USF Libraries
USF Digital Collections

Novel sol-gel titania-based hybrid organic-inorganic coatings for on-line capillary microextraction coupled to high-perf...

MISSING IMAGE

Material Information

Title:
Novel sol-gel titania-based hybrid organic-inorganic coatings for on-line capillary microextraction coupled to high-performance liquid chromatography
Physical Description:
Book
Language:
English
Creator:
Kim, Tae-Young
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
In-tube SPME
CME-HPLC
CME-GC
Gradient elution
Isocratic elution
Titania-PDMS
Titania-silica-TESP-PEO
PAHs
Ketones
Alkylbenzenes
Aldehydes
Anilines
Phenols
Fatty acids
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Novel sol-gel titania-poly(dimethylsiloxane) (TiO2-PDMS) and titania-silica-N-(triethoxysilylpropyl)-O-polyethylene oxide urethane (TiO2-SiO2-TESP-PEO) coatings were developed for capillary microextraction (CME) to perform on-line preconcentration and HPLC analysis of trace impurities in aqueous samples. Due to chemical inertness of titania, effective covalent binding of a suitable organic ligand to its surface is difficult via conventional surface modification methods. In this research, sol-gel chemistry was employed to chemically bind hydroxy-terminated poly(dimethylsiloxane) (PDMS) and N-(triethoxysilylpropyl)-O-polyethylene oxide urethane (TESP-PEO) to sol-gel titania and sol-gel titania-silica network, respectively. A method is presented describing in situ preparation of the titania-based sol-gel PDMS and TESP-PEO coatings and their immobilization on the inner surface of a fused-silica microextraction capillary.^ To perform on-line CME-HPLC, the sol-gel TiO2-PDMS or TiO2-SiO2-TESP-PEO capillarywas installed in the HPLC injection port as an external sampling loop, and a conventionalHPLC separation column was used for the liquid chromatographic separation. The sol-gel TiO2-PDMS-coated microextraction capillary was used for on-line CME-HPLC analysis of non-polar and moderately polar analytes, and the sol-gel coatings showed excellent pH (1-13), and solvent (acetonitrile and methanol) stabilities under elevated temperatures (150 C) over analogous non-sol-gel silica-based coatings. Extraction of highly polar analytes, especially from aqueous phases is not an easy task. However, the sol-gel TiO2-SiO2-TESP-PEO-coated capillaries showed excellent capability of extracting underivatized highly polar analytes from aqueous samples.^ This opens the possibility to employ sol-gel titania-based polar coatings for solvent-free extraction and trace analysis of target analytes in environmental and biomedical matrices. To our knowledge, this is the first research on the use of sol-gel titania (or titania-silica)-based organic-inorganic materials as a sorbent in capillary microextraction. The newly developed sol-gel titania (or titania-silica)-based organic-inorganic hybrid extraction media provides an effective solution to coupling CME with HPLC (CME-HPLC), and this can be expected to become a powerful analytical tool in environmental investigations, proteomic research, early disease diagnosis and biomarker research. Being a combination of a highly efficient solvent free sample preconcentration technique (CME) and a powerful separation method (HPLC), CME-HPLC poses to become a key analytical tool in solving complex chemical, environmental, and biomedical problems involving complex matrices.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2006.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Tae-Young Kim.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 213 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001920062
oclc - 187088980
usfldc doi - E14-SFE0001833
usfldc handle - e14.1833
System ID:
SFS0026151:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001920062
003 fts
005 20080103141347.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 080103s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001833
035
(OCoLC)187088980
040
FHM
c FHM
049
FHMM
090
QD31.2 (ONLINE)
1 100
Kim, Tae-Young.
0 245
Novel sol-gel titania-based hybrid organic-inorganic coatings for on-line capillary microextraction coupled to high-performance liquid chromatography
h [electronic resource] /
by Tae-Young Kim.
260
[Tampa, Fla] :
b University of South Florida,
2006.
3 520
ABSTRACT: Novel sol-gel titania-poly(dimethylsiloxane) (TiO2-PDMS) and titania-silica-N-(triethoxysilylpropyl)-O-polyethylene oxide urethane (TiO2-SiO2-TESP-PEO) coatings were developed for capillary microextraction (CME) to perform on-line preconcentration and HPLC analysis of trace impurities in aqueous samples. Due to chemical inertness of titania, effective covalent binding of a suitable organic ligand to its surface is difficult via conventional surface modification methods. In this research, sol-gel chemistry was employed to chemically bind hydroxy-terminated poly(dimethylsiloxane) (PDMS) and N-(triethoxysilylpropyl)-O-polyethylene oxide urethane (TESP-PEO) to sol-gel titania and sol-gel titania-silica network, respectively. A method is presented describing in situ preparation of the titania-based sol-gel PDMS and TESP-PEO coatings and their immobilization on the inner surface of a fused-silica microextraction capillary.^ To perform on-line CME-HPLC, the sol-gel TiO2-PDMS or TiO2-SiO2-TESP-PEO capillarywas installed in the HPLC injection port as an external sampling loop, and a conventionalHPLC separation column was used for the liquid chromatographic separation. The sol-gel TiO2-PDMS-coated microextraction capillary was used for on-line CME-HPLC analysis of non-polar and moderately polar analytes, and the sol-gel coatings showed excellent pH (1-13), and solvent (acetonitrile and methanol) stabilities under elevated temperatures (150 C) over analogous non-sol-gel silica-based coatings. Extraction of highly polar analytes, especially from aqueous phases is not an easy task. However, the sol-gel TiO2-SiO2-TESP-PEO-coated capillaries showed excellent capability of extracting underivatized highly polar analytes from aqueous samples.^ This opens the possibility to employ sol-gel titania-based polar coatings for solvent-free extraction and trace analysis of target analytes in environmental and biomedical matrices. To our knowledge, this is the first research on the use of sol-gel titania (or titania-silica)-based organic-inorganic materials as a sorbent in capillary microextraction. The newly developed sol-gel titania (or titania-silica)-based organic-inorganic hybrid extraction media provides an effective solution to coupling CME with HPLC (CME-HPLC), and this can be expected to become a powerful analytical tool in environmental investigations, proteomic research, early disease diagnosis and biomarker research. Being a combination of a highly efficient solvent free sample preconcentration technique (CME) and a powerful separation method (HPLC), CME-HPLC poses to become a key analytical tool in solving complex chemical, environmental, and biomedical problems involving complex matrices.
502
Dissertation (Ph.D.)--University of South Florida, 2006.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
500
Title from PDF of title page.
Document formatted into pages; contains 213 pages.
Includes vita.
590
Adviser: Abdul Malik, Ph.D.
653
In-tube SPME.
CME-HPLC.
CME-GC.
Gradient elution.
Isocratic elution.
Titania-PDMS.
Titania-silica-TESP-PEO.
PAHs.
Ketones.
Alkylbenzenes.
Aldehydes.
Anilines.
Phenols.
Fatty acids.
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1833



PAGE 1

Novel Sol-Gel Titania-Based H ybrid Organic-Inorganic Coat ings for On-Line Capillary Microextraction Coupled to HighPerformance Liquid Chromatography by Tae-Young Kim A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry College of Arts and Sciences University of South Florida Major Professor: Abdul Malik, Ph.D. Kirpal S. Bisht, Ph.D. Milton D. Johnston, Jr., Ph.D. Dean F. Martin, Ph.D. Date of Approval: November, 2006 Keywords: In-tube SPME, CME-HPLC, CME-GC, Gradient elution, Isocratic elution, Titania-PDMS, Titania-Silica-TESP-PEO, PA Hs, Ketones, Alkylbenzenes, Aldehydes, Anilines, Phenols, Fatty acids Copyright 2006, Tae-Young Kim

PAGE 2

DEDICATION To my family and relatives for their overwhelming support, concern, and love. Kim, Keum Hun, Yang, Sung Ae, Dr. Kim, Soo Chul, Kim, Chae Kyoung, Kwon, Young Hee, Kim, Kwanyoung Christopher, and many others …

PAGE 3

ACKNOWLEDGMENTS I would like to present my sincere thanks to many people for their support, guidance, and encouragement during my school years. I would like to express my thanks to my major professor, Dr. Abdul Malik for his supervisi on, patience, and encouragement. I am also very grateful to my dissertation committee members: Dr. Kirpal S. Bisht; Dr. Milton D. Johnston, Jr.; and Dr. Dean F. Martin for their valuable support, advice, and encouragement. I would like to extend thanks to all my former and current colleagues, Dr. Khalid Alhooshani, Dr. Abuzar Kabir, Dr. Wen Li, Li Fang, Same er Kulkarni, Anne Marie Shearrow, Erica Turner, and Scott Segro for their continuous advice, assistance, encouragement, and friendship dur ing my graduate school life. I acknowledge the Department of Chemistr y for their long-term financial support throughout my graduate study and the US Na val office (N00014-98-10848) for partial research support. Finally, I would like to express my great appreciation to my family and relatives for their unlimited patience, c oncern, and endless dedication.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES vii LIST OF FIGURES x LIST OF SCHEMES xiii LIST OF SYMBOLS AND ABBREVIATIONS xiv ABSTRACT xvii CHAPTER ONE: AN OVERVIEW ON SOLID-PHASE MICROEXTRACTION (SPME) 1 1.1 Introduction to Sample Preparation 1 1.2 Fundamentals of Sample Extraction 4 1.2.1 Extraction of analytes from solid sample matrices 6 1.2.1.1 Soxhlet extraction 6 1.2.1.2 Supercritical fluid extraction (SFE) 6 1.2.1.3 Microwave-assisted extraction (MAE) 7 1.2.1.4 Accelerated solven t extraction (ASE) 7 1.2.2 Extraction of analytes fro m liquid sample matrices 8 1.2.2.1 Liquid-liquid extraction (LLE) 8 1.2.2.2 Solid-phase extraction (SPE) 8 1.2.2.3 Membrane extraction techniques 9 1.2.3 Extraction of analytes from gaseous sample matrices 9 1.2.3.1 Static headspace analysis 10 1.2.3.2 Dynamic headspace analysis 10 1.3 Solid-Phase Microextraction (SPME) 11

PAGE 5

ii 1.3.1 Principles of SPME 14 1.3.2 Conventional SPME coatings 20 1.3.2.1 Fiber SPME coatings 20 1.3.2.1.1 Direct (immersion) SPME 25 1.3.2.1.2 Headspace SPME (HS-SPME) 25 1.3.2.2 In-tube SPME 25 1.4 Sol-Gel Capillary Mi croextraction (CME) and Sample Preconcentration 32 1.4.1 Principles of CME 33 1.5 References for Chapter One 38 CHAPTER TWO: SOL-GEL TEC HNOLOGY IN SOLID-PHASE MICROEXTRACTION 43 2.1 Introduction to Sol-Gel Technology 43 2.2 Fundamentals of Sol-Gel Chemistry 46 2.2.1 Sol-gel precursors 49 2.2.2 Sol-gel solvent system 51 2.2.3 Sol-gel catalysts/inhibitors 52 2.3 Chemical Reactions of Transi tion Metal Alkoxides during the Sol-Gel Process 56 2.3.1 Hydrolysis 58 2.3.2 Condensation reaction 60 2.4 Sol-Gel Coatings for Capill ary Microextraction (CME) 62 2.4.1 Pre-treatment of fused-silica capillary 62 2.4.2 Preparation of sol solution 65 2.4.3 Sol-gel coating technology 65 2.4.4 Further treatment of sol-gel-coated CME capillary 66 2.5 Characterization of Sol-Gel Stat ionary Phase and Its Morphology 67 2.6 The Application of Sol-Gel-Coated Microextraction Capillary in

PAGE 6

iii SPME and CME 68 2.7 Sol-Gel Monoliths in Separation Science 74 2.8 References for Chapter Two 79 CHAPTER THREE: HIGHAND LOW-pH-RESISTANT, SURFACEBONDED SOL-GEL TITANIA HYBRID ORGANIC-INORGANIC COATING FOR ON-LINE CME-HPLC 86 3.1 Introduction 86 3.1.1 Titania as a chromatographic support in separation science 88 3.1.1.1 Titania as a chromatographic column support in HPLC 89 3.1.1.1 Titania as a chromatographic column support in CE 92 3.2 Other Applications of Titania 92 3.3 Sol-Gel Titania as an Ex traction Sorbent in CME 95 3.4 Experimental 96 3.4.1 Equipment 96 3.4.2 Chemicals and materials 96 3.4.3 Preparation of the sol solution 97 3.4.4 Preparation of sol-gel TiO2-PDMS-coated microextraction capillary 99 3.4.5 Sol-gel titania coatings in capillary microextraction (CME) for on-line CME-HPLC analysis 99 3.4.5.1 Treatment of sol-gel titania-PDMS-coated capillaries with 0.1 M NaOH solution 102 3.4.5.2 Treatment of sol-gel titania-PDMS-coated capillaries with 0.1 M HCl solution 102 3.4.5.3 Treatment of sol-gel titania-PDMS-coated capillaries with HPLC solvents at high temperature 102 3.4.5.4 Safety precuations 103

PAGE 7

iv 3.5 Results and Dicussion 103 3.5.1 Sol-gel reactions for the preparation of sol-gel TiO2PDMS coating 104 3.5.2 Scanning electron microscopy (SEM) of surface-bonded sol-gel TiO2-PDMS coating 110 3.5.3 Deactivation of the sol-gel TiO2-PDMS coating 112 3.5.4 Fourier transform infrar ed (FTIR) spectroscopic investigation of the crea ted sol-gel titan ia sorbent 112 3.5.5 Applications of sol-gel TiO2-PDMS-coated microextraction capillary 114 3.5.6 Extraction kinetic profile for sol-gel TiO2-PDMS-coated microextraction capillary 123 3.5.7 High pH stability of sol-gel TiO2-PDMS coating 125 3.5.8 Stability of sol-gel TiO2-PDMS coating under highly acidic conditions 130 3.5.9 Stability of sol-gel TiO2-PDMS coating in HPLC solvents under elevated temperatures 133 3.6 Conclusion 140 3.7 References for Chapter Three 140 CHAPTER FOUR: SOL-GEL TITANI A-SILICA HYBRI D ORGANICINORGANIC COATING FOR THE EXTR ACTION OF POLAR ANALYTES WITH ON-LINE CME-HPLC AND OFF-LINE CME-GC 147 4.1 Introduction 147 4.1.1 Polyethylene glycols (PEGs) as sorbent in solid-phase microextraction (SPME) 148 4.1.2 Applications of low molecular weight PEG, N -(triethoxy -silylpropyl)-O-polyethylene oxide urethane (TESP-PEO) 149 4.2 Experimental 150

PAGE 8

v 4.2.1 Equipment 150 4.2.2 Chemicals and materials 151 4.2.3 Preparation of the sol solution 152 4.2.4 Preparation of sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary 154 4.2.5 Capillary microextraction (CME) and on-line CMEHPLC analysis 154 4.2.6 Off-line CME-GC analysis 155 4.2.7 Safety precautions 158 4.3 Results and Discussion 158 4.3.1 Sol-gel reactions for the preparation of sol-gel TiO2-SiO2 -TESP-PEO coating 158 4.3.2 Scanning electron microscopy (SEM) of sol-gel titaniasilica-TESP-PEO coatings bonded to the inner surface of a fused-silica capillary 163 4.3.3 Fourier transform infrared (FTIR) spectroscopy of the sol-gel titania-silica-TESP-PEO surface 165 4.3.4 Applications of sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary 167 4.3.5 Extraction kinetic profile of sol-gel TiO2-SiO2-TESPPEO-coated microextraction capillary 182 4.4 Conclusion 184 4.5 References for Chapter Four 184 APPENDICES 188 Appendix A: High pH-resistant, su rface-bonded sol-ge l titania hybrid organic-inorganic coating for effective on-line hyphenation of capillary microe xtraction (in-tube solidphase microextraction) w ith high-performance liquid chromatography 189

PAGE 9

vi Appendix B: Sol-gel approach to in situ creation of high pH-resistant surface-bonded organic-inorganic hybrid zirconia coating for capillary microe xtraction (in-tube SPME) 199 Appendix C: United States Patent Application Publication 213 ABOUT THE AUTHOR End Page

PAGE 10

vii LIST OF TABLES Table 1.1 Conventional sample preparation techniques 5 Table 1.2 Commercially available fiber coatings for SPME and their applications 21 Table 1.3 Applications of in-tube SPME techniques for various samples 30 Table 2.1 List of common alkoxide -based sol-gel precursors 50 Table 2.2 Summary of sol-gel so rbent used in SPME and CME 70 Table 3.1 Names, functions, and chemi cal structures of the coating solution ingredients used to prepare sol-gel TiO2-PDMS-coated microextraction capillaries 98 Table 3.2 Physical properties and chem ical structures of PAHs used to prepare aqueous samples for CME-HPLC analysis employing a sol-gel TiO2-PDMS-coated microextraction capillary 116 Table 3.3 Physical properties and chemi cal structures of ketones used to prepare aqueous samples for CME-HPLC analysis employing a sol-gel TiO2-PDMS-coated microextraction capillary 119 Table 3.4 Physical properties and chem ical structures of alkylbenzenes used to prepare aqueous samples for CME-HPLC analysis employing a sol-gel TiO2-PDMS-coated microextraction capillary 121 Table 3.5 Peak area repeat ability and limits of de tection (LOD) data for PAHs, ketones, and alkylbenzenes obtained in CME-HPLC experiments using sol-gel TiO2-PDMS-coated microextraction capillaries 122 Table 3.6 Peak area repeat ability and limits of de tection (LOD) data for PAHs obtained on a sol-gel TiO2-PDMS-coated microextraction capillary before and after treatment with 0.1 M NaOH for 12h 129

PAGE 11

viii Table 3.7 CME-HPLC peak area rep eatability and limits of detection (LOD) data for a mixture of ke tones and PAHs obtained on a sol-gel TiO2-PDMS-coated microextraction capillary before and after treatment with 0.1 M HCl for 12 h 132 Table 3.8 Peak area repeatab ility and limits of detection (LOD) data for a mixture of ketones and PAHs using a sol-gel TiO2-PDMS -coated microextraction capillary before and after the extraction capillary was filled w ith the mixture of ACN/water (50/50, v/v) and heated at 150 C for 12 h 137 Table 3.9 Peak area repeat ability and limits of de tection (LOD) data for a mixture of ketones and PAHs using a sol-gel TiO2PDMS-coated microextraction cap illary before and after the extraction capillary was filled w ith 100% ACN and heated at 150 C for 12 h 138 Table 3.10 Peak area repeat ability and limits of de tection (LOD) data for a mixture of ketones and PAHs using a sol-gel TiO2PDMS-coated microextraction cap illary before and after the extraction capillary was filled with 100% MeOH and heated at 150 C for 12 h 139 Table 4.1 Names, functions, and chemical structures of the coating solution ingredients used to prepare sol-gel TiO2-SiO2-TESPPEO-coated microextraction capillaries 153 Table 4.2 Physical properties and chem ical structures of aldehydes extracted from an aqueous sample using a sol-gel TiO2-SiO2TESP-PEO-coated microextraction capillary 170 Table 4.3 Physical properties and ch emical structures of aniline derivatives extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary 172 Table 4.4 Physical properties and chemical structures of substituted phenols extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary 175 Table 4.5 Physical properties and chemical structures of substituted phenols extracted from an aqueous sample using a sol-gel TiO-SiO2-TESP-PEO-coated microextraction capillary 177

PAGE 12

ix Table 4.6 Physical properties and ch emical structures of fatty acids extracted from an aqueous sample using a sol-gel TiO2-SiO2TESP-PEO-coated microextraction capillary 180 Table 4.7 Peak area repeat ability and limits of de tection (LOD) data for aldehydes, aniline derivatives, and substituted phenols in CME-HPLC, and fatty acids in CME-GC using a sol-gel TiO2SiO2-TESP-PEO-coated microextraction capillary 181

PAGE 13

x LIST OF FIGURES Figure 1.1 Steps in an analytical process 3 Figure 1.2 Design of the first commercial SPME device made by Supelco 13 Figure 1.3 Modes of SPME operation: (A ) direct (immersion) SPME, (B) headspace SPME (HS-SPME) 19 Figure 1.4 Graphical scheme fo r choosing SPME fiber coating 24 Figure 1.5 Schematic of coatings in (A) fiber-based SPME and (B) in-tube SPME 28 Figure 1.6 Extraction of analytes by (A) fiber SPME and (B) in-tube SPME 29 Figure 1.7 Classification of sa mple preparation techniques 31 Figure 1.8 Extraction and preconcentra tion of analytes by sol-gel CME capillary 37 Figure 2.1 Shape of different products available through processing by solgel technology 45 Figure 2.2 Overview of the sol-gel process 48 Figure 2.3 Structures of bridgi ng and chelating ligands, R=Pri: (A) bidentate bridging ligand, (B) chelating lig and, and (C) two chelating agents 55 Figure 2.4 Schematic of a homemade capillary filling/purging device 64 Figure 2.5 Chromatographic separation of five –blocking drugs on a silica rod column at different flow rate 76 Figure 2.6 SEM images of a cross-section from a Chromolith structure 77 Figure 2.7 SEM of a sol-gel monolithic column 78 Figure 3.1 Crystal structures of two crystallographic modifications of titanium dioxide: (A) an atase and (B) rutile 87

PAGE 14

xi Figure 3.2 Preparation of “bonded” titan ia-based stationary phase for HPLC via silanization/hydrosilylation 90 Figure 3.3 Application of TiO2 in the field of catalysis 94 Figure 3.4 Schematic diagram of the on-line CME-HPLC setup 101 Figure 3.5 Scanning electron microscopic (SEM) images of a 320-m i.d. fused-silica capillary with sol-gel TiO2-PDMS coating 111 Figure 3.6 FTIR spectra of the sol-gel TiO2-PDMS coating 113 Figure 3.7 CME-HPLC analysis of PAHs 115 Figure 3.8 CME-HPLC analysis of ketones 118 Figure 3.9 CME-HPLC analysis of alkylbenzenes 120 Figure 3.10 Illustration of the Extraction kinetic profile of fluorene ( ), and hexanophenone ( ) obtained on a 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMS-coated microe xtraction capillary using 100 and 300 ng/mL aqueous solutions, respectively. 124 Figure 3.11 Chromatograms representing CME-HPLC analysis of PAHs using a sol-gel TiO2-PDMS-coated microextrac tion capillary before (A) and after (B) rinsing the mi croextraction capillary with a 0.1 M NaOH solution (pH=13) for 12h 126 Figure 3.12 Chromatograms representing CME-HPLC analysis of PAHs using a segment of commercial PDMS-based GC column as the microextraction capillary before (A) and after (B) rinsing the microextraction capillary with a 0.1 M NaOH solution (pH=13) for 12h 128 Figure 3.13 Chromatograms representing CME-HPLC analysis of ketones and PAHs using a sol-gel TiO2-PDMS-coated microextraction capillary before (A) and after (B) rinsing the microextraction capillary with a 0.1 M HCl solution (pH=1) for 12h 131 Figure 3.14 Chromatograms representing CME-HPLC analysis of ketones and PAHs using a sol-gel TiO2-PDMS-coated microextraction capillary before (A) and after (B ) the microextraction capillary filled with the mixture of ACN/water (50/50, v/v), sealed by mini union connector, and heated at 150 C for 12h 135

PAGE 15

xii Figure 4 1 Gravity-fed sample delivery sy stem (1.0 x 60 cm, Glass barrel only) for capillary microextraction 157 Figure 4.2 Scanning electron microscopic image of a 250m i.d. fusedsilica capillary with sol-gel TiO2-SiO2-TESP-PEO coating 164 Figure 4.3 FTIR spectrum of the sol-gel TiO2-SiO2-TESP-PEO coating 166 Figure 4.4 CME-HPLC analysis of aldehydes 169 Figure 4.5 CME-HPLC analysis of aniline derivatives 171 Figure 4.6 CME-HPLC analysis of substituted phenols 174 Figure 4.7 CME-HPLC analysis of other substituted phenols 176 Figure 4.8 CME-GC analysis of fatty acids 179 Figure 4.9 Illustration of the extraction ki netic profile of acridine ( ), and 2-chlorophenol ( ) obtained on a 40 cm 0.25 mm i.d.x 0.2 m sol-gel TiO2-SiO2-TESP-PEO-coated micr oextraction capillary using 25 and 500 ng/mL aqueous solutions, respectively 183

PAGE 16

xiii LIST OF SCHEMES Scheme 2.1 Sol-gel hydrolysis and condensation reactions 57 Scheme 2.2 Hydrolysis of metal alkoxides in the sol-gel process 59 Scheme 2.3 Sol-gel condensati on reactions of metal alkoxides 61 Scheme 3.1 (A) Hydrolysis of titanium (IV) isopr opoxide, and (B) polycondensation of hydrolysis pr oduct, titanium hydroxide 106 Scheme 3.2 (C) Polycondensation of hydroxyl-terminated PDMS with the evolving sol-gel network 107 Scheme 3.3 (D) Chemical anchoring of the sol-gel polymer to the inner walls of the capillary 108 Scheme 3.4 Deactivation of surface-bonded sol-gel TiO2-PDMS coating with HMDS and PMHS taking place during thermal treatment of the coated microextraction capillary at 150 C 109 Scheme 4.1 (A) Hydrolysis of titanium (IV) isopropoxide and the alkoxysilane compounds 160 Scheme 4.2 (B) Polycondensation and chemical incorporation of the hydrolysis products with the evolving sol-gel network 161 Scheme 4.3 (C) Chemical anchoring of the sol-gel TiO2-SiO2 hybrid polymer to the inner walls of the capillary 162

PAGE 17

xiv LIST OF SYMBOLS AND ABBREVIATIONS a Activity of analytes AN Nucleophilic Addition ACN Acetonitrile AFM Atomic Force Microscopy ASE Accelerated Solvent Extraction bp Boiling Point BMA Butyl Methacrylate BTEX Benzene, Toluene, Ethylbenzene, and Xylene C18-TMS N-octadecyldimethyl[3-(tri methoxysilyl)propyl] ammonium chloride CAR Carboxen CE Capillary Electrophoresis CEC Capillary Electrochromatography CERAMERS Ceramic Polymers CME Capillary Microextraction CN-PDMS Cyanopropyl-Poly(dimethylsiloxane) CW Carbowax d Density D Translational diffusion coefficient DATEG/OH-TSO -Diallyltriethylene Glycol/Hydr oxyl-Terminated Silicon Oil DCCA Drying Control Chemical Additive DI Deionized DB (or DVB) Divinylbenzene EOF Electroosmotic Flow EPA Environmental Protection Agency EXAFS X-Ray Absorption Fine Structure Spectroscopy FID Flame Ionization Detector FTIR Fourier Transform Infrared Spectroscopy GAA Glacial Acetic Acid GC Gas Chromatography HMDS 1,1,1,3,3,3-Hexamethyldisilazane Viscosity of a solvent HPLC High-Performance Liquid Chromatography HS-SPME Headspace Solid-Phase Microextraction

PAGE 18

xv IC-CD Ion Chromatography with Conductivity Detection i.d. Inner Diameter LC-MS Liquid Chromatography-Mass Spectrometry kB Boltzmann’s constant K Distribution constant LLE Liquid-Liquid Extraction LOD Limit of Detection MAE Microwave-Assisted Extraction MeOH Methanol MIP Molecularly Imprinted Polymers MS mass spectrometry MTMOS Methyltrimethoxysilane MW Molecular Weight nD Refractive Index NMR Nuclear Magnetic Resonance ODS Octadecylsilane OH-DB14C4 Hydroxy-Terminated Dibenzo-14-Crown-4 ORMOCERS Organically Modified Ceramics ORMOSILS Organically Modified Silicates OTCs Open-Tubular Capillary Columns OTEC Open-Tubular Electrochromatography OTLC Open-Tubular Liquid Chromatography PA Polyacrylate PAHs Polycyclic Aromatic Hydrocarbons PCBs Polychlorinated Biphenyls PDMDPS Poly(dimethyldiphenylsiloxane) PDMS Poly(dimethylsiloxane) PEEK Polyetheretherketone PEG Polyethylene Glycol PheDMS Phenyldimethylsilane PLE Pressurized Liquid Extraction PMHS poly(methylhydrosiloxane) POLYCERAM Polymeric Ceramics PPY Polypyrrole PVA Poly(vinyl alcohol) RPLC Reversed-Phase Liquid Chromatography RPM Revolution per Minute RSD Relative Standard Deviation S/N Signal to Noise

PAGE 19

xvi SN Nucleophilic Substitution Se Surface concentration of adsorbed analytes in the solid extracting phase SAM Self-Assembled Monolayer SEM Scanning Electron Microscopy SFC Supercritical Fluid Chromatography SFE Supercritical Fluid Extraction SPE Solid-Phase Extraction SPME Solid-Phase Microextraction TEOS Tetraethylorthosilicate (or Tetraethoxysilane) TESP-PEO N-(Triethoxysilylpropyl)-O-Pol yethylene Oxide Urethane TFA Trifluoroacetic Acid TMOS Tetramethylorthosilicate (or Tetramethoxysilane) TR Templated Resin UV Ultraviolet VOC Volatile Organic Compound XANES X-Ray Absorption Near Edge Spectroscopy XPS X-Ray Photoelectron Spectroscopy

PAGE 20

xvii Novel Sol-Gel Titania-Based H ybrid Organic-Inorganic Coat ings for On-Line Capillary Microextraction Coupled to HighPerformance Liquid Chromatography Tae-Young Kim ABSTRACT Novel sol-gel titania-poly(dimethylsiloxane) (TiO2-PDMS) and titania-silicaN (triethoxysilylpropyl)-O-polyeth ylene oxide urethane (TiO2-SiO2-TESP-PEO) coatings were developed for capillary microextraction (CME) to pe rform on-line preconcentration and HPLC analysis of trace impurities in aqueous samples. Due to chemical inertness of titania, effective covalent bindi ng of a suitable organic ligand to its surface is difficult via conventional surface modification methods. In this research, sol-gel chemistry was employed to chemically bind hydroxy-termin ated poly(dimethylsiloxane) (PDMS) and N -(triethoxysilylpropyl)-O-polyethylene oxide ur ethane (TESP-PEO) to sol-gel titania and sol-gel titania-silica network, respectively. A met hod is presented describing in situ preparation of the titania-based sol-gel PDMS and TESP-PEO coatings and their immobilization on the inner surface of a fu sed-silica microextraction capillary. To perform on-line CME-HPLC, the sol-gel TiO2-PDMS or TiO2-SiO2-TESP-PEO capillary was installed in the HPLC injection port as an external sampling loop, and a conventional HPLC separation column was used for the li quid chromatographic separation. The sol-gel

PAGE 21

xviii TiO2-PDMS-coated microextraction capillary wa s used for on-line CME-HPLC analysis of non-polar and moderately polar analytes, a nd the sol-gel coatings showed excellent pH (1-13), and solvent (acetonitrile and methanol ) stabilities under elevated temperatures (150 C) over analogous non-sol-gel silica-based coatings. Extraction of highly polar analytes, especially from aqueous phases is not an easy task. Howe ver, the sol-gel TiO2SiO2-TESP-PEO-coated capillaries showed excellent capability of extracting underivatized highly polar analytes from aqueou s samples. This opens the possibility to employ sol-gel titania-based polar coatings fo r solvent-free extraction and trace analysis of target analytes in environmental and bi omedical matrices. To our knowledge, this is the first research on the use of sol-gel tita nia (or titania-silica)-b ased organic-inorganic materials as a sorbent in capillary microext raction. The newly developed sol-gel titania (or titania-silica)-based organic-inorganic hybrid extraction media provides an effective solution to coupling CME with HPLC (CME-HPLC), and this can be expected to become a powerful analytical tool in environmental investigations, proteomic research, early disease diagnosis and biomarker research. Being a combination of a highly efficient solvent free sample preconcentration techni que (CME) and a powerful separation method (HPLC), CME-HPLC poses to become a key anal ytical tool in solving complex chemical, environmental, and biomedical pr oblems involving complex matrices.

PAGE 22

1 CHAPTER ONE AN OVERVIEW OF SOLID-PHAS E MICROEXTRACTION (SPME) 1.1 Introduction to Sample Preparation Traditional sample preparation is a labor-i ntensive analytical process, and for a long time it has not been consider ed as a field of analytical chemistry [1]. More than 75% of analysis time is spent on sampling and samp le preparation steps [2]. However, due to increasing awareness of environmental and health concerns, and demands on fast and cost-effective analyses of trace organic com pounds, the analytical procedure for sample extraction and preconcentration from various environmental matrices has become very important. There are several steps in typical analytical procedur es as shown in Figure 1.1: sampling, sample preparation, separation, quan titation, statistical evaluation, and decision making. As emphasized in Figure 1.1, analytical steps follow one after another, and the next one cannot start until the preceding one ha s been completed. Therefore, the overall speed of the analytical procedur e depends on the slowest step. The chemical properties of the analytes are important parameters for the sample extraction, as are the properties of the liquid medium in which it is dissolved and the gaseous, liquid, supercritical fluid, or solid extr actant used to effect a separation. Of all the relevant solute properties, five chemical properties ar e fundamental to understanding extraction theory: vapor pressure, solubility, molecular weight, hydrophobicity, and acid

PAGE 23

2 dissociation. These essential pr operties determine the transport of chemicals in the human body, the transport of chemicals in the air-w ater-soil environmenta l compartments, and the transport between immiscible pha ses during analyti cal extraction.

PAGE 24

3 Figure 1.1 Steps in an analytical process. Reproduced from ref. [3] with permission.

PAGE 25

4 1.2 Fundamentals of Sample Extraction The fundamental concept of a sample extrac tion is to convert a real matrix into a sample format that is suitable for analysis various analytical techniques. This can be accomplished by employing different treatments and enrichment procedures with following common goals: Removing potential interferences from the sample to increase the selectivity of the method. Increasing the concentration of the anal yte hence the sensitivity of the method. Converting the analyte into a more suitabl e form for detection or separation, if necessary. Developing a robust and reproducible method which is independent of variations in the sample matrix. Conventional sample preparation met hods involve time-consuming and labor intensive processes and multi-step procedur es that often use la rge amounts of toxic organic solvents and are prone to analyt e losses. These characteristics make such methods very difficult to integrate with sa mpling and separation methods, especially for hyphenation and automation. Various traditional sample preparation techniques are used in analytical practice. These include Soxhlet extraction [4], liquid-liquid extraction (LLE) [5] accelerated solvent extraction (ASE) [6], microwave-as sisted extraction (MAE) [7], solid-phase extraction (SPE) [8], supercritical fluid extraction (SFE) [9], purge-and-trap [10], membrane extraction [11], static headspace an alysis [12], and others [13-16]. Table 1.1 summarizes conventional samp le preparation me thods for various sample matrices.

PAGE 26

5 Table 1.1 Conventional sample pr eparation techniques. Sample preparation technique Type of sample matrixa Solvent-free Ref. Soxhlet extraction s no [4] Accelerated solvent extraction (ASE) s no [6] Microwave-assisted extraction (MAE)s no [7] Supercritical fluid extraction (SFE) s yes [9] Liquid-liquid extraction (LLE) l no [5] Solid phase extraction (SPE) l no [8] Membrane extraction l no [11] Purge-and-trap extractio n s, l, g yes [10] Static headspace analysis s, l, g yes [12] as: solid samples; l: liquid sa mples; g: gaseous samples

PAGE 27

6 1.2.1 Extraction of analytes fr om solid sample matrices Many solid samples, such as soils, environmental solids, plant material, and polymers are highly insoluble an d usually cannot be directly studied. Strong acid is often used to digest the samples, but this pro cedure often decomposes the target analytes. Therefore, it is necessary to use a better techni que to extract the analyte of interest from the sample matrix with high efficiency, specificity, and selectivity to simplify the subsequent separation processes. 1.2.1.1 Soxhlet extraction Soxhlet extraction technique [4] has b een the most widely used method for exhaustive extraction analytes from solid samples. Soxhlet extraction involves placing the solid sample in a porous cellulous sample thimble placed in thimble holder. During operation the thimble is filled with fresh or ganic solvent from a distillation flask. During the process of extraction, the extracted analytes accumulate in the solvent, and are automatically siphoned into a distillation fl ask on a regular basis. These steps are repeated until exhaustive extr action of the analytes is achieved. Although the Soxhlet technique uses inexpensive equipment to ope rate, the processes i nvolved are quite slow and may require the use of large amounts of hazardous organic solvents to ensure complete extraction process; these are hazar dous to the environm ent and human health and are highly costly to dispose of. 1.2.1.2 Supercritical fluid extraction (SFE) Supercritical fluid extraction (SFE) is a fast and selective technique. Unlike Soxhlet extraction, which requires significan t amounts of organic solvents, SFE uses compressed carbon dioxide characterized by low viscosity, high volatility and high solute

PAGE 28

7 diffusion rates. However, the relatively low pol arity of carbon dioxide is one of the major problems, making it unsuitable for most polar samples including pharmaceuticals and drugs. Therefore, SFE often requires the use of polar organic modifiers to extract analytes of higher polarity. In addition, using high pur ity carbon dioxide can be expensive; SFE equipment is generally heavy which makes the technique incompatible with field analyses [17,18]. Also the high pressures required have often caused troubles for automation. 1.2.1.3 Microwave-assisted extraction (MAE) Microwave-assisted extraction (MAE) u tilizes microwaves to facilitate the digestion of solid samples by focusing energy into the sample, resulting both in heating and increased agitation [19]. High efficien cy is the major advantage of MAE, which offers fast mass transfer and short extraction time with less solvent. However, MAE often shows limitations in using solvents because, some of them do not absorb microwave. Moreover, cooling and filtration after extraction delays the ov erall process. Since MAE is an exhaustive technique, the extract often contains impurities that requires cleanup prior to analysis and also requir es the use of environmentall y hazardous organic solvents. 1.2.1.4 ccelerated solvent extraction (ASE) Accelerated solvent extraction (ASE) [6 ] is also known as pressurized fluid extraction (PFE) or pressurize d liquid extractio n (PLE). It uses high temperature (100 to 180 C) and high pressure (1500 to 2000 psi) to accelerate the extraction of organic analytes from solid matrices and to e nhance the extraction efficiency. The major advantage of ASE is that it uses minimal am ount of organic solvent and is fast, fully automated, and easy to use. However, th e equipment of ASE is expensive, and

PAGE 29

8 concentration and/or cleanup processes are often required prior to analysis. 1.2.2 Extraction of analytes fr om liquid sample matrices Traditionally, partitioning into an immiscible solvent, trapping, or evaporation [5] has been used to isolate analytes from liqui d matrices. Collected li quid samples must be representative and maintain compositional integr ity prior to analysis. Because of the great diversity of liquid samples, there is no unive rsal sampling technique or typical sampling equipment. Liquid samples can be taken from diverse sources: surface waters, groundwaters, drinking water, industrial waters, physiological fluids, and so on. 1.2.2.1 Liquid-liquid extraction (LLE) Liquid-liquid extraction (LLE) [5], or so lvent extraction, is one of the oldest techniques, but most widely used. It involves the distri bution of sample components between two immiscible liquid phases. The most common LLE method for a liquid matrix is to use a separatory funnel to extract any organic compounds from aqueous phase into a nonpolar organic phase by phase separation. This method typically requires using a large volume of organic solvent, and the extraction has to be repeated several times. In addition, drying and cleanup proces ses are often required, which make the overall process slow and costly. 1.2.2.2 Solid-phase extraction (SPE) The historical development of solid-pha se extraction (SPE) has been traced by various authors [20,21]. The most commonly cite d benefits of SPE methods that led to early advances relative to LLE are reduced the amount of solvent required, shorter extraction time, and lower cost. SPE uses the disposable pre-packaged cartridge for the extraction of analytes in solution [20,22,23], achieved through non-equilibrium,

PAGE 30

9 exhaustive removal and accumulation of analyt es from a flowing liquid sample matrix via retention on the solid sorbent contained in the cartridge. However, SPE is still a multistep process that is prone to loss of analyte if it is not fully automated and still requires organic solvents for the elution step. SPE is limited to semi-volatile compounds, and the boiling points of the anal ytes must be above those of the solvents [24]. Also, SPE cartridges are not reusable because the cartrid ges are usually dicarded after one extraction. In addition, SPE suffers from poor reproduc ibility and high carry over problems [25]. 1.2.2.3 Membrane extraction techniques Membrane extraction techniques [11] i nvolve two simultaneous processes where the analytes are extracted from the matrix by the membrane, and then analytes are removed from the membrane using a stripping phase. This method is only applicable for nonpolar volatile and semi-vol atile compounds, but has lim ited application to polar compounds as well. The other limitations of th is technique include system carryover due to slow response of the membrane to changes in concentration, and difficulty to interface this technique with separation instruments. 1.2.3 Extraction of analytes fr om gaseous sample matrices Generally volatile organic compounds (VOCs) do not require much sample preparation. They can be directly separate d and analyzed by gas chromatography (GC) [26,27]. The whole sample is either gas or solid and liquid matric es containing volatile organic compounds. Unlike othe r extraction techniques, th e volatile analytes do not equilibrate between the gas phase and sample matrix, instead, a fl owing gas continuously transfers the analytes from the sample matrix. However, the analytes of interest are often at low concentrations and near their limits of detection. Also, the high diffusion rates in

PAGE 31

10 gases create the problem that the sample is hard to maintain from the collection point to the analyzer. Typical gas phase extractions include static head space and dynamic headspace (also known as purge-and-trap) analyses. 1.2.3.1 Static headspace analysis In static headspace [12], also known as equilibrium headspace extraction or simply as headspace, the volatile analytes in the sample matrix diffuse into the headspace in a vial, and the concentration of the anal yte in the headspace reaches equilibrium with concentration in the sample matrix. Once th e equilibrium is established after some time determined from the calibration plot, a small vo lume of the headspace gas is injected into a GC. Using static headspace method is advant ageous due to its eas e in initial sample preparation. It is more convenient for qua litative analysis since the sample can be placed directly into the headspace vial a nd analyzed with no such additional preparation. However, this technique is less sensitive because there is no mechanism for sample preconcentration. Also, this technique only applies to volatile samples. 1.2.3.2 Dynamic headspace analysis In dynamic headspace, also known as purge-a nd-trap [10], a carrier gas is bubbled through either solid or liquid samples cont aining volatile organic compounds (VOCs), and VOCs captured in a sorbent trap. The sorbent trap is then heated to desorb the analytes from the trap for GC analysis. Like static headspace analysis, dynamic headspace technique also relies on the volat ility of the analytes for extraction from sample matrices. However, equilibration is no t achieved between the analytes and sample matrix. Instead, the analytes are removed from the sample matrix continuously by a flowing gas. The major disadvantages of dynamic headspace analysis are instrument

PAGE 32

11 carryover and incompatibility with separation instruments. 1.3 Solid-Phase Microextraction (SPME) Conventional sample preparation techni ques are time-consuming, labor-intensive, involve multistep processes, and often require large quantities of organic solvents. In addition, most of the conventional sample prep aration techniques are not suitable for field analysis, and often require a dditional processing. To overcome these inherent problems of conventional sample preparation techniques, solid-phase micr oextraction (SPME), a solven t-free sample preparation technique, was developed in the early 1990s by Pawliszyn and co-workers [28,29]. In SPME, the extracting phase coated on the surface of a fused-silica fiber or capillary plays an important role in the extr action process. The extraction in SPME is an equilibrium process between the extracting pha se and the analyte to be extracted. Once the equilibrium is established, the extracti ng phase cannot accumulate the analyte from the sample matrix, and there is a direct rela tionship between sample concentration and the amount of analyte extracted. The simple format of the classical SPME device based on the HamiltonTM 7000 series microsyringe was introduced as th e first SPME device. The metal rod in the microsyringe is replaced with stainless stee l microtubing having a sl ightly larger inner diameter than the outer diameter of the fu sed-silica rod. Generally, the first 5 mm of the protective external coating is burned off from 1.5 cm long fused-silica rod, and this end is installed in the microsyri nge using high temperature e poxy glue to protect from mechanical damage. In this case, fiber is only exposed for extrac tion and desorption

PAGE 33

12 processes when the plunger is moved. Figure 1.2 illustrates the confi guration of the first commercial SPME device [30]. A small piece of fused-silica fiber is coated with a polymeric sorbent covering a small segment (~ 1 cm) of it at one of the ends. The thickness of the SPME coating generally ranges between 10 a nd 100 m. Thermally stable polar or nonpolar polymeric sorbents that allow fast solute diffusion are commonly used in SPME. Such polymers include poly(dimethylsiloxane) (PDMS), divinylbenzene (DB), polyacrylate (PA), Carboxen (CAR; a carbon molecular sieve), and Carbowax (CW; polyethylene glycol). Generally, there are two steps in the SPME process: (a) extraction of the analytes on the fiber coating and (b) desorption of he extracted molecules into an analytical instrument for analysis. Using a polymeric sorbent coated SPME fiber, the analytes present in the sample medium are directly extracted on the coated sorbent of the SPME fiber in the process of reaching extraction equilibrium with the sample matrix. The extracted analytes are then desorbed into an instrument for separation and analysis. The desorption process is typically done by placing the fiber in a GC injection port. It can be also performed in an high-performance liquid chromatography (HPLC) by introducing SPME-HPLC interface. The whole process can be automated and coupled to GC [31,32] or HPLC [33]. There are two basic modes of extraction in SPME: direct (immersion) SPME, and headspace SPME (HS-SPME) [3]. For the analys is of gaseous and relatively clean liquid samples, direct SPME can be applied. Howe ver, HS-SPME is better suited for the analysis of solid samples and dirtier li quid samples containing volatile analytes. When SPME was introduced, it was used to analyze relatively volatile

PAGE 34

13 Figure 1.2 Design of the first commercial SPM E device made by Supelco. Reproduced from ref. [30] with permission.

PAGE 35

14 compounds in environmental samples. Now th e application of SPME has extended to a wide variety of sample matrices and analyt es. To date, SPME has been used successfully to analyze various analytes in solid, liquid, or gaseous sample matrices, such as pesticides [34,35], phenols [36,37], polychlorinated biphe nyls (PCBs) [38,39], polycyclic aromatic compounds (PAHs) [34,40], and so me inorganic compounds [41]. SPME is very simple, fast, easily automate d, portable, sensitive, and inexpensive. Unlike conventional sample preparation met hods SPME does not require the use of toxic organic solvents, and only sm all volumes of sample are ne eded for analysis. Also SPME can be easily automated with analytical inst ruments, such as gas chromatography (GC) [31,32], high-performance liquid chromato graphy (HPLC) [33,42], and capillary electrochromatography (CEC) [43,44]. 1.3.1 Principles of SPME The principles of SPME have been pres ented by Pawliszyn and co-workers [4547]. SPME can be used for aqueous or gaseous samples. In both cases, there is proportional relationship (so cal led, the distribution constant ) between the concentration of analyte in the sample and the amount of analyte extracted by the extracting phase when the latter is at equilibrium with sample matrix. In direct (immersion) SPME, the mathem atical relationship of the distribution constant (Kfs) for aqueous samples is described as following: Kfs = CfCs(1-1) where, Kfs: distribution constant between extracti ng phase and aqueous sample matrix,

PAGE 36

15 Cf : equilibrium analyte concentra tion in the extracting phase, Cs: equilibrium analyte concentration in the aqueous sample. The amount of analyte in the extracting phase on the fiber (nf) is given by [3]: nf = Kfs V s V fC0KfsVf + Vs(1-2) where, nf: number of moles of the analyte(s) extracted by the extracting phase, Vf: volume of the extracting phase, Vs: volume of the aqueous sample, C0: initial concentration of a given analytes in the sample. Since the volume of the sample (Vs) in the aqueous phase is very large or practically infinite (KfsVf << Vs) compared to the volume of the extracting phase (Vf), the term, KfsVf, in the denominator can be ignore d. Therefore, the amount of analyte extracted on the fiber coating can be simply expressed as: nf = K fs V fC0(1-3) Equation (1-3) shows that once the establ ishment of equilibrium is reached, the amount of analytes extracted in the extracting phase will be di rectly related to the initial concentration of the analytes in the samp le and will not depend on the sample volume. During the extraction process, the concentratio n of the analytes in the extracting phase rapidly increases first, then more slowly until equilibrium is reached. The amount of extracted analytes on the fiber is proportion al to the volume of extracting phase. Therefore, the thicker the extr acting phase, the more analytes will be extracted onto the

PAGE 37

16 SPME fiber, however the equilibration time te nds to be delayed due to the longer diffusion time for the analytes from the sample to the extracting phase. The extraction efficiency and sensitivity depend on the distribution constant. To achieve higher selectivity and sensitivity, a hi gh distribution constant is desirable. The distribution constant does not simply depend on the fiber coating material, but also depends on various operating parameters including temperature, pressure, and sample matrix conditions such as pH, salt concentra tion, and concentration of organic component, which need to be optimized for maximum tran sfer of analytes to the sorbent phase for extraction. In headspace SPME (HS-SPME), aqueous sample matrices containing volatile compounds are placed in a sealed container at a constant temperature until equilibrium of the analytes between gaseous and aqueous phases is reached in the closed container. Then the SPME fiber is inserted into the sealed container and exposed to the headspace above the aqueous sample matrix for a certain period of time to extract the analytes in the gaseous phase. The mass of an analyte extracted by the ex tracting phase on the fiber is related to the overall equilibrium of the analyte in the three-phase system. Since the total mass of an analyte should remain constant during the extraction: Cf Vf + C0Vs = Ch Vh + Cs Vs(1-4) where, C0: the initial concentration of th e analyte in the sample matrix, : at equilibrium, Cf : analyte concentration in the extracting phase,

PAGE 38

17 Vf: the volume of the extracti ng phase on the SPME fiber, Vh: the volume of the gaseous phase (or headspace), Ch: the concentration of the analyte(s) in the gaseous phase (or headspace), Cs: the concentration of the an alyte in the aqueous phase, Vs: the volume of the sample matrix. The distribution constant (Kfh) between the extracting phase and the headspace can be defined as: Kfh = Cf Ch (1-5) And, the distribution constant (Khs) between the headspace and the sample matrix can be defined as: Khs = Ch Cs (1-6) The mass of the analyte extr acted by the sorbent phase, nf = Cf Vf can be expressed as: nf = K fh K hsVfC0VsKfhKhsVf + KhsVh +Vs(1-7) Also, Kfs = KfhKhs = KfgKgs, since the extracting pha se/headspace distribution constant, Kfh, can be approximated by the extractin g phase/gas distribution constant, Kfg. The headspace/sample distribution constant, Khs, can be also approximated by the gas/sample distribution constant, Kgs. If the moisture effect in the gaseous headspace is neglected, the equation (1-3 ) can be rewritten as: nf = K fsVfC0VsKfsVf + KhsVh + Vs(1-8)

PAGE 39

18 In contrast to the equation (1-2), the distribution of the analyte between the gaseous (headspace) and the aqueous phases (s ample matrix) is considered. Equation (18) shows that the amount of extracted analyte is independent of the location of the fiber in the system. It may be placed in the gaseous phase or directly in the aqueous phase as long as all other parameters, such as the vol ume of the extracting phase, headspace, and the sample matrix, remain constant. Direct SPME and HS-SPME are illustrated in Figure 1.3.

PAGE 40

19 Figure 1.3 Modes of SPME operation: (A) direct (immersion) SPME, (B) headspace SPME (HS-SPME). Adapted from ref. [48].

PAGE 41

20 1.3.2 Conventional SPME coatings To achieve good selectivity for the analytes of interest, the choice of the most suitable coating is essential. The principle of “like dissolves lik e” can be applied to SPME fiber selection. Conventional SPME extracting phases (or sorbent coatings) are generally nonbonded, partially cross-linked, or highly cross-linked. They are held on the outer surface of a fused-silica fiber (or on the inner surface of a fused-silica capillary for in-tube SPME) and serve as the extraction medium in which the analytes are preferentially sorbed an d/or preconcentrated. 1.3.2.1 Fiber SPME coatings Currently several coatings with different thicknesses are commercially available: poly(dimethylsiloxane) (PDMS), poly acrylate (PA), the mixed phases of poly(dimethylsiloxane)-poly(divinyl benzene) (PDMS-DVB), Carboxenpoly(dimethylsiloxane) (CAR-PDMS), Carbow ax-poly(divinylbenzene) (CW-DVB), and Carbowax-templated resin (CW-TR). Table 1.2 shows the summar y of commercially available SPME coatings on fibers. Poly(dimethylsiloxane) (PDMS) and polyacrylate (PA) were the first coated fibers to be used in SPME. Especially, PD MS fibers are the most popular SPME coatings to date. Due to the difficulties in stabilizi ng thick coatings through cross-linking reaction, the PDMS fiber with 7 m coating thickne ss is the only commercially available crosslinked one. The cross-linked SPM E coating is very rugged and is stable up to about 340 C while other two are not. PDMS is a nonpolar polymer which usually extracts nonpolar analytes such as BTEX compounds (benzene, toluene, ethylbenzene, and xylene) [46,49], alkanes [50,51], polycycl ic aromatic hydrocarbons (PAHs) [52,53],

PAGE 42

21 Table 1.2 Commercially available fiber coati ngs for SPME and their applications. Adapted from ref. [2,54,55]. Fiber Coating Coating Thickness (m) Max. Temp. (for GC use) (C) Applications Poly(dimethylsiloxane) (PDMS) 100c 30c 7a 280 280 340 GC/HPLC, Nonpolar organic compounds such as VOCs, PAHs and BTEX Polyacrylate (PA) 85b 320 GC/HPLC, Polar organic compounds such as triazines and phenols PDMS-divinylbenzene (PDMS-DVB) 65b 60b,d 270 GC/HPLC, PAHs, aromatic amines, VOCs Carboxen-PDMS (CAR-PDMS) 85b 75b 320 320 GC/HPLC, VOCs and hydrocarbons Carbowaxdivinylbenzene (CW-DVB) 70b 65b 265 265 GC/HPLC, Polar analytes such as alcohols and polar compounds Carbowax-templated resin (CW-TR) 50b HPLC, Anionic surfactants and aromatic amines DivinylbenzeneCarboxen-PDMS (DVB-CAR-PDMS) 50/30b 270 GC/HPLC, Odors and flavors aBonded phase; bPartially cross-linked phase; cNon-bonded phase; dGC application only.

PAGE 43

22 organic compounds (VOCs) [56,57] and some pe sticides [58,59]. However, it may also extract more polar analytes if the extracti ng conditions are optimized such as pH, salt concentration, and temperature. Polyacrylate (PA) is a more polar coati ng and suitable for the extraction of more polar compounds such as alcohols [6062], diols [60,63], aldehydes [62,64,65], ketones [62,65], esters [62,64], amines [66,67], acids [60,63], phenols and their derivatives [37,68] and some pesticides [37,42,69]. However, the extraction time tends to be longer because diffusion coefficients in PA are smaller than in PDMS [37,68,70]. The mixed phase coatings have been more recently introduced. They have complementary properties compared with PDMS and PA, and are prepared with a blend of two phases in which the porous particles are embedded in the partially cross-linked polymeric phase. For example, porous particles of poly(divinylbenzene) (DVB), or Carboxen (CAR) are blended in PDMS, to produce PDMS-DVB and CAR-PDMS, respectively. Similarly, por ous particles of poly(diviny lbenzene) are blended in Carbowax (CW) to prepare CW-DVB, which are suitable for ex tracting more polar analytes such as alcohols and ethers [71]. In addition, CARPDMS fibers possess a larger surface area and show good extr action capability for organi c analytes, such as low molecular weight VOCs from the air [57]. Carbowax-templated resin (CW-TR) is a partially cross-linked phase with bi-polar properties. CW-TR is used for the extraction of surfactants from aqueous samples. Generally speaking, the mixed phase coatings demonstrate better affinity for polar analytes, and provide very high selectivity due to the existence of two sorbents in the coating, but they tend to possess lower mechanical stability. Figure 1.4 demonstrates a graphical scheme for choosing a SPME fiber based on polarity and

PAGE 44

23volatility of the analytes of interest. The first SPME fibers were developed for GC applications, and they often created problems when applied to HPLC [40]. Currently, some coating fibers have been developed for HPLC applications. The analyte desorption process in HPLC can only be performed when the fiber coating is stable to the addition of organic mobile phases, and can perform without dissolution and swelling. Among commercially available SPME fibers, only the bonded phases are compatible with all organic solven ts, and are recommended for use with HPLC.

PAGE 45

24 Figure 1.4 Graphical scheme for choosing SPME fiber coating. Reproduced from ref. [3] with permission.

PAGE 46

25 1.3.2.1.1 Direct (immersion) SPME In the direct immersion SPME (or simply direct SPME), the sorbent coated fiber is directly immersed into the sample matrix, and the analytes of interest are extracted directly from the sample to the coating. Ofte n some type of agitation, such as rapid vial movement, stirring, and sonication, need to be employed to enhance the extraction from aqueous sample matrix due to slower diffusi on coefficients compared to gaseous samples [57,72]. 1.3.2.1.2 Headspace SPME (HS-SPME) Headspace SPME (HS-SPME) would be cons idered for the same analytes as static or dynamic headspace extractions descri bed in section 1.3. In HS-SPME, the volatile analytes first need to be equilibrate d between the sample matrix (either solid or liquid) and the headspace in the closed cont ainer. Once the pre-equilibrium is reached, the coated fiber is inserted to extract the an alytes in the headspace until the equilibrium is reached between the coating and the headspac e. HS-SPME is preferred to protect the fiber coating from especially complex or dirty samples, such as those at very high or low pH, or those with large molecules. 1.3.2.2 In-tube SPME In conventional fiber-based SPME, still ther e exist a number of shortcomings that need to be overcome. These include inadequate thermal and solvent stability of conventionally prepared sorbent coatings [52], low sample capacity, difficulties associated with the immobilization of thic k coatings, susceptibility of the fiber (especially the coated end) to mechanical damage [73,74], and technical difficulties associated with coupling of fi ber-based SPME to li quid-phase separati on techniques [42,

PAGE 47

26 75]. In-tube SPME methods [33,76] present a c onvenient means for coupling SPME to other analytical instruments, such as HPLC, CE C, MS, FTIR etc. It is also suitable for automation, which not only reduces the total an alysis time, but prov ides better accuracy and precision compared to traditional methods. There are two types of in-tube SPME: static in-tube SPME and dynamic in-tube SPME (also called capillary microextraction (C ME) [77]). In static in-tube SPME, the extracting phase is not exposed directly to th e sample matrix. Instead, it is contained in the protective tubing (needle) without a ny flow of the sample through it, and the extraction occurs through the static gas pha se present in the needle. However, in dynamic in-tube SPME, the sample is brought in to direct contact w ith the extracting phase by means of a continuous flow of the sample through the tube. Coupling of in-tube SPME to HPLC is especially important for the analysis of a wide range of less volatile or thermally lab ile compounds [78] that are not amenable to GC separation. In the open tubular format of SPME, a sorbent coating is applied to the inner surface of a capillary as shown in Figure 1.5. This a lternative format provides an effective solution to the problem associated with the mechanical damage of sorbent coating frequently encountered in conventi onal fiber-based SPME where the coating is applied on the outer surface of the fiber. In th is new format of SPME, a segment of wallcoated capillary GC column is commonly used [33, 76,78] for the dire ct extraction of organic analytes from an aqueous medium, a nd the analytes in the aqueous samples are transported into the capillary, so the analytes can be directly extracted and concentrated in the extracting phase by repeated draw and eject cycles of the sample solution. Figure

PAGE 48

27 1.6 illustrates the comparison of extraction process for the transf er of the analytes in fiber SPME and in-tube SPME. In order to develop the in-tube SPMEHPLC method, extraction and desorption parameters need to be optimized, including the selection of the extracting phase, pH, extraction flow rate, desorption solvent, and separation conditions. In-tube SPME methods were mainly hyphenated with HP LC-UV and LC-MS. To perform HPLC analysis, the extracted analytes are transfe rred to the HPLC column by desorbing them with an appropriate mobile phase. In addition, the hyphena tion with ion chromatography with conductivity detection (IC-CD) [79] a nd GC with flame ionization detection (GCFID) [80] were also reporte d. Table 1.3 shows a list of capill aries and their applications for in-tube SPME. Classification of sample pr eparation techniques is illustrated in Figure 1.7.

PAGE 49

28 Figure 1.5 Schematic of coatings in (A) fibe r-based SPME and (B) in-tube SPME. Adapted from ref. [81].

PAGE 50

29 Figure 1.6 Extraction of analytes by (A) fiber SPME and (B) in-tube SPME. Reproduced from ref. [82] with permission.

PAGE 51

30 Table 1.3. Applications of in-tub e SPME technique for various samples. Adapted from ref. [82]. Capillary Detection Application Omegawas 250 HPLC-UV LC-MS Phenylureas and carbamates -blockers, amphetamine s, ranitidines, and heterocyclic amines DB-WAX LC-MS Phenoxy acid herbicides PPY-coated HPLC-UV LC-MS IC-CD PAHS and aromatic amines Organoarsenic compounds, amphetamines, phenylureas and carbamates, -blockers, amphetamines, catechins, and caffeine Inorganic anions BP-20 PEG GC-FID BTEX and phenols Fiber-packed PEEK™ (Polyetheretherketone) HPLC-UV Phthalates Supel-Q PLOT HPLC-UV LC-MS Phthalates, phenols, and isoflavones Trimethyllead, thiethyllead, and benzodiazepines MIP-packed HPLC-UV Propranolol Wire-packed DB-1 HPLC-UV Tricyclic antidepressants Fiber-packed DB-5 CE-UV Tr icyclic antidepressants

PAGE 52

31 Figure 1.7 Classification of sample prepar ation techniques. Reproduced from ref. [83] with permission.

PAGE 53

32 1.4 Sol-Gel Capillary Microextraction (CME) and Sample Preconcentration Although in-tube SPME (or CME) has great pr ospects in trace an alysis, especially in liquid-phase sample matrices, the technolo gy needs further improvements in a number of areas to achieve its full an alytical potential. First, segments of GC columns that are commonly used for sample preconcentration have thin coatings that limit the sorption capacity, and hence, the extr action sensitivity of in-tube SPME. Second, the sorbent coatings in such microext raction capillaries usually are not chemically bonded to capillary inner walls, which limits thei r thermal and solvent stabilities. Third, conventionally prepared GC coatings that ar e used in in-tube SPME capillaries inherently possess poor pH stability. This places serious limitations on the range of solutes amenable to CME-HPLC analysis. Low pH stability of in-tube SPME coatings practically excludes the applicab ility of the technique to high-pH samples or analytes that require high-pH solvent systems for desorp tion from the microextraction capillary. Therefore, development of methodologies for the creation of high pHand solventresistant sorbent coatings is an important area in the future development of in-tube SPME, which is expected to play a major role in effective hyphenation of this sample preconcentration technique w ith liquid-phase separation t echniques that commonly use organo-aqueous mobile phases with a wide range of pH conditions [84]. Sol-gel chemistry offers a great promise to overcome the inherent shortcomings of conventional SPME and in-tube SPME. Sol-gel chemistry provides a simple and convenient way of developing new forms of sol-gel extraction media for CME analysis via chemical incorporation of organic compone nts into the inorganic polymeric structures. Chemically bonded sol-gel extracting phas es can withstand harsh experimental

PAGE 54

33 conditions, such as high temperatures, extrem e pHs, and organic solvents. In addition, a simple sol-gel coating technology allows in situ creation of chemically bonded thick and thin coatings. The sol-gel chemistry offers several promising directions in the SPME extracting phase (open tubular coatings and monolithic bed) technology. 1.4.1 Principles of CME For all chemical extraction methods, the fundamental thermodynamic principle is common and involves distributio n of analytes between the sample matrix and the extracting phase. When a liquid is used as the extracting phase then the distribution constant (Kes) defines the equilibrium conditions and enrichment factors achievable in the technique. Kes = aeas= CeCs (1-9) where, Kes: distribution constant between liquid extracting phase and sample matrix, ae : activities of analytes in th e liquid extracting phase, as: activities of analytes in the sample matrix, Ce : analytes concentration in the liquid extracting phase, Cs: analytes concentration in the sample matrix. For solid extracting phase adsorption equilibria can be expressed as: Ks es = SeCs (1-10) where, Ks es: distribution constant between solid extracting phase an d sample matrix,

PAGE 55

34 Se : surface concentration of adsorbed anal ytes in the solid extracting phase. The expression (1-10) is similar to (1-9 ) with the exception that extraction phase concentration is replaced with surface concen tration. The interstiti al linear extraction phase velocity (ue) can be expressed as [85]: ue = us(1 + k0) (1-11) and us = L t0(1-12) where, ue: interstitial linear extraction phase velocity (m/s), us: chromatographic linear velo city of the sample thr ough the tube (sample flow rate) (m/s), k0: ratio of the intraparticulate void volume to the interstitial void space (partition ratio), L: length of the extractio n vessel (capillary) (m), t0: time required to remove one void volum e of the extracting phase from the extraction vessel in chromatography (can also be rewritten as te in in-tube SPME) (sec). Since the partition ratio (k) can be defined as: Ks V eVv (1-13) k =es where, Ve: volume of the extracting phase (cm3), Vv: void volume of the tubing cont aining the extracting phase (cm3), Kes s: extracting phase/sample matr ix distribution constant.

PAGE 56

35 By plugging the equations (1 -12) and (1-13) into (111), the equation (1-11) can be rewritten in terms of te: Ks V eVv (1-14) 1 + te = ueL es The equation (1-14) shows that the time for the extraction required to reach the equilibrium is proportional to the length of the capillary and inversely proportional to the linear flow rate of the sample. The increases of the extracting phase/sample distribution constant and the volume of the extractin g phase also prolong the extraction time. However, extraction time decreases with an increase of the void volume of the capillary. In other words, thicker coatings with a smaller capillary void volume will take a longer time to establish extraction equilibrium comp ared to the thinner coatings on a capillary with the same internal diameter. The CME technique is somewhat different from the conventional SPME methods in terms of the equilibrium process, except dynamic in-tube SPME. In static SPME methods, an equilibrium between the sample matrix and the extracting phase is achieved which directly represents the concentration of the analytes in the sample. However, in flow-through techniques such as dynamic in -tube SPME and CME, the mass transfer equilibrium is achieved between the extracting phase and the analytes in sample matrix, and the analytes can be preconcentrated in the extracting phase. In addition, the sensitivity and selectivity of the extracted an alytes in the sorbent phase may be different depending on the type of the extracting phas e. Therefore, sorben t coating technology plays an important role in the SPME. All other paramete rs such as temperature, pH,

PAGE 57

36 pressure, and sample matrix conditions, as described in the c onventional SPME methods, also affect the mass transfer and distributi on for CME applications. Figure 1.8 illustrates mass transfer of the analytes in the sol-gel CME capillary.

PAGE 58

37 Figure 1.8 Extraction and preconcentration of analytes by sol-gel CME capillary.

PAGE 59

38 1.5 References for Chapter One [1] F. Settle, Handbook of Instrumental Te chniques for Analytical Chemistry, Prentice Hall PTR, Upper Saddle River, NJ, 1991. [2] M.D.F. Alpendurada, J. Chromatogr. A 889 (2000) 3. [3] J. Pawliszyn, Solid -Phase Microextraction: Theory and Practice, Wiley-VCH, New York, NY, 1997. [4] F. Soxhlet, Dinger’s Polyt. J. 232 (1879) 461. [5] I.M. Kolthoff, E.B. Sandell, E.J. Meeha n, S. Bruckenstein, Quantitative Chemical Analysis, 4th ed., Macmillan, New York, 1969. pp. 335-375. [6] B.E. Richter, B.A. Jones, J.L. Ezzel, N. L. Porter, N, Abdalovic, C. Pohl, Anal. Chem. 68 (1996) 1033. [7] A. Zlotorzynski, Crit. Re v. Anal. Chem. 25 (1995) 43. [8] K. Coulibaly, I.J. Jeon, Food Rev. Int. 12 (1996) 131. [9] S.B. Howthorne, Anal. Chem. 62 (1990) 633A. [10] M.M. Minnich, J.H. Zimmerman, B.A. Schumacher, J. AOAC Int. 79 (1996) 1198. [11] J.A. Jonsson, L. Mathiasson, J. Sep. Sci. 24 (1002) 495. [12] B. Kolb, L.S. Ettre, Static HeadspaceGas Chromatography: Theory and Practice, Wiley, New York, NY, 1997. [13] S.K. Poole, T.A. Dean, J.W. Oudsema, C.F. Poole, Anal. Chim. Acta 236 (1990) 3. [14] T.S. Reighard, S.V. Olesik, Crit. Rev. Anal. Chem. 26 (1996) 61. [15] E.V.D. Vlis, M. Mazereeuw, U.R. Tjaden, H. Irth, J.V.D. Greef, J. Chromatogr. A 687 (1994) 333. [16] M. Gilar, E.S.P. Bouvier, B.J. Comp ton, J. Chromatogr. A 909 (2001) 111. [17] M.A. McHugh, V.J. Krukonis, Supercritical Fluid Extraction: Principles and Practice, Butterworths, London, 1986.

PAGE 60

39 [18] R.M. Smith, J. Chromatogr. A 856 (1999) 83. [19] V. Lopez-Avila, in: D. Barcelo (Ed.), Sample Handling and Trace Analysis of Pollutants, Elsevier, Amsterdam, 1999. [20] I. Liska, J. Chromatogr. A 885 (2000) 3. [21] N.J.K. Simpson (Ed.), Solid-Phase Extr action: Principles, Techniques, and Applications, Marcel Dekker, New York, 2000. [22] J.S. Fritz, Analytical Solid-Phas e Extraction, Wiley–VCH, New York, 1999. [23] C.F. Poole, A.D. Gunatillekam, R. Se thuraman, J. Chromatogr. A 885 (2000) 17. [24] Z. Zhang, M.J. Yang, J. Pawliszyn, Anal. Chem. 66 (1994) 844A. [25] T.S. Sun, J. Jia, D. Zhong, Y. Wang, Anal. Sci. 22 (2006) 293. [26] H. Hachenberg, A.P. Schmidt, Gas Chromatographic Headspace Analysis, Heyden, London, 1977. [27] H.M. McNair, E.J. Miller, Basic Ga s Chromatography, Wiley, New York, 1997. [28] R.P. Belardi, J.B. Pawliszyn, Water Pollut. Res. J. Can. 24 (1989) 179. [29] C.L. Arthur, J. Pawliszyn, Anal. Chem. 62 (1990) 2145. [30] H. Lord, J. Pawliszyn, J. Chromatogr. A 885 (2000) 153. [31] R. Eisert, K. Levsen, J. Chromatogr. A 737 (1996) 59. [32] R. Eisert, J. Pawliszyn, J. Chromatogr. A 776 (1997) 293. [33] R. Eisert, J. Pawliszyn, Anal. Chem. 69 (1997) 3140. [34] H. Daimon, J. Pawliszyn, Anal. Chem. 34 (1997) 365. [35] I. Vlor, J.C. Molto, D. Apraiz, G. Font, J. Chromatogr. A 767 (1997) 195. [36] K.D. Buchholz, J. Pawliszyn, Anal. Chem. 66 (1994) 160. [37] M.R. Lee, Y.C. Yeh, W.S. Hsiang, B. H. Hwang, J. Chromatogr. A 806 (1998) 317. [38] Y. Yang, D.J. Miller, S.B. Hawthor ne, J. Chromatogr. A 800 (1998) 257.

PAGE 61

40 [39] Y. Yang, S.B. Hawthorne, D.J. Miller, Y. Liu, M.L. Lee, Anal. Chem. 70 (1998) 1866. [40] M.R. Negrao, M.F. Alpendurada, J. Chromatogr. A 823 (1998) 221. [41] T. Gorecki, J. Pawliszyn, Anal. Chem. 68 (1996) 3008. [42] K. Jinno, T. Muramatsu, Y. Saito, Y. Kiso, S. Magdic, J. Pawliszyn, J. Chromatogr. A 754 (1996) 137. [43] S. Li, S.G. Weber, Anal. Chem. 69 (1997) 1217. [44] C.W. Whang, J. Pawliszyn, Anal. Commun. 35 (1998) 353. [45] D. Louch, S. Motlagh, J. Pawlis zyn, Anal. Chem. 64 (1992) 1187. [46] C.L. Arthur, L.M. Killam, K.D. Buchholz, J. Pawlisz yn, J.R. Berg, Anal. Chem. 64 (1992) 1960. [47] Z. Zhang, J. Pawliszyn, Anal. Chem. 65 (1993) 1843. [48] J. Pawliszyn, B. Pawliszyn, M. Pawliszyn, Chem. Educator [Online] 1997, 2(4); DOI 10.1007/s00897970137a. [49] I. Valor, C. Cortada, J.C. Molto, J. High Resolut. Chro matogr. 472 (1996) 472. [50] A. Saraullo, P.A. Martos, J. Pawliszyn, Anal. Chem. 69 (1997) 1992. [51] B. Schafer, P. Hennig, J. High Resolut. Chromatogr. 20 (1997) 217. [52] S.L. Chong, D. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [53] K. Kolar, M. Ciganek, J. Malech a, J. Chromatogr. A 1029 (2004) 263. [54] A. Penalver, E. Pocurull, F. Borrull, R.M. Marce, Trends Anal. Chem. 18 (1999) 557. [55] S. Mitra, Sample Preparation Techni ques in Analytical Chemistry, WileyInterscience, Hoboken, NJ, 2003. [56] F.J. Santos, M.T. Galceran, D. Fr aisse, J. Chromatogr. A 742 (1996) 181. [57] M. Chai, J. Pawliszyn, Envir on. Sci. Technol. 29 (1995) 693.

PAGE 62

41 [58] J. Dugay, C. Miege, M.C. Hennion, J. Chromatogr. A 795 (1998) 27. [59] Z. Zhang, J. Pawliszyn, Anal. Chem. 67 (1995) 34. [60] H.W. Chin, R.A.M. Rosenberg, J. Food Sci. 61 (1996) 1118. [61] E. Matisova, J. Sedlakova, M. Slezackova, T. Welsch, J. High Resolut. Chromatogr. 22 (1999) 109. [62] M.J. Jordan, K.L. Goodner, M. Castill o, J. Laencina, J. Sci. Food Agric. 85 (2005) 1065. [63] D.D.L.C. Garcia, M. Reichenbacher, K. Da nzer, C. Hurlbeck, C. Bartzsch, K.H. Feller, J. High Resolut. Chromatogr. 20 (1997) 665. [64] F. Augusto, A.L.P. Valente, E.D.S. Tada S.R. Rivellino, J. Chromatogr. A 873 (2000) 117. [65] S. Rocha, L. Maeztu, A. Barros, C. Ci d, M.A. Coimbra, J. Sci. Food Agric. 84 (2004) 43. [66] H.V. Doorn, C.B. Grabanski, D.J. Miller, S.B. Hawthorne, J. Chromatogr. A 829 (1998) 223. [67] B.I. Escher, M. Berg, J. Muhlemann, M.A.A. Schwarz, J.L.M. Hermens, W.H.J. Vaes, R.P. Schwarzenbach, Analyst 127 (2002) 42. [68] P. Bartak, L. Cap, J. Chromatogr. A 767 (1997) 171. [69] B.D. Page, G. Lacroix, J. Chromatogr. A 757 (1997) 173. [70] C.G. Zambonim, F. Palmisano, Analyst 123 (1998) 2825. [71] R. Shirey, V. Mani, M. Butle r, Supelco Reporter 14 (1995) 4. [72] S. Motlagh, J. Pawliszyn, Anal. Chim. Acta 284 (1993) 265. [73] I. Rodriguez, M.P. Llompart, J. Chromatogr. A 885 (2000) 291. [74] L. Muller, T. Gorecki, J. Pawliszyn, J. Fresenius, Anal. Chem. 364 (1999) 610. [75] A.A. Boyd-Boland, J. Pawliszyn, Anal. Chem. 68 (1996) 1521. [76] H. Kataoka, J. Pawliszyn, Chromatographia 50 (1999) 532.

PAGE 63

42 [77] S. Bigham, J. Medlar, A. Kabir, C. Sh ende, A. Alli, A. Malik, Anal Chem. 74 (2002) 752. [78] J. Chen, J. Pawliszyn, Anal. Chem. 67 (1995) 2530. [79] J. Wu, W. Xie, H.L. Lord, J. Pawliszyn, Analyst 125 (2000) 391. [80] B.C.D. Tan, P.J. Marriott, H.K. L ee, P.D. Morrison, Analyst 124 (1999) 651. [81] J.V. Hinshaw, LC-GC Europe, December (2003) 2. [82] H. Kataoka, Anal. Bioanal. Chem. 373 (2002) 31. [83] J. Pawliszyn, Can. J. Chem. 79 (2001) 1403. [84] J. Wu, J. Pawliszyn, Anal. Chem. 73 (2001) 55. [85] J. Pawliszyn, Sampling and Sample Pr eparation for Field and Laboratory, Elsevier, Amsterdam, 2002.

PAGE 64

43 CHAPTER TWO SOL-GEL TECHNOLOGY IN SOLI D-PHASE MICROEXTRACTION 2.1 Introduction to Sol-Gel Technology The sol-gel process [1] represents a power ful synthetic route whereby a colloidal liquid phase (sol) is formed from sol-ge l precursors, typically through hydrolytic polymerization reactions that ultimately lead to the formation of a solid phase (gel). In recent years, this term has been used for any solution process involving hydrolysis and formation of a gel starting from precursor mate rials. In general, hyd rous metal oxides or hydroxides can be formed by chem ically process of the precur sor. The main advantage of the sol-gel approach is the controllability of the entire sol-gel process from the sol-gel precursor to the end product, so called “tailor-made” materials. In addition, due to the inherent flexibility of sol-gel processing th e development of highly specialized materials is possible by variation of the sol-gel com ponents, and/or processing conditions [2]. In 1844, Ebelman [3] was the fi rst scientist to describe sol-gel synthesis. He prepared transparent solid using silica ester by slow hydrolysis process at the room temperature. After this, it took almost a centu ry to use the sol-gel technology. Geffcken [4] used alkoxides to prepare oxide films. Later, the Schott Glass Company in Germany developed this process, and which well reviewed by Schroede r [5]. Hurd [6] showed a polymeric structure of silicic acid with continuous liquid phase, which became widely accepted research for the demonstration of th e network structure of silica gels in the 1930s. In the 1950s, Roy used sol-gel met hod to prepare homogeneous powders in

PAGE 65

44 ceramics research [7,8], however that work di d not fully explain the concepts of sol-gel reaction mechanisms. A few years later, Disl ich [9], and Levene and co-workers [10] developed multicomponent glasses independen tly using alkoxides by controlling the solgel hydrolysis and condensation reactions. Yo ldas [11-13], and Yamane and co-workers [14] demonstrated in their papers that mono liths could be prepared by careful drying of gels, which devoted to gain large atte ntion in sol-gel research to date. The new field, the synthesis of organic-i norganic hybrid materials, was initiated in the 1980s by the pioneering wo rk of Schmidt at the Frau nhoffer Institute [15,16], and it was one of the major developments in solgel processing. Hybrid materials have been called ORMOSILS ( OR ganically MO dified SIL icates), ORMOCERS ( OR ganically MO dified CER amics) and CERAMERS ( CERA mic poly MERs ) or POLYCERAM ( POLY meric CERAM ics). Later, Mackenzie and co -workers [17] pointed out many inherent advantages to sol-ge l process, such as better homogeneity and purity, easier controllability, enhanced ma nageability, and so on. Due to unique combinations of properties and numerous inherent advantages, sol-gel process has found growing interest in diverse research areas including nanopart icles, coatings, fibers, monoliths, or bulk materials as shown in Figure 2.1 [18]. Schmidt’s work made a commitment to th e new sol-gel application in chemistry, which was started less than two decades ago. In 1987, Cortes and co-workers [19] at Dow Chemical Company reported that they creat ed porous monolithic ceramic beds within small-diameter capillaries using sol-gel t echnology by polymerizing solutions containing potassium silicate to apply as separation column in liquid ch romatography (LC). In 1993, Crego and co-workers [20] re ported a procedure for the preparation of sol-gel open-

PAGE 66

45 Figure 2.1 Shape of different products ava ilable through pro cessing by sol-gel technology. Adapted from ref. [18].

PAGE 67

46 tubular capillary columns (OTCs) using tetraethyl orthosilicate (TEOS), Si(OC2H5)4, with octadecylsilane (ODS) moieties for reversed -phase liquid chromatography (RPLC). Guo and Colon [21] developed a sol-gel st ationary phase for open tubular liquid chromatography (OTLC) and open tubular electrochromatography (OTEC). Malik and co-workers prepared sol-gel coated columns for capillary gas chromatography (GC) [22], and sol-gel coated fibers for solid-p hase microextraction (SPME) [23,24]. Recently, Tanaka and co-workers [25-27] developed sol-gel monolithic columns for high-performance liquid chromatography (HPLC), and showed high permeability and high column efficiency compared to the conventional packed co lumn. Their sol-gel monolithic work opened up a new direction of sol-gel research in separation science. The sol-gel process offers many advantages such as tunable specificity, reactivity, homogeneity and purity, and controllable po rosity, so it has found ever increasing application in a variety of disciplines, such as ceramics [28,29], sensors [30-32], optics [33,34], nanotechnology [35-37], and different areas of chemistry [38-43]. The sol-gel process also provides an effective means for the control of the surface area and surface characteristics (i.e., hydrophilic/hydrophobic proper ties, positive/negative charge, etc.) of sol-gel materials which enables one to creat e chemically bonded stationary phases on the inner surface of fused-silica capillary with high stability and efficiency for the applications in sample prepara tion [23,44-48] and separation [22,49,50] 2.2 Fundamentals of Sol-Gel Chemistry In general, by the simultaneous hydrolys is and polycondensation reactions of

PAGE 68

47 metal alkoxide precursors, M(OR)x, followed by aging and drying under ambient conditions are the major key reactions of the so l-gel process as illust rated in Figure 2.2. During the sol-gel process, sol-gel precursors form colloidal particles in a liquid, a sol, which then transforms into a thr ee dimensional rigid network with pores of submicrometer dimensions and polymeric chains a gel. Flory [51] classified “gel”s in four categories: (1) well-ordered lamellar stru ctures; (2) completely disordered covalent polymeric networks; (3) predominantly diso rdered polymer networks formed through physical aggregation; and (4) part icular disordered structures. When the pore liquid is removed as a gas phase from the three dimensional sol-gel network under hypercritical conditions, the ne twork does not collapse and a low density aerogel is produced. When the solvent is evap orated by thermal evaporation at or near ambient pressure and shrinkage occurs, a xe rogel is generated. This process is called drying. A gel is defined as dried when the physically adsorbed water is completely evacuated, which typically occurs between 100 C and 180 C. The porous gel is transformed to a dense ceramic material when all pores are eliminated. This process is completed under elevated temperatures (generally above 1000 C). Typically, there are four components to prepare a sol solution: (1) a sol-gel precursor, usually a metal alkoxide M(OR)x; (2) a solvent system; (3) a catalyst or a inhibitor; and (4) water.

PAGE 69

48 Figure 2.2 Overview of the sol-gel pro cess. Adapted from ref. [1].

PAGE 70

49 2.2.1 Sol-gel precursors In general, silica-based and/or non-sil ica-based metal alko xides are commonly used as a starting material (precursor) for solgel process. The stabil ity of metal alkoxide decreases as the electonegativity of the me tal increases, which is from left to right direction across the periodic table [52]. A lthough tetrafunctional al koxide precursors are most commonly used to incorporate into th e sol solution, one or more functional group substituted alkoxides may also be used in so l-gel process. While there are many sol-gel precursors containing different metal elements, su ch as Si, Ti, Al, V, Zr, and Ge, silica is the most convenient from many aspects: well-known chemistr y, stability of Si-O bond, well-documented sol-gel methodology [45,46,48], facility of characterizations and commercially available star ting materials [53]. Some common silicaand non-silicabased sol-gel precursors are shown in Table 2.1.

PAGE 71

50 Table 2.1 List of common alkoxide-b ased sol-gel precursors.a Name/Structure MW bp ( C) nD (20 C) d (g/mL) (20 C) Solubility Tetramethoxysilane (TMOS) Si H3CO OCH3 OCH3 OCH3 152.2 121 1.3688 1.02 Alcohols Tetraethoxysilane (TEOS) Si OC2H5 C2H5O OC2H5 OC2H5 208.3 169 1.3838 0.93 Alcohols Tetran -propoxysilane [Si(n-C3H7O)4] Si C3H7O OC3 H 7 OC3H7 OC3H7 264.4 224 1.401 0.916 Alcohols Tetran -butoxysilane [Si(n-C 4 H 9 O) 4 ] Si OC4H9 C4H9O OC4H9 OC4H9 320.5 115 1.4126 0.899 Alcohols Titanium (IV) Isopropoxide Ti O O O O CHCH3 CH3 CHCH3 H3CHC Ch3HC H3C CH3 CH3 284.22232 1.464 0.96 Alcohols Zirconium (IV) n -Butoxide CH3(CH2)3O ZrO(CH2)3CH3 O(C H 2)3C H 3 O(CH2)3CH3 383.68117 1.466 1.049 (25 C) Alcohols *MW: Molecular weight; bp: Boiling point; nD: Refractive index; d: Density. aAdapted from ref. [1,54,55].

PAGE 72

51 2.2.2 Sol-gel solvent system The sol solution does not just contain the sol-gel precursor(s), it may also contain polymer, catalyst(s) (or inhibito r(s)), and water. Usually the solvent is chosen depending on its ability to generate a homogeneous sol and progression of the gelation process. The solvent should also be chemically inert so it does not participate in the sol-gel reactions that could lead to the creation of undesirable side products. The selection of the solvent may depend on the nature of the alkoxy substituents present on the central metal atom. For example, the hydrolysis by-product of TMOS is methanol; however, typically any miscible al cohol other than meth anol would often be used as the solvent to prevent any liquidliquid phase separation during the sol-gel hydrolysis reactions. Since the hydrolysis -by-product of TMOS is methanol, using methanol as the solvent would suppress th e hydrolysis reaction by Le Chtelier’s principle. Often more than one solvent syst em (e.g., mixture of methanol and methylene chloride) may be used depending on the solubility of the sol-gel precursor(s) and the copolymer. In addition, the amount of solvent( s) used for sol-gel processing may affect the speed of the sol-gel reaction. Using high er amounts of solvent(s) may dramatically increase gelation time. A drying control chemical additive (DCCA) (e.g., formamide, glycerin, dimethyl ether, and oxalic acid) as a co-solvent may be also play another crucial role in the sol-gel reaction [56,57]. As the solvent escapes from w ithin the gel or as the pore size changes during these processes, stress is develope d in the sol-gel network, which may cause cracking. To overcome this, DCCA is incorporat ed in the starting mixture before gelation [1,58,59] so it may facilitate a more effective way of drying the gel. Unfortunately, the

PAGE 73

52 exact way in which the DCCA improves th e drying process is still unclear [59]. In addition, the water/precursor(s) ratio in the sol solution may also affect the speed of reaction as well as th e physical properties of the obtained sol-gel materials. So this ratio can be carefully cont rolled to fine tuned porous stru ctures of sol-gel materials. Constantin and Freitag [60] reported that there is an op timum content of water (about 200 %) that facilitates the formation of uniform and porous sol-gel products. 2.2.3 Sol-gel catalysts/inhibitors Among all constituents used in sol-gel proc esses, the catalyst plays an important role in a silica-based system due to its slow reaction rate. Acid or base catalysts can accelerate both the hydrolysis and condensation reactions and influence the structure of the resulting sol-gel materials [17]. Various catalysts (e.g., acetic acid [61], hydrochloric acid [62], trifluoroacetic acid [23], ammonia [63], amines [63], and potassium hydroxide [64]) have been used to cause faster sol-gel processes for silicon alkoxides. Under acidic conditions, the hydrolysis reaction rates of silicon alkoxide precursors are significantly faster than the conde nsation reaction [65]. It is likely that an alkoxide group is protonated under acidic cond itions in a rapid first step followed by SN2type water molecule attack to form a transiti on state [66]. As a result the resulting sol-gel material is weakly branched polymers with a predominant more microporous structure [67]. Unlike the acid-catalyzed reaction, under basic conditions the condensations reaction rates of silicon alkoxide precursors are higher than the hydrolysis reaction rates. It is likely that water molecu les dissociate to produce nuc leophilic hydroxyl anions (OH-) in a rapid first step followed by the form ation of transition state and then the displacement of alkoxy (OR-) group by SN2 reaction mechanism. As a result, the

PAGE 74

53 resulting sol-gel material is highly branch ed polymers with more mesoporous structure [68]. Generally, the sol-gel precu rsor is less reactive in hydrolysis and condensation reactions if the alkoxide group a ttached to a central metal at om is longer and/or bulkier [69]. Obviously, changing the pr ecursor and/or its concentration is on e of the ways to control the rates of the hydrol ysis and condensation reacti ons. In addition, a judicious selection of the sol-ge l catalyst is also important in determining the morphology of the sol-gel materials in silica-based systems. The addition of acidic catalyst decrease s the gel time of silicon alkoxide systems. However, some alkoxide precurs ors do not require the use of catalyst in so l-gel reaction due to their higher reactivity. Sol-gel proce ss with titanium and zirconium alkoxides, for example, requires the use of a chelating agent to inhibit sol-gel reac tion. Livage and coworkers [70] and Sanchez and co-workers [71] have performed several experiments including X-ray absorption near edge spec troscopy (XANES), extended X-ray absorption fine structure spectroscopy (EXAFS), and nuclear magnetic resonance (NMR) to understand the behavior of titanium isopropoxide, Ti(OPri)4, in glacial acetic acid (GAA). Ti atom in monomeric Ti(OPri)4 has a coordination number of 4 ( N = 4). XANES experiment showed that N increased to 6 after the addition of glacial acetic acid. 13C NMR showed that alteration of the chemical environments of the carboxylic and methyl carbons in GAA after reaction with Ti(OPri)4 was caused by bonding with titanium. It also showed that both te rminal and bridging OPri groups were present in the NMR spectra. Livage and co-workers [70] showed that OPri groups were hydrolyzed on a preferential basis, whereas the bridging acetate ligands remain bonded much longer to titanium

PAGE 75

54 throughout the condensation process and thereb y slowed down the gelation process [72]. Because they are not hydrolyzed, bridging acetate ligands change the condensation process possibly promoting the forma tion of linear polymeric products. The 1H and 13C NMR [71] chemical shifts st udies showed that corresponding acetylacetone (acac) was not free in th e equimolar mixture of acac and Ti(OPri)4, but was bonded to titanium. The XANES spectrum also i ndicated that the a cac reaction caused the coordination of Ti to increase from 4 to 5. EXAFS data shows no Ti-Ti correlation. This data explains a chelated titanate precursor, Ti(OPri)3acac for a 1:1 mixture of acac and Ti(OPri)4, or an octahedrally coordinated dich elated precursor for a 2:1 mixture of acac and Ti(OPri)4. Figure 2.3 shows the structures of bridging and chelating ligands with titanium alkoxide.

PAGE 76

55C C C H3C CH3 O O Ti OR RO OR C C C H3C CH3 O O Ti OR RO C C C CH3 H3C O O CH O Ti OR OR O Ti OR OR RO RO CH3CH O O H H H (A) (B) (C) CH3 Figure 2.3 Structures of bridging a nd chelating ligands, R = Pri: (A) bidentate bridging ligand [73], (B) chelating ligand [74], and (C) two chelating agents [75].

PAGE 77

56 2.3 Chemical Reactions of Transition Met al Alkoxides during the Sol-Gel Process Metal alkoxides, M(OR)n, are versatile sol-gel precu rsors, which are known for most of the transitional metal elements including the lanthanides [76]. For silica based systems, the sol-gel chemistry is well-know n and involves hydrolysis and condensation reactions. Moreover, there is a lack of data concerning the sol-gel process of transition metal alkoxides. The main differences between silicon alkoxides and transition metal alkoxides are: (1) the lower el ectronegativity of transition elements leads to a much higher electrophilic character of the metal and (2) lack of fully satisfied coordination in the molecular precursors due to the several possible coordination numbers for transition metals. As a result, transition metal alkoxide s are much more reactive compared with silicon alkoxides, more moisture sensitive, an d they readily form precipitates rather than gels when water is added. Typical sol-gel process produ ces inorganic or organicinorganic hybrid materials through the growth of the solgel network incorporating conde nsation residues from both precursors and organic compone nts with sol-gel-active site s. In general, two main reactions are involved in the sol-gel proces s: (1) hydrolysis and (2) water or alcohol condensation reactions. Typical sol-gel hydrol ysis and condensation reactions of metal alkoxide precursor are illustr ated in Scheme 2.1 [55].

PAGE 78

57Hydrolysis: H2OROH Alcohol condensation: ROH Water condensation: H2O + + + + + M OR OR RO OR M OR OH RO OR M OR OH RO OR M OR OR HO OR M OR O RO OR M OR OR OR + M OR OR RO OR M OR OR HO OR M OR O RO OR M OR OR OR where R is an alkyl group, and M is a metal atom (e.g. Si, Al, Ti, Zr, Ge, etc.) Scheme 2.1 Sol-gel hydrolysis and condensation r eactions. Adapted from ref. [55].

PAGE 79

58 2.3.1 Hydrolysis Generally, hydrolysis reaction of metal alkoxides occurs u pon addition of water or a mixture of water and alcohol gene rating reactive hydroxo group, M-OH, since electronegative alkoxo groups, OR, make the metal atom highly pr one to nucleophilic attack. Scheme 2.2 shows a three-step mechanism which is typically proposed in the literature [77]. The first step (a) is a solvation of me tal cations by water molecule, so-called a nucleophilic addition, which leads to a tran sition state (b), where the coordination number of metal atom has increased by one. The second step involves a proton transfer within (b) leading to the intermediate (c), where a proton from the water molecule is transferred to the negatively charged oxygen of an adjacent OR group making the water molecule more acidic. The third step is the departure of leaving gr oup which should be the most positively charged species within the tr ansition state. The entire process, (a) to (d), follows a nucleophilic s ubstitution mechanism. [52].

PAGE 80

59 O H H M where M is a metal atom (e.g. Al, Ti, Zr, B, etc.), and R is an alkyl group. Overall: M(OR)n + n H2OM(OH)n+ n ROH Three-step mechanism: + O H H M OR MOR HO (a) (b) (c) (d) O R H RO H MOH + Scheme 2.2 Hydrolysis of metal alkoxides in the so l-gel process. Adapted from ref. [77].

PAGE 81

60 2.3.2 Condensation reaction Condensation is also a complex process and can occur as soon as hydroxo groups are generated. It can proceed by either nucleophilic substitution (SN) when the preferred coordination is satisfied or nucleophilic addition (AN) when the preferred coordination is not satisfied. Depending on experimental c onditions, three competitive mechanisms have to be considered: (1) oxolation, (2) alkoxolation, and (3) olation. Scheme 2.3 illustrates three different condensation mechanisms. Oxolation is a reaction by which an oxo bridge ( O ) is formed between metal atoms through the elimination of a water molecule. Ba sically, the mechanism is the same as for condensation process with metal atom repl acing hydrogen atom in the entering group. When the metal is unsaturated in terms of coordination number, oxolation occurs by nucleophilic addition (AN) with rapid kinetics (> 105M-1s-1) [78]. Alkoxolation follows pretty much the sa me mechanisms as oxolation except the leaving group is an alcohol molecule. Olation forms a hydroxy bridge, and can occur by nucleophilic substitution (SN) where the hydroxy group is the nucleophile and water molecule is the leaving group. In this case bridging hydroxo (M OH) groups can be generated through the elimination of a solvent molecule (either H2O or ROH) depending on the water concentration in the medium. The transformation of a so l-gel precursor into an oxide network may be governed by hydrolysis and condensation reactions. Theref ore, these reactions play an important role in the structure and morphology of the re sulting oxide. Their ro les can be optimized if the experimental conditions are carefully adjusted, and are related to both internal

PAGE 82

61 O H M where, M is a m e t al a t o m (e. g A l,Ti,Zr,B,e t c.),a n d Risa n alk y l g rou p + O H M OR MOR O R H ROH MO +(2) Alkoxolation: (1) Oxola t io n : (3) Olation: M M MO M O H M + O H M OH MOH O H H H2O MO +M M MO M MOH +H2O MO +M H : 2(OH)1MOH2 M O +H2O M OH +: 3(OH)1MOH2 M M M H2OM OH HO M OH2 M H O O H M 2H2O+: 2(OH)2+ M H O O H M H2O+: 2(OH)3 OH M H O M OH O O H H2 Scheme 2.3 Sol-gel condensat ion reactions of metal alkoxides. x(OH)y defines the number of M atoms linked by a single OH (x) and the number of bridges between these x metal atoms (y). Reproduced from ref. [52] with permission.

PAGE 83

62 (nature of the metal atom and alkyl groups, structure of the molecular precursors) and external (water/alkoxide ratio, catalyst, co ncentration, solvent, and temperature) parameters. 2.4 Sol-Gel Coatings for Cap illary Microextraction (CME) In capillary microextraction (CME, also called in-tube SPME), a sorptive coating on the inner surface of a fused-silica capillary serves as the extraction medium in which the analytes are preferentially sorbed and preconcentrated from various sample matrices. It will cover the development and application of sol-gel coatings and/or monolithic beds as CME. Due to inherent flexibility of solgel technology, various chromatographic applications, such as gas chromat ography [22,23], high-performance liquid chromatography [79-81], and capillary elec trochromatography [79,80,82], have been conducted. Preparation of the sol-gel extract ing phase coatings for CME involves four steps: (1) pre-treatment of the fused-silica capillary, (2) preparation of the sol solution, (3) sol-gel coating process, and (4) treatment of the coated capillary. 2.4.1 Pre-treatment of fused-silica capillary The main purpose of fused-silica capillary pre-treatment is to activate and increase the number of silanol (-OH) groups as much as possible on the inner surface of capillary to facili tate the effective bonding of the solgel sorbent (coating or monoliths) materials for in situ creation of sol-gel stationary pha se. Pre-treatment also cleans the surface from possible contaminants. Hayes and Malik [83,84] described a method of hydrothermal pre-treatment of the

PAGE 84

63 fused-silica capillary in their papers. For this, a homemade gas-pressure-operated capillary filling/purging device (Figure 2.4) was used for rins ing and coating processes. The fused-silica capillary (250 or 320 m i.d.) was sequentially rinsed with two organic solvents of different polarities (e.g., CH2Cl2 and CH3OH) and deionized water to clean the inner surface of the capillary from orga nic and inorganic contaminants. This was followed by a pressurized helium purge for a predetermined period of time (e.g., 10 min). This process was meant for expelling most of the water in the capillary leaving only a thin layer of water on the capillary inner surface. Both ends of the capillary were then sealed using an oxyacetylene torch, and the s ealed capillary was further conditioned in a GC oven by temperature programming from 40 to 250 C at a rate of 5 C min-1 with a final temperature hold time of 2 hours. The n, both sealed ends of the capillary were cut open using an alumina wafer and the capillary was subjected to further thermal conditioning using the same temperature progr am, but under helium purge. In the first step hydroxyl groups are genera ted due to hydrolysis of the siloxane bridges (Si-O-Si), but the second step was done to moderate the si lanol concentration of the surface (i.e., to achieve a uniform surface con centration of silanols). Constantin and Freitag [60] described an alternative method for the pretreatment of fused-silica capillary. For conditioning, the capillary was sequentially rinsed with 1 M NaOH for 60 min at 5 bar, with 0.1 M of HC l for 15 min at 5 bar, and finally with deionized water for 15 min at 5 bar. Then, the capillary was purged with argon and dry hexane for 10 min and placed in the vacuum oven for 12 hours (35 C, 20 mbar). This method was used by other researchers to prepare sol-gel columns [85-87].

PAGE 85

64 Figure 2.4 Schematic of a homemade capillary filling/purging device. Reproduced from ref. [82] with permission.

PAGE 86

65 2.4.2 Preparat i on of sol solution The selection of the ingredients and prep aration of the sol solution are the most critical steps to create the desired sol-gel so rbent. Typically the sol solution ingredients include sol-gel precursor(s), a sol-gel active organic polymer, a solvent system, a catalyst (or inhibitor), and water. Si nce the homogeneity of the so l system is very important, careful selection of a solvent system which is compatible with the used precursor(s) and polymer is essential. In addition to the typical ingredients in the sol solution, various additives are often used, such as a drying control chem ical additive (DCCA), and/or a surface deactivating reagent. As mentioned earlier, a DCCA may be used to minimize the shrinkage and cracking during conversion of the wet gel to dry gel, and also it helps to increase the porosity of the sol-gel material. A surface deactivating reagent may be used to deactivate residual silanol groups on the created sol-gel material to reduce possible adsorpti ve effects. Malik and coworkers reported the use of various deactiva tion reagents for sol-gel stationary phases, such as phenyldimethylsilane (PheDMS) [83] [88], 1,1,1,3,3,3-hexamethyldisilazane (HMDS) [44,46-48], and poly(methylhydr osiloxane) (PMHS) [22,23,44,46-48]. 2.4.3 Sol-gel coating technology Malik and co-workers developed the solgel SPME fiber coating procedure [23]. Prior to coating, the protective polyimide laye r was removed from a 1 cm segment of the fused-silica fiber using a cigarette lighter, followed by cleaning with methanol and drying. Then the cleaned end f the fiber was held insi de the sol solution by ve rtically dipping for about 20 min, so that a sol-gel coating was fo rmed on the bare oute r surface of the fiber

PAGE 87

66 end. This coating process was repeated until the desired coating thickness was achieved. The fiber was then conditioned under helium in the GC injection port. The sol-gel SPME coating technology has been us ed and modified by other SPM E research groups [89-91]. A sol-gel in-tube SPME (cap illary microextraction, CME) coating technology was first introduced by Malik and co-workers [22,92]. A hydrothermally treated fusedsilica capillary (1 m x 250 or 320 m i.d.) installed in a homemade capillary filling/purging device (Figure 2.4), and filled us ing the top clear portion of a sol solution obtained after centrifugation. After filling, th e sol solution was kept inside the capillary for a controlled period of time (typically 15-30 min) to facilitate the formation of a solgel coating on the capillary inner surface. During this process, a sol-gel organic-inorganic hybrid network was evolving within the so l solution portion of this network got chemically bonded to the inner walls of th e capillary via condens ation with surface silanol groups. After this, th e free unbonded portion of the so l solution was expelled from the capillary under helium pressure (e.g., 50 ps i) leaving behind a surface-bonded sol-gel coating within the capillary. The sol-gel coated capillary was further purged with helium for additional 30 min to facilitate the evap oration of the remaining volatile organic solvents. 2.4.4 Further treatment of sol-gel-coated CME capillary The sol-gel coated capillary was thermally conditioned in a GC oven using temperature-programmed heating from 40 to 320 C, for example, at a rate of 1 C min-1 with a final temperature hold time of 2 hours under helium purge where the final temperature being determined by thermal stabil ity of organic component used to prepare the sorbent coating. The capillary was cooled down to room temperature and sequentially

PAGE 88

67 rinsed with organic solvents (e.g., methylene chloride and methanol) to clean the coated surface. Finally, the capillary was installe d in the GC oven for drying and further conditioning using the same te mperature-programmed heating, except that this time the capillary was held at the final temperature for 30 min. The conditioned capillary was then cut into small pieces (e.g., 10-40 cm) that were used to perform capillary microextraction. 2.5 Characterization of Sol-Gel St ationary Phase and Its Morphology To understand the morphology of sol-gel organic-inorganic hybrid materials, various material characterization methods are performed. Scanning electron microscopy (SEM) [84,93] is a power tool and the most widely used techniques to study the morphology of sol-gel materials. SEM images show the uniformity of the sol-gel coating thickness and structural deta ils through cross sectional an d surface views of sol-gel coating. In addition to SEM, atomic force microscopy (AFM) [94] and X-ray photoelectron spectroscopy (XPS) [94-96 ] are also used to study the morphology of solgel materials. However, these techniques, ex cept XPS, do not provide the information of chemical bonds within the sol-gel structure. To investigate the chemical bonds in sol-gel stru cture, Fourier transform infrared spectroscopy (FTIR) [97-99] and nuclear magnetic resonance (NMR) [100,101] have been used. FTIR spectroscopy is a simple tool to study sol-gel process in its evolution with time and to identify specific chemical bonds within sol-gel coatings. Another powerful analytical technique is nuclear ma gnetic resonance (NMR) to investigate the structural features present in the sol-gel materials. The connectivity of the inorganic network has been studied by 29Si, 27Al, and 17O NMR techniques [1 02-105]. The latter

PAGE 89

68 technique has been useful to study the exis tence of Si-O-Ti and Si-O-Zr bonds in liquid sol solutions [103,104] and verify the homogene ous distribution within hybrid systems. Since the properties of the sol-gel capillary are based on the species present in sol-gel coating, it is essential to understand the morphology and chemical bonds of sol-gel coating in details. 2.6 The Application of Sol-Gel-Coated Mi croextraction Capillary in SPME and CME The extracting phase coating in a capillary microextrac tion (CME), capillary must be able to survive under harsh operati ng conditions, such as high temperature and pressure, wide pH ranges, and occasional elev ated temperature with organic solvents. The chemical bonds between the inner surface of the fused-silica capillary and the sol-gel sorbent coating provide both thermal [91,10 6] and solvent stabilities [91,107] with reproducible performance of the coating. This attribute of sol-gel coatings provides an effective means to couple CME with HPLC [108,109] as well as GC [44,89] and CEC [82,88]. In addition, the thickness of the coating can be contro lled using sol-gel coating technology [109,110] to enhance th e detection limits by using a variety of sol-gel active organic ligands. To perform CME-HPLC, the sol-gel coat ed capillary can be installed in the injection port as a sampling loop, and then online sample extraction [47] and separation can be performed in one place without additional instrumental modification to desorb the extracted analytes from CME capillary coating and transfer them to the separation column. This is accomplished by simply switching the injection valve from the “load” to

PAGE 90

69 “inject” position. The injected analytes are then separated on the separation column by either under isocratic or gradient conditions. Since sol-gel technology can provide tuna ble selectivity and efficiency, it is possible to create organic-i norganic hybrid porous material s through manipulation of the sol solution compositions. It significantly increases the surface area of the extracting phase and provides acceptable so rbent loading, sample capacit y, and faster mass transfer using thinner coatings [23]. Table 2.2 summarizes orga nic components used to prepare sorbents reported in the literatures.

PAGE 91

70 Table 2.2 Summary of sol-gel sorbent used in SPME and CME. Sorbent Structure Ref. Hydroxy-terminated poly (dimethylsiloxane) (PDMS) HOSi O CH3 Si O C H 3 CH3 Si OH CH3 C H 3 C H 3 n [23] Silanol-terminated poly (dimethyldiphenylsiloxane) (PDMDPS) Si CH3 CH3 xSi y O O Si O H CH3 CH3 HO [48] Phenyl-terminated dendrimer with a triethoxysilyl root. [45] Hydroxy-terminated dibenzo14-crown-4 (OH-DB14C4) [110]

PAGE 92

71 Amide bridged-calix[4]arene [111] Divinyl benzene (DVB) CH2CH CH CH2 [106] Poly(vinyl alcohol) (PVA) CH2 CH OH n [112] Superox-4 (PEG) C HO H H C H H O H m [89] Hydroxyfullerene (fullerol)OH-TSO C60OH n [113] A. Methoxypoly(ethylene glycol)-silane (PEG 1) H3CO(CH2CH2O) (CH2)2N C N O H (CH2)3 H Si OC2 H 5 OC2H5 OC2H5 n [44] [50] B. Poly(ethylene glycol)-bissilane (PEG 2) (CH2CH2O)(CH2)2N C N O H (CH2)3 H Si OC2 H 5 OC2H5 OC2H5 nN C O N H H Si OC2H5 OC2H5 C2H5O (CH2)3 [44] [50]

PAGE 93

72 5,11,17,23-Tetratert -butyl-25, 27-diethoxy-26,28-dihydroxycalix[4]arene [91] -Diallyltriethylene glycol H2CCH CH2 O CH2 CH CH2 n [114] Dihydroxy-terminated benzo15-crown-5 (DOH-B15-C5) OO O O O HO(H2C)3OCH2 HO(H2C)3OCH2 [107] [115] A. 4-Allyldibenzo-18-crown-6 B. 3-Allyldibenzo-15-crown-5 C. Allyloxyethoxymethyl-18crown-6 O O O O O O O O O O O O CH2CH CH2 O CH2CH CH2 O O O O O O CH2OCH2CH2OCH2CH CH2 A B C [116] Poly(methylphenylvinylsiloxane) OSi O H3C Si O CH3 CH 3 Si CH 3 HC CH2 x y z [117]

PAGE 94

73 5,11,17,23-Tetratert -butyl-25, 27-diethyxy-26,27-dihydroxycalix[4]arene t-Bu t-B u t-Bu t-B u OEt OH OEt HO [118] Allyloxy bisbenzo 16-crown-5 trimethoxysilane [119]

PAGE 95

74 2.7 Sol-Gel Monoliths in Separation Science For decades, chromatographers have been obtaining better efficiency by reducing the size of column packing materials in HPLC. However, the maximum obtainable number of theoretical plates by using a separation column has been steady at 10,00025,000 in HPLC due to high pressure drop caused by small-size column packing materials under the upper limit of opera ting pressure (5000 psi. with current instrumentation). Increasing the column permeability can increase the column performance. One possible way to attain higher column permeabili ty is to use a sol-gel monolithic column. The sol-gel monolith is a single piece of c ontinuous bed composed of chemically bonded network structures, which is often called as rod column or fritless column [19]. The main advantages of the sol-gel monolith are: (1) higher permeability and column efficiency, (2) free from back pressure problem due to the ab sence of retaining end frits, and (3) higher surface areas. In late 1980s, Cortes and co-workers [19] reported the preparation of the sol-gel silica-based monolithic capillary columns fo r liquid chromatography for the first time. However, the most successful preparation of sol-gel based monolithic column (or silicabased rod column) was reported by Nakanish i and co-workers [ 79,120] in early 1990s. They prepared bimodal pore silica gel m onolithic columns (macropores and mesopores) by sol-gel process of using TMOS and a watersoluble polymer. Late r the usefulness of this method was reported by Tanaka and co-w orkers [81] and Lubda and co-workers [121], who demonstrated that the sol-gel monolithic columns possessed many advantages, such as higher efficiency, lowered the backpressure, and reduced separation time

PAGE 96

75 compared to the conventional packed columns. Merck KGaA (Darmstadt, Germany) introduced this silica monolithic column, Chromolith, in the market, which opened the new era in column technology. Cabrera and co -workers [122] reported the use of this commercial monolithic column for fast analysis in HPLC with up to 9 mL/min flow rate (Figure 2.5). The SEM images of the commercial sol-gel monolithic column (Chromolith) are shown in Figure 2.6 [123]. The Tanaka group [124] prepared sol-ge l monolithic column in a fused-silica capillary using TMOS for CEC application. However, the efficiency of column was not as good as the polymer-based monolithic co lumns reported by Svec’s group [125]. Later Tanaka and co-workers [126] reported the us e of a low pressure-assisted operation for fast CEC separations to achieve better colu mn efficiency and faster analysis. Since monolithic column possesses highly porous struct ure and excellent permeability, it allows the use of pressure-driven flow. Hayes and Malik [83] introduced the single-step preparation of sol-gel monolithic columns in the fused-silica capillary for CEC using a commercially available sol-gel precursor, N -octadecyldimethyl[3-(trimethoxysilyl )propyl] ammonium chloride (C18TMS) to create a C18 moiety in a monolithic bed. They reported highly porous structure, sufficient stationary phase fo r chromatographic interactions and enhanced permeability of the monolithic column. SEM images of the sol-gel monolithic column are shown in Figure 2.7 [83]. Fujimoto [127] prepared the sol-gel mono lithic CEC column. Later, Takeuchi and co-workers [128] used a modified procedure of Fujimoto to prepare sol-gel silicabased monolithic beds, and use it as a pre-column extrac tion loop for on-line sample

PAGE 97

76 Figure 2.5 Chromatographic separation of five -blocking drugs on a silica rod column at different flow rates: Column: silica r od column RP 18e, 50 x 4.6 mm; mobile phase: acetonitrile/0.1% TFA in water (20/80; v/v ); flow rate: 1 9 mL /min, detection: UV 254 nm, samples: 1. atenolol, 2. pindolol, 3. metoprolol, 4. celipro lol, 5. bisoprolol. Reproduced from ref. [122] with permission.

PAGE 98

77 Figure 2.6 SEM image of a cross-section from a Chromolith structure. Reproduced from ref. [123] with permission.

PAGE 99

78 (A) (B) (C) Figure 2.7 SEM of a sol-gel monolithic column: (A) cross-sectional view (1800), (B) longitudinal view (7000), and (C) longitudina l view (15000). Adapted from ref. [83].

PAGE 100

79 preconcentation in HPLC. Sol-gel monolithic column format is probably the one major approach used in developing new types of stationary phases for both HPLC and CEC. A single piece of porous material allows the preparation of frit-free columns, which eliminates the bandbroadening and operation problems often caused by the formation of bub bles. In addition, enhanced column permeability enables to us e higher flow rates, which allows faster analysis, and provides better column efficiency. 2.8 References for Chapter Two [1] C. Brinker, G. Scherer, Sol-Gel Science: The Physics and Chemistry of SolGel Processing, Academic Press, Boston, MA, 1990. [2] J.D. Hayes, Sol-Gel Chemistry-Mediated Novel Approach to Column Technology for Electromigration Separations Ph.D. dissertation, Department of Chemistry, University of South Florida, Tampa, FL 2000. [3] M. Ebelman, Ann. Chim. Phys. 16 (1846) 129. [4] W. Geffcken, E. Berger, German Patent 736,411, May 1939. [5] H. Schroeder, Phys. Thin Films, 6 (1969) 87. [6] C.B. Hurd, Chem. Rev. 22 (1938) 403. [7] R. Roy, J. Am. Ceram. Soc. 39 (1956) 145. [8] R. Roy, J. Am. Ceram. Soc. 52 (1969) 344. [9] H. Dislich, Angew Chem. 10 (1971) 363. [10] L. Levene, I.M. Thomas, U.S. Patent 3,640,093, February 8, 1972. [11] B.E. Yoidas, J. Mater. Sci. 10 (1975) 1856. [12] B.E. Yoldas, Bull. Am. Ceram. Soc. 54 (1975) 286.

PAGE 101

80 [13] B.E. Yoidas, J. Mater. Sci. 12 (1977) 1203. [14] M. Yamane, A. Shinji, T. Sakai no, J. Mater. Sci. 13 (1978) 865. [15] H. Schmidt, J. Non-Cryst. Solids 73 (1985) 681. [16] H. Schmidt, Mater. Res. Soc. Symp. Proc. 32 (1984) 327. [17] J.D. Mackenzie, D.R. Ulrich, Ultrastruc ture Processing of Advanced Materials, Wiley & Sons, New York, NY 1988. [18] G. Schottner, Chem. Mater. 13 (2001) 3422. [19] H.J. Cortes, C.D. Pfeiffer, B.E. Richte r and T.S. Stevens, J. High Resolut. Chromatogr. Chromatogr Commun. 10 (1987) 446. [20] A.L. Crego, J.C. Diez-Masa, M.V. Dabrio, Anal. Chem. 65 (1993) 1615. [21] Y. Guo, L.A. Colon, Anal. Chem. 67 (1995) 2511. [22] D.X. Wang, S.L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [23] S.L. Chong, D. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [24] A. Malik, S.L. Chong. In: J. Pawlisz yn (Ed.), Application of Solid-Phase Microextraction, Royal Society of Chemistry, Cambridge, UK, 1999, pp 73. [25] H. Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N. Tanaka, J. Chromatogr. A 762 (1997) 135. [26] N. Ishizuka, H. Minakuchi, K. Nakanishi, N. Soga, H. Nagayama, K. Hosoya, N. Tanaka, Anal. Chem. 72 (2000) 1275. [27] N. Tanaka, H. Kobayashi, K. Nakanish i, H. Minakuchi, N. Ishizuka, Anal. Chem. 73 (2001) 420A. [28] A.C. Pierre, Ceram. Int. 23 (1997) 229. [29] D. Hreniak, E. Zych, L. Kepinski, W. Strek, J. P hys. Chem. Solids 64 (2003) 111. [30] J. Wang, Anal. Chim. Acta 399 (1999) 21. [31] L.R. Allain, Z. Xue, Anal. Chim. Acta, 433 (2001) 97.

PAGE 102

81 [32] Z. Zhang, H. Liao, H. Li, L. Nie, S. Yao, Anal. Biochem 336 (2005) 108. [33] N. Ozer, Sol. Energy Mater. Sol. Cells 43 (1996) 319. [34] D. Levy, Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 354 (2000) 159. [35] B. Munro, S. Kramer, P. Zapp, H. Krug, J. Sol-Gel Sci. Technol. 13 (1998) 673. [36] R. Badheka, A. Johnson, C.C. Perry, P.A. Sermon, A. Taylor, M.S.W. Vong, J. Mater. Chem. 12 (2002) 3278. [37] W. Li, S. Seal, E. Megan, J. Ramsdell, K. Scrammon, G. Lelong, L. Lachal, K. Richardson, J. Appl. Phys. 93 (2003) 9553. [38] J. Lin, C.W. Brown, Trends Anal. Chem. 16 (1997) 200. [39] J. Zarzycki, J. Sol-Gel Sci. Technol. 8 (1997) 17. [40] E. Bescher, J. D. Mackenzie, Mater. Sci. Eng. C 6 (1998) 145. [41] J.D. Mackenzie, J. Sol-Gel Sci. Technol. 26 (2003) 23. [42] J. Livage, New J. Chem. 25 (2001) 1. [43] R.J.P. Corriu, New J. Chem. 25 (2001) 2. [44] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002 ) 752. [45] A. Kabir, C. Hamlet, A. Mali k, J. Chromatogr A 1047 (2004) 1. [46] A. Kabir, C. Hamlet, K.S. Yoo, G.R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [47] T.-Y. Kim, K. Alhooshani, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165. [48] K. Alhooshani, T.-Y. Kim, A. Kabir, A. Malik, A. J. Chromatogr. A 1062 (2004) 1. [49] N. Ishizuka, H. Kobayashi, M.H. Mina kuchi, K. Nakanishi, K. Hirao, K. Hosoya, T. Ikekami, N. Tanaka, J. Chromatogr. A 960 (2002) 85. [50] C. Shende, A. Kabir, E. Townsend, A. Malik, Anal. Chem. 75 (2003) 3518.

PAGE 103

82 [51] P.J. Flory, Principles of Polymer Chemis try, Cornell University Press, Ithaca, NY, 1953. [52] J. Livage, M. Henry, C. Sanches, Prog. Solid St. Chem. 18 (1988) 259. [53] T. Ogoshi, Y. Chujo, Compos. Interfaces 11 (2005) 539. [54] R. Anderson, B. Arkles, C.L, Larson, Pe trarch Systems Silanes and Silicones (Petrarch Systems, 1987). [55] W. Li, Electrically Charged Sol-Gel Coa tings for On-Line Preconcentration and Analysis of Zwitterionic Biomolecule s by Capillary Electrophoresis. Ph.D. dissertation, Department of Chemistry, Un iversity of South Florida, Tampa, FL 2006. [56] T. Adachi, S. Sakka, J. Non-Cryst. Solids 99 (1988) 118. [57] A. Balamurugan, S. Kannan, S. Rajeswari, Mater. Lett. 57 (2003) 1244. [58] G. Orcel, L.L. Hench, I. Artaki, J. Jonas, T.W. Zerda, J. N on-Cryst. Solids 105 (1988) 223. [59] N. De La Rosa Fox, L. Esquivias, J. Mater. Sci. Lett. 10 (1991) 1237. [60] S. Constantin, R. Freitag, J. Sol Gel. Sci. Technol. 28 (2003) 71. [61] E. Stathatos, P. Lianos, Langmuir 19 (2003) 7587. [62] Y. Yan, S.R. Chaudhuri, D.G. Chen, A. Sarkar, Chem. Mater. 7 (1995) 2007. [63] S. Yada, S. Ohshima, K. Ka miya, A. Kawai, K. Uchida, A. Gotoh, F. Ikazaki, J. Non-Cryst Solids 208 (1996) 191. [64] E. Suzuki, M. Akiyama, Y. Ono, J. Chem. Soc., Chem. Commun. (1992) 136. [65] C. Tilgner, P. Fischer, F.M. Bohnen, H. Rehage and W.F. Maier, Microporous Mater. 5 (1995) 77. [66] M.M. Collinson, Crit. Rev. Anal. Chem. 29 (1999) 289. [67] C.J. Brinker, K.D. Keefer D.W. Schaefer, C.S. Ashl ey, J. Non-Cryst Solids 48 (1982) 47. [68] M.I. Sarwar, Z. Ahmad, Eu r. Polym. J. 36 (2000) 89.

PAGE 104

83 [69] S.S. Kistler, J. Phys. Chem. 36 (1932) 52. [70] J. Livage, C. Sanchez, M. Henry, S. Do euff, Solid State Ionics 32/33 (1989) 633. [71] C. Sanchez, F. Babonneau, S. Doeuff, A. Leaustic, in: J.D. Mackenzie, D.R. Urlich (Eds.), Ultra struct ure Processing of Advanced Ceramics, Wiley, New York, NY, 1988. [72] S. Doeuff, M. Henry, C. Sanchez, J. Livage, J. Non-Cryst. Solids 89 (1987) 206. [73] I. Lindqvist, Arkiv. Kemi, 2 (1950) 349. [74] T.J.R. Weakley, Polyhedron, 1 (1982) 17. [75] W.W. Day, M.F. Fredrich, W.G. Klemperer, J.W. Shum, J. Am. Chem. Soc. 99 (1977) 952. [76] D.C. Bradley, R.C. Mehrotra, D.P. Ga ur, Metal Alkoxides, Academic press, London, 1978. [77] C. Sanchez, J. Livage, M. Henry, F. Babonneau, J. Non-Cryst. Solids 70 (1985) 301. [78] G. Schwarzenback, J. Meier, J. Inorg. Nucl. Chem. 8 (1958) 302. [79] K. Nakanishi, N. Soga, J. Am. Ceramic. Soc. 74 (1991) 2518. [80] Y. Guo, L.A. Colon, J. Microcolumn Sep. 7 (1995) 485. [81] H. Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N. Tanaka, Anal. Chem. 68 (1996) 3498. [82] J.D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3. [83] J.D. Hayes, A. Malik, Anal. Chem. 72 (2000) 4090. [84] J.D. Hayes, A. Malik, Anal. Chem. 73 (2001) 987. [85] Z. Chen, T. Hobo, Anal. Chem. 73 (2001) 3348. [86] J. Kang, D. Wistuba, V. Schuri g, Electrophoresis 23 (2002) 1116. [87] Y. Zhao, R. Zhao, D. Shangguan, G. Liu, Electrophoresis 23 (2002) 2990. [88] W. Li, D. Fries, A. Alli, A. Malik, Anal. Chem. 76 (2004) 218.

PAGE 105

84 [89] Z. Wang, C. Xiao, C. Wu, H. Ha n, J. Chromatogr. A 893 (2000) 157. [90] J. Cai, J. Xing, L. Dong, C. Wu, J. Chromatogr. A 1015 (2003) 11. [91] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [92] A. Malik, US Patent 6,783,680 B2, August 31, 2004. [93] A. Fidalgo, L. M. Ilharco, J. So l-Gel Sci. Technol. 26 (2003) 357. [94] X. Zuo, K. Wang, L. Zhou, S. Hu ang, Electrophoresis 24 (2003) 3202. [95] R.M. Almeida, M.I.D.B Marques, X.J. Or ignac, Sol-Gel. Sci. Technol. 8 (1997) 293. [96] M. Spirkova, J. Brus, D. Hlavata, H. Ka misova, L. Matejka, A. Strachota, J. Appl. Polym. Sci. 92 (2004) 937. [97] Y. Yan, Y. Hoshino, Z. Kuan, S.R. Chaudhuri, Chem. Mater. 9 (1997) 2583. [98] P. Innocenzi, J. Non-Cryst. Solids 316 (2003) 309. [99] A. Fidalgo, L.M. Ilharco, J. Non-Cryst. Solids 283 (2001) 144. [100] M. Pursch, A. Jaeger, T. Schneller, R. Brindle, K. Albert, E. Lindner, Chem. Mater. 8 (1996) 1245. [101] S.A. Rodrigues, L.A. Col on, Chem. Mater. 11 (1999) 754. [102] M. Templin, U. Wiesner, H.W. Spiess, Adv. Mater. 9 (1997) 814. [103] F. Babonneau, Mater. Res. Soc. Symp. Proc. 346 (1994) 949. [104] V. Gualandris, J. Maquet, F. Babonneau, P. Florian, D. Massiot, Mater. Res. Soc. Symp. Proc. 576 (1999) 21. [105] D. Hoebbel, M. Nacken, H. Schmidt, J. Sol-Gel Sci. Technol. 12 (1998) 169. [106] M.M. Liu, Z. Zeng, C.L. Wang, Y.J. Ta n, H. Liu, Chromat ographia 58 (2003) 597. [107] D. Wang, J. Xing, J. Peng, C. W u, J. Chromatogr. A 1005 (2003) 1. [108] J.J. Kirkland, J.W. Henderson, J.J. DeStefano, M.A. van Straten, H.A. Claessens, J. Chromatogr. A 762 (1997) 97.

PAGE 106

85 [109] T.P. Gbatu, K.L. Sutton, J.A. Caruso, Anal. Chim. Acta 402 (1999) 67. [110] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [111] X. Li, Z. Zeng, J. Zhou, S. Gong, W. Wang, Y. Chen, J. Chromatogr. A 1041 (2004) 1. [112] A.L. Lopes, F. Augusto, J. Chromatogr. A 1056 (2004) 13. [113] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [114] L. Yun, Anal. Chim. Acta 486 (2003) 63. [115] D. Wang, J. Peng, J. Xing, C. Wu, Y. Xu, J. Chromatogr. Sci. 42 (2004) 57. [116] L. Cai, Y. Zhao, S. Gong, L. Dong, C. Wu, Chromatographia 58 (2003) 615. [117] M. Yang, Z.R. Zeng, W.L. Qiu, Y.L. Wang. Chromatographia 56 (2002) 73. [118] X. Li, Z. Zeng, Y. Chen, Y. Xu, Talanta, 63 (2004) 1013. [119] J. Wu, C. Wu, J. Xing, J. Chromatogr. A 1036 (2004) 101. [120] K. Nakanishi, N. Soga, J. Non-Cryst. Solids 139 (1992) 1. [121] D. Lubda, K. Cabrera, H. Minakuchi, K. Na kanishi, J. Sol-Gel Sci. Technol. 23 (2002) 185. [122] K. Cabrera, D. Lubda, H.-M. Eggenweiler, H. Minakuchi, K. Nakanishi, J. High. Resolut. Chromatogr. 23 (2000) 93. [123] D. Lubda, K. Cabrera, K. Nakanishi, W. Linder, Anal. Bioanal. Chem. 377 (2003) 892. [124] N. Ishizuka, H. Minakuchi, K. Nakanish i, N. Soga, K. Hosoya, N. Tanaka, J. High Resolut. Chromatogr. 21 (1998) 477. [125] E.C. Peters, M. Petro, F. Svec, J.M. J. Frechet, Anal. Chem. 70 (1998) 2296. [126] H. Kobayashi, C. Smith, K. Hosoya, T. Ikegami, N. Tanaka, Anal. Sci. 18 (2002) 89. [127] C. Fujimoto, J. High Resolut. Chromatogr. 23 (2000) 89. [128] L.W. Lim, K. Hirose, S. Tatsumi, H. Uzu, M. Mizukami, T. Takeuchi, J. Chromatogr. A, 1033 (2004) 205.

PAGE 107

86 CHAPTER THREE HIGHAND LOW-pH-RESISTANT, SU RFACE-BONDED SOL-GEL TITANIA HYBRID ORGANIC-INORGANIC CO ATING FOR ON-LINE CME-HPLC 3.1 Introduction Titania (TiO2) is a white pigment that has f ound various applications in many branches of industry such as plastics, enamel s, artificial fibers, el ectronic materials, and rubber [1]. Titania is also the most efficien t photocatalyst in some cases such as water decomposition [2,3]. In c ontrast to silica (SiO2), titania is mainly us ed as a support due to its excellent mechanical propertie s, inertness, and low price. Typically, titania is encountered in three different crystallographic forms: anatase, rutile, and brookite [4]. Due to inherently lo w stability of brookite, it does not seen to offer significant practical importance. Anat ase is the one which is thermodynamically stable up to 800 C and conversion to rutile ta kes place if the temperature is higher than 800 C. Both anatase and rutile form crystal structures in a tetr agonal lattice. The coordination number of titanium is 6, and that of oxygen, 3. The two modifications differ in number of common edges of the TiO6 octahedra: 4 for anatase and 2 for rutile. Both anatase and rutile have a bout 10 hydroxyl groups per nm2 of surface area [5]. In general, anatase is widely used for catalytic purpos es. Figure 3.1 shows crystal structures of anatase and rutile [6].

PAGE 108

87 Figure 3.1 Crystal structures of tw o crystallographic modificati ons of titanium dioxide: (A) anatase and (B) rutile. Reproduced from ref. [6] with permission.

PAGE 109

88 3.1.1 Titania as a chromatographic support in separation science Among chromatographic supports, silica (SiO2) is traditionally the most commonly used in separation science due to its highly favorable properties, such as mechanical strength to withstand high pressu res, ability to provide rapid mass transfer during chromatographic separations, the ease of surface modification due to the presence of highly reactive silanol groups (SiOH), and commercial availability [7-9]. Unfortunately, silica supports suffer many limitations for use in chromatography, especially in HPLC. The pres ence of the residual silanol groups on the silica surface leads to low chromatographic performance [ 8,10]. In addition, sili ca possesses stability only in the pH range of 2-8 [11,12]. Below pH 2, siloxane (-Si-O -Si-) linkages undergo hydrolytic attack and bonded ligands are slow ly removed from the surface, which leads to deterioration in chromatographic perfor mance [7,13]. Above pH 8, silica begins to dissolve leading to a collapse of the column bed [8,14]. Therefore, to overcome these inherent limitations of sili ca support, alternative inorgani c support materials have been introduced such as alumina [15], zi rconia [16] and titania [17,18]. Titania was introduced by Kawahara and co -workers [17] in the late 1980s and early 1990s for the chromatographic applicat ions to overcome inhe rent drawbacks of silica supports. In contrast to silica, titania materials show many advantages, such as high chemical and mechanical stabilities, that ma ke it an attractive choice to use titania as support material in separation science [17,19]. Titania also possesses excellent stability in the wide ranges of pH, which enables to separate analytes under extreme pH cond itions [20-22]. In addition, titania has anionexchange properties at acidic pH and cation-exchange prope rties at basic pHs [23],

PAGE 110

89 whereas silica has only a cation-exchange prope rty. Several studies have been performed on the chromatographic applications of titania material due to its good prospect as a new chromatographic support. 3.1.1.1 Titania as a chromatographic column support in HPLC Traditionally, surface-derivatization of tita nia was limited to the preparation of octadecylsilyl derivatives. Trudinger and co-w orkers [19] synthesized porous amorphous titania particles using titanyl chloride and otadecyltrimethoxysilane by a sol-gel process, and used them as reversed-phase packing materials for HPLC. Tani and co-workers [20,24] prepared titania-based reversed-phase packings with octadecyltriethoxysilane by the sol-gel method for normal-phase liquid ch romatography. Pesek and co-workers [25] reported an alternative surface derivatization method for tita nia with triethoxysilane to prepare bonded titania-based stationary phases via silani zation/hydrosilylation (Figure 3.2) [26], and the resulting surface-deriva tized titania was investigated by NMR spectroscopy. Sato and co-workers [27] reported the sol-gel preparation of spherical titaniabased packing materials for HPLC using t itanium isopropoxide, and demonstrated the possibility of pore size control in titania using stearic aci d as a pore regulating reagent. Ikeguchi and Nakamura [28] used a titania precolumn to selectively trap organic phosphates for online preconcentration under acidic conditions. Later, Fadeev and McCarthy [29] reported the covalent reaction of hydridosilanes with titanium surfaces by Ti-O-Si bonds, and Shafi and co-workers [30] reported a fast and efficient method for coating octadecyltrihydrosilanes [CH3(CH2)17SiH3] on rutile surfaces by sonochemistry.

PAGE 111

90 Figure 3.2 Preparation of ‘‘bonded’’ titania-based stationary phases for HPLC via silanization/hydrosilylation. Reproduced from ref. [25] with permission.

PAGE 112

91 In 1997, a German company, Sachtleben, introduced porous titania sorbent (Sachtopore) for HPLC. Winklet and Marme [3 1] reported the use of “Sachtopore for the application of normal-phase liquid chromatogr aphy to separate basic molecules such as amines. The conventional silica-based packings show degradation at temperatures under 100 C [32]. Moreover, the solubi lity of silica in water increases abov e 100 C, which is a problem for faster analysis and Kephart and Dasgupta [33] demonstrated the thermal stability and mechanical strength by using HPLC columns packed with C18-modified TiO2, which was capable to operate at 200 C and pressures up to 10,000 psi for high speed capillary liquid chromatography. Miyazaki and co-workers [34] develope d novel titania-coated monolithic silica columns for liquid chromatography to separate phosphorous-containing compounds. They mentioned that titania-coated mono lithic silica column s possess excellent selectivity for phosphorylated substances with low pressure drop compared to the conventional packed column. Lucy and co-workers [35] compared silica, zirconia, and titania columns for their ability to separate diesel samples by superc ritical fluid chromatography (SFC), and they found that a titania column coupled in series to a silica column was found to provide the highest overall grouptype resolutions. Recently, conventional and comprehensive two-dimensional (2D) HPLC systems using the combination of packed-titania and monolithic columns were established by Ueda and co-workers [36] for the online an alysis of phosphopeptides, which would be useful for online phosphoproteome analyses in future.

PAGE 113

92 3.1.1.2 Titania as a chromatographic column support in CE The use of titania as a so rbent in liquid chromatography has been described in the proceeding section. Although applications of titania in HPLC have been widely investigated, little attention ha s been given to the applicabil ity in capillary electrophoresis (CE) and capillary electrochr omatography (CEC). Tsai and co -workers [21] prepared solgel titania coatings in fused-silica capillarie s to separate proteins by CE. Fujimoto [37] created titania coatings on the inner wall of the fused-silica ca pillaries with a solution of a titanium peroxo complex for CE and CEC app lications. The titania-coated capillaries were found to possess both directions of elect roosmotic flow (EOF) and low solubility in aqueous solutions between pH 3 and 12, which is indicative of long term stability of titania compared to silica-based bonded phases. Xu and co-workers [38] also reported the preparation of sol-gel titania-coated capi llary with switchable EOF for nonaqueous CE separation of widely different mobilities. 3.2 Other Applications of Titania Traditionally, titania-silica composite materi als were prepared in the form of thin layer coating by several methods such as flame hydrolysis [39], chemical vapor deposition [40], electron-beam evaporati on [41], and sol-gel processes [42]. Titania typically resides in its applicatio n as pigment to provide whiteness and/or opacity in paints, plastics, and paper [43] due to its excellent optical properties due to high refractive index, lack of absorption of visible light, stab ility, nontoxicity, and malleability in the desired size range. Titani a is also used as a material in other applications such as ceramic membranes [ 44,45], adsorbents [46,47], and catalyst support

PAGE 114

93 [48]. The use of titania in th e area of catalysis is presented in Figure 3.3 [49]. The properties of the titania surf ace are decisive for its catalyt ic activity and selectivity depending on the type and concentration of diff erent active sites. In this case, the surface affects the formation of definite structur es of the active phase of titania-supported catalysts. Recently, the fine particles of titania have found applications as an advanced semiconductor material for solar cells [ 50], a luminescent material [51], and a photocatalyst for photolysis of water [52] or organic compounds [53] and as a bacteriocide [54]. Kunitake and Lee [55] prepar ed ultrathin titania gel films via sol-gel process for molecular imprinting application, and mentione d several advantages of the metal oxide films such as thermal stability, formation of multi-functional sites, and simplicity of operation. Schubert [56] summar ized the use of several ch emical additives for the chemical modification of titanium alkoxide pr ecursors using their Lewis acid properties. Currently, titania has become the subject of intensive research efforts in view of the potential and promise of titania nanotube s in bone growth and regeneration [57], environmental applications [58], dielectrics [ 59], optoelectronics [60], and sensors [61].

PAGE 115

94 Figure 3.3 Application of TiO2 in the field of catalysis. Reproduced from ref. [48] with permission.

PAGE 116

95 3.3 Sol-Gel Titania as an Extraction Sorbent in CME Sol-gel chemistry has been recently ap plied to solid-phase microextraction (SPME) [62-66] and capillary microextraction (CME) [6 7] to create sili con-based hybrid organic-inorganic coatings. The sol-gel te chnique provided chemically bonded coatings on the inner surface of fused-silica capillaries, and easily solved the coating stability problems as described in the proceeding chapter. Although sol-gel technique helped overcom e some significant shortcomings of SPME or in-tube SPME techniques by prov iding an effective means of chemical immobilization of the sorbent coatings, an im portant problem inherent in silica-based material systems (commonly used in SPME or CME) still remains to be solved: silicabased materials possess a narrow window of pH stability [68]. In the context of SPME, it pertains to the stability of silica-based fibers and coatings. The development of alternative materials possessing superior pH stability and better mechanical strength should provide SPME with additiona l ruggedness, and versatility. To date, very little (if any) resear ch has been done on the development and application of titania-based coatings in anal ytical microextraction techniques, although titania possesses many attractive pr operties such as superior pH stability and mechanical strength compared with sili ca [20-22,31,37]. Moraes and co-workers [69] used a two-step sol-gel process to synthesize a silica-titania hybrid material. The hybrid materials were employed as sorbents for solid-phase ex traction (SPE) for the investigation of carcinogenic N-containing compounds from a queous samples followed by GC analysis. In this chapter, the preparation of sol-gel TiO2-PDMS-coated capillaries will be presented and th e possibility of online CME-HPLC operation using sol-gel TiO2-PDMS

PAGE 117

96 microextraction capillaries to provide a significant improvement in pH and solvent stabilities, as well as enhancement in ex traction sensitivity will be demonstrated. 3.4 Experimental 3.4.1 Equipment On-line CME-HPLC experiments were car ried out on a Micro-Tech Scientific (Vista, CA) Ultra Plus HPLC system with a variable wavelength UV detector (Linear UVIS 2000). A Nicolet model Avatar 320 FTIR (Thermo Nicolet, Madison, WI) was used for FTIR measurements. A reversedphase ODS column (Supelco, 25 cm x 4.6 mm i.d., 5 m dp) and Betabasic 8 (Thermo Elect ron Co., 10 cm x 4.6 mm i.d., 5 m dp) were used for HPLC separation of the extracted analytes. A Fisher model G-560 Vortex Genie 2 system (Fisher Scientific) was used fo r thorough mixing of the sol solutions. A Microcentaur model APO 5760 centrifuge (Acc urate Chemical and Scientific Corp., Westbury, NY) was used for centrifugation of sol solutions. A Barnstead model 04741 Nanopure deionized water system (Barnst ead/Thermolyne, Dubuque, IA) was used to obtain 16.0 M -cm water. A JEOL model JSM-35 sca nning electron microscope (SEM) was used for the investigation of surface mo rphology of the sol-gel titania-PDMS-coated capillaries. On-line data collection and processing were done using ChromPerfect (version 3.5 for Windows) computer software (Justice Laboratory Software, Denville, NJ). 3.4.2 Chemicals and materials Fused-silica capillary (250 and 320 m i.d.) was purchased from Polymicro

PAGE 118

97 Technologies Inc. (Phoenix, AZ). A commercia l polysiloxane-based GC column (30 m x 0.25 mm i.d., 0.25 m film thickness) was used for comparison with sol-gel titaniaPDMS-based microextraction capillary in pH stability studies. Titanium (IV) isopropoxide (99.999 %), 1-butanol (99.4+ %), Poly(methylhydrosiloxane) (PMHS), 1,1,1,3,3,3-hexamethyldisilazane (HMDS), trif luoroacetic acid (TFA), polycyclic aromatic hydrocarbons (PAHs) (acenaphthylene, fluorene, phenanthrene, fluoranthene, pyrene), ketones (butyrophenone, valer ophenone, hexanophenone, heptanophenone), and alkylbenzenes (toluene, ethylbenzene, cu mene, n-propylbenzene, n-butylbenzene, amylbenzene) were purchased from Aldrich (Milwaukee, WI). Hydr oxy-terminated poly (dimethylsiloxane) (PDMS) was purchased from United Chemical Technologies, Inc. (Bristol, PA). HPLC-grade solvents (acet onitrile, methylene chloride, and methanol) were purchased from Fisher Sc ientific (Pittsburgh, PA). 3.4.3 Preparation of the sol solution The sol solution was prepared by thoroughly vortexing the follow ing reagents in a 2-mL polypropylene centrifuge tube: a solgel-active organic component (hydroxyterminated PDMS, 50 mg), a sol-gel precursor [titanium(IV) isopropoxide, 50 L], two solvents (methylene chloride and 1-butanol, 200 L each), a mixture of two surface deactivation reagents (HMDS, 8 L and PMHS, 2 L), and a sol-gel chelating agent (27 % TFA in H2O, 18 L). The content of the tube was then centrifuged for 5 min (at 13000 rpm; 15682 x g ). Finally the top clear solution was transferred to another clean vial by decantation, and was further used for coati ng the fused-silica microextraction capillary. The chemical ingredients used in the sol-ge l coating solutions are represented in Table 3.1.

PAGE 119

98 Table 3.1 Names, functions, and chemical structur es of the coating solution ingredients used to prepare sol-gel TiO2-PDMS-coated microextrtaction capillaries. Ingredient Function Chemical structure Titanium (IV) isopropoxide Sol-gel precursor Ti OCH(CH3)2 OCH(CH3)2 (H3C)2HCO OCH(CH 3 ) 2 Poly(dimethylsiloxane), hydroxy terminated (PDMS) Sol-gel-active polymer Si CH3 CH 3 n O H HO Trifluoroacetic acid (TFA) Bridging (chelating) ligand HO O CF3 1,1,1,3,3,3-Hexamethyldisilazane (HMDS) Deactivating reagent H3CSi H3C H 3 C N Si CH3 CH 3 CH3 H Poly(methylhydrosiloxane) (PMHS) Deactivating reagent Si O C H 3 CH3 H3C Si O C H 3 CH3 Si O H C H 3 Si CH3 CH 3 C H 3 m n

PAGE 120

993.4.4 Preparation of sol-gel TiO2-PDMS-coated microextraction capillary A 1-m long hydrothermal treated [70] fused-silica capillary (250 or 320 m i.d.) was installed on an in-house built gas pressu re-operated capillary filling/purging device [71] shown in Figure 2.4, and th e capillary was filled with th e prepared sol solution under 10 psi helium pressure. After filling, the sol solution was kept inside the capillary for 15 min to facilitate the creati on of a surface-bonded coating through sol-gel reactions taking place in the coating solution inside the capillary. Following this, the unbonded portion of the sol solution was expelled from the capilla ry under helium pressure (20 psi), and the capillary was further purged with helium for 30 min. The coated capillary was then conditioned in a GC oven by progr amming the temperature from 40 C to 320 C at 1 C/min under helium purge. Th e capillary was held at 320 C for 180 min. Finally, the capillary was cooled down to room temperat ure and rinsed with methylene chloride and methanol (3 mL each). Following this, the cap illary was installed in the GC oven for drying and further thermal conditioning und er temperature-programmed heating as described above, except that this time the capi llary was held at the final temperature for 30 min. 3.4.5 Sol-gel titania coatings in capilla ry microextraction (CME) for on-line CMEHPLC analysis A schematic of the CME-HPLC setup fo r on-line capillary microextraction and HPLC analysis is presented in Figure 3.4. An ODS column (25 cm 4.6 mm i.d., 5 m dp) was previously installed in the HPLC syst em and pre-equilibrated with the mobile phase consisting of an acetonitrile /water mixture (80:20, v/v). A 40-cm segment of the

PAGE 121

100 sol-gel TiO2-PDMS-coated microextraction ca pillary was mounted on the HPLC injection port as an external sampling loop. Analytes were pr econcentrated in the sol-gel TiO2-PDMS coating by passing the aqueous samp le from a gravity-fed dispenser [67] through this sol-gel titaniaPDMS-coated microextraction capillary for 40 min. Using a syringe, the sampling loop was flushed out with deionized water to remove the sample matrix. The analytes extracted in the sol-gel TiO2-PDMS coating of the sampling loop were then transferred into the HPLC column by desorbing with 100 % acetonitrile for 30 seconds. This was accomplished by simply switching the injection valve from the “load” to “inject” position. The injected analytes were then separated on the ODS column under gradient elution conditions by programmi ng acetonitrile composition in the organoaqueous mobile phase from 80 % (v/v) to 100 % in 15 min.

PAGE 122

101 Pump APump B Waste Waste Con t ainer Pump Controller Six-Port Injection Valve UV Detector Solvent Mixer HPLC Column Gravi t y -Fe d Sample Dispenser Computer PEEKTM Tubing Deactivated Fused Silica Capillary Deactivated Glass Column Sample Containing the Analytes of Interest Sol-Gel TiO2-PDMS Capillary (40-cm) as a Sampling Loop Computer for Data Acquisition and Processing Figure 3.4 Schematic diagram of the on-line CME-HPLC setup.

PAGE 123

1023.4.5.1 Treatment of sol-gel titania-PD MS-coated capillaries with 0.1 M NaOH solution A 40-cm segment of the sol-gel TiO2-PDMS-coated microextraction capillary was directly installed on the gravit y-fed sample dispenser, and co ntinuously rinsed with 0.1 M NaOH solution (pH = 13) for 12 hours. The capillary was then flushed out with deionized water for 30 minutes, and mounted back on the HPLC injection port. The target analytes (PAHs) were extracted on-line for 40 min, follo wed by their HPLC analysis as described in section 3.4.5. Using the same procedure, a 40-cm segment of the commercial GC capillary was treated with a 0.1 M NaOH solution. CME perfor mances of the used capillaries were evaluated both before and after the alkaline tr eatment to explore pH stability of the used coatings. 3.4.5.2 Treatment of sol-gel titania-PD MS-coated capillaries with 0.1 M HCl solution A new 40-cm segment of the sol-gel TiO2-PDMS-coated microextraction capillary was directly installed on the gravit y-fed sample dispense r, and continuously rinsed with 0.1 M HCl solution (pH = 1) for 12 hours. The capillary was then treated as described in 3.4.5.1, and used to extract the ta rget analytes (PAHs and ketones) on-line for 40 min, followed by their HPLC analysis. 3.4.5.3 Treatment of sol-gel titania-PDMS-coat ed capillaries with HPLC solvents at high temperature A 40-cm segment of the sol-gel TiO2-PDMS-coated microextraction capillary was filled with a mixture of aceton itrile and deionized water (50:50, v/v), and both ends were

PAGE 124

103 sealed by a mini union connector. The sealed capillary was subjected to high temperature (150 C) in the GC oven for 12 hours. The capi llary was then washed with DI water, purged with helium gas, and mounted back on the HPLC injection po rt as an external sampling loop. The target analytes (PAHs and ketones) were extrac ted on-line for 40 min, followed by their HPLC analysis. In similar fashion, the sol-gel TiO2-PDMS coating stability was also studied using either 100% acetonitrile or 100% methanol. 3.4.5.4 Safety precautions The presented work involved the use of va rious chemicals (organ ic and inorganic) and solvents that might be environmentally hazardous with adverse health effects. In accordance with material safety data sheet (M SDS) proper safety measures were taken in handling strong acids, bases and organic solven ts such as methanol, methylene chloride, and acetonitrile. All used chemicals were disposed in the proper waste containers to ensure personnel and environmental safety. 3.5 Results and Discussion The goal of this research was to deve lop high pH-resistant, surface-bonded sol-gel titania coatings for capillary microextraction to facilita te effective hyphenation of CME with HPLC. Judicious utilizati on of unique attributes of solgel chemistry allowed us to create a surface-bonded hybrid organic-inorganic titania coating on the inner walls of a fused-silica capillary providing an opportunity to exploit advanced material properties of titania-based sorben ts [17,19] in capilla ry microextraction. The sol-gel coating technique was fast, straight forward, and highly effective. Unlike the conventional multi-step coating technology [73-76], the sol-gel approach involved a single-step procedure to

PAGE 125

104 accomplish the sorbent coating, its chemical immobilization, and deactivation [77]. 3.5.1 Sol-gel reactions for the preparation of sol-gel TiO2-PDMS coating In sol-gel chemistry, alkoxides are comm only used as sol-gel precursors. Titanium alkoxides differ significantly from sili con alkoxides in term s of their chemical reactivity and complex-forming ability. Th ese differences dictate the adoption of different strategies for the creation of titaniabased sol-gel sorbents compared with those for silica-based analogs. While sol-gel reactions in a silica-based system are rather slow and often require the use of catalysts to accelerate the process [78], titania-based (transition metal oxide-based in general) sol-ge l reactions are very fast. This is explained by the fact that titanium alkoxides are very re active toward such nucleophilic reagents as water [79]. They readily undergo hydrolysis whic h results in a very fa st sol-gel process. Because of this, titania-based sol-gel reactions need to be decelerated by a suitable means to allow for the sol-gel process to be conducte d in a controlled manner. This is usually accomplished through the use of suitable chelat ing agents that form complexes with the sol-gel precursors (or repl ace the reactive alkoxy group with a less reactive group), thus hindering their participation in the sol-gel re actions. Without such a chelating agent, the gelation takes place instantaneously as the sol-gel solution ingredients are mixed with water. Chelating agents such as acetic acid [ 80,81], trifluoroacetic acid [82], or metal diketonates [83] are often used for this purpose. In the present work, sol-gel TiO2-PDMS-coated capillaries were prepared through hydrolytic polycondensation reactions performe d within fused-silica capillaries followed by thermal conditioning of the created coatings to achieve porous coating structures of enhanced surface area. In this work, TFA served as a bridging (or chelating) agent [82],

PAGE 126

105 and decelerated the gelation pr ocess for the creation of TiO2-PDMS coating. It has been shown by infra-red spectroscopy [80,82,84] that the acetate ion can serve as a bidentate ligand (chelating and bridging) to the tran sition metal alkoxides, such as Ti(OR)4 or Zr(OR)4. The sol-gel process for the generation a nd chemical immobili zation of the hybrid organic-inorganic surface co ating involved: (A) hydrolysis of the titanium alkoxide precursor [85] (Scheme 3.1A), (B) polycondens ation of the hydrolysis products into a three-dimensional sol-gel network [86,87] (S cheme 3.1B), chemical incorporation of hydroxy-terminated PDMS in the sol-gel ne twork [88,89] (Scheme 3.2), and chemical anchoring of the sol-gel hybrid polymer to the inner walls of the capillary [86,87] (Scheme 3.3). Finally, Scheme 3.4 represents chemical reactions involving HMDS [90] and PMHS [29,91] to deactivate the sol-gel TiO2-PDMS coating. There exists a marked distinction between the silica-based and transition metal oxide-based (including titania-b ased) sol-gel systems in terms of reaction kinetics. While sol-gel reactions in a silica-based system are rather slow, and require the use of catalysts [92], titania-based (transition metal oxide-based in general) sol-gel reactions are very fast [79]. Here the sol-gel reactions need to be slowed down to achieve controllable rates. This is usually accomplished through the us e of suitable chelati ng agents that form complexes with the sol-gel precursors (or replace the reactive alkoxy group with a less reactive group), thus hindering their participa tion in the sol-gel reactions. Without such a chelating agent, the gelation takes place instan taneously as the sol-gel precursor comes in contact with water during pr eparation of the sol solution by mixing the ingredients.

PAGE 127

106TiOCH(CH3)2 OCH(CH3)2 (H3C)2HCO 4 H2O Chelating Agentn (CH3)2CHOH Titanium (IV) Isopropoxide Titanium Hydroxide + + OCH(CH3)2 TiOH OH HO OH TiOH OH HO OH +n Ti(O O HO O Ti)nO O O ( A ) (B) TiOH (H3C)2HCO 4-n nwhere, n = 1, 2, 3, or 4 n: number of hydrolyzed butoxide ligands in the precursor molecule 4-n: number of intact butoxide ligands in the precursor molecule H2O TiOCH(CH3)2 OH HO OH TiOH OH HO OH +n Ti(O O HO O Ti)nO O O (CH3)2CHOH and/or Water condensation Alcohol condensation Scheme 3.1 (A) Hydrolysis of titanium (IV) is opropoxide, and (B) polycondensation of hydrolysis product, titanium hydroxide. Com position of the sol solution: a sol-gel precursor (titanium (IV) isopropoxide, 50 L), two solvents (methylene chloride and 1butanol, 200 L each), and a sol-gel chela ting agent (27 % TFA in H2O, 18 L) (at room temperature).

PAGE 128

107 H2O Ti)n(O O OH O Ti HO O O + (Si O Si OH CH3 PDMS CH3 CH3 CH3 O)m Si OH CH3 CH3 Ti)n(O O O Ti HO O O (Si O Si O CH3 CH3 CH3CH3O)m Sol-gel TiO2-PDMS polymer Si O CH3 CH3 (C) Scheme 3.2 (C) Polycondensation of hydroxyl-termi nated PDMS with the evolving solgel network. Composition of the sol solution: a sol-ge l precursor (titanium (IV) isopropoxide, 50 L), two solvents (methylene chloride and 1-butanol, 200 L each), a sol-gel chelating agent (27 % TFA in H2O, 18 L), and a sol-gel-active organic component (hydroxy-terminated PDMS 50 mg) (at room temperature).

PAGE 129

108 OH OH OH + Ti)n(O O O Ti OH O O (Si O Si O CH3 CH3 CH3 CH3 O)m Si O CH3 CH3 OH O OH Ti)n(O O O Ti O O (Si O Si O CH3 CH3 CH3 CH3 O)m Si O CH3 CH3 Wall bonded TiO2-PDMS coating H2O Capillary wall (D) Scheme 3.3 (D) Chemical anchoring of the sol-ge l polymer to the inner walls of the capillary. Composition of the sol solution: a sol-gel precursor (tita nium (IV) isopropoxide, 50 L), two solvents (methylene chloride and 1-butanol, 200 L each), a sol-gel chelating agent (27 % TFA in H2O, 18 L), and a sol-gel-active organic component (hydroxy-terminated PDMS, 50 mg ) (at room temperature).

PAGE 130

109 OH O OH Ti)n(O OH OH Ti OH O (Si O Si O CH3 CH3 CH3 CH3 O)m Si OH CH3 CH3 Wall bonded TiO2-PDMS coating (Si O Si CH3 CH3 CH3 CH3 O)q Si CH3 CH3 CH3 + (Si CH3 H O)p CH3 OH O O Ti)n(O O O Ti O O (Si O Si O CH3 CH3 CH3 CH3 O)m Si O CH3 CH3 (Si O Si CH3 CH3 CH3 CH3 O)q Si CH3 CH3 CH3 (Si CH3 O)p CH3 Deactivated, wall bonded TiO2-PDMS coating N Si (H3C)3 A PMHS, and B. HMDS A. B. H Si (CH3)3 Si (CH3)3Si (CH3)3 -H2, -NH3 Scheme 3.4 Deactivation of surface-bonded sol-gel TiO2-PDMS coating with HMDS and PMHS taking place during thermal treatment of the coated microextraction capillary at 150 C.

PAGE 131

110 3.5.2 Scanning electron microscopy (S EM) of surface-bonded sol-gel TiO2-PDMS coating Figure 3.5 represents two scanning elect ron micrographs showing the fine structural features of a 320m i.d. fused-silica capill ary with sol-gel TiO2-PDMS coating on the inner surface. As is evident from these images, the sol-gel TiO2-PDMS coating in the microextraction capillary acquires a por ous structure, providing enhanced surface area and sorption ability. Based on the SEM data, the thickness of the sol-gel TiO2PDMS coating was estimated at ~0.5 m. These images also show remarkable coating thickness uniformity in the sol-gel TiO2-PDMS-coated microextraction capillaries.

PAGE 132

111 Figure 3.5 Scanning electron microscopic (SEM) images of a 320m i.d. fused-silica capillary with sol-gel TiO2-PDMS coating: (A) cross-sectional view (500 ) and (B) surface view (10000 ).

PAGE 133

1123.5.3 Deactivation of the sol-gel TiO2-PDMS coating Deactivation of the sol–gel coatings can be expected to take place mainly during thermal conditioning of the capillary, through derivatization of th e free hydroxyl groups in the coating structure with HMDS [90] and PMHS [29,91] get incorporated in the solgel titania hybrid coating duri ng its formation from the solution. To control the gelation time and to obtain a transparent gel, it was essential to find an optimum ratio (v/v) of HMDS and PMHS. In the present study, this ratio was found to be 4:1 (HMDS:PMHS, v/v). 3.5.4 Fourier transform infrared (FTIR) sp ectroscopic investigation of the created sol-gel titania sorbent The formation of Ti-O-Si bonds in the pr epared sol-gel sorbent was examined by using a FTIR. The FTIR experiments were performed by passing IR radiation through a thin layer of sol-gel TiO2-PDMS coating material that was used in the fused-silica capillary. This was done in separate experiment s outside the fused-sili ca capillary. It has been reported [93,94] that a characteristic IR band repres enting Si-O-Ti bonds is located at 940-960 cm-1. Figure 3.6 shows FTIR spectra of th e sol-gel Ti-PDMS coating with a specific band at 952.63 cm–1. This is indicative of the pr esence of Si-O-Ti bonds in the sol-gel sorbent used in the fused-silica mi croextraction capillaries to perform on-line CME-HPLC analysis.

PAGE 134

113 Figure 3.6 FTIR spectra of the sol-gel TiO2-PDMS coating.

PAGE 135

1143.5.5 Applications of sol-gel TiO2-PDMS-coated microextraction capillary Sol-gel technology is quite versatile, a nd allows for the control of coating thickness either by manipulating the reacti on time or composition of the sol solution. Zeng and co-workers [65] has recently reported the preparation of 70 m thick silicabased sol-gel coating on conventional SPME fiber. It should be possible to create such thick coatings (either silica-based or transiti on metal oxide-based) on the inner surface of fused-silica capillaries as well. Use of thicker coatings should enhance the sample capacity and extraction sensitivity in CME with titania-based sol-gel coatings. Figure 3.7 presents a chromatogram illustrating CME-HPLC analysis of polycyclic aromatic hydrocarbons (PAHs) (ace naphthylene, fluorene, phenanthrene, and fluoranthene) from an aqueous sample using a sol-gel TiO2-PDMS-coated microextraction capillary. The extraction was performed on a 40 cm x 0.32 mm i.d. fused-silica microextraction capillary for 40 min using a gravity-fed sample delivery system at room temperature. The concentra tions of PAHs were in 20-500 ng/mL range. The extracted PAHs are list ed in Table 3.2. Th e nonpolar nature of the sol-gel TiO2PDMS coating showed high affinity and detec tion limits for these low polar analytes. In this case, the run-to-run p eak area repeatability was less than 9.74 % RSD. Detection limits for the extracted PAHs ranged be tween 0.15 ng/mL for phenanthrene to 3.07 ng/mL for acenaphthylene in conjunction with UV detection. Figure 3.8 presents a chromatogram illustrating CME-HPLC analysis of moderately polar aromatic ketones extracted from an aqueous sample using a 0.32 mm i.d. sol-gel TiO2-PDMS-coated microextraction capill ary. Compared to PAHs samples, ketones sample with higher analyte concentrations (300 ng/mL – 1 g/mL) were to be

PAGE 136

115 Figure 3.7 CME-HPLC analysis of PAHs Extraction conditions: 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMS-coated microextraction cap illary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution with mobile phase composition programmed from 80:20 (v/v) ACN/water to 100 % ACN for 20 min; 1 mL/m in flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) acenaphthyl ene (500 ng/mL), (2) fluorene (100 ng/mL), (3) phenanthrene (20 ng/mL), and (4) fluoranthene (100 ng/mL).

PAGE 137

116Table 3.2 Physical properties and chemical struct ures of PAHs used to prepare aqueous samples for CME-HPLC analys is employing a sol-gel TiO2-PDMS-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Adapted from ref. [95]. Name of Compound MW mp (C) bp (C) d (g/mL) at 25 C Structure of Compound Acenaphthylene 154.21 95 279 1.02 Fluorene 166.21 116.5 295 1.202 Phenanthrene 178.22 101 340 1.179 Fluoranthene 202.26 105 380 1.252 *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 138

117 used for CME-HPLC analysis. This may be explained by the nonpolar nature of the solgel TiO2-PDMS coating, higher solubility of ket ones in water due to higher polarity compared with PAHS, and the working principl es of UV detection. In this case, the runto-run peak area repeatability was less th an 7.90 % RSD. Detection limits for the extracted ketones ranged between 2.47 ng/mL for heptanophenone to 11.60 ng/mL for valerophenone in conjunction with UV detecti on. From the presented results it is evident that sol-gel TiO2-PDMS coating is able to extract both nonpolar and moderately polar analytes with good extraction sensitivity. Such an ability of the used sol-gel coating may be due to the presence of two different types of domain s (a nonpolar organic domain based on PDMS and a more polar inorganic do main based on sol-gel titania material) in such coatings [96]. The extracte d ketones are listed in Table 3.3. Figure 3.9 illustrates on-line CME-HPLC analysis of alkylbenzenes using a solgel TiO2-PDMS-coated microextraction capillary. Excellent detection limits were also achieved for these analytes (0.65 – 5.45 ng /mL), using UV detection. Like PAHs, alkylbenzenes are less polar analytes than ar omatic ketones, and they are well extracted by a sol-gel TiO2-PDMS coating in a microextracti on capillary providing low ng/mL and sub-ng/mL level detection limits. The alkylben zenes extracted from an aqueous sample are listed in Table 3.4. Table 3.5 summarizes the peak area repeatability and detection limit data for PAHs, ketones, and alkylbenzenes.

PAGE 139

118 Figure 3.8 CME-HPLC analysis of ketone s. Extraction conditions: 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMS-coated microextraction cap illary; extracti on time, 40 min (gravity-fed at room temperature). Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution with mobile phase composition programmed from 80:20 (v/v) ACN/water to 100 % ACN in 15 min; 1 mL/m in flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) butyrophenone (1 g/mL), (2) valerophenone (1 g/mL), (3) hexanophenone (500 ng/mL), and (4) heptanophenone (300 ng/mL).

PAGE 140

119Table 3.3 Physical properties and chemical struct ures of ketones used to prepare aqueous samples for CME-HPLC analys is employing a sol-gel TiO2-PDMS-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Adapted from ref. [95]. Name of Compound MW mp (C) bp (C) d (g/mL) at 25 C Structure of Compound Butyrophenone 148.2 12 221 0.988 O Valerophenone 162.23 9 106 0.988 O Hexanophenone 176.26 28 265 0.957 O Heptanophenone 190.29 17 155 0.946 O *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 141

120 Figure 3.9 CME-HPLC analysis of alkylben zenes. Extraction conditions: 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMS-coated microextractio n capillary; extraction time, 40 min (gravity-fed at room temperature) Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution with mobile phase composition programmed from 80:20 ACN/water to 100 % ACN in 15 min; 1 mL /min flow rate; UV de tection at 205 nm; ambient temperature. Peaks: (1) toluene ( 600 ng/mL), (2) ethylbenzene (200 ng/mL), (3) cumene (50 ng/mL), (4) n-propylbenzene ( 50 ng/mL), (5) n-butylbenzene (50 ng/mL), and (6) amylbenzene (50 ng/mL).

PAGE 142

121Table 3.4 Physical properties and chemical struct ures of alkylbenzenes used to prepare aqueous samples for CME-HPLC an alysis employing a sol-gel TiO2-PDMS-coated microextraction capillary. Data obt ained from www.sigmaaldrich.com. Name of Compound MW mp (C) bp (C) d (g/mL) at 25 C Structure of Compound Toluene 92.13 95 111 0.86 CH3 Ethylbenzene 106.16 95 136 0.867 CH2CH3 Cumene (isopropyl benzene) 120.19 96 151 0.86 H C H3C CH3 n-Propylbenzene 120.19 99 159 0.862 CH2CH2CH3 n-Butylbenzene 134.22 88 183 0.86 CH2CH2CH2CH3 Amylbenzene 148.25 75 205 0.86 CH2CH2CH2CH2CH3 *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 143

122Table 3.5 Peak area repeatability and limits of detection (LOD) data for PAHs, ketones, and alkylbenzene obtained in CME-HPLC experiments using sol-gel TiO2-PDMS-coated microextraction capillaries.a Peak area repeatability (n = 3) Mean peak areaRSD LOD (ng/mL) Chemical class Name (Arbitrary Unit)(%) (S/N = 3) PAH Acenaphthylene 23.52 9.50 3.07 Fluorene 12.17 8.87 1.40 Phenanthrene 19.91 8.78 0.15 Fluoranthene 21.42 9.74 0.84 Ketone Butyrophenone 48.64 3.92 9.62 Valerophenone 27.72 4.64 11.60 Hexanophenone 27.91 3.51 4.35 Heptanophenone 21.60 7.90 2.47 Alkylbenzene Toluene 20.28 1.93 5.45 Ethylbenzene 23.93 1.64 1.24 Cumene 12.28 6.09 0.74 n-Propylbenzene 13.62 4.52 0.65 n-Butylbenzene 14.41 9.93 0.84 Amylbenzene 9.44 7.43 1.07 aExtraction conditions: 40 cm x 0. 32 mm i.d. x 0.5 m sol-gel TiO2-PDMS-coated microextraction capillary; extraction time: 40 min (gravity-fed at room temperature); HPLC conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution from 80:20 (v/v) ACN/water to 100% ACN in 15 min (20 min for PAHs); 1 mL/min flow rate; UV detection at 254 nm (at 205 nm for alkylbenzenes); ambient temperature.

PAGE 144

1233.5.6 Extraction kinetic profile for sol-gel TiO2-PDMS-coated microextraction capillary Figure 3.10 illustrates the extraction kinetic profile for: (A) fluorene (nonpolar analyte) and (B) hexanophenone (moderate ly polar analyte) on a sol-gel TiO2-PDMScoated microextraction capillary. The microe xtraction experiments were performed using aqueous samples containing 100 ng/mL and 300 ng/mL concentrations of fluorene and hexanophenone, respectively. Experimental da ta for the two kinetic profile curves extraction were obtained by individually perf orming capillary microe xtraction for each of the solutes. A series of capillary microextr action experiments were conducted to vary the extraction time for each of the two analytes that were extracted from their standard solutions. Three replicate extr actions of each analyte were performed for 1, 5, 10, 20, 30, 40, 50, and 60 min. The average peak area was th en plotted against th e extraction time to obtain Figure 3.10. For both fluorene and he xanophenone, extraction equilibrium was reached within 40 min as is evidenced by th e plateau on the extraction curve. Since PDMS has nonpolar charac teristics, the TiO2-PDMS coating tends to extract a nonpolar analyte, in this case fluorene, better than a more polar analyte, hexanophenone, which has higher affinity for the aqueous medium.

PAGE 145

124 Figure 3.10 Illustration of the extraction kinetic profile of fluorene ( ), and hexanophenone ( ) obtained on a 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMScoated microextraction cap illary using 100 and 300 ng/mL aqueous solutions, respectively. Extraction conditions: 40 cm 0.32 mm i.d. x 0.5 m sol-gel TiO2-PDMScoated microextraction capill ary; extraction time, 40 mi n (gravity-fed at room temperature). Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); Isocratic elution 85:15 (v/v), and 90:10 (v/v) acetonitrile/ water, respectively; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature.

PAGE 146

1253.5.7 High pH stability of sol-gel TiO2-PDMS coating The sol-gel titania-PDMS coatings dem onstrated excellen t pH stabilities compared with conventionally created coatings like thos e used in commercial GC capillary columns. Figure 3.11 illustra tes the CME performance of a TiO2-PDMS-coated microextraction capillary (250 m i.d.) in CME-HPLC analysis of PAHs before (Figure 3.11A) and after (Figure 3.11B) rinsing th e capillary with a 0.1 M NaOH solution (pH=13) for 12h. Analogously obtained data for a piece of commercial GC column are presented in Figures 3.12A and 3.12B, resp ectively. Chromatogram 3.11B was obtained on the sol-gel TiO2-PDMS-coated microextraction cap illary after it was thoroughly rinsed with deionized water following the NaOH treatment. The extraction of PAHs was performed under the same set of conditions as in the Figu re 3.7, except the extraction capillary with 0.25 mm i.d was used instead. From the comparison of peak profiles and peak heights in Figures 3.11A and 3.11B, it is evident that the sol-gel TiO2-PDMS coating in the microextraction capillary re mained unaffected even after the prolonged rinsing with 0.1 M NaOH solution at pH 13. On the other hand, the stati onary phase coating in th e commercial GC capillary showed significantly less extraction sensitivit y as is evident from the comparison of peak heights in Figure 3.12 (A vs. B). It also faile d to survive the harsh conditions of rinsing with 0.1 M NaOH solution, which is evidenced by a dramatic decrease in the extraction sensitivity after the NaOH treatment (com pare Figures 3.12A obtained before NaOH treatment and 3.12B obtained after NaOH treatme nt). These results show that a sol-gel TiO2-PDMS-coated microextraction capillary po ssesses excellent stability under high pH conditions and retains its extr action ability even after bei ng subjected to extreme pH

PAGE 147

126 Figure 3.11 Chromatograms representing CME-HPLC analysis of PAHs using a sol-gel TiO2-PDMS-coated microextraction capillary before (A) and after (B) rinsing the microextraction capillary w ith a 0.1 M NaOH solution (pH=13) for 12h. Extraction conditions: 40 cm 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution with mobile phase composition programmed from 80:20 (v/v) ACN/water to 100 % ACN in 20 min; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) acenaphthylene (500 ng/mL), (2) unknown, (3) phenanthrene (20 ng/mL), and (4) fluoranthene (100 ng/mL).

PAGE 148

127 conditions, while conventionally prepared GC coatings were found to be unstable under such extreme pH conditions [97,98]. Table 3.6 shows repeatability and detec tion limit data for CME-HPLC analysis using sol-gel TiO2-PDMS-coated microextra ction capillaries For a 0.25 mm i.d. sol-gel TiO2-PDMS microextraction capillary, the RSD va lue in peak area remained within 9.22 %, and using a UV-detector, detection limits in the range of 0.28 5.37 ng/mL were achieved.

PAGE 149

128 Figure 3.12 Chromatograms representing CME-HPLC analysis of PAHs using a segment of commercial PDMS-based GC column as th e microextraction capil lary: (A) before and (B) after rinsing the microextraction capill ary with a 0.1 M NaOH solution (pH=13) for 12h. Extraction conditions: 40 cm 0.25 mm i.d. x 0.25 m commercial GC capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution with mobile phase composition programmed from 80:20 (v/v) ACN/water to 100 % ACN in 20 min; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) acenapht hylene (500 ng/mL), (2) unknown, (3) phenanthrene (20 ng/mL), an d (4) fluoranthene (100 ng/mL).

PAGE 150

129 Table 3.6 Peak area repeatability and limits of detecti on (LOD) data for PAHs obtained on a sol-gel TiO2-PDMS-coated microextraction capillary before and after treatment with 0.1 M NaOH for 12 h.a Before Rinsing After Rinsing LOD (ng/mL) (S/N = 3) Extracted PAHs Mean Peak Area* (A1) RSD (%)Mean Peak Area* (A2)RSD (%) % change in peak area (|A2-A1)/A1| x 100%) Before rinsing After rinsing Acenaphthylene 17.65 1.71 16.24 3.55 5.14 5.37 4.39 Phenanthrene 13.94 6.20 14.20 1.03 1.44 0.28 0.24 Fluoranthene 16.52 9.22 16.20 2.29 1.23 1.32 1.00 *n=3, arbitrary unit, aExtraction conditions: 40 cm x 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microext raction capillary; extraction time: 40 min (gravity-fed at room temperature); HPLC conditions: 25 cm x 4.6 mm i.d. ODS column (5 m dp); gradient elution from 80:20 (v/v) ACN/water to 100% ACN in 20 min; 1 mL/min flow rate; UV de tection at 254 nm; ambient temperature.

PAGE 151

1303.5.8 Stability of sol-gel TiO2-PDMS coating under highly acidic conditions As demonstrated in the pr evious section, sol-gel tita nia-PDMS coatings possess excellent stabilities unde r highly alkaline conditions comp ared with the PDMS coatings in commercial GC capillary columns. In this section, st ability of a TiO2-PDMS-coated microextraction capillary under highly acidic conditions is illustrated. Figures 3.13 shows the CME performance of a sol-gel TiO2-PDMS-coated microextraction capillary (250 m i.d.) in CME-HPLC analysis of f our compounds (butyrophenone, valerophenone, phenanthrene, and pyrene) before (Figure 3.13A) and after (Figur e 3.13B) rinsing the capillary with a 0.1 M HCl solution (pH = 1) for 12h. Chromatogram 3.13B was obtained on the sol-gel TiO2-PDMS-coated microextraction cap illary after it was thoroughly rinsed with deionized water following the HC l treatment. The extracti on of the mixture of ketones and PAHs was performe d under the same set of condi tions as in the Figure 3.11A. From the comparison of peak profiles and peak heights in Figur es 3.13A and 3.13B, it is also evident that the sol-gel TiO2-PDMS coating in the microextraction capillary remained unaffected even after the prolonged rinsing with 0.1 M HC l solution of pH 1. These results again show that a sol-gel TiO2-PDMS-coated microextraction capillary possesses excellent stability under highly acidi c conditions and re tains its extraction ability under low pH conditions. Table 3.7 shows peak area repeatability and detection limit data obtained in CMEHPLC analysis using a sol-gel TiO2-PDMS-coated microextraction capillary before and after rinsing with 0.1 M HCl for 12 h. For a 0.25 mm i.d. sol-gel TiO2-PDMS microextraction capillary, the RSD value in p eak area remained within 4.09 %, and using UV-detection, detection limits in the range of 0.18-10.45 ng/mL were achieved.

PAGE 152

131 Figure 3.13 Chromatograms representing CME-HPLC analysis of ketones and PAHs using a sol-gel TiO2-PDMS-coated microextraction capi llary before (A) and after (B) rinsing the microextracti on capillary with a 0.1 M HCl solution (pH=1) for 12h. Extraction conditions: 40 cm 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 70:30 (v /v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) butyrophenone (500 ng/mL), (2) valerophenone (500 ng/mL), (3) phenanthrene (1 0 ng/mL), and (4) pyrene (50 ng/mL).

PAGE 153

132 Table 3.7 CME-HPLC peak area repeatability and limits of detection (L OD) data for a mixture of ke tones and PAHs obtained on a sol-gel TiO2-PDMS-coated microextraction capillary before and after treatment with 0.1 M HCl for 12 h.a Before Rinsing After Rinsing LOD (ng/mL) (S/N = 3) Extracted ketones and PAHs Mean Peak Area* (A1)RSD (%)Mean Peak Area* (A2)RSD (%) % change in peak area (|A2-A1)/A1| x 100%) Before rinsing After rinsing Butyrophenone 6.82 3.32 6.72 3.62 1.47 7.81 7.9 Valerophenone 4.91 3.87 4.61 4.01 6.11 9.90 10.45 Phenanthrene 7.68 1.47 7.59 1.03 1.17 0.18 0.19 Pyrene 8.80 2.14 9.03 2.29 2.61 1.02 0.87 *n=3, arbitrary unit, aExtraction conditions: 40 cm x 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microext raction capillary; extraction time: 40 min; HPLC conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 70:30 (v/v) ACN/water; 1 mL /min flow rate; UV detection at 254 nm.

PAGE 154

1333.5.9 Stability of sol-gel TiO2-PDMS coating in HPLC solvents under elevated temperatures There are some distinct efficiency advant ages in chromatographic separations if high temperature of the eluent is used, and is becoming increasingly accepted as a separation parameter. The effect of temper ature has been discu ssed in the 1970s [99]. HPLC operation under elevated column temper atures increases the mass transfer rates and decreases column back-pressure due to decrease in mobile pha se viscosity with increase in temperature, which markedly reduces total analysis time [100]. Antia and Horvath [101] theoretically showed that the optimum velocity of the mobile phase is shifted to higher flow rates as a conse quence. Since the diffusion coefficient is proportional to the absolute temperature an d inversely proportiona l to the viscosity according to Stokes-Einstein relationship [102-104], HPLC separations under elevated temperatures allow HPLC operation at higher flow rates and using a longer column for faster analysis and better se paration efficiency [105,106]. The Stokes-Einstein relation as the simple hydrodynamic diffusion model can be expressed as following: (3-1) D = kB T 6where,r D : kB: T : r : the translational diffusion coefficient for a solute, radius of a solute in a solvent, viscosity of a solvent, Boltzmann's constant, temperature. Although the advantages us ing high temperature in HPLC have been widely discussed [107-109] using high temperature HPLC is limited due to the lack of

PAGE 155

134 temperature-resistant stationary phases. Gene rally, silica-based sta tionary phases are not recommended to use at temperatures above 80 C for a prolonged period [110,111], especially if the aggressive additives are used in the mobile phase such as phosphate buffers, or the column is run at very high or low pH. Recently, zirconia-based stationary phase [112] was introduced to use as a alternative column packing materials, and it ha s been shown to be st able in long-term use up to 200 C [113,114]. The sol-gel zirconia-PD MDPS-coated microext raction capillary [115] was developed by to show excellent pH and temperature stabilities for CME applications. Chemically bonded sol-gel TiO2-PDMS-coated capillaries demonstrated excellent stabilities under operational conditions i nvolving high temperature HPLC solvent environment. Figure 3.14 shows HPLC chro matograms of CME-HPLC analysis of a mixture of ketones and PAHs before (Figure 3.14A) and after (Figure 3.14B) the microextraction capillary was treated with a mixture of acetonitrile/w ater (50:50, v/v) at 150 C for 12h. These results show that the newly developed sol-gel titania and zirconia hybrid CME sorbents may also serve as stab le stationary phases in high temperature HPLC operations. Table 3.8 shows repeatability and detec tion limit data obtained in CME-HPLC analysis using a sol-gel TiO2-PDMS-coated microextraction capillary. For a 40 cm x 0.25 mm i.d. sol-gel TiO2-PDMS microextraction capillary, the RSD value in peak area remained within 5.01 %, and using a UV-detect or, detection limits in the range of 0.10 8.62 ng/mL were achieved. In addition, % change in peak area remained within 6.43 %, which demonstrates excellent solvent and temperature stabilities and extraction capability

PAGE 156

135 Figure 3.14 Chromatograms representing CME-HPLC analysis of ketones and PAHs using a sol-gel TiO2-PDMS-coated microextraction capilla ry before (A) and after (B) the microextraction capillary was filled with the mixture of ACN/water (50/50, v/v), sealed using a mini union connector, and heated at 150 C for 12h. Extraction conditions: 40 cm 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 70:30 (v/v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) butyrophenone (500 ng/mL), (2) valerophenone (500 ng/mL), (3) phenanthrene (10 ng/mL), and (4) pyrene (50 ng/mL).

PAGE 157

136 of the sol-gel TiO2-PDMS-coated microextraction capillary. Sol-gel TiO2-PDMS coating stability at elevat ed temperature was also studied using 100% of acetonitrile and 100% of meth anol. Tables 3.9 (with 100% acetonitrile) and 3.10 (with 100% methanol) show the high temperature solvent stability results using sol-gel TiO2-PDMS-coated capillaries.

PAGE 158

137 Table 3.8 Peak area repeatability and limits of detection (LOD) data for a mixture of ketones and PAHs using a sol-gel TiO2-PDMScoated microextraction capillary before a nd after the extraction capilla ry was filled with the mixture of ACN/water (50/50, v/v ) and heated at 150 C for 12 h.a Before Heating After Heatin g LOD (ng/mL) (S/N = 3) Extracted ketones and PAHs Mean Peak Area* (A1)RSD (%)Mean Peak Area* (A2)RSD (%) % change in peak area (|A2-A1)/A1| x 100%) Before Heating After Heating Butyrophenone 5.91 4.52 5.53 5.01 6.43 7.61 8.08 Valerophenone 5.12 4.91 5.43 4.28 6.05 8.62 8.12 Phenanthrene 7.08 2.40 6.88 2.79 2.80 0.10 0.19 Pyrene 9.03 2.86 8.70 3.12 3.65 0.82 0.93 *n=3, arbitrary unit, aExtraction conditions: 40 cm x 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microext raction capillary; extraction time: 40 min (gravity-fed at room temperature); HPLC conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with a mobile phase composition of 70:30 (v/v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature.

PAGE 159

138 Table 3.9 Peak area repeatability and limits of detection (LOD) data for a mixture of ketones an d PAHs using a sol-gel TiO2-PDMScoated microextraction capillary before a nd after the extraction capilla ry was filled with 100% ACN and heated at 150 C for 12 h.a Before Heating After Heatin g LOD (ng/mL) (S/N = 3) Extracted ketones and PAHs Mean Peak Area* (A1)RSD (%)Mean Peak Area* (A2)RSD (%) % change in peak area (|A2-A1)/A1| x 100%) Before Heating After Heating Butyrophenone 5.81 5.10 5.32 6.41 8.43 7.51 7.83 Valerophenone 5.03 5.54 4.73 7.12 5.96 8.40 7.12 Phenanthrene 7.29 2.13 6.84 3.20 6.17 0.13 0.19 Pyrene 9.11 3.21 8.60 4.04 5.59 0.78 0.90 *n=3, arbitrary unit, aExtraction conditions: 40 cm x 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microext raction capillary; extraction time: 40 min (gravity-fed at room temperature); HPLC conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase com position of 70:30 (v/v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature.

PAGE 160

139 Table 3.10 Peak area repeatability and limits of detection (LOD) da ta for a mixture of ketones and PAHs using a sol-gel TiO2-PDMScoated microextraction capillary before a nd after the extraction capilla ry was filled with 100% MeOH and heated at 150 C for 1 2 h.a Before Heating After Heatin g LOD (ng/mL) (S/N = 3) Extracted ketones and PAHs Mean Peak Area* (A1)RSD (%)Mean Peak Area* (A2)RSD (%) % change in peak area (|A2-A1)/A1| x 100%) Before Heating After Heating Butyrophenone 5.71 4.90 5.42 6.93 5.08 7.31 7.96 Valerophenone 5.20 4.48 4.70 8.00 9.62 8.76 7.34 Phenanthrene 7.23 3.12 6.62 4.91 8.43 0.10 0.34 Pyrene 8.92 3.68 8.41 5.63 5.72 0.81 1.03 *n=3, arbitrary unit, aExtraction conditions: 40 cm x 0.25 mm i.d. x 0.25 m sol-gel TiO2-PDMS-coated microext raction capillary; extraction time: 40 min (gravity-fed at room temperature); HPLC conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase com position of 70:30 (v/v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature.

PAGE 161

1403.6 Conclusion To the best of our knowledge, this is th e first report on the cr eation and use of a sol-gel TiO2-PDMS coating in solid-phase microextraction. Sol-gel TiO2-PDMS-coated microextraction capillaries po ssess excellent pH stabilities and retain their extraction characteristics intact even after prolonged treatment with highly alkaline (pH=13) NaOH or highly acidic (pH=1) HCl solu tions. In addition, sol-gel TiO2-PDMS-coated microextraction capillaries demonstrated ex cellent temperature and solvent stabilities. Direct chemical bonding of th e sol-gel coatings to capilla ry inner walls provides these coatings with excellent solvent resistance, and makes sol-gel TiO2-PDMS-coated microextraction capillaries very much suit able for on-line sample preconcentration in CME-HPLC analysis. The newly developed sol-gel TiO2-PDMS coating was effectively used for the extraction of different classes of analytes with good extraction sensitivity, and run-to-run repeatabilit y. Low ng/mL and sub-ng/mL level (0.15 ng/mL – 11.60 ng/mL) detection limits were achieved for PA Hs, ketones, and alkylbenzenes in CMEHPLC analysis using the ne wly constructed sol-gel TiO2-PDMS-coated microextraction capillary in conjunction with UV detection. Through proper optimization of experimental conditions for sol-gel coating and the capilla ry microextraction pr ocesses it should be possible to further enhance the extraction se nsitivity. For volatile and thermally stable analytes, use of sol-gel TiO2-PDMS-coated capillaries in CME-GC should provide significant enhancemen t in sensitivity. 3.6 References for Chapter Three [1] J. Whitehead, Titanium Compounds Inorgani c in Kirk Othmer Encyclopedia of

PAGE 162

141 Chemical Technology, 3rd ed., Wiley, NY, 1983. [2] M. Anpo, Res. Chem. Intermed. 11 (1989) 67. [3] A. Mills, R.H. Davies, D. Worsley, Chem. Soc. Rev. 22 (1993) 417. [4] B. Aiken, W.P. Hsu, E. Matijevic, J. Mater. Sci. 25 (1990) 1886. [5] T. Rentschler, Eur. Coatings J. 10 (1997) 939. [6] Millennium Chemical, Titanium diox ide crystal types, www.millenniumchem. com, (2005). [7] A. Berthod, J. Chromatogr. 549 (1991) 1. [8] J. Nawrocki, J. Chromatogr. A 779 (1997) 29. [9] C.P. Jaroniec, M. Jaroniec, M. Kruk, J. Chromatogr. A 797 (1998) 93. [10] A. Nahum, C. Horvath, J. Chromatogr. 203 (1981) 53. [11] J.J. Kirkland, M.A.V. Straten, H.A. Claessens, J. Chroma togr. A 691 (1995) 3. [12] J.J. Kirkland, J. Chromatogr. Sci. 34 (1996) 309. [13] K.K. Unger, Porous Silica, its Propert ies and Use as Support in Column Liquid Chromatography, 3rd ed ., Elsevier, NY, 1986. [14] R.K. Iler, The Chemistry of Silica, Wiley-Interscience, NY, 1979. [15] J.J. Pesek, J.E. Sandoval, M. Su, J. Chromatogr. 630 (1993) 95. [16] J.W. Li, P.W. Carr, An al. Chem. 69 (1997) 2193. [17] M. Kawahara, H. Nakamura, T. Naka jima, J. Chromatogr. 515 (1990) 149. [18] K. Murayama, H. Nakamura, K. Takahashi, A. Yoshida, Anal. Sci. 10 (1994) 815. [19] U. Trudinger, G. Muller, K.K. Unger, J. Chromatogr. 535 (1990) 111. [20] K. Tani, Y. Suzuki, J. Chromatogr. A 722 (1996) 129. [21] P. Tsai, C.T. Wu, C.S. Lee, J. Chromatogr. B 657 (1994) 285.

PAGE 163

142 [22] Z.T. Jiang, Y.M. Zuo, Anal. Chem. 73 (2001) 686. [23] K.A. Kraus, H.O. Phillips, J. Am. Chem. Soc. 78 (1956) 249. [24] K. Tani, Y. Suzuki, J. Liq. Chro matogr. Rel. Technol. 19 (1996) 3037. [25] J.J. Pesek, M.T. Matyska, J. Ramakrishnan, Chromatographia 44 (1997) 538. [26] M.R. Buchmeiser, J. Chromatogr. A 918 (2001) 233. [27] S. Sato, S. Oimatsu, R. Takahashi, T. Sodesawa, F. Nozaki, Chem. Commun. 22 (1997) 2219. [28] Y. Ikeguchi, H. Nakamura Anal. Sci. 13 (1997) 479. [29] A.Y. Fadeev, T.J. McCarthy, J. Am. Chem. Soc. 121 (1999) 12184. [30] K.V.P.M. Shafi, A. Ulman, X. Yan, N. L. Yang, M. Himmelhaus, M. Grunze, Langmuir 17 (2001) 1726. [31] J. Winkler, S. Marme, J. Chromatogr. A 888 (2000) 51. [32] K.D. Nugent, W.G. Burton, T.K. Slatte ry, B.F. Johnson, L.R. Snyder, J. Chromatogr. 443 (1988) 381. [33] T.S. Kephart, P.K. Dasgupta, Anal. Chim. Acta 414 (2000) 71. [34] S. Miyazaki, M.Y. Miah, K. Morisato, Y. Shintani, T. Kuroha, K. Makanishi, J. Sep. Sci. 28 (2005) 39. [35] R.E. Paproski, J. Cooley, C.A. Lucy, J. Chromatogr. A 1095 (2005) 156. [36] K. Hata, H. Morisaka, K. Hara, J. Mi ma, N. Yumoto, Y. Tatsu, M. Furuno, N. Ishizuka, M. Ueda, Anal. Biochem. 350 (2006) 292. [37] C. Fujimoto, Electrophoresis 23 (2002) 2929. [38] L. Xu, Y.Q. Feng, Z.G. Shi, S.L. Da Y.Y. Ren, J. Chromatogr. A, 1028 (2004) 165. [39] M. Kawachi, M. Yasu, T. Edahiro, Electron. Lett. 19 (1983) 583. [40] H. Henry, R.F. Kazarinov, H.J. Lee, K.J. Orlowsky, L.E. Katz, Appl. Opt. 26 (1987) 2621.

PAGE 164

143 [41] H. Sankur, W. Gunning, J. Appl. Phys. 66 (1989) 4747. [42] R.R.A. Syms, A.S. Holmes, J. Non-Cryst. Solids 170 (1994) 223. [43] T.A. Egerton, in D. Bloor, R.J. Brook, M.C. Flemings, S. Mahajan (eds.), Encyclopedia of Advanced Mate rials, Pergamon, Oxford, 1994. [44] K.N.P. Kumar, K. Keizer, A.J. Bur ggraaf, J. Mater. Chem. 3 (1993) 1141. [45] L.Q. Wu, P. Huang, N.P. Xu, J. Shi, J. Membr. Sci. 173 (2000) 263. [46] K. Tani, Y. Suzuki, Ch romatographia 38 (1994) 291. [47] K. Tani, Y. Suzuki, Ch romatographia 46 (1997) 623. [48] S. Matsuda, A. Kato, Ap pl. Catal. 8 (1983) 149. [49] K.I. Hadjiivanov, D.G. Klissurski, Chem. Soc. Rev. 25 (1996) 61. [50] B. O’Regan, M. Gratzel, Nature 353 (1991) 737. [51] T. Sekiya, S. Kamei, S. Kurita, J. Lumin 87-89 (2000) 1140. [52] T. Ohno, F. Tanigawa, K. Fujihara, S. Izumi, M. Matsumura, J. Photochem. Photobiol. A 127 (1999) 107. [53] J. Grzechulska, M. Hamerski, A.W. Morawski, Water Res. 34 (2000) 1638. [54] H. Zheng, P.C. Maness, D.M. Blake, E.J. Wolfrum, S.L. Smolinski, W.A. Jacoby, J. Photochem. Photobiol. A 130 (2000) 163. [55] T. Kunitake, S.W. Lee, Anal. Chim. Acta 504 (2004) 1. [56] U. Schubert, J. Mater. Chem. 15 (2005) 3701. [57] S. Kubota, K. Johkura, K. Asanuma, Y. Okouchi, N. Ogiwara, K. Sasaki, T. Kasuga, J. Mater. Sci. 15 (2004) 1031. [58] X. Quan, S. Yang, X. Ruan, H. Zhao, Environ. Sci. Technol. 39 (2005) 3770. [59] N.I. Tatarenko, A. Mozalev, Solid State Electron. 45 (2001) 1009. [60] Z. Tun, J.J. Noel, D.W. Shoesmith, J. Electrochem. Soc. 146 (1999) 988. [61] O.K. Varghese, G.K. Mor, C.A. Grimes, M. Paulose, N. Mukherjee, J. Nanosci.

PAGE 165

144 Nanotechnol. 4 (2004) 733. [62] S.L. Chong, D. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [63] A. Malik, S.L. Chong, In J. Pawlisz yn (Ed.), Application of Solid-phase Microextraction, Royal So ciety of Chemistry: Cambridge, U.K, 1999, pp. 73-91. [64] Z. Wang, C. Xiao, C. Wu, H. Ha n, J. Chromatogr. A 893 (2000) 157. [65] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [66] Z. Zeng, W. Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [67] S. Bigham, J. Medlar, A. Kabir, C. Sh ende, A. Alli, A. Malik, Anal Chem. 74 (2002) 752. [68] R.K. Iler, J. Colloid Interf. Sci. 55 (1976) 25. [69] S.V.M.D. Moraes, J.B. Passos, P. Schossler, E.B. Caramao, C.C. Moro, T.M.H. Costa, E.V. Benvenutti, Talanta 59 (2003) 1039. [70] J.D. Hayes, Ph.D. Dissertation, Univer sity of South Florida, USA, 2002. [71] J.D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3. [72] S. Bigham, J. Medlar, A. Kabir, C. Sh ende, A. Alli, A. Malik, Anal Chem. 74 (2002) 752. [73] C.F. Poole, S.K. Poole, In E. Heftma nn (Ed.) Chromatography, fifth ed., Part A: Fundamentals and Techniques, Journal of Chromatography Library 51A, Elsevier, Amsterdam, 1992. [74] C.L. Woodley, R.C. Kong, B.E. Richter, M. L. Lee, J. High Resolut. Chromatogr. Chromatogr. Commun. 7 (1984) 329. [75] J. Bouche, M. Verzele, J. Gas Chromatogr. 6 (1968) 501. [76] L.G. Blomberg, J. Microcolumn Sep. 2 (1990) 62. [77] D.X. Wang, S.L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [78] C.J. Brinker, G.W. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing, Academic Press, New York, NY, 1990.

PAGE 166

145 [79] D.C. Bradley, R.C. Mehrotra, D.P. Ga ur, Metal Alkoxides, Academic Press, London, 1978. [80] S. Doeuff, M. Henry, C. Snachez, J. Li vage, J. Non-Cryst. Solids 89 (1987) 206. [81] I. Moriguchi, Y. Tsujigo, Y. Teraoka, S. Kagawa, J. Phys. Chem. B 104 (2000) 8101. [82] J. Livage, M. Henry, C. Sanchez, Prog. Solid St. Chem. 18 (1988) 259. [83] R.C. Mehrotra, R. Bohra, D.P. Gaur, In Metal -Diketonates and Allied Derivatives, Academic Press, London, 1978. [84] M. Emili, L. Incoccia, S. Mobilio, G. Fa gherazzi, M. Guglielmi, J. Non-Cryst. Solids 74 (1985) 11. [85] M.M. Haridas, S. Datta, J.R. Be llare, Ceram. Int. 25 (1999) 601. [86] N.T. McDevitt, W.L. Baun, Spectrochim. Acta 20 (1964) 799. [87] L. Tellez, J. Rubio, F. Rubio, E. Morales, J. L. Oteo, J. Mater. Sci. 38 (2003) 1773. [88] J. Wen, J.E. Mark, J. Appl. Polym. Sci. 58 (1995) 1135. [89] Q. Chen, N. Miyata, T. Kokubo, T. Nakamu ra, J. Biomed. Mate r. Res. 51 (2000) 605. [90] N.L. Wu, S.Y. Wang, I.A. Ru sakova, Science 285 (1999) 1375. [91] A.Y. Fadeev, R. Helmy, S. Ma rcinko, Langmuir 18 (2002) 7521. [92] S. Sakka, K. Kamiya, J. Non-Cryst. Solids 42 (1980) 403. [93] M.R. Boccuti, K.M. Rao, A. Zecchina, G. Leofanti, G. Petrini, Stud. Surf. Sci. Catal. 48 (1989) 133. [94] D.C.M. Dutoit, M. Shneider, A. Baiker, J. Catal. 153 (1995) 165. [95] K.R. Alhooshani, Sol-Gel Zirconiaand Titania-Based Surface-Bonded Hybrid Organic-Inorganic Coatings for Sample Preconcentration and Analysis via Capillary Microextraction in Hyphena tion with Gas Chromatography (CMEGC). Ph.D. dissertation, Department of Chemistry, University of South Florida, Tampa, FL 2005.

PAGE 167

146 [96] A.B. Brennan, G.L. Wilkes, Polymer 32 (1991) 733. [97] J.L. Glajch, J.J Kirkland, L. K ohler, J. Chromatogr. 384 (1987) 81. [98] J.J. Kirkland, J.L.Glajch, R.D. Farlee, Anal. Chem. 61 (1989) 2. [99] H. Engelhardt, Hochdruck-Flussigkeits-C hromatographie, second revised ed., Springer, Berlin, 1977. [100] N.M. Djordjevic, P.W.J. Fowler, F. Houdiere, J. Microc olumn Sep. 11 (1999) 403. [101] F.D. Antia, C. Horvath, J. Chromatogr. 435 (1988) 1. [102] G. Stokes, Trans. Cambridge Philos. Soc. 9 (1856) 5. [103] A. Einstein, Ann. Phys. (Leipzig) 324 (1906) 289. [104] A. Einstein, Ann. Phys. (Leipzig) 339 (1911) 591. [105] B. Yan, J. Zhao, J.S. Brown, J. Black well, P.W. Carr, Anal. Chem. 72 (2000) 1253. [106] P.W. Carr, J.D. Thompson, Anal. Chem. 74 (2002) 1017. [107] T. Greibrokk, T. Anderson, J. Chromatogr. A 1000 (2003) 743. [108] Y. Yang, L.J. Lamm, P. He, T. Kondo, J. Chromatogr. Sci. 40 (2002) 107. [109] J.W. Dolan, LC-GC Eur. 16 (2002) 394. [110] Y. Yang, LC-GC Eur. 2 (2003) 37. [111] S.J. Marin, B.A. Jones, W.D. Felix, J. Clark, J. Chroma togr. A 1030 (2004) 255. [112] J.A. Blackwell, P.W. Carr, J. Chromatogr. 549 (1991) 43. [113] C. McNeff, L. Zigan, K. Johnson, P.W. Carr, A. Want, A.M. Weber-Main, LCGC 18 (2000) 514. [114] ZirChrom Column Care Companion, ZirChrom Separations, Anoka, MN, 1999. [115] K. Alhooshani, T.-Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1.

PAGE 168

147 CHAPTER FOUR SOL-GEL TITANIA-SILICA HYBRID ORGANIC-INORGANIC COATING FOR THE EXTRACTION OF POLAR ANALY TES WITH ON-LINE CME-HPLC AND OFF-LINE CME-GC 4.1 Introduction In General, polar analytes such as alcohols, phenols, and carboxylic acids are hydrophilic and show pronounced affinity towa rd aqueous media. This makes their isolation and preconcentration from an aqueous environment by conventional solid-phase microextraction (SPME) a diffe rent analytical problem. For efficient extraction of such hydrophilic analytes from aqueous samples, the SPME coating must possess high polarity. However, it is not an easy task to immobilize polar stationary phases (or sorbent coatings) on silica substrates using conventional techni ques [1]. If the SPME coating and the fusedsilica fiber are not chemically bonded, as is ge nerally the case in conventionally prepared SPME coatings, they often show low thermal a nd/or solvent stabilities [2]. Use of such coatings for the extraction of polar analytes from the aqueous sample matrices commonly leads to desorption and sample carryover problems. In 1995, Pan and co-workers [3] demonstr ated the extracti on of underivatized short chain volatile fatty aci ds from aqueous media usi ng headspace SPME (HS-SPME) with in-fiber derivatization on a poly(acrylate) (PA)-coated fiber. Lopes and Augusto [4] prepared the sol-gel-based polydimethylsil oxane/divinylbenzene (PDMS/DVB) sorbent, and reported the improved thermal stability of the composite phase compared to sol-gel

PAGE 169

148 PDMS sorbent. Other types of polymer blends such as Ca rboxen/polydimethylsiloxane/ (CAR/PDMS) [5,6], Carbowax/divinylbenzen e (CW/DVB) [6-8], polydimethylsiloxane/ Carboxen/divinylbenzene (PDMS/ CAR/DVB) [6] have been al so used for extraction of highly polar compounds. Yun [9] reporte d the use of the open crown ether diallyltriethylene glycol/hydr oxyl-terminated silicon oil ( DATEG/OH-TSO) as extracting phases in SPME using sol-gel coating technol ogy and cross-linking reaction and showed enhanced extraction efficiency for both pol ar and nonpolar analytes compared to PDMS and CW/DVB-coated fibers. Liu and co-wor kers [10] reported the use of hydroxylterminated silicon oil-butyl methacryl ate-divinylbenzene (OH-TSO-BMA-DVB) copolymers for the extraction of polar alco hols and fatty acids, and nonpolar esters. Recently, Campins-Falco and co-workers [11] reported the use of SPME fiber with a Carbowax-templated resin for on-fiber deriva tization of trimethylamine in water and air samples for HPLC analysis. Kulkarni a nd co-workers [12] developed a sol-gel cyanopropyl-PDMS (CN-PDMS) coating contai ning highly polar cyanopropyl (CN) and nonpolar poly(dimethylsiloxane) (PDMS) for extraction and preconc entration of both nonpolar and highly polar analytes from a queous sample media without additional derivatization process. 4.1.1 Polyethylene glycols (PEGs) as sorb ent in solid-phase microextraction (SPME) Extraction of polar compounds from aque ous sample matrices is typically challenging due to the hydrophilic affinity of analytes towards water. Since the hydrophilic nature of PEGs may be a suitable choice for the extraction of polar analytes, high molecular weight PEGs such as Carbowax 20M, were used as composite coatings in SPME [5,7]. However, in SPME-GC applica tions low thermal stability of PEG fiber

PAGE 170

149 coating was a major shortcoming responsible for incomplete desorption and sample carryover problems. Recently, Wang and co-workers [13] reported on utilizing Superox-4 (polyethylene glycol, PEG) in sol-gel t echnology for SPME-GC, and demonstrated many advantageous features over the conventi onal SPME fibers, incl uding high thermal stability, long life time, and faster mass tr ansfer rate. Malik and co-workers [14,15] reported an effective immobilization of solgel PEGs on inner walls of fused-silica capillaries and used these sol-gel PEG coati ngs as GC stationary phase in separation columns and as extracting media for capillary microextraction (CME also called in-tube SPME). Silva and Augusto [16] prepared SPME fibers with Carbowax 20M-modified Ormosil (organically modified silica) using sol-gel process, and de monstrated superior extraction efficiencies compar ed with commercial SPME fibers coated with PDMS and CW/DVB. They also mentioned in their pa per that keeping the PEG fiber at high temperature for a long period (at 230 C, 50 hr) improved the precision of analysis evidenced by lower RSD values. This phenomenon was pointed out by Sato and coworkers [17] that the presence of PEGs in so l-gel matrix contribute s in controlling pore size distribution to give a porous structure which drastically in creases the surface area of the extracting phase. 4.1.2 Applications of low molecular weight PEG, N-(triethoxysilylpropyl)-Opolyethylene oxide urethane (TESP-PEO) Successful immobilization of comparativel y high molecular weights polyethylene glycols (PEGs) by sol-gel process has been re ported in several papers [13-15]. However, high molecular weight PEGs have lower polarities than their low molecular weight

PAGE 171

150 counterparts, which may limit their suitability for extracting highly polar analytes from aqueous matrices. Recently, N -(triethoxysilylpropyl)-O-poly ethylene oxide urethane (TESP-PEO) was used on silica surfaces to achieve effective surface passivation of microfluidic devices in orde r to avoid cell adhesion and ad sorption to the silicon and glass surfaces via self-assemb led monolayer (SAM) [18,19]. For sol-gel CME applications, the pres ence of sol-gel-active trialkoxysilane groups at one end in TESP-PEO will allow for the creation of sol-gel organic-inorganic hybrid titania coatings on inner walls of fused-silica capillaries In addition, hydrophilic nature of PEO chains in the sol-gel TiO2-SiO2-TESP-PEO coatings will allow extracting moderately polar and highly polar analyt es from aqueous samples for capillary microextraction hyphenated with GC and HPLC. 4.2 Experimental 4.2.1 Equipment On-line CME-HPLC experiments with sol-gel TiO2-SiO2-TESP-PEO-coated capillaries were carried out on a Micro-Tech Scientific (V ista, CA) Ultra Plus HPLC system with a variable wavelength UV det ector (Linear UVIS 2000). Off-line CME-GC experiments were performed on a Shimadzu Model 17A GC (Shimadzu, Kyoto, Japan) equipped with flame ionization detection (FID) system and a split-splitless injector. Online data collection and processing were done using Chrom-Perfect (version 3.5 for Windows) computer software (Justice Laboratory Software, Denville, NJ). An in-house built gas pressure-operated capillary filling/purging device [20] was used to rinse the fused-silica capillary with solv ents, fill the extraction capilla ry with sol solution, expel

PAGE 172

151 the sol solution from the capillary at the e nd of sol-gel coating process, and purge the capillary with helium. An in-house-designed liquid sample dispenser [14] was used to facilitate gravity-fed flow of aqueous samples through the sol-gel microextraction capillary. A Barnstead model 04741 Nanopure deionized water system (Barnstead/Thermolyne, Dubuque, IA) was used to obtain 16.0 M -cm water. A Fisher model G-560 Vortex Genie 2 system (Fisher Scientific) was used for thorough mixing of the sol-gel ingredient in the coating solution. A Mi crocentaur model APO 5760 centrifuge (Accurate Chemical and Scien tific Corp., Westbury, NY) was used for centrifugation of sol soluti ons. A Chemcadet model 5984-5 0 pH meter (Cole-Palmer Instrument Co., Chicago, IL) equipped with a TRIS-specific pH electrode (SigmaAldrich, St. Louis, MO) was used to measur e the buffer pH. A Nicolet model Avatar 320 FTIR (Thermo Nicolet, Madison, WI) was us ed for FTIR measurem ents. A JEOL model JSM-35 scanning electron microscope (SEM) wa s used for the investigation of surface morphology of the sol-gel TiO2-SiO2-TESP-PEO-coated capillaries. A reversed-phase Betasil-C8 (Thermo Electron Co., 12.5 cm x 4.0 mm i.d., 5 m dp) and Betabasic 8 (Thermo Electron Co., 10 cm x 4.6 mm i.d., 5 m dp) columns were used for HPLC separation of the extracted analytes. 4.2.2 Chemicals and materials Fused-silica capillary (250 m i.d.) was purchased from Polymicro Technologies Inc. (Phoenix, AZ). HPLC-grade solvents (a cetonitrile, dichlorometh ane, and methanol), Kimwipes, polypropylene microcentrifuge tubes (2.0 mL), and 7.0 mL borosilicate vials (used to store standard solutions) were purchas ed from Fisher Scientific (Pittsburgh, PA).

PAGE 173

152 Titanium (IV) isopropoxide (99.999 %), 1butanol (99.4+ %), methyltrimethoxysilane (MTMOS, 98%), trifluoroace tic acid (TFA, 99%), potassium phosphate monobasic (KH2PO4, 99.99%), aromatic aldehydes ( p -anislaldehyde, benzaldehyde, 4isopropylbenzaldehyde), aniline deri vatives (benzanilide, acridine, N N -dimethylaniline, N -butylaniline), substituted phenols (2 -chlorophenol, 2-methoxy-4-methylphenol, 4chloro-3-methylphenol, 4-tert-butylphenol, 2,4-dichlo rolphenol, 2,4,6-trichlorophenol, pentachlorophenol), and free aliphatic fatty acids (octanoic acid, nonanoic acid, decanoic acid, and undecanoic acid), were purchased from Aldrich (Milwaukee, WI). N (triethoxysilylpropyl)-O-polyethylene oxide urethane (TESP-PEO, 95%) was purchased from Gelest (Morrisville, PA). 4.2.3 Preparation of the sol solution The sol solution was prepared by thoroughly vortexing the follow ing reagents in a 2-mL polypropylene centrifuge tube: a sol-gel-active organic polymer ( N (triethoxysilylpropyl)-O-polyethylene oxide ur ethane (TESP-PEO), 50 mg), two sol-gel precursors (titanium (IV) isopropoxide, 25 L and methyltrimethoxysilane (MTMOS), 50 L), two solvents (methylene chloride and 1-butanol, 200 L each), and a sol-gel catalyst/chelating agent (TFA containing 10% H2O, 100 L). The content of the tube was then centrifuged for 4 min (at 13000 rpm; 15682 x g ). Finally the top clear solution was transferred to another clean vial by decanta tion, and was further used for coating the fused-silica microextraction capil lary (250 m i.d.). The chemical ingredients used in the sol-gel coating solutions ar e represented in Table 4.1.

PAGE 174

153Table 4.1 Names, functions, and chemical structur es of the coating solution ingredients used to prepare sol-gel TiO2-SiO2-TESP-PEO-coated microextrtaction capillaries. Ingredient Function Chemical structure Titanium (IV) isopropoxide Sol-gel precursor Ti OCH(CH3)2 OCH(CH3)2 (H3C)2HCO OCH(CH 3 ) 2 Methyltrimethoxysilane (MTMOS) Sol-gel precursor SiOCH3 OCH 3 H3CO CH3 N -(triethoxysilylpropyl)-Opolyethylene oxide urethane (TESP-PEO) Sol-gel-active polymer NH C O O (CH2CH2O)4-6H SiOC2H5OC2H5 C2H5O Trifluoroacetic acid (TFA) Catalyst/Bridging (chelating) ligandHO O CF3

PAGE 175

1544.2.4 Preparation of sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary A 60-cm long hydrothermal treated [ 15,21] fused-silica capillary (250 m i.d.) was installed on an in-house built gas pressu re-operated capillary filling/purging device [20] (Figure 2.4 in Chapter 2), and the capillary was filled with the prepared sol solution under 10 psi helium pressure. After filling, the sol solution was kept inside the capillary for 15 min to facilitate the creation of a su rface-bonded coating due to sol-gel reactions taking place in the coating and on the capillary inner surface. Following this, the unbonded portion of the sol solution was expe lled from the capillary under helium pressure (20 psi), leaving behind a surface-bond ed sol-gel coating on the inner surface of the fused-silica capillary. The capillary was further purged and dried with helium for 30 min under the same pressure. The coated microe xtraction capillary wa s then installed and conditioned in a GC oven by temper ature-programmed heating from 40 C to 300 C at 1 C/min under helium purge. The capillary was held at 300 C for 60 min. Finally, the capillary was cooled down to room temperat ure and rinsed with methylene chloride and methanol (3 mL each). Following this, the cap illary was installed in the GC oven for drying and further thermal conditioning und er temperature-programmed heating as described above, with exception that this time the capillary was held at the final temperature for 30 min. The conditioned capi llary was then cut in to 40-cm and 12-cm long pieces that were further used in CMEHPLC and CME-GC analyses, respectively. 4.2.5 Capillary microextraction (CME ) and on-line CME-HPLC analysis Selected analytes were prepared in me thanol (10 mg/mL) and stored in surfacedeactivated amber glass vials fo r use as stock solutions. Ea ch stock soluti on was further

PAGE 176

155 diluted 100 times (100 g/mL) in methanol, and stored in separate surface-deactivated amber glass vials before to prepare the a queous sample solutions. For extraction, fresh aqueous samples were prepared by further diluting these stock solutions in deionized water to 1 g/mL or lower concentrations. A schematic of the CME-HPLC setup for online capillary microextrac tion and HPLC analysis is presented in Figure 3.4 in Chapter 3. The HPLC column was pre-equilibrated with selected mobile pha se consisting of a mixture of acetonitrile and water (or buffer solution), for example. A 40-cm segment of the sol-gel TiO2-SiO2-TESP-PEO-coated microextractio n capillary was mounted on the HPLC injection port as an external sampling loop. Analytes were preconcentrated in the sol-gel TiO2-SiO2-TESP-PEO coating by passing the aqueous sample from a gravity-fed dispenser [14] throug h this sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary for 40 min. The analytes extracted in the sol-gel TiO2-SiO2-TESP-PEO coating of the sampling loop were then transferred in to the HPLC column by desorbing with the mobile phase. This was accomplished by simply switching the injection valve from the “load” to “inject” position. Th e injected analytes were then separated on the HPLC analysis column under isocratic elution conditions until all peaks representing each analyte were eluted, typica lly in 10 or 15 minutes. 4.2.6 Off-line CME-GC analysis Selected free fatty acid samples and stoc k solutions were prepared as described earlier. For extraction, fresh aqueous samples were prepared by further diluting the stock solutions in deionized water to 1 g/mL or lower concentrations. A Chromaflex AQ column (Knotes Glass, Vineland, NJ) was modi fied as described in Figure 4.1 and used for gravity-fed sample delivery in capillary microextraction. A 12-cm long piece of

PAGE 177

156 thermally conditioned sol-gel TiO2-SiO2-TESP-PEO-coated micr oextraction capillary (250 m i.d.) was vertically connected to th e lower end of the sample dispenser. The aqueous sample containing trace amounts of fatty acids was poured into the dispenser from its top end and allowed to flow through the microextraction cap illary under gravity. The extraction was carried out for 40 min for equilibrium to be established. The capillary was then detached from the dispenser and th e residual sample droplets were removed by touching one of the ends of microextraction ca pillary with Kimwipe tissue. After this, the capillary was installed in the GC injection port, keeping ~ 3 cm of its lower end protruding into the GC oven. This end was th en interfaced with the inlet of a GC capillary column using a deactivated two-way press–fit quartz connector. Under splitless conditions, the extracted analytes were then thermally desorbed from the capillary by rapidly raising the temperature of the injection port (from 30 oC to 300 oC at 60 oC/min), while keeping the GC oven temperature at 35 oC. Such a rapid temperature program of the injection port facilitated effective desorpti on of the extracted analytes from the sol-gel TiO2-SiO2-TESP-PEO microextraction capillary and their focusing at the inlet of the GC analysis column. Following this, the GC oven was temperature programmed from 35 oC to 300 oC at rate of 20 oC/min to achieve separation of th e focused analytes on the GC column. A flame ionization detect or (FID) maintained at 350 oC was used for analyte detection.

PAGE 178

157 Figure 4.1 Gravity-fed sample delivery system (1.0 x 60 cm, Glass barrel only) for capillary microextraction. Adapted from ref. [22].

PAGE 179

1584.2.7 Safety precautions The presented work involved the use of various chemicals (org anic and inorganic precursors) and solvent that might be envi ronmentally hazardous with adverse health effects. Proper safety measures were taken in handling organic solvents such as methanol, methylene chloride, and acetonitrile. All chemicals once used, were disposed of in the proper waste containers to ensure personnel and environmental safety. 4.3 Results and Discussion The main purpose of this research was to develop surfacebonded sol-gel titaniabased polar coatings to facilita te effective extraction of polar analytes in aqueous sample matrices as well as hyphenati on of capillary microextractio n (CME) with HPLC. Sol-gel chemistry allowed us to create a surface-bonde d hybrid organic-inorga nic titania coating of desired properties on the inner walls of a fu sed-silica capillary in CME. As a versatile tool, sol-gel technology has been effectively utilized to create su rface-bonded coatings on the outer surface of conventional SPME fibers [23] as well as on the inner walls of fusedsilica capillary for use in CME (or in-tube SPM E) [14,24-26]. In the present work, sol-gel chemistry was utilized to create a hybrid coating based on titania and a low molecular weight TESP-PEO on the inner surface of fusedsilica capillary for efficient extraction and CME-HPLC analysis of trace amounts of polar analytes from aqueous samples. 4.3.1 Sol-gel reactions for the preparation of sol-gel TiO2-SiO2-TESP-PEO coating As described earlier in Chapter 3, ch emical reactivity and complex-forming ability of titanium alkoxides differ from silic on alkoxides. It is we ll-known that silicon alkoxides are capable of undergoing hydrol ytic polycondensation reactions in the

PAGE 180

159 presence of a sol-gel catalyst [27]. However, compared with silica-based systems, titaniabased sol-gel reactions are much faster [28] and often require a chelating (or bridging) agent to control and decelerate the sol-gel process [29]. Because of this, developing titania-based organic-inorgani c hybrid materials usually requ ires the use of the optimum ratio of catalyst (or inhibitor) and water. Otherwise, the gelation or precipitation often takes place instantaneously as the sol-ge l solution ingredients are mixed together, especially in hybrid systems. In this res earch, trifluoroacetic acid [30] was used as chelating (or bridging) agents for titanium alkoxide, but cata lyst for silicon alkoxides. In the present work, sol-gel TiO2-SiO2-TESP-PEO-coated capillaries were prepared through hydrolytic pol ycondensation reac tions performed within fused-silica capillaries followed by thermal conditioning of the created coatings to achieve fine surface structures. The N-(triethoxysilylpr opyl)-O-polyethylene oxide urethane (TESPPEO) ligand possesses polar polyethylene oxide repeating units at one end and sol-gelactive triethoxysilane groups at the other. Via sol-gel active groups, TESP-PEO was chemically bonded to the other sol-gel precurs ors as well as to th e inner surface of the fused-silica capillary. The sol-gel process fo r the generation and ch emical immobilization of the coating involves: (A) hydrolysis of th e titanium alkoxide precursor [31] and the alkoxysilane compounds, MTMOS and TESP-PEO (Scheme 4.1), (B) polycondensation and chemical incorporation of the hydrolysis products into a three-dimensional sol-gel network [32-35] (Scheme 4.2), a nd (C) chemical anchoring of the sol-gel hybrid polymer to the inner walls of the fused-si lica capillary [32,33] (Scheme 4.3).

PAGE 181

160TiOCH(CH3)2 OCH(CH3)2 (H3C)2HCO 4 H2O TiOH OH HO Chelating Agent4 (CH3)2CHOH Titanium (IV) Isopropoxide Titanium Hydroxide ++ OH OCH(CH3)2 SiOCH3 OCH3 H3CO CH3 3 H2O 3 CH3OH SiOH OH HO CH3 Methyltrimethoxysilane (MTMOS)Catalyst 3 H2O 3 C2H5OH N-(triethoxysilylpropyl)-Opolyethylene oxide urethane (TESP-PEO) NH C O O (CH2CH2O)4-6H SiOH OH HO NH C O O (CH2CH2O)4-6H SiOC2H5OC2H5 C2H5O Catalyst(A) Scheme 4.1 (A) Hydrolysis of titanium (IV) isopropoxide and the alkoxysilane compounds.

PAGE 182

161Ti O HO O O Ti O Si O CH3 O SiO O O SiO OH z x SiOH OH HO CH3 NH C O O (CH2CH2O)4-6H Ti OH O NH C O O (CH2CH2O)4-6H NH C O O (CH2CH2O)4-6H SiOH OH HO n H2O TiOH OH HO OH y OH OH (B) Scheme 4.2 (B) Polycondensation and chemical inco rporation of the hydrolysis products with the evolving sol-gel network.

PAGE 183

162 Ti O O O Ti O Si O O SiO O O SiO O NH C O O (CH2CH2O)4-6H Ti OH O NH C O O (CH2CH2O)4-6H O Si (C) O O O Si O O Si O Si O Si O Wall bonded TiO2-TESP-PEO coating OH OH OH Si OH O O Si OH O Si O Si O Si O Ti O HO O O Ti O Si O CH3 O SiO O O SiO OH NH C O O (CH2CH2O)4-6H Ti OH O NH C O O (CH2CH2O)4-6H OH OH + Si OH O Si Si OH O O O Si Scheme 4.3 (C) Chemical anchoring of the sol-gel TiO2-SiO2 hybrid polymer to the inner walls of the capillary.

PAGE 184

1634.3.2 Scanning electron microscopy (SEM ) of sol-gel titania-silica-TESP-PEO coatings bonded to the inner surface of a fused-silica capillary Figure 4.2 represents the scanning electron micrograph showing the side view of a 250m i.d. fused-silica capillary with sol-gel TiO2-SiO2-TESP-PEO coating with 10000x magnification. As is evident from these images, the sol-gel TiO2-SiO2-TESPPEO coating in the microextraction capillary acquires a fine and thin coating structure, providing remarkable capability of sorption with such a thin coating. Based on the SEM data, the thickness of the sol-gel TiO2-SiO2-TESP-PEO coating was estimated at 0.2 m. This image also shows uniformity of coating thickness in a consistent way in the sol-gel TiO2-SiO2-TESP-PEO-coated microext raction capillaries.

PAGE 185

164 Figure 4.2 Scanning electron microscopic image of a 250m i.d. fused-silica capillary with sol-gel TiO2-SiO2-TESP-PEO coating: side view (10000 ).

PAGE 186

1654.3.3 Fourier transform infrared (FTIR) spectroscopy of the sol-gel titania-silica TESP-PEO surface The formation of Ti-O-Si and Si-O-Si bonds in the prepared sol-gel sorbent was examined by FTIR by performing separate expe riments outside the fu sed-silica capillary. The FTIR experiments were performed by pass ing IR radiation through a thin layer of sol-gel TiO2-SiO2-TESP-PEO coating material that was used in the fused-silica capillary. It has been reported that the characteristic IR band representing Si -O-Ti bonds is located at 940-960 cm-1 [36,37], and Si-O-Si bonds is located at 1000-1200 cm-1 [37-40]. Figure 4.3 shows a FTIR spectrum of the sol-gel Ti-T ESP-PEO coating with a specific band at 946.61 cm–1 for Si-O-Ti bonds and 1070 cm–1 for Si-O-Si bonds. A band for Si-O-Si may overlap with existing C-O-C band wh ich is also located at 1000-1300 cm–1 [41,42]. The FTIR result is indicative of the presence of both Si-O-Ti and Si-O-Si bonds in the sol-gel sorbent used in the fused-silica microextra ction capillaries to perform on-line CMEHPLC analysis.

PAGE 187

166 Figure 4.3 FTIR spectrum of the sol-gel TiO2-SiO2-TESP-PEO coating.

PAGE 188

1674.3.4 Applications of sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary Sol-gel TiO2-SiO2-TESP-PEO-coated capillaries showed excellent affinity for the polar analytes in the aqueous phase, thus enab led the extraction of analytes belonging to moderately or highly polar ch emical classes, such as aldehydes, anilines, phenols, and fatty acids. Due to the environmental a nd toxicological sign ificance, analysis of aldehydes is important in general, and in aquatic and atmo spheric oxidation processes, in particular. More recently, low molecular mass of aldehyde s have been found to be major organic byproducts in disinfection, oxi dation, and ozonation of natura l waters to produce drinking water [43,44]. In addition, aromatic aldehyde products are important intermediate of pharmaceuticals [45], pesticides [46], and dyestuffs [47]. Figure 4.4 illustrates a chromatogram illustrating CME-HPLC analysis of aromatic aldehydes ( p -anislaldehyde, benzaldehyde, 4isopropylbenzaldehyde) using a sol-gel coated TiO2-SiO2-TESP-PEO microextraction capillary. The extraction was performed on a 40 cm x 0.25 mm i.d. fused-s ilica microextraction capillary for 40 min using a gravity-fed sample delivery system at room temperature. The concentrations of aldehydes were in 50-500 ng/mL range. The polar nature of the sol-gel TiO2-SiO2-TESPPEO coating showed good affin ity and detection limits for m oderately polar analytes. In this case, the run-to-run p eak area repeatability was less than 6.91 % RSD. Detection limits for the extracted aldehydes ranged between 0.98 ng/mL for 4isopropylbenzaldehyde to 11.92 ng/mL for p -anisaldehyde in conjunction with UV detection. The aldehydes extracted from an aqueous sample are listed in Table 4.2. Aromatic amines are commonly used as intermediates in the photographic,

PAGE 189

168 pesticide, dye, and pharmaceutical industries [48-50]. Aromatic amin es are also found in air, water and soil [51,52], and many of them have been classifi ed as mutagenic and carcinogenic [53]. Therefore, detection of tracelevel contents of aromatic amines in the environment and in drinking water is very important. Figure 4.5 presents a chromatogram illust rating CME-HPLC analysis of aniline derivatives extracted from an aqueous samp le using a 0.25 mm i.d. sol-gel coated TiO2SiO2-TESP-PEO capillary. Compared to modera tely polar aldehyde samples, aniline derivatives needed lower analyte concen trations (50 ng/mL 300 ng/mL) for CMEHPLC analysis. For aromatic amine compounds the run-to-run peak area repeatability was less than 6.08 % RSD. Detection limits for the extracted anilines ranged from 0.53 ng/mL for acridine to 3.48 ng/mL for N N -dimethylaniline in conjunction with UV detection. From the presented result s it is evident that sol-gel TiO2-SiO2-TESP-PEO coating is able to extract moderately polar analytes with good extr action sensitivity. The extracted ketones from an aqueous sample are listed in Table 4.3. Chlorophenols are present as importan t group of highly toxi c pollutants [54]. Chlorophenols have been widely used in va rious applications such as preservatives, pesticides, antiseptics, and disinfectants [ 55]. They are often f ound in waters [56,57], soils, and sediments [58] and are formed as a result of hydrolysis, oxidation, and microbiological degradation of chlorinated pesticides. Chlorophenols are environmental important compounds.

PAGE 190

169 Figure 4.4 CME-HPLC analysis of aldehyde s. Extraction conditions: 40-cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-SiO2-TEPS-PEO-coated microextra ction capillary; extraction time, 40 min (gravity-fed at room temper ature). Other conditions : 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 70:30 (v/v) ACN/water; 1 mL/min flow rate; UV de tection at 214 nm; ambient temperature. Peaks: (1) p -anislaldehyde (500 ng/mL), (2) benzaldehyde (500 ng/mL), and (3) 4isopropylbenzaldehyde (50 ng/mL).

PAGE 191

170Table 4.2 Physical properties and chemical stru ctures of aldehydes extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Name of Compound MW mp (C) bp (C) d (g/mL) at 25 C Structure of Compound p -Anisaldehyde 136.15 1 248 1.12 H O H3CO Benzaldehyde 106.12 26 178.1 1.0415 H O 4-Isopropylbenzaldehyde 148.21 N/A 234 0.977 H O *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 192

171 Figure 4.5 CME-HPLC analysis of aniline de rivatives. Extraction conditions: 40-cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-TEPS-PEO-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 60:40 (v/v) ACN/water; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature. Peaks: (1) benzanilide ( 70 ng/mL), (2) acridine (50 ng/mL), (3) N N dimethylaniline (300 ng/mL), and (4) N -butylaniline (70 ng/mL).

PAGE 193

172Table 4.3 Physical properties and chemical struct ures of aniline derivatives extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Name of Compound MW mp (C)bp (C) pKa Structure of Compound Benzanilide 197.24 162 117 13.52 C O H N Acridine 179.22 108 346 5.60 N N N Dimethylaniline 121.18 2 193 4.70 N H3C CH3 N -Butylaniline 149.24 14.4 239 5.05 N HCH2CH2CH2CH3 *MW: Molecular weight; mp: Melting point; bp: Boiling point.

PAGE 194

173 Figure 4.6 illustrates on-lin e CME-HPLC analysis of substituted phenols using a TiO2-SiO2-TESP-PEO-coated microextraction capillary. A 50 mM of potassium biphosphate buffer (UV cutoff: < 200 nm (0.1 %)) solution was prepared, and then pH was adjusted to 7 by adding a concentrated NaOH solution to control the pH of the mobile phase based on p Ka of each phenolic sample. Low ng/mL level detection limits were also achieved for these analytes (11.61 – 22.34 ng/mL), using UV detection. Unlike moderately polar analytes, such as aldehyde s and anilines, substituted phenols are highly polar compounds and they favor the aque ous media. However, a sol-gel TiO2-SiO2TESP-PEO extraction capillary can successfu lly compete with the aqueous media towards polar analytes and extract the phenols. The extracted substituted phenols from an aqueous sample are listed in Table 4.4. Figure 4.7 illustrates other example of online CME-HPLC analysis of substituted phenols using a TiO2-SiO2-TESP-PEO-coated microextra ction capillary. Acceptable detection limits were also achieved for these analytes (23.20 – 50.18 ng/mL) using UV detection. The extracted substituted phenols from an aqueous sample are listed in Table 4.5. Fatty acids are carboxylic acids with a long hydrocarbon chain, generally straight, which are important key metabolites and inte rmediates in biological processes [59]. Especially a high level of saturated fatty acids in the diet raises blood cholesterol levels. Due to the hydrophilic characte ristics of underivatized shortchain fatty acids, extraction of fatty acids, especially in aqueous sample matrices, is extremely challenging. Here, solgel TiO2-SiO2-TESP-PEO-coated microextraction ca pillaries demonstrated effective extraction of underivatized fatty acids in aque ous sample media. Due to the lack of

PAGE 195

174 Figure 4.6 CME-HPLC analysis of substituted phenols. Extraction conditions: 40-cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-TEPS-PEO-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 12.5 cm x 4.0 mm i.d. Betasil-C8 column (5 m dp); isocratic elution with mobile phase composition of 60:40 ACN/KH2PO4 (50 mM, pH=7); 1 mL/min flow rate; UV detection at 280 nm; ambient temperature. Peaks: (1) pentachlorophenol (600 ng/mL), (2) 2,4,6trichlorophenol (600 ng/mL), (3) 2-chlorophenol (1.0 g/mL), and (4) 2,4-dichlorophenol (1.5 g/mL).

PAGE 196

175Table 4.4 Physical properties and chemical structures of substituted phenols extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Name of Compound MW mp (C)bp (C) pKa Structure of Compound Pentachlorophenol 266.34175 310 4.92 OH C l Cl Cl Cl Cl 2,4,6-Trichlorophenol 197.4565 246 6.23 OH C l Cl Cl 2-Chlorophenol 128.568 175 8.52 OH Cl 2,4-Dichlorophenol 163.0042 209 7.85 OH C l Cl *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 197

176 Figure 4.7 CME-HPLC analysis of other subst ituted phenols. Extraction conditions: 40cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-TEPS-PEO-coated microextraction capillary; extraction time, 40 min (gravity-fed at room temperature). Other conditions: 12.5 cm x 4.0 mm i.d. Betasil-C8 column (5 m dp); isocratic elution with mobile phase composition of 65:35 ACN/KH2PO4 (25 mM, pH=8); 1 mL/min flow rate; UV detection at 280 nm; ambient temperature. Peak s: (1) 2-methoxy-4-methylphenol (5.0 g/mL), (2) 4-chloro-3-methylphenol (3.0 g/mL), and (3) 4tert -butylphenol (3.0 g/mL).

PAGE 198

177Table 4.5 Physical properties and chemical structures of substituted phenols extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Name of Compound MW mp (C)bp (C) pKa Structure of Compound 2-methoxy-4methylphenol 138.175 221 10.27 OH CH3 OCH3 4-chloro-3methylphenol 142.5964 235 9.549 OH C l CH3 4tert -butylphenol 150.2298 237 10.43 OH CH3 CH3 H3C *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 199

178 sensitive chromophores in aliphatic free fatty acids, CME-GC experiment was performed instead. Figure 4.8 illustrates on-line CME-GC analysis of fatty acids using. Using a TiO2-SiO2-TESP-PEO-coated microextraction capi llary for CME-GC with FID detection enabled to achieve excellent detection limits for these analytes (2.26 – 6.76 ng/mL). Free fatty acids are highly polar analytes and not eas y to extract if they are in aqueous media, however they were well extracted by using a sol-gel TiO2-SiO2-TESP-PEO-coated extraction capillary with low ng/mL level detection limits. The free fatty acids extracted from an aqueous sample are listed in Table 4.6. Table 4.7 summarizes the peak area repeatability and detection limit data for aldehydes, substituted phenols, aromatic amines, and free fatty acids.

PAGE 200

179 Figure 4.8 CME-GC analysis of fatty acids Extraction conditions: 12-cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-TEPS-PEO-coated microextraction capillary; extraction time, 40 min (gravity-fed at ambient temperature) Other conditions: 5 m 250 m i.d. sol–gel PDMS column; splitless injection; injector temperature: initial 30 oC, final 300 oC, programmed at a rate of 60 oC/min; GC oven temperature programmed from 35 oC to 300 oC at a rate of 20 oC/min; helium carrier gas; FID temperature 350 oC. Peaks: (1) octanoic acid (500 ng/mL), (2) nonanoic acid (200 ng/mL), (3) decanoic acid (150 ng/mL), and (4) undecanoic acid (150 ng/mL).

PAGE 201

180Table 4.6 Physical properties and chemical structures of fatty acids extracted from an aqueous sample using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary. Data obtained from www.sigmaaldrich.com. Name of Compound MW mp (C)bp (C) d (g/mL) Structure of Compound Octanoic acid 144.21 16 237 0.91 CH3(CH2)5CH2C O OH Nonanoic acid 158.24 9 268 0.906 CH3(CH2)6CH2C O OH Decanoic acid 172.26 30 270 0.893 CH3(CH2)7CH2C O OH Undecanoic acid 186.29 29 249 0.86 CH3(CH2)8CH2C O OH *MW: Molecular weight; mp: Melting point; bp: Boiling point; d: Density.

PAGE 202

181Table 4.7 Peak area repeatability a nd limits of detection (LOD) data for aldehydes, aniline derivatives, and substituted phenols in CME-HPLC, and fatty acids in CME-GC* using a sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary.a Peak area repeatability (n = 3) Mean peak areaRSD LOD Chemical class Name (Arbitrary Unit)(%) (ng/mL) (S/N = 3)Aldehydes p -Anisaldehyde 11.32 6.91 11.92 Benzaldehyde 14.31 5.92 9.43 4-Isopropylbenzaldehyde13.69 4.84 0.98 Anilines Benzanilide 22.54 5.70 0.94 Acridine 25.30 4.22 0.53 N N -Dimethylaniline 23.22 4.78 3.48 N -Butylaniline 15.43 6.08 1.23 Phenols Pentachlorophenol 13.90 7.62 11.61 2,4,6-Trichlorophenol 13.82 7.14 11.73 2-Chlorophenol 16.89 6.86 15.92 2,4-Dichlorophenol 18.11 5.60 22.34 2-Methoxy-4-methylphenol 26.87 4.11 50.18 4-Chloro-3-methylphenol34.91 3.40 23.20 4tert -Butylphenol 34.11 3.02 22.34 Fatty acids* Octanoic acid 20.03 1.78 6.76 Nonanoic acid 18.37 1.93 2.93 Decanoic acid 15.72 2.44 2.58 Undecanoic acid 17.88 1.76 2.26 aExtraction conditions: 40-cm (12-cm for fatty acids*) x 0.25 mm i.d. x 0.2 m sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capill ary; extraction time, 40 min (gravityfed at ambient temperature).

PAGE 203

1824.3.5 Extraction kinetic profile of sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillary Figure 4.9 illustrates the extraction kinetic profile for: (A) acrid ine and (B) 2chlorophenol on a sol-gel TiO2-SiO2-TESP-PEO-coated micr oextraction capillary. Experimental data for these curves represen ting extraction kinetic profiles were obtained by individually performing capillary microe xtraction for each of the solutes. The microextraction experiments were perfor med using aqueous samples containing 25 ng/mL and 500 ng/mL concentrations of acrid ine and 2-chloropheno l, respectively. A series of capillary microext raction experiments were cond ucted to vary the extraction time for each of the two analytes that were ex tracted from their standard solutions. Three replicate extractions of each analyte were performed for 1, 5, 10, 20, 30, 40, 50, and 60 min. The average HPLC peak area was then pl otted against the extraction time to obtain Figure 4.8. For both acridine and 2-chloroph enol, extraction equilibrium was reached within 40 min as is evidenced by th e plateau on the extraction curve.

PAGE 204

183 Figure 4.9 Illustration of the extraction kinetic profile of acridine ( ), and 2chlorophenol ( ) obtained on a 40 cm 0.25 mm i.d. x 0.2 m sol-gel TiO2-SiO2-TESPPEO-coated microextraction capillary using 25 and 500 ng/mL aqueous solutions, respectively. Extraction conditions are the same as in the Fi gure 4.4. Other conditions: 10 cm x 4.6 mm i.d. Betabasic 8 column (5 m dp); isocratic elution with mobile phase composition of 60:40 (v/v) ACN/water for acridine: 12.5 cm x 4.0 mm i.d. Betasil-C8 column (5 m dp); isocratic elution with m obile phase composition of 60:40 ACN/KH2PO4 (50 mM, pH=7) for 2-chlo rophenol; 1 mL/min flow rate; UV detection at 254 nm; ambient temperature.

PAGE 205

1844.4 Conclusion This is the first report on the creation and use of a sol-gel TiO2-SiO2-TESP-PEO coating in solid-phase microe xtraction to extract various classes of polar compounds. Solgel TiO2-SiO2-TESP-PEO-coated microextraction cap illaries possess excellent extraction capability for highly polar analytes. Direct ch emical bonding of the coating to capillary inner walls provides these coatings with excel lent solvent resistan ce, and makes sol-gel TiO2-SiO2-TESP-PEO-coated capillaries very much suitable for on-line sample preconcentration in CME-HPLC analys is. The newly developed sol-gel TiO2-SiO2TESP-PEO coating was effectively used for the extraction of different classes of analytes with good extraction sensitivity, and run-to-ru n repeatability in CME-HPLC and CMEGC analyses. Low ng/mL level detection lim its were achieved for aldehydes, aniline derivatives, and substituted phenols in CMEHPLC-UV, and free fatty acids in CMEGC-FID analyses using the ne wly constructed sol-gel TiO2-SiO2-TESP-PEO-coated microextraction capillaries. Through optimization of experimental conditions for sol-gel coating procedure and the capillary microext raction process it should be possible to further enhance the extraction sensitivity. 4.5 References for Chapter Four [1] L.G. Blomberg, J. Microcol. Sep. 2 (1990) 62. [2] S.L. Chong, D. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [3] L. Pan, M. Adams, J. Pawliszyn, Anal. Chem. 67 (1995) 4396. [4] A.L. Lopes, F. Augusto, J. Chromatogr. A 1056 (2004) 13.

PAGE 206

185 [5] V. Mani, Applications of Solid-Phase Microextraction, Royal Society of Chemistry, Cambridge, UK, 1999. [6] M. Abalos, J.M. Bayona, J. Pawliszyn, J. Chromatogr. A 873 (2000) 107. [7] B.J. Hall, J.S. Brodbelt, J. Chromatogr. A 777 (1997) 275. [8] O.E. Mills, A.J. Broome, ACS Symp. Ser. 705 (1998) 85. [9] L. Yun, Anal. Chim. Acta, 486 (2003) 63. [10] M. Liu, Z. Zeng, Y. Tian, Anal. Chim. Acta 540 (2005) 341. [11] C. Chafer-Pericas, P. Campins-Falco, R. Herraez-Hernandez, Talanta, 69 (2006) 716. [12] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [13] Z. Wang, C. Xiao, C. Wu, H. Ha n, J. Chromatogr. A 893 (2000) 157. [14] S. Bigham, J. Medlar, A. Kabir, C. Sh ende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [15] C. Shende, A. Kabir, E. Townsend, A. Malik, Anal. Chem. 75 (2003) 3518. [16] R.G.D.C. Silva, F. Augusto, J. Chromatogr. A 1072 (2005) 7. [17] S. Sato, T. Murakata, T. Suzuki, T. Ohgawara, J. Mater. Sci. 25 (1990) 4880. [18] J.D. Cox, M.S. Curry, S.K. Skirboll, P. L. Gourley, D.Y. Sa saki, Biomaterials, 23 (2002) 929. [19] T.T. Razunguzwa, J. Lenke, A.T. Timperman, Lab Chip, 5 (2005) 851. [20] J.D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3. [21] J.D. Hayes, Ph.D. Dissertation, Univer sity of South Florida, USA, 2002. [22] K.R. Alhooshani, Sol-Gel Zirconiaand Titania-Based Surface-Bonded Hybrid Organic-Inorganic Coatings for Sample Preconcentration and Analysis via Capillary Microextraction in Hyphena tion with Gas Chromatography (CMEGC). Ph.D. dissertation, Department of Chemistry, University of South Florida, Tampa, FL 2005.

PAGE 207

186 [23] D.X. Wang, S.L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [24] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [25] T.-Y. Kim, K. Alhooshani, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165. [26] K. Alhooshani, T.-Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [27] S. Sakka, K. Kamiya, J. Non-Cryst. Solids 42 (1980) 403. [28] D.C. Bradley, R.C. Mehrotra, D.P. Ga ur, Metal Alkoxides, Academic Press, London, 1978. [29] C.J. Brinker, G.W. Scherer, Sol-Gel Science: The Physics and Chemistry of Sol-Gel Processing, Academic Press, New York, NY, 1990. [30] J. Livage, M. Henry, C. Sanchez, Prog. Solid St. Chem. 18 (1988) 259. [31] M.M. Haridas, S. Datta, J.R. Be llare, Ceram. Int. 25 (1999) 601. [32] N.T. McDevitt, W.L. Baun, Spectrochim. Acta 20 (1964) 799. [33] L. Tellez, J. Rubio, F. Rubio, E. Morales, J. L. Oteo, J. Mater. Sci. 38 (2003) 1773. [34] J. Wen, J.E. Mark, J. Appl. Polym. Sci. 58 (1995) 1135. [35] Q. Chen, N. Miyata, T. Kokubo, T. Nakamu ra, J. Biomed. Mate r. Res. 51 (2000) 605. [36] M.R. Boccuti, K.M. Rao, A. Zecchina, G. Leofanti, G. Petrini, Stud. Surf. Sci. Catal. 48 (1989) 133. [37] D.C.M. Dutoit, M. Shneider, A. Baiker, J. Catal. 153 (1995) 165. [38] C.A. Muller, M. Maciejewski, T. Mallat A. Baiker, J. Catal. 184 (1999) 280. [39] M.A. Holland, D.M Pickup, G. Mountjoy, E.S.C. Tsang, W. Wallidge, R.J. Newport, M.E. Smith, J. Mater. Chem. 10 (2000) 2495. [40] J.M. Fraile, J.I. Garcia, J.A. Mayor al, E. Vispe, J. Catal. 233 (2005) 90. [41] D.L. Pavia, G.M. Lampman, G.S. Kriz Introduction to Spectroscopy, Saunders H.B.J, Orlando, FL, 1979.

PAGE 208

187 [42] H.L. Cheng, K.F. Lin, J. Polym. Res. 6 (1999) 123. [43] H. Yamada, I. Somiya, Ozone Sci. Eng. 11 (1989) 125. [44] W.H. Glaze, M. Koga, D. Cancilla, Environ. Sci. Technol. 23 (1989) 838. [45] Z. Li, L.K. Jacobus, W.P. Wuelfing, M. Golden, G.P. Martin, R.A. Reed, J. Chromatogr. A 1104 (2006) 1. [46] W. Li, I. Kubo, Bioorg. Med. Chem. 12 (2004) 701. [47] J. Pissarra, S. Lourenco, A. M. Gonzalez-Paramas, N. Ma teus, C. Santos-Buelga, V.D. Freitas, Anal. Chim. Acta 513 (2004) 215. [48] M.S. Reish, Chem. Eng. News 66 (1988) 7. [49] G. Sabbioni, H.G. Neumann, Carcinogenesis 11 (1990) 111. [50] J. Szadowski, J. Dyes Pigments 14 (1990) 217. [51] G. Birner, H.G. Neumann, Ar ch. Toxicol. 62 (1988) 110. [52] P.D. Santo, G. Moneti, M. Salvadori, C. Saltutti, A.D. Rosa, P. Dolara, Cancer Lett. 60 (1991) 245. [53] ACGIH, Threshold Limit Values and Biological Exposure Indices, ACGIHPress, Cincinnati, OH, 1999. [54] L.H. Keith, W.A. Telliard, Envi ron. Sci. Technol. 13 (1979) 416. [55] V.H. Kitunen, R.J. Valo, M.S. Salkainija -Salonen, Environ. Sci. Technol. 21 (1987) 96. [56] D. Puig, D. Barcelo, Trends Anal. Chem. 15 (1996) 362. [57] M. Llompart, M. Lourido, P. Landin, C. Garcia-Jares, R. Cela, J. Chromatogr. A 963 (2002] 137. [58] M.R. Lee, Y.C. Yeh, W.S. Hsiang, B.H. Hwang, J. Chromatogr. A 806 (1998) 317. [59] E. Fogelqvist, B. Josefsson, C. Roos, J. High Resolut. Chromatogr. Chromatogr. Commun. 3 (1980) 568.

PAGE 209

188 APPENDICES

PAGE 210

189 Appendix A

PAGE 211

190 Appendix A (Continued)

PAGE 212

191 Appendix A (Continued)

PAGE 213

192 Appendix A (Continued)

PAGE 214

193 Appendix A (Continued)

PAGE 215

194 Appendix A (Continued)

PAGE 216

195 Appendix A (Continued)

PAGE 217

196 Appendix A (Continued)

PAGE 218

197 Appendix A (Continued)

PAGE 219

198 Appendix A (Continued)

PAGE 220

199 Appendix B

PAGE 221

200 Appendix B (Continued)

PAGE 222

201 Appendix B (Continued)

PAGE 223

202 Appendix B (Continued)

PAGE 224

203 Appendix B (Continued)

PAGE 225

204 Appendix B (Continued)

PAGE 226

205 Appendix B (Continued)

PAGE 227

206 Appendix B (Continued)

PAGE 228

207 Appendix B (Continued)

PAGE 229

208 Appendix B (Continued)

PAGE 230

209 Appendix B (Continued)

PAGE 231

210 Appendix B (Continued)

PAGE 232

211 Appendix B (Continued)

PAGE 233

212 Appendix B (Continued)

PAGE 234

213 Appendix C

PAGE 235

ABOUT THE AUTHOR Tae-Young Kim was born in Seoul, South Korea. He received a bachelor of health science degree in Me dical Technology from Yonsei University in 1992. He transferred to the University of Minnesota for dual bachelor degrees in chemistry and microbiology. During his senior year, he c onducted undergraduate research on zirconia columns for one year with a graduate student of Dr. Peter Carr, which motivated him to start his career in analytic al chemistry. He joined S UNY; college at Buffalo for a master’s degree, and did research in organi c synthesis under the direction of Dr. Subodh Kumar. However, he decided to move to th e Department of Chemistry, University of South Florida, to pursue a doctorate and to co ntinue his career in analytical chemistry. He joined Dr. Abdul Malik’s research group in 2001 and worked on developing sol-gel titania-based organic-inor ganic hybrid coatings for on-line CME-HPLC. He has 2 publications in international journa ls and 1 US patent application.