Figure 4.2, except NaCl concentration, 0.05 M ...... 115

4.7 (a) Chiral separation of BOH. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV) with 0.1 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); detection, 214 nm. Chiral separation of temazepam. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV) with 0.1 M NaCl; pressure injection, 0.9 psi for 7 s; background electrolyte, 25 mM Tris and 25 mM Na2B4O7 (pH 8.5); detection, 220 nm. (c) Chiral separation of secobarbital, and (d) Chiral separation of pentobarbital. Conditions for (c) and (d): 3 bilayers; 0.5% (w/v) PDADMAC and 2.5% (w/v) poly (L-SULV) with 0.2 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 300 mM H3BO3 and 30 mM Na2HPO4 (pH 7.2); detection, 214 nm. Other conditions are the same as Figure 4.2 ...... 117

xii

5.1 Dependence of electroosmotic mobility on the NaOH flushing time. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV) with 0.1 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); applied voltage, 30 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm ...... 128

5.2 Dependence of selectivity on the NaOH flushing time. Conditions: same as Figure 5.1 ...... 129

5.3 Coating Regeneration - Dependence of electroosmotic mobility on the NaOH flushing time. Conditions: same as Figure 5.1 ...... 131

5.4 Dependence of selectivity on the NaOH flushing time. Conditions: same as Figure 5.1 ...... 132

5.5 The z-Section images of a 20-bilayer capillary (left) and a fused-silica capillary (right). Optical slice thickness: 3.3 µm. z-Step size 1.65 µm. Every fourth image is shown. All images are 204 x 512 pixels (460.6 µm x 115.1 µm) ...... 134

5.6 z-Section images of a portion of the same 20-bilayer capillary, as in Figure 5.5. Exposure to 1.0 M NaOH. Optical slice thickness: µm. z-Step size 1.65 µm. Every third image is shown. All images are 512 x 512 pixels (65.8 µm x 65.8 µm) ...... 135

5.7 LSCM images of different regions within the same 20-bilayer capillary, as in Figure 5.6. Optical slice thickness: 3.3 µm. Both images are 512 x 820 pixels (115.1 µm x 184.4 µm) ...... 136

5.8 LSCM image of the same 20-bilayer capillary, as in Figure 5.7. Longer exposure to 1.0 M NaOH. Optical slice thickness: 3.3 µm. The image is 512 x 2048 pixels (115.1 µm x 460.6 µm) ...... 137

6.1 Structures of the six β-blocker analytes ...... 142

xiii

LIST OF ABBREVIATIONS

Abbreviation Name

AFM atomic force microscopy

BGE background electrolyte

BNP 1,1’-binaphthyl-2,2’-dihydrogenphosphate

BOH 1,1’-bi-2-naphthol

CDCPC cellulose tris(3,5-dichlorophenylcarbamate)

CE capillary electrophoresis

CEC capillary electrochromatography

CGE capillary gel electrophoresis

CIEF capillary isoelectric focusing

CITP capillary isotachophoresis

CMC critical micellar concentration

CTAB cetyltrimethylammonium bromide

CTAC cetyltrimethylammonium chloride

C18-TEOS n-octadecyltriethoxysilane

CZE capillary zone electrophoresis

DAPS N-dodecyl(N,N-dimethyl-3-ammonio-1-propane sulfonate)

Dioxo[13]aneN4 1,4,7,10-tetraazacyclotridecane-11,13,dione

DMPCC 3,5-dimethylphenylcarbamoyl cellulose

DS dextran sulfate

DTAB dodecyltrimethylammonium bromide

ELP extended light path

xiv

EOF electroosmotic flow

FESI field-enhanced sample injection

FRET fluorescence resonance energy transfer

GC gas chromatography

H3BO3 boric acid

HCl hydrochloric acid

HPLC high performance liquid chromatography

H2TPFPP 5,10,15,20-tetrakis(penta fluorophenyl)porphyrin

LC liquid chromatography

LODs limits of detection

LSCM laser scanning confocal microscopy

MEKC micellar electrokinetic chromatography

MIPs molecular imprinted polymers

Mono (L-SUG) mono (sodium N-undecenoyl-L-glycinate)

MS mass spectrometry

NA numerical aperture

Na2B4O7 sodium borate

NaCl sodium chloride

Na2HPO4 sodium phosphate

NaOH sodium hydroxide

NIR near-infrared

NSOM near-field scanning optical microscopy

ODS octadecylsilane

xv

OT-CEC open-tubular capillary electrochromatography

OT-CLC open-tubular capillary liquid chromatography

Packed-CEC packed capillary electrochromatography

PAHs polycyclic aromatic hydrocarbons

PAPS lithium 3’-phosphoadenosine 5’-phosphosulfate

PB polybrene

PDADMAC poly (diallyldimethylammonium chloride)

PEI polyethyleneimine

PEM polyelectrolyte multilayer pI isoelectric point

PMBC para-methylbenzoyl cellulose

Poly (L-SUG) poly (sodium N-undecanoyl-L-glycinate)

Poly (L-SULV) poly (sodium N-undecanoyl-L-leucylvalinate)

Poly (SUS) poly (sodium undecylenic sulfate)

PSS poly (styrene sulfonate)

PVA poly (vinylamine)

R6G rhodamine 6G

RSD relative standard deviation

SDS sodium dodecyl sulfate

SFC supercritical fluid chromatography

SMIL successive multiple ionic-polymer layer

TEOS n-octadecyltriethoxysilane

THF tetrahydrofuran

xvi

Tris tris(hydroxymethyl)aminomethane

1B-3MI-TFB 1-butyl-3-methylimidazolium tetrafluoroborate

1E-3MI-HFP 1-ethyl-3-methyl-1H-imidazolium hexafluorophosphate

xvii

ABSTRACT

The goal of the research reported in this dissertation is to develop packed and open-tubular capillary electrochromatographic methods for improved achiral and chiral separations of various classes of analytes. The first part of this research involves the separation of seven benzodiazepines by the use of the packed mode of capillary electrochromatography (CEC) and a 40 cm packed bed of Reliasil 3 µm C18 stationary phase. Optimal conditions were established by varying the mobile phase, the amount of organic modifier, the buffer concentration, the applied voltage, and the column temperature. The second part of this research focuses on the open-tubular mode of CEC and the polyelectrolyte multilayer

(PEM) coating approach. In the first study of this part, poly (diallyldimethylammonium chloride), PDADMAC, was used as the cationic polymer and poly (sodium N-undecanoyl-

L-glycinate), poly (L-SUG), was used as the anionic polymer for the construction of the

PEM coating. The performance of the modified capillaries as a separation medium was evaluated by use of seven benzodiazepines as analytes. In the second study, the anionic polymeric surfactant poly (sodium N-undecanoyl-L-leucylvalinate), poly (L-SULV), was used as the chiral discriminator for the separations of several drug analytes.

Reproducibility of the PEM coating was evaluated by computing the relative standard deviation (RSD) values of the electroosmotic flow (EOF). The PEM-coated capillaries were remarkably robust with excellent reproducibilities and high stabilities against extreme pH values. The stability of the capillary surface was further investigated after exposure to

NaOH solutions. The structural changes of these coatings were monitored using laser scanning confocal microscopy (LSCM). These changes were discussed in terms of separations using open-tubular CEC (OT-CEC). In addition, the electropherograms

xviii

obtained from the chiral separation of 1,1’-binaphthyl-2,2’-dihydrogenphosphate (BNP) in

OT-CEC allowed the measurements of both selectivity and electroosmotic mobility changes after long exposure to NaOH.

xix CHAPTER 1.

INTRODUCTION

1.1 Capillary Electrophoresis

Capillary electrophoresis (CE) is a family of related techniques that use narrow-bore fused-silica capillaries to perform high efficiency separations of both small and large molecules. Although CE is originally considered for the analysis of biological macromolecules, it has also been utilized for the separation of other compounds such as chiral drugs, vitamins, pesticides, dyes, inorganic ions, organic acids, and surfactants [1,

2]. CE offers a number of advantages when compared with chromatographic techniques:

(1) extremely small amount of sample; (2) high separation efficiency and resolution; (3) rapid and quantitative separation; (4) automated instrumentation; (5) various modes to vary selectivity; and (6) simple separation mechanism [2, 3].

As mentioned above, one of the main advantages of CE is the overall simplicity of the instrumentation. Figure 1.1 illustrates a schematic diagram of a typical CE system. The basic components of this system are a fused-silica capillary whose ends are placed in buffer reservoirs, a high-voltage power supply, a UV lamp, a photodiode-array detector, a sample reservoir, and two electrodes that are used to make electrical contact between the high-voltage power supply and the capillary. In CE, the separation of analytes is achieved by replacing the inlet buffer reservoir with a sample reservoir, and by applying either an electric field or an external pressure. After the sample is loaded onto the capillary, the sample is replaced by the inlet buffer reservoir, the electric field is applied, the ions in the sample move along the capillary, and the separation is performed.

1 Anode High-Voltage Cathode Power Supply Diode-Array Detector

Capillary

(+) Electrode (-) Electrode

UV-Lamp

Inlet Buffer Reservoir Sample Reservoir Outlet Buffer Reservoir

Figure 1.1 Schematic diagram of a CE system.

The utility of CE is also derived from its six different modes of operation. These modes of CE include capillary zone electrophoresis (CZE), capillary gel electrophoresis (CGE), capillary isoelectric focusing (CIEF), capillary isotachophoresis (CITP), micellar electrokinetic chromatography (MEKC), and capillary electrochromatography (CEC). CZE is the simplest form and the most widely used technique in CE [1, 2]. The separation principle of CZE is the difference in charge-to-size ratio and the difference in electrophoretic mobilities of solutes that result in different velocities. Electrophoretic

2 mobility is the mobility of an ion when an electric field is applied across the capillary. The velocity of an ion is expressed by:

ν = µ e E (1.1)

ν µ where is the ion velocity, e is the electrophoretic mobility, and E is the applied electric field. The electric field is a function of the applied voltage and the capillary length with units of V/cm. When a constant electric field is applied, ionic species undergo an

electrostatic force, Fe :

= Fe qE (1.2) where q is the charge of a particular ion. This force causes ions to accelerate towards the oppositely charged electrode. In a viscous medium the electrostatic force is balanced by its

frictional force, F f , that slows the mobility of ions. According to Stokes’ law for spherical particles, the frictional force can be given by Equation 1.3:

= πη ν F f 6 r (1.3) where η is the dynamic viscosity of the solution, and r is the radius of the particle or the ion. During electrophoresis a steady state is achieved, and at this point the forces are equal.

The combination of the above equations yields an equation that describes the electrophoretic mobility as follows:

q µ = (1.4) e 6πηr

The electrophoretic mobility is a property of a given ion and medium, and it is a physical constant for that ion. Equation 1.4 demonstrates that small species with high charges have high mobilities, whereas large, minimally charged species have low

3 mobilities. In addition, neutral species do not possess an electrophoretic mobility ( q = 0 ).

Therefore, they are not separated by use of CZE. Another important characteristic that is derived from the above equation is that the electrophoretic mobility of ionic species decreases as the viscosity of the background electrolyte (BGE) increases [1, 5, 6].

A significant factor that causes the movement of both neutral and charged species in solution is the electroosmotic flow (EOF). The origin of this flow is the electrical double layer that is formed at the solid-liquid interface, between the bulk liquid and the inner capillary surface (Figure 1.2). Under alkaline conditions, the inside wall of a fused-silica capillary is negatively charged due to ionization of the surface silanol groups to negatively charged silanoate groups. Electrolyte cations accumulate adjacent to the negative surface of the capillary, they absorb on it by electrostatic attraction, and they form an immobilized layer that is called the Stern layer. The remaining ions constitute the diffuse layer, which extends into the bulk liquid. This arrangement of ions develops the electric double layer.

When an electric field is applied across the column, the cations, which predominate in the diffuse layer, migrate in the direction of the cathode. Since ions are solvated by water, the bulk solution in the capillary is dragged along by the migrating charge.

This generates the EOF, which can be described as follows:

εζ ν =   EOF  E (1.5)  η 

ν ζ ε where EOF is the electroosmotic velocity, is the zeta potential, and is the dielectric constant of the BGE. The zeta potential is the corresponding potential across the layers, and it depends on the thickness of the diffuse layer and the surface charge on the capillary wall. The EOF can be also expressed in terms of mobility by the equation:

4 εζ µ = (1.6) EOF η

µ where EOF is the electroosmotic mobility.

Surface of capillary Stern layer

Diffuse layer

EOF Bulk solution

Figure 1.2 Origin of EOF – Electrical Double Layer.

A unique characteristic of the EOF is that it is uniformly distributed along the capillary.

This results in a flat plug-like profile (Figure 1.3a) rather than a parabolic profile that is generated by a pressure-driven system (Figure 1.3b), such as in high performance liquid chromatography (HPLC). A flat profile of EOF reduces band broadening, enhances fast elution rates, and yields highly efficient peaks.

Another distinctive feature of the EOF under normal conditions, where the capillary surface is negatively charged, is that it causes movement of nearly all species in the same direction, regardless of charge. In CZE, the migration of charged species is

5 determined by the EOF and by the electrophoretic mobility of the solutes. This can be expressed by the following equation:

µ = µ + µ αpp e EOF (1.7)

µ where αpp is the apparent mobility. The value of electrophoretic mobility of an ion depends on the charge of that ion, and it can be either positive or negative. Therefore, cations migrate fastest since both EOF and electrophoretic mobility are in the same direction. Neutral molecules move along with the EOF, because they do not have electrophoretic mobility, and they are not separated. Finally, anions migrate slowest since their electrophoretic mobility is in the opposite direction of EOF. However, they still move toward the detector, because the magnitude of EOF is much greater than their electrophoretic mobility. This is clearly illustrated in Figure 1.4.

EOF Pressure Driven Flow

Flat profile Parabolic profile

Figure 1.3 (a) EOF flow profile, (b) Pressure driven flow profile and their corresponding solute zone profiles.

6

Capillary electrophoretic separations are also performed using another mode of CE, called CGE. This form of CE is performed in a porous gel polymer matrix, and the separation of analytes is based on their charge and size. CGE has been mostly employed for the analysis of large biomolecules, such as proteins and nucleic acids. The most common sieving gels used in CGE are cross-linked polyacrylamide and agarose. The former gel has smaller pore size, and it is used for protein separations, while the latter, with a larger pore size, is mostly applied to DNA analysis [2, 4, 5].

N N N N EOF

N N

Figure 1.4 Solute migration.

CIEF is an electrophoretic technique used to separate amino acids, peptides, and proteins based on their pI (isoelectric point) values. In CIEF, a pH gradient is formed within the capillary using ampholytes. Ampholytes, or else zwitterions, are molecules that contain both an acidic and a basic moiety. They have pI values between pH 3 and 9. The

7 gradient is formed when the capillary is filled with sample and ampholytes. A basic solution is placed at the cathode, and an acidic solution is placed at the anode. When the electric field is applied, the charged ampholytes and the sample components migrate through the medium until they reach their pI value, at which they become uncharged. This procedure is known as “focusing”, and it is indicated by a decrease in current. After this, the contents of the capillary are mobilized to the detector by pressure or by adding an electrolyte into one of the reservoirs [1, 4, 5].

CITP is another electrophoretic technique, in which a combination of two electrolytes causes all analyte bands to migrate at the same velocity. This technique can separate either cations or anions, but not both at the same time. During a separation procedure, an analyte mixture is injected between a leading electrolyte containing ions of higher mobility (Cl-) than any of the analyte ions and a terminating electrolyte with ions that have lower mobility (heptanoate) than the sample ions. When an electric field is applied, the analyte ions migrate in bands according to their unique mobilities. After all analyte ions are separated into different bands, they move at the same velocity [1, 4].

MEKC and CEC are hybrid techniques that combine the best features of both electrophoresis and chromatography. Both electrophoretic techniques can be used for the separation of neutral as well as charged compounds. MEKC involves the introduction of a surfactant at a concentration above the critical micellar concentration (CMC), at which micelles are formed. The separation is based on the hydrophobic and ionic interactions of the analytes with the micelles that act as a moving stationary phase, or else a pseudostationary phase. CEC uses a stationary phase rather than a micellar pseudostationary phase. As in MEKC, the mechanism of separation depends on the

8 partitioning between the two phases. However, in the case of CEC, the partitioning occurs between the packed or coated stationary phase and the mobile phase. In addition, when the analytes are charged, the separation also depends on their electrophoretic mobilities [3-7].

Both approaches are discussed in detail in the following sections.

1.2 Micellar Electrokinetic Chromatography

MEKC, which was first introduced by Terabe et al. in 1984 [8], is the second most commonly used CE technique. Although MEKC is a form of CE, its separation principle is more similar to HPLC than to CE. As stated earlier, the separation mechanism is based on the electrophoretic mobility and the partitioning of the analytes between the mobile phase and the micellar pseudostationary phase. However, for neutral species, it is only the partitioning that effects the separation. The driving force for the partitioning of analytes is hydrophobicity. In addition, hydrogen bonding, dipole-dipole, and dispersive interactions can contribute to the solute partitioning between the two phases. The most commonly used pseudostationary phase in MEKC is the micelle [1, 5, 9-11]. A more detailed discussion about surfactants and micelles is given in the succeeding paragraphs.

Surfactants, which are also known as amphiphiles, are organic compounds that consist of a polar or ionic head group and a hydrocarbon tail group (Figure 1.5). The head group has hydrophilic properties, and it is compatible with the aqueous environment, while the tail group is hydrophobic and not friendly with the aqueous environment. Surfactants are classified into different groups according to the nature of their head group (anionic, cationic, nonionic, and zwitterionic). Table 1.1 reports these four groups along with the structures of different head groups and a few examples.

9 The surfactant molecules, at a low concentration, are dispersed in solution.

However, at a concentration above the CMC, and above a certain temperature, known as the Krafft temperature, the surfactant molecules begin to build up their own structures.

They aggregate together to form micelles in the interior of the aqueous solution and monolayers at the surface of the solution (Figure 1.6).

Hydrophilic Head Group

Hydrophobic Tail Group

Figure 1.5 Structure of a surfactant molecule.

The CMC is a very important characteristic of each surfactant. Its value needs to be determined in order to understand some of the characteristics and the properties of a surfactant or a micelle. The CMC value may be attained graphically by plotting a physicochemical property of the solution, i.e. surface tension, conductivity, turbidity, osmotic pressure, fluorescence, and light scattering, versus the concentration of the surfactant. The value of the CMC is the discontinuity in the plot, or the sudden variation in the above relation. In an aqueous medium, above the CMC, surfactant molecules aggregate together to form a micelle. As the surfactant concentration increases, the shape of the ionic

10 micelle changes from spherical to cylindrical, to hexagonal, and finally to lamellar. For nonionic micelles, as the concentration increases the shape changes from spherical directly to lamellar [6]. The number of aggregated molecules in a micelle is called the aggregation number, n , and it is usually around 50 to 100. It can be determined by several analytical techniques, such as light scattering, fluorescence, and NMR sedimentation.

Table 1.1 Classification of surfactant molecules used in MEKC.

Hydrophilic group Structure Examples

- Anionic R-O-SO 3 SDS

+ Cationic R-N (CH3)3 CTAB, CTAC, DTAB

Nonionic R-O-Glucose Octyl glucoside

R-Cyclohexyl-O-PEG Triton X-100-RS

+ - Zwitterionic R-N (CH3)2-(CH2)3-SO 3 PAPS, DAPS

R: alkyl group For abbreviations see pages xiv, xvi

The MEKC separation process for neutral analytes is only based on their partitioning in and out of the micelle. When anionic micelles are used, the more the analyte interacts with the micelle the longer its migration time. This is because anionic micelles have electrophoretic mobilities that are in the opposite direction of the EOF. The more

11 hydrophobic analytes interact more strongly with the micelle, and they are retained longer since the micelle partially offsets the effects of the EOF.

Surfactant Monolayer

Micelle

Figure 1.6 Equilibria between a surfactant molecule, a surface monolayer, and a micelle.

The separation mechanism in MEKC is chromatographic, and it can be described by the use of modified chromatographic relationships. How effective a chromatographic column is in separating two analytes partially depends on the relative rates at which the analytes are eluted. These rates are determined by the magnitude of the equilibrium constants (partition coefficients), K , which are defined as:

c K = S (1.8) cM

where cS is the molar concentration of the analyte in the micellar pseudostationary phase

and cM is its molar concentration in the mobile phase. The retention factor, or capacity factor, k' , describes the migration rates of analytes in the column, or the ratio of moles of analyte in the micelle to those in the mobile phase:

12 ()t − t  V  k'= r 0 = K S  (1.9)  t  V   − r   M  t0 1   tm 

where tr is the retention time of the analyte, t0 is the retention time of the unretained

analyte that moves with EOF, tm is the retention time of the micelle, VS is the volume of

the micellar phase, and VM is the volume of the mobile phase. The above equation is partly different from the normal chromatographic k' because of the movement of the micellar

pseudostationary phase. If the micelle becomes the stationary phase the term tm becomes infinite, and the Equation 1.9 is converted into the conventional chromatographic equation.

The selectivity factor α of a column for two analytes A and B is determined by the equation:

()t − t α = r B 0 (1.10) ()− tr A t0 () () where tr B and tr A are the retention times of the more strongly and less strongly retained analytes B and A, respectively. This definition dictates that the selectivity is always greater than or equal to unity.

The resolution, RS , of a column describes its ability to separate two analytes. This column resolution is defined as:

  t    1−  0   1/ 2 '    N α −1 k  tm  R =    2    (1.11) S    '   4  α  k +1  t     2 1−  0 k '     1    tm  

13 where N is the number of theoretical plates. This number of theoretical plates is widely used as a quantitative measure of column efficiency, and is defined by use of the following equation:

2  t  = r N 5.54  (1.12) W1/ 2 

where W1/ 2 is the width of the peak at half its maximum height.

In Equation 1.11, the first derivative corresponds to efficiency, the second to selectivity, and the third and fourth to retention. This suggests that resolution can be improved by optimizing efficiency, selectivity, and/or the capacity factor. Since the capacity factor is proportional to the concentration of the surfactant, it can be easily modified by varying this concentration. In addition, resolution can be improved by

increasing the elution range, or time window, which is the time between t0 and tm (Figure

1.7). Neutral analytes elute between this elution range. The analytes that elute with the

EOF are hydrophilic and do not interact with the micelles. The analytes that are completely

t retained by the micelles elute with the micelles. Maximum resolution is obtained when 0 tm is very small (large time window). Selectivity can also be easily modified by varying the size, charge, and geometry of the micelle, and particularly the surfactant [1, 3].

In chromatography and electrophoresis, resolution is given by:

2[()t − ()t ] R = r B r A (1.13) S + WA WB

where WA and WB are the widths of the base of peaks A and B, respectively. A resolution of 1.5 gives a baseline separation of the two analytes. The resolution can be improved by

14 increasing the length of the column, thus, increasing the number of theoretical plates

(column efficiency). However, this results in an increase in separation time [4].

Time window

Analytes Micelle

EOF

0t0 tr1 tr2 tr3 tm

Time

Figure 1.7 Elution time window for neutral analytes in MEKC.

1.3 Capillary Electrochromatography

In recent years, CEC has become an important member of the arsenal of tools available in separation science. Although the CEC approach was first introduced by

Pretorius et al. [12] in 1974, it received renewed interest during the 1990s [13-19]. This revival of CEC is because it is a microcolumn electroseparation technique that combines the best features of HPLC with those of CE. In essence, CEC mainly couples the high

15 selectivity of HPLC and the high separation efficiency of CE [20-26]. This is because in

CEC, the mobile phase solvent is transported through a capillary by use of the EOF. In addition, CEC provides high resolution, short analysis time, ruggedness, and low sample and mobile phase consumption. It also offers wider selectivity and facilitates the separation of both neutral and charged compounds.

As discussed earlier, the latter characteristic is because the separation of analytes is based on their interactions with the stationary phase and, when charged, their electrophoretic mobility [15, 25-32]. Therefore, the apparent mobility for a charged analyte in CEC is influenced by its electrophoretic mobility, its partitioning interaction and the

EOF. However, for a neutral analyte, the apparent mobility depends only on the partitioning interaction with the stationary phase, while its elution is driven by the EOF.

Several manuscripts have described the capacity factor for separation of an analyte

′ in CEC [33-36]. Briefly, the capacity factor in CEC ( kCEC ) can be expressed as:

 µ  k ' −  e   µ  k ' =  EOF  (1.14) CEC  µ  1+  e   µ   EOF 

This equation incorporates both the electrophoretic and chromatographic separation mechanisms in CEC. It should be noted that the mechanism of separation in CEC is highly

µ ′ ′ dependent upon the analyte. For neutral analytes, e is zero, and kCEC is equal to k .

Thus, neutral analytes are solely separated on the basis of a chromatographic mechanism.

Finally, charged analytes are separated by a combination of chromatographic and electrophoretic mechanisms, provided these analytes interact with the stationary phase.

16 In CEC, as in every chromatographic system, the capillary column is the most important component. It serves as a vessel to transport the mobile phase and as a separation channel [15, 28]. Therefore, preparation of the column, and particularly the stationary phase packed or immobilized on the inner walls of the capillary, is critical for

CEC. The capillary columns in CEC are usually classified into three main formats [21, 22,

28, 37-44]: (i) packed-CEC [23-27, 45-52], (ii) open-tubular CEC (OT-CEC) [39-43, 53-

64], and (iii) monolithic (or continuous rod) CEC [20, 38, 65-85]. In this dissertation, the first two main formats are discussed in detail.

1.3.1 Packed-Capillary Electrochromatography

In packed-CEC, a fused-silica capillary is filled with a typical HPLC packing material (i.e. octadecyl silica). The mobile phase in packed-CEC is driven by the EOF that is induced by applying an electric field across the capillary column. The origin of this flow is the electrical double layer that is formed at the solid-liquid interface of a charged surface in contact with the BGE (Figure 1.8).

In a capillary packed with silica particles the surfaces of the capillary wall and the particles are negatively charged due to the dissociation of silanol groups. The velocity of

EOF in packed-CEC is shown in the following equation:

1/ 2  ε ε RT  σ  o r  E 2cF 3 µ =   (1.15) EOF η

σ ε where , which is proportional to zeta potential, is the charge density at the surface, o is

-12 2 -1µ-2 ε the permittivity of vacuum (8.85 x 10 C N ), r is the dielectric constant of the mobile phase, R is the gas constant, T is temperature, c is the concentration of the

17 electrolyte, F is Faraday’s constant, η is the viscosity of the mobile phase, and E is the electric field strength. The above equation illustrates that the velocity of EOF depends on the charge density on the surfaces of the capillary walls and silica particles, the dielectric constant, the viscosity, and the concentration of the BGE, and the temperature.

However, there are some problems that need to be solved in order for this conventional form of CEC to be a viable alternative to both CE and HPLC. One of the problems of packed-CEC is the requirement to fabricate frits, which are needed to retain the packed particles inside the capillary column. Another problem in packed capillaries is the tendency to form bubbles around the packing material or at the frit. This problem often results in an unstable baseline, variable migration times, and current breakdown.

Pressurization of both ends of the column is required, and the mobile phase must be thoroughly degassed in order to reduce the possibility of bubble formation. Another major drawback of conventional CEC is that the packing procedure is generally more difficult than for HPLC due to the narrow inner bore of the capillary and the small diameter of the particles (5 µm or less). Finally, basic compounds are difficult to separate in packed-CEC due to the presence of silanol groups that are needed to generate an adequate EOF [21, 39-

41, 86-93]. In order to circumvent the problems mentioned above, continuous bed-type columns, i.e. open-tubular and monolithic columns, have been suggested as alternatives to packed-CEC.

1.3.2 Open-Tubular Capillary Electrochromatography

In the OT-CEC format, the stationary phase is deposited on the inner walls of the capillary. Preparation of the stationary phase, or coating, is a crucial step in any chromatographic system. In OT-CEC, this coating needs to be stable in order to provide

18 efficient chromatographic separations and a reproducible EOF [92]. In OT-CEC, there are six general approaches used for modifying the capillary column. These include (i) adsorption, (ii) covalent bonding and/or crosslinking, (iii) porous silica layers, (iv) chemical bonding after etching, (v) sol-gel, and (vi) molecular imprinting [39-42, 91-94].

Although OT-CEC is considered an alternative to packed-CEC, its phase ratio and sample capacity are relatively low due to the typical small surface area of the coating. Several options have been proposed to increase the surface area, and therefore, to increase the interactions between the analyte and the coated phase. These options include polymer coatings, porous silica layers, etching, and sol-gel techniques [22, 41, 89, 94].

Electrical double layers Capillary wall

Stationary phase particles

EOF

Figure 1.8 Origin of EOF in a packed capillary.

19 1.3.2.1 Adsorption

Adsorption of chemicals to solid surfaces is a common phenomenon observed in separation science. The adsorption of proteins, peptides, and basic compounds on the silica capillary wall usually causes serious problems, such as loss in efficiency, peak tailing, unstable baseline, nonreproducible migration times, and reduction of capillary lifetime. In order to reduce or eliminate the analyte-wall interactions, different methods have been developed. The most common methods shield the negatively charged silanol groups with another layer, the adsorbed stationary phase [28, 39, 61, 95]. According to the strength of adsorption, the adsorbed stationary phases can be divided into two groups: dynamically adsorbed and physically adsorbed stationary phases [28, 30, 92]. If the adsorption of the modifier on the capillary wall is weak, the stationary phase is called dynamically adsorbed stationary phase. In this case, the modifier is added to the mobile phase. In contrast, if the modifier is strongly adsorbed, the stationary phase is termed a physically adsorbed stationary phase. In this case, the addition of the modifier to the mobile phase is not necessary. Several modifiers have been used for the preparation of adsorbed stationary phases [17, 18, 39, 59, 61, 62, 92, 93, 95-121]. These modifiers can be grouped into three main categories: (i) cationic surfactants, (ii) polymeric surfactants, and (iii) charged polymers.

(i) Cationic Surfactants

In 1990, Pfeffer and Yeung [17] developed a method for the separation of several polycyclic aromatic hydrocarbons (PAHs) by use of open-tubular capillary liquid chromatography (OT-CLC). The capillaries were first coated with a polymer solution of

0.9% PS-264 (polyvinylsiloxane). The velocity of the EOF on these columns was low due

20 to the polymer coating that blocked the silanol groups on the capillary wall. The cationic surfactant cetyltrimethylammonium bromide (CTAB) was then added to the mobile phase at low concentrations (µM). When this eluent was used in a polymer coated, reversed- phase open-tubular column, a substantial EOF was developed. This, in turn, provided highly efficient and fast separations of neutral compounds. In 1991, Pfeffer and Yeung

[18] used OT-CLC for the separation of anions. This separation involved partitioning of the anions with the help of the ion-pairing agent, tetrabutylammonium (TBA) cation, being adsorbed onto the surface of an open-tubular column. The separation of the anions was based on differences in retention, rather than differences in electrophoretic mobility. The ion pairing agent, which was added in the mobile phase at low concentrations (600 µM), was used to control the affinity of the anions for the stationary phase.

Garner and Yeung [95] proposed the use of an ion-exchange mechanism for the separation of compounds with similar mobilities, such as 4-amino-1-naphthalenesulphonic acid and 5-amino-2-naphthalenesulphonic acid. Capillaries coated with a hydrophobic stationary phase were shown to be dynamic ion exchangers when the quaternary ammonium compound, CTAB, was added to the mobile phase. CTAB was dynamically adsorbed to the column, forming a charged double layer. Sample ions were then retained due to their coulombic attraction to the double layer. In this study, they showed that the retention of the sample ions could be varied in this system by changing various parameters such as the concentration of the buffer ion, the addition of an organic modifier, and the concentration of the cationic surfactant CTAB.

Liu et al. [105-107] developed a novel preparation method for a physically adsorbed stationary phase in OT-CEC for the separation of both achiral and chiral

21 compounds. The adsorbed stationary phases were prepared by simply rinsing the capillary with a mobile phase containing a cationic surfactant, such as CTAB or a basic chiral selector, such as protein, peptide or amino acid. After adsorption of the stationary phase, the capillary was rinsed again with a mobile phase, which did not contain either CTAB or a chiral selector. Five alkyl substituted benzenes were baseline separated within 6 min [105].

The run-to-run reproducibility of retention time was good with relative standard deviation

(RSD) values of less than 2.3%. However, the day-to-day reproducibility was not as good with RSD values of less than 10% [106, 107].

Baryla et al. [100] described the adsorption mechanisms and aggregation properties of the cationic surfactants CTAB and didodecyldimethylammonium bromide (DDAB) that were used for the preparation of dynamically adsorbed stationary phases. Atomic force microscopy (AFM) was used for the elucidation of the coating morphology. In addition, separation of the basic proteins lysozyme, ribonuclease A, α-chymotrypsinogen A, cytochrome c, and myoglobin was performed. Two of the five proteins were irreversibly adsorbed to the wall. All three peaks that were present were broad and tailed due to wall adsorption. The use of a CTAB-coated capillary gave a baseline separation of three of the proteins in less than 15 min with high efficiencies. When a DDAB-coated capillary was used all five proteins were separated in less than 6 min. This was due to the increased surface coverage provided by DDAB.

(ii) Polymeric Surfactants

Although micelles have been successfully used in separations, they have some significant limitations that cause problems in MEKC. The most important problem associated with conventional micelles is the dynamic equilibrium between the surfactant

22 molecules and the micelle. The presence of dynamic equilibrium lowers the stability of the pseudostationary phase. PAHs, which are considered extremely hydrophobic compounds, are difficult to separate because they co-migrate with the micelle. This problem can be solved by either increasing the concentration of the surfactant or by adding an organic modifier in the BGE solution. However, organic modifiers usually perturb the micelle structure. In addition, an increase in the concentration of the surfactant may cause longer migration times and Joule heating, which has detrimental effects on MEKC separations [6,

7, 11]. In order to compensate the above problems, new pseudostationary phases have been developed by use of chiral and achiral polymeric surfactants.

In 1994, Wang and Warner [120] reported the use of a polymeric surfactant added to the BGE in MEKC. Polymeric surfactants offer several distinct advantages over conventional micelles [121-126]. Firstly, polymerization of the surfactant eliminates the dynamic equilibrium due to the formation of covalent bonds between the surfactant aggregates. This, in turn, enhances stability and improves resolution. Secondly, polymeric surfactants can be used at low concentrations because they do not depend on the CMC.

This usually provides higher efficiencies and rapid analysis.

Harrell et al. [119] developed a dynamic coating by use of a novel nonionic micelle polymer, poly (n-undecyl-α-D-glucopyranoside). They applied this coating to the separation of seven tricyclic antidepressants. Although physical and dynamic adsorptions have simple and rapid coating procedures and good reproducibilities, they usually have short lifetimes and limited pH ranges [92, 112, 113]. Both coatings are adsorbed to the capillary wall via electrostatic interactions and hydrogen bonding. These interactions are weaker than covalent bonds. However, multiple electrostatic interactions within a coating

23 that involves a layer-by-layer deposition process provides greater stability and longer lifetime [92, 93, 112-114, 121].

A layer-by-layer coating is termed a polyelectrolyte multilayer (PEM), which is usually constructed in situ by use of alternating rinses of positively and negatively charged polymers [114, 127-130], where the negatively charged polymer may be a polymeric surfactant [92, 93, 121]. A layer of polymer adds to the oppositely charged surface, reversing the surface charge and priming the film for the addition of the next layer via electrostatic forces. The advantages of such coatings are two-fold. First, less consumption of the polymer is required, since it is adsorbed onto the capillary wall. Second, there is less detection interference between the polymer and the analyte of interest, which in turn, makes the system more amenable to coupling with mass spectrometry (MS).

Although the basic idea of the PEM coating is simple, a theoretical description is relatively complex due to the long range of the coulombic interaction between layers. In addition, several techniques have been employed (neutron reflectometry, atomic force microscopy, infrared spectrometry) in order to better understand the mechanism of the

PEM coating formation. However, it is still not very well understood. The PEM coating is constructed by the use of the simple procedure described above. The first layer should have high adsorptivity in order to strongly attach the multilayer to the substrate. Several studies have shown that after the first four or five layers, the thickness of the coating increases linearly with the number of layers, and the zeta potential has a value of around zero. The thickness of each layer depends on the amount of salt added to the polymeric solutions, and not on the surface charge of the substrate. However, the total thickness of the coating increases with a surface charge increase. In addition, it is observed that salt has the

24 strongest effect on the polyelectrolyte layer thickness. Polymer concentration, polymer type, molar mass, salt type, deposition time, and solvent are less important variables [127,

131, 132].

All polymer deposition solutions contain an amount of salt, usually NaCl. Salt moderates the electrostatic interactions that occur between the oppositely charged polyelectrolyte sections, Pol+ and Pol-, according to the following chemical equation:

+ - + - + - - + Pol Pol m + Na aq + Cl aq ↔ Pol Cl m + Pol Na m where m refers to the multilayer phase. The charges on a polymer can be balanced by either those on the oppositely charged polymer or by the salt counterions within the film.

In the first case, where the positive charge of one polymer is balanced by the negative charge of another polymer, the compensation is called intrinsic (Figure 1.9a). In the second case, the compensation is extrinsic, since much of the polymer charge is balanced by the salt ions that are added in the polymer deposition solution (Figure 1.9b) [127, 131].

In this dissertation, the PEM coating approach is explored, and its performance is evaluated by use of different drug compounds. For the achiral studies, the polymeric surfactant poly (sodium N-undecanoyl-L-glycinate), poly (L-SUG), was used as the anionic polymer and poly (diallyldimethylammonium chloride), PDADMAC, was used as the cationic polymer [92]. A diagrammatic scheme of the PEM coated capillary is illustrated in

Figure 1.10. This diagram is not intended to give an actual structural representation of the bilayer, but rather a representation of the order of polymer deposition. More details concerning this work are given in Chapter 3.

25 ------+ + - - + + + + + + + + + + + + (a) Intrinsic

- - - Na+ - --Na+ -- - - - + - Na - - Na+ - Cl - - + + + Cl Cl- Cl + + + + ++ +++ + + (b) Extrinsic

Figure 1.9 (a) Intrinsic, (b) Extrinsic charge compensation.

Kamande et al. [93] investigated the use of a PEM coating for the separation of phenols and benzodiazepines. For their studies, they used the polymeric surfactant, poly

(sodium undecylenic sulfate), poly (SUS), and PDADMAC in a single bilayer PEM coating. The run-to-run and capillary-to-capillary reproducibilities were very good, and the

RSD values of EOF were less than 1.5%. The endurance of the coating was more than 100 runs. In addition, this study demonstrated the importance of the PEM coating by comparing the separation of benzodiazepines using CZE and MEKC.

26 Capillary wall

Silicate surface Si Si Si Si Si Si Si Si Si (deprotonated O O O O O O O O ------silanol groups) O O O O O O O O O

+ + + + + + + N N N N N+ N+ N+ N N+ N N

+NaO- +NaO- +NaO- +NaO- +NaO- C O C O C O C O C O + - + - + - + - + - O NH NaO O NH NaO O NH NaO O NH NaO O NH NaO +NaO- C O C C +NaO- C O C C +NaO- C O C C +NaO- C O C C +NaO- C O C C Bilayer O O O O O NH O NH O NH O NH O NH O C NH C NH C NH C NH C NH C C C C C O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O O O O O C C C C C C C C C C O H NH O O H NH O O H NH O O H NH O O H NH O + - C N + - C N + - C N + - C N + - C N NaO C NaO C NaO C NaO C NaO C O O O O O O - + O - + O - + O - + O - + C ONa C ONa C ONa C ONa C ONa N C N C N C N C N C H O H O H O H O H O O O O O O HN C HN C HN C HN C HN C C O C O C O C O C O O O O O O +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C HN - + HN - + HN - + HN - + HN - + C ONa C ONa C ONa C ONa C ONa C C C C C HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O O O O O O C HN C HN C HN C HN C HN - + - + - + - + - + ONa C ONa C ONa C ONa C ONa C O -ONa + O -ONa+ O -ONa+ O -ONa+ O -ONa +

Figure 1.10 Scheme of the PEM-coated capillary.

(iii) Charged Polymers

Erim et al. [104] reported the use of a simple method for the preparation of a polyethyleneimine (PEI) coating on the inner surface of fused-silica capillaries. The PEI layer was constructed by flushing the capillary with a solution that contained high- molecular-mass PEI. These authors investigated the reproducibility and the long-term stability of the PEI coating by measuring the plate numbers and retention times of some basic proteins. The RSD values in migration times for run-to-run reproducibilities ranged from 0.5 to 1.5%, and for column-to-column reproducibilities ranged from 1.9 to 2.8%.

The stability of the coating was tested by flushing the capillary with a solution at pH 11.0 for 60 hours. The migration times of all proteins after 60 hours slightly increased, and the efficiencies were not altered.

A stable modification of the inner wall was also developed by using a simple coating procedure, i.e. successive multiple ionic-polymer layer (SMIL) coating. Katayama

27 et al. [112, 113] used polybrene (PB) as the cationic polymer, and dextran sulfate (DS), alginic acid, or hyaruronic acid as an anionic polymer. In their first study [112], they established an anion-modified capillary (SMIL-DS capillary) by first attaching the cationic polymer to the capillary wall, and then the anionic polymer to the cationic polymer layer.

Efficient separation of acidic proteins was achieved even at pH values under 7.4. The ability to prevent the adsorption of the acidic proteins to the SMIL-DS capillary wall was due to the presence of sulfonic groups in the DS layer. These groups provided strong repulsion between the protein and the capillary wall. However, the SMIL-DS capillary was not suitable for the separation of basic proteins since the capillary wall was negatively charged through pH 2-11. In a pH range of 2-11, the SMIL-DS capillary exhibited a pH- independent EOF from anode to cathode. The endurance of the SMIL-DS coated capillary was more than 100 runs and the capillary-to-capillary variation was less than 1%. In their next study [113], this group further modified the SMIL coating and developed a cation- modified capillary (SMIL-PB capillary) by attaching the cationic polymer to the anionic polymer layer. The SMIL-PB capillary was applied to the separation of basic proteins. The separation was possible even when the pH of the mobile phase was near the pI of the protein. The SMIL-PB capillary demonstrated strong endurance with the achievement of

600 runs and excellent chemical stability against 1 M sodium hydroxide (NaOH) and 0.1

M hydrochloric acid (HCl). The RSD values of the run-to-run, day-to-day, and capillary- to-capillary reproducibilities were below 1%.

Poly (styrene sulfonate), PSS, is another anionic polymer that has been used in a

PEM coating procedure. Graul et al. [114] used this coating for the separation of a series of basic proteins. The coatings used for protein separations and reproducibility studies

28 consisted of 6.5 layer pairs. A layer pair is a layer of cationic polymer plus a layer of anionic polymer, also termed a bilayer. The polymer deposition solutions for the construction of the first 3.5 bilayers contained no salt, and for the last 3 contained 0.5 M sodium chloride (NaCl). They observed an excellent run-to-run stability over a wide range of pH. These capillaries were also proven to be very stable to extremes of pH (pH 12.0 and pH 2.0) and ionic strength, and to dehydration/rehydration. The RSD values of capillary- to-capillary reproducibilities were less than 2% when both pH 6.0 and pH 4.0 were used.

In addition to the reproducibility control of EOF, these authors achieved stable flow rates immediately after exposure of the column to mobile phase, as well as reverse flow.

1.3.2.2 Covalent Bonding and/or Crosslinking

Fixation of a layer by covalent bonding and/or crosslinking is another approach used for modifying the capillary [58, 63, 101, 102, 133-142]. Although this approach offers a long capillary lifetime, it usually requires a more complicated coating procedure.

Rehder et al. [58] studied the electrochromatographic separation of bovine β-lactoglobulin variants A and B by covalently attaching DNA aptamer, which form a G-quartet conformation to a capillary surface. Aptamers are short, single-stranded oligonucleotides that are recognized for their high affinity and specific binding to target molecules. Their use as a stationary phase offers many advantages. They are stable, easy and simple to produce, manipulate, and attach to silica surfaces, and they are compatible with a wide range of separation conditions and mobile phases. For the preparation of the stationary phase, oligonucleotides were covalently attached to the inner capillary surface using the organic linker molecule, sulphosuccinimidyl 4-(N-maleimidomethyl) cyclohexane-1- carboxylate.

29 A porphyrin derivative, 5,10,15,20-tetrakis(penta fluorophenyl)porphyrin,

H2TPFPP, was also used as a capillary wall modifier in OT-CEC by Charvatova et al.

[102]. H2TPFPP was physically adsorbed or covalently bonded to the capillary surface for the separation of aromatic carboxylic acids. In both covalently bonded and physically adsorbed phases, the capillaries were filled with a solution of the porphyrin derivative in dichloromethane. In the first case, both ends of the capillary were closed and the capillary was left overnight at room temperature. In the second case, the capillary was left in the vacuum oven for 2 hours at 60 °C. The synthetic strategy used for the preparation of chemically bonded phases was based on the generation of anionic silanol groups on the capillary wall, which were then employed for nucleophilic substitution at the para position of H2TPFPP. The main advantage of H2TPFPP is that it is capable of increasing the EOF.

The results from this study showed that coating of the capillary with H2TPFPP gave a better resolution at all pH values tested. Although both covalent bonding and physical adsorption resulted in similar results at pH 8.5 and pH 6.0, covalently bonded capillaries offered better results. At pH 5.0, physically adsorbed coatings lead to results, which were not reproducible.

In another study, Chiari et al. [101] reported the use of poly (vinylamine), PVA, for the separation of polyanionic acids. The coating procedure consisted of two steps: (i) dynamic adsorption of PVA, and (ii) crosslinking and derivatization of PVA by nucleophilic addition of primary amino groups to the conjugated double bonds of N, N- methylenebisacrylamide and N, N, N-trimethylaminoethylacrylamide. The best stability of this coating was achieved with the crosslinked PVA. There was no change in the EOF even after 120 hours of continuous use at pH 4.0. RSD values and average transit times of ten

30 injections were taken every 6 hours during the 120-hour study. The values between 0.2 and

1% indicated excellent reproducibility.

Calixarene-coated capillaries were also used in OT-CEC. Such capillaries are expected to provide more efficient separations of larger compounds [139-143]. Most recently, Wu et al. [140] have successfully used p-tert-butylcalix[8]arene bonded capillaries for the separation of o-, m-, and p-benzenediols, α- and β-naphthols, and α- and

β-naphthylamines. These capillaries were prepared with the use of γ- glycidoxypropyltrimethoxysilane as a bridge. The bonded capillaries showed good separation selectivity, suggesting significant interactions between the analytes and the bonded phase. In addition, for the evaluation of the stability of the capillary surface, five batches prepared over a period of six months were tested. All capillaries exhibited high stability and reproducibility.

1.3.2.3 Porous Layers

Another approach to increasing the phase ratio, the surface area, and the loading capacity of a capillary is by use of porous-layer open-tubular columns [13, 143-146].

Huang et al. [13] introduced the utility of such columns with a positively charged hydrocarbonaceous porous layer as the stationary phase for the separation of basic proteins and peptides. The porous layer was crosslinked and prepared by in situ polymerization of vinylbenzyl chloride and divinylbenzene in the presence of 2-octanol as a porogen inside a silanized fused-silica capillary. The chloromethyl groups at the surface of the porous layer were reacted with N, N-dimethyldodecylamine to obtain a positively charged surface with fixed C12 alkyl chains. Stability was monitored daily by measuring the electroosmotic mobility. They observed that the migration times remained constant for more than a week.

31 A porous silica layer chemically modified with C18 groups was used as a stationary phase by Crego et al. [143]. The 9.60 µm i.d. capillary used in this work had a thin porous silica layer (0.70 µm) with C18 groups chemically bonded onto the prepared layer. These authors investigated the effects of several experimental parameters, such as the volume fraction and the type of organic modifier in the mobile phase, the concentration, type and pH of the buffer on the electroosmotic mobility, the retention behavior of test analytes, and the column efficiency. In their studies, they used a group of PAHs as a test mixture.

Consequently, they were able to optimize the separation of this group of analytes by varying all of the operating parameters identified above.

The separation of charged analytes was also examined on porous-layer open- tubular capillaries [146]. The preparation of the stationary phase was performed by use of the sol-gel technique. This technique is discussed in detail later in this section. The successful separation of acidic diuretics and N-alkylanilines as basic compounds at low pH indicated that it is possible to separate charged compounds on porous-layer open-tubular columns. Neutral aryl alkyl ketones were also successfully separated with high efficiencies. However, basic pharmaceutical drugs demonstrated severe peak tailing and were poorly separated by use of these columns. In addition, strongly acidic analytes were not detected due to their high electrophoretic mobilities.

1.3.2.4 Chemical Bonding After Etching

Etching is another approach to overcoming the major problems associated with OT-

CEC. The etching process increases the surface area of the capillary by a factor of up to

1000 [147]. Therefore, more stationary phase can be attached to the wall, which in turn, increases the loading capacity of the column. In addition, during the etching process,

32 dissolution and redeposition of silica material occurs. This creates radial extensions from the wall that decrease the distance an analyte has to travel in order to interact with the stationary phase [88]. The etching process was first introduced by Onuska et al. [147] for gas chromatography (GC), and it was then modified for OT-CEC by Pesek et al. [14, 39,

41, 88-91, 148-156].

The entire procedure includes etching followed by chemical bonding. The etching process uses ammonium hydrogen difluoride as the etching agent which, under controlled conditions of temperatute and reaction time, can increase the surface area. The method of bonding an organic moiety to the etched surface of a capillary utilizes the silanization/hydrosilation method. In the first step, which involves silanization, the etched surface is reacted with triethoxysilane (TES) to produce a hydride layer that is deposited on the surface. In the second step, an organic moiety is attached to the hydride intermediate via hydrosilation. This is accomplished by passing a solution that contains a terminal alkene and a catalyst, typically hexachloroplatinic acid, through the capillary [92].

In a 1997 study, Pesek et al. [148] attached two organic moieties, octadecyl and diol, to the etched capillary wall. Peptide (angiotensins) and protein samples were used for comparison studies between these two columns and a bare capillary. The migration times for each of the analytes increased in the following order: bare

33 modified capillaries varied due to differences in the chemical properties of the two organic moieties.

Another study from Pesek et al. focused on a comparison of the migration behavior of two antibiotics (ampicillin and gentamycin) between several types of etched open- tubular capillaries [90]. The types of columns used included (i) etched and unmodified, (ii) etched and bonded with an octadecyl moiety, (iii) etched and bonded with a cholesterol moiety, (iv) etched and coated with anionic fluorosurfactant (FSA), and (v) etched and coated with neutral zwitterionic fluorosurfactant (FSK). Each of the antibiotic samples consisted of several components, and the resolution capabilities of the different types of columns were compared. Their studies showed that each type of column gave a different elution pattern due to the stationary phase effect. A number of variables had to be optimized in order to get the maximum separation for these samples. Optimization was achieved by varying several experimental conditions, such as the type of stationary phase, the pH and composition of the buffer, and the applied voltage.

Matyska et al. [89] investigated the performance characteristics of n-octadecyl and cholesterol- modified capillaries for the analyses of synthetic peptides. The results indicated that the resolution and selectivity of peptides could be affected by varying the type of chemically bonded group and the nature of the surface chemistry used to modify the capillary wall. In addition, other experimental variables, such as the type and percentage of organic solvent modifier, and the pH of the buffer system can affect the retention of peptides. For the evaluation of the reproducibility and the stability of both types of open-tubular columns, four batches were prepared over a period of two years. For this study, a mixture of two small basic compounds, serotonin and tryptamine, and a

34 mixture of two basic proteins, chicken and turkey lysozyme, were used. The measurement of the selectivities and efficiencies of these analytes showed that both the reproducibility and stability were excellent. All of the etched modified capillaries used in this study still performed well after more than 200 runs.

1.3.2.5 Sol-Gel Technique

As noted earlier, the preparation procedure for columns with expansive surface areas and large loading capacities involves two steps: (i) etching or laying down a porous layer, and (ii) attaching functional groups to the intermediate layer by chemical bonding

[30]. However, this two-step procedure may be time consuming. Another approach, which involves the construction of polymeric stationary phases, involves good column stability and high surface ratio. However, the main problem associated with these columns is the poor column efficiency as a result of the slow diffusivity of analytes in the layers [145]. In order to circumvent these drawbacks, the sol-gel technique was introduced for the synthesis of organic-inorganic hybrid materials that are used as stationary phases in OT-

CEC. Recently, the sol-gel technique, which combines the synthesis of a bonded phase and a supporting porous silica film in a single step, has gained much attention [29, 40, 145,

146, 157-164].

The advantages of the sol-gel processed stationary phases include high stability, high mass loadability, great column efficiency, large surface area leading to higher retentions, and a relatively simple preparation procedure [157, 159-161]. The sol-gel process consists of five major steps: (i) hydrolysis, (ii) condensation and polycondensation of sol-gel precursors, (iii) casting of the sol, (iv) aging, and (v) drying. In the first step, a metal alkoxide is hydrolyzed under acidic or basic conditions to form the corresponding

35 metal hydroxide, which condenses and polycondenses to form “sol” particles. These “sol” particles are then crosslinked to form a wet “gel,” which can take any desired shape. A dry

“gel,” or a “xerogel” is finally formed after the loss of solvent through aging and drying.

By controlling different processing parameters, such as temperature, pressure, alcohol solvent, and pH during the different steps of the process, one can fabricate sol-gel derived materials with specific structures [15, 157, 160-162].

In 1995, Guo and Colon [145] introduced the organosilicon sol-gel fabricated coatings in OT-CEC. They synthesized a porous glass film in such a way that the stationary phase (octyl groups) is incorporated into the glass matrix during the glass formation process. This film was then used in capillaries, and functioned as a stationary phase. The performance of this sol-gel derived stationary phase was evaluated by use of a group of PAHs. Its stability was also studied under acidic and basic conditions. In order to test the stability at low and high pH values, the capillary was washed with 1% trifluoroacetic acid (pH ≈ 0.3) and with the mobile phase methanol/1 mM phosphate buffer

(70:30) at pH 11.4. Their results indicated that a retentive layer was still on the surface of the capillary, even after exposure to high- and low-pH conditions. Finally, the mass loadability of the column was evaluated by injecting different concentrations of naphthalene in the capillary. The fact that the efficiency started to degrade at a concentration of 100 mM indicated overloading. In a conventional column, the efficiency deteriorated at concentrations below 10 mM.

Rodriguez and Colon [162] used n-octadecyltriethoxysilane (C18-TEOS) and tetraethoxysilane (TEOS) as the precursors to fabricate an organic-inorganic hybrid material by use of the sol-gel method. Fused-silica capillaries were coated with the C18-

36 TEOS/TEOS mixture and were tested using OT-CEC. A test mixture containing toluene, ethylbenzene, biphenyl, dimethylnaphthalene and amybenzene was separated by using two columns coated with (i) TEOS, and (ii) C18-TEOS/TEOS sol-gel derived materials.

Additionally, eight PAHs were separated by use of a C18-TEOS/TEOS-coated column.

These compounds were baseline separated only in the columns containing the organically modified composite (C18-TEOS/TEOS).

A novel sol-gel approach for fabrication of open-tubular columns was described by

Hayes and Malik [158]. A surface-bonded octadecylsilane (ODS) stationary phase coating was created by sol-gel chemistry. The use of a deactivating reagent, phenyldimethylsilane

(PheDMS), in the sol-gel solution was evaluated. The performance of both nondeactivated and deactivated sol-gel columns was also evaluated with the test mixtures of PAHs, benzene derivatives, and aromatic aldehydes and ketones. The column deactivation resulted in a higher separation efficiency and better resolution. In addition, the RSD values for the run-to-run reproducibility of the sol-gel columns were less than 0.70% and 0.80% for the aromatic hydrocarbons and the aromatic carbonyl compounds, respectively.

Wang et al. [161] used 1,4,7,10-tetraazacyclotridecane-11,13, dione

(dioxo[13]aneN4) for the first time in the sol-gel approach for the preparation of dioxo[13]aneN4 modified capillaries. In comparison with columns prepared by the sol-gel process with just TEOS, the sol-gel derived macrocyclic dioxopolyamine columns were able to give better separations of a mixture of isomeric nitrophenols and benzenediols, a mixture of isomeric aminophenols and diaminobenzenes, and a mixture of four neurotransmitters. The reproducibilities of migration times and plate numbers were

37 satisfactory with RSD values of less than 2% and less than 8.5% for the respective run-to- run and column-to-column reproducibilities.

1.4 Chirality

Chirality is the geometric property that is responsible for the nonidentity of an object with its mirror image [165]. In other words, an object is chiral if it is non- superimposable on its mirror image. In 1848, Pasteur [166] marked the beginning of chirality and chiral separation. He discovered that the resolution of the racemic mixture of ammonium sodium tartrate yielded two enantiomorphic crystals. The separation was perfomed by use of a pair of tweezers and a hand lens. He also observed that the crystals gave a left and a right rotation of polarized light. In addition, he proposed that since the difference of the optical rotation was examined in solution, the molecules are the ones that are mirror images of each other [167].

Chiral compounds are molecules that relate to each other like a pair of hands. The word chiral was derived from the Greek word cheir, which means hand. The chiral molecules are also termed optically active because they have the ability to rotate plane- polarized light. Optically active molecules that rotate light to the left are called levorotatory (L), and they have the negative sign (-). If the molecules rotate the light to the right, they are said to be dextrorotatory (D) or positive (+). In addition, chiral compounds are also termed as molecules with non-superimposable mirror images. Molecules that are mirror images of each other are also called enantiomers or optical isomers. These molecules have an asymmetric tetrahedral carbon atom with four different substituents. If the priority of the substituents is in a clockwise direction, the configuration is called R

(right or rectus). If the priority is in a counterclockwise direction, the configuration of the

38 chiral center is S (left or sinister). These differences are due to the asymmetric element in the chiral molecule that can be a center, an axis, or a plane of asymmetry. An equimolar mixture of two enantiomers is called a racemate, and it is designated by the symbol (±) [6,

165, 168].

Although enantiomers rotate plane-polarized light in opposite directions, these molecules have identical physical properties such as boiling and melting points, and spectroscopic properties such as nuclear magnetic resonance (NMR) spectra. However, enantiomers in a racemic drug usually have different biological activities. For example, dextromethorphan is an over-the-counter cough suppressant, whereas levomethorphan is a controlled narcotic. Thalidomide is a sedative drug that was prescribed to pregnant women in the early 1960’s. When this drug was taken during the early stages of pregnancy, it prevented the normal growth of the fetus. This resulted in serious birth defects in thousands of children around the world. It was later observed that the R-enantiomer was a sedative, whereas the S form was the one that caused the foetal abnormalities. The examples mentioned above illustrate the different pharmakokinetic characteristics and pharmacological activities of each enantiomer in a racemic drug. Therefore, the development of analytical techniques for the separation of chiral bioorganic molecules has become very important [6, 168, 169].

In order to distinguish between two enantiomers in a racemic drug, a chiral selector has to be added to the BGE solution. However, the mechanism of chiral discrimination is not very well understood. In general, the “three point rule,” which was illustrated by

Easson and Stedman [170], describes the interactions that are necessary for chiral discrimination. Chiral separation can be achieved if a minimum of three simultaneous

39 interactions occur between the chiral selector and one of the enantiomers. Due to spatial restrictions, the other enantiomer should attain at least two of these interactions. These interactions can be hydrophobic interactions between the hydrophobic core of the polymer and the analyte, and electrostatic interactions between the polar head group of the polymer and the analyte. Another type of interaction is the dipole-dipole forces, such as hydrogen bonding between the polar group of the chiral selector and the analyte. In addition, other secondary interactions can occur, such as π-π interactions, ion-dipole bonds, and Van der

Waals forces.

1.5 Chiral Selectors in Open-Tubular Capillary Electrochromatography

Chiral separation has received considerable attention in many industries, particularly the pharmaceutical industry. Several analytical techniques have already been developed for the separation of chiral bioorganic molecules. One of them is OT-CEC, where the chiral selector is coated and immobilized onto the inner surface of a capillary. In

1992, Mayer and Schurig [171] reported the first enantiomeric separation in OT-CEC by using capillaries coated with immobilized Chirasil-Dex (permethyl-β-cyclodextrin chemically linked to dimethylpolysiloxane). In addition to derivatized cyclodextrins [164,

171-178], cellulose [55, 60], proteins [42, 105-107, 179-183], molecular imprinted polymers (MIPs) [184-195], and polymeric surfactants [121] were also used.

1.5.1 Derivatized Cyclodextrins

Mayer et al. [171-174] used an immobilized Chirasil-Dex for the separation of a number of chiral compounds. Chirasil-Dex is a permethyl-β- or γ-cyclodextrin covalently bonded to a dimethylpolysiloxane via an octamethylene spacer. Chirasil-Dex modified capillaries proved to be buffer-resistant and stable under neutral and acidic conditions.

40 Schurig et al. [175, 176] also investigated the use of an immobilized Chirasil-Dex

(mono-6-O-octamethylenepermethyl-β-cyclodextrin chemically linked to dimethylpolysiloxane) in chromatography. The concept of unified enantioselective chromatography was demonstrated for the chiral separation of hexobarbital by use of GC, supercritical fluid chromatography (SFC), liquid chromatography (LC), and CEC [175].

The results from this study indicated that CEC is superior to GC, SFC and LC in regards to several parameters, which are important in chiral separation. These parameters include the chiral separation factor, α, peak resolution, Rs, and efficiency, N. In addition, Chiralsil-Dex demonstrated a long lifetime, as well as configurational and thermal stability.

A related approach was also reported by Armstrong et al. [177] for the separation of racemic mephobarbitol by OT-CEC. The capillary was coated with permethylated-β- cyclodextrin that was covalently linked to dimethylpolysiloxane. This stationary phase appeared to be stable and relatively unchanged by high temperature. Further, it could not be removed from the capillary by conventional polar and nonpolar solvents or by supercritical fluids.

Another approach for immobilization of β-cyclodextrin was developed by Pesek et al. [178]. These authors etched the capillaries with ammonium hydrogendifluoride and then modified them with a chiral selector by use of the silanization/hydrosilation method, which was described earlier. The types of selectors evaluated include lactone, β- cyclodextrin, and naphthylethylamine. For the etched cyclodextrin-modified capillary (2- hydroxy-3-methacryloyloxypropyl-β-cyclodextrin), benzodiazepines were the only class of compounds resolved among the test analytes used. They obtained partial resolution of the two isomers for oxazepam, temazepam, chlorodiazepoxide, and diazepam. The stability of

41 the column was also good since it was used for at least 200 injections with no observed changes in enantiomeric resolution.

Wang et al. [164] coated columns with 2,6-dibutyl-β-cyclodextrin (DB-β-CD) using the sol-gel technique. This new technology provided columns that demonstrated good resolution, stability, and reproducibility on separation of some positional isomers and chiral compounds. The chiral compounds used in this study were ibuprofen and binaphthol. The two isomers of both analytes gave partial resolution.

1.5.2 Cellulose

Capillaries coated with cellulose derivatives were demonstrated by Francotte et al.

[55] for the separation of a number of chiral pharmaceuticals. Their columns were prepared by coating 3,5-dimethylphenylcarbamoyl cellulose (DMPCC) or para-methylbenzoyl cellulose (PMBC) on 50 µm i.d. fused-silica capillaries. The capillaries were coated at 35-

40 °C and 0.3 mbar. A film thickness of about 0.025 µm was obtained by using a filtered solution of the cellulose derivative in dichloromethane or tetrahydrofuran (THF) for

DMPCC or PMBC, respectively. Although the enantioselectivity of the coating was excellent, the column stability was low, and its lifetime was relatively short. Unfortunately, column lifetime rarely exceeded one hundred injections. These problems were largely due to the fact that the cellulose coating was not immobilized on the inner wall of the capillary.

Recently, another study was performed by Wakita et al. [60], in order to examine the enantioseparation ability of cellulose tris(3,5-dichlorophenylcarbamate), (CDCPC), in

OT-CEC. The capillary was filled with a solution containing CDCPC derivative, styrene, dry THF, and a solution of 2,2-azobisisobutyronitrile. The column was then heated at 60

°C for 20 hours to produce copolymerization. The covalently bound CDCPC was used as a

42 chiral selector for the separations of trans-stilbeneoxide, laudanosine, etozolin, and piprozolin. The enantiomers of the first two analytes were partially resolved and the enantiomers of the last two analytes were baseline resolved on this CDCPC coated capillary.

1.5.3 Proteins

Liu et al. [105-107] prepared chiral stationary phases for OT-CEC by use of a method based upon the physical adsorption of a basic chiral selector such as protein, peptide, and amino acid on the wall of a capillary. A number of chiral compounds were successfully resolved using this system. The adsorbed protein, lysozyme, showed high chiral selectivity, and the resolution ranged from 1.74 to 2.05.

Liu, Otsuka, and Terabe [42, 180-182] studied chiral separations by OT-CEC with avidin as a chiral stationary phase. This coating was prepared by use of the physical adsorption method that was proposed by Liu et al. [105-107]. A total of sixteen enantiomeric compounds were separated with avidin adsorbed on the capillary wall [42,

180]. Among them, ten pairs of enantiomers were baseline separated, while the other enantiomers were partly resolved with resolutions of 0.8-1.2. However, due to the low phase ratio of OT-CEC, only enantiomers that have strong interactions with the stationary phase can be separated. The RSD values for the run-to-run, day-to-day, and column-to- column reproducibilities were less than 2.2%, 2.3%, and 1.1%, respectively. The column- to-column reproducibility was better than the day-to-day reproducibility due to the small loss of the stationary phase during the day-to-day reproducibility measurements. Therefore, the capillary was rinsed daily with an avidin solution for 10 min prior to experiments.

43 The feasibility of OT-CEC for quantitative analysis of enantiomers was also examined with an avidin adsorbed capillary [181]. With conventional injection, the lowest detectable concentration was 10-6 M. The limits of detection (LODs) for enantiomers were below 1%. The detection sensitivity was improved by use of the field-enhanced sample injection (FESI) technique. In this technique, a water plug is introduced hydrodynamically into the capillary inlet end, and then, the sample solution is introduced with electrokinetic injection. Using the FESI technique, the LOD was reduced to a 10-9 M level, and the enantiomeric ratio was improved to 0.3%. In addition to the sensitivity, both the efficiency and resolution were improved.

Another approach to improving the sensitivity of OT-CEC is to use an extended light path (ELP) capillary, which has a bubble cell at the detection point. A physically adsorbed avidin stationary phase was used to evaluate an ELP capillary and an etched capillary [182]. With an ELP capillary, the peak height was enhanced by a factor of as much as 10. However, the peak efficiency and resolution decreased. When an etched capillary was used, the phase ratio was slightly enhanced by a factor of 1.64, as compared to an unetched capillary.

1.5.4 Molecular Imprinted Polymers

Over the last few years, molecular imprinting has attracted considerable attention as an approach for the preparation of chiral stationary phases in OT-CEC. This is due to the high enantioselectivity and predetermined elution order that characterizes the MIPs [44].

The selectivity of the resultant MIP is predetermined by the choice of template (imprint) molecule used for the imprint preparation. Monomers are selected for their ability to interact with the template molecule by mainly electrostatic interactions, hydrogen bonding,

44 dipole-dipole interactions, and hydrophobic interactions. The template molecule can either be the analyte of interest or a structural analogue [194-196]. MIPs are prepared by polymerization of a mixture that contains a template molecule, functional monomers, a crosslinking agent, and an initiator in a nonpolar solvent (porogen). In the prepolymerization mixture, the functional monomers interact with the imprint molecule, and orient around it in a specific way to form complexes. During polymerization, in the presence of a crosslinking agent, a rigid polymer is produced. After polymerization, the imprint molecule can be removed by solvent extraction. This gives rise to a material that is filled with cavities, and one that is complementary in size, shape, and chemical functionality to the imprint molecule. These cavities enable the polymer to rebind the original template molecule [189, 191, 193-197].

Schweitz and coworkers [189-192, 194, 195] prepared chiral stationary phases by using molecular imprinting of the (R) and (S) enantiomers of propranolol. In 1997, Nilsson et al. [189] performed the enantiomeric separation of β-adrenergic antagonists by using three different capillary electrochromatographic methods. In these methods, they used different cyclodextrins added to the mobile phase, a crosslinked protein gel, and a molecularly imprinted ((R) enantiomer of propranolol) superporous polymer as chiral selectors. All three methods were able to resolve or partially resolve all β-adenergic antagonists into their enantiomers. Very rapid enantiomer separations of propranolol were also studied by the use of short super-porous monolithic MIPs [194]. MIP stationary phases were synthesized by use of an in situ photo-initiated polymerization reaction.

Propranolol enantiomers were resolved in less than 1 min. In addition, more MIP coatings were synthesized by the use of a surface-coupled radical initiator [195]. A

45 prepolymerization mixture was prepared by dissolving the template molecule (S) propranolol, the functional monomer methacrylic acid, and the crosslinking monomer

1,1,1-tris(hydroxymethyl)propane trimethacrylate in either toluene, dichloromethane, or acetonitrile. The prepolymerization mixture was then introduced into the capillary and the ends were sealed. The polymerization was performed by illuminating the capillary with a

UV lamp. The use of different solvents facilitates control of the coating, regarding its morphology and appearance. Furthermore, all MIP coatings synthesized using the different solvents, provided enantiomeric separation of propranolol when the (S) enantiomer was used as the template molecule.

Tan et al. [194] reported a method for in situ preparation of molecular imprint polymers as thin films inside 25 µm i.d. fused-silica capillaries. Methacrylic acid and 2- vinyl pyridine were used as the functional monomers. Ethylene dimethacrylate or trimethylol propane trimethacrylate was used as the crosslinker and toluene as the porogen.

Chiral separations of the enantiomers D- and L- dansyl phenylalanines were achieved in both OT-LC and OT-CEC. The resolution in OT-CEC was much higher than in OT-LC.

1.5.5 Polymeric Surfactants

In this dissertation, the chiral separations of 1,1’-binaphthyl-2,2’- dihydrogenphosphate (BNP), 1,1’-bi-2-naphthol (BOH), secobarbital, pentobarbital and temazepam are investigated by using the PEM coating [121]. However, the PEM coating procedure used in the achiral studies [92] needed to be modified in order to achieve chiral separations. PDADMAC was used as the cationic polymer, and the polymeric surfactant poly (sodium N-undecanoyl-L-leucylvalinate), poly (L-SULV), was used as the anionic polymer. Thus far, poly (L-SULV) has shown the best chiral discrimination ability for a

46 number of pharmaceutical compounds [126]. The PEM coating approach applied to chiral separations is discussed in detail in Chapter 4.

1.6 Laser Scanning Confocal Microscopy

Since the beginning of observational science, the need to observe objects smaller than those visible by eye alone has driven the development of microscopy. Conventional or

“far-field” light microscopy is the oldest form of microscopy. Very early in the development of microscopy, it was realized that the spatial resolution that could be achieved was limited by diffraction phenomena. This limit is best known as the diffraction limit, or Abbe’s limit. Abbe showed that the theoretical resolution limit for an objective lens is approximately half the wavelength, λ , of light employed. This relationship is demonstrated in the following equation:

0.61λ R = (1.16) NA where NA is the numerical aperture of the objective lens. The numerical aperture of an objective lens is a measure of the angular size of the focusing cone of light and is defined as:

NA = nsinθ (1.17) where θ is the acceptance angle of the lens, and n is the refractive index of the medium

(usually air, n=1.0) between the sample and objective lens. Therefore, the resolution in far- field optical microscopes depends on the wavelength of light used. The use of a shorter wavelength light can improve the spatial resolution. However, when ultraviolet light is used several difficulties arise. These low wavelengths are usually absorbed by atmospheric

47 gases, liquids and many of the materials from which conventional optical elements are constructed [198, 199].

In order to eliminate the problems caused by the far-field optics, and to improve the optical resolution, near-field scanning optical microscopy (NSOM) was developed. Synge, in 1928, was the first to propose the development of NSOM as a scanned probe microscopy. He proposed the use of a light source formed from a subwavelength-size aperture in a metal screen (Figure 1.11). When light passes through the backside of the aperture, fields are produced in and near the front side of the hole. Only the sample region directly beneath the aperture is illuminated. Images can be formed by moving the aperture and sample relative to one another (raster scanning). The signal is observed by the use of a single-element detector. The resolution is determined by the size of the aperture and its distance from the sample, rather than the wavelength of the light [198].

Confocal microscopy is a valuable tool for obtaining three-dimensional images of thick objects. It is widely used in the fluorescence mode for imaging biological species and in the reflection mode for imaging objects of different forms. In the fluorescence mode, the objects are stained with fluorescent dyes. Fluorescent dyes are molecules that absorb light at one wavelength and emit light at another wavelength. When these molecules absorb light at a specific absorption wavelength, the electron rises to a higher energy level, called the excited state. In this state, electrons are unstable; thus, they return to the ground state by releasing energy in the form of heat and light. This emission of energy is called fluorescence. Due to some loss in energy in the form of heat, the emitted light has less energy. Therefore, the light is emitted at a longer wavelength than the absorbed light [200,

201].

48 Incident Plane Waves

Y

Z X

Aperture << λ Metal Screen

Near Field << λ

Sample

Far Field λ

Figure 1.11 Model for the aperture-based near-field optical microscope.

Figure 1.12 is a schematic diagram of a confocal microscope. This microscope differs from a conventional optical microscope in that it images points of light rather than large volumes of light. In general, it illuminates and images the object one point at a time through a pinhole. Light from a laser source passes through a small pinhole, which acts as a spatial filter, to an objective lens. This lens focuses the light onto a small spot on the object, at the focal plane of the objective lens. The light that is reflected back from the illuminated object is collected by the objective lens, and it is partially reflected by a beam splitter. Then, it is directed at a confocal aperture, or pinhole, that is placed in front of the detector (i.e. photomultiplier tube). The pinhole rejects light that did not originate from the

49 focal plane of the microscope objective lens. Therefore, only a very small volume of light is focused at any time. This is what gives the system its confocal property [200, 202].

The laser scanning confocal microscope (LSCM) is the most widely used confocal microscope in the areas of biomedicine and material science. It works in both transmission and reflection modes. In addition, it gives higher resolution and thinner optical sections than those obtained by use of the conventional microscope. By scanning a lot of thin sections through the sample, a very clear three-dimensional image of the specimen can be produced. Scanning is important in LSCM because only a small volume is illuminated at a time, but a larger area should be collected for producing a better image. Scanning can be accomplished by either beam scanning or stage scanning. The advantage of beam scanning is that a total image can be generated from the small volumes of light without moving the specimen. During beam scanning the specimen is stationary, and it is scanned by flying spots of light. However, the advantage of stage scanning over beam scanning is that the total light intensity is much greater.

The most important components of a typical LSCM system are a light source, an objective lens, a scanning device, pinhole lens, a beam splitter, a detector, an analog signal processor, a digital signal processor, an image analyzer, and a computer. In LSCM, the light source should be a stable source of spatially coherent light. Therefore, the single mode laser is most commonly used for this purpose. For fluorescence applications, multi- line lasers, such as Ar, KrAr, and HeNe, can be used. The objective lens is used both to illuminate and receive the reflected light from the illuminated spot on the specimen. After the laser beam is converged by the objective lens and focused on the specimen, the scanner scans the focused spot in an x, y raster pattern in order to produce an image of the sample

50 point by point. The reflected light from the specimen is collected by the objective lens, and reflected off by the beam splitter to the pinhole and the detector. The pinhole is an

2nλ important component for the determination of both the axial ( R(z) = ) and lateral NA2

0.4λ ( R(x, y) = ) resolution of the microscope. When the pinhole is large, it transmits NA more light to the detector. This, in turn, generates a larger signal with lower resolution.

However, a smaller pinhole results in a higher resolution and a lower signal-to-noise ratio.

After the light passes through the pinhole, it is directed to the detector, which generates an electrical signal. Then, the electrical signal is amplified and sent to a scan converter, where it is combined with the x and y position signals in order to produce an image for display

[203-205].

1.7 Scope of Dissertation

In this dissertation, both packed and open-tubular capillary electrochromatographic methods were developed for the achiral and chiral separations of various classes of analytes. Chapter 2 is a report of studies on the packed mode of capillary electrochromatography (CEC). The goal of this work was to separate seven benzodiazepines by the use of a 40 cm packed bed of Reliasil 3 µm C18 stationary phase.

Optimal conditions were established by varying the mobile phase, the amount of organic modifier, the buffer concentration, the applied voltage, and the column temperature. In addition, the method developed was applied to the characterization of oxazepam in a standard urine sample.

51 Photodetector

Source Pinhole

Detector Pinhole

Light Source Beam Splitter

In-focus light Objective Lens

Out-of-focus light

Object Focal Plane

Figure 1.12 Schematic diagram of a confocal microscope.

In Chapter 3, the open-tubular mode of CEC, which is an alternative approach to conventional CEC, is described. In this approach, a polyelectrolyte multilayer (PEM) coating procedure was used to construct thin films on the capillary walls of fused silica capillaries. For the fabrication of this coating both positively and negatively charged polymers were utilized. Poly (diallyldimethylammonium chloride), PDADMAC, was used as the cationic polymer and poly (sodium N-undecanoyl-L-glycinate), poly (L-SUG), was

52 used as the anionic polymer. These capillaries were evaluated by use of seven benzodiazepines as analytes. The run-to-run, day-to-day, week-to-week and capillary-to- capillary reproducibilities were evaluated by computing the relative standard deviations

(RSDs) of electroosmotic flow (EOF). In addition, the chromatographic performance of the monomeric form of the polymeric surfactant was compared for the separation of these analytes.

Chapter 4 is an outline of an investigation of the chiral separations of 1,1’- binaphthyl-2,2’-dihydrogenphosphate (BNP), 1,1’-bi-2-naphthol (BOH), secobarbital, pentobarbital and temazepam by using the PEM coating approach and the polymeric surfactant poly (sodium N-undecanoyl-L-leucylvalinate), poly (L-SULV), as the chiral discriminator. However, the PEM coating procedure used in the achiral studies (Chapter 3) needed to be modified in order to achieve chiral separations. The quaternary ammonium salt PDADMAC was used as the cationic polymer, and the polymeric surfactant poly (L-

SULV) was used as the anionic polymer. Optimal conditions were established by varying the additive (sodium chloride, 1-ethyl-3-methyl-1H-imidazolium hexafluorophosphate, 1- butyl-3-methylimidazolium tetrafluoroborate) in the polymer deposition solutions, the salt concentration, the column temperature, and the bilayer number. Reproducibilities were also evaluated by using the RSD values of the EOF and the first peak (R-(+)-BNP).

Chapter 5 details an approach that utilizes open-tubular capillary electrochromatography (OT-CEC) and laser scanning confocal microscopy (LSCM) to further investigate the stability of the PEM coating after exposure to 0.1 M and 1.0 M

NaOH. The multilayer coatings used for these studies consisted of two and twenty layer pairs. The structural changes of these coatings were monitored and imaged using LSCM

53 after flushing the capillaries with 1.0 M NaOH. This technique also allowed a study of the uniformity and discontinuity of the coating. Using OT-CEC, both electroosmotic mobility and selectivity changes were measured after flushing the capillaries with 0.1 M and 1.0 M

NaOH. In addition, a correlation between flushing time and change in electroosmotic mobility and selectivity was examined.

1.8 References

1. Heiger, D. N. In High Performance Capillary Electrophoresis – An Introduction, 2nd ed.; Hupe, P., Rozing, G., McManigill, D., van de Goor, T., Swedberg, S., Jegle, U., Verhoef, M., Jarke, L., Ford, J., Krull, B., Eds.; Hewlett-Packard Company: France, 1992.

2. Terabe, S.; Otsuka, K.; Nishi, H. J. Chromatogr. A 1994, 666, 295.

3. Chiari, M.; Nesi, M.; Righetti, P. G. In Capillary Electrophoresis in Analytical Biotechnology; Righetti, P. G., Ed.; CRC Press, Inc.: Boca Raton, FL, 1996.

4. Skoog, D. A.; Holler, J. F.; Nieman, T. A. In Principles of Instrumental Analysis, 5th ed.; Sherman, M., Bortel, J., Messina, F., Eds.; Harcourt Brace College Publishers: Orlando, FL, 1998.

5. Chankvetadze, B. In Capillary Electrophoresis in Chiral Analysis; John Wiley & Sons Ltd: London, UK, 1997.

6. Williams, C. C.; Shamsi, S. A.; Warner, I. M. In Advances in Chromatography; Brown, P. R., Grushka, E., Eds.; Marcel Dekker, Inc.: New York, NY, 1997, 363- 423.

7. Yarabe, H. H.; Billiot, E.; Warner, I. M. J. Chromatogr. A 2000, 875, 179.

8. Terabe, S.; Otsuka, K.; Ichikawa, K.; Tsuchiya, A.; Ando, T. Anal. Chem. 1984, 56, 111.

9. Terabe, S.; Otsuka, K.; Ando, T. Anal. Chem. 1985, 57, 834.

10. Terabe, S.; Ozaki, K.; Otsuka, K.; Ando, T. J. Chromatogr. 1985, 332, 211.

11. Haynes, J. L.; Shamsi, S. A.;Warner, I. M. Anal. Chem. 1999, 18, 317.

12. Pretorius, V.; Hopkins, B. J., Schieke, J. D. J. Chromatogr. 1974, 99, 23.

54 13. Huang, X.; Zhang, J.; Horvath, C. J. Chromatogr. A 1999, 858, 91.

14. Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 1996, 736, 255.

15. Malik, A. Electrophoresis 2002, 23, 3973.

16. Knox, J. H.; Grant, I. H. Chromatographia 1991, 32, 317.

17. Pfeffer, W. D.; Yeung, E. E. Anal. Chem. 1990, 62, 2178.

18. Pfeffer, W. D.; Yeung, E. S. J. Chromatogr. 1991, 557, 125.

19. Hindocha, D.; Smith, N. W. J. Chromatogr. A 2000, 904, 99.

20. Chen, Z.; Hobo, T. Electrophoresis 2001, 22, 3339.

21. Liu, Z.; Otsuka, K.; Terabe, S.; Motokawa, M.; Tanaka, N. Electrophoresis 2002, 23, 2973.

22. Tsuda, T.; Matsuki, S.; Munesue, M.; Kitagawa, S., Oehi, H. J. Liq Chromatogr. & Rel. Techn. 2003, 26, 697.

23. Chen, X.; Zou, H.; Ye, M.; Zhang, Z. Electrophoresis 2002, 23, 1246.

24. Deng, Y.; Zhang, J.; Tsuda, T.; Yu, P. H.; Boulton, A. A.; Cassidy, R. M. Anal. Chem. 1998, 70, 1586.

25. Henry, C. W.; McCarroll, M. E.; Warner, I. M. J. Chromatogr. A 2001, 905, 319.

26. Thiam, S.; Shamsi, S. A.; Henry, C. W.; Robinson, J. W.; Warner, I. M. Anal. Chem. 2000, 72, 2541.

27. Colon, L. A.; Guo, Y.; Fermier, A. Anal. Chem. 1997, 69, 461A.

28. Liu, Z.; Wu, R.; Zou, H. Electrophoresis 2002, 23, 3954.

29. Wu, J. T.; Huang, P.; Li, M. X.; Qian, M. G.; Lubman, D. M. Anal. Chem. 1997, 69, 320.

30. Zou, H.; Ye, M. Electrophoresis 2000, 21, 4073.

31. Lammehofer, M.; Lindner, W. J. Chromatogr. A 1999, 839, 167.

32. Lammehofer, M.; Lindner, W. J. Chromatogr. A 1998, 829, 115.

55 33. Wu, J. T.; Huang, P.; Li, M. X.; Lubman, D. M. Anal. Chem. 1997, 69, 2908.

34. Ye, M.; Zou, H.; Liu, Z.; Ni, J.; Zhang, Y. Science in China B 1999, 42, 639.

35. Walhagen, K.; Unger, K. K.; Olsson, A. M.; Hearn, M. T. W. J. Chromatogr. A 1999, 853, 263.

36. Zou, H.; Ye, M. Electrophoresis 2000, 21, 4073.

37. Paces, M.; Kosek, J.; Marek, M.; Tallarek, U.; Seidel-Morgenstern, A. Electrophoresis 2003, 24, 380.

38. Wistuba, D.; Schurig, V. Electrophoresis 2000, 21, 3152.

39. Pesek, J. J.; Matyska, M. T.; Sentellas, S.; Galceran, M. T.; Chirai, M.; Pirri, G. Electrophoresis 2002, 23, 2982.

40. Constantin, S.; Freitag, R. J. Chromatogr. A 2000, 887, 253.

41. Matyska, M. T.; Pesek, J. J.; Katrekar, A. Anal. Chem. 1999, 71, 5508.

42. Liu, Z.; Otsuka, K.; Terabe, S. J. Sep. Sci. 2001, 24, 17.

43. Jinno, K.; Sawada, H.; Trend in Anal. Chem. 2000, 19, 664.

44. Wistuba, D.; Schurig, V. J. Chromatogr. A 2000, 875, 255.

45. Tobler, E.; Lammerhofer, M.; Lindner, W. J. Chromatogr. A 2000, 875, 341.

46. Ye, M.; Zou, H.; Lei, Z.; Wu, R.; Liu, Z.; Ni, J. Electrophoresis 2001, 22, 518.

47. Tegeler, T.; Rassi, Z. E. Electrophoresis 2002, 23, 1217.

48. Huber, C. G.; Choudhary, G.; Horvath, C. Anal. Chem. 1997, 69, 4429.

49. Bailey, C. G.; Yan, C. Anal. Chem. 1998, 70, 3275.

50. Pirogov, A. V.; Buchberger, W. J. Chromatogr. A 2001, 916, 51.

51. Lim, L. T.; Zare, R. N.; Bailey, C. G.; Rakestraw, D. J.; Yan, C. Electrophoresis 2000, 21, 737.

52. Wen, E.; Asiaie, R.; Horvath, Cs. J. Chromatogr. A 1999, 855, 349.

53. Charvatova, J.; Deyl, Z.; Kasicka, V.; Kral, V. J. Chromatogr. A 2003, 990, 159.

56 54. Fujimoto, C. Electrophoresis 2002, 23, 2929.

55. Francotte, E.; Jung, M. Chromatographia 1996, 42, 521.

56. Charvatova, J.; Kral, V.; Kasicka, V. Collection Symposium Series 2001, 4, 109.

57. Pesek, J. J.; Matyska, M. T. J. Capillary Electrophoresis 1997, 4, 213.

58. Rehder, M. A.; McGown, L. B. Electrophoresis 2001, 22, 3759.

59. Tan, Z. J.; Remcho, V. T. Anal. Chem. 1997, 69, 581.

60. Wakita, T.; Chankvetadze, B.;Yamamoto, C.; Okamoto, Y. J. Sep. Sci. 2002, 25, 167.

61. Charvatova, J.; Matejka, P.; Kral, V.; Deyl, Z. J. Chromatogr. A 2001, 921, 99.

62. Sun, P.; Landman, A.; Barker, G. E.; Hartwick, R. A. J. Chromatogr. A 1994, 685, 303.

63. Mingxian, H.; Guoliang, Y.; Bradshow, J. S.; Milton, L. J. Microcol. Sep. 1993, 5, 199.

64. Guan, N.; Zeng, Z.; Wang, Y.; Fu, E.; Cheng, J. Anal. Chim. Acta 2000, 418, 145.

65. Kato, M.; Dulay, M. T.; Bennett, B.; Chen, J. R.; Zare, R. N. Electrophoresis 2000, 21, 3145.

66. Fujimoto, C.; Fujise, Y. Anal. Chem. 1996, 68, 2753.

67. Kornysova, O.; Owens, P. K.; Maruska, A. Electrophoresis 2001, 22, 3335.

68. Koide, T.; Ueno, K. J. Chromatogr. A 2000, 893, 177.

69. Gusev, I.; Huang, X.; Horvath, C. J. Chromatogr. A 1999, 855, 273.

70. Kang, J.; Wistuba, D.; Schurig, V. Electrophoresis 2002, 23, 1116.

71. Vegvari, A.; Foldesi, A.; Hetenyi, C.; Kocnegarova, O.; Schmid, M. G.; Kudirkaite, V.; Hjerten, S. Electrophoresis 2000, 21, 3116.

72. Schmid, M. G.; Grobuschek, N.; Tuscher, C.; Gubitz, G.; Vegvari, A.; Machtejevas, E.; Maruska, A.; Hjerten, S. Electrophoresis 2000, 21, 3141.

73. Chen, Z.; Hobo, T. Anal. Chem. 2001, 73, 3348.

57 74. Lammerhofer, M.; Peters, E. C.; Yu, C.; Svec, F.; Frechet, J. M. J.; Lindner, W. Anal. Chem. 2000, 72, 4614.

75. Lammerhofer, M.; Svec, F.; Frechet, J. M. J.; Lindner, W. Anal. Chem. 2000, 72, 4623.

76. Koide, T.; Ueno, K. Anal. Sci. 2000, 16, 1065.

77. Koide, T.; Ueno, K. J. Chromatogr. A 2001, 909, 305.

78. Peters, E. C.; Petro, M.; Svec, F.; Frechet, J. M. J. Anal. Chem. 1997, 69, 3646.

79. Peters, E. C.; Petro, M.; Svec, F.; Frechet, J. M. J. Anal. Chem. 1998, 70, 2288.

80. Peters, E. C.; Petro, M.; Svec, F.; Frechet, J. M. J. Anal. Chem. 1998, 70, 2296.

81. Que, A. H.; Novothy, M. V. Anal. Chem. 2002, 74, 5184.

82. Wu, R.; Zou, H.; Ye, M.; Lei, Z.; Ni, J. Electrophoresis 2001, 22, 544.

83. Wu, R.; Zou, H.; Fu, H.; Jin, W.; Ye, M. Electrophoresis 2002, 23, 1239.

84. Ping, G.; Schmitt-Kopplin, P.; Hertkorn, N.; Zhang, W.; Zhang, Y.; Kettrup, A. Electrophoresis 2003, 24, 958.

85. Legido-Quigley, C.; Marlin, N. D.; Melin, V.; Manz, A.; Smith, N. W. Electrophoresis 2003, 24, 917.

86. Baltussen, E.; van Dedem, G. W. K. Electrophoresis 2002, 23, 1224.

87. Smith, N. W.; Carter-Finch, A. S. J. Chromatogr. A 2000, 892, 219.

88. Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 2000, 887, 31.

89. Matyska, M. T.; Pesek, J. J.; Boysen, R. I.; Heam, M. T. W. Anal. Chem. 2001, 73, 5116.

90. Pesek, J. J.; Matyska, M. T.; Tran, H. J. Sep. Sci. 2001, 24, 729.

91. Jinno, K.; Sawada, H.; Catabay, A. P.; Watanabe, H.; Sabli, N. B. H.; Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 2000, 887, 479.

92. Kapnissi, C. P.; Akbay, C.; Schlenoff, J. B.; Warner, I. M. Anal. Chem. 2002, 74, 2328.

58 93. Kamande, M. W.; Kapnissi, C. P.; Zhu, X.; Akbay, C.; Warner, I. M. Electrophoresis 2003, 24, 945.

94. Xie, M. J.; Feng, Y. Q.; Da, S. L.; Meng, D. Y.; Ren, L. W. Anal. Chim. Acta 2001, 428, 255.

95. Garner, T. W.; Yeung, E. S. J. Chromatogr. 1993, 640, 397.

96. Breadmore, M. C.; Macka, M.; Avdalovic, N.; Haddad, P. R. Anal. Chem. 2001, 73, 820.

97. Breadmore, M. C.; Palmer, A. S.; Curran, M.; Macka, M.; Avdalovic, N.; Haddad, P. R. Anal. Chem. 2002, 74, 2112.

98. Breadmore, M. C.; Macka, M.; Haddad, P. R.; Avdalovic, N. Analyst 2000, 125, 1235.

99. Boyce, M. C.; Breadmore, M.; Macka, M.; Doble, P.; Haddad, P. R. Electrophoresis 2000, 21, 3073.

100. Baryla, N. E.; Melanson, J. E.; McDermott, M. T.; Lucy, C. A. Anal. Chem. 2001, 73, 4558.

101. Chiari, M.; Ceriotti, L.; Crini, G.; Morcellet, M. J. Chromatogr. A 1999, 836, 81.

102. Charvatova, J.; Kasicka, V.; Deyl, Z.; Kral, V. J. Chromatogr. A 2003, 990, 111.

103. Breadmore, M. C.; Boyce, M.; Macka, M.; Avdalovic, N.; Haddad, P. R. J. Chromatogr. A 2000, 892, 303.

104. Erim, F. B.; Cifuentes, A.; Poppe, H.; Kraak, J. C. J. Chromatogr. A 1995, 708, 356.

105. Liu, Z.; Zou, H.; Ni, J. Y.; Zhang, Y. Anal. Chim. Acta 1999, 378, 73.

106. Liu, Z.; Zou, H.; Ye, M.; Ni, J.; Zhang, Y. Chinese J. Chromatogr. 1999, 17, 245.

107. Liu, Z.; Zou, H.; Ye, M.; Ni, J.; Zhang, Y. Electrophoresis 1999, 20, 2891.

108. Cifuentes, A.; Poppe, H.; Kraak, J. C.; Erime, F. B. J. Chromatogr. B 1996, 681, 21.

109. Gilges, M.; Kleemiss, M. H.; Schomburg, G. Anal. Chem. 1994, 66, 2038.

110. Cordova, E.; Gao, J.; Whitesides, G. M. Anal. Chem. 1997, 69, 1370.

59 111. Chiu, R. W.; Jimenez, J. C.; Monnig, C. A. Anal. Chim. Acta 1995, 307, 193.

112. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 2254.

113. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 5272.

114. Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007.

115. Sawada, H.; Jinno, K. Electrophoresis 1999, 20, 24.

116. Bendahl, L.; Hansen, S. H.; Gammelgaard, B. Electrophoresis 2001, 22, 2565.

117. Kohr, J.; Engelhardt, H. J. Microcol. Sep. 1991, 3, 491.

118. Li, M. X.; Liu, L.; Wu, J. T.; Lubman, D. M. Anal. Chem. 1997, 69, 2451.

119. Harrell, C. W.; Dey, J.; Shamsi, S. A.; Foley, J. P.; Warner, I. M. Electrophoresis 1998, 19, 712.

120. Wang, J.; Warner, I. M. Anal. Chem. 1994, 66, 3773.

121. Kapnissi, C. P.; Valle, B. C.; Warner, I. M. Anal. Chem. 2003, 75, 6097.

122. Yarabe, H. H.; Billiot, E.; Warner, I. M. J. Chromatogr. A 2000, 875, 179.

123. Palmer, C. P.; Tanaka, N. J. Chromatogr. A 1997, 792, 105.

124. Palmer, C. P.; Terabe, S. Anal. Chem. 1997, 69, 1852.

125. Wang, J.; Warner, I. M. J. Chromatogr. A 1995, 711, 297.

126. Shamsi, S. A.; Valle, B. C.; Billiot, F.; Warner, I. M. Anal. Chem. 2003, 75, 379.

127. Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626.

128. Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968.

129. Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552.

130. Rmaile, H. H.; Schlenoff, J. B. Langmuir 2002, 18, 8263.

131. Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153.

132. Castelnovo, M.; Joanny, J.-F. Langmuir 2000, 16, 7524.

60 133. Huang, X.; Horvath, C. J. Chromatogr. A 1997, 788, 155.

134. Cifuentes, A.; de Frutos, M.; Santos, J. M.; Diez-Masa, J. C. J. Chromatogr. 1993, 655, 63.

135. Schmalzing, D.; Piggee, C. A.; Foret, F.; Carrilho, E.; Karger, B. L. J. Chromatogr. 1993, 652, 149.

136. Towns, J. K.; Regnier, F. E. J. Chromatogr. 1990, 516, 69.

137. Bao, J. J. J. Liq. Chromatogr. & Rel. Tecnol. 2000, 23, 61.

138. Towns, J. K.; Bao, J. M.; Regnier, F. E. J. Chromatogr. 1992, 599, 227.

139. Zeng, Z.; Xie, C.; Li, H.; Han, H.; Chen, Y. Electrophoresis 2002, 23, 1272.

140. Wu, X.; Liu, H.; Liu, H.; Zhang, S.; Haddad, P. R. Anal. Chim. Acta 2003, 478, 191.

141. Li, H. B.; Zeng, Z. R.; Xie, C. H.; Chen, Y. Y. Chromatographia 2002, 55, 591.

142. Wang, Z.; Chen, Y.; Yuan, H.; Huang, Z.; Liu, G. Electrophoresis 2000, 21, 1620.

143. Crego, A. L.; Martinez, J.; Marina, M. L. J. Chromatogr. A 2000, 869, 329.

144. Bruin, G. J. M.; Tock, P. P. H.; Kraak, J. C.; Poppe, H. J. Chromatogr. 1990, 517, 557.

145. Guo, Y.; Colon, L. A. Anal. Chem. 1995, 67, 2511.

146. Dube, S.; Smith, R. M. Chromatographia 2003, 57, 485.

147. Onuska, F. I.; Comba, M. E.; Bistricki, T.; Wilkinson, R. J. J. Chromatogr. 1977, 142, 117.

148. Pesek, J. J.; Matyska, M. T.; Mauskar, L. J. Chromatogr. A 1997, 763, 307.

149. Catabay, A. P.; Sawada, H.; Jinno, K.; Pesek, J. J.; Matyska, M. T. J. Cap. Elec. 1998, 005:1/2, 89.

150. Pesek, J. J.; Matyska, M. T.; Cho, S. J. Chromatogr. A 1999, 845, 237.

151. Pesek, J. J.; Matyska, M. T.; Swedberg, S.; Udivar, S. Electrophoresis 1999, 20, 2343.

61 152. Matyska, M. T.; Pesek, J. J.; Yang, L. J. Chromatogr. A 2000, 887, 497.

153. Matyska, M. T.; Pesek, J. J.; Sandoval, J. E.; Parkar, U.; Liu, X. J. Liq. Chrom. & Rel. Technol. 2000, 23, 97.

154. Matyska, M. T.; Pesek, J. J.; Boysen, R. I.; Hearn, M. T. W. J. Chromatogr. A 2001, 924, 211.

155. Matyska, M. T.; Pesek, J. J.; Boysen, I.; Hearn, T. W. Electrophoresis 2001, 22, 2620.

156. Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 1996, 736, 313.

157. Narang, P.; Colon, L. A. J. Chromatogr. A 1997, 773, 65.

158. Hayes, J. D.; Malik, A. Anal. Chem. 2001, 73, 987.

159. Zhao, Y.; Zhao, R.; Shangguan, D.; Liu, G. Electrophoresis 2002, 23, 2990.

160. Rodriguez, S. A.; Colon, L. A. Anal. Chim. Acta 1999, 397, 207.

161. Wang, Y. C.; Zeng, Z. R.; Xie, C. H.; Guan, N.; Fu, E. Q.; Cheng, J. K. Chromatographia 2001, 54, 475.

162. Rodriguez, S. A.; Colon, L. A. Applied Spectroscopy 2001, 55, 472.

163. Crosnier de Bellaistre, M.; Mathieu, O.; Randon, J.; Rocca, J. L. J. Chromatogr. A 2002, 971, 199.

164. Wang, Y.; Zeng, Z.; Guan, N.; Cheng, J. Electrophoresis 2001, 22, 2167.

165. Jacques, J.; Collet, A.; Wilen, S. H. In Enantiomers, Racemates, and Resolutions; Krieger Publishing Co: Malabar, FL, 1981.

166. Pasteur, L. Ann. Chim. Phys. 1848, 24, 442.

167. Feibush, B.; Grinberg, N. In Chromatographic Chiral Separations – The History of Enantiomeric Resolution, 2nd ed.; Zief, M., Crane, L. J., Eds.; Marcel Dekker, Inc.: New York, NY, 1988.

168. Ahuja, S. In Chiral Separations – Applications and Technology; American Chemical Society: Washington, DC, 1997.

169. http://chirality.ouvaton.org/homepage.htm

62 170. Easson, E. H.; Stedman, E. J. Biochem. 1933, 27, 1257.

171. Mayer, S.; Schurig, V. J. High Res. Chromatogr. 1992, 15, 129.

172. Mayer, S.; Schurig, V. J. Liq. Chromatogr. 1993, 16, 915.

173. Mayer, S.; Schurig, V. Electrophoresis 1994, 15, 835.

174. Mayer, S.; Schleimer, M.; Schurig, V. J. Microcol. Sep. 1994, 6, 43.

175. Schurig, V.; Jung, M.; Mayer, S.; Fluck, M.; Negura, S.; Jakubetz, H. J. Chromatogr. A 1995, 694, 119.

176. Jakubetz, H.; Czesla, H.; Schurig, V. J. Microcol. Sep. 1997, 9, 421.

177. Armstrong, D. W.; Tang, Y.; Ward, T.; Nichols, M. Anal. Chem. 1993, 65, 1114.

178. Pesek, J. J.; Matyska, M. T.; Menezes, S. J. Chromatogr. A 1999, 853, 151.

179. Yang, J.; Hage, D. S. Anal. Chem. 1994, 66, 2719.

180. Liu, Z.; Otsuka, K.; Terabe, S. Chromatogr. 2000, 21, 302.

181. Liu, Z.; Otsuka, K.; Terabe, S. Electrophoresis 2001, 22, 3791.

182. Liu, Z.; Otsuka, K.; Terabe, S. J. Chromatogr. A 2002, 961, 285.

183. Hofstetter, H.; Hofstetter, O.; Schurig, V. J. Microcol. Sep. 1998, 10, 287.

184. Lin, J. M.; Uchiyama, K.; Hobo, T. Chromatographia 1998, 47, 625.

185. Quaglia, M.; de Lorenzi, E.; Sulitzky, C.; Massolini, G.; Sellergren, B. Analyst 2001, 126, 1495.

186. Bruggemann, O; Freitag, R.; Whitcombe, M. J.; Vulfson, E. N. J. Chromatogr. A 1997, 781, 43.

187. Lin, J. M; Nakagama, T.; Uchiyama, K.; Hobo, T. J. Liq. Chromatogr. & Rel. Techn. 1997, 20, 1489.

188. Lin, J. M.; Nakagama, T.; Uchiyama, K.; Hobo, T. J. Pharmac. Biomed. Analysis 1997, 15, 1351.

189. Nilsson, S.; Schweitz, L.; Peterson, M. Electrophoresis 1997, 18, 884.

63 190. Schweitz, L.; Spegel, P.; Nilsson, S. Analyst 2000, 125, 1899.

191. Schweitz, L.; Andersson, L. I.; Nilsson, S. Anal. Chem. 1997, 69, 1179.

192. Spegel, P.; Schweitz, L.; Nilsson, S. Electrophoresis 2001, 22, 3833.

193. Tan, Z. J.; Remcho, V. T. Electrophoresis 1998, 19, 2055.

194. Schweitz, L.; Andersson, L. I.; Nilsson, S. Anal. Chim. Acta 2001, 435, 43.

195. Schweitz, L. Anal. Chem. 2002, 74, 1192.

196. de Boer, T.; Mol, R.; de Zeeuw, R. A.; de Jong, G. J.; Sherrington, D. C.; Cormack, P. A. G.; Ensing, K. Electrophoresis 2002, 23, 1296.

197. Quaglia, M.; de Lorenzi, E.; Sulitzky, C.; Caccialanza, G.; Sellergren, B. Electrophoresis 2003, 24, 952.

198. Higgins, D. A.; Mei, E. In Near-Field Scanning Optical Microscopy; Bonnell, D., Eds.; Wiley-VCH, Inc: New York, NY, 2001, 371-420.

199. http://www.jasco.co.uk/nearfield.htm

200. Sheppard, C. J. R. In Multidimensional Microscopy; Cheng, P. C., Lin, T. H., Wu, W. L., Wu, J. L., Eds.; Springer-Verlag New York Inc.: New York, NY, 1994, 1-31.

201. http://dept.kent.edu/projects/cell/FLUORO.HTM

202. Rochow, T. G.; Tucker, P. A. In Introduction to Microscopy by Means of Light, Electrons, X rays, or Acoustics, 2nd ed.; Plenum Press: New York, NY, 1994.

203. Sheppard, C. J. R. In Multidimensional Microscopy; Cheng, P. C., Lin, T. H., Wu, W. L., Wu, J. L., Eds.; Springer-Verlag New York Inc.: New York, NY, 1994, 53-71.

204. Corle, T. R.; Kino, G. S. In Confocal Scanning Optical Microscopy and Related Imaging Systems; Academic Press: San Diego, CA, 1996.

205. http://www.cs.ubc.ca/spider/ladic/intro.html

64

CHAPTER 2.

SEPARATION OF BENZODIAZEPINES BY USE OF CAPILLARY ELECTROCHROMATOGRAPHY

2.1 Introduction

Benzodiazepines are effective medicinal compounds, which are used primarily for the treatment of anxiety and sleep disorders [1]. In fact, these drugs are among the most widely prescribed of all psychoactive drugs. They are important in forensic toxicology having hypnotic, tranquillizing and anticonvulsant properties. Therefore, they are often encountered in casework involving road traffic offences or drug overdose [2-4]. They enter the brain rapidly and work by binding to a specific type of receptor protein, which is widely distributed in groups of nerve cells involved in anxiety, memory, sedation, and coordination. In recent years, there has been a growing interest regarding the side effects of benzodiazepines. These effects include dizziness, adverse interaction with alcohol, and risk of dependence after long-term use [5]. For these reasons, the analysis of such compounds is vital to areas such as pharmaceutical analysis, therapeutic drug monitoring, and forensic toxicology [6-8]. A variety of methods have been developed for the separation of benzodiazepines, but only some of them have been applied to their identification and determination in complex matrices such as blood [9-21], urine [22-27] and hair [28, 29].

The methods most frequently used for controlling and monitoring these drugs in blood or urine are voltametry [30, 31], radioimmunoassay [6, 15], spectrophotometry [32], supercritical fluid chromatography (SFC) [33], thin layer chromatography (TLC) [25, 34], gas chromatography (GC) [10, 11, 26-28] and high-performance liquid chromatography

(HPLC) [2, 16-21, 34-38]. Among these techniques, GC and HPLC have been the most

65

popular due to their selectivity. However, GC analysis is complicated and time consuming, due to the need for derivatization and the thermal instability of some drugs such as oxazepam [39]. According to the literature, the separation of benzodiazepines is mainly performed by liquid chromatography (LC) using reversed-phase systems composed of silica support materials and chemically bonded alkyl chains (octyl or octadecyl) [2, 34]. In comparison with existing electrokinetic techniques such as capillary zone electrophoresis

(CZE) and micellar electrokinetic chromatography (MEKC), HPLC has low column efficiency due to the need for pressure-driven flow.

CZE and MEKC are well established for the separation of many classes of compounds. However, CZE is unable to resolve benzodiazepines, because the majority of them are neutral and are of similar hydrophobicity. MEKC has been proposed as an alternative approach. In this approach, micelles are introduced into the background electrolyte and the separation of some neutral species is achieved [40, 41]. MEKC has been successfully used to separate benzodiazepines [13, 22-24, 39, 42]. Boonkerd et al. [42] have studied the migration behavior of a series of benzodiazepines in MEKC using three kinds of surfactants, i.e., sodium dodecyl sulfate (SDS), dodecyl trimethylammonium bromide (DoTAB) and bile salts. Renou-Gonnord and David [39] have studied the effects of SDS, β-cyclodextrin (β-CD), urea, organic solvents and applied voltage on the separation of nine benzodiazepines using MEKC. Although easy to implement, MEKC lacks the selectivity and the variety of stationary phases that HPLC offers [43, 44].

Another disadvantage of MEKC is its incompatibility with mass spectrometry detection due to the high concentrations of surfactants used [45].

66

CEC is considered as an alternative approach to MEKC. It uses a stationary phase rather than a micellar pseudo-stationary phase. Solutes are separated according to their partitioning between the mobile and stationary phase and, when charged, their electrophoretic mobility [46, 47]. The mobile phase in CEC is driven by electroosmotic flow (EOF) induced by applying an electrical field over the column [45]. The advantages that CEC offers over HPLC include higher efficiency, the possibility of using small particles in beds that would create high back pressure in HPLC, and the unique selectivity due to the superimposition of chromatographic and electrophoretic effects [48].

CEC has been successfully used for the analysis of neutral drugs [49-51]. However, only a few studies have been performed on the separation of benzodiazepines using CEC

[44, 45, 52, 53]. Cahours et al. [45] have investigated the influence of temperature, ionic strength and organic modifier content on electrophoretic, chromatographic and separation performances of five benzodiazepines using CEC on a phenyl-bonded silica column. Jinno et al. [52] have compared the separation behavior of a series of benzodiazepines in packed

CEC and open-tubular CEC using a cholesteryl silica-bonded phase.

In this chapter, the influence of several experimental parameters is described in order to obtain improved selectivity and efficiency for the separation of seven benzodiazepine standards. This is accomplished by use of an octadecyl silica (ODS) stationary phase in CEC. The optimized method proved to be effective in characterizing oxazepam in a urine sample.

67

2.2 Experimental

2.2.1 Reagents and Chemicals

Oxazepam, lorazepam, nitrazepam, clonazepam, temazepam, flunitrazepam, diazepam and tris(hydroxymethyl)aminomethane·hydrochloride (Tris·HCl) were purchased from Sigma Chemical Company (St Louis, MO, USA). Acetonitrile (ACN), methanol

(MeOH) and tetrahydrofuran (THF) were obtained from Fisher (Springfield, NJ, USA).

Hydrochloric acid was purchased from Mallinckrodt & Baker (Paris, KY, USA).

Polyimide-coated fused-silica capillary columns of 100 µm i.d. × 365 µm o.d. were obtained from Polymicro Technologies (Phoenix, AZ, USA). The columns were packed with 3 µm Reliasil CEC C18 stationary phase that was purchased from Column Engineering

(Ontario, CA, USA). The DAT Multi-Drug High Urine Calibrator was donated by Earl K.

Long hospital (Baton Rouge, LA, USA).

2.2.2 Instrumentation and Conditions

The CEC experiments reported here were conducted using an HP3DCE capillary electrophoresis system (Agilent Technologies, Wilmington, DE, USA). Data were collected by use of HP3DCE Chemstation software (Agilent Technologies). The dimensions of the capillary were 48 cm × 100 µm i.d. (40 cm to the detector). All CEC separations were performed by pressurizing both the inlet and outlet buffer vials with 12 bar of nitrogen to prevent bubble formation. Unless stated otherwise, the temperature of the capillary cassette was maintained at 25 °C by the instrument thermostating system.

Detection of the analytes was performed using a photodiode array detection system set to

220 nm. All sample solutions were injected electrokinetically (15 kV for 5 s).

68

2.2.3 Sample and Buffer Preparation

Analytical standard benzodiazepine stock solutions were prepared at concentrations of 2 mg/ml in MeOH. A 400 µl aliquot of each analyte was mixed and the final concentration of each benzodiazepine in the test mixture was ∼0.3 mg/ml. A buffer solution of 100 mM Tris was prepared by dissolving the appropriate amount of Tris buffer in 10 ml of deionized water and the pH was adjusted to 8.0 using 1M HCl. An appropriate percentage of ACN, MeOH or THF (v/v) was added to an appropriate percentage of the aqueous Tris buffer solution (v/v), and then, the final volume was adjusted with deionized water depending on the mobile phase being studied. The final solution was filtered using a polypropylene nylon filter with 0.45 µm pore size and sonicated for 15 min. Finally, the urine sample was injected into the CEC capillary without any preparation.

2.2.4 Preparation and Conditioning of Packed Capillary Columns

The CEC columns were packed in our laboratory according to a standard procedure developed elsewhere [54, 55]. The 3 µm Reliasil CEC C18 silica stationary phase was slurried in acetone at a concentration of 0.2 g/ml. After sonication, 1 ml of slurry was injected through a Rheodyne injector connected to a stainless steel reservoir. The injector was connected to the pump that was a Knauer pneumatic HPLC pump (Berlin, Germany) and the slurry reservoir was connected to the capillary. The other end of the capillary was connected to a union containing a 0.5 µm frit. The pump pressure was set to 400 bar. When the capillary was filled with stationary phase, the pump was turned off and the excess slurry was removed from the reservoir. The capillary was reconnected and the pump was set back to 400 bar for two more hours. While the pump was on, the first frit was fabricated using an electrically heated Nichrome wire. The bottom union from the capillary

69

was removed and the excess stationary phase was flushed out with 200 bar pump pressure.

The last procedure was repeated after the preparation of the second frit. The detection window was placed adjacent to the outlet frit by burning off the polyimide coating. The packed capillary column was flushed with the mobile phase for one hour and then installed in a capillary cartridge. The column was further conditioned by applying both pressure of

12 bar to the inlet side and potential in 5 kV increments for 10 min up to 25 kV. Finally, both the inlet and the outlet vials were pressurized, and the voltage was set to 30 kV until the current was stabilized. This procedure was used whenever a new mobile phase was tested. Between injections, the CEC capillary columns were conditioned for 5 min using 10 kV.

2.3 Results and Discussion

The chemical structures and the numerical designations for each of the seven benzodiazepines used in this study are shown in Figure 2.1. The influence of operating parameters, such as nature and amount of organic modifier, buffer electrolyte concentration, applied voltage, and temperature was studied to optimize the CEC separation of these benzodiazepines.

2.3.1 Effect of Organic Modifier

Mobile phases comprised of Tris (10 mM, pH 8) modified with 60%, 70% and 45% of either ACN, MeOH or THF, respectively, were used to study the influence of the modifier on the separations of benzodiazepines. The organic solvent content was fixed in order to have mobile phases with the same elution strength. A nomograph, which provides the interconversion of reversed-phase mobile phases having the same strength, can be found elsewhere [56, 57]. Vertical lines in this figure intersect mobile phases having the

70

same strength. For example, 70% MeOH has the same strength as 60% ACN or 45% of

THF. The buffer Tris was chosen because its low mobility would more closely match that of the analytes, when compared to more conventional buffers such as borate and phosphate. The low mobility of Tris also allows higher concentrations of buffer to be used without significantly increasing the current. In addition, the relatively high ionic strength of the buffer leads to sharper and more defined peaks. The ionic strength of each mobile phase in this study was constant (10 mM). A pH of 8 was chosen because the EOF at this pH had a greater stability and the analysis time was shorter as compared to lower pH values.

O O O O H H H H N N N N

OH OH

N N N N Cl Cl O2N O2N

Cl Cl

1. Oxazepam 2. Lorazepam 3. Nitrazepam 4. Clonazepam

H3C H3C H3C O O O N N N

OH

N N N Cl O2N Cl

F

5. Temazepam 6. Flunitrazepam 7. Diazepam

Figure 2.1 Structures of the seven benzodiazepine analytes.

71

µ For a given capillary, the EOF ( EOF ) is defined as:

L L µ = d t EOF (2.1) Vt0

where Ld is the distance from injector to detector, Lt is the total capillary length, t0 is the migration time of the electroosmotic flow marker and V is the applied voltage. The

relative EOF can be monitored by use of the values of t0 , since all the other factors are

constant as is evident from Equation 2.1. In the studies reported here, the values of t0 were measured with MeOH.

Figure 2.2 demonstrates electrochromatograms for the separation of benzodiazepines on a C18 stationary phase using Tris-ACN (40:60), Tris-MeOH (30:70) and Tris-THF (55:45) binary mixtures. A reduction in electroosmotic mobility was observed from 1.78×10-4 for the ACN-buffer mixture to 4.64×10-5 cm2V-1s-1 for THF– buffer mixture. When ACN-buffer mixture was used, the total separation time decreased and the peak efficiencies increased at the expense of lower resolution and retention factor values. Using Tris-THF (55:45), the total separation was very long and the peak efficiencies were low. Although the analytes in the mixture were not well resolved using

Tris-ACN (40:60), they eluted in 20 min. Taking into consideration the peak efficiency and the speed of analysis, the Tris-ACN (40:60) binary mixture was used to further optimize separation conditions for the benzodiazepines.

The retention mechanism of benzodiazepines on a C18 stationary phase with a mobile phase of ACN/H2O is based on the differential partitioning of the analytes into the alkyl-bonded phase. Their retention is determined by hydrophobic interactions between the

C18 stationary phase and the nonpolar moiety of each analyte, and by interactions between

72

the polar mobile phase and the sample molecules. The migration order of benzodiazepines

1,2 3,4 5 6 7 for the ACN/H2O mobile phase was tr < tr < tr < tr < tr . However, the elution order changed when MeOH and THF were used. The elution order for the MeOH/H2O mobile

3,4 2 1 6 5 7 1,2 5 6 3 4 7 phase is tr < tr < tr < tr < tr < tr and tr < tr < tr < tr < tr < tr for the THF/H2O mobile phase.

2.3.2 Effect of Mobile-Phase Composition

In an attempt to achieve baseline separation, the content of the aqueous buffer (10 mM Tris, pH 8) was increased from 30% to 60%. As depicted in Figure 2.3, both selectivity and retention time of the analytes increase with decreasing ACN concentration.

A decrease from 70% to 40% ACN increased the selectivity between peaks 6 and 7 from

1.31 to 1.99. The increase in selectivity is due to changes in the partition coefficients as a result of the increased polarity of the mobile phase. As the mobile phase becomes more polar, the analytes partition more into the stationary phase, and they are significantly retained. As a consequence of the latter, the migration times are longer, the resolution is higher, and the efficiency is lower. A slight increase in electroosmotic mobility, at higher concentrations of ACN, was also observed from 1.57x10-4 to 3.56x10-4 cm2V-1s-1, probably due to an increase in the ratio of the dielectric constant to buffer viscosity.

2.3.3 Effect of Applied Voltage

The effect of the applied voltage on the CEC separation of benzodiazepines was then investigated using a mobile phase of 10 mM Tris (pH 8)-ACN (60:40). As expected, retention times decreased when a higher voltage was applied. Figure 2.4 demonstrates the electrochromatograms obtained when 30 kV, 20 kV and 15 kV were applied. At 30 kV, the analytes elute faster with lower resolution and higher efficiency. Although the resolution

73

between analytes 2 and 3, and 5 and 6 is higher at 15 kV, the total separation time is

longer. Based on these results, 20 kV was applied to further optimize the separation

conditions.

m A U 1,2 80 60 3,4 6 0 % A C N 40 5 6 7 20

0 20 40 60 80 100 120 min m A U 3,4 25 20 2 1 6 5 7 7 0 % M e O H 15

10 5 0 20 40 60 80 100 120 min mAU 4 5 % T H F 40 1,2 5 6 3 4 7 20 0 -20

20 40 60 80 100 120 min

Figure 2.2 Effect of the nature of organic modifier on the CEC separation of benzodiazepines. Conditions: C18 stationary phase; 40 cm packed x 100 µm i.d.; electrolyte, 10 mM Tris (pH 8)-ACN (40:60), 10 mM Tris (pH 8)- MeOH (30:70), 10 mM Tris (pH 8)-THF (55:45); applied voltage, 30 kV; electrokinetic injection, 15 kV for 5 s; temperature, 25 °C; UV detection, 220 nm.

74

m A U 1,2,3,4 A C N = 7 0 % 60 40 5 6 7 20 0

20 40 60 80 min

m A U 1,2 80 A C N = 6 0 % 60 3,4 7 40 5 6 20 0 20 40 60 80 min m A U 1 2 5,6 40 3 4 7 A C N = 5 0 % 30 20 10 0 2 0 1 2 3 4 40 60 80 min m A U 5 6 A C N = 4 0 % 7 10 5 0 20 40 60 80 min

Figure 2.3 Effect of mobile-phase composition on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the mobile phase composition (ACN-Tris) was varied.

2.3.4 Effect of Tris Concentration

The influence of the ionic strength of the mobile phase was also evaluated using 10 mM, 20 mM, and 30 mM Tris (pH 8)-ACN (60:40). At a constant ACN content, as the ionic strength increased from 10 mM to 30 mM, the retention time of the analytes increased (Figure 2.5). In addition, the electroosmotic mobility decreased from 1.76x10-4 to 1.30x10-4 cm2V-1s-1 with increasing Tris concentration due to the interactions of Tris

75

with the silica interface, which reduces the zeta potential. As mentioned in Chapter 1, the electroosmotic mobility in packed-CEC is given by:

1/ 2  ε ε RT  σ  o r  E 2cF 3 µ =   (2.2) EOF η

σ ε where , which is proportional to zeta potential, is the charge density at the surface, o is

-12 2 -1µ-2 ε the permittivity of vacuum (8.85 x 10 C N ), r is the dielectric constant of the mobile phase, R is the gas constant, T is temperature, c is the concentration of the electrolyte, F is Faraday’s constant, η is the viscosity of the mobile phase, and E is the electric field strength. According to the above equation, the electroosmotic mobility is inversely proportional to the square root of the buffer concentration. Resolution also increases upon increasing ionic strength due to improved stacking during electrokinetic injection. An increase in ionic strength from 10 mM to 30 mM increased the resolution between the analytes 5 and 6 from 1.04 to 1.39. Therefore, 30 mM Tris was used to further optimize the separation.

2.3.5 Effect of Column Temperature

For this component of our study the temperature was varied from 45 °C to 15 °C, and the binary mixture 30 mM Tris (pH 8)-ACN (60:40) was used as the mobile phase. As in CE, the electroosmotic mobility increases upon increasing temperature in CEC, due to the decrease in viscosity of aqueous-organic solvent system. When the temperature increased from 15 °C to 45 °C, the electroosmotic mobility increased from 1.24x10-4 to

1.82x10-4 cm2V-1s-1. As is also typical for liquid chromatography, retention factors, retention times, and resolution decrease at higher temperature. The effects of these

76

parameters by temperature variation are shown in Figure 2.6. It is also shown that by decreasing the temperature to 15 °C all benzodiazepines were baseline resolved.

m A U 1 23 4 12.5 V = 3 0 k V 1 0 6 7 7. 5 5 5 2. 5 0 -2.5 20 40 60 80 100 120 min m A U 1 2 3 V = 2 0 k V 30 4 5 6 7 20

10

0

-10

20 40 60 80 100 120 min 1 2 m A U 2 0 3 4 5 6 V = 1 5 k V 7 15 1 0 5 0 -5 20 40 60 80 100 120 min

Figure 2.4 Effect of applied voltage on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the applied voltage was varied; electrolyte, 10 mM Tris (pH 8)-ACN (60:40).

77

m A U 12 3 1 0 m M T r i s 30 4 5 6 7

20

10

0

-10 20 40 60 80 100 120 min 1 2 mAU 2 0 m M T r i s 3 4 5 6 7 30 20 10 0 -10 2 0 1 40 60 80 100 120 min m A U 2 3 3 0 m M T r i s 40 4 5 6 7 30 20 10 0 -10 20 40 60 80 100 120 min

Figure 2.5 Effect of Tris concentration on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the Tris buffer concentration was varied; electrolyte, Tris (pH 8)-ACN (60:40); applied voltage, 20 kV.

2.3.6 CEC Separation of Drugs from a Urine Sample

The sample used in this study was a urine calibrator that contained 1000 ng/ml oxazepam and other drugs, such as acetaminophen, amphetamines, imipramine, morphine, cocaine, etc. Without any sample preparation the urine sample was injected into the CEC capillary and the optimum conditions were applied. Figure 2.7a illustrates that the trace

78

amount of oxazepam was able to be detected and separated from the other drugs. The urine

sample was injected at 15 kV for 5 s. We also proved the presence of oxazepam in the

urine sample by spiking with the standard analyte (Figure 2.7b.). For this experiment, the

urine sample was injected at 20 kV for 10 s, and the standard analyte at 15 kV for 5 s.

m A U 2,3 T = 4 5° C 40 1 4 5,6 7 20 0

20 40 60 80 100 min 1 2 3 4 mAU 5,6 7 20 T = 3 5° C 10 0 -10 -20 20 40 60 80 100 min 1 2 3 4 mAU T = 2 5° C 30 5 6 7 20 10 0 -10 20 40 1 2 60 80 100 min m A U 3 4 T = 1 5° C 20 5 6 7 10 0 -10 20 40 60 80 100 min

Figure 2.6 Effect of column temperature on the CEC separation of benzodiazepines. Separation conditions are the same as in Figure 2.2, except the temperature was varied; electrolyte, 30 mM Tris (pH 8)-ACN (60:40); applied voltage, 20 kV.

79

? (a) m A U 60 ?

50

40

30 ? 20

10 1 0

-10 10 20 30 40 50 min

mAU (b) ?

800

600

40 0 1

200

? ? 0

0 10 20 30 40 50 mi n

Figure 2.7 CEC separation of drugs from a urine sample. (a) Without spiking. Conditions: C18 stationary phase; 40 cm packed x 100 µm i.d.; electrolyte, 30 mM Tris (pH8)-ACN (60:40); applied voltage, 20 kV; electrokinetic injection, 15 kV for 5 s; temperature, 15 °C; UV detection, 220 nm. (b) With spiking. Separation conditions are the same as above, except electrokinetic injection (urine sample), 20 kV for 10 s; electrokinetic injection (2 mg/ml standard oxazepam), 15 kV for 5 s.

80

2.4 Conclusion

In this study, we have developed a new CEC method for the baseline resolution of benzodiazepines. The optimized method proved to be effective in separating and identifying oxazepam in urine samples that contain various concentrations of other drugs.

We conclude that this CEC method is promising for determining benzodiazepines in forensic and clinical drug analysis without any sample preparation. We have also been able to understand how several parameters affect the electrophoretic and separation mechanisms of hydrophobic analytes on an ODS stationary phase. The volume fraction of ACN and the temperature are inversely proportional to migration time and resolution. However, electroosmotic mobility increases upon increasing the volume fraction of ACN or the temperature. In contrast, the ionic strength of the electrolyte is proportional to the migration time and resolution, and inversely proportional to electroosmotic mobility.

Finally, higher resolutions were obtained at 15°C when a binary mixture of 30 mM Tris

(pH 8)-ACN (60:40) was used.

2.5 References

1. Linnoila, M. In Molecular Biology to Clinical Practice; Costa, E., Ed.; Raven Press: New York, NY, 1983, 267-278.

2. Gill, R.; Law, B.; Gibbs, J. P. J. Chromatogr. 1986, 356, 37.

3. Sternbach, L. H.; Randall, L. O.; Banziger, R.; Lehr, H. In Drugs affecting central nervous system; Nutley, N. J., Ed.; Marcel Dekker, Inc.: New York, NY, 1968, 237-264.

4. Sioufi, A.; Dubois, J. P. J. Chromatogr. 1990, 531, 459.

5. Doble, A; Martin, I. L. Trends in Pharmacological Sciences 1992, 13, 76.

6. Drummer, O. H. J. Chromatogr. B 1998, 713, 201.

81

7. Smyth, W. F.; McClean, S. Electrophoresis 1998, 19, 2870.

8. Brettell, T. A.; Inman, K.; Rudin, N.; Saferstein, R. Anal. Chem. 2001, 73, 2735.

9. Tomita, M.; Okuyama, T. J. Chromatogr. B 1996, 678, 331.

10. Douse, J. M. F. J. Chromatogr. 1984, 301, 137.

11. Drouet-Coassolo, C.; Aubert, C.; Coassolo, P.; Cano, J. P. J. Chromatogr. 1989, 487, 295.

12. Evenson, M. A.; Wiktorowicz, J. E. Clin. Chem. 1992, 38, 1847.

13. Imazawa, M.; Hatanaka, Y. Journal of Pharmaceutical and Biomedical Analysis 1997, 15, 1503.

14. De Silva, J. A.; Bekersky, I.; Puglisi, C. V.; Brooks, M. A.; Weinfeld, R. E. Anal. Chem. 1976, 48, 10.

15. Dixon, W. R.; Earley, J.; Postma, E. J. Pharm. Sci. 1975, 64, 937.

16. Lloyd, J. B. F.; Parry, D. A. J. Chromatogr. 1988, 449, 281.

17. Mascher, H.; Nitsche, V.; Schutz, H. J. Chromatogr. 1984, 306, 231.

18. Klockowski, P. M.; Levy, G. J. Chromatogr. 1987, 422, 334.

19. Vree, T. B.; Baars, A. M.; Hekster, Y. A.; Van der Kleijn, E. J. Chromatogr. 1981, 224, 519.

20. Minder, E. I.; Schaubhut, R.; Minder, C. E.; Vonderschmitt, D. J. J. Chromatogr. 1987, 419, 135.

21. Aakerman, K. K.; Jolkkonen, J.; Parviainen, M.; Penttilae, I. Clin. Chem. 1996, 42, 1412.

22. Schafroth, M.; Thormann, W.; Allemann, D. Electrophoresis 1994, 15, 72.

23. Tomita, M.; Okuyama, T.; Sato, S.; Ishizu, H. J. Chromatogr. 1993, 621, 249.

24. Wernly, P.; Thormann, W. Anal. Chem. 1992, 64, 2155.

25. Inoue, T.; Niwaguchi, T. J. Chromatogr. 1985, 339, 163.

26. Mule, S. J.; Casella, G. A. J. Anal. Toxicol. 1989, 13, 179.

82

27. Maurer, H.; Pfleger, K. J. Chromatogr. 1987, 422, 85.

28. Yegles, M.; Mersch, F.; Wennig, R. Forensic Science International 1997, 84, 211.

29. Tagliaro, F.; Smyth, W. F.; Turrina, S.; Deyl, Z.; Marigo, M. Forensic Science International 1995, 70, 93.

30. Brooks, M. A.; D’Arconte, L.; Hackman, M. R.; De Silva, J. A. F. J. Anal. Toxicol. 1997, 1, 179.

31. Brooks, M. A.; Bruno, J. J. B.; De Silva, J. A. F.; Hackman, M. R. Anal. Chim. Acta 1975, 74, 367.

32. Werner, I. A.; Altorfer, H.; Perlia, X. Chromatographia 1990, 30, 255.

33. Smith, R. M.; Sanagi, M. M. J. Chromatogr. 1989, 483, 51.

34. Valko, K.; Olajos, S.; Cserhati, T. J. Chromatogr. 1990, 499, 361.

35. Mura, P.; Piriou, A.; Fraillon, P.; Papet, Y.; Reiss, D. J. Chromatogr. 1987, 416, 303.

36. Guillaume, Y.; Guinchard, C. J. Liq. Chromatogr. 1993, 16, 3457.

37. Guillaume, Y.; Guinchard, C. J. Liq. Chromatogr. 1994, 17, 1443.

38. Choma, I.; Dawidowicz, A. L.; Lodkowski, R. J. Chromatogr. 1992, 600, 109.

39. Renou-Gonnord, M. F.; David, K. J. Chromatogr. A 1996, 735, 249.

40. Nishi, H.; Tsumagari, N.; Kakimoto, T.; Terabe, S. J. Chromatogr. 1989, 465, 331.

41. Terabe, S.; Otsuka, K.; Ichikawa, K.; Tsuchiya, A.; Ando, T. Anal. Chem. 1984, 56, 111.

42. Boonkerd, S.; Detaevernier, M. R.; Vindevogel, J.; Michotte, Y. J. Chromatogr. A 1996, 756, 279.

43. Yan, C.; Dadoo, R.; Zhao, H.; Zare, R. N.; Rakesraw, D. J. Anal. Chem. 1995, 67, 2026.

44. Catabay, A. P.; Sawada, H.; Jinno, K.; Pesek, J. J.; Matyska, M. T. J. Capillary Electrophor. 1998, 5, 89.

45. Cahours, X.; Morin, P.; Dreux, M. J. Chromatogr. A 1999, 845, 203.

83

46. Thiam, S.; Shamsi, S. A.; Henry, C. W.; Robinson, J. W.; Warner, I. M. Anal. Chem. 2000, 72, 2541.

47. Henry, C. W.; McCarroll, M. E.; Warner, I. M. J. Chromatogr. A 2001, 905, 319.

48. Moffatt, F.; Cooper, P. A.; Jessop, K. M. J. Chromatogr. A 1999, 855, 215.

49. Smith, N. W.; Evans, M. B. Chromatographia 1994, 38, 649.

50. Reilly, J.; Saeed, M. J. Chromatogr. A 1998, 829, 175.

51. Wang, J.; Schaufelberger, D. E.; Guzman, N. A. J. Chromatogr. Sci. 1998, 36, 155.

52. Jinno, K.; Sawada, H.; Catabay, A. P.; Watanabe, H.; Haji Sabli, N. B.; Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 2000, 887, 479.

53. Matyska, M. T.; Pesek, J. J.; Katrekar, A. Anal. Chem. 1999, 71, 5508.

54. Knox, J. H.; Grant, I. H. Chromatographia 1987, 24, 135.

55. Knox, J. H.; Grant, I. H. Chromatographia 1991, 32, 317.

56. Schoenmakers, P. J.; Billiet, H. A. H.; De Galan, L. J. Chromatogr. 1979, 185, 179.

57. Schoenmakers, P. J.; Billiet, H. A. H.; De Galan, L. J. Chromatogr. 1981, 218, 261.

58. Dittmann, M. M.; Rozing, G. P. J. Chromatogr. A 1996, 744, 63.

84

CHAPTER 3.

ANALYTICAL SEPARATIONS USING POLYMERIC SURFACTANTS IN OPEN- TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY

3.1 Introduction

Capillary electrochromatography (CEC) is a hybrid electroseparation technique that couples the selectivity of high performance liquid chromatography (HPLC) and the separation efficiency of capillary electrophoresis (CE) [1-5]. CEC also provides high resolution, short analysis time, smaller sample and buffer consumption, and efficiencies five to ten times higher than HPLC. The separation in CEC is based upon the electrophoretic mobility of the solutes and their partitioning between the stationary and mobile phases.

In the development of CEC, both packed and open-tubular column configurations have been reported [1-10]. The packed mode of CEC utilizes a fused-silica capillary with a typical internal diameter of 50-100 µm. This capillary is packed with a typical HPLC stationary phase such as an octadecyl silica (ODS) stationary phase [2, 4]. However, there are several problems that need to be solved in order for packed-CEC to be a viable alternative to either CE or HPLC. The limitations of conventional CEC include the necessity to fabricate frits, the tendency that packed capillaries have to form bubbles around the packing material or at the frit, the difficult packing procedure, and the difficult separation of basic compounds. The problems mentioned above are discussed in detail in

Chapter 1.

Open-tubular CEC (OT-CEC) is an alternative approach to packed-CEC [9]. None of the problems mentioned above are likely to be encountered in an open-tubular format. In

85

this CEC format, a stable coating needs to be constructed on the inner walls of the capillary in order to provide efficient chromatographic separations and reproducible EOF [10]. The most commonly used approaches to wall coatings for modifying the capillary include: (i) dynamic coating performed by adding the cationic or neutral modifier to the electrolytes

[11, 12]; (ii) adsorbed cationic modifier on the capillary wall by physical adsorption [13-

16]; and (iii) fixation of the hydrophilic layer by covalent bonding and/or cross linking

[17-22]. Harrell et al. [23] achieved a baseline separation of seven tricyclic antidepressants by use of a novel nonionic micelle polymer, poly (n-undecyl-α-D-glucopyranoside) as a dynamic coating. However, dynamic coating is known to cause problems when CE is coupled to mass spectrometry (CE/MS). In addition, the presence of the nonvolatile buffer constituents may deteriorate the ionization of the analytes [24, 25]. Although physical adsorption has a simple and rapid coating procedure and good reproducibility, it has been shown to have a short lifetime and limited pH range [24, 26]. In contrast, some of the covalent bonding and/or cross-linking have a long lifetime, but require a more complicated coating procedure [24, 26]. Obviously, an ideal coating procedure would be one that is both simple and stable.

In this chapter, an alternative to covalent linking of a polymer to silica beads is explored. In our approach, the polymeric surfactant poly (sodium N-undecanoyl-L- glycinate), poly (L-SUG), is used in a simple coating procedure, which involves a layer-by- layer deposition process [27, 28]. This coating is a polyelectrolyte multilayer (PEM), and it is constructed in situ by alternating rinses of positively and negatively charged polymers

[29-33]. Via electrostatic forces, a layer of polymer adds to the oppositely charged surface, reversing the surface charge and priming the film for the addition of the next layer. Such

86

coatings have been found to be robust, and thus, highly resistant to charge and deterioration during use [24, 26, 32]. The advantages of our PEM coating are two-fold.

First, since the polymeric surfactant is coated electrostatically onto the capillary, less consumption of the reagent is required. Second, with the polymeric surfactant coated on the capillary, there is less detection interference with the analyte of interest, which in turn makes the system more amenable to coupling with mass spectrometry or other detectors where the polymeric surfactant reagent interferes.

This PEM coating approach used for fabricating columns for use in OT-CEC is described below. The performance of the modified capillaries as a separation medium is evaluated by use of seven benzodiazepines as analytes. The coating was found to be remarkably stable with excellent performance for more than 200 runs.

3.2 Experimental

3.2.1 Apparatus and Conditions

Separations were performed on a Beckman P/ACE MDQ capillary electrophoresis system with UV detection (Fullerton, CA). The fused-silica capillary, 57 cm (50 cm effective length) x 50 µm i.d., was purchased from Polymicro Technologies (Phoenix, AZ) and mounted in a Beckman capillary cartridge. Unless stated otherwise, the cartridge temperature was maintained at 25 °C by use of liquid coolant. UV detection was performed at 214 nm and the samples were injected by pressure (0.1 psi; 1 psi=6894.76 Pa) for 1 sec.

3.2.2 Reagents and Chemicals

Flunitrazepam, temazepam, diazepam, oxazepam, lorazepam, clonazepam and nitrazepam were purchased from Sigma Chemical Company (St Louis, MO). The structures of the analytes used in this study are shown in Figure 2.1 (Chapter 2). However,

87

the elution order of benzodiazepines changes to flunitrazepam1

4 5 6 7

NaH2PO4), hydrochloric acid (HCl) and sodium chloride (NaCl) were all obtained from

Fisher Scientific (Fair Lawn, NJ). Poly (diallyldimethylammonium chloride), PDADMAC

(Mw=200,000-350,000) was obtained from Aldrich (Milwankee, WI). Other chemicals, including L-glycine, undecylenic acid and N-hydroxysuccinimide, were also purchased from Sigma.

3.2.3 Sample and Buffer Preparation

Analytical standard benzodiazepine stock solutions were prepared in methanol- water (1:1) at concentrations of about 0.15 mg/ml each. A buffer solution of 50 mM

Na2HPO4 was prepared by dissolving the appropriate amount of Na2HPO4 in 10 ml of deionized water. The solution was filtered using a polypropylene nylon filter with 0.45 µm pore size and sonicated for 15 min before use.

3.2.4 Synthesis of Monomeric and Polymeric Surfactant

The surfactant monomer of sodium N-undecenoyl-L-glycinate, mono (L-SUG), was synthesized from the N-hydroxysuccinimide ester of undecylenic acid according to a previously reported procedure [34]. A 100 mM sodium salt solution of the monomer was then polymerized by use of 60Co-γ radiation. After irradiation, the polymer was dialyzed by use of a 2000 molecular mass cut-off and then lyophilized to obtain the dry product.

Structures of the monomeric and polymeric surfactants are illustrated in Figure 3.1.

3.2.5 Procedure for Polyelectrolyte Multilayer (PEM) Coating

PEM coating was achieved by deposition of the polymer solutions using the rinse function on the Beckman CE system. Each polymer deposition solution contained 0.5%

88

(w/v) polymer in 0.2 M aqueous NaCl solution. It was observed that the addition of NaCl to the polymer solution resulted in enhanced thickness for each polyelectrolyte layer [32].

The capillary was conditioned before coating using a 5-min rinse of water in order to remove any contaminants originating from the capillary drawing process. The column was then conditioned with 1 M NaOH for 60 min. Pure deionized water was flushed through the capillary for 15 more min. The first monolayer of polymer (PDADMAC) was deposited by rinsing the solution of the cationic polymer through the capillary for 20 min followed by a 5-min water rinse. All other polymer depositions were done with 5-min rinses followed by 5-min water rinses. A diagrammatic scheme of the PEM-coated capillary is shown in Figure 3.2. This diagram is not provided to give an actual structural representation of the bilayer, but only to represent the order of polymer deposition. The multilayer coatings used for the separation of benzodiazepines and the reproducibility studies consisted of ten layer pairs (a layer pair is a layer of cationic polymer plus a layer of anionic polymer; also termed a bilayer). The capillary was then flushed with buffer until a stable current was achieved. The columns were conditioned with buffer for 2 min between injections.

3.3 Results and Discussion

3.3.1 Endurance of PEM Coating

An important aspect of this approach, which must be considered, is the lifetime of the stationary phase. Thus, we examined the stability of our coating by use of the following procedure. A 10-layer pair multilayer was constructed, and nearly 50 separations were performed within 5 days. Each separation was done with applied voltages of 15 kV to 30 kV at 25 °C. The electrolyte concentration was 50 mM of Na2HPO4 and a pH range of 9.2-

89

11.0 was used. After these experiments, the capillary was removed from the instrument, and the tips of the column were placed in water vials for one week. When the capillary was placed back in the instrument, 50 replicate runs were performed using an applied voltage of 20 kV, a temperature of 25 °C, and a 50-mM phosphate buffer (pH 9.2). The capillary was again removed from the instrument and the tips were placed in water vials for an additional week, after which the column was placed back into the instrument and more than 100 runs were performed. In these studies, both the applied voltage and the temperature were varied from 15 kV to 30 kV and from 15 °C to 35 °C, respectively.

Therefore, the aggregate performance of the PEM-coated capillary was evaluated for more than 200 runs.

Na+O- (a) Na +O- (b) C O + - CO O NH Na O Na+O- C O C C O NH O C NH NH C O -ONa+ O C O C O C H NH O + - C N Na O C O O - + C ONa N C H O O HN C C O O Na+O- C HN - + C ONa C HN O C O O O C HN - + ONa C O -ONa+

Figure 3.1 Structures of the (a) monomeric SUG and (b) polymeric SUG.

90

Capillary wall, silanol groups SiO- SiO- SiO- SiO- SiO- SiO- SiO- SiO- SiO- + + + + + + N N N N N N N+ N+ N+ N+ N+ N+ N+ Cationic polymer, PDADMAC

+NaO- +Na O- +Na O- +Na O- +Na O- +Na O- +Na O- +Na O- C O C O C O C O C O C O C O C O +NaO- +NaO- +NaO- +NaO- +NaO- +NaO- +NaO- +NaO- O NH O NH O NH O NH O NH O NH O NH O NH +Na O- C O C C +NaO- C O C C +Na O- C O C C +Na O- C O C C +NaO- C O C C +Na O- C O C C +Na O- C O C C +Na O- C O C C O O O O O O O O NH O NH O NH O NH O NH O NH O NH O NH O C NH C NH C NH C NH C NH C NH C NH C NH C C C C C C C C O - ONa+ O - ONa+ O - ONa+ O - ONa+ O - ONa+ O -ONa+ O -ONa+ O - ONa+ O C O C O C O C O C O C O C O C C C C C C C C C O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O + - C N + - C N + - C N + - C N + - C N + - C N + - C N + - C N Na O C Na O C NaO C Na O C Na O C Na O C Na O C NaO C O O O O O O O O O - + O - + O - + O - + O - + O - + O - + O - + C ONa C ONa C ONa C ONa C ONa C ONa C ONa C ONa N C N C N C N C N C N C N C N C H O H O H O H O H O H O H O H O Anionic polymer, poly(L-SUG) O HN C O HN C O HN C O HN C O HN C O HN C O HN C O HN C C O C O C O C O C O C O C O C O O O O O O O O O +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C HN - + HN - + HN - + HN - + HN - + HN - + HN - + HN - + C ONa C ONa C ONa C ONa C ONa C ONa C ONa C ONa C C C C C C C C HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O O C HN O C HN O C HN O C HN O C HN O C HN O C HN O C HN -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+

+NaO- +Na O- +Na O- +Na O- +Na O- +Na O- +Na O- +NaO- C O C O C O C O C O C O C O C O + - + - + - + - + - + - + - + - O NH NaO O NH NaO O NH NaO O NH NaO O NH NaO O NH NaO O NH NaO O NH NaO +Na O- C O C C +NaO- C O C C +Na O- C O C C +Na O- C O C C +NaO- C O C C +Na O- C O C C +Na O- C O C C +Na O- C O C C O O O O O O O O NH O NH O NH O NH O NH O NH O NH O NH O C NH C NH C NH C NH C NH C NH C NH C NH C C C C C C C C O - ONa+ O - ONa+ O - ONa+ O - ONa+ O - ONa+ O -ONa+ O - ONa+ O - ONa+ O C O C O C O C O C O C O C O C C C C C C C C C O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O O H NH O + - C N + - C N + - C N + - C N + - C N + - C N + - C N + - C N Na O C Na O C NaO C Na O C Na O C Na O C Na O C Na O C O O O O O O O O O O O O O O O O C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ N C N C N C N C N C N C N C N C H O H O H O H O H O H O H O H O O O O O O O O O HN C HN C HN C HN C HN C HN C HN C HN C C C C C C C C C O O O O O O O O O O O O O O O O +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C +NaO- C HN - + HN - + HN - + HN - + HN - + HN - + HN - + HN - + C ONa C ONa C ONa C ONa C ONa C ONa C ONa C ONa C C C C C C C C HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O HN O C O O O C HN O C HN O C HN O C HN O C HN O C HN O C HN O C HN -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C -ONa+ C O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+ O -ONa+

+ + + + + + + + N N N N N N N N N+ N+ N+ N+ N+

Figure 3.2 Scheme of the PEM-coated capillary.

3.3.2 Stability of PEM Coating

Another important factor to consider when using PEM-coated capillaries is the stability of the capillary surface, especially after exposure to solutions with extreme pH values. To evaluate this parameter, two more phosphate buffers of pH 11.0 and 3.0 were prepared. First, 30 replicate runs were performed with 50 mM phosphate buffer (pH 9.2).

th -3 2 -1 -1 The 10 run (Figure 3.3a) yielded an electroosmotic mobility of 2.39x10 cm V s . The capillary was then flushed with 50 mM phosphate buffer (pH 11.0) for 100 min and 50 mM phosphate buffer (pH 9.2) for 30 min. One of the electropherograms obtained after the exposure to pH 11.0 (Figure 3.3b) gave an electroosmotic mobility of 2.37x10-3 cm2V-1s-1.

After this, the same procedure was followed with the 50 mM phosphate buffer (pH 3.0).

One of the runs that was performed after the last exposure (Figure 3.3c) yielded an electroosmotic mobility of 2.37x10-3 cm2V-1s-1. Therefore, the PEM coating was demonstrated to have extraordinary stability under extreme values of pH.

91

3.3.3 Reproducibilities

Reproducibility of the PEM coating is also an important consideration. The reproducibilities were evaluated by computing the relative standard deviations (RSDs) [36] of the EOF, which are reported in Table 3.1. The run-to-run RSD was obtained from 50 consecutive electrophoresis runs; both the day-to-day and capillary-to-capillary RSDs were obtained by use of five replicate analyses; the week-to-week RSD was obtained by use of three replicate analyses. All RSDs of the EOF were below 1%, and thus, very good reproducibilities were observed.

3.3.4 Voltage Study

The PEM-coated capillary was applied to the separation of seven benzodiazepines.

The influence of the applied voltage on the efficiency, resolution, and analysis time of the benzodiazepines was evaluated using a mobile phase of 50 mM Na2HPO4 at 25 °C. As expected, a higher voltage decreased the retention times. At 30 kV, the analytes eluted faster with higher efficiency and lower resolution. In contrast, at 15 kV, the migration times were longer, the resolution was higher, and the efficiency was lower. However, an applied voltage of 15 kV did not have a major impact on analyte resolution, compared to the electropherogram obtained when a 20 kV voltage was applied (Figure 3.4).

3.3.5 Temperature Study

The effect of temperature on the separation of benzodiazepines was also studied.

The temperature for this study was varied from 35 °C to 15 °C. As shown in Figure 3.5, the retention time decreased at higher temperature, and peak efficiency decreased at lower temperature. In addition, electroosmotic mobility increased when temperature increased, likely due to a decrease in electrolyte viscosity.

92

0.010 a 123 4 5 6 7 0.008

0.006 A U 0.004

0.002 (a) run 10

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.010

0.008

0.006 AU 0.004

0.002 (b) after exposure to pH 11.0 (run 40)

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.010

0.008

0.006 AU 0.004

0.002 (c) after exposure to pH 3.0 (run 110)

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

Figure 3.3 Stability studies of PEM coating. Conditions: 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SUG) with 0.2 M NaCl; pressure injection, 0.1 psi for 1 s; electrolyte, 50 mM Na2HPO4 (pH 9.2); applied voltage, 20 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm.

93

Table 3.1 Reproducibilities of PEM capillary coating. Conditions: same as Figure 3.3.

EOF Average (min) RSD (%)

run-to-run 9.990 0.78

day-to-day 9.982 0.81

week-to-week 9.930 0.86

capillary-to-capillary 9.950 0.93

3.3.6 Comparison Between Monomeric and Polymeric Surfactants

Another important consideration for this study is whether molecular micelles are needed to form an effective and stable PEM, or can the same be achieved by use of monomeric surfactants. In an effort to compare the chromatographic performance of mono

(L-SUG) and poly (L-SUG) for the separation of hydrophobic analytes, benzodiazepines were used as test solutes. The separation of benzodiazepines using 50 mM Na2HPO4 (pH

9.2) as the electrolyte and 0.5% (w/v) poly (L-SUG) as the anionic polyelectrolyte for the construction of the PEM coating is shown in Figure 3.6a. Figure 3.6b is the electropherogram of the benzodiazepines under the same conditions as in Figure 3.6a.

However, the anionic surfactant used for PEM coating construction in this figure is the monomeric (nonpolymerized) surfactant at the concentration of 0.5% (w/v) (19 mM).

Almost no separation is noted, even though the monomeric surfactant concentration is significantly above the normal CMC of the nonpolymerized surfactant (7 mM). Thus, it is

94

clear from this study that the molecular micelle allows better discrimination of the hydrophobic analytes than the conventional micelle.

In a normal (nonpolymerized) micellar system, the dynamic equilibrium that exists between the monomers and micellar aggregates, has been demonstrated to be a disadvantage for separations.33 This dynamic equilibrium will likely reduce the stability of the PEM coating in OT-CEC. In such a case, poor analyte separation will be observed as seen here. In contrast, polymeric micelles do not have such problems because the covalent bonds formed between monomers eliminate dynamic equilibrium. Thus, the coating that is produced will be more stable as is observed in this study.

3.4 Conclusion

A stable modified capillary has been developed by use of a simple PEM coating procedure employing a molecular micelle. Excellent run-to-run, day-to-day, week-to-week and capillary-to-capillary reproducibilities in separation were observed since the RSD values of electroosmotic flow were below 1% in all cases. In addition, the PEM-coated capillaries exhibited high stability against extreme pH values. The stability of the capillaries allowed us to perform over 200 runs. This approach also shows that highly efficient and reproducible peaks can be obtained from stationary phases prepared by such a simple procedure. In addition, the chromatographic performance of the monomeric form of the molecular micelle was compared for the separation of benzodiazepines. This study confirmed that the polymeric surfactant allows better discrimination of hydrophobic analytes than the nonpolymerized surfactant. We conclude that this method is a promising alternative to conventional CEC and should prove very useful in both chiral and achiral separations.

95

0.010

30 kV 0.008 123 4 5 6 7

0.006 AU 0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.010

0.008 20 kV

0.006 AU 0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.010

0.008 15 kV

0.006 AU 0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

Figure 3.4 Effect of applied voltage on the OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except applied voltage was varied.

96

0.006

123 4 5 6 7 35°C

0.004 AU

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.010 ° 0.008 25 C

0.006 AU 0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

0.004

15°C 0.003

0.002 AU

0.001

0.000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 Minutes

Figure 3.5 Effect of temperature on the OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except the temperature was varied.

97

0.010 (a) poly(L-SUG) 0.008 123 4 5 6 7

0.006 AU

0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 Minutes

0.010 (b) mono(L-SUG) 0.008

0.006 AU

0.004

0.002

0.000 0 2 4 6 8 10 12 14 16 18 20 Minutes

Figure 3.6 Comparison between monomeric and polymeric surfactants for OT-CEC separation of benzodiazepines. Conditions: same as Figure 3.3, except applied voltage, 30 kV. (a) 0.5% (w/v) poly (L-SUG) (b) 0.5% (w/v) mono (L-SUG).

3.5 References

1. Liu, Z.; Zou, H.; Ye, M.; Ni, J.; Zhang, Y. Electrophoresis 1999, 20, 2891.

2. Henry, C. W.; McCarroll, M. E.; Warner, I. M. J. Chromatogr. A 2001, 905, 319.

3. Ye, M.; Zou, H.; Liu, Z.; Ni, J.; Zhang, Y. Anal. Chem. 2000, 72, 616.

4. Thiam, S.; Shamsi, S. A.; Henry, C. W.; Robinson, J. W.; Warner, I. M. Anal. Chem. 2000, 72, 2541.

5. Jorgenson, J. W.; Lukacs, K. D. Anal. Chem. 1981, 53, 1298.

98

6. Jinno, K.; Sawada, H.; Catabay, A. P.; Hiroshi, W.; Sabli, N. B. H.; Pesek, J. J.; Matyska, M. T. J. Chromatogr. A 2000, 887, 479.

7. We, J.; Huang, P.; Li, M. X.; Qian, M. G.; Lubman, D. M. Anal. Chem. 1997, 69, 320.

8. Dulay, M. T.; Quirino, J. P.; Bennett, B. D.; Kato, M.; Zare, R. N. Anal. Chem. 2001, 73, 3921.

9. Matyska, M. T.; Pesek, J. J.; Katrekar, A. Anal. Chem. 1999, 71, 5508.

10. Hayes, J. D.; Malik, A. Anal. Chem. 2001, 73, 987.

11. Gilges, M.; Kleemiss, M. H.; Schomburg, G. Anal. Chem. 1994, 66, 2038.

12. Cifuentes, A.; Poppe, H.; Kraak, J. C.; Erim, F. B. J. Chromatogr. B 1996, 681, 21.

13. Erim, F. B.; Cifuentes, A.; Poppe, H.; Kraak, J. C. J. Chromatogr. A 1995, 708, 356.

14. Chiu, R. W.; Jimenez, J. C.; Monnig, C. A. Anal. Chim. Acta 1995, 307, 193.

15. Preisler, J.; Yeung, E. S. Anal. Chem. 1996, 68, 2885.

16. Cordova, E.; Gao, J.; Whitesides, G. M. Anal. Chem. 1997, 69, 1370.

17. Towns, J. K.; Regnier, F. E. J. Chromatogr. 1990, 516, 69.

18. Figeys, D.; Aebersold, R. J. Chromatogr. B 1997, 695, 163.

19. Bruin, G. J. M.; Chang, J. P.; Kuhlman, R. H.; Zegers, K.; Kraak, J. C.; Poppe, H. J. Chromatogr. 1989, 471, 429.

20. Hjerten, S.; Johansson, M. K. J. Chromatogr. 1991, 550, 811.

21. Cobb, K. A.; Dolnik, V.; Novotony, M. Anal. Chem. 1990, 62, 2478.

22. Huang, X.; Horvath, C. J. Chromatogr. A 1997, 788, 155.

23. Harrell, C. W.; Dey, J.; Shamsi, S. A.; Foley, J. P.; Warner, I. M. Electrophoresis 1998, 19, 712.

24. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 5272.

25. Niessen, W. M. A.; Tjaden, U. R.; Geef, J. J. Chromatogr. 1993, 636, 3.

99

26. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 2254.

27. Decher, G.; Schmitt, J. Prog. Colloid Polym. Sci. 1992, 89, 160.

28. Decher, G. Science 1997, 277, 1232.

29. Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626.

30. Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552.

31. Schlenoff, J. B.; Li, M. Ber. Bunsen-Ges. Phys. Chem. 1996, 100, 943.

32. Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007.

33. Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968.

34. Wang, J.; Warner, I. M. Anal. Chem. 1994, 66, 3773.

35. Dittmann, M. M.; Rozing, G. P. J. Chromatogr. A 1996, 744, 63.

36. Skoog, D. A.; Holler, J. F.; Nieman, T. A. In Principles of Instrumental Analysis, 5th ed.; Sherman, M., Bortel, J., Messina, F., Eds.; Harcourt Brace College Publishers: Orlando, FL, 1998.

100

CHAPTER 4.

CHIRAL SEPARATIONS USING POLYMERIC SURFACTANTS AND POLYELECTROLYTE MULTILAYERS IN OPEN-TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY

4.1 Introduction

The separation of chiral compounds has been of great importance in many industries, particularly the pharmaceutical industry. This interest is due to the different pharmakokinetic characteristics and pharmacological activities of each enantiomer in a racemic drug [1, 2]. Thus, there has been a great demand for the development of analytical techniques for the separation of chiral bioorganic molecules.

In recent years, different modes of capillary electrophoresis (CE) have proven to be very efficient methods for the separation and analysis of enantiomeric compounds [3-5].

Micellar electrokinetic chromatography (MEKC), which is one of the most recent methods employed for chiral separations, uses a pseudostationary phase that involves the addition of a chiral micellar selector in the background electrolyte (BGE).

In 1994, Wang and Warner reported the use of a synthetic chiral polymeric surfactant added to the BGE for chiral separations. These polymeric amino acid-based surfactants offer several advantages over conventional micelles [6-9]. As stated in Chapter

1, the polymerization of the surfactant eliminates the dynamic equilibrium between monomers and micelles. This, in turn, results in better chiral recognition and enantiomeric resolution. In addition, the polymeric surfactant can be used at very low concentrations since it does not depend on the critical micelle concentration. This usually results in higher efficiencies and rapid analysis in CE.

101

Recently, the use of polymeric surfactants with two amino acid functional groups

(dipeptides) at the polar head region has attracted considerable attention [10]. Various polymeric dipeptide chiral surfactants have been synthesized and used as chiral discriminators in MEKC [11, 12]. So far, poly (sodium N-undecanoyl-L-leucyl-valinate), poly (L-SULV), has shown the best chiral discrimination ability for a wide variety of drug compounds [8]. One major drawback to using these polymeric surfactants as additives in

BGE is that they are ultimately flushed out, and a fresh BGE is required for each separation. To circumvent this problem, alternatives to MEKC have been explored [13-19].

Open-tubular capillary electrochromatography (OT-CEC) is an alternative approach to MEKC. It requires the construction of a stable coating on the inner walls of the capillary. This, in turn, provides efficient chromatographic separations and reproducible electroosmotic flow (EOF) [20].

In 1995, Erim et al. [13] investigated the performance of a polyethyleneimine (PEI) layer as a coating for the separation of basic proteins and peptides. The PEI coating was constructed by just filling the capillary with a solution containing high-molecular-mass PEI and flushing the capillary after a certain time. Katayama et al. [14, 15] reported the use of a new simple coating procedure that is called successive multiple ionic-polymer layer coating. In their studies, they developed both anion-modified and cation-modified capillaries for the separation of acidic and basic proteins, respectively. The anion-modified capillary was prepared by first attaching the cationic polymer to the capillary wall, and then the anionic polymer to the cationic polymer layer. The cation-modified capillary was established by attaching the cationic polymer to the anionic polymer layer. In 1999, Graul and Schlenoff [16] used poly (styrene sulfonate) as the anionic polymer and poly

102

(diallyldimethylammonium chloride) as the cationic polymer in a polyelectrolyte multilayer (PEM) coating procedure, which involved an electrostatic layer-by-layer deposition process [21, 22]. This coating was constructed in situ by alternating rinses of positively and negatively charged polymers [23-30], and it was applied to the separation of a series of basic proteins. In addition, in 2003, Rmaile and Schlenoff [17] demonstrated that the use of optically active PEMs for chiral membrane separations allows very high enantiomer permeation rates with encouraging selectivity. In our laboratory, a PEM coating has been successfully employed as a separation medium for the achiral separation of benzodiazepines (Chapter 1) and phenols [18, 19].

In this chapter, the PEM coating is applied to chiral separations by the use of the anionic polymer poly (L-SULV) as the chiral discriminator. An examination of the influence of several parameters that are used to obtain more efficient and more reproducible chromatographic separations is also discussed. The PEM-coated capillary was found to be remarkably robust with a performance of more than 230 runs.

4.2 Experimental

4.2.1 Apparatus and Conditions

All experiments were conducted using a Beckman P/ACE MDQ capillary electrophoresis system with UV detection (Fullerton, CA). Fused-silica capillaries, 57 cm

(50 cm effective length) x 50 µm i.d., were purchased from Polymicro Technologies

(Phoenix, AZ). The applied voltage was 30 kV. The samples were injected by pressure.

1,1’-binaphthyl-2,2’-dihydrogenphosphate (BNP), 1,1’-bi-2-naphthol (BOH), secobarbital and pentobarbital were injected at 0.5 psi (1 psi = 6894.76 Pa) for 3 s, and temazepam at

0.9 psi for 7 s. Unless stated otherwise, the temperature of the cartridge was maintained at

103

25 °C using a liquid coolant. UV detection was performed at 214 nm for BNP, BOH, secobarbital and pentobarbital, and 220 nm for temazepam.

4.2.2 Reagents and Chemicals

The analytes BNP, BOH, secobarbital, pentobarbital and temazepam, and tris(hydroxymethyl)aminomethane (Tris) were purchased from Sigma Chemical Co. (St.

Louis, MO). Sodium borate (Na2B4O7) and sodium chloride (NaCl) were obtained from

Fisher Scientific (Fair Lawn, NJ). Boric acid (H3BO3) was purchased from Alfa Products

(Danvers, MA), and sodium phosphate (Na2HPO4) from Mallinckrodt & Baker, Inc. (Paris,

KY). The ionic liquids 1-ethyl-3-methyl-1H-imidazolium hexafluorophosphate (1E-3MI-

HFP) and 1-butyl-3-methylimidazolium tetrafluoroborate (1B-3MI-TFB) were purchased from Aldrich Chemical Co. (Milwaukee, WI) and Chemada Fine Chemicals Ltd. (Nir

Itzhak, D. N. HaNegev 85455, Israel), respectively. The dipeptide L-leucylvalinate, undecylenic acid, and N-hydroxysuccinimide were purchased from Sigma. Poly

(diallyldimethylammonium chloride) (PDADMAC; Mw = 200,000-350,000) was obtained from Aldrich.

4.2.3 Sample and Background Electrolyte Preparation

The BGE at pH 10.0 consisted of 100 mM Tris and 10 mM Na2B4O7, the BGE at pH 8.5 consisted of 25 mM Tris and 25 mM Na2B4O7, and the BGE at pH 7.2 consisted of

300 mM H3BO3 and 30 mM Na2HPO4. In all cases, the pH values were adjusted by using either 1 M sodium hydroxide (NaOH) or 1 M hydrochloric acid (HCl). All solutions were filtered using 0.45 µm polypropylene nylon filters and sonicated for 15 min before use.

The analytes BNP, BOH and secobarbital were dissolved in 50:50 methanol/water, and the analytes pentobarbital and temazepam were dissolved in 80:20 methanol/water. The final

104

analyte concentrations were 0.1 mg/ml for BNP and BOH, 0.2 mg/ml for secobarbital, and

0.5 mg/ml for pentobarbital and temazepam.

4.2.4 Synthesis of Polymeric Surfactant

The monomeric surfactant of sodium N-undecenoyl-L-leucylvalinate was synthesized from the N-hydroxysuccinimide ester of undecylenic acid using a procedure reported by Wang and Warner [7]. A 100 mM sodium salt solution of the monomer was then polymerized by use of 60Co-γ radiation. The structure of poly (L-SULV) is shown in

Figure 4.1.

O O H N C O-Na+ N C O C H

[CH-CH] x

Figure 4.1 Structure of poly (L-SULV).

105

4.2.5 Procedure for Polyelectrolyte Multilayer Coating

PEM coatings were constructed in our laboratory according to a procedure described in Chapter 3 [18]. The coatings were achieved by flushing the cationic and anionic polymer deposition solutions through the capillaries. Polymer deposition solutions contained 0.5% and 2.5% (w/v) polymer and 0.0-0.2 M NaCl or 0.01 M of the ionic liquid

1E-3MI-HFP or 1B-3MI-TFB. The capillary was first conditioned with water for 5 min, with 1 M NaOH for 60 min, and then again with water for 15 min. The first layer of the cationic polymer was deposited by flushing the PDADMAC solution through the capillary for 5 min followed by a 5-min water rinse. All polymer depositions were done with 5 min rinses followed by 5 min water rinses. The PEM coatings used for all the studies reported here consisted of two layer pairs. However, for the chiral separations of the analytes secobarbital and pentobarbital three layer pairs were constructed. Each coated capillary was then flushed with the BGE until a stable current was achieved.

4.3 Results and Discussion

Several experimental parameters such as the type of additive in the polymer deposition solutions, NaCl concentration, column temperature, and bilayer number were studied to optimize the chiral separation of BNP. The influence of each parameter on chiral separation was examined in detail as reported below. For the purpose of this work two polymeric surfactants were examined: poly (sodium N-undecanoyl-L-valinate), poly (L-

SUV), and poly (L-SULV). The use of the first polymeric surfactant in the chiral separation of BNP resulted in a single peak. Therefore, all studies were performed using the polymeric surfactant poly (L-SULV).

106

4.3.1 Effect of Additives in Polymer Deposition Solutions

Each polymer deposition solution contained a polymer and an additive. NaCl was one of the additives used for this study. It has been reported that the addition of NaCl to the polymer solution resulted in enhanced thickness for each layer [16, 25]. Several studies have also shown that salt concentration is approximately proportional to the thickness of each polyelectrolyte layer [31-33].

The ionic liquids 1E-3MI-HFP and 1B-3MI-TFB were also used as additives in the solutions used for deposition. Ionic liquids are liquid electrolytes composed entirely of ions with melting points at ambient temperature. In recent years, there has been a growing interest in ionic liquids due to their distinct properties. They are good solvents for several organic, inorganic and polymeric materials, and they have excellent chemical and thermal stability. They are also good electrical conductors, and nonvolatile [33-35]. With these properties in mind, the ionic liquids were explored for possible enhanced separations.

Figures 4.2a, 4.2b and 4.2c demonstrate the chiral separation of BNP using 0.01 M

NaCl, and the ionic liquids 1E-3MI-HFP or 1B-3MI-TFB as additives in the polymer solutions, respectively. The addition of either ionic liquid increased the resolution of BNP from 0.83 to 0.88 and 0.90. However, increasing the concentration of the ionic liquid further resulted in the formation of aggregates in the cationic polymer solution. Therefore,

NaCl was used to further optimize the separation conditions.

4.3.2 Effect of NaCl Concentration

Polymer deposition solutions composed of 0.5% (w/v) polymer and 0.00 M, 0.01

M, 0.05 M, and 0.10 M NaCl were used to study the influence of NaCl concentration on the chiral separation of BNP. As shown in Figure 4.3, the retention times of the analyte

107

peaks slightly increased when NaCl was added to the solutions, probably due to an increase in the thickness of the coating. In addition, an increase in resolution at higher concentrations of NaCl was observed. The electroosmotic mobility decreased from 3.89 x

10-4 to 3.37 x 10-4 cm2V-1s-1 with increasing NaCl concentration. Based on these results,

0.1 M NaCl was used to further optimize the chiral separation of BNP.

4.3.3 Effect of Column Temperature

The influence of column temperature was also investigated. Figure 4.4 illustrates the chiral separation of BNP at 35 °C, 25 °C and 15 °C. At a constant applied voltage, with temperature decreasing, the retention times of R-(+)-BNP and S-(-)-BNP increased mainly due to an increase in electrolyte viscosity. In addition, the electroosmotic mobility decreased from 4.24 x 10-4 to 2.79 x 10-4 cm2V-1s-1 with decreasing temperature. In Figure

4.4, it is also shown that BNP was successfully resolved at a column temperature of 15 °C.

4.3.4 Effect of Bilayer Number

The effect of the number of bilayers on the resolution and analysis time of BNP was also examined. It has been shown that an increase in the bilayer number resulted in an enhanced film thickness [27, 36]. In this study, 2, 4, 8, and 12 bilayers were constructed in the PEM coating. As shown in Figure 4.5, the analysis time, the resolution, and the selectivity increased when more bilayers were constructed. However, two and four bilayers did not demonstrate a change in selectivity, even though there was an increase in retention time.

108

0.000

O O R-(+)-BNP S-(-)-BNP (a) 0.01 M NaCl -0.002 P H O O α=1.022 AU -0.004

-0.006

-0.008

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

0.001

0.000 (b) 0.01 M 1E-3MI-HFP

-0.001 α=1.023 AU -0.002

-0.003

-0.004

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

0.000 (c) 0.01 M 1B-3MI-TFB -0.002 α=1.0235 -0.004 AU

-0.006

-0.008

-0.010 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

Figure 4.2 Chiral separation of BNP. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV); pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); applied voltage, 30 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm. (a) 0.01 M NaCl; (b) 0.01 M 1E-3MI-HFP; (c) 0.01 M 1B-3MI-TFB.

109

0.010

0.00 M NaCl 0.005

AU α=1.021 0.000

-0.005

-0.010 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

0.000

-0.002 0.01 M NaCl

AU -0.004 α=1.022

-0.006

-0.008

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

0.000

-0.002 0.05 M NaCl

AU -0.004 α=1.025 -0.006

-0.008

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

-0.001

-0.002 0.10 M NaCl AU α=1.026 -0.003

-0.004

-0.005 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

Figure 4.3 Effect of NaCl concentration on the chiral separation of BNP. Conditions: same as Figure 4.2, except NaCl concentration was varied.

110

0.002 35 °C

α=1.019 A 0.000 U

-0.002

-0.004 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

-0.001 25 °C -0.002 α=1.026 A U -0.003

-0.004

-0.005 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

0.002

15 °C 0.000 α AU =1.044

-0.002

-0.004

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Minutes

Figure 4.4 Effect of column temperature on the chiral separation of BNP. Conditions: same as Figure 4.2, except temperature was varied.

111

-0.003 2 Bilayers -0.004

AU α=1.026 -0.005

-0.006

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Minutes

0.001

0.000 4 Bilayers AU α=1.026 -0.001

-0.002

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Minutes

0.001 8 Bilayers

0.000 AU α=1.040

-0.001

-0.002

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Minutes 0.006

0.004 12 Bilayers

0.002 α AU =1.043

0.000

-0.002

0 2 4 6 8 10 12 14 16 18 20 22 24 26 Minutes

Figure 4.5 Effect of bilayer number on the chiral separation of BNP. Conditions: same as Figure 4.2, except bilayer number was varied.

112

4.3.5 Reproducibilities

Reproducibility of the PEM coating is an important factor for the evaluation of column performance. Reproducibilities were evaluated by using the relative standard deviation (RSD) [37] values of the EOF and the R-(+)-BNP peak. In both cases, the run-to- run RSD values were obtained from 10 consecutive electrophoresis runs. For each NaCl concentration added to the polymer solutions (0.00 M, 0.01 M, 0.05 M, and 0.10 M), three capillaries were coated. The RSD values of the EOF ranged from 0.13% to 0.49%, and the

RSD values of the R-(+)-BNP peak ranged from 0.24% to 0.99% (Table 4.1). Figure 4.6 is an illustration of this excellent reproducibility, and it demonstrates 10 electropherograms obtained from 10 replicate runs when 0.05 M NaCl was used.

The capillary-to-capillary reproducibilities were obtained from 30 runs (10 consecutive runs performed in 3 different capillaries for each NaCl concentration). It was observed that all RSD values were below 0.5% (Table 4.2). In addition, 10 more replicate analyses were performed in 6 consecutive days. For the first two days, the RSD values were above 0.2% (Table 4.3). However, after the second day, all RSD values dropped below 0.2%. Therefore, conditioning of the coated capillaries significantly improves the reproducibility.

4.3.6 Chiral Separation of Analytes

As shown in Figure 4.7, four more analytes were able to be resolved with resolution (Rs) values ranging from 1.07 to 1.55. BOH, which is a partially anionic analyte at pH 10, was almost baseline separated with a resolution value of 1.24 (Figure 4.7a).

Temazepam is a neutral chiral compound in the benzodiazepine class of analytes. The racemate of this analyte was successfully and easily resolved with the resolution value of

113

1.41 (Figure 4.7b). The barbiturates secobarbital and pentobarbital are partially anionic compounds in a buffer solution of pH 7.2. The enantiomeric separation of both analytes was performed using a three-bilayer PEM coating, 2.5% poly (L-SULV), and polymer deposition solutions with 0.2 M NaCl. The chiral separations of both secobarbital (Figure

4.7c) and pentobarbital (Figure 4.7d) resulted in resolution values of 1.07 and 1.55, respectively.

Table 4.1 Run-to-run reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration was varied.

NaCl Concentration (M) RSD of EOF (%) RSD of R-BNP (%)

0.13 0.40 0.00 0.18 0.38

0.21 0.24

0.13 0.30

0.01 0.26 0.39

0.26 0.54

0.29 0.65

0.05 0.35 0.96 0.49 0.99

0.15 0.35 0.10 0.30 0.61

0.32 0.90

114

0.06 0.05 M NaCl

0.04

AU 0.02

0.00

4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0 12.0 Minutes

Figure 4.6 Illustration of a run-to-run reproducibility for the chiral separation of BNP. Conditions: same as Figure 4.2, except NaCl concentration, 0.05 M.

115

Table 4.2 Capillary-to-capillary reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration was varied.

NaCl Concentration (M) RSD of the EOF (%)

0.00 0.16

0.01 0.22

0.05 0.40

0.10 0.37

Table 4.3 Reproducibilities of PEM capillary coating. Conditions: same as Figure 4.2, except NaCl concentration, 0.1 M.

RSD of the EOF (%)

Day 1 0.30

Day 2 0.23

Day 3 0.09

Day 4 0.12

Day 5 0.10

Day 6 0.10

116

-0.0028 (a) α=1.029

-0.0030

AU -0.0032 BOH

-0.0034

10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 15.0 M inutes

H 3 C (b) O 0.0004 N α=1.032 OH

Cl N

0.0003 AU Temazepam 0.0002

0.0001

11.0 11.2 11.4 11.6 11.8 12.0 12.2 12.4 12.6 12.8 13.0 13.2 13.4 13.6 13.8 14.0 14.2 14.4 14.6 M inutes

0.004 (c) 0.003 α=1.105

AU 0.002 Secobarbital 0.001

0.000 5.1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 6.0 6.1 6.2 Minutes

0.010 (d) 0.008 α=1.184

0.006 AU 0.004 Pentobarbital

0.002

0.000

5.0 5.2 5.4 5.6 5.8 6.0 6.2 6.4 6.6 6.8 7.0 7.2 7.4 Minutes

Figure 4.7 (a) Chiral separation of BOH. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L-SULV) with 0.1 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); detection, 214 nm. (b) Chiral separation of temazepam. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L- SULV) with 0.1 M NaCl; pressure injection, 0.9 psi for 7 s; background electrolyte, 25 mM Tris and 25 mM Na2B4O7 (pH 8.5); detection, 220 nm. (c) Chiral separation of secobarbital, and (d) Chiral separation of pentobarbital. Conditions for (c) and (d): 3 bilayers; 0.5% (w/v) PDADMAC and 2.5% (w/v) poly (L-SULV) with 0.2 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 300 mM H3BO3 and 30 mM Na2HPO4 (pH 7.2); detection, 214 nm. Other conditions are the same as Figure 4.2.

117

4.4 Conclusion

In this study, we were able to develop a simple PEM coating procedure for chiral separations. The anionic polymer poly (L-SULV) proved to be a good chiral discriminator for the separation of several analytes. The enantiomeric separations of BNP, BOH, and temazepam were achieved using a two-bilayer PEM coating, 0.5% poly (L-SULV), and 0.1

M NaCl in the polymer deposition solutions. On the other hand, secobarbital and pentobarbital were separated using a three-bilayer PEM coating, 2.5% poly (L-SULV), and polymer deposition solutions with 0.2 M NaCl. In addition, the endurance of the PEM coated capillary was more than 230 runs with RSD values of less than 1%. In all cases, the run-to-run and capillary-to-capillary RSD values of EOF were less than 0.5%, and the run- to-run RSD values of the R-(+)-BNP peak were less than 1%. It should be noted that the chiral separation coating had to be modified from the procedure previously used for achiral separations [18]. In our laboratory, further studies are ongoing to separate more chiral compounds, and to better understand the structure of the coating.

4.5 References

1. Wang, J.; Warner, I. M. Anal. Chem. 1994, 66, 3773.

2. Jamali, F.; Mehvar, R.; Pasutto, F. M. J. Pharm. Sci. 1989, 78, 695.

3. Snopek, J.; Jelinek, I.; Smolkova-Keulemansova, E. J. Chromatogr. 1992, 609, 1.

4. Haddadian, F.; Billiot, E. J.; Shamsi, S. A.; Warner, I. M. J. Chromatogr. A 1999, 858, 219.

5. Shamsi, S. A.; Warner, I. M. Electrophoresis 1997, 18, 179.

6. Palmer, C. P.; Terabe, S. Anal. Chem. 1997, 69, 1852.

7. Wang, J.; Warner, I. M. J. Chromatogr. 1995, 711, 297.

118

8. Shamsi, S. A.; Valle, B. C.; Billiot, F.; Warner, I. M. Anal. Chem. 2003, 75, 379.

9. Pundlett, K. L.; Armstrong, D. W. Anal. Chem. 1996, 68, 3493.

10. Shamsi, S. A.; Macossay, J.; Warner, I. M. Anal. Chem. 1997, 69, 2980.

11. Billiot, F. H.; Billiot, E. J.; Warner, I. M. J. Chromatogr. A 2001, 922, 329.

12. Billiot, E.; Warner, I. M. Anal. Chem. 2000, 72, 1740.

13. Erim, F. B.; Cifuentes, A.; Poppe, H.; Kraak, J. C. J. Chromatogr. A 1995, 708, 356.

14. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 2254.

15. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 5272.

16. Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007.

17. Rmaile, H. H.; Schlenoff, J. B. J. Am. Chem. Soc. 2003, 125, 6603.

18. Kapnissi, C. P.; Akbay, C.; Schlenoff, J. B.; Warner, I. M. Anal. Chem. 2002, 74, 2328.

19. Kamande, M. W.; Kapnissi, C. P.; Zhu, X.; Akbay, C.; Warner, I. M. Electrophoresis 2003, 24, 945.

20. Hayes, J. D.; Malik, A. Anal. Chem. 2001, 73, 987.

21. Decher, G.; Schmitt, J. Prog. Colloid Polym. Sci. 1992, 89, 160.

22. Decher, G. Science 1997, 277, 1232.

23. Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626.

24. Sui, Z.; Salloum, D.; Schlenoff, J. B. Langmuir 2003, 19, 2491.

25. Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968.

26. Caruso, F.; Lichtenfeld, H.; Donath, E.; Möhwald, H. Macromolecules 1999, 32, 2317.

27. Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552.

28. Smith, R. N.; Reven, L.; Barrett, C. J. Macromolecules 2003, 36, 1876.

119

29. Rmaile, H. H.; Schlenoff, J. B. Langmuir 2002, 18, 8263.

30. Liu, Y.; Fanguy, J. C.; Bledsoe, J. M.; Henry, C. S. Anal. Chem. 2000, 72, 5939.

31. Dubas, S. T.; Schlenoff, J. B. Macromolecules 1999, 32, 8153.

32. Lösche, M.; Schmitt, J.; Decher, G.; Bouwman, W. G.; Kjaer, K. Macromolecules 1998, 31, 8893.

33. Armstrong, D. W.; He, L.; Liu, Y. Anal. Chem. 1999, 71, 3873.

34. Dupont, J.; De Souza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002, 102, 3667.

35. Yanes, E. G.; Gratz, S. R.; Baldwin, M. J.; Robison, S. E.; Stalcup, A. M. Anal. Chem. 2001, 73, 3838.

36. Schlenoff, J. B.; Dubas, S. T. Macromolecules 2001, 34, 592.

37. Skoog, D. A.; Holler, J. F.; Nieman, T. A. In Principles of Instrumental Analysis, 5th ed.; Sherman, M., Bortel, J., Messina, F., Eds.; Harcourt Brace College Publishers: Orlando, FL, 1998.

120

CHAPTER 5.

INVESTIGATION OF THE STABILITY OF POLYELECTROLYTE MULTILAYER COATINGS IN OPEN-TUBULAR CAPILLARY ELECTROCHROMATOGRAPHY USING LASER SCANNING CONFOCAL MICROSCOPY

5.1 Introduction

In recent years, layer-by-layer deposition process of polyelectrolytes on hydrophilic surfaces via electrostatic interactions has attracted considerable attention. This method was first developed by Decher et al. [1, 2]. In this approach, a substrate is alternately exposed to solutions of positively and negatively charged polyelectrolytes. The alternate adsorption of these oppositely charged polyelectrolytes produces fairly uniform thin films that have been used in the areas of light-emitting devices, sensors, coatings, nonlinear optics, catalysis, patterning, diagnostics, bioadhesion, and drug delivery [3-6].

The layer-by-layer method has also been involved in analytical separations [7-11].

This coating involves a polyelectrolyte multilayer (PEM), constructed in a fused-silica capillary by alternating rinses of cationic and anionic polyelectrolytes [12-22]. The first layer of the cationic polymer adds to the negatively charged surface of the capillary. The construction of the cationic layer reverses the charge on the surface and primes the film for the addition of the next layer of anionic polymer. These layer-by-layer coatings have shown strong stability and high resistance to charge and deterioration during use [7, 8, 23,

24].

Katayama et al. [23, 24] established a stable modification of the inner walls of the capillary by use of the successive multiple ionic-polymer layer (SMIL) coating. Their first study focused on anion-modified capillaries that were achieved by first attaching the

121

cationic polymer to the inner walls of the capillaries and then, the anionic polymer to the cationic polymer layer [23]. The endurance of the coated capillary was more than 100 runs, and this is due to the multiple attachment of the anionic polymer. The coating was also tolerant to some organic and alkaline solvents. However, the coated capillary was unstable after 0.1 M HCl and CH3CN rinsing, and the coating was detached after 1 M NaOH rinse.

In their second study, these authors developed a cation-modified capillary by attaching the cationic polymer to the anionic polymer layer [24]. This coated capillary endured 600 replicate runs, and showed strong stability against 1 M NaOH and 0.1 M HCl with degradation ratios of less than 2%. Degradation ratios are changes in electroosmotic flow

(EOF).

Graul and Schlenoff [7] coated fused-silica capillaries with thin films of charged polymers by use of the PEM coating procedure. They demonstrated that these columns were stable to extremes of pH and ionic strength, and to dehydration/rehydration. For the stability studies against exposure to extreme pH values and ionic strengths, the 6.5-layer pair-coated capillary was flushed twice with 0.01 M (pH 12) for 10 min, and twice with

0.01 M HCl (pH 2) for 10 min.

In the study reported in this chapter, we describe an approach that uses open- tubular capillary electrochromatography (OT-CEC) and laser scanning confocal microscopy (LSCM) to further investigate the stability of the PEM coating approach that was previously described. In this approach, the polymeric surfactant poly (sodium N- undecanoyl-L-leucylvalinate), poly (L-SULV), was used in the construction of the coating.

Using OT-CEC, both electroosmotic mobility and selectivity changes were measured after flushing the capillaries with 0.1 M and 1.0 M NaOH. In addition, a correlation between

122

flushing time and change in electroosmotic mobility and selectivity was investigated. The structural changes of these coatings were also monitored and imaged after exposure to 1.0

M NaOH with LSCM. LSCM has proven to be a useful and valuable tool for characterization of packed columns, coatings, tablets, capsules, living cells, and other surfaces [25-39].

5.2 Experimental

5.2.1 Apparatus and Conditions

(i) Capillary Electrochromatography

All experiments were performed with a Beckman P/ACE MDQ capillary electrophoresis system using UV detection (Fullerton, CA). Fused-silica capillaries of 57 cm (50 cm effective length) x 50 µm i.d., were purchased from Polymicro Technologies

(Phoenix, AZ). The capillaries were thermostated at 25 °C by use of liquid coolant. The applied voltage was 30 kV, and the samples were injected by pressure at 0.5 psi (1 psi =

6894.76 Pa) for 3 s. UV detection was done at 214 nm.

(ii) Laser Scanning Confocal Microscopy

An inverted confocal microscope (Axiovert 100, LSM 510 scan module, Carl Zeiss

Inc., Thornwood, NY) was used to image the PEM-coated capillaries. The setup for imaging has been described elsewhere [25]. Briefly, a 4-cm coated capillary was placed between a microscope slide and a coverslip supported by two spacers consisting of two short pieces of fused-silica capillary. The window of the coated capillary was immersed in a refractive index matching fluid. Refractive index matching is important for obtaining minimally distorted images by reduction of refraction and scattering from the capillary walls. This arrangement is then placed on the microscope stage.

123

The imaging system was equipped with a helium-neon laser (Uniphase, 1674P) that provided excitation at 543 nm. All confocal images were obtained with a 20x (NA 0.5) dry objective. Band-pass or long-pass filters were used for selection of the spectral range of fluorescence emission. Fluorescence was directed through an adjustable pinhole to a photomultiplier tube.

5.2.2 Reagents and Chemicals

The analyte 1,1’-binaphthyl-2,2’-dihydrogenphosphate (BNP), and the buffer tris(hydroxymethyl)aminomethane (Tris) were purchased from Sigma Chemical Co. (St.

Louis, MO). Sodium borate (Na2B4O7), sodium chloride (NaCl), and sodium hydroxide

(NaOH) were obtained from Fisher Scientific (Fair Lawn, NJ). Rhodamine 6G (R6G) and

PDADMAC (Mw = 200,000-350,000) were obtained from Aldrich (Milwaukee, WI).

Refractive index matching oil (n=1.51) was purchased from Carl Zeiss Microimaging, Inc.

(Thornwood, NY). The dipeptide L-leucylvalinate, undecylenic acid, and N- hydroxysuccinimide were also obtained from Sigma.

5.2.3 Sample and Background Electrolyte Preparation

The analyte BNP was dissolved in 50:50 methanol/water. Its final concentration was 0.1 mg/ml. The background electrolyte (BGE) consisted of 100 mM Tris and 10 mM

Na2B4O7. The pH was adjusted to 10 by use of 1.0 M NaOH. The solution was filtered using 0.45 µm polypropylene nylon filters and sonicated for 15 min before use. The fluorescent dye R6G was dissolved in deionized water at concentrations of 10 µM and 100

µM.

124

5.2.4 Synthesis of Polymeric Surfactant

The monomeric surfactant of sodium N-undecenoyl-L-leucylvalinate was synthesized in our laboratory using the N-hydroxysuccinimide ester of undecylenic acid according to a procedure previously described by Wang and Warner [40]. A 100 mM sodium salt solution of the monomer was then polymerized by use of 60Co-γ radiation to form the polymeric surfactant poly (L-SULV). The structure of poly (L-SULV) is shown in

Figure 4.1 (Chapter 4).

5.2.5 Procedure for Polyelectrolyte Multilayer Coating

The procedure for construction of the PEM coating has been described in Chapters

3 and 4 [8]. The cationic and anionic polymer deposition solutions were alternately flushed through the capillary by use of the rinse function on the Beckman CE system. Both solutions contained 0.5% (w/v) polymer in 0.1 M aqueous NaCl solution. First, the fused- silica capillary was conditioned with water for 5 min, 1.0 M NaOH for 60 min, and again with water for 15 min. Then, the first layer of the polymer PDADMAC was constructed by rinsing the polymer solution through the capillary for 5 min followed by a water rinse for another 5 min. All other polymer depositions were performed by flushing the capillary with the polymer solutions for 5 min followed by 5-min water rinses. The multilayer coatings used for the OT-CEC and LSCM studies consisted of two and twenty layer pairs.

For the LSCM experiments, after each coating was constructed, the capillary was flushed with 100 µM R6G for 10 min in order to incorporate the fluorescent probe into the polymer coating. For the chromatographic experiments, after the construction of the coating, the capillary was conditioned with the BGE until a stable current was achieved.

125

5.3 Results and Discussion

5.3.1 Open-Tubular Capillary Electrochromatography

Experiments performed in OT-CEC studied the stability of capillary surface, the capillary recovery, and the coating regeneration. These parameters were evaluated by

µ α computing the electroosmotic mobility, EOF , and the selectivity factor, .

As stated in Chapter 2, the electroosmotic mobility is described by the following equation:

L L µ = d t eo (5.1) Vt0

where Ld is the distance from injector to detector; Lt is the total length of the capillary; t0 is the migration time of the EOF marker; and V is the applied voltage. Methanol was used as the EOF marker in this study since it is not retained by the PEM coating.

Selectivity is another important parameter that was used to study the stability and regeneration of the coating after exposure to NaOH. The selectivity factor of a capillary column for two species A and B, as already discussed, is defined as:

()t − t α = r B 0 (5.2) ()− tr A t0 () () where tr B and tr A are the retention times of B and A, respectively. In this study, B and

A are the enantiomers S-(-)-BNP and R-(+)-BNP, respectively.

5.3.1.1 Coating Stability

To study the stability of the capillary surface, two bilayers were constructed. First, the EOF was measured before the coating was deposited on the capillary walls of the fused-silica capillary. The measured electroosmotic mobility was 5.04 x 10-4 cm2V-1s-1.

126

The EOF was again measured after the two bilayers were constructed in the PEM coating.

In this case, the electroosmotic mobility decreased to 3.31 x 10-4 cm2V-1s-1. Then, 0.1 M and 1.0 M NaOH solutions were flushed through the PEM-coated capillary, and the magnitude of the EOF was measured. Figure 5.1 demonstrates the dependence of electroosmotic mobility on the NaOH flushing time. In addition, this figure illustrates graphically how NaOH can increase the value of electroosmotic mobility, as compared to the 2-bilayer capillary that is not treated with the strong base. As shown, 0.1 M NaOH did not have a major impact on electroosmotic mobility, as compared to 1.0 M NaOH, even after flushing the coated capillary for 100 min. A 5-min 1.0 M NaOH flushing time slightly increased the value of electroosmotic mobility to 3.35 x 10-4 cm2V-1s-1. However, fifteen more minutes demonstrated a significant increase that was still observed upon continuous rinse with 1.0 M NaOH. A total of 230 min resulted in an electroosmotic mobility of 5.03 x 10-4 cm2V-1s-1. In addition, a total of 350 min gave the same EOF value obtained before deposition of the coating (5.04 x 10-4 cm2V-1s-1). This clearly demonstrates that the coating was completely desorbed from the capillary column. It should be noted that each data point in this graph and the graphs that follow is the average value obtained from running each experiment multiple times. Standard deviation values were also calculated for each data point. However, these values are very small, and not perceptible on the graphs.

Figure 5.2 demonstrates the dependence of selectivity on the NaOH flushing time.

Before polymer deposition, the selectivity was 1.0 since no separation between the two enantiomers of BNP was observed. The construction of two bilayers gave a selectivity value of 1.026, which was relatively stable during the coating’s exposure to 0.1 M NaOH.

A slight decrease was observed when 1.0 M NaOH solution was flushed through the

127

coated capillary. A total flush time of 170 min with 1.0 M NaOH resulted in a single BNP peak. This observation also confirmed that the coating was removed after a long exposure to NaOH.

5.5 5.5 0.1 M NaOH 1.0 M NaOH

5 5 ) 4 -

4.5 4.5 x 10 1 - s 1 - V 2 4 4 EOF (cm EOF 3.5 3.5

3 3

0 20406080100120Time (minutes) 0 50 100 150 200 250 300 350 400

time (minutes)

Figure 5.1 Dependence of electroosmotic mobility on the NaOH flushing time. Conditions: 2 bilayers; 0.5% (w/v) PDADMAC and 0.5% (w/v) poly (L- SULV) with 0.1 M NaCl; pressure injection, 0.5 psi for 3 s; background electrolyte, 100 mM Tris and 10 mM Na2B4O7 (pH 10.0); applied voltage, 30 kV; temperature, 25 °C; capillary, 57 cm (50 cm effective length) x 50 µm i.d.; detection, 214 nm.

The stability was further evaluated by calculating the change in EOF, also termed

“degradation ratio” [23, 24]. This ratio was calculated by use of the following equation:

EOF − EOF Degradation ratio = 1 2 ×100 (5.3) EOF1

128

where EOF1 and EOF2 were measured before and after flushing the capillary with the

-4 NaOH solution, respectively. In this case, EOF1 is as mentioned above, i.e. 3.31 x 10 cm2V-1s-1. These degradation ratio values are reported in Table 5.1. The PEM-coated capillary was relatively stable after rinsing with 0.1 M NaOH for 110 min, followed by a rinse with 1.0 M NaOH for 5 min. The degradation ratios obtained until the last exposure

(1.0 M NaOH for 5 min) were all below 2%. However, after additional 1.0 M NaOH rinse, the degradation ratios were above 2%. The coating was finally detached from the capillary wall after long exposure to 1.0 M NaOH as evidenced by α and the EOF value.

1.05 1.05 0.1 M NaOH 1.0 M NaOH

1.04 1.04

1.03 1.03

1.02 1.02 Selectivity Selectivity

1.01 1.01

1 1

0.99 0.99

0 20 40Time (minut 60 es) 80 100 120 0 50 100 150 200 250 300 350 400 time (minutes)

Figure 5.2 Dependence of selectivity on the NaOH flushing time. Conditions: same as Figure 5.1.

129

Table 5.1 Stability of PEM capillary coating. Conditions: same as Figure 5.1.

Total Flushing Time EOF2 Average (min) Degradation Ratio (min) (x 10-4 cm2 V-1s-1) (%)

0.1 M NaOH

5.0 3.31 0.00

20.0 3.31 0.00

50.0 3.31 0.00

110.0 3.26 1.51

1.0 M NaOH

5.0 3.35 1.21

20.0 3.65 10.27

50.0 3.86 16.61

110.0 4.39 32.63

170.0 4.45 34.44

230.0 5.03 51.96

350.0 5.04 52.27

5.3.1.2 Capillary Recovery and Coating Regeneration

When the coating was completely detached, and a selectivity value of 1.0 was obtained, a 2-bilayer coating was reconstructed in the same fused-silica capillary. Figure

130

5.3 graphically demonstrates how electroosmotic mobility gradually changes after again flushing the coated capillary with NaOH solutions. The trend observed in this figure is the same as obtained when the first 2-bilayer coating was constructed. The EOF after the last coating deposition was 3.32 x 10-4 cm2V-1s-1, as compared to the value of 3.31 x 10-4 cm2V-

1s-1 that was measured after the first coating. In addition, a similar trend was observed with the selectivity values, as shown in Figure 5.4.

5.5 5.5

0.1 M NaOH 1.0 M NaOH

5 5 ) 4 -

4.5 4.5 x 10 1 - s 1 - V 2 4 4 EOF (cm EOF 3.5 3.5

3 3

0 20 40Time (minutes) 60 80 100 120 0 50 100 150 200 250 300 350 400

time (minutes)

Figure 5.3 Coating Regeneration - Dependence of electroosmotic mobility on the NaOH flushing time. Conditions: same as Figure 5.1.

131

1.05 1.05

0.1 M NaOH 1.0 M NaOH 1.04 1.04

1.03 1.03

1.02 1.02 Selectivity Selectivity

1.01 1.01

1 1

0.99 0.99 020406080100120Time (minutes) 0 50 100 150 200 250 300 350 400

time (minutes)

Figure 5.4 Dependence of selectivity on the NaOH flushing time. Conditions: same as Figure 5.1.

Another important consideration for this study is the amount of time that is needed to completely detach the coating for capillary recovery. This allows us regeneration of the coating by use of this simple PEM coating procedure. For purpose of this study, two and twenty bilayers were constructed. Both electroosmotic mobility and selectivity measurements illustrated that a 2-bilayer and a 20-bilayer PEM coating (data not shown) could be completely detached from the capillary walls after approximately 3.5 and 9.5 hours, respectively of continuous exposure to 1.0 M NaOH.

132

5.3.2 Laser Scanning Confocal Microscopy

Using the results from the above study, one can conclude that the PEM coating constructed using polymeric surfactants as anionic polymers affects separations [8-10, 44].

However, we desired an imaging technique that would allow visualization of the PEM- coated capillaries for characterization of the surface morphology. In this part of our study,

LSCM was used as an imaging technique to examine the coating. LSCM was also used to image the structural changes of the coating after its exposure to NaOH. For the experiments reported here, R6G proved to be a suitable fluorescent dye to examine the coating since this positively charged probe prefers the environment of the coated phase.

Figure 5.5 is a comparison of a 20-bilayer capillary (left) and a fused-silica capillary (right) side by side in a single image. This comparison is used to visually demonstrate the difference between a coated and bare capillary. In both sections of the capillaries, 10 µM R6G in water was flushed, and then, air was injected in order to wash the excess solution out and dry the capillary. This figure illustrates z-section images that were collected at different depths of focus of the LSCM. The pinhole size of the LSCM was adjusted such that the optical slice was 3.3 µm. From Figure 5.5a to Figure 5.5f, the z- sectioning goes deeper into the PEM-coated capillary. The focal plane in Figure 5.5a is at the outer edge of the capillary wall. Figure 5.5f slices through the outer wall of the capillary on the opposite side. Figures 5.5c and 5.5d slice though regions of the capillary near the center plane. As shown in these images, the walls of the 20-bilayer capillary are very bright as compared to the fused-silica capillary. Since the R6G should be retained to a greater extent in the coated phase than on the wall of the fused-silica capillary, this difference confirms the existence of a coating on the capillary wall.

133

(a) (b)

20-bilayer capillary Fused-silica capillary

(c) (d)

(e) (f)

Figure 5.5 The z-section images of a 20-bilayer capillary (left) and a fused-silica capillary (right). Optical slice thickness: 3.3 µm. z-Step size 1.65 µm. Every fourth image is shown. All images are 204 x 512 pixels (460.6 µm x 115.1 µm).

134

The same 20-bilayer capillary was exposed to 1.0 M NaOH. It was then flushed again with 10 µM R6G and injected with air. Using the confocal microscope, we observed a change in the structure of the coating, i.e. large portions of the polymer coating have been removed from the wall. This observation is clearly illustrated in Figure 5.6. The images of this figure represent again z-section images that were taken as the excitation beam traversed deeper into the coated capillary.

(a) (b) (c)

(d) (e) (f)

Figure 5.6 The z-section images of a portion of the same 20-bilayer capillary, as in Figure 5.5. Exposure to 1.0 M NaOH. Optical slice thickness: 3.3 µm. z- Step size 1.65 µm. Every third image is shown. All images are 512 x 512 pixels (65.8 µm x 65.8 µm).

135

Additional LSCM images of larger portions of the capillary at different locations demonstrate similar discontinuities in the coating (Figure 5.7). The area depicted in Figure

5.7a includes the discontinuity and a large region where the coating on the walls of the capillary does not appear to be damaged to the same extent. Figure 5.7b includes several discontinuities along the wall of the capillary where the coating has been greatly affected.

(a) (b)

Figure 5.7 LSCM images of different regions within the same 20-bilayer capillary, as in Figure 5.6. Optical slice thickness: 3.3 µm. Both images are 512 x 2048 pixels (115.1 µm x 460.6 µm).

136

When the same 20-bilayer capillary was exposed to additional 1.0 M NaOH, the majority of the coating appears to be removed, leaving only several bright islands on the wall (Figure 5.8). In addition, this last image demonstrates that the coating is removed after long exposure to 1.0 M NaOH. The same procedure was followed with the 2-bilayer capillary (data not shown).

Figure 5.8 LSCM image of the same 20-bilayer capillary, as in Figure 5.7. Longer exposure to 1.0 M NaOH. Optical slice thickness: 3.3 µm. The image is 512 x 820 pixels (115.1 µm x 184.4 µm).

5.4 Conclusion

The stability of PEM coatings was further investigated in this study after exposure to NaOH solutions. The OT-CEC study, which allowed measurements of electroosmotic mobility and selectivity, suggests that the PEM coating is relatively stable against 0.1 M

NaOH. In addition, both OT-CEC and LSCM demonstrated the ability to recover the

137

capillaries by flushing them with 1.0 M NaOH. We were also able to visualize the coating within the capillary, and to study its structural changes by using LSCM. This helped us prove that the coating exists, allowed us to see the effects of NaOH on the coating, and demonstrated that the coating can be removed after long exposure to 1.0 M NaOH.

Although LSCM provided new information, it did not provide us with all the details regarding the structure of the coating.

5.5 References

1. Decher, G.; Schmitt, J. Prog. Colloid Polym. Sci. 1992, 89, 160.

2. Decher, G. Science 1997, 277, 1232.

3. Schlenoff, J. B.; Dubas, S. T.; Farhat, T. Langmuir 2000, 16, 9968.

4. Smith, R. N.; Reven, L.; Barrett, C. J. Macromolecules 2003, 36, 1876.

5. Schlenoff, J. B.; Ly, H.; Li, M. J. Am. Chem. Soc. 1998, 120, 7626.

6. Farhat, T. R.; Schlenoff, J. B. Langmuir 2001, 17, 1184.

7. Graul, T. W.; Schlenoff, J. B. Anal. Chem. 1999, 71, 4007.

8. Kapnissi, C. P.; Akbay, C.; Schlenoff, J. B.; Warner, I. M. Anal. Chem. 2002, 74, 2328.

9. Kamande, M. W.; Kapnissi, C. P.; Zhu, X.; Akbay, C.; Warner, I. M. Electrophoresis 2003, 24, 945.

10. Kapnissi, C. P.; Valle, B. C.; Warner, I. M. Anal. Chem. 2003, 75, 6097.

11. Rmaile, H. H.; Schlenoff, J. B. J. Am. Chem. Soc. 2003, 125, 6602.

12. Müller, M. Biomacromolecules 2001, 2, 262.

13. Laurent, D.; Schlenoff, J. B. Langmuir 1997, 13, 1552.

14. Schlenoff, J. B.; Dubas, S. T. Macromolecules 2001, 34, 592.

138

15. Caruso, F.; Lichtenfeld, H.; Donath, E.; Möhwald, H. Macromolecules 1999, 32, 2317.

16. Rmaile, H. H.; Schlenoff, J. B. Langmuir 2002, 18, 8263.

17. Castelnovo, M.; Joanny, J.-F. Langmuir 2000, 16, 7524.

18. Dubas, S. T.; Schlenoff, J. B. Macromolecules 2001, 34, 3736.

19. Dubas, S. T.;Schlenoff, J. B. Macromolecules 1999, 32, 8153.

20. Hayes, J. D.; Malik, A. Anal. Chem. 2001, 73, 987.

21. Farhat, T.; Yassin, G.; Dubas, S. T.; Schlenoff, J. B. Langmuir 1999, 15, 6621.

22. Sui, Z.; Salloum, D.; Schlenoff, J. B. Langmuir 2003, 19, 2491.

23. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 2254.

24. Katayama, H.; Ishihama, Y.; Asakawa, N. Anal. Chem. 1998, 70, 5272.

25. Lowry, M.; He, Y.; Geng, L. Anal. Chem. 2002, 74, 1811.

26. Zhu, H.; Ji, J.; Tan, Q.; Barbosa, M. A.; Shen, J. Biomacromolecules 2003, 4, 378.

27. Moya, S.; Donath, E.; Sukhorukov, G. B.; Auch, M.; Bäumler, H.; Lichtenfeld, H.; Möhwald, H. Macromolecules 2000, 33, 4538.

28. Guo, H. X.; Heinämäki, J.; Yliruusi, J. International J. Pharmaceutics 1999, 186, 99.

29. Silvano, D.; Krol, S.; Diaspro, A.; Cavalleri, O.; Gliozzi, A. Microscopy Research and Technique 2002, 59, 536.

30. Georgieva, R.; Moya, S.; Leporatti, S.; Neu, B.; Bäumler, H.; Reichle, C.; Donath, E.; Möhwald, H. Langmuir 2000, 16, 7075.

31. Voigt, A.; Lichtenfeld, H.; Sukhorukov, G. B.; Zastrow, H.; Donath, E.; Bäumler, H.; Möhwald, H. Ind. Eng. Chem. Res. 1999, 38, 4037.

32. Lamprecht, A.; Schäfer, U. F.; Lehr, C.-M. European J. Pharmaceutics and Biopharmaceutics 2000, 49, 1.

33. Stricker, S. A.; Whitaker, M. Microscopy Research and Technique 1999, 46, 356.

139

34. Gao, C.; Leporatti, S.; Donath, E.; Möhwald, H. J. Phys. Chem. B 2000, 104, 7144.

35. Buttino, I.; Ianora, A.; Carotenuto, Y.; Zupo, V.; Miralto, A. Microscopy Research and Technique 2003, 60, 458.

36. Khopade, A. J.; Caruso, F. Langmuir 2003, 19, 6219.

37. Astrakharchik-Farrimond, E.; Shekunov, B. Y.; Sawyer, N. B. E.; Morgan, S. P.; Somekh, M. G.; See, C. W. Particle and Particle Systems Characterization 2003, 20, 104.

38. Missert, N.; Copeland, R. G.; Barbour, J. C.; Mikkalson, J. E. Electrochemical Society Proceedings 2001, 22, 687.

39. Caponetti, G.; Hrkach J. S.; Kriwet, B.; Poh, M.; Lotan, N.; Colombo, P.; Langer, R. J. Pharm. Sci. 1999, 88, 136.

40. Wang, J.; Warner, I. M. J. Chromatogr. 1995, 711, 297.

41. Chiari, M.; Nesi, M.; Righetti, P. G. In Capillary Electrophoresis in Analytical Biotechnology; Righetti, P. G., Ed.; CRC Press, Inc.: Boca Raton, FL, 1996.

42. Chankvetadze, B. In Capillary Electrophoresis in Chiral Analysis; John Wiley & Sons Ltd: London, UK, 1997.

43. Skoog, D. A.; Holler, J. F.; Nieman, T. A. In Principles of Instrumental Analysis, 5th ed.; Sherman, M., Bortel, J., Messina, F., Eds.; Harcourt Brace College Publishers: Orlando, FL, 1998.

44. Zhu, X.; Kamande, M. W.; Thiam, S.; Kapnissi, C. P.; Mwongela, S. M.; Warner, I. M. Electrophoresis, 2004, 25, 562.

140

CHAPTER 6.

CONCLUSIONS AND FUTURE STUDIES

In this work, two different modes of capillary electrochromatography (CEC) were used for the achiral and chiral separations of various classes of analytes. In Chapter 2 of this dissertation, the packed mode of CEC is described. A 40-cm packed bed of Reliasil 3

µm C18 stationary phase was able to separate seven benzodiazepines. The separation conditions were optimized by varying the mobile phase, the amount of organic modifier, the buffer concentration, the applied voltage, and the column temperature. A mobile phase composition of Tris.HCl (pH 8)-acetonitrile (60:40), an electrolyte concentration of 30 mM, and a temperature of 15 °C with an applied voltage of 20 kV proved to be optimum.

In addition, the packed capillary electrochromatographic method developed here was applied to the characterization of oxazepam in a standard urine sample that contained various concentrations of other drugs, such as acetaminophen, amphetamines, imipramine, morphine, cocaine, etc.

Some preliminary data were also obtained for the separation of a mixture of six β- blockers using packed-CEC with octadecyl silica stationary phase and co-polymerized micelle SUS/SUG (sodium undecylenic sulfate/sodium N-undecanoyl-L-glycinate)

“pseudostationary phase.” β-blockers are clinically important drugs, used for the treatment of several disorders such as angina pectoris, cardiac arrhythmias, and hypertension. They have also been used as doping agents for athletes in order to improve their athletic performance in cases where high psychological pressure may cause the heart to race. The structures of the β-blockers used in this study are shown in Figure 6.1. Co-polymerized micelles are produced by irradiating a micellar solution of surfactants with 60Co

141

irradiation. Their main advantage is that they can be used in a wider pH range than the conventional polymeric surfactants, since they are soluble at and below pH 6.8. The studies that have been performed until now showed that the presence of the co- polymerized micelle SUS/SUG improved the resolution of the analytes. The resolution was also proportional to the concentration of the co-polymerized micelle and the percentage of

SUS in the co-polymerized micelle. The studies on the separation of β-blockers will be continued, and when optimum conditions are established, the method will be applied to a real sample (urine or blood).

O OH

CH3SO2NH CHCH2NHCH(CH3)2 H2NCCH2 OCH2CHCH2NHCH(CH3)2

OH

1. Sotalol O 2. Atenolol

H2NC OH OH

CH OCH CH OCH CHCH NHCH(CH ) 3 2 2 2 2 3 2 HO CHCH2NHCHCH2CH2

CH3

3. Metoprolol 4. Labetalol

OH O OCH 2CHCH 2NHC H(CH3)2 CCH3 O OH

CH3CH2CH2CNH OCH2CHCH2NHCH(CH3)2

5. Propranolol 6. Acebutolol

Figure 6.1 Structures of the six β-blocker analytes.

142

Chapter 3 is an examination of the open-tubular mode of CEC, which is an alternative approach to conventional CEC. The primary advantage of open-tubular CEC

(OT-CEC) is the elimination of problems associated with frits and silica particles in conventional CEC. In this approach, fused silica capillaries coated with thin films of physically adsorbed charged polymers were developed by use of a polyelectrolyte multilayer (PEM) coating procedure. The PEM coating was constructed in situ by alternating rinses with positively and negatively charged polymers, where the negatively charged polymer was a polymeric surfactant. In this study, poly

(diallyldimethylammonium chloride), PDADMAC, was used as the cationic polymer and poly (sodium N-undecanoyl-L-glycinate), poly (L-SUG), was used as the anionic polymer for PEM coating. The performance of the modified capillaries as a separation medium was evaluated by use of seven benzodiazepines as analytes. The run-to-run, day-to-day, week- to-week and capillary-to-capillary reproducibilities of electroosmotic flow (EOF) were very good with relative standard deviation (RSD) values of less than 1% in all cases. In addition, the chromatographic performance of the monomeric form of the polymeric surfactant was compared for the separation of these analytes. The anionic surfactant used for PEM coating construction in this experiment was the monomeric (nonpolymerized) surfactant at the concentration of 0.5% (w/v) (19 mM). Almost no separation was noted, even though the monomeric surfactant concentration was above the normal critical micellar concentration of the nonpolymerized surfactant (7 mM). Therefore, the molecular micelle allows better discrimination of the hydrophobic analytes than the conventional micelle. Furthermore, the PEM-coated capillary was remarkably robust with more than 200

143

runs accomplished in this study. Strong stability against extreme pH values (pH 11.0 and pH 3.0) was also observed.

In Chapter 4, the fused-silica capillaries were again modified using the PEM coating procedure. The quaternary ammonium salt PDADMAC was used as the cationic polymer, and the polymeric surfactant poly (sodium N-undecanoyl-L-leucylvalinate), poly

(L-SULV), was used as the anionic polymer. In this study, the PEM coating was applied to investigate the chiral separations of 1,1’-binaphthyl-2,2’-dihydrogenphosphate (BNP),

1,1’-bi-2-naphthol (BOH), secobarbital, pentobarbital and temazepam. However, the PEM coating procedure used in the achiral studies needed to be modified in order to achieve chiral separations. Optimal conditions were established by varying the additive (sodium chloride, 1-ethyl-3-methyl-1H-imidazolium hexafluorophosphate (1E-3MI-HFP), 1-butyl-

3-methylimidazolium tetrafluoroborate (1B-3MI-TFB)) in the polymer deposition solutions, the salt concentration, the column temperature, and the bilayer number. The ionic liquids 1E-3MI-HFP and 1B-3MI-TFB were explored for possible enhanced separations. The addition of either ionic liquid to the polymer deposition solutions increased the resolution of BNP from 0.83 to 0.88 and 0.90. The effects that ionic liquids have on separations are not very well understood. Therefore, fluorescence studies should be performed to better understand the mechanisms that take place in separations when ionic liquids are used as additives. Reproducibilities were also evaluated by using the RSD values of the EOF and the first peak (R-(+)-BNP). In all cases, the run-to-run and capillary-to-capillary RSD values of EOF were less than 0.5%, and the run-to-run RSD values of the R-(+)-BNP peak were less than 1%. In addition, more than 230 runs were performed in the PEM coated capillary.

144

The studies described above for both chiral and achiral separations have shown that

PEM-coated capillaries have excellent reproducibilities, remarkable endurance, and strong stabilities against extreme pH values when used in open tubular capillary electrochromatography (OT-CEC). In Chapter 5, the stability of the capillary surface was further investigated after exposure to 0.1 M and 1.0 M NaOH. The multilayer coatings used for the studies reported in this chapter consisted of two and twenty layer pairs. A layer pair, i.e. a bilayer, is one layer of a cationic polymer and one layer of an anionic polymer. PDADMAC was used as the cationic polymer, and the polymeric surfactant poly

(L-SULV) was used as the anionic polymer. The structural changes of these coatings were monitored using laser scanning confocal microscopy (LSCM) after flushing the capillaries with NaOH. This technique also allowed a study of the uniformities and discontinuities of the coatings. The observed structures were discussed in terms of separations using OT-

CEC. In addition, the electropherograms obtained from the chiral separation of BNP in

OT-CEC showed a decrease in selectivity and an increase in electroosmotic mobility after long exposure to NaOH. To study the stability of the capillary surface, two bilayers were constructed. The stability was evaluated by computing the change in EOF (degradation ratio). The degradation ratios obtained after rinsing the PEM-coated capillary with 0.1 M

NaOH for 110 min, and with 1.0 M NaOH for 5 min were all below 2%. However, after additional 1.0 M NaOH rinse, the degradation ratios were above 2%. Finally, the ability to recover the capillaries by direct exposure to NaOH was demonstrated. Measurements of both electroosmotic mobility and selectivity showed that a 2-bilayer and a 20-bilayer PEM coating could be completely removed from the capillary surface after approximately 3.5 and 9.5 hours, respectively, of continuous exposure to 1.0 M NaOH.

145

The studies discussed earlier verify that the PEM coating has a significant effect on separations. The next goal is to characterize the coating using various characterization techniques, such as nuclear magnetic resonance (NMR) spectroscopy, ellipsometry, X-ray diffraction, Fourier transform infrared (FTIR) spectroscopy, neutron reflectometry, fluorescence resonance energy transfer (FRET), and near-infrared (NIR) multispectral imaging spectrometry. Different kinds of microscopic techniques, such as atomic force microscopy (AFM), scanning electron microscopy (SEM), transmission electron microscopy (TEM), scanning tunneling microscopy (STM), and near-field scanning optical microscopy (NSOM) can also be used for the characterization of the coating. In particular, the NIR multispectral imaging technique, which can simultaneously record the spectral and spatial information of a sample [1, 2], will be used to study the uniformity and discontinuity of the coating, and to find the optimized flush times of both PDADMAC and polymeric surfactant. Then, FRET and some microscopic techniques will be used to measure the thickness of the PEM coating. FRET is a distance-dependent interaction between the electronic excited states of two fluorophores. Excitation is transferred from a donor fluorophore (i.e., fluorescein) to an acceptor fluorophore (i.e., tetramethylrhodamine) without emission of a photon. The efficiency of FRET is inversely proportional to the sixth power of the intermolecular separation.

Although LSCM was able to demonstrate the existence of the coating and the change in structure after rinsing the capillary with NaOH, it did not provide enough details regarding the structure of the coating. Therefore, the techniques mentioned above can be used to provide information about film thickness, interlayer spacings, structure, elemental content etc [3-14].

146

6.1 References

1. Fischer, M.; Tran, C. D. Anal. Chem. 1999, 71, 2255.

2. Khait, O.; Smirnov, S.; Tran, C. D. Anal. Chem. 2001, 73, 732.

3. Kerimo, J.; Adams, D. M.; Brbara, P. F.; Kaschak, D. M.; Mallouk, T. E. J. Phys. Chem. 1998, 102, 9451.

4. Barberi, R.; Bonvent J. J.; Bartolino, R.; Roeraade, J.; Capelli, L.; Righetti, P. G. J. Chromatogr. B 1996, 683, 3.

5. Pullen, P. E.; Pesek, J. J.; Matyska, M. T.; Frommer, J. Anal. Chem. 2000, 72, 2751.

6. Mahltig, B.; Müller-Buschbaum, P.; Wolkenhauer, M.; Wunnicke, O.; Wiegand, S.; Gohy, J.-F.; Jerome, R.; Stamm, M. J. Colloid and Interface Science 2001, 242, 36.

7. Perez-Salas, U. A.; Faucher, K. M.; Majkrzak, C. F.; Berk, N. F.; Krueger, S.; Chakof, E. L. 2003, 19, 7688.

8. Styrkas, D. A.; Bütün, V.; Lu, J. R.; Keddie, J. L.; Armes, S. P. Langmuir 2000, 16, 5980.

9. Oehlke, J.; Birth, P.; Klauschenz, E.; Wiesner, B.; Beyermann, M.; Oksche, A.; Biener, M. Eur. J. Biochem. 2002, 269, 4025.

10. Kharlampieva, E.; Sukhishvili, S. A. Langmuir 2003, 19, 1235.

11. Preisler, J.; Yeung, E. S. Anal. Chem. 1996, 68, 2885.

12. Kaupp, S.; Wätzig, H. J. Chromatogr. A 1997, 781, 55.

13. Kaupp, S.; Steffen, R.; Wätzig, H. J. Chromatogr. A 1996, 744, 93.

14. Advincula, R.; Wang, Y.; Park, M.; Youk, J. H.; Mays, J.; Yang, J.; Kaneko, F.; Baba, A.; Knoll, W. Polymer Reprints 2001, 42, 281.

147

VITA

Constantina P. Kapnissi was born in Nicosia, Cyprus, on March 22, 1977. She is the first-born child in the family, and she has three sisters. The fourth one is ten years old.

In June of 1995, she graduated from a public high school. Two months after that she attended the University of Cyprus. She majored in general chemistry, and she obtained a

Bachelor of Science degree in May, 1999. However, before graduation, in the summer of

1998, she started her undergraduate thesis work at ETH University in Zurich, Switzerland.

After spending a month there, she moved back to Cyprus to continue her work with the help and direction of her advisor Dr. Epameinondas Leontidis. Her undergraduate thesis title was “Morphology of Gold Colloids from Cationic Surfactant Solutions.” On July 11,

1999, she got engaged to Andreas Christodoulou, who had already been a student at

Louisiana State University in Baton Rouge, Louisiana. After the engagement, she moved to United States with him to pursue a degree of Doctor of Philosophy in chemistry at the same university, under the direction of Dr. Isiah M. Warner. Her dissertation focuses on the development of packed and open-tubular capillary electrochromatographic methods for improved achiral and chiral separations of various classes of analytes.

The articles she published, and the patent she submitted from her research work are listed below:

• Warner, I. M., Kapnissi, C. P., Kamande, M., Valle, B. C., Schlenoff, J. B. Analytical Separations With Polyelectrolyte Layers, Molecular Micelles, or Zwitterionic Polymers, U.S. Patent, Submitted.

• Kapnissi, C. P., Agbaria, R. A., Lowry, M., Geng, L., Warner, I. M. Investigation of the Stability of Polyelectrolyte Multilayer Coatings Using Laser Scanning Confocal Microscopy and Open-Tubular Capillary Electrochromatography, Submitted for publication.

148

• Kamande, M. W., Zhu, X., Kapnissi-Christodoulou, C. P., Warner, I. M. Chiral Separations Using a Polypeptide, a Polymeric Dipeptide Surfactant, and a Polyelectrolyte Multilayer Coating in Open-Tubular Capillary Electrochromatography, Submitted for publication.

• Kapnissi, C. P., Warner, I. M. Separation of Benzodiazepines by Use of Capillary Electrochromatography, Journal of Chromatographic Science, 2004, 42, 238-244.

• Zhu, X., Kamande, M. W., Thiam, S., Kapnissi, C. P., Mwongela, S. M., Warner, I. M. Open-Tubular Capillary Electrochromatography/Electrospray Ionization- Mass Spectrometry Using Polymeric Surfactant as a Stationary Phase Coating, Electrophoresis, 2004, 25, 562-568.

• Kapnissi, C. P., Valle, B. C., Warner, I. M. Chiral Separations Using Polymeric Surfactants and Polyelectrolyte Multilayers in Open-Tubular Capillary Electrochromatography, Analytical Chemistry, 2003, 75, 6097-6104.

• Kapnissi-Christodoulou, C. P., Zhu, X., Warner, I. M. Analytical Separations in Open-Tubular Capillary Electrochromatography, Electrophoresis, 2003, 24, 3917- 3934.

• Kamande, M. W., Kapnissi, C. P., Akbay, C., Zhu, X., Agbaria, R. A., Warner, I. M. Open-Tubular Capillary Electrochromatography Using a Polymeric Surfactant Coating, Electrophoresis, 2003, 24, 945-951.

• Kapnissi, C. P., Akbay, C., Schlenoff, J. B., Warner, I. M. Analytical Separations Using Molecular Micelles in Open-Tubular Capillary Electrochromatography, Analytical Chemistry, 2002, 74, 2328-2335.

During her years in graduate school, she presented her work at a number of scientific conferences including:

• “Chiral Separations Using Polymeric Surfactants and Polyelectrolyte Multilayers in Open-Tubular Capillary Electrochromatography - Morphologic Elucidation of Capillaries Using Laser Scanning Confocal Microscopy” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Chicago, IL, March 2004. Kapnissi-Christodoulou, C. P., Agbaria, R. A., Lowry, M., Valle, B. C., Geng, L., Warner, I. M.

• “Chiral Separations Using a Polymeric Dipeptide Surfactant in Open-Tubular Capillary Electrochromatography” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Chicago, IL, March 2004. Kamande, M. W., Kapnissi- Christodoulou, C. P., Zhu, X., Warner, I. M.

149

• “Chiral Separations Using Polymeric Surfactants and Polyelectrolyte Multilayers in Open-Tubular Capillary Electrochromatography” American Chemical Society Meeting, New York, NY, September 2003. Kapnissi, C. P., Agbaria, R. A., Lowry, M., Valle, B. C., Geng, L., Warner, I. M.

• “Analytical Separations Using Molecular Micelles in Open-Tubular Capillary Electrochromatography: Morphologic Elucidation of Capillaries Modified by Polyelectrolyte Multilayers” American Chemical Society Meeting, New Orleans, LA, March 2003. Kapnissi, C. P., Agbaria, R. A., Lowry, M., Valle, B. C., Geng, L., Warner, I. M.

• “Separation of Phenols and Benzodiazepines Using Poly (Sodium Undecylenic Sulfate) and Open-Tubular Capillary Electrochromatography” American Chemical Society Meeting, New Orleans, LA, March 2003. Kamande, M. W., Kapnissi, C. P., Zhu, X., Akbay, C., Agbaria, R. A., Warner, I. M.

• “Morphologic Elucidation of Capillaries Modified by Polyelectrolyte Multilayers for Analytical Separations” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Orlando, FL, March 2003. Kapnissi, C. P., Agbaria, R. A., Lowry, M., Valle, B. C., Geng, L., Warner, I. M.

• “Elution Behavior of Hydrophobic Analytes by Use of Various Stationary Phases in Capillary Electrochromatography” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Orlando, FL, March 2003. Igbinosun, O. J., Kapnissi, C. P., Kimberly Y. H., Warner, I. M.

• “Open-tubular Capillary Electrochromatography Using a Polymeric Surfactant Coating” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Orlando, FL, March 2003. Kamande, M. W., Kapnissi, C. P., Zhu, X., Akbay, C., Agbaria, R. A., Warner, I. M.

• “Open-Tubular Capillary Electrochromatography Using a Polymeric Surfactant Coating” 16th International Symposium on Microscale Separations and Analysis, San Diego, CA, January 2003. Kamande, M. W., Kapnissi, C. P., Zhu, X., Akbay, C., Warner, I. M.

• “Novel Separations Using Polymeric Surfactants in Open-Tubular Capillary Electrochromatography” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, New Orleans, LA, March 2002. Kapnissi, C. P., Schlenoff, J. B., Warner, I. M.

• “Separation of Beta-Blockers Using Capillary Electrochromatography” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, New Orleans, LA, March 2002. Kapnissi, C. P., Igbinosun, O. J., Agbaria, R. A., Warner, I. M.

150

• “Separation of Benzodiazepines by Capillary Electrochromatography” Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, New Orleans, LA, March 2001. Kapnissi, C. P., Thiam, S., Warner, I. M.

• “Separation of Benzodiazepines by Capillary Electrochromatography” American Chemical Society Meeting, New Orleans, LA, December 2000. Kapnissi, C. P., Thiam, S., Warner, I. M.

151