USF Libraries
USF Digital Collections

Surface-bonded sol-gel sorbents for on-line hyphenation of capillary microextraction with high-performance liquid chroma...

MISSING IMAGE

Material Information

Title:
Surface-bonded sol-gel sorbents for on-line hyphenation of capillary microextraction with high-performance liquid chromatography
Physical Description:
Book
Language:
English
Creator:
Segro, Scott
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Sol-Gel
Silica
Titania
Germania
CME
SPME
In-Tube SPME
High-Performance Liquid Chromatography
Solvent Resistance
PH Stability
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: High-performance liquid chromatography (HPLC) is the most widely used analysis technique. However, its sensitivity is limited. Sample preconcentration methods, such as fiber-based solid-phase microextraction (SPME) and in-tube SPME (capillary microextraction) offer improved detection limits. It is, however, difficult to couple fiber SPME on-line with HPLC due to the need for complicated desorption devices. Such coupling is further complicated due to the limited solvent stability of the extracting phase both in the fiber and in-tube formats of SPME. In this research, surface-bonded sol-gel sorbents were developed to provide the solvent stability required for effective on-line hyphenation of capillary microextraction (CME) with HPLC. These sol-gel sorbents were prepared using (1) silica-based, (2) titania-based, and (3) germania-based sol-gel precursors. Sol-gel reactions were performed within fused silica capillaries to create a number of organic-inorganic hybrid sorbents in the form of surface-bonded coatings: (1) alkyl (methyl, octyl, octadecyl), (2) polydimethyldiphenylsiloxane, (3) titania poly(tetrahydrofuran), and (4) germania tri-block polymer. The sol-gel coated microextraction capillaries were capable of efficiently extracting a wide variety of analytes, including polycyclic aromatic hydrocarbons, ketones, aldehydes, aromatic compounds, amines, alcohols, and phenols with ng/L to pg/L detection limits. The sol-gel methyl coating demonstrated a counterintuitive ability to extract polar analytes. Sol-gel polydimethyldiphenylsiloxane coatings were found to be resistant to high temperature solvent exposure (150°C and 200°C), making them suitable for use in high-temperature liquid phase separations. To better understand how extraction takes place, effects of alkyl chain length and sol-gel precursor concentration were evaluated in the study on sol-gel alkyl coatings. The sol-gel titania poly(tetrahydrofuran) coating was also capable of extracting underivatized aromatic acids and polypeptides at pHs near their respective isolectric points. The sol-gel titania poly(tetrahydrofuran) coatings and the sol-gel germania tri-block polymer coatings demonstrated impressive resistance to extreme pH conditions, surviving prolonged exposure to 1.0 M HCl (pH ≈ 0.0) and 1.0 M NaOH (pH ≈ 14.0) with virtually no change in extraction behavior. Sol-gel germania tri-block polymer coatings were also stable under high temperature solvent conditions (200°C). In addition, for the first time, the analyte distribution constants between a sol-gel germania coating and the aqueous samples (Kcs) were determined.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2010.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Scott Segro.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains X 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:
usfldc doi - E14-SFE0003321
usfldc handle - e14.3321
System ID:
SFS0027637:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Surface-Bonded Sol-Gel Sorbents for On-Line Hyphenation of Capillary Microextraction with High-Performance Liquid Chromatography by Scott S. Segro 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. Milton D. Johnston, Ph.D. Roman Manetsch, Ph.D. Xiao Li, Ph.D. Date of Approval: March 24, 2010 Keywords: Sol-Gel, Silica, Titania, Germania, CME, SPME, In-Tube SPME, HighPerformance Liquid Chromatography, Solvent Resistance, pH Stability Copyright 2010, Scott S. Segro

PAGE 2

DEDICATION To my loving family, my parents Johnny Segro and Carol Segro, my sister Shirley Wojciak, my brother-in-law Joe Wojciak, my niece Mackenzie Wojciak, my nephew Anthony Wojciak, and my grandmothers Mary Segro and Helen Hall.

PAGE 3

ACKNOWLEDGEMENTS I would like to express my gratitude to my major prof essor, Dr. Abdul Malik, for his supervision, patience and encouragement duri ng the past five years. I also want to thank my dissertation committee members Dr. Milton Johnston, Dr. Roman Manetsch, and Dr. Xiao Li for their valuable advice, thoughtful comments and support. Also, I would like to thank Dr. Scott Campbell for serving as the chair of my dissertation defense. I wish to thank my high school chemistry teacher, Frank Lock, for inspring me to major in chemistry. I also wish to thank a ll my former and current colleagues, Dr. TaeYoung Kim, Dr. Sameer M. Kulkarni, Dr. Anne M. Shearrow, Dr. Li Fang, Erica Turner, MinhPhuong Tran, Abdullah Alhendal, Chenlia ng Jiang, and Sheshanka Kesani. I would also like to thank Yaniel Cabezas, Judy Trip lett, Alberto Aguilera, and Kathryn Balance for their support and assistan ce. I want to thank God fo r the giving me the opportunity, ability, patience, and encouragem ent to complete this resear ch and disserta tion. Finally, I want to extend thanks to the organizati ons which financially supported my doctoral education, including the Genshaft Family Fe llowship, the USF Diversity Fellowship, the Alice F. Smith Scholarship, the Cora Str oh Scholarship, the Florida Swimming Pool Association, the Guy and Gloria Muto Foundation, the Gulf Coast Community Foundation of Venice, the Richard Bowman Jr Memorial Scholarship, and the Charles and Mrs. Gray Memorial Scholarship.

PAGE 4

i TABLE OF CONTENTS LIST OF TABLES ............................................................................................................. xi LIST OF FIGURES ......................................................................................................... xiv LIST OF SCHEMES...................................................................................................... xxiii LIST OF ABBREVIATIONS ........................................................................................ xxiv LIST OF SYMBOLS ..................................................................................................... xxvi ABSTRACT .................................................................................................................. xxv ii CHAPTER ONE: TRADITIONAL AND SOLVENT-FREE SAMPLE PREPARATION………. ...............................................................................................1 1.1 Introduction ........................................................................................................1 1.2 Traditional Sample Preparation Methods ..........................................................2 1.2.1 Liquid-Liquid Extraction ....................................................................2 1.2.2 Soxhlet Extraction ...............................................................................3 1.3 Sample preparation methods with reduced solvent consumption ......................4 1.3.1 Solid Phase Extraction ........................................................................4 1.3.2 Supercritical Fluid Extraction .............................................................5 1.4 Solvent-Free Sample Preparation ......................................................................6 1.4.1 Gas-Phase Extraction ..........................................................................6 1.4.2 Membrane Extraction........................................................................10 1.4.3 Solid Phase Microextraction (SPME) ...............................................11

PAGE 5

ii 1.4.4 Further Development in SPME .........................................................23 1.4.5 Solid Phase Dynamic Extraction ......................................................24 1.4.6 SPME with Rotation of the Microfiber .............................................26 1.4.7 Microwave Assisted Headspace SPME ............................................28 1.4.8 Matrix SPME ....................................................................................28 1.4.9 Membrane Protected SPME ..............................................................29 1.4.10 Stir Bar Sorptive Extraction (SBSE)...............................................29 1.4.11 Thin Film Microextraction ..............................................................31 1.4.12 In-Tube SPME or Capillary Microextraction (CME) .....................32 1.5 Applications of Micr oextraction Techniques...................................................35 1.6 Drawbacks of SPME ........................................................................................36 1.7 References for Chapter One .............................................................................40 CHAPTER TWO: SOL-GEL COAT INGS AND MONOLITHS IN ANALYTICAL SAMPLE PREPARATION ..............................................................45 2.1 Introduction ......................................................................................................45 2.2 Sol-Gel Chemistry ...........................................................................................46 2.3 Physico-Chemical Characteriza tion of Sol-Gel Materials ..............................47 2.4 Classification of Sol-Gel Coatings in Solid Phase Microextraction ...............50 2.4.1 Fiber Preparation and Pr etreatment Procedures ................................50 2.4.2 Sol-Gel Coating Procedures for SPME.............................................51 2.4.3 Sol-Gel Polysiloxane-Based Monofunctional Coatings for SPME ...............................................................................................53

PAGE 6

iii 2.4.3.1 Silica-Based Sol-Gel Polydimethylsiloxane Coatings for SPME ............................................................53 2.4.3.2 Non-Silica-Based Sol-Gel Polydimethylsiloxane Coatings for SPME ............................................................57 2.4.4 Polysiloxane-Based Multifunctional Sol-Gel Fiber Coatings ...........58 2.4.4.1 PDMS/Poly(vinylalcohol) Sol-Gel Fiber Coatings ...........58 2.4.4.2 Polymethylphenylvinyls iloxane Sol-Gel Fiber Coatings .............................................................................59 2.4.4.3 C11 PDMS Sol-Gel Fiber Coating ....................................61 2.4.4.4 Polymethylphenylvinyls iloxane Sol-Gel Fiber Coating ...............................................................................61 2.4.4.5 Sol-Gel Anilinemethyltriethoxysilane/PDMS Coating for SPME ..............................................................61 2.4.4.6 Sol-Gel Silicone Polydi vinylbenzene Copolymeric Coating for SPME ..............................................................62 2.4.4.7 Sol-Gel Aminopropylsilica/PDMS Coating for SPME .................................................................................62 2.4.4.8 Sol-Gel 3-Aminopropyl triethoxysilane/PDMS Coating for SPME ..............................................................63 2.4.4.9 Sol-Gel Amino-Functiona lized PDMS Coating for SPME .................................................................................63 2.4.5 Sol-Gel Coatings with Cavity Ligands ................................................63 2.4.5.1 Crown Ether Sol-Gel Fiber Coatings .................................63

PAGE 7

iv 2.4.5.2 Sol-Gel Calixarene Coatings for SPME ............................67 2.4.5.3 Sol-Gel Cyclodextrin Coating for SPME ..........................71 2.4.5.4 Sol-Gel Hydroxyfullerene Coating for SPME ...................72 2.4.5.5 Sol-Gel Quinoxaline-Bridged Cavitand Coating for SPME .................................................................................73 2.4.6 Non-Polysiloxane Sol-Gel Coatings .................................................73 2.4.6.1 Sol-Gel Polyethylene Glycol Coatings for SPME .............73 2.4.6.2 Sol-Gel Acrylate Coatings for SPME ................................74 2.4.6.3 Other Silica-Based Non-Polysiloxane Sol-Gel SPME Coatings ..................................................................77 2.4.6.4 Other Non-Silica-Based Non-Polysiloxane Sol-Gel Fiber Coatings ......................................................80 2.5 Sol-Gel Materials in In-Tube SPME (C apillary Microext raction (CME)) ......83 2.5.1 Introduction .......................................................................................83 2.5.2 Capillary Pretreatment Procedures ...................................................83 2.5.3 Silica-Based Sol-Gel Coatings in CME ............................................84 2.5.3.1 Polysiloxane-Based Monofunctional Sol-Gel Capillary Coatings .............................................................84 2.5.3.2 Polysiloxane-Based Multifunctional Sol-Gel Coating in CME ...............................................................................88 2.5.3.3 Sol-Gel Coating with Cavity Ligands for CME ................90 2.5.3.4 Electrically Charged Sol-Gel Coatings for CME ...............93 2.5.3.5 Non-Polysiloxane-Based Sol-Gel Coatings for CME ........95

PAGE 8

v 2.5.4 Non-Silica-Based Sol-Gel Coatings in CME ..................................100 2.5.4.1 Titania-Based Sol-Gel Coatings in CME .........................100 2.5.4.2 Zirconia-Based Sol-Gel Coatings in CME ......................101 2.5.4.3 Germania-Based Sol-Gel Coatings in CME ....................103 2.6 Sol-Gel Materials in Stir Bar Sorptive Extraction .........................................105 2.6.1 Sol-Gel PDMS Coated Stir Bars .....................................................105 2.6.2 Sol-Gel Multifunctional Stir Bar Coatings .....................................107 2.7 Sol-Gel Monoliths in Analytical Microextraction .........................................108 2.8 Conclusion .....................................................................................................119 2.9 References for Chapter Two ..........................................................................119 CHAPTER THREE: SOL-GEL MET HYL COATING IN CAPILLARY MICROEXTRACTION HYPHENA TED ON-LINE WITH HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY. COUNTERINTUITIVE EXTRACTION BEHAVIOR FOR POLAR ANALYTES ................................................................................................127 3.1 Introduction ....................................................................................................127 3.2 Experimental ..................................................................................................131 3.2.1 Equipment .......................................................................................131 3.2.2 Chemicals and Materials .................................................................131 3.2.3 Pretreatment of Fused Silica Capillary ...........................................132 3.2.4 Preparation of the Sol-Gel Me thyl Coated Microextraction Capillary ..........................................................................................132 3.2.5 On-Line CME-HPLC Analysis .......................................................133

PAGE 9

vi 3.3 Results and Discussion ..................................................................................136 3.4 Conclusion .....................................................................................................158 3.5 References for Chapter Three ........................................................................158 CHAPTER FOUR: SOL-GEL COATINGS WITH COVALENTLY ATTACHED METHYL-, OCTYL-, AND OCTADE CYLLIGANDS FOR CAPILLARY MICROEXTRACTION. E FFECTS OF ALKYL CHAIN LENGTH AND PRECURSOR CONCENTRATION ON EXTRACTION BEHAVIOR ..................162 4.1 Introduction ....................................................................................................162 4.2 Experimental ..................................................................................................166 4.2.1 Equipment .......................................................................................166 4.2.2 Chemicals and Materials .................................................................166 4.2.3 Pretreatment of Fused Silica Capillary ...........................................167 4.2.4 Preparation of the Sol-Gel Capillaries ............................................167 4.2.4.1 Preparation of Sol-Gel Coated Microextraction Capillaries using Constant Mola r Concentration of the Precursors .............................................................167 4.2.4.2 Preparation of Sol-Gel Coated Microextraction Capillaries using Varied Molar Concentrations of Sol-Gel Precurs or and Constant Solution Volume ..........170 4.2.5 On-Line CME-HPLC Analysis using Sol-gel C8 and C18-bonded Microextraction Capillaries ...........................................173 4.3 Results and Discussion ..................................................................................175

PAGE 10

vii 4.3.1 Microextraction Capillaries Prepared using Sol Solution with Constant Mo lar Concentration of the Precursor .....................175 4.3.2 Microextraction Capillaries Prepared using Sol Solution with Vari ed Molar Concentration of C8TMS Precursor .................192 4.4 Conclusion .....................................................................................................203 4.5 References for Chapter Four ..........................................................................204 CHAPTER FIVE: SOLVEN T-RESISTANT SOL-GEL POLYDIMETHYLDIPHENYLSILOXA NE COATING FOR ON-LINE HYPHENATION OF CAPILLARY MI CROEXTRACTION WITH HIGHPERFORMANCE LIQUID CHROMATOGRAPHY ..............................................209 5.1 Introduction ....................................................................................................209 5.2 Experimental ..................................................................................................212 5.2.1 Equipment .......................................................................................212 5.2.2 Chemicals and Materials .................................................................213 5.2.3 Pretreatment of Fused Silica Capillary ...........................................213 5.2.4 Preparation of the Sol-Gel Polydimethyldiphenylsiloxane Coated Microextraction Capillary ...................................................214 5.2.5 On-Line CME-HPLC Analysis .......................................................215 5.3 Results and Discussion ..................................................................................218 5.4 Conclusion .....................................................................................................243 5.5 References for Chapter Five ..........................................................................243

PAGE 11

viii CHAPTER SIX: ULTRA-HIGH-STABILITY, pH-RESISTANT SOL-GEL TITANIA POLY(TETRAHYDROFURAN) COATING FOR CAPILLARY MICROEXTRACTION ON-LINE C OUPLED TO HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY .............................................................................247 6.1 Introduction ....................................................................................................247 6.2 Experimental ..................................................................................................250 6.2.1 Equipment .......................................................................................250 6.2.2 Chemicals and Materials .................................................................251 6.2.3 Preparation of the Sol-Gel Titania Poly(tetrahydrofuran) Coated Capillary ..............................................................................252 6.2.3.1 Fused Silica Capillary Pretreatment.................................252 6.2.3.2 Preparation of the Sol-Gel Titania Poly(tetrahydrofuran) Sol Solution ..................................252 6.2.3.3 Sol-Gel Coating of the Fused Silica Capillary .................253 6.2.3.4 Capillary Conditioning ....................................................254 6.2.4 On-Line CME-HPLC Analysis .......................................................254 6.3 Results and Discussion ..................................................................................257 6.4 Conclusion .....................................................................................................291 6.5 References for Chapter Six ............................................................................292 CHAPTER SEVEN: SOL-GEL GERMANIA TRI-BLOCK POLYMER COATINGS OF EXCEPTIONAL pH STABILITY IN CAPILLARY MICROEXTRACTION ON-LI NE COUPLED TO HIGH-PERFORMANCE LI QUID CHROMATOGRAPHY....................................296

PAGE 12

ix 7.1 Introduction ....................................................................................................296 7.2 Experimental ..................................................................................................298 7.2.1 Equipment .......................................................................................298 7.2.2 Chemicals and Materials .................................................................299 7.2.3 Surface Cleaning and Hydrothermal Pretreatment of Fused Silica Capillary ................................................................................300 7.2.4 Preparation of the Sol Solution .......................................................300 7.2.5 Coating and Conditioning of the Capillary .....................................301 7.2.6 Preparation of Aqueous Samples for CME-HPLC analysis ...........302 7.2.7 CME-HPLC Analysis of Aqueous Samples ...................................302 7.2.8 Conversion of Peak Areas to Amounts Extracted (ng) ...................306 7.3 Results and Discussion ..................................................................................306 7.4 Conclusion .....................................................................................................339 7.5 References for Chapter Seven ........................................................................340 APPENDICES .................................................................................................................344 Appendix A: Sol-Gel Methyl Coatin g in Capillary Microextraction Hyphenated On-Line with High-Performance Liquid Chroma tography. Counterintuitive Ex traction Behavior for Polar Analytes ...................................................................................345 Appendix B: Solvent-Resistant SolGel Polydimethyldiphenylsiloxane Coating for On-Line Hyphenation of Capillary Mi croextraction with Hi gh-Performance Liquid Chromatography ...............................................................................355

PAGE 13

x Appendix C: Sol-Gel Coatings with Cova lently Attached Methyl-, Octyl-, and Octade cylLigands for Capillary Microextraction. Effects of Al kyl Chain Length and Precursor Concentration on Extraction Behavior ..........................................................................365 Appendix D: Ultra-High-Stability, pH-Resistant Sol-Gel Titania Poly(tetrahydrofuran) Coating for Capillary Microextraction On-Line Coupled to High-Performance Liquid Chromatography ...............................................................................374 ABOUT THE AUTHOR ................................................................................... END PAGE

PAGE 14

xi LIST OF TABLES Table 3.1 HPLC peak area repeatability and detection limit data for PAHs, ketones, phenols, alcohols, and amines in CME-HPLC using a sol-gel methyl coated mi croextraction capillary ..................................................142 Table 3.2 Peak area comparison for m -toluidine, 3,4-dimethylphenol, benzhydrol, 4’phenylacetophenone, a nd naphthalene on the sol-gel methyl coated capillary deactivat ed with PMHS, the undeactivated sol-gel methyl coated capill ary, and an uncoated capillary .................................155 Table 4.1 Compositions of the sol solutions used to prepare the sol-gel alkyl coated capillaries with constant molar concentration of the sol-gel precursor ..............................................................................................................169 Table 4.2 Compositions of the sol soluti ons used to prepare the sol-gel octyl coated capillaries with vari ed molar concentration of sol-gel precursor ..............................................................................................................172 Table 4.3 CME-HPLC-UV peak area repeatab ility and detection limit data for PAHs, ketones, phenols, and amines fo r the sol-gel octadecyl, octyl, and methyl coated microextraction cap illaries (0.514 M sol-gel precursor concentration) ......................................................................................................186 Table 4.4 Capillary-to-capillary CME-HPLC -UV peak area repeatability for the sol-gel octadecyl, octyl, and methyl coated microextraction capillaries .............188

PAGE 15

xii Table 4.5 HPLC peak area comparison of amines, phenols, ketones, and PAHs using sol-gel octyl coated microext raction capillaries with varied sol-gel precursor concentrations ..........................................................................197 Table 5.1 HPLC peak area repeatability and detection limit data for PAHs, ketones, nonpolar aromatic compoun ds, aromatic amines, and aldehydes in CME-HPLC using a sol-gel pol ydimethyldiphenylsiloxane coated microextraction capillary .....................................................................................226 Table 5.2 Capillary-to-capillary peak ar ea reproducibility in CME-HPLC for the sol-gel PDMDPS coated capillaries ...............................................................235 Table 5.3 High-temperature solvent stab ility of the sol-gel PDMDPS coated capillary................................................................................................................242 Table 6.1 HPLC peak area repeatability a nd detection limit data for phenols, alcohols, and amines in CME-HPLC using a sol-gel titania poly-THF coated microextraction capillary ..........................................................................265 Table 6.2 HPLC peak area repeatability and detection limit data for acids, ketones, and PAHs in CME-HPLC using a sol-gel titania poly-THF coated microextraction capillary ..........................................................................271 Table 6.3 Capillary-to-capillary peak ar ea reproducibility in CME-HPLC for the sol-gel titania poly-THF coated capillaries ....................................................275 Table 6.4 HPLC peak area comparison of acids, amines, phenols, alcohols, ketones, and PAHs before and af ter exposing the sol-gel titania poly-THF coated capil lary to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 18 h ..................................................................................281

PAGE 16

xiii Table 6.5 HPLC peak area comparison of acids, amines, phenols, alcohols, ketones, and PAHs before and af ter exposing the sol-gel silica poly-THF coated capil lary to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 18 h ..................................................................................286 Table 7.1 HPLC peak area repeatability and detection limit data and distribution constant (Kcs) values for PAHs, ketones, amines, alchols, and phenols in CME-HPLC using a sol-gel ge rmania tri-block polymer coated microextraction capillary .....................................................................................316 Table 7.2 HPLC peak area comparison of PAHs, ketones, amines, alcohols, and phenols before and after exposin g the sol-gel germania tri-block polymer coated capillary to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 20 h and 5 days ...........................................................................324 Table 7.3 HPLC peak area comparison of PAHs, Ketones, Amines, Alcohols, and Phenols Before and After Exposi ng the Sol-Gel Germania Tri-Block PEO-PPO-PEO Coated Microext raction Capillary to ACN/H20 (50/50, v/v) for 2 h at 200C .......................................................................................................329 Table 7.4 Capillary-to-capillary peak ar ea reproducibility in CME-HPLC for the sol-gel germania tri-block polymer coated capillaries ...................................332

PAGE 17

xiv LIST OF FIGURES Figure 1.1 Sealed vial used for the st atic headspace extr action technique ..........................8 Figure 1.2 Schematic of an SPME extraction device ........................................................13 Figure 1.3 SPME field sampling device ............................................................................16 Figure 1.4 Comparison of direct SPME and headspace SPME .........................................21 Figure 1.5 Solid phase dynamic extraction. .......................................................................25 Figure 1.6 Different sorption methods used for SPME .....................................................27 Figure 1.7 The stir bar sorptive extraction process ............................................................30 Figure 1.8 Comparison of in-tube SPME with fiber SPME ..............................................34 Figure 1.9 Schematic of a desorption de vice used to couple fiber SPME with HPLC .....................................................................................................................38 Figure 2.1 Example of an extraction profile ......................................................................49 Figure 2.2 Sol-gel fiber pretreatme nt and coating procedure. ...........................................52 Figure 2.3 Schematic diagram of SPME and back-extraction procedures ........................70 Figure 2.4 A cross-sectional scanning elec tron microscopic image of a zirconia hollow fiber ............................................................................................................81 Figure 2.5 A schematic of a gravity-fed sample dispensing unit used for capillary microextraction .......................................................................................87 Figure 2.6 A schematic diagram of the experimental set-up used for in-tube SPME-HPLC analysis ............................................................................................92

PAGE 18

xv Figure 2.7 The structure of the surfac e-bonded sol-gel dendrimer coating ......................97 Figure 2.8 Excellent stability of so l-gel germania-PDMS coating under highly acidic conditions demonstrat ed through CME-GC analysis of aldehydes using a germania-PDMS co ated microextraction capillary before (A) and after (B) continuous ly rinsing the capillary with a 0.05 M HCl (pH 1.3) solution for 24 h. ...............................................................104 Figure 2.9 Extraction of a sample contai ning a PAH and an alcohol on (A) a sol-gel ODS monolith and (B) a sol-gel ODS coated capillary ...........................110 Figure 2.10 Scanning electron microgra phs of the 320 m i.d. sol-gel monolithic precolumn ..........................................................................................113 Figure 3.1 Experimental se tup used to perform the sol-gel CME-HPLC experiments ..........................................................................................................135 Figure 3.2 Extraction profiles of anthracene (5 x 104 ng/L), 1-[1,1’-biphenyl]-4-ylethanone (5 x 104 ng/L), 3,5-dimethylphenol (5 x 105 ng/L), benzhydrol (5 x 105 ng/L), and diphenylamine (5 x 104 ng/L). ....................................................................140 Figure 3.3 A chromatogram representing online CME-HPLC analysis of PAHs using a sol-gel methyl co ated microextraction capillary .....................................144 Figure 3.4 A chromatogram representing online CME-HPLC of ketones using a sol-gel methyl coated capillary ............................................................................146 Figure 3.5 A chromatogram representi ng on-line CME-HPLC analysis of phenols using a sol-gel methyl coated capillary ..................................................148

PAGE 19

xvi Figure 3.6 A chromatogram representi ng on-line CME-HPLC analysis of alcohols using a sol-gel methyl coated capillary .................................................150 Figure 3.7 A chromatogram representing online CME-HPLC analysis of amines using a sol-gel me thyl coated capillary ................................................................152 Figure 3.8 Chromatograms representing on-li ne CME-HPLC anal ysis of polar and nonpolar compounds using a so l-gel methyl coated capillary deactivated with PMHS, a sol-ge l methyl coated cap illary prepared without a deactivator, and an uncoated capillary .................................................154 Figure 3.9 Durability of sol-gel methyl coatings .............................................................157 Figure 4.1 The experimental setup used to perform the sol-gel CME-HPLC experiments .........................................................................................................174 Figure 4.2 Extraction profile of 5 x 104 ng/L naphthalene on sol-gel octadecyl, octyl, and methyl coated capillaries ...................................................179 Figure 4.3 Extraction profile of 5 x 104 ng/L 1-[1,1’-biphenyl]-4-yl-ethanone on sol-gel octadecyl, octyl, and methyl coated capillaries ...................................180 Figure 4.4 Extraction profile of 2 x 105 ng/L 3,4-dimethylphenol on sol-gel octadecyl, octyl, and methyl coated capillaries ...................................................181 Figure 4.5 Chromatograms representingC ME-HPLC-UV analysis of amines, phenols, ketones, and PAHs usi ng sol-gel alkyl coated capillaries. ....................183 Figure 4.6 Capillary-to-capillary CMEHPLC-UV reproducibility of amines, phenols, ketones, and PAHs using three sol-gel octadecyl coated capillaries .............................................................................................................189

PAGE 20

xvii Figure 4.7 Capillary-to-capillary CMEHPLC-UV reproducibility of amines, phenols, ketones, and PAHs using thr ee sol-gel octyl coat ed capillaries ............190 Figure 4.8 Capillary-to-capillary CMEHPLC-UV reproducibility of amines, phenols, ketones, and PAHs using thr ee sol-gel methyl coated capillaries .........191 Figure 4.9 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, ketones, and PAHs usi ng sol-gel octyl coated capillaries prepared using 0.257 M concentration of C8TMS in the sol solution .................193 Figure 4.10 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, ketones, and PAHs usi ng sol-gel octyl coated capillaries prepared using 0.514 M concentration of C8TMS in the sol solution. ................194 Figure 4.11 A chromatograms representing CME-HPLC-UV analysis of amines, phenols, ketones, and PAHs usi ng sol-gel octyl coated capillaries prepared using 1.028 M concentration of C8TMS in the sol solution .................195 Figure 4.12 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, ketones, and PAHs using so l-gel octyl coated capillaries prepared using 1.542 M concentration of C8TMS in the sol solution ................................196 Figure 4.13 Scanning electron microscopic (SEM) image of the sol-gel octyl coated microextraction capillary prepared using 0.247 M concentration of C8TMS in the sol solution ...............................................................................200 Figure 4.14 Scanning electron microscopic (SEM) image of the sol-gel octyl coated microextraction capillary prepared using 0.514 M concentration of C8TMS in the sol solution ...............................................................................201

PAGE 21

xviii Figure 4.15 Scanning electron microscopic (SEM) image of the sol-gel octyl coated microextraction capillary prepared using 1.542 M concentration of C8TMS in the sol solution ...............................................................................202 Figure 5.1 Sol-gel CME-HPLC experimental setup ........................................................217 Figure 5.2 Extraction profiles of naphthalene (1 x 105 ng/L), biphenyl (2 x 104 ng/L) 4’phenylacetophenone (5 x 104 ng/L), and m -tolualdehyde (2 x 105 ng/L) on the sol-gel PDMDPS coated capillary ....................................................................................................224 Figure 5.3 A chromatogram representing online CME-HPLC analysis of PAHs using a sol-gel PDMDPS coat ed microextraction capillary ................................228 Figure 5.4 A chromatogram representi ng on-line CME-HPLC analysis of aromatic compounds using a solgel PDMDPS coated microextraction capillary................................................................................................................23 0 Figure 5.5 A chromatogram representi ng on-line CME-HPLC analysis of aldehydes and ketones usi ng a sol-gel PDMDPS coated microextraction capillary .....................................................................................233 Figure 5.6 Durability of sol-gel PDMDPS coated capillaries..........................................237 Figure 5.7 Chromatograms representing online CME-HPLC analysis of a mixture of polar and moderately polar compounds using a sol-gel PDMDPS coated microextra ction capillary, (4.7 A) before rinsing and heating, (4.7 B) after rinsing with 120 mL of 50 /50 ACN/water (v/v) at 150 C for 2 h, and (4.7 C) after ri nsing capillary 4.7 B with 120 mL of 50/50 ACN/water (v/v) at 200 C for 2 h ........................................................241

PAGE 22

xix Figure 6.1 Experimental setup used to carry out the CME-HPLC experiments using the sol-gel titania po ly-THF coated capillary .............................................256 Figure 6.2 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, ketone s and PAHs using a sol-gel titania coated capillary without poly-THF ......................................................................260 Figure 6.3 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, ketone s and PAHs using a sol-gel titania poly-THF coated capillary ...................................................................................261 Figure 6.4 Extraction profiles of m -toluidine, otoluic acid, 1-[1,1’-biphenyl]-4ylethanone, benzhydrol, 3,4-dimethyl phenol, and naphthalene for the sol-gel titania poly-THF coated capillary ............................................................263 Figure 6.5 A chromatogram representing CM E-HPLC-UV analysis of phenols and alcohols using a sol-gel tita nia poly-THF coated capillary. ................................266 Figure 6.6 A chromatogram representing CM E-HPLC-UV analysis of amines and ketones using a sol-gel tit ania poly-THF coated capillary...................................267 Figure 6.7 A chromatogram representing CM E-HPLC-UV analysis of acids using a sol-gel titania poly-THF coated capillary .........................................................270 Figure 6.8 A chromatogram representing CM E-HPLC-UV analysis of acids and PAHs using a sol-gel titania poly-THF coated capillary .....................................273 Figure 6.9 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, ketone s and PAHs using a sol-gel titania poly-THF coated capillary before exposure to acidic or basic conditions ...........277

PAGE 23

xx Figure 6.10 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, ketone s and PAHs using a sol-gel titania poly-THF coated capillary after 18 h exposure to 1.0 M NaOH. ........................278 Figure 6.11 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel titania poly-THF coated capillary af ter 18 h exposure to 1.0 M HCl ................................279 Figure 6.12 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohol s, ketones and PAHs using a sol-gel silica poly-THF coated capillary before exposure to acidic or basic conditions ...........283 Figure 6.13 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, keto nes and PAHs using a sol-gel silica poly-THF coated capillary after 18 h exposure to 1.0 M NaOH .........................284 Figure 6.14 A chromatogram representing CME-HPLC-UV analysis of acids, amines, phenols, alcohols, keto nes and PAHs using a sol-gel silica poly-THF coated capillary after 18 h exposure to 1.0 M HCl .............................285 Figure 6.15 CME-HPLC-UV analysis of pol y-arginine-tyrosine (4:1) using a sol-gel titania poly-THF coated capillary ............................................................289 Figure 6.16 A chromatogram represen ting CME-HPLC-UV analysis of poly-glutamic acid-tyrosine (1 :1) using a sol-gel titania poly-THF coated capillary ....................................................................................................290 Figure 7.1 Experimental setup used to carry out the gravity-fed CME-HPLC experiments using sol-gel germania tr i-block polymer coated capillaries ...........303

PAGE 24

xxi Figure 7.2 Experimental se tup used to carry out the higher flow rate CME-HPLC experiments using solgel germania tri-block polymer coated capillaries ..................................................................................................305 Figure 7.3 FTIR spectrum of germanium dioxide ...........................................................311 Figure 7.4 FTIR spectrum of poly(ethylen e oxide)-block-poly(propylene oxide)block-poly(ethylene oxide). .................................................................................312 Figure 7.5 FTIR spectrum of the sol-gel germania tri-block polymer coating. ...............313 Figure 7.6 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones, and PAHs using a sol-gel germania tri-block polymer coated capillary ......................................................................................317 Figure 7.7 Scanning electron image of the sol-gel germania tr i-block polymer coated microextraction capillary ..........................................................................320 Figure 7.8 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel germania tri-block polymer coated capillary af ter 20 h of exposure to 1.0 M HCl ...........................322 Figure 7.9 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel germania tri-block polymer coated capillary afte r 5 days of exposure to 1.0 M HCl ........................323 Figure 7.10 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel germania tri-block polymer coated capillary after 20 h of exposure to 1.0 M NaOH. .......................326

PAGE 25

xxii Figure 7.11 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel germania tri-block polymer coated capillary afte r 5 days of exposure to 1.0 M NaOH ....................327 Figure 7.12 A chromatogram representing CME-HPLC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel germania tri-block polymer coated capillary after 2 h exposure to ACN/H20 (50/50, v/v) at 200 C ...................................................................................................................33 0 Figure 7.13 Extraction profiles of m -toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans -chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for gravity-fed extraction (~0.2 mL/min flow rate). .....................................................................................334 Figure 7.14 Extraction profiles of m -toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans -chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for HPLC pump driven extraction (1.0 mL/min flow rate) ........................................................................337 Figure 7.15 Extraction profiles of m -toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans -chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for HPLC pump driven extraction (2.5 mL/min flow rate). .......................................................................338

PAGE 26

xxiii LIST OF SCHEMES Scheme 3.1 Chemical reactions involved in the formation of the sol-gel methyl coating ......................................................................................................138 Scheme 4.1 Chemical reactions involved in the formation of the sol-gel octadecyl, octyl, and methyl coatings ..................................................................177 Scheme 5.1 Hydrolysis of sol-ge l precursor (MTMS) followed by polycondensation of the hydrolyzed precursor and chemical bonding of the sol-gel active polymer (PDMDPS) to the evolving sol-gel network .............221 Scheme 5.2 Chemical anchoring of th e evolving sol-gel PDMDPS network to the inner walls of a fused silica capillary .........................................................222 Scheme 6.1 Chemical reactions involved in the formation of th e sol-gel titania poly-THF coated capillary ...................................................................................258 Scheme 7.1 Chemical reactions involved in the formation of the sol-gel germania tri-block polymer coated capillary .......................................................308 Scheme 7.2 Chemical anchoring of the evolving sol-gel germania tri-block Polymer network to the inner wa lls of a fused-silica capillary ............................309

PAGE 27

xxiv LIST OF ABBREVIATIONS ACN Acetonitrile AIBN Azobis(isobutyronitrile) APS 3-aminopropyltrimethoxysilane BMA Butyl Methacrylate BPA Bisphenol-A BTEX Benzene, Toluene, Ethyl-Benzene, Xylene BTX Benzene, Toluene, Xylene CE Capillary Electrophoresis CEC Capillary Electrochromatography CME Capillary Microextraction CW Carbowax C8TEOS n -octyltriethoxysilane C8TMS Octyltrimethoxysilane C18TEOS n -octadecyltriethoxysilane C18TMS Octadecyltrimethoxysilane DA Domoic Acid DAD Diode Array Detection DATEG Diallyltriethylene Glycol DM Dextromethorphan DP Dextrorphan DVB Divinylbenzene ECD Electron Capture Detector FID Flame Ionization Detector GC Gas Chromatography HMDS 1,1,1,3,3,3-Hexamethyldisilazane HPLC High-Performan ce Liquid Chromatography HS-SPME Headspace Solid Phase Microextraction ICP-MS Inductively Coupled Plasma Mass Spectrometry i.d. Internal Diameter KH-550 3-aminopropyltriethoxysilane KH-560 3-(2-cyclooxypropoxyl)propyltrimethoxysilane KH-570 Methacryloxypropyltrimethoxysilane LC Liquid Chromatography MAHs Monocyclic Aromatic Hydrocarbons MIP Molecularly Imprinted Polymer MIS Molecularly Imprinted Silica MPTMS Mercaptopropyltrimethoxysilane MS Mass Spectrometry

PAGE 28

xxv MTMS Methyltrimethoxysilane O.D. Outer Diameter ODS Octadecyl Silane OH-TSO Hydroxy-terminated Silicone Oil PAHs Polycyclic Aromatic Hydrocarbons PBPs Persistent and Bioaccumulative Pollutants PBS Phosphate-Buffered Saline PCBs Polychlorinated Biphenyls PDMS Polydimethylsiloxane PDMDPS Polydimethyldiphenylsiloxane PEEK Poly ether ether ketone PEG Polyethylene Glycol PEO-PPO-PEO Poly(ethylene oxide)block-poly(propylene oxide)-blockpoly(ethylene oxide) PMHS Polymethylhydroxiloxane PMPVS Poly(methylphenylvinylsiloxane) poly-THF Polytetrahydrofuran PTFE Polytetrafluoroethylene PTMOS Phenyltrimethoxysilane PVA Poly(vinyl alcohol) RSD Relative Standard Deviation SBSE Stir Bar Sorptive Extraction SEM Scanning Electron Microscope SFC Supercritical Fluid Chromatography SPE Solid Phase Extraction SPDE Solid Phase Dynamic Extraction SPME Solid Phase Microextraction TEOS Tetraethoxysilane TFA Trifluoroacetic Acid THF Tetrahydrofuran TMOS Tetramethoxysilane TMSPMA (Trimethoxysilyl)propyl Methacrylane TNBG Tetran -butoxygermane UV Ultraviolet VTEOS Vinyltriethoxysilane

PAGE 29

xxvi LIST OF SYMBOLS Kfs distribution constant of the solu te between the fiber coating and the sample n the number of moles of analyte th at are extracted by the fiber at equilibrium Vf the volume of the coa ting on the fiber in liters Vs the volume of the sample in liters Co the molar concentration of th e analyte in the original sample k retention factor te extraction time L length of the capillary holding the extracting phase U laminar flow rate of the fluid

PAGE 30

xxvii SURFACE-BONDED SOL-GEL SORBENTS FOR ON-LINE HYPHENATION OF CAPILLARY MICROEXTRACTION WITH HIGH-PERFORMANCE LIQUID CHROMATOGRAPY Scott S. Segro ABSTRACT High-performance liquid chromatography (HPLC) is the most widely used analysis technique. However, its sensitiv ity is limited. Sample preconcentration methods, such as fiber-based solid-phase microextraction (SPME) and in-tube SPME (capillary microextraction) offer improved de tection limits. It is, however, difficult to couple fiber SPME on-line with HPLC due to the need for complicated desorption devices. Such coupling is further complicated due to the limited solvent stability of the extracting phase both in the fibe r and in-tube formats of SPME. In this research, surfacebonded sol-gel sorbents were developed to provide the solvent stability required for effective on-line hyphenation of capillary micr oextraction (CME) with HPLC. These solgel sorbents were prepared using (1) silica-based, (2) t itania-based, and (3) germaniabased sol-gel precursors. Sol-gel reactions we re performed within fused silica capillaries to create a number of organic-inorganic hybr id sorbents in the form of surface-bonded coatings: (1) alkyl (methyl, octyl, octad ecyl), (2) polydimethyldiphenylsiloxane, (3) titania poly(tetrahydrofuran), and (4) german ia tri-block polymer. The sol-gel coated microextraction capillaries were capable of efficiently extracting a wide variety of

PAGE 31

xxviii analytes, including polycyclic aromatic hydr ocarbons, ketones, aldehydes, aromatic compounds, amines, alcohols, and phenols with ng/L to pg/L detection limits. The solgel methyl coating demonstrated a counterintui tive ability to extract polar analytes. Solgel polydimethyldiphenylsiloxane coatings were found to be resistant to high temperature solvent exposure (150C and 200C), making th em suitable for use in high-temperature liquid phase separations. To better understand how extractio n takes place, effects of alkyl chain length and sol-gel precursor concentration were evaluated in the study on solgel alkyl coatings. The sol-gel titania poly( tetrahydrofuran) coating was also capable of extracting underivatized aroma tic acids and polypeptides at pHs near their respective isolectric points. The sol-gel titania poly( tetrahydrofuran) coa tings and the sol-gel germania tri-block polymer coatings demonstr ated impressive resistance to extreme pH conditions, surviving prolonged exposure to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) with virtually no change in extrac tion behavior. Sol-gel germania tri-block polymer coatings were also stable under hi gh temperature solvent conditions (200C). In addition, for the first time, the analyte distri bution constants between a sol-gel germania coating and the aqueous samples (Kcs) were determined.

PAGE 32

1 CHAPTER ONE TRADITIONAL AND SOLVENTFREE SAMPLE PREPARATION 1.1 Introduction Sample preparation is the most time consum ing and tedious step in analysis. It is also a major source of inaccuracy and imprecision of an analytical method [1]. In traditional sample preparation methods, which include, liquid-liquid extraction [2], soxhlet extraction [3], accelerated solvent ex traction [4], microwave-assisted solvent extraction [5], and purge and trap [6], larg e amounts of hazardous organic solvents are typically consumed. Today, chemists are look ing for ways to minimize or even eliminate the use of organic solvents in the preparation of samples for analysis. Other sample preparation methods that reduce the amount of solvent consumed include solid-phase extraction [7] and supercritical fluid extraction [8]. The pr eparation of samples without the use of solvents is known as solvent-free sample preparation. Solvent-free sample preparation techniques include gas phase extraction (GPE) [7], membrane extraction [7], and solid-phase microextraction (SPME) [9]. In this chapter, traditional sample preparation methods and solvent-free sample preparation techniques wi ll be reviewed.

PAGE 33

2 1.2 Traditional sample preparation methods The two most commonly utilized traditional sample preparation methods are liquid-liquid extraction and Soxhlet extraction. Both of these sample preparation methods involve the use of large amounts of organic solvents. 1.2.1 Liquid-liquid extraction Liquid-liquid extraction involves the use of water-immiscible organic solvents to extract analytes from aqueous samples [2]. The aqueous sample solution and organic phases are then mixed and shaken, and during th is process the analytes dissolve into the organic phase. The solvent with the extrac ted analytes separates out from the aqueous phase when the shaking is stopped. Liquid-liq uid extraction is also useful for sample cleanup, since constituents that are insoluble in the organic solvent, such as salts and biological macromolecules, remain in the a queous solution. Typical solvents used in liquid-liquid extraction include hexane, diethyl ether, chloroform, pentane, methylene chloride, and ethyl acetate. Liquid-liquid ex traction is an exhaustive sample preparation method. In exhaustive sample preparation methods, the analytes should be extracted quantitatively from the sample and into the organic solvent. For efficient liquid-liquid extraction, the volume of organic solvent used is typically equivalent to or even larger than the original sample volume. For pr econcentration purposes, the organic solvent which contains the extracted analytes is often evaporated to dryness and then reconstituted into a small volume of solvent [2 ]. Liquid-liquid extraction is most suitable for large processing capacities, such as t hose used in the oil industry [10]. Other modifications of liquid-liquid extraction, such as liquid-liquid-liquid microextraction [11]

PAGE 34

3 and liquid phase microextraction [11], ha ve been introduced for more efficient preconcentration and samp le cleanup [12]. 1.2.2 Soxhlet extraction Soxhlet extraction is a frequently used li quid-solid extraction me thod [3]. It is among the oldest methods of sample pretreatment and has been in use for over a century [13,14]. In Soxhlet extraction, a solid sample is typically placed into a cellulose extraction thimble [15]. The extraction thimble is subsequently placed into a Soxhlet assembly, where the thimble gets filled with warm organic solvent, which comes from a distillation flask. The analytes get extracted into this solvent, which automatically siphons back to the distillation flask. A co ffee maker is a type of Soxhlet extraction device. The extraction and siphoni ng process is repeated until all of the analyte has been extracted into the solvent (exhaustive extraction) [13]. Since it is repeated and the same solvent is reused, Soxhlet extraction is considered a hybrid co ntinuous-discontinuous technique [13]. Soxhlet extraction offers the advantages of being inexpensive and very simple to operate – no specialized trai ning is required. It is also not dependant on the sample matrix [13]. Soxhlet ex traction is not without its drawb acks. Soxhlet extraction is an exhaustive extraction procedure and it is a sl ow process (typically 18-24 hours) [16,17]. Soxhlet extraction, like liquid-liquid extracti on, requires the use of large quantities of organic solvents [15]. Since such larg e amounts of organic solvents are used, evaporation/concentration steps are also required [13].

PAGE 35

4 1.3 Sample preparation methods wi th reduced solvent consumption To reduce the amounts of hazardous solven ts consumed in sample preparation techniques, some alternative sample prepara tion methods have been introduced. These methods, which include solid-phase extraction and supercritical flui d extraction, require reduced amounts of organic solvents. 1.3.1 Solid phase extraction Sorbents can also be used to extr act trace organic compounds from aqueous samples. These sorbents can also be used to extract analytes from air and soil samples [7]. These sorbents include chemically bonded reversed-phase silica, co-polymers, carbon-based, ion-pair and ion-exchange, mi xed mode, normal phase, restricted access matrix, molecularly imprinted polymers, and me tal loaded sorbents [18]. In solid-phase extraction, a sample is passed thr ough a plastic tube or disk that contains sorbent that is dispersed on a particulate support [7]. The main goals of solid phase extraction include trace concentration, matrix si mplification, and medium exchange [19]. The idea behind solid phase extraction is that sorbents with a strong affinity towards organic compounds can retain and concentrate these compounds from very di lute gaseous or aqueous samples. Many types of sorbents are availabl e, each specially suited to extract different groups of organic compounds with varied selectiv ities. During this process, the analytes, as well as other interfering compounds in the sample, are extr acted by the sorbent material, which is known as the solid phase [ 19]. Next, a selective solvent is used to rinse out the interfering compounds Finally, another solvent is used to elute the target analytes. Solid phase extraction offers the a dvantages of being inexpensive, simple, and requiring only small amounts of solvent. Dr awbacks of solid phase extraction include

PAGE 36

5 low recovery of analytes, blocking from solid or oily components, and poor reproducibility due to carryover problems and variations of the sorbent from batch-tobatch. Solid phase extraction is also limited to the extraction of se mi-volatile compounds with boiling points substantially hi gher than the solvents used to desorb them [7]. Solid phase extraction became widely used in th e 1980s with the introduction of commercially available disposable cartridges [19]. 1.3.2 Supercritical fluid extraction Organic solvents are not consumed fo r the extraction of nonpolar analytes in supercritical fluid extraction [20]. In superc ritical fluid extraction, a supercritical fluid, typically compressed carbon dioxide near its cri tical point, is used as an extracting phase for the removal of nonvolatile organic compounds from samples at ambient temperatures. This is useful for the analysis of thermally unstable compounds [7]. Other supercritical fluids that can be used include nitrous oxide, ethane, propane, n -pentane, ammonia, fluoroform, sulphur hexafluoride, and water [8]. Carbon dioxide offers the advantages of being inexpensive, readily available, highpurity, non-toxic, having a readily accessible critical point [21]. Supercritical fluid extr action has been applied to the extraction of nonpolar and moderately polar compounds [22]. Supercritical fluids offer the advantages of gas-like diffusion and viscos ity and liquid-like solvent ch aracteristics. Supercritical fluid extraction offers the advantages of se lectivity, accuracy, and on-line integration of sample preparation and detection [8]. Superc ritical fluid extraction offers comparable or better sensitivity compared to traditional S oxhlet and liquid-liquid extraction techniques [23]. The disadvantages include lack of a universal method, difficulty in extracting polar and ionic compounds, where organic solvents ar e often added to the supercritical fluid,

PAGE 37

6 difficulty in extracting liquid samples, slow adoption of method by regulatory authorities, inefficiency in sample cleanup, and unreliab le equipment [8]. Supercritical fluid extraction requires the use of expensive, high-pressure supercri tical fluid delivery systems and tanks of high-purity gas. This limits the field applications of the technique [7]. As a result, supercritical fluid extr action has not had the expected impact in environmental research, and some manufact urers have stopped pr oducing supercritical fluid extraction equipment [8]. 1.4 Solvent-free sample preparation Due to the large amounts of hazardous solvents that are consumed in traditional sample preparation techniques, there was a pus h for the scientific community to develop sample preparation techniques that do not use solvents. The demand for solvent-free sample preparation methods was also fuel ed by stricter environmental regulation regarding the disposal of hazardous organic solvents. Solvent-free sample preparation techniques include gas phase extrac tion, membrane extraction, and SPME. 1.4.1 Gas phase extraction Sample preparation methods that involv e the partitioning of analytes into a gas phase are known as gas phase extraction [7]. During the partitioning process, nonvolatile compounds with high molecular weights are eliminated, thus preventing contamination of separation columns when used in conjunction with chromatographic separation techniques. Gas phase extrac tion is accomplished via headspace sampling, which can be either static or dynamic. Stat ic headspace sampling is the most frequently applied solvent-free sample preparation method [7]. In the static headspace technique,

PAGE 38

7 the sample, which could be a liquid or a solid, is placed in a vial which is later sealed. The vial is then heated and the volatil e samples enter the headspace, forming an equilibrium between the headspace and the samp le matrix. A portion of the vapor from the headspace is then removed a nd injected into a GC [24]. It has even been applied to field analysis [7]. Static headspace sampling ha s been used extensively in the analysis of volatile organic compounds in clinical [7], and food samples [25]. Aside from sampling from the headspace in a sealed vial contai ning a liquid sample [26], static headspace sampling can also be used for collecting ai r samples in the field, a process known as “sniffing” [7]. A diagram of static h eadspace sampling is shown in figure 1.1 [24].

PAGE 39

8 Figure 1.1 Sealed vial used for th e static headspace extraction technique. An extraction equilibrium of the analyte is established be tween the sample solution and the headspace [24].

PAGE 40

9 Static headspace sampling does not concen trate the analytes, and therefore has a significant sensitivity disadvant age. It is generally a nonexhaustive extraction technique, except in the case of extremely volatile gasse s, and therefore requi res careful calibration to accurately determine the concentration of an alytes in the original solution. To address these drawbacks, dynamic headspace sampli ng, also known as purge and trap sampling, was developed [7]. Dynamic headspace sampling has been applied to the analysis of volatile organic compounds in water [27], f ood [28], and airborne pollutants [27]. In dynamic headspace extraction, it is possible to quantitatively remove volatile organic compounds from the sample solution to accurate ly determine their concentrations [7]. Volatile organic compounds are generally low molecular weight aliphatic and aromatic compounds with low boiling points, many of which are halogenated [29]. Dynamic headspace sampling involves two steps. In the first step, a carrier gas bubbles through an aqueous sample to purge the volatile orga nic compounds from the solution. During the second step, these compounds are collected us ing a cold or sorbent trap. Dynamic headspace sampling still has some drawbacks, including instrument carryover. It is important to note that all gas phase extracti on methods are limited to the analysis of volatile compounds. Less volatile compounds can sometimes be analyzed by heating the samples [7]. Overall, the advantages of headspace sampling include low cost, simple instrumentation, and elimination of organic so lvents [30]. The primary disadvantage is low sensitivity [30].

PAGE 41

10 1.4.2 Membrane extraction Membrane extraction was al so introduced as a solvent-free sample preparation technique [7]. In membrane extraction, two simultaneous processes occur. Analytes are extracted from the sample matrix by the membrane material and the analytes are extracted from the membrane material by th e stripping phase. Membrane extraction was developed for use in mass spectrometry, but it has not been extensively applied to chromatographic separation techniques. T ypically, when applied to chromatographic separation techniques, nitrogen was used to strip the analytes from a polymeric membrane to a bed of activated charcoal. After switching a valve, the analytes were desorbed into a gas chromatograph for analys is. Early methods of membrane extraction used supported membrane sheets [7]. When supported membrane sheets are used in membrane extraction, a three phase system is used, where an organic phase is sandwiched between two aqueous phases. The organic phase is immobilized in a porous hydrophobic membrane. Membrane extraction us ing supported membrane sheets can be compared to two-step liquid-liquid extrac tion with dialysis [31]. More recent developments involve the use of hollow membra ne fibers [7]. Hollow membrane fibers are self supporting and simpler to make. The hollow fiber format also offers the advantage of enhanced surface area to volume ratio, making the stripping gas more efficient in desorbing the extracted analyt es. Membrane extraction can be directly combined with gas chromatography (GC) [7], mass spectrometry (MS) [7], highperformance liquid chromatography (HPLC) [32], and capillary el ectrophoresis (CE) [33]. High selectivity and clean analyte extracts can be obtained using membrane extraction [31]. Also, membrane extraction virtually eliminates the use of organic

PAGE 42

11 solvents [31]. Sensitivity in membrane extr action can be enhanced using a sorbent trap [34]. Disadvantages of membra ne extraction include limitation to the analysis of volatile or semi-volatile compounds [35] and slow response of the membranes to changes in concentration, which results in system carryove r [7]. Also, hollow membrane fibers are fragile and difficult to clean after use [33]. 1.4.3 Solid phase microextraction (SPME) Solid phase microextraction is a relati vely new sample preparation technique developed by Belardi and Pawliszyn in 1989 [9]. It is solvent-free, portable, inexpensive, and integrated. [36,37] Solid phase micr oextraction is typi cally coupled with chromatographic analysis of the extracted anal ytes. It has some unique capabilities in that it can be used in the chromatographic anal ysis of dilute solutions in difficult matrices in both the liquid and gaseous phases. It can also be used in the headspace SPME of solids. In general, SPME is used to extrac t organic analytes from gaseous or aqueous sample matrices and is not applied to the an alysis of organic matr ices, such as solvent impurities. Solid phase microextraction consists of two discrete steps. Step one, sampling mode, involves the sorptio n of an analyte from the samp le matrix into a layer of silicone or related sorptive material. Step two, desorption mode, consists of transferring the sorbed analytes into a chromatography in let system by thermal or solvent desorption. Solid phase microextraction can be used fo r GC [38], HPLC [39], supercritical fluid chromatography (SFC) [40,41], and (CE) [42]. By using a sorbent material, solid phas e microextraction essentially eliminates solvent consumption since the sorbent material takes the place of the extracting solvent. This reduces overall solvent consumption in the laboratory, and elimin ates the problem of

PAGE 43

12 having to dispose of used solvents, many of which are hazardous to the environment and to health in general [43]. SPM E can transfer analytes easily to a GC inlet [36]. It was used primarily for environmental water analys is at first, but its applications have dramatically increased [44] since it was inve nted. SPME involves basic equipment that is relatively simple. Figure 1.2 depicts a basic SPME sampling device [36].

PAGE 44

13 Figure 1.2 Schematic of an SPME extraction device [36].

PAGE 45

14 In the SPME device, a fused silica rod is connected to a stainless steel tube that can be withdrawn, using a plunger, into a syri nge needle after samp ling. It was designed to be used with reusable and replaceable fi ber assemblies [43]. The fused silica rod is coated with relatively thin films of polymic stationary phases, whic h are usually used as coating materials in chromatography. The co ating is typically 100 m thick and extends 1 cm from the end of the fiber. This film concentrates the organic analytes during sorption from the sample matrix and can be reused many times. The affinity of the fiber coating for an analyte is the most importan t factor in SPME [36]. SPME fiber coatings are not uniformly sensitive to all compounds [45]. The selected fiber must be of suitable polarity and thickness for the an alyte under investigation [36] The most commonly used nonpolar SPME coatings are polydimethyls iloxane (PDMS) [46-48]. Polar SPME coatings containing polyesters or acrylates will enrich polar analytes and discriminate against nonpolar ones [49-51]. Coatings w ith active carbon cons tituents will retain volatile components more strongly than coat ings made of nonpolar dimethylsiloxanes. Samples can also be converted to more stable analogs before extracti on. It definitely is very important to think about desorption when choosing a fiber coating. If the analyte is held very strongly to the fiber coating, it might be too difficult to transfer the analyte off of the coating for analysis [43]. Fiber SPME can be applied to gaseous, liquid, semiliquid, and solid samples [52]. Prior to sampling, the fiber should be cl eaned to remove any contaminants that may give a high background in the chroma togram. During sampling, the plunger is pushed down, exposing the fiber from the syringe needle. The fiber is then inserted into an aqueous sample solution [36]. The sample may be agitated to assist the extraction

PAGE 46

15 process. The fiber coating is exposed to a st irred sample and the analytes are sorbed into the fiber coating. Once the sampling is complete the fiber is typically withdrawn into the syringe and then transferred to a heated gas chromatogra phy inlet [36]. It is very important to withdraw the fiber when tran sferring because, once the fiber is removed from the sample environment, the extracted an alytes will immediately start to desorb into its surroundings. This desorption is fairly low for many solute s, but more volatile solutes can experience significant losse s. Also, the SPME coating can easily pick up non-sample components from the air. Usually, in a labor atory, the transfer time is short enough that the losses are insignificant. During exte nded transportation times, enclosing the SPME fiber will be necessary. The SPME devices can be enclosed in commercially available sealing systems [43]. To fac ilitate the use of solid-phase microextraction (SPME) for field sampling, a new field sampler was desi gned and tested [53]. This sampler is versatile and user-friendly. The needle is pr otected within a shield at all times, which eliminates the risk of operator injury and fiber damage. A replaceable Teflon cap is used to seal the needle to preserve the sample [53]. A schematic of a field SPME sampler is depicted in figure 1.3 [53].

PAGE 47

16 Figure 1.3 The new SPME field sampler. Parts: (a) and (b) fiber holder, (c) commercialized fiber assembly. (d-1) cross view of the adjustable cylinder, (d-2) side view of the adjustable cylinder, (e) prot ecting shield, (f) repl aceable Teflon cap [53].

PAGE 48

17 The analytes are then thermally desorbed from the fiber and transferred into the column. SPME does not require any sp ecial thermal desorption equipment or modification of the gas chromatography inlet. In other words, SPME has the power to combine the sample preparation and the injec tion of the sample into a chromatograph in only one step [37]. Solid phase microextrac tion is a non-exhaustive type of extraction. Not everything is picked up from the sample, but an equilibrium is established between the fiber and the matrix over time. The concentr ation of the analyte in the original sample can be calculated from this equilibrium [54]. Fiber SPME is definitely a very powerfu l sample preparation technique, but it does have some disadvantages. In fiber SPME, the fibers are very fragile and can be easily damaged. Also, the fiber coating may be damaged during insertion and agitation. The needle of the SPME device can also be bent A problem for all types of SPME is that high molecular weight compounds, such as protei ns, may irreversibly sorb onto the fiber, which would change the properties of the stat ionary phase and make it unusable [55]. The short length of the coated segment of th e fiber only allows a small amount of sorbent loading available for extracti on. This results in low sample capacity of the fiber and imposes limitations on the sens itivity of the technique. Other problems include ghost peaks due to septum particles and fiber gl ue, as well as the memory effect caused by incomplete analyte desorption at the highest allowable temp erature [56]. Also, in SPME, possible matrix effects could affect the sorp tion of the target analyte. For example, SPME samples of tert-butyl ether and tert-but yl alcohol were affected by the presence of high levels of monoaro matic compounds [57].

PAGE 49

18 SPME sampling is based on the principle of partitioning the analytes between the fiber coating on the SPME device and the sample matrix. Once equilibrium is reached between the fiber coating and the sample, which typically takes about 30 minutes, the concentration of the analyte in the original sample can be calculated using the equation: 0C V V K V V K ns f fs s f fs Equation 1.1 Kfs = the distribution constant of the solute between the fiber coating and the sample n = the number of moles of analyte that are extracted by the fiber at equilibrium Vf = the volume of the coating on the fiber Vs = the volume of the sample in liters C0 = the molar concentration of the analyte in the original sample. Usually, however, the volume of the sample is much greater than the volume of the coating on the fiber multiplied by the distribution constant. In this case, the KfsVf in the denominator of the equation can be ignored, wh ich allows the volume of the sample to be eliminated from the equation. This is very valuable because the original volume of the sample does not even need to be known. Th is comes in extremely handy when collecting environmental samples, such as the water in a lake [58]. If the volume of the original sample is eliminated from the equa tion, the equation is simplified to: n = KfsVfCo Equation 1.2 If it is not known that the equilibrium will establish in thirty minutes, and it cannot be obtained from the l iterature, an experiment will need to be conducted which involves allowing the fiber to extract for di fferent time intervals and determining for which time interval the amount of analyte sorbed on the fiber reaches a maximum and

PAGE 50

19 starts to level off [43]. If an unreasonab ly long sampling time is required to reach equilibrium, it is possible to perform SPME wit hout reaching equilibrium. In this case, it is very important to make sure that th e same SPME sampling time is used for each sample and this amount of time should be as long as possible [43] For compounds with a high Kfs value, the sample volume can contribute significantly to th e amount extracted. The sensitivity of SPME is also affected by th e volume of the fiber coating. Increasing the volume of the fiber coating is not feasible because it is difficult to make a thicker coating on the fiber and because th e fiber must fit inside the sy ringe needle in order to be easily injected into a gas chromatography inlet. The best way to increase sensitivity of SPME is to increase the affinity of the analyt e for the fiber coating. This can be done by either changing the chemical nature of the fiber coating, by agitating the sample, by modifying the sample matrix through adjusti ng the pH or temperature, or by adding salt [55]. As mentioned previously, the solvent-free sample prep aration technique of solidphase microextraction was developed origin ally for water samples. In order to accommodate solid samples, a modification of SPME, known as the headspace SPME technique, was developed. This technique combines both headspace and SPME techniques. Regular SPME is used typically for cleaner samples, while headspace SPME is used for dirtier samples [59]. Since the development of headspace SPM E, more applications for SPME have been studied, including applications in pharmaceutical samples, food samples, and different environmental samples. In headspace SPME, a solid sample is put into a headspace vial and then the vial is sealed. It is typically best to minimize the volume of

PAGE 51

20 the headspace in order to maximize the extrac tion potential [60]. Sometimes the vial is heated to increase the vapor pr essure of the target analytes from the sample. The fiber may be cooled as well to allow for maximum extraction. An extraction equilibrium is established between the solid sample and the vapor headspace. For sampling, an SPME fiber is inserted into the headspace without contacting the sample. The fiber coating on the end of the rod of the SPME device absorbs the analytes from the headspace. In the end, a three-way equilibrium is established between the sample, the headspace, and the fiber. In headspace SPME, volatile samples ar e readily concentrated in the headspace. For semi-volatile analytes, the mass transfer from the matrix to the headspace will be much slower. Also, for solid samples, the diffusion is much slower. The slow volatilization can limit the speed of the extr action, causing longer extr action times [60]. Figure 1.4 compares the direct SPME technique (a) and the headspace SPME technique (b) [43].

PAGE 52

21 Figure 1.4 Comparison of direct SPME (a ) and headspace SPME (b) [43].

PAGE 53

22 The sensitivity of headspace SPME can never be any higher than that of direct immersion SPME. For volatile an alytes, the sensitivity of SPM E is generally not affected by using headspace SPME. However, for semi-volatile analytes, the sensitivity of the headspace SPME technique is greatly reduced. Overall, the headspace SPME technique, when used for solid samples, is slower and less sensitive than the classical SPME technique [59]. Generally, other traditional sample preparation techniques for the analysis of organic pollutants in water require the extraction of thes e pollutants with the use of harmful organic solvents. Solid phase micr oextraction is a solvent-free non-exhaustive extraction method, based on equilibrium. With the proper calibration, SPME can allow for quantitative analysis of organic pollutant s at a good sensitivity without the use of any organic solvents. This solvent free sample prep aration technique can take as little as 30 minutes for sample preparation and analytical separations. SPME can even be easily automated. Solid phase microextraction is id eally suited for use in the field. Recently, subcritical water extraction has even been combined with so lid phase microextraction to allow for a very rapid, organic solvent free method that is used to determine organic pollutants in soils and sludges. One of the best features of SPME is that analyte extraction and pre-concentration of the analytes are co mbined in one step. Since SPME is such a powerful technique, it has been commer cialized by Supelco, Varian, and Leap Technologies. It certainly has numerous appli cations in the analysis of many compounds in a vast array of matrices [43].

PAGE 54

23 1.4.4 Further development in SPME Since SPME was invented, there has b een further development of new SPME coating materials, such as molecularly imprinted polymers. Molecularly imprinted polymers are polymers that are created in order to leave specific cavities in the polymeric matrix. These cavities correspond to the shapes and sizes of the target analyte molecules. Molecularly imprinted polymers are incredibly sp ecific to the target analyte. Molecularly imprinted polymers are used as both fiber co atings and as coati ngs in in-tube SPME techniques [37]. The efficiency of anodized aluminum wire was also investigated as a new fiber for SPME. Anodized aluminum wire works well for the sampling of organic compounds from gaseous samples due to the layer of porous aluminum oxide. This study is an example of the search for firm SPME fibers with long life spans that can replace the delicate silica fibers [61]. Overall, SPME is now widely recognized and used, especially in clinical, pharmaceutical, environm ental, and food analysis [37,55,62]. Some alternative microext raction techniques have also been developed, including solid-phase dynamic extraction [55] using an internally coated needle, solid-phase microextraction with rotation of the microfiber [63], micr owave assisted headspace solid phase microextraction [64], and matrix so lid phase microextraction [65], membraneprotected solid phase microext raction [66], stir-bar-sorptiv e extraction using a coated magnetic stir bar [55], and thin-film microe xtraction [67]. These techniques can be coupled to gas chromatography, GC mass spectrometry, high-performance liquid chromatography, and capillary electrophores is for the analysis of many complex mixtures. These new techniques are also so lvent free sample preparation methods, which

PAGE 55

24 saves solvent purchase and dispos al costs. Most of these t echniques can be performed in less time than traditional sample preparation methods and they can be applied to pharmaceutical and biomedical analysis. 1.4.5 Solid phase dynamic extraction Solid-phase dynamic extraction, or SPDE, is an inside-needle technique that is used for vapor and liquid sampling. This met hod uses stainless steel needles, about 8 cm long, that are coated with a 50 mm film of polydimethylsiloxane (PDMS) and a 10% activated carbon [68]. An SPD E device can be compared to an SPME fiber. In this method, dynamic sampling is performed by passing the headspace through a tube using a syringe. The volume of the stationary phase of the SPDE needle is about 5.99 mm3. In SPDE, the analytes are concentrated onto PDMS and activated carbon coated onto the inside walls of the stainless steel needle of a 2.5 mL gas tight syringe. SPDE sampling can be done under dynamic conditions while keeping the headspace volume constant. The analytes that are trapped can be recovered by heat desorption directly into a GC inlet. The main advantage of SPDE over SPME is the mechanical strength of the capillary. The SPDE device is much stronger than the frag ile fibers used in SPME. In fact, it is nearly impossible to mechanically damage a SPDE device. The main disadvantage to SPDE is that it may have carryover, in which the analytes tend to remain inside the needle wall after the heated desorption into th e GC inlet. This technique has not yet been extensively applied [69]. In figure 1.5, the SPDE extraction process is shown [55].

PAGE 56

25 Figure 1.5 Solid phase dynamic extraction. The n eedle device is inserted into the headspace, and the syringe is moved back a nd forth to extract the analytes from the headspace onto the extracting phase on the inner surface of the SPDE device [55].

PAGE 57

26 1.4.6 Solid phase microextraction with rotation of the microfiber A new technique has been developed th at helps diminish the absorption time needed during solid phase microextracti on. This method relie s on rotation of the microfiber in order to accelerate the absorpti on process. The absorption efficiency of the rotation technique is better than standard (static) solid phase mi croextraction. This technique can also be automated [63]. Theoretical extraction times for the mixing of aqueous samples can only be reached approximately and are dependent on th e mixing technique that is applied. Three mixing techniques that have been used incl ude magnetic stirring, intrusive stirring, and sonication. Magnetic stirring only exhibits low mixing ef ficiency. Intrusive mixing allows efficient agitation, but causes heating of the sample. Sonication, the third tested technique, is the most effici ent method, but it also causes samp le heating and is difficult to automate. Gepperts suggested vibration of th e fiber, as an altern ative to sonication. This method has been patented. It significan tly increases the precision and sample output [63]. The fiber rotation method was tested ag ainst the vibration method. It was found that, using the fiber rotation method, the abso rption process accelerates up to a similar time as the fiber vibration method. Therefor e, the fiber rotation method is a suitable alternative to the existing vi brational and stirring techniqu es. The fiber rotation method is easy to handle, smooth, and rugged. It s eems to have the potential to replace the existing techniques [63]. Figure 1.6 depicts four different sorption methods used for SPME [63].

PAGE 58

27 Figure 1.6 Different sorption methods used for SPM E. Part A shows static sorption, with no movement of the fiber. Part B shows rota tion of the fiber. Part C shows magnetic stirring of the sample. Part D s hows vibration of the fiber [63].

PAGE 59

28 1.4.7 Microwave assisted headspace SPME Microwave assisted headspace solid phase microextraction was developed as a simple and effective method for fast samp ling of volatile organi c compounds [64]. During microwave heating, a simple shielding device made of aluminum foil is used to protect the SPME fiber from microwave irra diation while allowing the sample to be heated. A room temperature water bath is al so used to allow micr owave heating to be conducted in a more controlled manner. Th e inner heating from the microwave radiation accelerates the emission of volatile organic compounds from a sample without a marked change in headspace temperature in the sa mple vial. Under optimum conditions, the extraction efficiencies obtained from microwave assisted headspace solid phase microextraction are much higher than t hose obtained without microwave heating. This improvement of extraction efficiency allo ws more volatile organic compounds to be detected with a more balanced extraction of volatile organic compounds of lower and higher molecular weights. When coupled with GC with flame ioniza tion detection (FID), this technique is useful for quantitative an alysis of individual volatile compounds [64]. 1.4.8 Matrix SPME In 2000, a study was done to measure dissolved concentrations of persistent and bioaccumulative pollutants (PBPs) in sediment porewater [65]. The method used for measuring PBPs in the sediment porewater is called matrix SPME because it utilizes the entire sediment matrix as a reservoir for an equilibrium extraction. A glass fiber with a coating of poly(dimethylsiloxane) was placed in a sediment sample until the PBPs reached their equilibrium distribution between the PDMS and the sediment matrix. The PBP concentrations were analyzed through gas chromatography and the porewater

PAGE 60

29 concentrations were calculate d. This method has a sensitivity in the pg /L to ng/L range. Matrix-SPME requires few materials and little operat ion time [65]. 1.4.9 Membrane protected SPME Membrane protected solid phase microextraction is typically used for samples that contain interferences larger than the target analyte [66]. In membrane protected SPME, the coating of the fiber is covered with a me mbrane. This membrane has holes in it that allow the analyte to enter the SPME system, but it restricts the larger, interfering molecules from entering the fiber [66]. 1.4.10 Stir bar sorptive extraction (SBSE) Stir bar sorptive extraction (SBSE) is a solvent-free sample preparation technique that is very promising for enriching solute s from aqueous samples [55,70,71]. The stir bar sorptive extraction technique uses a magnetic st ir bar coated with PDMS phase, which is similar to SPME, but in a thicker layer of 0.3 to 1.0 mm [70]. PDMS co ated stir bars are commercially available as Twister stir bars. These stir bars come in lengths of 10 mm and 40 mm. The 10 mm stir bars are coated with 55 mL of PDMS liquid phase and the 40 mL stir bars are coated with 219 mL of PDMS liquid phase. The 10 mm stir bars are best suited for stirring 10 to 50 mL samples. The 40 mm stir bars are best suited for samples up to 250 mL [55]. Figure 1.7 depicts the SBSE process [71].

PAGE 61

30 Figure 1.7 The stir bar sorptive extrac tion process. In A, the magnetic stir bar is allowed to spin in the analyte solution for extraction, in B the stir bar is insert ed in to a stir bar desorption device, and in C, the stir bar deso rption device is transferred to a GC inlet for thermal desorption [71].

PAGE 62

31 In SBSE, the sample is poured into a 20 mL headspace vial and stirred with a PDMS coated stir bar for 60 minutes at 1000 rpm. (part A of figure 1.7) After sampling, the stir bar is removed with tweezers, the residual water drop lets are removed with clean tissue paper, and the stir bar is placed in an empty glass tube that is 187 mm long with a 6 mm outside diameter and a 4 mm inside diamet er. (shown in part B of figure 1.7) This allows for thermal desorption of the sample (part C of figure 1.7) Sample volume and stirring speed are important in SBSE because they influence the extraction efficiency. The typical stirring times for an equilibrium to be establis hed, as in SPME, are 30 to 60 minutes. This extraction mechanism is si milar to SPME based on PDMS sorption, but SBSE uses a much higher mass of PDMS. This results in higher recoveries and higher sample capacities. The PDMS phase is not pola r, so it is not suitable for extracting polar compounds. Derivitization of the aqueous phase can allow for the possibility for sampling polar compounds. SBSE is compa tible with GC and HPLC [72]. SBSE combined with liquid desorption has allowed for applications for the analysis of high molecular mass compounds and thermolabile solutes. SBSE has been used for environmental, food, biomedical, and life scie nce applications [73] SBSE is more sensitive than SPME fibers for certain applic ations, but it requires a special desorption unit. This makes the process difficult to automate [74,75]. 1.4.11 Thin-film microextraction Pawliszyn and coworkers examined the properties of a thin sheet of poly(dimethylsiloxane) membrane as an extr action phase [67]. These properties were compared to those of PDMS coated SPME fi bers for their application to semi-volatile analytes in direct and headspace modes of SPME. This new approach has much higher

PAGE 63

32 extraction rates because of a larger surface area to extrac tion phase volume ratio of the thin film. The high extraction rate of the membrane SPME technique allows for larger amounts of analytes to be extracted within a shorter period of time than with typical thick coated fibers in SPME [67]. 1.4.12 In-tube SPME or capillary microextraction (CME) A new solvent-free sample preparation met hod has been developed that is related to SPME. This technique is known as in-tub e SPME. This technique uses fused-silica GC capillary columns for SPME [76]. The sorbent with the affinity for the target analyte is coated inside a capillary instead of just on the end of a fiber, as in standard SPME. Intube SPME is suitable for automation, which shortens analysis times and increases accuracy and precision relative to manual techniques. In-tube SPME also enhances selectivity since it the coating has a larger surface area than the surface area of a fiber coating. Furthermore, there is a reduced risk of breakage of the SPME unit using an intube SPME device when compared to a standard SPME device [76]. In-tube SPME can be coupled on-li ne with high performance liquid chromatography or gas chromatography [74]. In-tube SPME is suitable for automation. Extraction, desorption (removing the sample from the coating), and injection can be done continuously with an autosampler. Automa ted sampling procedures shorten the total analysis time and are actually more accura te and precise than manual techniques. Through using in-tube SPME for analysis the organic compounds in aqueous samples are extracted directly from the sample into the stat ionary phase that is co ated on the inside of the capillary, which has an affinity for the an alyte [55]. The coatings in in-tube SPME

PAGE 64

33 capillaries are very similar to those in co mmercially available SPME fibers [77]. Figure 1.8 compares in-tube SPME (b) with fiber SPME (b) [43]. In-tube SPME is also often referred to as capillary microext raction, a term which was coined by the introduction of sol-ge l capillary microextraction by Malik and coworkers in 2002 [78], where sol-gel extrac ting phase coatings were chemically bonded to the inner walls of fused silica capillari es. This surface bonding leads to enhanced thermal and solvent stability. Sol-gel capillary microextraction will be discussed in greater detail in chapter 2.

PAGE 65

34 Figure 1.8 In fiber SPME (a), the SPME extracting phase is coated on the outer surface of the fiber. In in-tube SPME (b), the SPME extracting phase is coated on the inner walls of a capillary [43].

PAGE 66

35 1.5 Applications of micr oextraction techniques Many SPME methods have been devel oped to extract drugs from various biological samples, including urine, serum, plasma, whole blood, saliva, and hair. Recently, there have been many publications about pharmaceutical and biomedical applications of SPME. These applications include many forensic and toxicological analyses, which indicate the ve rsatility of the method and the potential for using it to analyze other types of samples, includi ng clinical, metabolic, and pharmaceutical applications [55]. SPME extractions offer the potential for very clean analyses, with little or no interference from nonvolatile compounds. Th e application of SPME to low volatility drugs and metabolites in plasma may be limited to those with high therapeutic concentrations between 1 and 100 mg/mL due to the relatively low partition coefficients between polar drugs and the commerci ally available SPME fibers [55]. The in-tube SPME technique can be applie d to polar and non-polar drugs in liquid samples using a commercially available GC capillary column. It can be coupled with various analytical methods. Its application to the analysis of biological samples is increasing. The SPDE technique can be auto mated and coupled with GC and used for headspace extraction for the determination of il licit drugs in hair samples. The SBSE technique, using a magnetic stir bar covered with a thick layer of PDMS, can be applied to some drugs in biological samples through combination with GC. SPME and PDMS techniques can be used for the analysis of amphetamines in urine samples. SPME is also used for the analysis of resi dual solvents, such as ethanol, cyclohexane, toluene, benzyl chloride, and triethylamine which can be pr esent in pharmaceutical products [55]. These

PAGE 67

36 techniques can also be used to measure acet one present in breath, which is an indication of diabetes, by extracting from the exhaled air from a patient [79]. Volatile compounds in human blood samples that are indicators of disease can be analyzed using SPME techniques as well [80]. Solid phase microe xtraction and its related solvent-free sample preparation techniques have been widely used for the analysis of various contaminants, including pesticides, organometallic co mpounds, and volatile organic compounds in biological samples. These methods are extrem ely useful for the monitoring of biological toxicology and for enviro nmental chemistry [55]. The applications of SPME and its rela ted techniques are extremely vast. Recently, applications of SPME include chemical warfare agents, pharmaceutical process impurities, organochlorine pesticides in Chinese teas, volatile compounds in acidic media, volatile compounds in cheese, volatile phenols in wine, environmental pollutants in water samples, chloroanisoles in cork st oppers, volatile aliphatic amines in air, indoor air analysis, and phenylurea herb icides in aqueous samples [4 3]. Another application of SPME is the analysis of aldehydes in beer that involves on-fiber derivatization [81]. SPME can even be applied to the analysis of flavors in foods, such as in the ripening of fruit [82]. The volatile compounds of parmesan cheese have even been analyzed through SPME [83]. 1.6 Drawbacks of SPME A significant drawback of mo st SPME fibers is that they have relatively low operating temperatures in the range of 200-2700 C. This is lower than the upper temperature limit for these same materials used as stationary phases in GC columns. The

PAGE 68

37 coating on an SPME fiber is much thicker than the stationary phase film thickness of a GC column. It is much more difficult to st abilize this thicker film. SPME coatings are only physically held to the fiber, which pl aces limitations on their thermal and solvent stability. This, in turn, limits the applic ation of SPME to GC and HPLC, where higher desorption temperatures and organic solvents ar e used, respectively [56]. Furthermore, in the case of fiber SPME, special desorption devi ces must be used in order to couple the sample preparation technique with HPLC analysis [84-86]. The use of these desorption devices results in sample loss, increased e xperimental error, and a more complicated experimental procedure that cannot be easil y automated. A schematic of a desorption device that was used to coupl e fiber SPME to HPLC is s hown in figure 1.9 [86].

PAGE 69

38 Figure 1.9 Schematic of a desorption device used to couple fiber SPME with HPLC [86].

PAGE 70

39 In-tube SPME offers the advantage of being easily hyphenated with HPLC by connecting it as an external sampling loop. However, in-tube SPME is not practical for use in HPLC applications si nce, like fiber SPME, the extr acting phase is only physically held to the fiber. The organo-aqueous m obile phases used in HPLC strip away the extracting phase in in-tube SPME. The problems concerning the limited thermal and solvent stability of SPME coatings can be addressed using sol-gel chemis try. Sol-gel chemistry offers a simple and convenient pathway for the synthesis of advanc ed material systems and for applying them as surface coatings [57]. Sol-gel chemistry can provide efficient incorporation of organic components into inorganic polymeric stru ctures in solution under mild thermal conditions. Sol-gel technology ha s the advantages of low co st, the unique ability to achieve molecular level uniformity in the synt hesis of organic-inorganic composites, and a strong adhesion of the coating to the subs trate (fiber or capilla ry) due to chemical bonding. This chemical bonding provides solven t and thermal stability [57]. Sol-gel capillary microextraction ha s also been introduced [78]. In sol-gel capillary microextraction, a sol-gel extracting phase coating is chemically bonded to the inner walls of a fused silica capillary. Sol-gel cap illary microextraction coatings are thermally stable and resistant to solvents. Sol-ge l capillary microextraction also offers the advantage of ease in on-line hyphenation with HP LC [78]. The use of sol-gel coatings and monoliths in analytical sample preparati on are extensively reviewed in chapter 2. The development and use of novel surface-bonded sol-gel sorbents for capillary microextraction on-line hyphenated with HPLC are described in chapters 3 through 7.

PAGE 71

40 1.7 References for chapter one [1] C. W. Huck, G. K. Bonn, J. Chromatogr. A 885 (2000) 51. [2] S. Pedersen-Bjergaard, K. E. Rasmusse n, T. G. Halvorsen, J. Chromatogr. A. 902 (2000) 91. [3] C. W. Huie, Anal. Bioanal. Chem. 373 (2002) 23. [4] B. E. Richter, LC-GC 17 (1999) S22. [5] V. Camel, Trends Anal. Chem.19 (2000) 229. [6] A. Kumar, Gaurav, A. K. Malik, D. K. Tewary, B. Singh Anal. Chim. Acta 610 (2008) 1. [7] J. Pawliszyn, Trends Anal. Chem. 14 (1995) 113. [8] M. Zougagh, M. Valcarcel, A. Ri os, Trends Anal. Chem. 23 (2004) 399. [9] R. P. Belardi, J. B. Pawliszyn, Water Pollut. Res. J. Can. 24 (1989) 179. [10] P. M. Vashist, R. B. Beckmann, Ind. Eng. Chem. 60 (1968) 43. [11] T. S. Ho, S. Pedersen -Bjergaard, K. E. Rasmussen, J. Chromatogr. A 963 (2002) 3. [12] S. Pedersen-Bjergaard, K. E. Rasmussen, Anal. Chem. 71 (1999) 2650. [13] M. D. Luque de Castro, L. E. Ga rcia-Ayuso, Anal. Chim. Acta 369 (1998) 1. [14] J. R. Dean, An al. Comm. 33 (1996) 191. [15] N. Saim, J. R. Dean, M. P. Abdulla h, Z. Zakaria, J. Chromatogr. A 791 (1997) 361. [16] S. B. Hawthorne, C. B. Grabanski, E. Martin, D. J. Miller, J. Chromatogr. A 892 (2000) 421. [17] B. E. Richter, B. A. Jones, J. L. Ezzell, N. L. Porter, Anal. Chem. 68 (1996) 1033. [18] M. C. Hennion, J Ch romatogr. A 856 (1999) 3. [19] C. F. Poole, Tre nds Anal. Chem. 22 (2003) 362.

PAGE 72

41 [20] J. L. Hedrick, L. J. Mulcahey, L. T. Taylor, Mikrochim. Acta 108 (1992) 115. [21] E. D. Morgan, Na tural Prod. 3 (2000) 3451. [22] I. J. Barnabas, J. R. Dean, W. R. Tomlinson, S. P. Owen, Anal. Chem. 67 (1995) 2064. [23] J. L. Snyder, R. L. Grob, M. E. McNally, T. S. Oostdyk, Anal. Chem. 64 (1992) 1940. [24] D. Jentzsch, H. Kruger, G. Lebrecht, G. Dencks, J. Gut, Z. Fresenius Anal. Chem. 236 (1968) 96. [25] K. R. Christensen, G. A. Reineccius, J. Dairy Sci. 75 (1992) 2098. [26] N. H. Snow, G. C. Slac k, Trends Anal. Chem. 21 (2002) 608. [27] L. Pillonel, J. O. Bosset, R. Tabacch i, Lebensm. –Wiss. u.-Technol. 35 (2002) 1. [28] S. Mallia, E. Fernandez-Garcia, J. O. Bosset, Int. Dairy J. 15 (2005) 741. [29] D. L. Heikes, S. R. Jensen, M. E. Fl eming-Jones, J. Agric. Food Chem. 43 (1995) 2869. [30] J. C. F. Menendez, M. L. F. Sanchez, J. E. S. Uria, E. F. Martinez, A. Sanz-Medel, Anal. Chim. Acta 415 (2000) 9. [31] J. A. Jonsson, L. Mathiasson, Trends Anal. Chem. 18 (1999) 318. [32] J. A. Jonsson, L. Mathiasson, Trends Anal. Chem. 18 (1999) 325. [33] N. C. van de Merbel, J. Chromatogr. A 856 (1999) 55. [34] M. J. Yang, S. Harms, Y. Z. Lu o, J. Pawliszyn, Anal. Chem. 66 (1994) 1339. [35] G. Matz, G. Kibelka, J. Dahl, F. Lennemann, J. Chromatogr. A 830 (1999) 365. [36] C. L. Arthur, J. Paw liszyn, Anal. Chem. 62 (1990) 2145.

PAGE 73

42 [37] S. Risticevic, V. H. Niri, D. Vuc kovic, J. Pawliszyn, Anal. Bioanal. Chem. 393 (2009) 781. [38] F. Fytianos, N. Raikos, G. Theodorid is, Z. Velinova, H. Tsoukali, Chemosphere 65 (2006) 2090. [39] Y. Fan, Y. Q. Feng, J. T. Zhang, S. L. Da, M. Zhang, J. Chromatogr. A 1074 (2005) 9. [40] S. Santos, B. M. Simonet, A. Rios, M. Valcarcel, Electr ophoresis 28 (2007) 1312. [41] A. Panalver, E. Pocurull, F. Borrull, R. M. Marce, Trends Anal. Chem. 18 (1999) 557. [42] Z. Liu, J. Pawliszyn, J. Chromatogr. Sci. 44 (2006) 366. [43] J. V. Hinshaw, LC -GC N. Am. 21 (2003) 1056. [44] C. L. Arthur, M. Killam, S. Motlagh, M. Lim, D. W. Potter, J. Pawliszyn, Environ. Sci. Technol. 26 (1992) 979. [45] R. A. Murray, An al. Chem. 73(2001) 1646. [46] J. Koziel, M. Jia, A. Khaled, J. No ah, J. Pawliszyn, Anal. Chim. Acta 400 (1999) 153-162. [47] G. Ouyang, J. Pawliszyn, An al. Bioanal. Chem. 386 (2006) 1059. [48] D. A. Lambropoulou, V. A. Sakkas, T. A. Albanis, Anal. Bioanal. Chem. 374 (2002) 932. [49] E. M. J. Verbruggen, W. H. J. Vaes, T. F. Parkerton, J. L. M. Hermens, Environ. Sci. Technol. 34 (2000) 324. [50] W. H. J. Vaes, C. Hamwijk, E. U. Ramos, H. J. M. Verhaar, J. L. M. Hermens, Anal. Chem. 68 (1996) 4458.

PAGE 74

43 [51] J. H. Loughrin, J. Agric. Food Chem. 54 (2006) 3237. [52] Y. Leblanc., R. Gilber t, M. Duval, J. Hubert, J. Chromatogr. 633 (1993) 185. [53] Y. Chen, J. Pawliszyn, Anal. Chem. 76 (2004) 6823. [54] D. W. Potter, J. Pawliszyn, Environ. Sci. Technol. 28 (1994) 298. [55] H. Kataoka, Current Ph arm. Analysis, 1 (2005) 65. [56] S. L. Chong, D. Wang, J. D. Hayes, B. W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [57] L. Black, D. Fine, Envi ron. Sci. Technol. 35 (2001) 3190. [58] J. Pawliszyn, B. Pawliszyn, M. Pawliszyn, Chem. Educator 2 (1997) 1. [59] Z. Zhang, J. Pawliszyn, Anal. Chem., 65 (1993) 1843. [60] W. Wardencki, M. Mi chulec, J. Curyto, Int. J. Food Sci. Tech., 39 (2004) 703. [61] D. Djozan, Y. Assadi, S. H. Haddadi, Anal. Chem. 73 (2001) 4054. [62] G. Vas, K. Vekey, J. Mass Spectrom. 39 (2004) 233. [63] H. Geppert, Anal. Chem. 70 (1998) 3981. [64] G. Xiong, C. Goodridge, L. Wang, Y. Ch en, J. Pawliszyn, J. Agric. Food Chem. 51 (2003) 7841. [65] P. Mayer, W. H. J. Vaes, F. Wijnker, K. C. H. M. Legierse, K. C. H. M., R. H. Kraaij, J. Tolls, J. L. M. Hermens, Environ. Sci. T echnol. 34 (2000) 5177 [66] B. Vrana, P. Popp, A. Paschke, A., G. Schuurmann, Anal. Chem., 73 (2001) 5191. [67] I. Bruheim, X. Liu, J. Pa wliszyn, Anal. Chem. 75 (2003) 1002. [68] J. Lipinski, J. Fresen ius, Anal. Chem., 367 (2000) 445. [69] J. Lipinski, J. Fresen ius, Anal. Chem., 369 (2001) 57. [70] E. Baltussen, P. Sandra, F. David, C. Cramers, J. Microcolumn Sep. 11 (1999) 737.

PAGE 75

44 [71] B. Tienpont, F. David, K. Desmet, P. J. F. Sandra, Anal. Bioanal. Chem. 373 (2002) 46. [72] P. Popp, C. Bauer, L. Wennr ich, Anal. Chim. Acta 436 (2001) 1. [73] F. David, P. Sandra, J. Chromatogr. A 1152 (2007) 54. [74] E. Baltussen, C. A. Cramers, C.A., P. J. F. Sandra, Anal. Bioanal. Chem. 373 (2002) 3. [75] P. Popp, C. Bauer, B. Hauser, P. Keil, L. Wennrich, J. Sep. Sci. 26 (2003) 961. [76] R. Eisert, J. Pawlis zyn, Anal. Chem. 69 (1997) 3140. [77] Y. Saito, K. Jinno, Chromatogr., 22 (2001) 151. [78] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [79] C. Deng, J. Zhang, X. Yu, W. Zha ng, X. Zhang, J. Chromatogr. B., 810 (2004) 269. [80] C. Deng, X. Zhang, N. Li J. Chromatogr. B., 808 (2004) 269. [81] P. Vesely, L. Lusk, G. Basarova, J. S eabrooks, D. Ryder, J. Agric. Food Chem. 51 (2003) 6941. [82] X. Yang, T. Peppard, J. Agric. Food Chem. 42 (1994) 1925. [83] J. H. Lee, R. Diono, G. Y. Kim, D. B. Min, J. Agric. Food Chem. 51 (2003) 1136. [84] J. Chen, J. B. Pawliszyn, Anal. Chem. 67 (1995) 2530. [85] H. Kataoka, H. L. Lord, J. Pa wliszyn, J. Chroma togr. A 880 (2000) 35. [86] K. Jinno, M. Taniguchi, M. Hayashid a, J. Pharm. Biomed. Anal. 17 (1998) 1081.

PAGE 76

45 CHAPTER TWO SOL-GEL COATINGS AND MONOLI THS IN ANALYTICAL SAMPLE PREPARATION 2.1 Introduction Sol-gel materials are attracting pronoun ced attention in analytical sample preparation. However, there have been few review articles wr itten about sol-gel materials in analytical sample preparation. So l-gel [1] have been advantageously used as extraction media in various formats, incl uding (a) surface coati ngs [2-103] and (b) monolithic beds [104-117]. As surface coati ngs, sol-gel materials have been used on fibers for solid phase microextraction (SPME) [2-75], on capillary inner walls for in-tube SPME [76-96], and on stir bar sorptive extract ion [98-103]. Sol-gel monoliths have been used for sample preconcentr ation and extraction [104-117]. In this chapter, a comprehensive state of the art review of sol-gel extraction materials in analytical microextraction will be provided. It will include an introduction into sol-gel chemistry, classification of solgel materials, specific information on the preparation and characterizati on of sol-gel materials, dive rse applications of sol-gel materials in analytical extraction or sample preparation, instrumentation for the sol-gel microextraction techniques, types of compounds extracted and analy zed using the sol-gel materials, and analytical characteristics, incl uding extraction capabili ties, detection limits, relative standard deviat ion (RSD) values, percent recoveri es, and desirable properties.

PAGE 77

46 Real-world applications of the sol-gel mate rials to sample preparation will also be discussed for various fields of analysis, in cluding environmental, food, and biomedical areas. 2.2 Sol-gel chemistry Sol-gel precursors typically consist of silicon or transition metal alkoxides [1]. Transition metals whose alkoxides have been used as sol-gel precursors for analytical sample preparation include zirconium [75,94,95], aluminium [14,15], titanium [16,73,74,91-93], and germanium [96] based precursors. The sol-gel process involves mainly hydrolysis and condensation reactions [1]. The first step in the formation of a solgel material is the complete or partial hydrolys is of the sol-gel precurs or. This is usually accomplished under acid or base catalysis [1]. Next, the hydrolyzed precursor undergoes condensation reactions, forming a three-dime nsional polymeric network. During this time, any sol-gel active component in the so l-gel mixture can also condense with the evolving polymeric network. Also, particular ly in the formation of sol-gel fiber SPME coatings, sol-gel capillary coatings, and sol-gel monoliths, the evolving sol-gel network can condense with hydroxyl groups on the surf ace of the substrate, which is typically fused silica. This covalent bonding to the surface gives sol-gel materials their high thermal and solvent stability and makes them very successful as extracting phases [1,2].

PAGE 78

47 2.3 Physico-chemical characteri zation of sol-gel materials This chapter is focused primarily on sol-gel materials and their extraction properties and sorbent characteri stics. However, to have a better insight into the sorbent behavior at the molecular level, many of these sol-gel materials have also been characterized by various physico-chemical methods. Such methods include FTIR [8,9,14,16,17,36,43,53,54,58,59,61,69,87,89,91,94,96,103,112,116], Raman spectroscopy [89], elemental analysis [75], and X-ray studies [75]. Th ese studies are typically used to identify specific chemical bonds and functiona l groups in the sol-gel materials. Also, microscope images have been taken of the sol-gel materials to provide vivid illustration of the morphological aspects of these materials. Most often, a scanning electron microscope (SEM) is used to generate these images [2,6,8,14,17,30,39,54,57,58,64,69,70,75,76,78,84,87,89,91,94,95,98,100,103,106,109, 111,114,116]. A fluorescence micrograph can also be used [35]. These images show the physical appearance of the sol-gel material s and are extremely useful for examining properties such as porosity, crack ing, and coating thicknesses. Many of the studies on sol-gel extracti on materials developed to date involved extensive studies on the physical properties of the coatings th at served as the extracting phase and the performance of these coatings under different cond itions. An important aspect of these studies was the elucidation of extraction profiles for the sol-gel sorbents. In the extraction profile, samples representativ e of particular chemical types are extracted multiple times for different extraction times. The peak areas for each extraction time are then averaged and plotted against their re spective extraction times. The point on the graph in which the average peak area stops increasing with extrac tion time corresponds to

PAGE 79

48 the time required for equilibrium to be established between the sample solution and the sol-gel extracting phase. An example of an extraction profile is indicated in figure 2.1 [84].

PAGE 80

49 Figure 2.1 Example of an extractio n profile. The point on the graph where the average peak area stops increasing with extraction time corresponds to the time required for an equilibrium to be established between the anal yte in the sample and the sol-gel sorbent. This example illustrates the ex traction profiles of nonanal (120 g/L), hexanophenone (50 g/L), 1-nonanol (100 g/L), and nonanoic acid (100 g/L) analytes extracted on a 12 cm 250 m I.D. sol–gel PEG coated capillary from aqueous samples. Extraction conditions: triplicate extraction for 10 20, 30, 40, 50, 60, and 70 min. GC analysis conditions: 5 m 250 m I.D. sol–gel PDMS column; splitless injection; injector temperature: initial 30 C, final 340 C, pr ogrammed at a rate of 60 C/min; GC oven temperature programmed from 35 C (5 min) to 320 C at a rate of 20 C/min; helium carrier gas; FID temp erature 350 C [84].

PAGE 81

50 Other studies that have been done incl ude extraction temper ature profiles, pH studies, solvent tolorences, and lifetime studies. These results are typically presented in graphs and are conducted in order to determ ine the optimal extraction conditions for the sol-gel materials. 2.4 Classification of sol-gel coatings in solid phase microextraction (SPME) Sol-gel technology allowed for the creation of many di fferent types of sol-gel coatings for analytical microextraction. Su ch coatings were developed both for fiber SPME [2-65] and in-tube SPME (capillary mi croextraction) [76-96], with different extraction capabilities and a pplications. The reported so l-gel SPME coatings can be classified into (a) polysiloxane-based monofunctional sol-gel fiber coatings, (b) polysiloxane-based multifunctional sol-gel fiber coatings, (c) sol-gel coatings with cavity ligands, and (d) non-polysiloxa ne sol-gel coatings. 2.4.1 Fiber preparation and p retreatment procedures Most sol-gel fiber SPME coatings are made by coating the surface of a fused silica fiber. Some important pretreatment steps must firs t be performed. Fused silica fibers typically have a protective polyimide coating on them, which must be removed in order to expose the fused silica surface on the por tion of the fiber at one of its ends that is to be coated. One of the simplest ways to accomplish this is to burn the polyimide coating off of the fused silica fiber, using ei ther a match or a cigarette lighter. After burning off the polyimide layer, the fused silica surface is rinsed in methanol [2,6]. Other ways to remove the protective coating are to dip the end of the fiber in acetone

PAGE 82

51 [21,27,31,35,37-39,42,43,53,54,57,59,61,70] dichoromethane [69], or concentrated sulfuric acid [9,17] for several hours. Fused silica fiber contains a small numbe r of silanol groups on its surface. The chemical bonding of a sol-gel coating to a fiber is accomplished through condensation reactions between hydroxyl groups in the so l-gel network and silanol groups on the surface of the fused silica fiber. Therefore, it would be advantageous to have more silanol groups on the fused silica surface to allow for more chemical bonding between the sol-gel coating and the surface of the fiber. The formation of more silanol groups on the fused silica surface can be accomplished by soaki ng the fiber in a strong base, such as 1.0 M NaOH [8-11,21,27,31,35,37-39,42,43,53,54,57,59,61,70]. After soaking in the base, the excess base is subsequentially neutra lized by soaking the fiber in 0.1 M HCl [6,8,11,17,31,39,42,43,53,58,59,61,70]. After cleaning, the fibe r is ready to be coated with the sol-gel solution. 2.4.2 Sol-gel coating procedures for SPME The sol-gel fiber coating procedure was developed by Malik and coworkers [2-4]. The fiber is coated by dipping vertically in to the sol solution fo r a specific amount of time, then removing it from the sol solution and allowing it to gel. Sometimes the dipping process is repeated several times until the desired coating thickness is obtained. In between dippings, the sol-gel fiber may be allowed to stand, heated, or dried with helium. Figure 2.2 illustrates the fiber pr etreatment and coating procedure.

PAGE 83

52 Figure 2.2 Sol-gel fiber pretreatment and coating procedure. The polyimide coating is removed from the fused silica capillary in the pretreatment step. The sol-gel coating is made by vertically dipping the bare end of the fused silica fiber into the sol solution for a period of time (typically 20-30 min). During th is time, the sol-gel material chemically bonds to the silanol groups on the fused silica surface of the fiber.

PAGE 84

53 After the coating is made, it is usua lly conditioned in a GC injection port by heating to a specific temperatur e under helium or nitrogen purge for a certain amount of time to facilitate th e sol-gel reactions to completion. The fibers may be cleaned or conditioned again prior to each use. Unless otherwise stated, all of the sol-gel SPME coatings discussed in this chapter were created following this general procedure. 2.4.3 Sol-gel polysiloxane-based mono functional coatings for SPME 2.4.3.1 Silica-based sol-gel polydimeth ylsiloxane coatings for SPME The first sol-gel SPME coating used for analytical microextraction was developed by Malik and coworkers [2-4] in 1997. A fused-silica fiber was coated with sol-gel polydimethylsiloxane (PDMS). The sol-gel PDMS coating solution was prepared by mixing 300 L of methyltrimethoxysilane (MTMS) which served as the sol-gel precursor, 180 L of hydroxy-terminated PDMS, which served as the coating polymer, 30 mg of polymethylhydrosiloxane (PMHS), whic h served as a deactiv ation reagent, and 200 L of 95% trifluoroacetic aci d, which served as a sol-gel catalyst. This mixture was vortexed for two min and centrifuged at 13,000 rp m for 5 min. The clear top sol solution was removed and used to make the coating [2-4]. The coating process was followed by an end-capping process, in which the fiber was dipped into a (4:1 v/v) trimethylmet hoxysilane/methanol solution for one min and then placed in a desiccator at room temperat ure. The sol-gel PDMS fiber was attached to a homemade SPME device and conditioned at 32 0 C for 1-2 hours under helium purge in a GC injection port. The conditioning pr ocess was then repeated, using 10 min

PAGE 85

54 conditioning cycles, until a stable GC base line was established. The sol-gel PDMS coating was found to be thermally st able to over 320 C [2-4]. GC-FID was used in the analysis of compounds extracted using the sol-gel PDMS coating. For extraction, the coated fiber was dipped in aqueous samples at room temperature. Magnetic stirring was used to de crease the equilibration time. Equilibrium was found to take from about 10 min fo r naphthalene to about one hour for N methylaniline. After the extraction, the fibe r was withdrawn into the needle of a homemade SPME syringe and immediately insert ed into the GC injection port held at 320 C for 5 min. The carrier gas flow trans ported the desorbed analytes into the GC column for separation and analysis. The deso rption step was conducte d in splitless mode, holding the GC column at a relatively low temp erature (40 C), which allowed for analyte focusing on the front end of the GC column [2-4]. Unlike traditional SPME PDMS coatings the sol-gel PDMS fiber exhibited practically no bleeding or decomposition at 320C. The sol-gel PDMS coating was successfully employed in extracting bot h polar and nonpolar compounds. Polar components (e.g., silanol groups) in the compos ition of the sol-gel network might have played a positive role in the extraction of polar compounds, such as dimethylphenol isomers, aliphatic alcohols, a nd aniline derivatives [2]. Caruso and coworkers [5] applied a sol-ge l PDMS fiber coating for the analysis of seleno amino acids using gas chromatogra phy with inductively coupled plasma mass spectrometry, with sub-ppb detection limits [5]. The method described by Malik and co -workers [2] was used by Bagheri and coworkers [6] to prepare a sol-gel SPME fibe r for the determination of dextromethorphan

PAGE 86

55 (DM) and dextrorphan (DP) in human plasma using headspace SPME with GC-MS [6]. This method was capable of detecting concentrations of 10, 50, and 500 ng/mL of DM and DP with accuracies between 99.86 and 104.5%. The observed intra-day run-to-run RSD values were between 3.29 and 4.81%. The observed inter-day run-to-run RSD values were between 3.38 and 5.04%. The au thors advocated that this method can be used to determine DM and DP in clinical plasma samples with sufficient sensitivity and reproducibility [6]. Bagheri and coworkers [7] also successfu lly used the PDMS sol-gel coating in the determination of fentanyl in human plasma by headspace SPME and GC-MS. The observed detection limit was 0.03 ng/mL. The inter-day and intra-day run-to-run RSD values were both less than 5%. This me thod was also applied successfully to the determination of fentanyl in human plasma after a volunteer applied a 50 g/h Duragesic fentanyl patch [7]. de Oliveira and coworkers [8] develope d a sol-gel coating using a thin glassceramic rod as a surface for solid phase micr oextraction. These thin glass-ceramic rod coatings were compared to fiber coatings usi ng the same sol-gel mixture. The thin, glassceramic rods were prepared by melting appropriate amounts of Li2CO3, ZrOCl2.8H2O, Ba(CH3CO2)2 and SiO2 for 3 hours at 1100C. Fused sili ca fibers were also used for comparison [8]. GC-FID was used with headspace SPME in the analysis of a BTEX sample. The sol-gel PDMS glass-ceramic-based fiber demo nstrated detection limits of 0.3 g/L for benzene, 0.7 g/L for toluene, 0.2 g/L for ethyl-benzene, and 0.3 g/L for xylene. The run-to-run RSD values were between 4.2 a nd 5.3%, and the fiber-to-fiber RSD values

PAGE 87

56 were between 6.1 and 7.2%. This glass-cera mic-based sol-gel fibe r coating produced a thicker coating and better extraction ability than the fused-silica coating [8]. Guan and coworkers [9,10] developed a sol-gel SPME fiber using poly(dimethylsiloxane) containi ng 3% vinyl group in the coa ting solution [9]. GC-FID was used for the analysis in this study. Headspace SPME was used in the extraction of BTEX compounds. Direct SPME was used for the extraction of organophosphorus pesticides in water, orange juice, and wine [9]. This coating was capable of extract ing the BTEX compounds with detection limits between 0.26 and 2.03 ng/mL with runto-run RSD values between 2.8 and 5.7%. For organophosphorus pesticides, the detec tion limits ranged from 0.4 to 19.9 ng/L in water, from 0.7 to 32.9 ng/L in orange juice, and from 0.5 to 38.2 ng/L in wine. The runto-run RSD values were between 1.0 and 27.2% in water, 5.1 and 27.6% in orange juice, and 1.1 and 19.2% in wine. In this coating, th e PDMS is entangled in the sol-gel network and also crosslinked within itself. It is ther mally stable to 290 C and has been used over 100 times without a significant decrease in coating thickness or performance [9]. Azenha and coworkers [11] develope d a sol-gel coating on an unbreakable titanium wire. A titanium wire was dipped in 1 M NaOH to promote the formation of titanol groups on the surface. The wire was then washed with 0.1M HCl, water, and methanol. The sol-gel PDMS coating was prepared as done by Malik et. al [2]. The titanium wire was attached to an SPME device. The wire was dipped into the sol solution for 20 min. The sol-gel coated titanium wi re was conditioned for 30 min at 300 C under nitrogen in a GC injection port. Headspace SPME was used to perform the extractions. The titanium sol-gel wire was exposed to the headspace at 40 C for 30 min, the time

PAGE 88

57 required for equilibrium to be established. The extracted analytes were then thermally desorbed into a GC injection port at 300 C for 1 min. This coating was capable of extracting benzene, toluene, ethylbenzen e, benzaldehyde, 2-octanone, acetophenone, and 2,5-dimethylphenol. It wa s resistant to organic solvents and thermally stable to 350 C. This fiber combined the advantages of solgel stability with the robustness of titanium wire as a new substrate. This could be a solution to the problems associated with fiber breakage [11]. Recently, the performance of th e sol-gel PDMS coated titanium wire was improved using a sol-gel/silica particle blend [12]. Recently, a novel SPME mode was developed [13]. For this, solid glass microspheres were coated with sol-gel PDMS and used to extract polycyclic aromatic hydrocarbons (PAHs). Using a homemade ther mal desorption unit, the extraction was coupled with GC-FID analysis. For PAHs the observed detection limits ranged from 0.01 to 0.045 ng/mL. The recoveries of PAHs from a river water sample ranged from 78 to 127% [13]. 2.4.3.2 Non-silica-based sol-gel polysiloxan e-based monofunctional coatings for SPME Zeng and coworkers [14] developed a so l-gel alumina-based PDMS coating for the solid phase microextraction of polar compounds. Headspace SPME was used to perform the extractions for subsequent analys is by GC-FID. This coating demonstrated run-to-run RSD values less than 5.4%, and fi ber-to-fiber RSD values less than 6.0%. This coating was also applied to the analysis of alcohols and fatty acids in beer samples, where extraction recoveries were between 85.7 and 104% with run-to-run RSD values below 9%. The sol-gel coating was found to be thermally stable to 400 C and resistant

PAGE 89

58 to high pH liquids. This coating survived 12 hours of soaking in pH 14 NaOH without significant loss in performance. Highly basic compounds and compounds in highly basic matrices can be successfully ex tracted using this sol-gel fi ber coating. The coating was used successfully in over 160 extractions and desorptions [14]. Zeng and coworkers [15] developed a so l-gel alumina and hydroxyl silicone oil fiber coating which has high extraction e fficiency of polar compounds. Zeng and coworkers [16] also developed a titania-hydr oxy-terminated silicone oil sol-gel coating for SPME of polar compounds [16]. Arom atic amines, phenols, and PAHs were extracted by headspace SPME followed by GC-FID analysis using nitrogen as the carrier gas. This titania-based sol-gel coating was f ound to be more efficient than an analogous in-house prepared sol-gel silica OH-TSO fibe r in the extraction of PAHs, amines, and phenols. It was resistant to extreme pH c onditions and was thermally stable to 320 C. This coating was applied to the analysis of aromatic amines in dye process wastewater. The detection limits ranged from 0.22 to 0.84 g/L with run-to-run RSD values between 5.9 and 7% and extraction recoveri es between 83.6 and 101.4% [16]. 2.4.4 Polysiloxane-based multifunction al sol-gel fiber coatings 2.4.4.1 PDMS/Poly(vinylalcohol) sol-gel fiber coatings In 2004, Augusto and coworker s [17] developed a sol-gel polydimethylsiloxane/poly(vinylalcohol) coated fiber for headspace SPME of polychlorinated biphenyls. GC -ECD using helium carrier ga s was used for the analysis of the extracted PCBs. This coating dem onstrated run-to-run peak area RSD values between 3.4 and 17%. This coating was th ermally stable to over 350 C with less

PAGE 90

59 degradation at high temperatur es than pure PDMS coatings. It was used successfully over 150 times without a significan t loss in performance [17]. This coating was applied to the SPME of pesticide residues in herbal infusions [18]. Chlorothlonil, met hyl parathion, malathion, -endosulfan and -endosulfan were analyzed. Headspace SPME with GC-ECD detec tion was used in this study. The sol-gel coating demonstrated detection li mits between 0.01 and 1.00 ng/mL for P. alanta infusions, 0.03 and 1.50 ng/mL for P. edulis infusions, and between 0.03 and 1.10 ng/mL for P. incarnate infusions. The run-to-run peak ar ea RSD values were all between 1.2 and 12.1%. The sol-gel coated fiber was used in more than 300 extractions over five months with no significant ch ange in performance [18]. Most recently, Augusto and coworkers [19] applied this sol-gel coating to the determination of organochlorine and organophosphorus pesticides in herbal infusions using fiber introduction mass spectrometry with detection limits between 0.3 and 3.9 ng/mL [19]. This coating was also used for the determination of pesticides in Passiflora alata infuses using HS-SPME with GC-ECD [20]. 2.4.4.2 Polymethylphenylvinylsiloxane sol-gel fiber coatings Zheng and coworkers [21] developed a sol-gel poly(methylphenylvinylsiloxane) (PMPVS) fiber coating for SPME with GC-FID analysis. In headspace SPME, the BTX analytes reached equilibrium quite fast, pr esumably taking only 60 seconds for benzene and toluene and 120 seconds for xylene at 15 C The analytes required a desorption time of 40 seconds at 150 C. The equilibrium ti me of extraction for polycyclic aromatic hydrocarbons ranged from 20 min to 240 min. Th is coating had detection limits ranging from 0.01 to 0.1 g/L for the BTX and from 0.05 to 0.29 g/L for polycyclic aromatic

PAGE 91

60 hydrocarbons. This sol-gel fiber was thermally stable to 350 C and used successfully in over 150 extractions/desorption cycl es [21]. This coating was applied to the analysis of toluene and xylenes in industrial wastewater samples from a paint mill [22]. Detection limits were between 0.01 and 0.1 g/L [22]. Recently, Zeng and coworkers [23] applied this sol-gel coating to the determinati on of methylphosphonates and phosphates in air samples, with detection limits ranging fr om 27.2 to 28.3 g/L and run-to-run RSD values between 4.7 and 6.8% [23]. Dong and coworkers [24] used a sol-gel polymethylphenylvinylsiloxane-coated fiber for the determination of organochlorine pesticides a nd their derivatio ns in water using headspace SPME with GC-ECD analysis. The observed detection limits for the organochlorine pesticides studied were be tween 0.835 and 13.0 ng/L. In lake water samples, recovery values be tween 71.5 and 115.5% were observed. The observed run-torun RSD values were all less than 11.8%. This sol-gel SPME method was adequate for the determination of organochlorine pesticides in water at ultra tra ce levels [24]. In 2006, Dong and coworkers [25] applied this so l-gel coating to the GC analysis of organophosphorus pesticides in water, with detection limits betw een 0.14 and 1.89 g/L [25] and to the GC analysis of organochlori ne pesticides in wate r using headspace SPME with detection limits below 35.3. ng/L [26]. Wu and coworkers [27] developed a so l-gel polyphenylmethylsiloxane coated fiber for the determination of organochlorine pesticides in Chinese teas using microwaveassisted SPME-GC-ECD. The observed detection limits were below 0.081 ng/L, the extraction recovery values were betw een 39.05 and 94.35%, and the run-to-run RSD values were less than 16% [27].

PAGE 92

61 2.4.4.3 C11 PDMS sol-gel fiber coating In 1998, Jinno and coworkers [28] used micro LC coupled with SPME for the analysis of benzodiazepines in human urine. Two traditional SPME fibers and a sol-gel C11 PDMS fiber were compared in performan ce. The technique required a short analysis time and very low solvent consumption [28]. 2.4.4.4 Polymethylphenylvinylsiloxane sol-gel fiber coating Zeng and coworkers [29] developed a sol-gel polymethylphenylvinylsiloxane/ hydroxyl-terminated silicone oil fiber for SPME-GCFID. Organophosphorus pesticides were anal yzed with detection limits between 1.5 and 2.1 g/L and RSD values below 7.4% [29]. 2.4.4.5 Sol-gel anilinemethyltriethoxysilane/PDMS coating for SPME Li and coworkers [30] developed a sol-gel anilinemethyltriethoxysilane/polydimet hylsiloxane coating for solid phase microextraction. For monocyclic aromatic hydrocarbons (MAHs), the observed detection limits ranged from 0.6 to 3.8 g/L and the obs erved run-to-run RSD values were between 6.9 and 13.6%. For PAHs, the observed detection limits ranged from 0.2 to 1.5 g/L and the observed run-to-run RSD values were be tween 2.8 and 9.5%. This sol-gel coating was also applied to the extraction of PAHs in river water, and the extraction recoveries ranged from 95.2 to 113% with run-to-run RSD values between 4.1 and 10.5%. The phenyl group present in this co ating makes it suitable for envi ronmental analysis of real water samples for the determination of arom atic pollutants. This coating was also

PAGE 93

62 thermally stable to 300 C and demonstr ated a lifetime of over 150 uses without significant change in performance [30]. 2.4.4.6 Sol-gel silicone polyvinylbenzene copolymeric coating for SPME Zheng and coworkers [31] developed a sol-gel-derived silicone-divinylbenzene co-polymer fiber for solid phase microextrac tion. It was capable of extracting dimethyl methylphosphonate, trimethyl phosphate, and tr ibutyl phosphate from water using headspace SPME or directly from air samples. For water samples, this coating provided detection limits and run-to-run RSD valu es of 0.34 g/mL and 5.04% for dimethyl methylphosphonate, 2.20 g/mL and 3.67% for trimethyl phosphate, and 0.01 g/mL and 6.44% for tributyl phosphate, respectively. Fo r air samples, this co ating gave detection limits and run-to-run RSD values of 1.11 g/mL and 4.29% for dimethyl methylphosphonate, 1.46 g/mL and 4.79% for trimethyl phosphate, and 1.64 g/mL and 5.88% for tributyl phosphate, respectively. This coating might be usef ul in the analysis of chemical warfare agents using SPME [31]. 2.4.4.7 Sol-gel aminopropylsilic a/PDMS coating for SPME Recently, Augusto and coworkers [32] developed sol-gel aminopropylsilica/PDMS coated SPME fibers and applied them to the analysis of beer headspace. These fibers offered superior ex traction efficienty for both polar and semipolar compounds when compared with conve ntional fibers and were capable of extracting a broad range of analyt es, including orga nic acids [32].

PAGE 94

63 2.4.4.8 Sol-gel 3-aminopropyltriethoxys ilane/PDMS coating for SPME Sol-gel 3-aminopropyltriethoxysilane/PD MS coated fibers were recently developed for the analysis of PAHs at trace concentrations [33]. This coating offered enhanced thermal stability, good fiber-to-fibe r reproducibility (RSD values lower than 6%), and low ng/L de tection limits [33]. 2.4.4.9 Sol-gel amino-functionalized PDMS coating for SPME Recently, Bagheri and coworkers [34] de veloped an amino-functionalized PDMS coating for SPME using 3-(trimethoxysilyl) Pr amine and PDMS. Chlorophenols were extracted using these coated fibers, follo wed by GC-MS analysis. The phenols were derivatized using acetic anhydr ide under alkaline conditions prior to extraction. The observed detection limits ranged from 0.02 to 0.05 ng/mL and the relative standard deviations were between 6.8 and 10%. This coating was also capable of extracting chlorophenols from spiked tap water samples, with recovery values of over 90% [34]. 2.4.5 Sol-gel coatings with cavity ligands 2.4.5.1 Crown ether sol-ge l fiber coatings Zeng and coworkers [35] developed a so l-gel-derived hydroxyl-crown ether fiber coating for SPME with GC-FID analysis. This coating demonstrated good limits of detection for phenols, ranging from 0.15 to 0.86 ng/mL and good run-to-run peak area RSD values between 2.9% and 4.6% [35]. The sol-gel crownether coating was thermally stable to 350 C and provided good de tection limits and reproducibility. Using this coating, it was determined that a wastew ater sample from a paper mill contained 1.3

PAGE 95

64 g/mL of phenol and 0.8 g/mL of 2,4-dime thylphenol, with extraction recoveries of 87.4% and 95.3%, respectively [35]. Zeng and coworkers [36] developed thr ee more sol-gel fibers based on crown ethers and applied them to the solid phase microextrac tion of monocyclic aromatic amines. The three fibers were made usi ng dihydroxydibenzo-14-crown-4 (OH-DB14C4), dihydroxy-substituted urushiol crown ether (DHSU14C4), and 3,5-dibutyl-unsymmetrydibenzo-14-crown-4 dihydroxy crown ethe r (DBUD14C4). In their study, it was determined that the sol-gel OH-DB14C4 coat ed fiber had the best affinity for aniline derivatives. This fiber demonstrated de tection limits of 0.96 ng/mL for aniline, 0.98 ng/mL for m -toluidine, 0.23 ng/mL for N N -diethylaniline, 0.17 ng/mL for N -ethylm toluidine, and 0.27 ng/mL for 3,4-dimethylaniline. The observed run-to-run RSD values were between 3.23% and 6.20% for all of th e compounds tested. This coating was applied to the analysis of a wastewater sample obtained from a pharmaceutical factory, and it was determined that the wastewater sample contained 0.7 g/mL of aniline and 0.3 g/mL of 3,4-dimethylaniline. All three of the sol-gel crown ether coatings exhibited excellent solvent resistance and therma l stability up to 340 C. The OH-DB14C4 demonstrated excellent detection limits for aromatic amine derivatives with good run-torun RSD values. No significant decline in extraction performan ce was observed even after more than 150 uses [36]. Yun [37] developed a sol-gel SPME coating using the open crown ether diallyltriethylene glycol/hydroxy-terminated s ilicone oil and vinyltriethoxysilane in the coating sol solution. Phenols reached equili brium within 40 min of extraction time and demonstrated detection limits between 0.32 and 2.6 ng/mL. Run-to-run RSD values for

PAGE 96

65 phenols were between 2.5% and 3.5%. BTX compounds reached equilibrium within 15 min of extraction time, with detection limits between 0.03 and 0.3 ng/mL and run-to-run RSD values between 2.1% and 5.2%. This coat ing was also capable of extracting high boiling compounds, such as phthalate esters. Direct SPME was used for the phthalate esters, which required between 60 and 90 min of extraction to reach equilibrium. A 10 min desorption time was used for these compounds. Detection limits ranged from 0.078 to 0.41 ng/mL for phthalate esters. Runto-run RSD values were below 10% for phthalate esters. The high temperature resist ance of this coating allowed it to extract phthalate esters successfully. Overall, this coating provi ded extractions of both polar and nonpolar analytes, using both headspace and direct SPME [37]. Wu and coworkers [38] developed a so l-gel dibenzo-18-crown-6 fiber for solid phase microextraction with a new derivatizing reagent for the determination of aliphatic amines in lake water and human urine [38]. Headspace SPME was used to extract derivatized aliphatic amine samples follo wed by GC-FID analysis. The observed detection limits for the amines were betw een 0.05 and 0.005 g/L. All run-to-run RSD values were below 6%. This sol-gel coa ting could be useful in environmental and biomedical analysis. This coating was also so lvent resistant and thermally stable to over 350 C [38]. Wu and coworkers [39] developed a be nzo-15-crown-5 sol-gel coating for SPME with GC-FID analysis. The benzo-crown ethe r sol-gel coating exhibited thermal stability to 350 C and experienced less cracking when viewed under a microscope than other solgel coatings that did not c ontain benzo-crown ether. Th is coating extracted BTEX compounds with detection limits between 0.01 and 0.05 g/L, and run-to-run RSD values

PAGE 97

66 below 4%. This coating was also capable of extracting phenols with detection limits between 0.05 and 1 g/L. This coating was used to extract 6 chlorobenzenes and 18 carcinogenic arylamines. Overall, this co ating demonstrated good selectivity towards both polar and nonpolar compounds [39]. Wu an d coworkers [40] applied this sol-gel coating to the analysis of trac e organochlorine pesticides in water, with detection limits between 0.01 and 0.5 ng/L [40]. Wu and coworkers [41] also developed new solid-phase microextraction fibers using sol-gel technology for the dete rmination of organophosphorus pesticide multiresidues in food. Allyloxy bisbenzo 16-crown-5 trimethoxysilane was synthesized and used as the sol-gel precursor. Honey, oran ge, juice, and vegetable samples were used in the SPME extraction. In honey, the de tection limits for the organophosphorus pesticides ranged from 0.004 to 0.70 ng/g w ith run-to-run RSD values between 2.1 and 15%. In juice, the detection limits for the organophosphorus pest icides ranged from 0.003 to 0.20 ng/g with run-to-run RSD values between 2.0 and 9.2%. In Pakchoi, the detection limits for the organopesticides ra nged from 0.10 to 1.0 ng/g with run-to-run RSD values between 2.3 and 9.1%. This fibe r demonstrated solven t resistance and was used over 200 times without a signifi cant decrease in performance [41]. Wu and coworkers [42] developed a solgel vinyl crown ethe r cross-linked SPME fiber for the determination of organophosporus pesticides in food samples using GC-FPD analysis. This coating demonstrated detection limits between 0.0015 and 0.081 ng/g in water samples, between 0.003 and 0.075 ng/g in apple juice, between 0.0032 and 0.09 ng/g in apple, and between 0.0042 and 0.076 ng/g in tomato. The observed run-to-run RSD values were between 1.8 and 5.3% in wa ter samples. Extrac tion recoveries of

PAGE 98

67 organophosphorus pesticides ranged from 71.5 to 104.6% in apple juice, from 70.5 to 104.9% in apple, and from 55.3 to 106.4% in toma to. The detection limits for this sol-gel coating using the described method we re below the maximum residue limits recommended by the European Union [42]. 2.4.5.2 Sol-gel calixarene coatings for SPME Zeng and coworkers [43] developed a sol-gel calix[4]arene coating for SPME with GC-FID analysis. The sol-gel calix[4]arene coating had a long lifetime, showing no significant decrease in performance after over 17 0 uses. It was thermally stable to 380 C and resistant to organic solvents. The coati ng demonstrated detection limits from 4.7 to 35.2 ng/L for BTEX compounds, from 1.2 to 51.4 ng/L for PAHs, and from 15.6 to 72.4 ng/L for aromatic amines. This coating wa s also capable of extracting PAEs through direct SPME [43]. The calix[4]arene sol-gel fiber was a pplied to the SPME of chlorophenols in deionized water, river wate r, and soil samples taken from lakes [44]. For the chlorophenols, detection limits were be tween 0.005 and 0.276 g/L. Run-to-run RSD values were under 6.8%. Rec overies of 2-chlorophenol, 2,4-dichlorophenol, and 2,4,6trichlorophenol were 94.7%, 97.4%, and 97% from river wate r, respectively, and 98.6%, 97%, and 90.2% from soil, respectively. Pe ntachlorophenol was r ecovered at 86.3% in river water and 81% in soil [44]. Zeng and coworkers [45] used a sol-gel fiber coating made with amide bridgedcalix[4]arene for the headspace SPME of aliphatic amines with GC-FID analysis. Six aliphatic amines in distilled water were anal yzed, their detection limits ranged from 0.19 to 39.51 g/L and their run-to-run RSD values ranged from 1.4 to 5.1%. This coating

PAGE 99

68 was used to determine the amount of trimethyl amine in fish samples, which accumulates as fish spoils, and can caus e cancer in humans [45]. The sol-gel calix[4]arene coating was also used for the determination of organochlorine pesticides and their metabolite s in radish samples using headspace SPME [46]. It was capable of extracting organochl orine pesticides from distilled water and radish matrix with detection limits ra nging from 0.159 to 21.7 ng/L. Run-to-run RSD values ranged between 6.83 and 13.1%. The extraction recoveries of organochlorine pesticides from radish matrix observed were between 78.39 and 112.5%. This coating demonstrated better extraction efficiency th an commercially available PDMS fibers and was considered adequate in the determination of organochlorine pesticides at ultra trace levels in complex matrices [46]. The same research group developed a sol-gel calix[4] ope n-chain crown ether fiber for the SPME-GC-FID of polar aromatic and aliphatic compounds. Detection limits were in the parts per billion and parts per trillion range, with RSD values less than 7% [47,48]. Zeng and coworkers [49] synthesized diglyc idyloxy-C[4] then used it to prepare a sol-gel diglycidyloxycalix[4]arene coating fo r SPME-GC-FID. It demonstrated detection limits between 0.07 and 1.72 ng/L for PAHs, 1.53 and 4.50 ng/L for aromatic amines, and 0.01 and 0.48 g/L for phenols. The coatin g had a lifetime of over 300 uses and was thermally stable to 350 C. Run-to-run RSD values were between 1.8 and 6.7% for phenols. It showed the potential for the analysis of trace contaminants in the environment [49]. This coating was also ap plied to the SPME of propranolol in human

PAGE 100

69 urine using GC-FID, with detection limits of 0.275 g/L in headspace SPME and 0.019 g/L in direct SPME [50]. The sol-gel diglycidyloxycalix[4]arene coating was also used to determine propranolol enantiomers in urine usi ng headspace SPME coupled with capillary electrophoresis [51]. To perfor m the extractions, 10 mL silanized amber vials were filled with 5 mL of spiked water, 1.5 g of NaCl, and a magnetic spin bar. Depending on the sample, the pH was adjusted to the appropria te value using NaOH. The vial was sealed and the fiber was exposed to the headspace at 90 C for 30 min, although the propranolol enantiomers required 50 min for equilibrium to be established. Quantification was possible under non-equilibrium conditions sinc e a linear relationship was found to exist between the amount of analyte adsorbed by th e SPME fiber and its initial concentration in the sample matrix in non-equilibrium conditions. The fiber was then removed and placed in a back-extraction device, (figure 2.3) which was in-house designed [51]. The chamber of a 100 L syringe was used to hol d the back-extraction solution. The fiber was exposed to the back-extraction solution fo r 15 min. After heating to 40 C and using ultrasonic agitation to accelerat e the back-extraction process, the fiber was removed from the syringe and the back-extraction solution was transferred by injection to a vial for CE analysis. After the back-extra ction, the fiber was conditioned in a GC injection port at 320 C for 2 min to eliminate any carryover pr oblems. CE with field amplified sample injection (FASI) was performed, with the capill ary temperature set at 25 C. Ultraviolet absorption detection was used at 214 nm for the propranolol samples [51].

PAGE 101

70 Figure 2.3 Schematic diagram of SPME and b ack-extraction procedures: (I) SPME facility; (II) back-extraction facility; and (III) back-extraction solution is transferred to an injection vial for CE analysis. (a), (b), (c), and (d) show the preparation procedures for a back-extraction chamber: (1) laboratory-made SPME syringe; (2) SPME vial; (3) facility for heating and stirring; (4) back-extraction chamber; (5) sealing septum; (6) plunger; (7) ultrasonator; and (8) 200 l vial for CE sampling [51].

PAGE 102

71 This sol-gel fiber was capable of su ccessfully extracting the propranolol enantiomers from both water and urine sample s. The observed detection limits in urine samples were 8 ng/mL for R -(+)-Propranol and 10 ng/mL for S -(-)-Propranol. The fiber demonstrated extraction recovery values between 94 and 104% for R -(+)-Propranol and between 86 and 107% for S -(-)-Propranol. The fiber was thermally stable to 350 C, resistant to high alkali cond itions, resistant to solvents, and was used over 100 times without any cracking or loss of performance. This fiber might find potential applications in drug testing. When combined with FA SI-CZE, it could be used for the routine analysis of -blockers in doping contro l laboratori es [51]. Zeng and coworkers [52] also applied the sol-gel diglycidyloxycalix[4]arene SPME coating to the analysis of chloroben zenes in soil using GC -ECD [52]. It was capable of successfully extracting nine chloro benzenes from soil samples with detection limits between 0.11 and 3.85 ng/g in kaleyard soil samples and between 0.011 and 0.200 ng/g in red clay soil samples. The observ ed run-to-run RSD values were between 4.4 and 7.6%. The recovery of chlorobenzenes from real kaleyard soil ranged from 64 to 109.6% [52]. 2.4.5.3 Sol-gel cyclodextrin coating for SPME Zeng and coworkers [53] us ed a sol-gel coating with -cyclodextrin for headspace SPME of ephedrine and methamphetamine in human urine. The detection limits were 0.33 ng/mL for ephedrine and 0.60 ng/mL fo r methamphetamine in GC-FID. The observed run-to-run RSD values were 3.9 and 5.0% for ephedrine and methamphetamine, respectively. The recoveries from human urine were 98% for ephedrine and 98.2% for methamphetamine. This sol-gel coating demons trated thermal stabil ity to 340 C without

PAGE 103

72 loss of extraction efficiency and was still st able and usable after 150 extractions. This coating showed potential in the extraction of ephedrine and methamphetamine in more difficult matrices, such as blood [53]. 2.4.5.4 Sol-gel hydroxyfullerene coating for SPME Wu and coworkers [54] developed a so l-gel hydroxyfullerene fiber for SPME. PCBs were analyzed using GC-ECD. De tection limits were between 0.013 and 0.051 ng/L for all of the PCBs analyzed. The obser ved run-to-run peak area RSD values were between 1.8% and 4.6%. This coating was used to determine the concentration of PCBs in a contaminated sediment sample. PAHs were analyzed by GC-FID. The observed detection limits for PAHs were between 0.0049 to 0.125 ng/mL with run-to-run peak area RSD values between 1.9% and 8.9%. The sol-ge l fullerene coating was applied to the analysis of a wastewater sample containi ng PAHs and was even capable of extracting aromatic amines [54]. The sol-gel hydroxyfullerene fiber was th ermally stable to 360 C. This was significant because it extended the use of SPM E to the analysis of compounds with higher boiling points. The fiber was also resi stant to polar and nonpolar solvents and was used over 190 times without a si gnificant loss in pe rformance [54]. This fiber was used in the determination of the leaching of phthalic diesters from polyvinyl chloride toys dipped in simulated saliva using SPME-GC. The detection limits ranged from 0.079 to 1.36 g/L, the extraction recovery values were from 88.4 to 107.7%, and the run-to-run peak area RSD values were all less than 8% [55].

PAGE 104

73 2.4.5.5 Sol-gel quinoxaline-bridged cavitand coating for SPME Recently, a quinoxaline-bridged cavitand coated fiber was developed for the SPME of benzene and chlorobenzenes [56]. Th ese fibers demonstrated excellent thermal stability and good batch-to-batch reproducibility with RSD values less than 6%. These coated fibers were capable of sub-ng/L dete ction limits. They were also applied to the extracting of chlorobenzene in river water samples with recovery values between 87.4 and 94.7% [56]. 2.4.6 Non-polysiloxane sol-gel coatings 2.4.6.1 Sol-gel polyethylene glycol coatings for SPME Wu and coworkers [57] developed highperformance polyethylene glycol-coated solid-phase microextraction fibers using so l-gel technology. First, BTEX was extracted and analyzed using the sol-gel PEG coati ng. The equilibrium tim e of extraction was found to be very short, approximately 30 sec onds for benzene and to luene, 40 seconds for ethylbenzene and p -xylene, and 90 seconds for o -xylene. Commercial PDMS fibers normally take much longer to reach equilibration. Analytes can be thermally desorbed within 20 seconds at 280 C. The short extraction a nd desorption times were attributed to the porous structure of the sol-gel PEG fibe r. Sample carryover and bleeding at high temperatures were not observed. This fiber coating remained stable and reusable after more than 150 desorptions at 280 C. This coating demonstrated detection limits of 50 pg/mL for benzene, 20 pg/mL for toluene, 20 pg/mL for ethylbenzene, 10 pg/mL for p xylene, and 10 pg/mL for o -xylene in GC-FID [57]. Phe nols were also extracted on the sol-gel PEG coating. Detection limits were 10 ng/mL for phenol, 0.5 ng/mL for 2-

PAGE 105

74 chlorophenol, 5 ng/mL for 2-nitrophenol 1 ng/mL for 2,4-dimethylphenol, and 0.1 ng/mL for 2,4-dichlorophenol and 2,4,6-trichl orophenol. The sol-gel PEG coating was also effective in the extracti on of phthalic diesters, naphthale ne congeners, and pesticides [57]. Augusto and coworkers [58] developed a highly porous sol-gel SPME fiber based on poly(ethyleneglycol)-modified ormosils. For the more volatile compounds tested, such as benzene and toluene, the sol-ge l Carbowax 20M fiber was comparable to commercial PDMS and Carbowax-DVB counter parts, but for heavier compounds, such as o -xylene, the sol-gel fiber was able to extract 230 to 540% of the mass of the commercial fibers [58]. This sol-gel coating is notably more porous than most other coatings, which provides faster equilibrium and desorption times. Carbowax 20M and related polymers are often used as additives to improve poros ity in other sol-gel SPME coatings. This coating was also applied to th e screening of contaminants released by plastic containers in microwave ovens and in the monitoring of contamination of ground water by fuel leakage from gas stations [58]. 2.4.6.2 Sol-gel acrylate c oatings for SPME Zeng and coworkers [59] developed a so l-gel derived 3-(tri methoxysilyl)propyl methacrylate (TMSPMA) hydroxyl-terminated s ilicone oil solid phase microextraction fiber for the headspace SPME of aroma compou nds in beer. This sol-gel coating was capable of simultaneously extracting analytes of diverse polarity (alcohols, esters, and fatty acids) from beer samples by headspace SPME. The detection limits were calculated based on the analysis of these compounds in vol atile-free beer samples. In volatile free-

PAGE 106

75 beer, the sol-gel coating ha d detection limits between 0.01 and 10.4 g/L for alcohols, 0.01 and 35.2 g/L for fatty acids, and 0.01 a nd 16.4 g/L for esters. The observed runto-run peak area RSD values were between 1.68 and 3.66% for alcohols, 3.00 and 6.18% for fatty acids, and 1.98 and 5.02% for esters. This method was used to determine volatile compounds in four beer types. The major aroma contributing substances in the beer were identified based on their odor description and odor thresholds [59]. Zeng and coworkers [60] also develo ped a sol-gel hydroxyl-terminated silicone oil-butyl methacrylate–divinylbenz ene (OH-TSO-BMA-DVB) copolymer methacryloxypropyltrimethoxysilane (KH-570) coating for the SPME of volatile compounds in red wine. This coating was capab le of extracting alcohols, esters, and fatty acids from wine samples. In volatile-free wine samples, the sol-gel coating provided detection limits between 0.0494 and 49.1 g/L for alcohols, 0.0504 and 0.523 g/L for fatty acids, and 0.0108 and 66.2 g/L for esters The observed run-to-run RSD values were between 0.22 and 3.59% for alcohols, 6.81 and 7.30% for fatty acids, and 1.73 and 5.16% for esters. This method was used to determine volatile compounds in store-bought wine samples without significant matrix effects [60]. Zeng and coworkers [61] developed a solgel acrylate/silicone co-polymer coating for the headspace SPME of 2-chloroethyl ethyl sulfide in soil. This sol-gel fiber coating was capable of extracting 2-ch loroethyl ethyl sulfide with a detection limit of 2.7 ng/g. The run-to-run RSD value for the extraction wa s 2.19%. This coating was also used to recover 2-chloroethyl ethyl sulfide from red cl ay, sandy soil, and agricultural soil. The observed extraction recoveries were 88.06% in agricultural soil, 92.61% in red clay, and 101.95% in sandy soil. This coating demonstr ated thermal stability to 350 C and

PAGE 107

76 excellent solvent stability si nce no appreciable change in performance was observed after 2 hours of soaking in methylene chloride acetonitrile, and a cetone, and 12 hours of soaking in distilled water. This coating wa s still stable after 150 extraction cycles. Since it is capable of successfully extracting 2-chloro ethyl ethyl sulfide, this coating might be suitable for the extraction of sulfur mustard agent, which is used in chemical weapons, due to its similarity in structure [61]. Zeng and coworkers [62] used a sol-ge l butylmethacrylate/silicone coating for SPME coupled with CE to determine ephedrine derivatives in water and urine. The analytes were desorbed by immersing the fibe r into a back-extracti on device (figure 2.3) [51]. The back extraction solution was then transferred to an injection vial for CE analysis. The sol-gel fiber was prepared fo r the next extraction by conditioning at 260 C for 5 min. Field amplification was performed to enhance the detection limits in CE. The observed detection limits were betwee n 0.003 and 0.005 g/mL for the ephedrine derivatives. The observed run-to-run RS D values were between 4.96 and 7.57% and extraction recoveries between 88.7 and 98.6%. This sol-gel fiber and method can detect ephedrine derivatives in huma n urine samples, despite natu ral variations in the urine matrix, with run-to-run RSD values between 4.38 and 7.76%. The fiber demonstrated high solvent stability, high extraction efficiency and a long lifetime. It is suitable for SPME with CE or HPLC analysis [62]. This coating was applied to the extraction of medium and long chain fatty acids and to the analysis of fatty acids in lung ti ssues using GC-MS [63]. Detection limits, runto-run RSD values, and extraction recoveries for the fatty acids ranged from 0.51 to 2000 g/L, 3.33 to 13.33%, and 76.35 to 107%, respectively. It was determined that lung

PAGE 108

77 tissue mostly contained C16:0, C18:0, and C24:0 fatty acids, with smaller amounts of other saturated and unsaturated fatty acids. Highe r levels of saturated fatty acids (0.01-66.60 g/mg) and lower levels of unsaturated fatty acids (<0.005-3.95 g /mg) were found in cancerous lung tissue co mpared to normal lung tissue (<0.005-63.74 g/mg for saturated fatty acids and <0.005-9.22 g/mg for unsaturated fatty acids). This study demonstrates that lung cancer could be poten tially diagnosed by the SPME analysis of fatty acids in lung tissue. The major objective of this work was to provide an approach to investigate whether fatty acids have any relationshi p with lung cancer, which could provide potentially useful information for preventing and curing lung cancer [63]. 2.4.6.3 Other silica-based non-polys iloxane sol-gel SPME coatings Caruso and coworkers [64] prepared se veral different sol-gel fiber coatings consisting of different molar ratios of n -octyltriethoxysilane (C8-TEOS) and methyltrimethoxysilane (MTMS). It was found that the 1:1 molar ratio of C8-TEOS : MTMS demonstrated both the best extraction capability and run-to-run repeatability [64]. HPLC was used for the analysis of or ganotin, organoarsenic, and organomercury compounds extracted using the sol-gel octyl-me thyl coating. After extraction, the fiber was rinsed in distilled wa ter and placed in the 200 L desorption chamber containing 1.0 g/mL benzoic acid dissolved in the mobile phase (80% ac etonitrile and 20% water). Static desorption was performed for five min. The desorption solvent was then directly injected into the HPLC [64]. Such a desorpti on process results in the loss of analytes and decreases the extraction sens itivity. Sol-gel capillary microextraction, which will be discussed later, can overcome these probl ems with on-line hyphenation with HPLC.

PAGE 109

78 Diphenylmercury, trimethylphenyltin, and tripheny larsine extractions were tested with these sol-gel fibers. The fiber-to-fiber (n=5 ) relative standard deviations observed for these compounds were 29% for triphenylarsine, 22% for trimethylphenyltin, and 21% for diphenylmercury. These large RSD values are due to problems with the uniformity of the coatings. This sol-gel octyl-methyl coa ting demonstrated detection limits of 80 g/L for triphenylarsine, 412 g/L for diphenylmercury and 647 g/L for trimethylphenyltin. The sol-gel octyl-methyl coating was resistant to solvents and solutions of 1% trifluoroacetic acid and 1% NaOH for periods of up to 16 hours [64]. Giardina and Olesik [65-67] developed low temperature glassy carbon films for SPME using a sol-gel process. The films were prepared by coating silica particles with a diethynyl oligomer precursor and heat curing at 300-1000 C. Using a sol-gel process, the coated particles were then immobilized onto stainless steel fibers, which were subsequently used for headspace and li quid extractions. Aromatic hydrocarbons, geosmin, 2-methylisoborneol, and 2,4,6-trichloroa nisole were efficiently extracted from water samples using these films [65]. Teng and Chen [68] coupled SPME with sol-gel-assisted laser desorption/ionization time-of-flight mass spectro metry. An optical fiber with a sol-gel derived 2,5-dihydroxybenzoic acid film was used to extract trace organics from aqueous samples. The fiber was then inserted into the mass spectrometer for the analysis of the extracted compounds. This method easily coupled SPME with laser desorption/ionization mass spectrometry [68]. Azenha and coworkers [69] developed a sol-gel ultrathin phenyl-functionalized fiber coating for SPME. These coatings were thermally stable to 320 C and were able to

PAGE 110

79 withstand 150 extraction cycles involving solvent and thermal treatments. The sol-gel coatings were most efficient in the extr action of long chain and aromatic compounds. The most successful sol-gel coating was the one prepared using a sol solution characterized by an MTMS/PTMOS ratio of 4:1, a water/precurso r ratio of 2:1, and NaOH as the catalyst. This sol-gel coating was able to extract the compounds in amounts similar to, or higher than, a commercial PDMS fiber. This sol-gel coating was also able to extract long aliphatic chain compounds bett er than a commercial CW/DVB fiber, but the extraction of polar and aromatic com pounds was better with the CW/DVB fiber. Observed run-to-run peak area RSD va lues were between 25 and 30% [69]. Lee and coworkers [70] synthesized h ydrophilic oligomers and developed sol-gel oligomer coatings for SPME coupled to GC-M S analysis. For organochlorine pesticides, triazine herbicides, estrogens, alkylphenol s and bisphenyl A this coating was found to have similar or considerably better extrac tion efficiencies comp ared to commercial SPME fibers. The extraction efficiencies for tr iazine herbicides were significantly better on this coating than on commercial coa tings. Detection limits between 0.001 and 0.005 g/L were observed, with run-to-run peak area RSD values between 5.0 and 11.0%. Natural water from a reservoir was also used as a matrix for the triazine herbicides. Extraction recoveries from the natural water were higher than 84% with run-to-run peak area RSD values less than 11%. The efficiency of extraction was not significantly altered by matrix effects [70]. Wang [71] developed sol-gel SE-54 fibers for solid phase microextraction. These fibers demonstrated temperature stability to 310 C and run-to-run peak area RSD values less than 2.7% in GC analysis [71].

PAGE 111

80 Recently, sol-gel carbon nanotube-containi ng SPME fibers were developed [72]. A stainless steel fiber was coated with a so l solution which contai ned carbon nanotubes. These fibers demonstrated high temperature resistance, mechanical strength, and a long service life [72]. 2.4.6.4 Other non-silica-based non-poly siloxane sol-gel fiber coatings Wei, et al.[73] developed a ti tania-based sol-gel dimethyl-3,7diaminobenzothiophene-5,5-diox ide-3,3,4,4,-diphenylsulfone tetr acarboxylic dianhydride fiber for the efficient determination of BT EXs and low-polar halocarbons using SPMEGC-FID [73]. Recently, a sol-gel titania pol y(ethylene glycol) coating was applied to anodized aluminum fibers and used for h eadspace SPME of aromatic hydrocarbons from water samples [73]. Lee and Xu [75] developed a zirconia ho llow fiber and applie d it to solid phase microextraction. A polypropylene hollow fibe r was used as the template. It was ultrasonically cleaned in acetone for 5 min. The fiber was removed from acetone and air dried. The sol solution was prepared by dissolving 6 g of ZrOCl2:8H2O into 30 mL of ethanol/water (5/3, v/v). Th e mixture was sonicated for 30 min. The fiber was immersed into the sol solution for one hour, then dried at 393 K for 2 hours. This immersion and drying process was repeated several times. Finally, the fiber was heated from room temperature to 873 K at 5 K/min and maintained at the highest temperature for 4 hours. This removed the template and crystallized the zirconia. The fiber was not installed into an SPME syringe [75]. A cross-sectional SE M image of the sol-gel zirconia fiber is shown in figure 2.4 [75].

PAGE 112

81 Figure 2.4 A cross-sectional scanni ng electron microscopic image of a zirconia hollow fiber [75].

PAGE 113

82 HPLC with mass spectrometry was us ed in the analysis of pinacolyl methylphosphonic acid, a nerve agent degredati on product of a chemical warfare agent. The pH of the sample solution was adjusted to 6.5. The extraction was performed by immersing the zirconia fiber into 10 mL of sa mple solution in a 12 mL vial. A shaker was used to facilitate the extraction process. The extractions were performed for 30 min, even though this time was not enough to reach equilibrium. After extraction, the fiber was removed and placed in ultrapure water to remove any surface contaminants. The fiber was dried and placed in 200 L of 1% ammonia solution for desorption, which took 30 min. A 20 L volume of this solution was directly injected into the LC-MS system for analysis. After each use, the zirconia fi ber was washed with concentrated ammonia solution for 10 min to remove any possible residu al analyte. The fiber can be used again with no carryover effects. This fiber was capable of extracting pinacolyl methylphosphonic acid with a detection limit of 0.07 ng/mL and a run-to-run peak area RSD value of 3.7%. A lake water sample was also analyzed, and the recovery of pinacolyl methylphosphonic acid was 94.2% [75]. This sol-gel zirconia fiber represents a new configuration in microextraction and may be suitable for the determination of chemical warfare agent residues in environm ental samples, as well as phosphorylated peptides and proteins in biological samples [75].

PAGE 114

83 2.5 Sol-gel materials in in-tube SPME (Capillary Microextraction (CME)) 2.5.1 Introduction In sol-gel SPME, the sample capacity is still limited by the relatively short segment of the fiber that is coated with the so l-gel sorbent. This limits the sensitivity of sol-gel SPME. Also, since sol-gel SPME coatin gs are coated on the outside of the fibers, mechanical damage is still possible. Furthermore, it is difficult to hyphenate sol-gel SPME to HPLC and CE applications, as a spec ial desorption device is needed [51], which results in sample loss. Sol-gel capillary microextraction (CME) wa s developed in order to address these problems [76,77]. In sol-gel CME, the sol-ge l coating is chemica lly bonded to the inner surface of a fused silica capilla ry. Since the sol-gel coating is chemically bonded to the inner surface of the fused silica capillary, sol-gel capillary microextraction offers increased solvent and thermal stability when compared to traditional in-tube SPME. Since a longer segment of capillary can be coated, this increases the sample capacity and sensitivity [76]. The coating is also protected from mechani cal damage and can easily be hyphenated to HPLC [79,90,91,93] and CE [81,83]. 2.5.2 Capillary pretreatment procedures The inner surface of a fused silica capill ary contains a small concentration of silanol groups. The chemical bonding of a solgel coating to the inne r surface of a fused silica capillary is accomplis hed through condensation react ions between silanol, alkoxy, and other sol-gel active groups in the sol-gel network a nd silanol groups on the inner surface of the fused silica capillary. Theref ore, like in sol-gel fiber SPME, it would be

PAGE 115

84 advantageous to have more silanol groups on the fused silica surf ace to allow for more chemical bonding between the sol-gel coating an d the surface of the capillary. This is usually accomplished through hydrothermal pr etreatment. This pretreatment is conducted by first cleaning the fused sili ca capillary by rinsing sequentially with methylene chloride, methanol, and deionized wa ter. Next, the ends are typically sealed with an oxyacetylene torch and the capillary is placed in a GC oven and heated between 250 C and 350 C for approximately 2 hours. Th e ends are then cut and the capillary is ready to be coated [76,81,83,85,87,91,93,94,96]. In some cases [89,95], the formation of more silanol groups on the fused silica surface has also been accomplished by rinsing sequentially with 1 M NaOH for 2 hours, water for 30 min, 1 M HCl for 2 hours, and water for 30 min. After cleaning, the fiber is r eady to be coated with the sol-gel material. The capillary is usually coated by filling it with the sol solution using either a gas pressure-operated filling system [97] or a syri nge. The sol solution is allowed to stand in the capillary for an appropriate amount of time (typically 2030 min), followed by purging. The coated capillary is subsequen tly conditioned in a GC oven under helium flow [76,81,83,85,87,91,93,94,96] or in a muffle furnace [89,95] by heating to a specific temperature for a certain amount of time to f acilitate completion of the sol-gel reactions. The coated capillary may be cleaned or conditioned again prior to each use. 2.5.3 Silica-based sol-gel coatings in CME 2.5.3.1 Polysiloxane-based monofunctional sol-gel capillary coatings Malik and coworkers [76,77] introduced sol-gel capillary mi croextraction as a solventless extraction technique for the precon centration of trace analytes. This was the

PAGE 116

85 first report on the use of sol-gel coated capil laries in analytical microextraction. A solgel polydimethylsiloxane (PDMS) coated capil lary was prepared in this study. The solgel PDMS solution used to make the capillary coating was prepared in the same manner as the sol-gel fiber PDMS coating [2]. A 250 m I.D. fused silica capillary was previously cleaned and hydrothermally treate d, then filled with the sol solution using a homemade helium-pressure operated filling/purging device. A sol-gel network evolved in the sol solution within the fused-silica capilla ry and a thin layer of the sol-gel material chemically bonded to the inner all of the cap illary through condensati on with the silanol groups on the capillary walls. After this ti me period, the remaini ng sol-gel solution was expelled from the capillary under helium pressure. Helium was purged through the capillary for an additional hour. The capillary was then placed in a GC oven and heated from 40 C to 350 C a rate of 1 C/min, holding it at the final temperature for 5 hours under helium purge. The final conditioning temperature was determined by the thermal stability of each sol-gel coating. The capilla ry was subsequently rinsed with methylene chloride and methanol and reheated with th e same temperature program in a GC, except that it was held at the final temp erature only for 30 min [76]. Extraction was carried out using gravity-f ed sample dispensers (figure 2.5) [76] made from modification of a Chromaflex AQ column in which the glass was deactivated by rinsing with 5% HMDS in methylene chlori de and heating to 250 C. For extraction, a previously conditioned sol-gel capillary was attached to the bottom end of the sample dispenser. The sample to be analyzed (25 mL) was added to the di spenser and allowed to drip through the capill ary for 30 min, the time required to reach equilibrium. After extraction, the capillary was inst alled inside the GC injector with its lower end connected

PAGE 117

86 to the inlet of the GC column via a two-wa y press-fit quartz connector. For this, the quartz wool was removed from the glass inse rt to accommodate the deactivated two-way fused silica connector within the insert. Be fore installing the micr oextraction capillary, the GC injector was first cooled down to r oom temperature. To connect the capillary with the GC column, the column nut at the bottom of the injector was loosened and the column was slid up through the top of the in jector. The extraction capillary and the GC column were press-fitted into the fused silica connector from the top and bottom, respectively. The column was pulled back down until the microextraction capillary disappeared below the septum support of the in jector. Finally, the se ptum was reinstalled and the injector nut and capi llary column nut were tighten ed. The extracted analytes were then thermally desorbed through heating the injection port from 30 C to 330 C at a 100 C/min ramp rate for 5 min. The column te mperature was held at 30 C to allow for analyte focusing. The GC analysis was perf ormed with the split remaining closed, using helium flow, and flame ionization detection [76].

PAGE 118

87 Figure 2.5 A schematic of a gravity-fed sample dispensing unit used for capillary microextraction [94].

PAGE 119

88 PAHs, aldehydes, and ketones were extr acted using the sol-gel PDMS capillary. Polycyclic aromatic hydrocarbons demonstrated excellent detection limits of 0.31 ppt for naphthalene to 0.94 ppt for phenanthrene. Aldehydes demonstrated detection limits between 28.36 ppt for n -decylaldehyde to 103.20 ppt for benzaldehyde. Ketones demonstrated detection limits between 32.67 ppt for anthraquinone to 215.7 ppt for valerophenone. The compounds tested had exce llent run-to-run peak area RSD values between 2.5% and 6% [76]. Sol-gel capillary microextraction offers th e advantage of creating thicker coatings than those used in in-tube SPME where a shor t segment of GC capillary column is used for extraction. These thicker coatings allow for increased sensitivity and detection limits in the part per quadrillion range Sol-gel coatings are also chemically bonded to the inner walls of the capillary, which allows for greater thermal stability and solvent resistance. Because of their excellent solv ent resistance, sol-gel capillary coatings are suitable for use in HPLC. Also, coating the inner surface of the capillary allows for a more uniform coating compared to sol-gel coated fibe rs. The development of sol-gel capillary microextraction promoted the creation of many different types of sol-gel coated capillaries, with different ex traction capabilities and appli cations, many of which will be discussed in this chapter [76]. 2.5.3.2 Polysiloxane-based multifunctional sol-gel coating in CME Malik and coworkers [78] deve loped a sol-gel immobilized cyanopolydimethylsiloxane coating for capillary mi croextraction of aqueous trace analytes ranging from polycyclic aromatic hydrocar bons to free fatty aci ds using GC-FID analysis. The detection limits ranged from 2.9 to 4.1 ng/L for PAHs, 12.0 to 22.4 ng/L

PAGE 120

89 for aldehydes, 2.3 to 7.0 ng/L for ketones, 10.4 and 114.9 ng/L for aromatic amines, 39.1 and 161.5 ng/L for phenols, 1.4 and 60.3 ng/L fo r alcohols, and 4.4 and 5.9% for free fatty acids, with run-to-run peak area RSD values between 1.3 and 5.3% and capillary-tocapillary RSD values (for peak area) between 1.8 and 8.7% for all analytes. The sol-gel CN-PDMS coating demonstrated excellent repe atability and low ng/L detection limits for both polar and nonpolar analytes from aqueous media without the need for derivitization, pH adjustment, or salting out procedures. This coating was found to be highly resistant to solvents and thermally stabile to 330 C [78]. Recently, Segro and Malik [79] developed a sol-gel polydimethyldiphenylsiloxane (PDMDPS) co ating for on-line hyphenation of capillary microextraction with HPLC using UV detect ion. A 40 cm long segment of 0.25 mm I.D. fused silica capillary was coated with th e sol-gel PDMDPS coating, provided with the appropriate fittings, and installed as an extern al sampling loop in an HPLC injection port. The bottom of a vertically held gravity-fed sa mple dispenser was fitted with a deactivated fused-silica capillary (520 m i.d.) using a ppropriate ferrules and connections. The other end of this capillary was fitted with a syri nge needle. The gravity-fed dispenser was filled with the sample solution containing trace amounts of PAHs, aromatic compounds, ketones, and aldehydes. The HPLC system was placed in the load position and the needle was inserted into the HPLC inject ion port and the sample was allowed to pass through the sol-gel capillary for the time required to reach equilibrium between the solution and the sol-gel coating. The samples were desorbed from the sol-gel capillary by putting the HPLC injection valv e into the inject position. The injected analytes were then separated on a commercial C-18 column using isocratic or gradient elution with

PAGE 121

90 acetonitrile/water mobile phases. This so l-gel coating was capable of efficiently extracting these analytes with low ng/L de tection limits, good run-to-run reproducibility, and capillary-to-capillary reproduc ibility. The test re sults also indicate that the sol-gel PDMDPS coated capillary is resistant to high temperature solvents. This makes the solgel PDMDPS coated capillary suitable for appli cations in high-tempe rature HPLC [79]. 2.5.3.3 Sol-gel coating with cavity ligands for CME Feng and coworkers [80] determined nonsteroidal anti-inflammatory drugs in urine samples using a sol-gel -cyclodextrin coated capillary for in-tube SPME coupled to HPLC. A sample loop was replaced with the sol-gel coated extraction capillary. A schematic of the in-tube SPME-HPLC system used is presented in figure 2.6 [80]. Before extraction, the analytical HPLC colu mn was equilibrated with the mobile phase using one pump. The mobile phase consisted of 70% methanol and 30% acetate buffer solution at pH 5.0. Urine samples were co llected and centrifuged at 4,500 rpm for 10 min to remove precipitates. The supernat ant of the urine was diluted 10 times with double distilled water and spiked with anal ytes. The sample solution (250 L) was pumped through the capillary at a flow rate of 300 L/min. This method required less than 1 min for the extraction pr ocess. After extraction, the injection valve was returned to the load position and desorption was perf ormed by allowing the mobile phase to pass through the capillary. The desorbed analytes were transferred to the in let of an analytical column for separation followed by UV detecti on at 223 nm. Ketoprofen (KEP), fenbufen (FEP), and ibuprofen (IBP) were extracted a nd analyzed with detection limits of 38, 18, and 28 ng/mL, respectively. The intra-day and inter-day peak area RSD values were less than 4.9 and 6.9%, respectively. This sol-gel capillary was used in over 250 extractions

PAGE 122

91 without a decrease in extraction efficiency. This sol-gel capillary and method were convenient for common laboratory use. This cap illary can also be used in the in-tube SPME of other biological samples a nd environmental pollutants [80].

PAGE 123

92 Figure 2.6 A schematic diagram of the experi mental set-up used for in-tube SPMEHPLC analysis [80].

PAGE 124

93 2.5.3.4 Electrically charged sol-gel coatings for CME Malik and coworkers [81,82] developed pos itively charged sol-gel coatings for on-line preconcentration of amino acids in ca pillary electrophoresis. The capillary was first sealed on one end using an oxyacetylene torch. It was then filled with the sol solution from the open end using a homemade helium pressure-operated filling device. Gas was trapped at the sealed end of the capillary, keeping the surface near the sealed end untouched by the sol solution. The sol solution was kept in the capillary for 20 min, then released from the capillary by the compressed gas pocket when the external pressure was removed. The open end was sealed and the capillary was conditioned at 150 C for 2 hours in a GC oven. The sealed ends were then cut open, and the capillary was purged with helium for 30 min, rinsed with 100% acetonitrile, deionized water, and the appropriate running buffer. The polyimide coating was burned off the outside of the capillary in the uncoated section to create an optical window for in-line UV detection [81]. In-line extraction and preconcentr ation was conducted in a capillary electrophoretic setup. This was accomplished by installing the sol-gel coated capillary on the CE system and filling it with running buffer. The pH of the sample solution was maintained above the p I of the amino acid in the sample to impart a negative charge on the amino acid. Next, the inlet end of the column was inserted into the sample vial and the sample was hydrodynamically injected for 3 min at 100 psi. This is an extended injection time, which allowed for more th an one column volume of sample to pass through the electrically ch arged sol-gel capillary whic h increased the extraction sensitivity. Electrostatic interaction betw een the positively charged sol-gel coating and

PAGE 125

94 the negatively charged amino acid molecules allowed the extraction to occur. The sample solution was then removed from the column by purging with deionized water or by reversed electroosmotic flow. The use of reversed electroosmo tic flow resulted in lower detection limits. The in let of the capillary was return ed to the buffer solution, and a high electric field was applied to perform the CE analysis [81]. This was the first report of an on-column extraction and preconcentration effect in capillary zone electrophoresis using a positively charged sol-gel column. The on-line extraction and preconcentration capabilities of the sol-gel coated column dramatically improved the detection limits of the amino acids using capillary electrophoresis. The observed detection limits were 139 nM for al anine, 98 nM for asparagine, 141 nM for phenylalanine, and 115 nM for tryptophan. A 150,000-fold enrichment effect for alanine was obtained using this solgel column and method. This method does not limit the injection volume or require modifi cations of the CE system [81]. Malik and coworkers [82,83] also devel oped a negatively charged sol-gel coated capillary with stable electroosmotic flow for on-line preconcentration of zwitterionic biomolecules in capillary electromigration sepa rations. Myoglobin and asparagine were preconcentrated using the negati vely charged sol-gel column and analyzed using CE with UV detection at 214 nm. For myoglobin, th e 74% MPTMS column demonstrated the best sensitivity enhancement factor of 973 by peak height and 3104 by corrected peak area. For asparagine, the 60% MPTMS column demonstrated a sens itivity enhancement factor of 4450 by peak height, a nd 7335 by corrected peak area [83].

PAGE 126

95 2.5.3.5 Non-polysiloxane-based sol-gel coatings for CME Malik and coworkers [76] developed a so l-gel polyethylenegl ycol (PEG) coating for CME-GC-FID analysis. Phenols, alcohols, and amines were extracted using the solgel PEG capillary. Phenols demonstrated detection limits between 6.0 ppt for 2,3dimethylphenol and 16.1 ppt for 2,6-dimethyl phenol. Alcohols and amines provided detection limits between 2.0 ppt for myristyl alcohol and 6.0 ppt for benzanilide. Recently, Malik and coworkers [84] deve loped a sol-gel immobilized short-chain poly(ethylene glycol) coati ng for capillary microextrac tion of underivatized polar analytes. GC-FID with helium carrier gas was used for the separation of extracted analytes. An in-house prepared sol-gel PD MS capillary GC column was used in the separation of the extracted analytes. The so l-gel short-chain PEG coating was used to extract polar analytes, including aldehydes, ke tones, aromatic amines, phenols, alcohols, and acids, for which detection limits were between 11.8 and 20.4 ng/L, 7.2 and 119 ng/L, 5.2 and 398.2 ng/L, 23.3 and 158.6 ng/L, 8.1 and 133.8 ng/L, 19.7 and 67.8 ng/L, respectively. For the same classes of analyt es, the run-to-run peak area RSD values were between 3.0 and 4.1%, 2.9 and 4.1%, 2.7 and 4.9% 2.6 to 4.3%, 2.5 to 3.9%, and 1.9 to 3.7%, respectively. This coating was thermally stable to 340 C. The performance of the sol-gel PEG coated capillary in CME remained practically unchanged after rinsing it with 50 mL of dichloromethane/methanol mixture (1 :1, v/v) over a 24 h period. This was the first report on the use of low-molecular-w eight PEG to prepare sol-gel coatings for microextraction capi llaries [84]. Malik and coworkers [85,86] developed a sol-gel dendrimer coating for capillary microextraction (figure 2.7). In-house prepar ed sol-gel PDMS and sol-gel PEG capillary

PAGE 127

96 GC columns were used in the separation of th e extracted analytes. The sol-gel dendrimer coating was used to extract analytes belonging to various chemical cl asses with detection limits between 2.1 and 19.4 ppt for PAHs, 3.3 and 44.3 ppt for aldehydes, 1.9 and 15.2 ppt for ketones, 9.88 and 16.03 for alcohols, and 10.53 and 12.33 ppt for phenols. PAHs, aldehydes, ketones, alcohols, and phenols de monstrated run-to-run peak area RSD values between 2.04 and 9.20%, 3.70 and 8.97%, 2.08 and 6.45%, 2.53 and 7.59%, and 2.50 and 6.18%, respectively. This sol-gel dendrimer capil lary was capable of low part per trillion detection limits for both polar and non-polar analytes. This coating demonstrated thermal and solvent stability, making it suitable for GC and HPLC analysis [85].

PAGE 128

97 Figure 2.7 The structure of the surface-bonded sol-gel dendrimer coating [85].

PAGE 129

98 Malik and coworkers [87,88] developed a sol-gel poly(tetrahydrofuran) (polyTHF) coating for high-sensitivity sample preconcentration using capillary microextraction. This coating was capab le of extracting polycyclic aromatic hydrocarbons (PAHs), aldehydes, ketones, phenol s, and alcohols, for which the detection limits were between 260 and 750 parts per quadrillion (ppq), 625 and 1000 ppq, 340 and 1000 ppq, 18 and 150 ppt, and 0.59 and 13 ppt, resp ectively. The capillary-to-capillary peak area RSD values for PAHs, aldehydes, ke tones, phenols, and alcohols were between 2.13 and 6.45%, 4.01 and 10.31%, 3.02 and 8.01%, 3.72 and 8.74%, and 1.21 and 11.75%, respectively. For the same classes of analytes of analytes, the run-to-run peak area RSD values were between 1.07 and 5.05%, 2.19 and 7.48%, 2.03 and 8.36%, 2.21 and 7.32%, and 2.21 and 6.78%, respectively. This coating was very successful in the extraction of highly polar, moderately polar and nonpolar compounds. It can even be used in the simultaneous extraction of polar moderately polar, and nonpolar compounds. This coating is thermally stable to 320 C and resistant to solv ents. It could be successfully coupled to HPLC and CEC analysis [87]. Hu and Zheng [89] developed a so l-gel 3-mercaptopropyltrimethoxysilane modified silica coating for cap illary microextraction on-line hyphenated with inductively coupled plasma atomic emission spectrometry fo r the determination of Cu, Hg, and Pb in biological samples. Inductively coupled pl asma mass spectrometry was used to perform the analysis in hyphenation with CME. The sol-gel MPTMS capillary was connected to the system. The flow rate used was 0 .25 mL/min. The aqueous sample solution containing Cu, Hg, and Pb (with pH previous ly adjusted to 6.0 with 0.01 M NaAc and 0.1 M HCl) was passed through the so l-gel capillary and into a wast e container. The analytes

PAGE 130

99 were adsorbed into the sol-ge l coating during this step. Fo r the elution, a 0.1 M HCl with 4% thiourea (m/v) solution was pumped thr ough the capillary with a flow rate of 0.2 mL/min. The analytes in the eluent were detected by ICP-MS [89]. This sol-gel capillary demonstrated detection limits of 0.17 ng/mL for Cu, 0.22 ng/mL for Hg, and 0.52 ng/mL for Pb. The rela tive standard deviations were 4.2, 2.6 and 3.8% (C = 4 ng mL 1, n = 7, sample volume = 1 mL) for Cu, Hg and Pb, respectively. This sol-gel capillary and method was also ap plied to the determination of Cu, Hg, and Pb in human serum, human urin e, and preserved egg. The anal ytical results were within the range of reported values for these reference materials. This was the first report of solgel MPTS-silica coated capillary microextr action coupled to ICP-MS for trace element analysis. This coating demonstrated excelle nt sensitivity with little solvent and sample consumption [89]. Recently, Segro and Malik [90] developed a sol-gel methyl coating in capillary microextraction hyphenated on-line with highperformance liquid chromatography. In the preparation of this capill ary, methyltrimethoxysilane was used as the sol-gel precursor [90]. The capillary was coated and installed as an external sampling loop, as described previously [79]. The methyl groups on the so l-gel precursor molecules ultimately turned into pendant groups within th e sol-gel network and were pr imarily responsible for the extraction of nonpolar analytes, including P AHs and ketones, where low and sub ng/L detection limits were observed [90]. This so l-gel methyl coating also demonstrated a counterintuitive extraction cap ability for polar analytes, including aromatic phenols, alcohols, and amines. This extraction capab ility can be attribut ed to a symbiotic extraction between methyl groups and silanol groups within the so l-gel coating and the

PAGE 131

100 analyte molecules. This study demonstrates th at sol-gel sorbents w ith short alkyl side chains have the potential to offer a polymer-fr ee alternative to traditional sol-gel coatings which are typically prepared using polymers in the sol solution. The elimination of polymers is conducive to improved thermal stab ility and solvent tole rance in the created sol-gel extracting phase and also makes the preparation process mo re facile and cost effective [90]. 2.5.4 Non-silica-based sol-gel coatings in CME 2.5.4.1 Titania-based sol-gel coatings in CME Malik and coworkers [91,92] developed a sol-gel titania hybrid organic-inorganic coating for capillary microextraction coupled to HPLC with UV detection. A capillary was coated and installed as an external sampling loop, as desc ribed previously [79]. This sol-gel capillary was successf ul in the extraction of polyc yclic aromatic hydrocarbons, ketones, and alkylbenzenes. The detec tion limits were between 0.2 and 3.1 ppb for polycyclic aromatic hydrocarbons, 2.5 and 11.6 ppb for ketones, and 0.7 and 5.5 ppb for alkylbenzenes. This sol-gel co ating was resistant to solvents and was successfully used in capillary microextraction coupled to HPLC It was also thermally stable to 320 C, making it suitable for use in GC analysis. This titania-based sol-gel coating demonstrated impressive resistance to hi gh pH. It was rinsed with 0.1 M NaOH for 12 hours with no significant change in performance. This is ve ry significant, as silica-based coatings are notoriously unstable in extrem e pH conditions [91]. Recently, Malik and coworkers [93] deve loped a sol-gel titania poly-THF coating for use in CME-HPLC with UV detection. Th is coating offered ppt detection limits for

PAGE 132

101 PAHs, ketones, phenols, alcohols, and amin es. It was even capable of extracting underivatized acids. This coating demonstrated incredible pH resistance, surviving 18 h exposures to 1.0 M NaOH (pH 14.0) and 1.0 M HCl (pH 0.0). The sol-gel titania poly-THF coated capillaries were applied to the extraction of polypeptides at pH values near their respective is oelectric points [93]. 2.5.4.2 Zirconia-Based Sol-Gel Coatings in CME Malik and coworkers [94] also devel oped a sol-gel organic-inorganic hybrid zirconia coating for capillary microextraction coupled to GC-F ID analysis. This coating successfully extracted PAHs, aldehydes, and ke tones. The observed detection limits were between 0.03 and 0.57 ng/mL for PAHs, 0.05 and 0.33 ng/mL for aldehydes, and 0.02 and 0.92 ng/mL for ketones. The run-to-run peak RSD values were between 2.45 and 7.25% for PAHs, 1.29 and 5.45% for aldehydes, and 1.24 and 5.57% for ketones. The capillary-to-capillary RSD values for GC peak areas were 4.60% for undecanal, 1.61% for hexanophenone, 5.40% for fluorene, and 4.91% for phenanthrene. This coating also demonstrated high pH stability. It was rins ed with 0.1 M NaOH (pH 13) for 24 hours. It survived this high pH treatment with only a 2.15 to 5.64% change in peak area for PAHs. Parts per trillion det ection limits were achieved for both polar and nonpolar analytes using this sol-gel coating [94]. Hu and coworkers [95] developed a so l-gel zirconia coating for capillary microextraction on-line hyphenated with i nductively coupled plasma mass spectrometry for the determination of Cr, Cu, Cd, and Pb in biological samples. The sol-gel zirconia capillary was connected to the system. The flow rate used was 0.25 mL/min. In the prefill stage, pump 1 with the injector valve in the fill position and pump 2 with the injector

PAGE 133

102 valve in the injection position were activated so that all of the tubing would be filled with solution. In step 1, pump 1 was still in the fi ll position, and the capillary was conditioned with 0.01 M NH4OH solution. In step 2, the valve was still in the fill position, but pump 1 was stopped and pump 2 was activated. The aqueous sample solution containing Cr, Cu, Cd, and Pb (with pH previous ly adjusted to 8.0 with 0.01 M NH4OH and 0.01 M HNO3) was passed through the sol-gel capillary and into a waste container. The analytes were adsorbed into the sol-gel coating during th is step. In step 3, the valve was still in the fill position, but pump 2 was stopped and pump 3 was activated. High purity deionized water was sucked through the sol-gel capillary to remove the residual matrix. In step 4, the valve was turned to the inject position and 0.5 M HNO3 solution was pumped to elute the sorbed analytes. The an alytes in the eluent were detected by ICPMS. The required desorption time was 60 sec onds. This analysis required a short analysis time of only 450 seconds. The capilla ry was used repeatedly after regeneration with high-purity deionized water and 0.01 M NH4OH solutions [95]. This sol-gel capillary demonstrated detection limits of 6.1 pg/mL for Cr, 13.6 pg/mL for Cu, 1.4 pg/mL for Cd, and 1.6 pg/mL fo r Pb. The precisions for nine replicate measurements of 1 ng/mL Cr, Cu, Cd and Pb were 4.9%, 2.2%, 2.0%, and 3.2% (RSD), respectively. This sol-gel cap illary and method was used in the analysis of human urine samples. The samples were spiked with the metals, and the coating and method demonstrated extraction recoveries be tween 89.2 and 101.8%. A NIES No.10-a Rice Flour-Unpolished certified reference mate rial and a BCR No. 184 Bovine Muscle certified reference material were also analy zed using this sol-gel capillary and method. The analytical results were within the range of reported values for these reference

PAGE 134

103 materials. This was the first report of solgel capillary microextra ction coupled to ICPMS for trace element analysis. Coupling of the sol-gel zirconia co ated capillary to ICPMS demonstrated high sensitivity and selectiv ity with low solvent consumption in the trace/ultra trace elemental anal ysis of various samples in complicated matrices [95]. 2.5.4.3 Germania-based sol-gel coatings in CME Recently, Malik and coworkers [96] deve loped a germania-based sol-gel hybrid organic-inorganic coating for capillary microextraction with gas chromatography analysis. Sol-gel germania PDMS coated capil laries also proved to be useful as GC columns [96]. The sol-gel germania PD MS coated microextraction capillaries demonstrated detection limits betwee n 11.4 and 83.8 ng/L for PAHs, 89.7 and 139.7 ng/L for aldehydes, and 30.6 and 92.3 ng/L for ketone s. The observed run-to-run peak area RSD values were between 2.33 and 5.05% for PAHs, 1.32 and 6.46% for aldehydes, and 2.38 and 7.18% for ketones. The capillary-to -capillary RSD values for GC peak area were between 5.4 and 5.9%. The sol-gel german ia PDMS capillary was able to withstand extreme pH values between 1.5 and 13.0. It wa s separately rinsed with 0.05 M HCl and 0.1 M NaOH for 24 hours. The rinsing actually improved the extraction characteristics of the capillary, as demonstrated by the chro matograms in figure 2.8 [96]. The sol-gel germania polydimethyldiphenylsiloxane capi llary was also capable of extracting chlorophenols. This was the first report on the use of sol-gel ge rmania coatings in microextraction [96].

PAGE 135

104 Figure 2.8 Excellent stability of sol-gel germ ania-PDMS coating under highly acidic conditions demonstrated through CME-GC an alysis of aldehydes using a germaniaPDMS coated microextraction capillary before (A) and after (B) con tinuously rinsing the capillary with a 0.05 M HCl (pH 1.3) soluti on for 24 h. Extraction conditions: 10 cm 0.25 mm i.d. microextraction cap illary and extraction time, 40 min (gravity-fed at room temperature). Other conditions: 5 m 0.25 mm i.d. sol-gel PDMS coated GC capillary column; splitless desorption splitless analyt e desorption by rapidl y increasing the GC injector temperature from 30 to 300 C, 5 min; GC oven temper ature programmed from 30 to 300 C at a rate of 20 C/min; helium carri er gas; FID 350 C. GC peaks: (1) nonanal, (2) decanal, (3) undecan al, and (4) dodecanal [96].

PAGE 136

105 Malik and coworkers [96] also develope d a sol-gel germania capillary using 3aminopropyltrimethoxysilane. This sol-ge l germania 3-aminopropyltrimethoxysilane capillary was capable of extracting highly polar compounds, such as free fatty acids [96]. 2.6 Sol-gel materials in stir bar sorptive extraction 2.6.1 Sol-gel PDMS coated stir bars Guan and coworkers [98] developed stir bars for sorptive extraction using sol-gel technology. A bare glass bar (30 mm x 1.8 mm o.d.) was used for the preparation of the sorptive stir bars. The sorptiv e stir bar contained an iron ba r inside the glass tube, while the ultrasonic sorptive bar contained no iron bar. These bars were sequentially cleaned with water and methylene chloride, then sequentially soaked in 1 M NaOH and 0.01 M HCl for 8 – 12 hours. The bars were washed again with water and purged with nitrogen flow at 120 C in a 40 mm x 10 mm i.d. stainle ss steel tube. The ba rs were coated by immersing them in the sol solution. After co ating, the bars were placed into a vacuum desiccator for 8 hours to allow coating gelati on to occur. The bars were then heated again at 120 C and purged with nitrogen fl ow in a 40 mm x 10 mm i.d. stainless steel tube. This tube was then put into a GC oven and heated under temperature programming from 40 C to 120 C at 1 C/min, holding at 120 C for 180 min, then heating to 240 C at the same rate, holding anot her 180 min, and finally heating to 300 C at the same rate, this time holding for 240 min. The bars were extracted by boiling methylene chloride for 4 hours. Before extraction, the bars were purged under nitrogen at 300 C for 2 hours [98].

PAGE 137

106 GC-FID was used for the analysis of n -alkanes and polycyclic aromatic hydrocarbons. Thermionic specified dete ction was used for the analysis of organophosphorus pesticides. For stir bar sorptiv e extraction, the stir bar was placed in a 50 mL sample vial at room temperature. For ultrasonic sorptive extraction, the bar was placed into a 50 mL sample vial which was put into an ultrasonic bath. Using stir-bar sorptive extraction with this coating, 90 min of extracti on time was required to reach equilibrium. Using supersonic sorptive ex traction, only 10 min of extraction time was required to reach equilibrium. For orga nophosphorus pesticides, 120 min of stir bar extraction was required. A special apparatus is required for thermal desorption, and it was made in the laboratory. It consists of a quartz liner and heaters, two gas flow controllers, temperature contro l units, and a sample transfer line. A 90 mm long stainless steel cylinder with a in ternal diameter of 6.7 mm and an external diameter of 12 mm was used to hold the desorption liner and transfer line. A gas flow c ontroller provided purge gas for the desorption from the top of the lin er, and another provided sweep gas from the bottom of the liner to prevent diffusion and ad sorption of the sample outside of the quartz liner. The transfer line was in serted 20 cm deep through the in jector septum into the GC pre-column. Thermal desorption of the analytes was accomplished by drying the stir bars with tissue paper and placing them into th e desorption apparatus for 2 min, the time required to reach the desorption temperature (2 60 C to 280 C), and then holding for an additional 5 min [98]. This stir-bar sorptive sol-gel coating provided detection lim its between 0.74 and 20.0 pg/mL for n -alkanes, 0.18 and 2.76 pg/mL for pol ycyclic aromatic hydrocarbons, and 0.3 and 8.0 pg/mL for organophosphorus pes ticides. This coating did not exhibit

PAGE 138

107 carryover problems when desorption temperat ures between 260 and 280 C were used. This sol-gel bar coating was also solvent resi stant and thermally stab le to 300 C [98]. Guan and coworkers [99] used this same sol-gel coated stir bar for the determination of organophosophorus pesticides in cucumber and potato samples. Extraction recovery values for the organophosphorus compounds in the extracts were between 93 and 105%. The observed detecti on limits were between 0.004 and 0.15 ng/g in cucumber samples and between 0.0012 and 0.098 ng/g in potato samples. The sol-gel coated stir bar was also applied to the dete rmination of dimethoate and parathion-methyl in real cucumbers and potatoes. Dimethoate and parathion-methyl were detected in cucumbers and parathion-methyl wa s detected in potatoes [99]. 2.6.2 Sol-gel multifunctional stir bar coatings Hu and Yu [100] used stir bar sorptive extraction coupled with ultrasonic assisted extraction for the determination of brominated flame retardants (BFRs) in environmental samples using HPLC. The stir bar was co ated with sol-gel polydimethylsiloxanecyclodextrin. Stir-bar sorptive extracti on with HPLC-UV was performed to analyze aqueous samples containing brominated flame retardants (BFRs). The detection limits ranged from 2.9 to 4.2 g/L for the BFRs. The bar-to-bar peak area RSD ranged from 1.3 to 15.7%. This sol-gel coated stir bar was also applied to the ex traction of BFRs from soil and dust samples. One soil sample and one dust sample were used in a recovery test. In the determination of BFRs in these samp les, the method provided extraction recoveries between 56 and 118% [100]. The same res earch group recently developed sol-gel poly(dimethylsiloxane)/poly(vinyl alc ohol) [101] and sol-gel poly(ethylene

PAGE 139

108 glycol)/poly(dimethylsiloxane)/po ly(vinyl alcohol) [102] coated stir bars for sorptive extraction. Li and coworkers [103] used a sol-gel poly(dimethylsiloxane)/ -cyclodextrin stationary phase for stir bar sorptive extrac tion. This stir-bar sorptive extraction was coupled with HPLC analysis. The observ ed detection limits were between 0.04 and 0.11 g/L for estrogens using UV detection and 8 ng/L for bisphenol A using fluorescence detection [103]. 2.7 Sol-gel monoliths in an alytical microextraction The focus of this chapter to this point has been sol-gel coatings in fiber SPME and in-tube SPME, or capillary microextraction. So l-gel coatings have b een used extensively in the extraction of many types of compounds. The majority of the sol-gel coatings were used in SPME for GC analysis of the extrac ted samples, but some SPME coatings have been used in conjunction w ith HPLC and CE as well. Continuous, porous sol-gel materials are known as sol-gel monoliths. So l-gel monoliths completely fill fused silica capillaries, and liquids can pass through the pores of the sol-gel monolith. Sol-gel monoliths have been used successfully as st ationary phases in HPLC and CE. While there have not been as many publications on sol-gel monoliths in analytical microextraction, they have also been used su ccessfully for this purpose. In this section they will be discussed in a ch ronological order. Future developments will likely lead to the more frequent use of sol-gel mono liths in analytical microextraction.

PAGE 140

109 Malik and coworkers [104] reported the preparation of porous sol-gel monolithic beds for use in analytical mi croextraction. Like sol-gel co atings, these monoliths are also chemically bonded to the silica substrate. A sol-gel ODS monolithic bed was prepared inside a fused silica capillary usi ng the sol-gel precursors TMOS and N -octadecyldimenthyl [3-(trimethoxysilyl)propyl] ammoni um chloride. Phenyldimethylsilane was used as the deactivation reagent and TFA was us ed as the catalyst. A sol-gel coating was also prepared using the same reagents for comparison. An aqueous sample containing a PAH and an alcohol was extracted under id entical conditions using both the monolithic capillary and the coated capilla ry and analyzed using GC-FID. The results indicate that the sample capacity for the monolithic capillary is at least two orders of magnitude higher than that of the sol-gel coated capillar y. Chromatograms comp aring the extraction sensitivity in capillary microextraction usi ng a sol-gel monolithic capillary and a sol-gel coated capillary are depict ed in figure 2.9 [105].

PAGE 141

110 Figure 2.9 Extraction of a sample containing a P AH and an alcohol on (A) a sol-gel ODS monolith and (B) a sol-gel ODS coated capillary [105].

PAGE 142

111 Zare and coworkers [106,107] developed a sol-gel monolithic stationary phase for capillary electrochromatography and noticed that it was capable of preconcentrating a variety of neutral and charged analytes [ 106,107]. They used a porous photopolymerized sol-gel monolith with solvent gradient and sample stacking for on-line preconcentration in capillary electrochromatography [108]. Through combining the preconcentration effects of the sol-gel monolith with solvent gradient and sample stacking, it was possible to preconcentrate 8 alkyl phenyl ketones, and 4 PAHs. Ultraviolet detection was used at wavelengths of 214 and 254 nm. Electric fiel d-enhanced sample injection yielded 1,000fold improvement in detection sensiti vity for five peptides [108]. Takeuchi and coworkers [109] used monolithic precolumns for sample enrichment in microcolumn liquid chromatogr aphy. The samples were loaded into the precolumn using a hand-made rubber band-dr iven pumping device. The precolumn allowed for enrichment of aqueous samples with volumes up to 1 mL. The desorption solvent was then allowed to flow through the precolumn and into the separation column. Phthalates were detected using a UV detector at 204 nm. The run-to-run peak area RSD values were between 0.92 and 5.5% for the capi llaries. The detection limits of phthalates in HPLC-UV with the sol-gel monolithic precolumns were between 0.21 and 0.87 ng/mL. A 150 m i.d. sol-gel monolithic precolumn was applied to the trace analysis of phthalates in laboratory distilled water and tap water samples [109]. Toyo’oka and coworkers [110] developed a sol-gel monolith for preconcentration of biogenic amines prior to their on-column derivitization for subsequent analysis by capillary electrochromatography. Both ultrav iolet and fluorescence detectors were used in this analysis. Derivitization was accomp lished by first passing the derivatizing agent

PAGE 143

112 ( o -Phthalaldehyde (OPA), 2-mercaptoethanol (2ME)) through the monolith in buffer solution, then passing buffer solution wit hout the derivatizing agent through the monolith. On-line preconcentr ation was performed by passing the amine sample solution in 9% sodium chloride through the monolith for up to 600 seconds. The derivitization produced an o -phthalaldehyde-thiolamine derivative which can be detected at 340 nm using a UV/visible detector and at 480 nm using a fluorescence detector. Using the fluorescence detector with the preconcentra tion provided detection limits of 0.1 M, resulting in a 1,000-fold increase in the de tection sensitivity compared to standard capillary electrochromatography wi th UV/visible detection [110]. Zhang and coworkers [111] developed a monolithic precolumn for on-line trapping and preconcentration of peptides in a multidimensional liquid chromatography system for proteome analysis. Both 320 m i.d. and 530 m i.d. precolumns were prepared. On-column frits were fabricated using Zorbax BP-SIL particles. The capillaries were packed with 5 m Hypersil C18 particles. Figure 2.10 is an SEM of the monolithic precolumn [111].

PAGE 144

113 Figure 2.10 Scanning electron micrographs of the 320 m i.d. sol-gel monolithic precolumn (a = 240 x, b = 1000 x) [111].

PAGE 145

114 These sol-gel monoliths were mechanically stable under the pressures encountered in HPLC. BSA tryptic digest lecucine enkaphalin, and oxytocin were preconcentrated using the sol-gel mono lithic precolumns in this study. High concentrations of salt buffers did not presen t a noticeable effect on the preconcentration abilities of the precolumns. The detecti on limits for lecucine enkephalin and oxytocin were 0.87 and 0.53 ng/L, which is equivalent to a 70-fold preconcentration factor. The precolumns were capable of recovering over 90% of the BSA tryptic digest (25 g/L), 98% of the lecucine enkaphalin (20 g/L), and 99% of the oxytocin (20 g/L). The day-to-day RSD values for recoveries of BSA peptides on a single precolumn ranged from 4.66 to 7.56%, while the column-to-column RSD values were 3.51 – 6.13% for recoveries of BSA peptides. The precolumns were resistant to continuous flushing with acidic loading buffer and had good reproducibility for back pre ssure, with back pressure RSD values between 2.68 and 3.05% and column -to-column back pressure RSD values between 1.22 and 1.26%. The precolumns were used in over 150 preconcentration/desorption cycl es with no significant change s in performance. These sol-gel monolithic precolumns appear to be suitable for the preconcentration of dilute peptides [111]. Li and coworkers [112] develope d a sol-gel poly(dimethylsiloxane)/ cyclodextrin membrane for so lid-phase microextraction. Extraction was accomplished through a membrane extraction procedure. Fo r the extraction, a 1 cm x 0.2 cm membrane was dipped into the sample solution with magnetic stirring at 1,080 rpm for the time required for equilibrium to be established. After extracting, the membrane was removed with tweezers and dried carefu lly with filter paper. The membrane was cut into small

PAGE 146

115 pieces and transferred into a sm all glass vial with 200 L of acetonitrile. The vial was agitated ultrasonically for desorp tion of the analytes from the membrane into acetonitrile. Desorption took about 5 min. The acetonitrile ex tracts were then direc tly injected into a GC-MS for analysis [112]. This sol-gel membrane was capable of extracting polycyclic aromatic hydrocarbons with detection lim its between 0.01 and 0.2 g/L. The run-to-run peak area RSD values in the extraction of PAHs were between 5.4 and 9.3%. This sol-gel membrane was also used to determine PAHs in a real river water sample. This membrane was capable of extracting phenols with detection limits between 0.02 and 1.5 g/L. The run-to-run peak area RSD values observed in the extraction of the phenols ranged from 4.9 to 13.1%. This membrane is stable in methanol, acetone, and acetonitrile solutions. It does not present carryover problems but can be disposable because of its low cost. The membrane was ma ny times less efficient in the extraction of PAHs and phenols than sol-gel fiber SPME co atings. This is most likely due to the incomplete injection of the extracted compounds since direct injection of a small amount of desorbed sample was used. The detecti on limits could be improved with complete injection of the extracted sa mples through large volume sa mple injection or thermal desorption into the GC. The membrane ha s a large surface area to extraction phase volume ratio [112]. Xue and coworkers [113] used sol-gel monoliths for optical determination of Cr(VI). Optically transparent monoliths were obt ained and stored in water before use. In order to perform the extraction of Cr(VI), each monolith was placed in a solution with a known concentration of Cr(VI) in 0.1 M HCl. The monolith was exposed to the solution

PAGE 147

116 for about 300 min, then removed from solution, rinsed thoroughly with deionized water, and analyzed by UV-vis spectroscopy. The detection limits could be decreased by placing the monolith in diphenylcarbazide soluti on prior to UV-vis spectroscopy analysis. The monolith was regenerated by soaking in 6.0 M HCl after soaking in diphenylcarbazide. To perform the UV-vi s spectroscopy, absorbance at 350 nm was measured. At ppm concentrations of Cr(VI), the absorbance at 350 nm could be used to detect the Cr(VI). At ppb concentrations of Cr(VI), there is not enough color change to detect the Cr(VI) at 350 nm. After imme rsing the monolith in diphenylcarbazide solution, a magenta color forms, which can be detected at 540 nm. Using the monolith to extract the Cr(VI) and subseque ntly immersing it in diphenylc arbazide solution resulted in the detection of Cr(VI) at concentrations as low as 10 ppb. This sol-gel monolith was capable of removing 77.3 to 81.7% of Cr(VI) fr om solution. A sample of lake water was spiked with 85 ppb Cr(VI) and the pH was adjusted to 1. The monolith was exposed to 20 mL of this solution for 6 hours, rins ed with deionized water, exposed to diphenylcarbazide solution for one hour, and an alyzed using UV-vis spectrometry. The concentration of Cr(VI) in the lake water wa s accurately determined. The matrix of the lake water did not present a problem in this analysis, making this sol-gel monolith and method suitable for environmental analysis [113]. Bergquist and coworkers [114] used a sol-gel monolith column for on-line biological sample clean-up for electrospra y mass spectrometry. Electrospray ionization with time of flight mass spectrometry was used to perform the analysis of peptides in this study. The exit end of the monolith was direc tly connected to the mass spectrometer to effectively hyphenate the elec trophoretic separation with electrospray mass spectrometry.

PAGE 148

117 Peptide stock solutions were prepared in wa ter at concentrations of 15 g/mL. The running buffer used consisted of 5 mM a mmonium acetate (pH 6.8) in 40% water and 60% acetonitrile. A washing buffer of 5 mM ammonium acetate (pH 6.8) in water was used. Prior to analysis, urine samples were filtered and stored at -20 C. A freshly thawed sample was used each day of analysis. The sol-gel monolith column was conditioned with washing buffer for 5 min using a slight over-pressure of 1 bar. Urine samples spiked with 15 g/mL concentrations of neurotensin, oxytocin, angiotensin II, leucine-enkephaline, and leutenizing hormone -releasing hormone were injected (1.4 L) onto the column at 1 bar pressure for 5 min. Washing buffer was then allowed to flow through the column for 15 min to remove salts and electrolytes The inlet of the column was then placed into a vial containing r unning buffer and a separation voltage of +20 kV was applied. A 70-fold preconcentration wa s achieved for all of the peptides in the sample. This column had a lifetime of about 30 runs, due to the analysis of crude urine samples. This monolithic column was capable of selectively preconcentrating the hydrophobic peptides while the hydrophilic specie s, such as urea and salt, passed through the column. The peak area RSD values were about 4% on the same day and 5% between days. This sol-gel monolith was capable of both on-line desalting and preconcentration of peptide samples [114]. Bergquist and Johannesson [115] used a si milarly prepared [ 114] sol-gel monolith column for the on-line extraction and quantific ation of escitalopram from urine [115]. Electrospray ionization with time of flight mass spectrometry detection was used to perform the analysis in this study. The obs erved detection limit for ascitalopram was 10 pg/mL. The intra day RSD value was less th an 6.3%. This sol-gel monolith was also

PAGE 149

118 used in the determination of the concentrati on of ascitalopram in a human urine sample [115]. Feng and coworkers [116] developed a so l-gel hybrid organic-inorganic octyl monolithic column for in-tube SPME coupled to capillary HPLC with UV detection. Two microflow pumps (pump A and pump B) were used. Pump A was combined with valve A and a 1 mL stainless steel sampling loop. Two segments were connected using polyether ether ketone (PEEK) tubing between valves A and B. The sample loop of valve B was replaced with the monolithic cap illary column. Before extraction, valve A was switched to the load positi on and the carrier solution was driven by pump A to flow through the monolithic capillary for conditioni ng at 0.05 mL/min. The stainless steel sampling loop was filled with sample solution using a syringe. Valve A was rotated to the inject position for extract ion. Extraction efficiency was increased by adding NaCl to the 40% methanol in the sample solution. Extraction efficiency improved with increasing extraction volume, but a volume of over 1 mL was undesira ble since a longer extraction time was required at a flow rate of 0.05 mL/min. After extraction, the valve was returned to the load position. Desorption was performed by switching valve B from load to inject, and 90 L of mobile phase (85% methanol, 15% wate r) was driven through the monolith by pump B. The desorbed analyt es were transferred to the inlet of an analytical column for separation follo wed by ultraviolet de tection [116]. This sol-gel monolith was capable of extracting PAHs [116]. The observed detection limits were 6.5 ng/mL for biphenyl 7.1 ng/mL for fluorene, 2.4 ng/mL for phenanthrene, and 8.1 ng/mL for fluoranthene. The intraand inte r-day recovery RSD values were less than 7.4 and 8.1%, respect ively. Standard solutions of 2 mg/L

PAGE 150

119 concentrations of PAHs were extracted usi ng this monolith, and th e observed recoveries ranged from 83.2 to 110.2% [116]. Recently, a sol-gel tetraethoxysilane-ba sed monolith was used as an in-line concentrator for the determination of methioni ne enkephalin in cerebra l spinal fluid using capillary electrophoresis hyphenated with mass spectrometry [117]. A 40-fold preconcentration was demonstrat ed. The observed detection lim it in cerebral spinal fluid was approximately 1 ng/mL [117]. 2.8 Conclusion Sol-gel materials have been very successf ul in the area of analytical sample preparation. The preceding examples are indi cative of this. This chapter should be consulted as a reference material for anyone interested in doing further research using sol-gel coatings, monoliths, or particles for any sample preparation purposes. Sol-gel materials are versatile and capable of a wide range of extraction capabilities for numerous types of analytes. They are generally very re sistant to solvents a nd high temperatures and demonstrate long lifetimes and can even handl e difficult matrices. This chapter should promote awareness of the different sol-gel materials that have been developed for analytical sample preparation and facilitate future research and development in this area. 2.9 References for chapter two [1] C. J. Brinker, G. W. Scherer, Sol-Ge l Science, Academic Press Inc., New York, 1990.

PAGE 151

120 [2] S. L. Chong, D. Wang, J. D. Hayes, B. W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [3] J. Pawliszyn, Applications of Solid Ph ase Microextraction, The Royal Society of Chemistry, Cambridge, UK, 1999. [4] A. Malik, A. D. Wang, Capillary Colu mn and Method of Making. PCT Int. Appl. 2000, 78 pp. [5] A. P. Vonderheice, M. Montes-Ba yon, J. A. Caruso, Analyst 127 (2002) 49. [6] H. Bagheri, A. Es-haghi, M. R. R ouini, M. R. J. Chromatogr. B 818 (2005) 147. [7] H. Bagheri, A. Es-haghi, F. Khalilian, M. R. Rouini, J. Pharm. Biomed. Anal. 43 (2007) 1763. [8] A. Fernandes de Oliveira, C. Berta da S ilveira, S. Denafre de Campos, E. Antonio de Campos, E. Carasek, Chromatographia 61 (2005) 277. [9] W. Liu, Y. Hu, J. Zhao, Y. Xu, Y. Guan J. Chromatogr. A 2006, 1102, 37-43. [10] Guan, Y.; Liu, W.; Zhao, J.; Xu, Y. Me thod for preparation of coating for solid phase microextraction. Faming Zhuanli Shenqing Gongkai Shuomingshu 2007, 9pp. [11] M. A. Azenha, P. J. Nogueira, A. F. Silva, Anal. Chem. 78 (2006) 2071. [12] M. Azenha, P. Nogueira, A. Fern ando-Silva, Anal. Chim. Acta 610 (2008) 205. [13] C. G.Yang, X. J. Ji, X. Z. Lan, Chin. Chem. Letters 19 (2008) 996. [14] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [15] Z. Zeng, M. Liu, Faming Zhua nli Shenqing Gongkai Shuomingshu 2006, 10pp. [16] X. Li, J. Gao, Z. Zeng, Anal. Chim. Acta 290 (2007) 26. [17] F. Augusto, A. L. Lopes, J. Chromatogr. A 1056 (2004) 13.

PAGE 152

121 [18] V. G. Zuin, A. L. Lopes, J. H. Ya riwake, F. Augusto, J. Chromatogr. A 1056 (2004) 21. [19] R. C. da Silva, V. G. Zuin, J. H. Yariwake, M. N. Eberlin, F. Augusto, J. Mass Spectrom. 42 (2007) 825. [20] G. A. da Silva, F. Augusto, R. J. Poppi, J. Chromatogr. A 1138 (2007) 251. [21] M. Yang, Z. R. Zeng, W. L. Qiu, Y. L. Wang, Chroma tographia 56 (2002) 73. [22] M. Yang, W. Qiu, Z. Zeng, X. Niu, Fenxi Shiyanshi 21 (2002) 40. [23] Y. Wang, Z. Zeng, C. Wang,. Jilin Daxue Xuebao Lixueban 45 (2007) 661. [24] C. Dong, Z. Zeng, M. Ya ng, Water Res. 39 (2005) 4204. [25] Y. Wang, M. Yang, C. Dong, F. S ong, Huazhong Shifan Daxue Xuebao Ziran Kexueban 40 (2006) 374. [26] C. Dong, W. Wang, Huaxue Fenxi Jiliang 15 (2006) 27. [27] L. Cai, J. Xing, L. Dong, C. Wu, J. Chromatogr. A 1015 (2003) 11. [28] K. Jinno, M. Taniguchi, H. Sawada, M. Hayashida, Analusis 26 (1998) M27. [29] Y. Wang, Z. Zeng, M. Yang, C. D ong, Z. Wang, Wuhan Ligong Daxue Xuebao 27 (2005) 37. [30] Y. Hu, Y. Fu, G. Li, Anal. Chim. Acta 567 (2006) 211. [31] M. M. Liu, Z. R. Zeng, C. L. Wang, Y. J. Tan, H. Liu, Chromatographia 58 (2003) 597. [32] A. F. P. Biajoli, F. Augusto, Anal. Sci. 24 (2008) 1141. [33] F. Bianchi, F. Bisceglie, M. Careri, S. Di Berardino, A. Mangia, M. Musci, J. Chromatogr. A 1196 (2008) 15. [34] H. Bagheri, E. Babanezhad, F. Khalilian, Anal. Chim. Acta 616 (2008) 49.

PAGE 153

122 [35] Z. Zeng, W. Qiu, Z. Huang, Z. Anal. Chem. 73 (2001) 2429. [36] Z. Zeng, W. Qiu, M. Ya ng, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [37] L. Yun, Anal. Chim. Acta 486 (2003) 63. [38] L. Cai, Y. Zhao, S. Gong, L. Do ng, C. Wu, Chromatographia 58 (2003) 615. [39] D. Wang, J. Xing, J. Peng, C. Wu, J. Chromatogr. A 1005 (2003) 1. [40] D. Wang, J. Peng, J. Xing, C. Wu, Y. Xu, J. Chro matogr. Sci. 42 (2004) 57. [41] J. Yu, C. Wu, J. Xing, J. Chromatogr. A 1036 (2004) 101. [42] L. Cai, S. Gong, M. Chen, C. Wu, Anal. Chim. Acta 559 (2006) 89. [43] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [44] X. Li, Z. Zeng, J. Zhou, Anal. Chim. Acta 509 (2004) 27. [45] X. Li, Z. Zeng, J. Zhou, S. Gong, W. Wang, Y. Chen, J. Chromatogr. A 1041 (2004) 1. [46] C. Dong, Z. Zeng, X. Li, Talanta 66 (2005) 721. [47] M. Liu, Z. Zeng, Y. Lei, H. Li, J. Sep. Sci. 28 (2005) 2306. [48] Z. Zeng, M. Liu, Faming Zhuan li Shenqing Gongkai Shuomingshu 2006, 9 pp. [49] X. Li, S. Gang, Z. Zeng, Chromatographia 62 (2005) 519. [50] X. Li, Z. Zeng, M. Hu, M. Mao, J. Sep. Sci. 28 (2005) 2489. [51] X. Zhou, X. Li, Z. Ze ng, J. Chromatogr. A 1104 (2006) 359. [52] X. Li, Z. Zeng, Y. Xu, An al. Bioanal. Chem. 384 (2006) 1428. [53] J. Zhou, Z. Zeng, Anal. Chim. Acta 556 (2006) 400. [54] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37.

PAGE 154

123 [55] Z. Yang, W. Zhang, C. Wang, D. X u, K. Pan, C. Wu, Chin. J. Chromatogr. 21 (2003) 617. [56] F. Bianchi, M. Mattarozzi, P. Betti, F. Bisceglie, M. Careri, A. Mangia, L. Sidisky, S. Ongarato, E. Dalcanale, Anal. Chem. 80 (2008) 6423. [57] Z.Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [58] R. Gomes da Costa Silva, F. Augusto, J. Chromatogr. A 1072 (2005) 7. [59] M. Liu, Z. Zeng, B. Xiong, J. Chromatogr. A 1065 (2005) 287. [60] M. Liu, Z. Zeng, Y. Tia n, Anal. Chim. Acta 540 (2005) 341. [61] M. Liu, Z. Zeng, H. Fa ng, J. Chromatogr. A 1076 (2005) 16. [62] H. Fang, M. Liu, Z. Zeng, Talanta 68 (2006) 979. [63] D. Cha, M. Liu, Z. Ze ng, Anal. Chim. Acta 572 (2006) 47. [64] T. P. Gbatu, K. L. Sutton, J. A. Caruso, Anal. Chim. Acta 402 (1999) 67. [65] M. Giardina, S. V. Olesik, Anal. Chem. 73 (2001) 5841. [66] M. Giardina, S. V. Olesik, Anal. Chem. 75 (2003) 1604. [67] M. Giardina, L. Ding, S. V. Ol esik, J. Chromatogr. A 1060 (2004) 215. [68] C. Teng, Y. Chen, Rapid Commun. Mass Spectrom. 17 (2003) 1092. [69] M. Azenha, C. Malheiro, A. F. Silva, J. Chromatogr. A 1069 (2005) 163. [70] C. Basheer, S. Jegadesan, S. Valiyaveet til, H. K. Lee, J. Chromatogr. A 1087 (2005) 252. [71] D. Wang, Fenxi Ce shi Xuebao 25 (2006) 11. [72] G. Ouyang, R. Jiang, F. Zhu, T. Luan, Faming Zhuanli Shenqing Gongkai Shuomingshu 2008, 14pp. [73] L. M. Wei, Q. Y. Ou, J. B. Li, Chin. Chem. Letters 15 (2004) 1127.

PAGE 155

124 [74] K. Farhadi, R. Tahmaebi, R. Maleki, Talanta 77 (2009) 1285. [75] L. Xu, H. K. Lee, Anal. Chem. 79 (2007) 5241. [76] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [77] A. Malik, A. Sample Pr econcentration Tubes with SolGel Surface Coatings and/or Sol-Gel Monolithic Beds. PCT Int. Appl. 2002, 71 pp. [78] S. Kulkarni, L. Fang, K. Alhoosha ni, .A. Malik, Chromatogr. A 1124 (2006) 205. [79] S. S. Segro, A. Malik, J. Chromatogr. A 1205 (2008) 26. [80] Y. Fan, Y. Feng, S. Da, Z. Wang, Talanta 65 (2005) 111. [81] W. Li, D. Fries, A. Alli, A. Malik, Anal. Chem. 76 (2004) 218. [82] A. Malik, J. D. Hayes, Sol-Gel Open Tubular ODS Columns with Charged Inner Surface for Capillary Electrochromatography. PCT Int. Appl. 2002, 49 pp. [83] W. Li, D. Fries, A. Malik, J. Sep. Sci. 28 (2005) 2153. [84] S. Kulkarni, A. M. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50. [85] A. Kabir, C. Hamlet, K. S. Yoo, G. R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [86] A. Malik, A. Kabir, G. R. Newkome, K. S. Yoo, K. S. Sol-Gel Dendron Separation and Extraction Capillary Column. U.S. Pat. Appl. Publ. 2005, 37 pp. [87] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [88] A. Malik, A. Kabir, A. Sol-Gel Poly tetrahydrofuran-Based Coating for Capillary Microextraction. U.S. Pat. Appl. Publ. 2006, 24 pp. [89] F. Zeng, B. Hu, B. Talanta 73 (2007) 372. [90] S. S. Segro, A. Malik, A. J. Chromatogr. A 1200 (2008) 62.

PAGE 156

125 [91] T. Y. Kim, K. Alhoosha ni, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165. [92] A. Malik, T. Y. Kim, Titania-Based So l-Gel Coating for Capillary Microextraction. U.S. Pat. Appl. Publ. 2006, 23 pp. [93] S. S. Segro, Y. Cabezas, A. Malik, J. Chromatogr. A. 1216 (2009) 4329. [94] K. Alhooshani, T. Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [95] Y. Wu, B. Hu, Z. Jiang, Y. Feng, P. Lu, B. Li, Rapid Commun. Mass Spectrom. 20 (2006) 3527. [96] L. Fang, S. Kulkarni, K. Alhooshani A. Malik, A. Anal. Chem. 79 (2007) 9441. [97] J. D. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3. [98] W. Liu, H. Wang, Y. Gua n, J. Chromatogr. A. 1045 (2004) 15. [99] W. Liu, Y. Hu, J. Zhao, Y. Xu, Y. Guan, J. Chromatogr. A. 1095 (2005) 1. [100] C. Yu, B. Hu, J. Chromatogr. A 1160 (2007) 71. [101] C. Yu, B. Hu, J. Sep. Sci. 32 (2009) 147. [102] C. Yu, X. Li, B. Hu, J. Chromatogr. A 1202 (2008) 102. [103] Y. Hu, Y. Zheng, F. Zhu, G. Li, J. Chromatogr. A 1148 (2007) 16. [104] A. Malik, S. Bigham, J. Medlar, C. Ashf ord, A. Kabir, A. Sol-Gel Approach to in situ Creation of Surface Coatings and Monolit hic Beds for Analyti cal Microextraction. Extech 2000 Advances in Extraction T echnologies, May 1-3, 2000, Department of Chemistry, University of Waterloo, Waterloo, Ontario, Canada. [105] Malik, A, Comp. Anal. Chem. 37 (2002) 1023. [106] M. T. Dulay, J. P. Quirino, B. D. Be nnett, M. Kato, R. N. Zare, Anal. Chem. 73 (2001) 3921.

PAGE 157

126 [107] J. P. Quirino, M. T. Dulay, B. D. Bennett, R. N. Zare, Anal. Chem. 73 (2001) 5539. [108] J. P. Quirino, M. T. Dulay, R. N. Zare, Anal. Chem. 73 (2001) 5557. [109] L. W. Lim, K. Hirose, S. Tatsumi, H. Uzu, M. Mizukami, T. Takeuchi, J. Chromatogr. A, 1033 (2004) 205. [110] S. Oguri, H. Tanagaki, M. Hamaya, M. Kato, T. Toyo’oka, Anal. Chem. 75 (2003) 5240. [111] X. Gu, Y. Wang, Z. Zhang, J. Chromatogr. A 1072 (2005) 223. [112] Y. Hu, Y. Yang, J. Huang, G. Li, Anal. Chim. Acta 543 (2005) 17. [113] N. A. Carrington, G. H. Thomas, D. L. Rodman, D. B. Beach, Z. Xue, Z. Anal. Chim. Acta 581 (2007) 232. [114] N. Johannesson, E. Pearce, M. Dulay, R. N. Zare, J. Bergquist, K. E. Markides, J. Chromatogr. B 842 (2006) 70. [115] N. Johannesson, J. Bergquist, J. Pharm. Biomed. Anal. 43 (2007) 1045. [116] M. Zheng, B. Lin, Y. Fe ng, J. Chromatogr. A 1164 (2007) 48. [117] R.Ramautar, C. K. Ratnayake, G. W. Somsen, G. J. de Jong, Talanta 78 (2009) 638.

PAGE 158

127 CHAPTER THREE SOL-GEL METHYL COATING IN CAPILLARY MICROEXTRACTION HYPHENATED ON-LINE WITH HIGH PERFORMANCE LIQUID CHROMATOGRAPHY. COUNTERINTUI TIVE EXTRACTION BEHAVIOR FOR POLAR ANALYTES 3.1 Introduction Solid-phase microextraction (SPME) is a fairly new sa mple preparation technique [1] that uses a solid or liquid sorbent coati ng to extract and concen trate target analytes from a sample. It completely eliminates th e use of hazardous organic solvents used in conventional extraction techniques [2]. SPM E is typically coupled with a separation technique, including gas chromatogra phy (GC) [3], high-performance liquid chromatography (HPLC) [4,5], supercritical fluid chromatography (SFC) [6], and capillary electrophor esis (CE) [7,8]. In the traditional SPME technique, a so rbent coating is applied on the end segment of a fiber. The fiber, typically a small-diameter fused silica rod (~100 m in diameter), is installed in a specially desi gned syringe-like device (SPME syringe). Here the fiber is connected to a stai nless steel tube, which, in tur n, is connected to the plunger of the SPME syringe. The sorbent coating on the fiber serves as the extracting phase. The extracting phases, depending on their compositions, have different affinities for different analytes [2]. Initially, SPME coa tings were prepared from organic polymeric materials since they provided (a) ease of physical attachment to the fused-silica fibers, (b) reasonably high temperature stability, (c) sim ilar thermal expansion coefficients, and (d)

PAGE 159

128 great structural diversity. They predomin antly included organosiloxane materials, particularly polysiloxanes (silic ones) with different functiona l groups introduced as side chains [9-12]. Non-silicone coatings included polyamide [13], polyimide [14], cellulose acetate [15], and polyvinyl chloride [15]. Molecularly imprinted polymers have recently been used as SPME coatings [16,17]. Inorganic coatings have been occasionally used for SPME, but structurally they are not as diverse as the organic coatings [18]. One of the main drawbacks of the trad itional fiber SPME technique is that the coated segment of the fiber is very fragile and can be easily damaged. The needle of the SPME device can easily be bent during operation resulting in the scraping and damage of the coating as the fiber is pushed through the syringe needle. The short length of the coated segment only allows for a small extrac ting phase loading on the fiber, resulting in low sample capacity and extraction sensitivit y of fiber SPME. Besides, fiber SPME is difficult to couple to HPLC. To address these problems, in-tube SPME was developed, in which a capillary with the extracting pha se coating on the inne r surface was used instead of a fiber coated on the outer surface. In this format, the extracting phase coating is secured within th e capillary and, therefor e, protected from mechanical damage. Extraction is performed by simp ly passing the sample through the coated capillary. Since a significantly longer segment of the capillary can be used for extraction, in-tube SPME is also in a position to provide greater extracting phase loadin g, enhanced sample capacity, and improved extraction sensitivity. Segments of GC capillary columns with thin (sub-micrometer) coatings have been commonly employed in in-tube SPME [19]. The use of such thin coatings is not conduc ive to achieving the desired improvement in sample capacity and extraction sensitivity. Stable thick coatings are difficult to prepare

PAGE 160

129 using conventional techniques. Since convent ional coatings are not chemically bonded to the capillary wall, they are unstable at hi gh temperatures (SPME-GC) or under organoaqueous mobile phases (SPME-HPLC) typically us ed to desorb the extracted analytes for chromatographic analysis. Low thermal stabili ty of conventional coatings results in incomplete analyte desorption often leading to undesirable carryover problems in SPMEGC [20]. It also limits the range of analytes that can be analyzed by SPME-GC, practically leaving high-bo iling compounds off-limit [20]. Poor solvent stability of conventional coatings represents a se rious hurdle to effective hyphenation and widespread application of SPME-HPLC. To combat these problems, sol-gel coating was developed [20] for SPME. In solgel microextraction a sol-gel extr acting phase coating is attached to the substrate (fiber or capillary) through chemical bondi ng. This effectively solves the problems associated with poor thermal stability of conventional SPME coatings [20]. The sol-gel coatings were also applied to the insi de of capillary tubes [21,22]. This technique is known as sol gel capillary microextraction (C ME) [21], which allows for the creation of covalently bonded extracting phase coating of greater thickness, providing better extraction sensitivity. Also, for sol-gel coated capillar ies the use of organic solvents to rinse the coated phase or high temperatures to therma lly desorb the extracted analytes does not present a problem [21,23]. This solv ent resistance makes sol-gel capillary microextraction suitable for hyphenation with HPLC. A variety of sol-gel extracting phases have been developed for SPME, both in the fiber and capillary formats [22-27]. Sol-gel extracting phases used in fiber SPM E include sol-gel PDMS [20], sol-gel PEG [24], sol-gel crown ether [28], sol-gel pol y(methylphenylvinylsilo xane) (PMPVS) [10],

PAGE 161

130 sol-gel hydroxy fullerene [27], sol-gel calix[4]a rene [29], oligomer-based sol-gels [30], and sol-gel -cyclodextrin [26]. Sol-gel extr acting phases used in capillary microextraction include sol-gel PDMS [21], sol-gel dendrimer [31], sol-gel poly(tetrahydrofuran) [22], and sol-gel cyan o-PDMS [32]. Sol-gel titania PDMS [33] and sol-gel -cyclodextrin [34] have b een used in CME-HPLC. The sol-gel coatings reported in the l iterature have almost exclusively been prepared using sol solutions containing sol-ge l active polymeric materials together with the sol-gel precursors. During the coating pr ocess, these polymeric materials presumably get bonded to the resulting sol-gel network and create the surfaceanchored extraction phase coating. Very few studies have been de voted to the use of short chain ligands to provide the extracting phase. In this pa per, we used methyltrimethoxysilane as a precursor for the creation of a sol-gel coati ng for capillary microextraction and evaluated its performance in CME-HPLC. The use of methyltrimethoxysilane as the sol-gel precursor resulted in a sol-ge l coating with pendant methyl groups in its structure. Considering the nonpolar nature of the methyl group, one would expect such a coating to be effective for the extraction of only nonpolar analytes. However, here we provide experimental results showing that the sol-ge l methyl coating can be effective in the extraction of both polar and nonpol ar analytes. We provide a plausible explanation for this non-intuitive extraction behavior of the sol-gel methyl coating.

PAGE 162

131 3.2 Experimental 3.2.1 Equipment On-line coupled sol-gel CME-HPLC expe riments were carried out using a MicroTech Scientific (Vista, CA, USA) Ultra Plus HPLC system equipped with a Linear UVIS 200 variable wavelength UV detector. For the HPLC separations, a reversed-phase Luna C-18 column (15 cm x 4.6 mm i.d.) was use d. For thorough mixing of the sol solution ingredients, a Fisher model G-560 Vortex Geni e 2 system (Fisher Scientific, Pittsburgh, PA, USA) was utilized. A Thermo IEC model Micromax microcentrifuge (Needham Heights, MA, USA) was employed for the centrifugation of sol solutions. Nanopure water (15 M ) was obtained from a Barnstead model 04741 Nanopure deionized water system (Barnstead/Thermodyne, Dubuque, IA USA). Chrom-Perfect version 3.5 for Windows computer software (Justice Laborat ory Software, Denville, NJ, USA) was used for on-line collection and processi ng of the CME-HPLC data. 3.2.2 Chemicals and materials Fused silica capillary (0.25 mm I. D.) was purchased from Polymicro Technologies (Phoenix, AZ, USA). Methyltrimethoxysilane (MTMS), poly(methylhydrosiloxane) (PMHS), polycycli c aromatic hydrocarbons (fluoranthene, fluorene, naphthalene, phenanthrene, and acenaphthene), ketones (benzophenone, coumarin, and trans-chalcone), phenols (2,3-dichlorophenol, 2-chlorophenol, 3,4dimethylphenol, and 3,5-dimethylphenol), and amines (caffeine and m -toluidine) were purchased from Aldrich (Milwaukee, WI, USA). Anthracene and 1-[1,1’-biphenyl]-4ylethanone were purchased from Eastman Kodak (Rochester, NY, USA). Diphenylamine was purchased from J.T. Baker (Phillips burg, NJ, USA). Alcohols (benzhydrol and

PAGE 163

132 catechol) were purchased from Sigma (St. L ouis, MO, USA). Reso rcinol was purchased from Spectrum (Gardena, CA, USA). Trifl uoroacetic acid was purchased from Acros (Morris Planes, NJ, USA). HPLC-grade ace tonitrile (ACN), methanol, and methylene chloride were purchased from Fisher Scientific (Pittsburgh, PA, USA). 3.2.3 Pretreatment of fused silica capillary Fused silica capillary was pretreated prio r to the creation of the sol-gel methyl coating. First, it was sequentially rinsed with 4 mL each of methylene chloride, methanol, and deionized water. Next, both en ds of the capillary were sealed using an oxy-acetylene torch. Following this, the seal ed capillary was placed in a GC oven and heated at 350 0C for two hours. It was then removed from the oven and allowed to cool. Subsequently, the ends of the capillary were cut open using an alumina wafer, and it was placed in the GC oven with continuous helium flow (1 mL/min). The temperature of the oven was programmed from 40 0C to 250 0C, at 5 0C/min. The fused silica capillary was held at the temperature of 250 0C for 2 h. 3.2.4 Preparation of the sol-gel methyl coated microextraction capillary The sol solution was prepared using the following procedure. First, 0.02 g of PMHS was weighed into a clean microcentr ifuge tube. Then, 200 L of MTMS were added to it and the mixture was vortexed for 1 min. Next, 18 L of methylene chloride were added to the microcentrifuge tube, a nd the mixture was vortexed again for 1 min. Finally, 120 L of trifluoroace tic acid, which contained 15% deionized water, was added to the mixture. The sol solution was vortexe d once more for 1 min, and then centrifuged for 4 min at 14 000 rpm (15 682 g ).

PAGE 164

133 After centrifugation, the top por tion of the liquid in the microcentrifuge tube was removed using a micropipette and transferred to a new clean microcentrifuge tube. A piece of hydrothermally pretreated fused sili ca capillary (60 cm x 0.25 mm I.D.) was then coated with this sol solution. To accomplish this, the sol solution was allowed to flow through the capillary under heli um pressure (50 psi) usi ng a capillary filling/purging device [35]. The exit end of th e capillary was then sealed us ing a piece of rubber septum and kept under helium pressure (50 psi) for 40 min. Then, the rubber septum was removed to expel the liquid from the capilla ry and to allow helium to flow through the capillary for 40 min at 50 psi. The capillar y was further placed in a GC oven purging it with a continuous helium flow. Following this, the temperature of the oven was programmed from 40 0C to 350 0C, at 5 0C/min. It was held at the final temperature for 2 h, followed by cooling down to room temperature. The capillary was rinsed with 2 mL of a (1:1, v/v) methanol/methylene chloride solution. The capillary was once more conditioned under helium purge by progr amming the temperature from 40 0C to 350 0C, at 5 0C/min, holding it at 350 0C only for 30 min. The capillary was then ready for use in CME-HPLC. Sol-gel methyl coated capillaries were al so prepared using no PMHS in the sol solution. These capillaries were prepared using the same procedure and conditioning steps, except that no PMHS was added and 187 L of trifluoroacetic acid containing 15% deionized water was used. 3.2.5 On-line CME-HPLC analysis Figure 3.1 is a schematic of the set-up used for the on-line sol-gel CME-HPLC analysis. Prior to th e extraction, a Luna C-18 HPLC column (15 cm x 4.6 mm I.D.) was

PAGE 165

134 equilibrated with the mobile phase composition used for the se paration of the analytes in the sample. To perform the capillary microe xtraction on-line, a 40 cm long piece of the sol-gel methyl coated capillary was installe d on the HPLC six-port injection valve using it as an external sampling loop. For this, the ends of the capillary were provided with PEEK tubing sleeves properly fitted with nuts and ferrules. The extraction was performed by allowing liquid samples (placed in the gravity-fed sample delivery system [21]) to pass through the sol-gel methyl coat ed capillary. A piece of deactivated 0.53 mm i.d. fused silica capillary was used as a transfer line to facili tate the sample flow from the sample delivery system into the HPLC inject or. Keeping the inje ctor in the “load” position, the sample was allowed to drip through the sol-gel coated capillary (used as the external sampling loop) to reach an extraction equilibrium (40-80 min extraction time, depending on the sample) with the sol-gel methyl coating on inner surface of the capillary. The extracted analytes were then transferred to the HPLC column by switching the injection valve to the “inject” position. The high pressure flow of the HPLC mobile phase desorbed the extracted analytes from the sol-gel capillary in to the HPLC column for separation. Isocratic elution was used with H2O/ACN mobile phase of appropriate compositions to achieve adequate separati on of the analytes in the sample.

PAGE 166

135 Figure 3.1 Experimental setup used to perform the sol-gel CME-HPLC experiments. During extraction the sample solution flows from the gravity-fed sample dispenser, through the sol-gel methyl capi llary, and into waste. Du ring analysis the valve is switched to the inject positi on, and the mobile phase flows through the sol-gel methyl capillary, desorbing the analytes and eluting them to the column for separation followed by UV detection.

PAGE 167

136 3.3 Results and discussion The sol-gel process i nvolves the formation, us ually through hydrolytic polycondensation of one or more sol-gel precur sors (typically alkoxi des), of a colloidal system, the sol, followed by its gelation to form a three-dimensional network in a continuous liquid phase, the gel [36]. The pr eparation of a sol-ge l coating typically involves the use of a number of ingredients in the sol solu tion. These are the sol-gel precursor, the sol-gel active organic polymer, the solvent, the deactiv ating reagent, water, and the catalyst [29,31-33,37-44]. Typical sol-ge l precursors are alkoxysilanes [45], such as tetramethoxysilane (TMOS), tetraethoxys ilane (TEOS), or methyltrimethoxysilane (MTMS). Sol-gel-active organi c components that ha ve been used in the preparation of sol-gel coatings include poly(dimethylsilo xane) (PDMS) [20,21], poly(ethylene glycol) (PEG) [24,46], poly(vinyl alcoho l) (PVA) [47], crown ethers [28], cyclodextrins [26], and calixarenes [29]. Appropriate solvents are used to thoroughly dissolve all of these ingredients into a sol soluti on and include common solvents such as methylene chloride [33], THF [30], isopropanol [ 37], and acetone [25]. Deactiva ting reagents are typically used to derivatize (or block) residual silanol groups on the fused si lica capillary surface or in the sol-gel network. Typical deactivating reagen ts include PMHS [20,21,23] and 1,1,1,3,3,3-hexamethyldisilazane (HMDS) [32,33,37] Typical catalysts used are acids (e.g., trifluoroacetic acid [ 20,21]), bases (e.g., ammonium hydroxide [30]) or fluorides (e.g., ammonium fluoride [48] ). Although sol-gel SPME coatings reported in the literature are predominantly silica-based, titani a[33], zirconia[41], alumina[37], and germania[49] based sol-gel coatings have b een shown to possess superior pH stability.

PAGE 168

137 Both silicaand nonsilica-based organic-i norganic hybrid SPME coatings have been exclusively prepared using an or ganic polymer in the sol solution. The sol-gel methyl coating described here was created from a sol-gel precursor, MTMS, alone. Through hydrolytic polycondens ation reactions, MTMS generated a solgel network with pendant methyl groups. Cova lent bonding of a part of this evolving solgel network to the inner walls of the fused s ilica capillary ultimately led to the formation of a surface-anchored sol-gel coating to serv e as the extracting phase in CME. This presumably occurred through the formation of covalent bonds between sol-gel active groups on the network (e.g., silanol or alkoxy groups) and silanol groups on the inner wall of the fused silica capillary. To facilita te this, the stock fuse d silica capillary was initially subjected to hydrothermal pretreat ment to promote the formation of silanol groups on its inner surface. Poly(methylhydros iloxane) was added to the sol-gel mixture in order to deactivate residual silanol groups on the sol-gel coating after it has been formed. Scheme 3.1 depicts the hydrolytic pol ycondensation reactions carried out within a hydrothermally pretreated fu sed silica capillary using trif luoroacetic acid (containing 15% H2O) as the sol-gel catalyst.

PAGE 169

138 Si CH3 OCH3 CH3O OCH3 H2O Si CH3 OH HO OH CH3OH H y d r o l y sis of t h e MTMSso l g e l p r ecu r so r :catalystSi CH3 OH HO OH Si CH3 OH n HO OH n H2O Polycondensation of the hydrolyzed products: Bonding to inner wall of fused-silica capillary: Si CH3 O HO (O Si)n CH3 O OH H2O OH OH OH OH OH inner capillary wall surface-bonded sol-gel methyl coating Si CH3 O O (O Si)n CH3 O OH Si CH3 O HO (O Si)n CH3 O OH + + + + + + Scheme 3.1 Chemical reactions involved in the fo rmation of the sol-gel methyl coating.

PAGE 170

139 Extraction profiles of five different compounds (each representative of a particular chemical class) were investigated on the sol-gel methyl capillary. For this, three replicate extraction experiments were performed for each compound for each of the following extraction periods: 10 min, 20 mi n, 40 min, 60 min, 80 min, 100 min, and 120 min. The average HPLC peak area for each extraction time was then plotted against the extraction time. The point on the graph at which the test compound stops increasing in peak area corresponds to the time required for the compound to reach equilibrium between the sample solution and the sol-gel coating representing the extraction medium (sorbent). Anthracene and 1-[1,1’-biphenyl ]-4-ylethanone reached equilibrium between 60 and 80 min of extraction. Diphenyl amine and 3,5-dimethylphenol reached equilibrium within 60 min of extraction. Be nzhydrol reached equilibrium within 40 min of extraction. The longer extraction time s required for nonpolar PAHs and moderately polar ketones are likely due to the presence of highly polar silanol groups in the coating. In all subsequent analyses, the aqueous samp les were extracted using the sol-gel methyl coated capillary for the amount of time require d for equilibrium to be established (80 min for PAHs and ketones, 60 min for phenols and amines, and 40 min for alcohols). The extraction profile is presented in figure 3.2.

PAGE 171

140 Figure 3.2 Extraction profiles of anthracene (5 x 104 ng/L), 1-[1,1’-biphenyl]-4ylethanone (5 x 104 ng/L), 3,5-dimethylphenol (5 x 105 ng/L), benzhydrol (5 x 105 ng/L), and diphenylamine (5 x 104 ng/L).

PAGE 172

141 The sol-gel methyl coating demonstrated excellent detection limits in CMEHPLC analysis with UV detection. The re peatability and detection limit data are presented in table 3.1. The HPLC peak area RSD values remained at 9.35% or less. Depending on the analyte types and their UV absorption characteristics, the observed detection limits were between 0.72 and 481 ng/L.

PAGE 173

142 Table 3.1 HPLC peak area repeatability and dete ction limit data for PAHs, ketones, phenols, alcohols, and amines in CMEHPLC using a sol-gel methyl coated microextraction capillary. Chemical Class Chemical Name Peak Area Repeatability (n = 3) Detection Limits (ng/L) Mean Peak Area (arbitrary unit) R.S.D. (%) (S/N = 3) PAHs anthracene 7.75.90.7 fluoranthene 5.56.51.0 fluorene 4.75.62.6 naphthalene 15.88.40.9 phenanthrene 9.21.71.2 acenaphthene 4.35.13.2 Ketones 4'phenylacetophenone7.14.45.0 benzophenone 2.51.55.8 coumarin 2.22.26.3 trans-chalcone 1.43.85.2 Phenols 2,3-dichlorophenol 3.18.746.0 2-chlorophenol 1.93.276.0 3,4-dimethylphenol 1.75.981.6 3,5-dimethylphenol 2.14.266.2 Alcohols benzhydrol 2.03.069.9 catechol 1.35.2107.0 resorcinol 0.91.7167.0 Amines caffeine 1.21.0481.0 diphenylamine 1.59.49.7 m -toluidine 1.66.789.0 Extraction conditions: 40 cm x 0.25 mm i.d. so l-gel methyl coated capillary; extraction times: 40 minutes for alcohols, 60 minutes for amines and phenols, 80 minutes for ketones and polycyclic aromatic hydrocarbons HPLC conditions: 15 cm x 4.6 mm i.d. Luna C18 column; isocratic elution 90/10 ACN/wa ter; 1 mL/min flow rate, UV detection at 254 nm for anthracene, fluoranthene, and phenanthrene, 217 nm for acenaphthene and naphthalene, 260 nm for fluorene, 200 nm fo r ketones, phenols, alcohols, and amines.

PAGE 174

143 The sol-gel methyl capillary showed ve ry impressive detection limits (0.7-3.2 ng/L) for polycyclic aromatic hydrocarbons, wi th peak area RSD values ranging from 1.73 to 8.41%. A chromatogram indicating the CME-HPLC analysis of three polycyclic aromatic hydrocarbons is shown in figure 3.3.

PAGE 175

144 Figure 3.3 A chromatogram representing on-line CME-HPLC analysis of PAHs using a sol-gel methyl coated microe xtraction capillary. Extrac tion conditions: 40 cm x 0.25 mm I.D. sol-gel methyl coated capillary, 80 minute gravity fed extraction at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 70:30 ACN/water, 1 mL/min flow rate, UV de tection at 217 nm, ambient temperature. Peaks: 1 = Naphthalene (2 x 103 ng/L), 2 = Acenaphthene (8 x 103 ng/L), 3 = Anthracene (1 x 104 ng/L).

PAGE 176

145 The sol-gel methyl capillary also dem onstrated excellent detection limits for moderately polar carbonyl compounds. Benzophenone, 1-[1,1’-biphenyl]-4-ylethanone coumarin, and trans-chalcone were tested using the sol-ge l methyl capillary in CMEHPLC using ultraviolet detection. All of these compounds have good UV absorbance at 200 nm. The detection limits for ketones ranged from 5.0 to 6.3 ng/L. The peak area RSD values ranged from 1.51 to 4.36% for ketones. A chromatogram representing the extraction of three ketones on the sol-gel met hyl coated capillary is shown in figure 3.4.

PAGE 177

146 Figure 3.4 A chromatogram representing on-line CME-HPLC of ketones using a sol-gel methyl coated capillary. Extraction conditi ons: 40 cm x 0.25 mm I.D. sol-gel methyl coated capillary, 80 minute gravity fed ex traction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 70:30 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambi ent temperature. Peaks: 1 = coumarin (2 x 104 ng/L), 2 = 1-[1,1’-biphe nyl]-4-ylethanone (2 x 104 ng/L), 3 = trans-chalcone (1 x 104 ng/L).

PAGE 178

147 Conventional wisdom might predict that th e sol-gel methyl coated capillary would be useful only for the extracti on of nonpolar analytes. Howe ver, contrary to this, our results show that the sol-gel methyl coated cap illary has the capabil ity of extracting polar analytes like phenols with detection limits ranging from 46.0 to 81.6 ng/L. The peak area RSD values observed for phenols ranged from 3.15 to 8.65%. The detection limits were good for phenols, but not as impressive as the detection limits observed for polycyclic aromatic hydrocarbons and ketones. This can be attributed to the UV absorbing characteristics of phenols, their higher polarit y (and therefore, their higher affinity for water) and the nonpolar nature of the methyl groups in the so l-gel capillary. One factor that is likely to play a positive role in the extraction of phenols on the sol-gel methyl coating is the molecular leve l interactions betw een the methyl groups in the sol-gel coating and the aromatic rings of the phenols. Any residual silanol and hydroxyl groups present in the sol-gel coating may also synergistically interact with the polar functional groups on the solute molecules and aid in the ex traction of polar analyt es like phenols. A chromatogram indicating the extraction of thr ee phenols on the sol-gel methyl capillary is shown in figure 3.5.

PAGE 179

148 Figure 3.5 A chromatogram representing on-line CME-HPLC analysis of phenols using a sol-gel methyl coated capillary. Extracti on conditions: 40 cm x 0.25 mm I.D. sol-gel methyl coated capillary, 60 minute gravity fe d extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 50:50 ACN/water, 1 mL/min flow rate, UV detection at 200 nm ambient temperature. Peaks: 1 = 2chlorophenol (2 x 105 ng/L), 2 = 3,4-dimethylphenol (1 x 105 ng/L), 3 = 2,3dichlorophenol (1 x 105 ng/L).

PAGE 180

149 The sol-gel methyl capillary also demonstrated the capab ility to extract aromatic alcohols with detection limits between 69.9 ng/L and 167.0 ng/L. At 200 nm, a detection limit of 69.9 ng/L was observed for benzhydrol. Since benzhydrol has two benzene rings and one hydroxyl group, it is not as polar as resorcinol and catechol which have only one aromatic ring and two hydroxyl groups. Surp risingly, the sol-gel me thyl coating showed the ability to extract both resorcinol and catechol with detection limits of 167.0 and 107.0 ng/L, respectively. Extraction is possible through interactions between the methyl groups in the coating and the aromatic rings in the alcohols. Residual silanol groups in the coating can also contribute to the extracti on of alcohols, particularly the more polar alcohols. A chromatogram depicting the extrac tion of three alcohols on the sol-gel methyl coated capillary is indicated in figure 3.6.

PAGE 181

150 Figure 3.6 A chromatogram representing on-line CME-HPLC analysis of alcohols using a sol-gel methyl coated capillary. Extr action conditions: 40 cm x 0.25 mm I.D. sol-gel methyl coated capillary, 40 minute gravity fe d extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 40:60 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambi ent temperature. Peaks: 1 = resorcinol (3 x 105 ng/L), 2 = catechol (4 x 105 ng/L), 3 = benzhydrol (2 x 105 ng/L).

PAGE 182

151 Amines were also extracted on the so l-gel methyl capillary. The observed detection limits for amines at 200 nm UV de tection were 9.7 ng/L for diphenylamine, 89.0 ng/L for m -toluidine, and 481 ng/L for caffeine. These detection limits make sense if the molecules are examined. Diphenylamine contains two phenyl rings with an amine group in the middle. The two phenyl rings are nonpolar and are extracted well by the methyl groups in the coating. The molecule of m -toluidine contains only one aromatic ring with an amine group on it, so it does not extract as well as diphenylamine on the solgel methyl coating. Caffeine has the highest detection limit of the amines and for all of the compounds tested on the sol-gel methyl capillary. This might be due to the four nitrogen and two oxygen atoms in the compound, which make it much more polar than diphenylamine and m -toluidine. Since it is more polar in nature, it is not extracted as well by the methyl groups in the sol-gel coa ting. The residual silanol groups in the coating can contribute to this extraction. A chromatogram depicting the extraction and separation of three amines is indicated in figure 3.7.

PAGE 183

152 Figure 3.7 A chromatogram representing on-line CM E-HPLC analysis of amines using a sol-gel methyl coated capillary. Extracti on conditions: 40 cm x 0.25 mm I.D. sol-gel methyl coated capillary, 60 minute gravity fe d extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 70:30 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambi ent temperature. Peaks: 1 = caffeine (8 x 105 ng/L), 2 = m-toluidine (2 x 105 ng/L), 3 = diphenylamine (2 x 104 ng/L).

PAGE 184

153 Differences in UV absorbing characteristic s of the analytes in the preceding examples also contribute to the observed differences in their detection limits. The sol-gel methyl coated capillary contained a small amount of PMHS deactivator (0.02 g), contributing additional methyl groups to the sol-gel coating. Therefore, to have an idea about the relative contribution of methyl groups from MTMS and PMHS, we performed some additional experi ments. For this, a sol-gel methyl coated capillary was prepared using no deactivator. The same sample, consisting of both polar and nonpolar analytes, each representing a partic ular chemical class, was then extracted on both capillaries. The sol-gel methyl coated capillary with PMHS deactivator extracted these analytes very similarly to the sol-gel me thyl coated capillary with no deactivator. This indicates that the MTMS-based sol-ge l network is primarily responsible for the extraction. Both sol-gel methyl coated capi llaries were capable of extracting both polar and nonpolar analytes from the same sample. Chromatograms in figure 3.8 illustrate these observations. A peak area co mparison is shown in table 3.2. Since the extraction of polar analytes on the sol-gel methyl coated capillary is counterintuitive, it was also necessary to verify that the peaks observed in the chromatograms actually came from extraction and not just from the liquid sample filling the capillary. To verify this, the same sample that was extracted using the sol-gel methyl coated capillaries in figure 9 was passed th rough an uncoated fused-silica capillary of the same size (40 cm x 0.25 mm i.d.) for the same amount of time and under the same conditions. Only three very small peaks ar e visible in chromatogram C in figure 3.8. The peak areas observed for the analytes usi ng the uncoated capillary are shown in table 3.2.

PAGE 185

154 Figure 3.8 A Figure 3.8 B Figure 3.8 C Figure 3.8 Chromatograms representing on-line CME-HPLC analysis of polar and nonpolar compounds using; (3.8 A) a sol-gel me thyl coated capillary deactivated with PMHS; (3.8 B) a sol-gel methyl coated cap illary prepared without a deactivator; (3.8 C) an uncoated capillary. Extraction c onditions: 40 x 0.25 mm I.D. capillary, 80 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm i.d. Luna C18 column, isocratic elution 50:50 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. Peaks: 1 = m -toluidine (2 x 105 ng/L), 2 = 3,4-dimethylphenol (1 x 105 ng/L), 3 = benzhydrol (1 x 105 ng/L), 4= 1-[1,1’-biphe nyl]-4-ylethanone (2 x 104 ng/L), and 5 = naphthalene (2 x 104 ng/L).

PAGE 186

155 Table 3.2 Peak area comparison for m -toluidine, 3,4-dimet hylphenol, benzhydrol, 4’phenylacetophenone, and naphthalene on the solgel methyl capillary deactivated with PMHS, the undeactivated sol-gel methyl coated capillary, and an uncoated capillary. Capillary: Sol-Gel Methyl PM HS Sol-Gel Methyl Uncoated Chemical Name: Peak Area (arbitrary unit) m -toluidine 6.16.0 0.7 3,4-dimethylphenol 4.54.6 0.4 benzhydrol 8.79.1 0.3 4'phenylacetophenone 31.830.7 no peak naphthalene 22.918.9 no peak Extraction conditions: 40 x 0.25 mm I.D. cap illary, 80 min gravity -fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm i.d. Luna C18 column, isocratic elution 50:50 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature.

PAGE 187

156 For the concentrations te sted on the sol-gel methyl coated capillary (2 x 103 ng/L to 8 x 105 ng/L) no significant peaks were observed from the liquid sample filling the uncoated capillary. At higher co ncentrations (well over 1 x 106 ng/L), more significant peaks could be observed. However, at concentr ations this high, it is not necessary to use capillary microextraction to preconcentrate sa mples, as direct injection would suffice. Aside from being an excellent extraction device, the sol-gel me thyl capillary was found to be very stable and rugged. It is ab le to withstand high temperatures: it was conditioned right to 350C with no bleeding problems. This could be a problem for capillaries prepared using polymers. It was also rinsed with organic solvents such as methylene chloride, methanol, and acetonitrile. This system allowed for an effective online hyphenation of capillary microextraction w ith HPLC. Also, this makes automation a possibility in the future since the extraction is performed on-line with HPLC. The sol-gel methyl coated capillary demonstrates ex cellent durability and prolonged operation lifetime. For example, one piece of the sol-ge l methyl coated capillary survived for one year and 8 months and 326 extractions wit hout a significant change in extraction performance. A sample of trans-chalcone was the fifteenth sample extracted on the solgel methyl coated capillary. One year and eight months, and 311 ex tractions later, the same sample was extracted on the sol-gel methyl coated capillary with only a 3.63% change in peak area. The chromatograms in figure 3.9 demonstrate the durability and longevity of the sol-gel me thyl coated capillary.

PAGE 188

157 Figure 3.9 A Figure 3.9 B Figure 3.9 On-line CME-HPLC analysis of trans-ch alcone using a sol-gel methyl coated capillary. (3.9 A) Extraction # 15; (3.9 B) Extrac tion # 326, one year and eight months later. Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel MTMS coated capillary, 80 min gravity -fed extraction at room temperat ure. HPLC conditions: 12.5 cm x 4.6 mm I.D. Keystone Betasil C8 column, isocratic elution 90:10 ACN/water, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. HPLC peak area RSD of 3.63%.

PAGE 189

158 3.4 Conclusion To the best of our knowledge, this is the first report on the use of a sol-gel methyl coating in capillary microext raction. This study indicates th at pendant methyl groups of the sol-gel coating created from methyltrim ethoxysilane is capable of extracting a wide range of analytes, both nonpolar and polar. Th e ability of the sol-ge l methyl capillary to extract polar analytes is count erintuitive. A synergistic inte raction of silanol and methyl groups of the sol-gel coating w ith polar group(s) and aromatic ring of polar analytes may be responsible for this apparently anomal ous extraction behavior. The sol-gel methyl coating demonstrated good run-to-run repe atability and detection limits (0.72 and 6.3 ng/L) for nonpolar and moderately polar analytes in CME-HP LC with UV detection. For the tested polar analytes, the sol-gel methyl coating also provided good run-to-run repeatability and detection limits (9.7– 481 ng/L) in CME-HPLC analysis with UV detection. The newly developed sol-gel methyl coating has the potential to become an excellent all-purpose sorbent for the simulta neous extraction of both polar and non-polar analytes from an aqueous sample. It is resi stant to solvents, very durable, and thermally stable to over 350C. The present work provide s a general guideline fo r facile creation of new sol-gel coatings with pendant groups fr om sol-gel precursors alone without the use of sol-gel active organic components or polymers. 3.5 References for chapter three [1] C. Arthur, J. Pawlisz yn, Anal. Chem. 62 (1990) 2145. [2] J. Hinshaw, LC-GC N. Am. 21 (2003) 1056. [3] R. Eisert, K. Levsen, J. Chromatogr. A 733 (1996) 143.

PAGE 190

159 [4] H. Lord, J. Chromatogr. A 1152 (2007) 2. [5] C. Zambonin, Anal. Bioa nal. Chem. 375 (2003) 73. [6] E. Lesellier, Analusis 27 (1999) 363. [7] P. Santos, B. Simonet, A. Rios, M. Valcarcel, Electr ophoresis 28 (2007) 1312. [8] A. Penaluer, E. Pocurull, F. Borrull, R. Marce, Trends Anal. Chem. 18 (1999) 557. [9] G. Jiang, M. Huang, Y. Cai, Z. Zongs han, J. Chromatogr. Sci. 44 (2006) 324. [10] M. Yang, Z. Zeng, W. Qiu, Y. Wang, Chromatographia 56 (2002) 73. [11] Y. Hu, Y. Yang, J. Huang, G. Li, Anal. Chim. Acta 543 (2005) 17. [12] D. Zhang, C. Wu, F. Ai, Chin. J. Chromatogr. 17 (1999) 10. [13] R. Yang, W. Xie, Fore nsic Sci. Int. 139 (2004) 177. [14] T. Ding, H. Lin, C. Whang, J. Chromatogr. A 1062 (2005) 49. [15] M. Farajzadeh, M. Hatami, J. Sep. Sci. 26 (2003) 802. [16] E. Koster, C. Crescenzi, W. Hoedt, K. Ensing, G. de Jong, Anal. Chem. 73 (2001) 3140. [17] W. Mullett, P. Martin, J. Pa wliszyn, Anal. Chem. 73 (2001) 2383. [18] F. Mangani, R. Cenciarini, Chromatographia 41 (1995) 678. [19] H. Kataoka, Anal. Bioanal. Chem. 373 (2002) 31. [20] S. Chong, D. Wang, J. Hayes, B. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [21] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [22] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [23] D. Wang, S. Chong, A. Ma lik, Anal. Chem. 69 (1997) 4566. [24] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157.

PAGE 191

160 [25] Z. Zeng, W. Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [26] J. Zhou, Z. Zeng, Anal. Chim. Acta 556 (2006) 400. [27] L. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [28] Z. Zeng, W. Qiu, Z. Hu ang, Anal. Chem. 73 (2001) 2429. [29] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [30] C. Basheer, S. Jegadesan, S. Valiyavee ttil, H. Lee, J. Chromatogr. A 1087 (2005) 252. [31] A. Kabir, C. Hamlet K. Yoo, G. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [32] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [33] T. Kim, K. Alhooshani, A. Kabir, A. Malik, J. Chromatogr. A 1047 (2004) 165. [34] Y. Fan, Y. Feng, S. Da, Z. Wang, Talanta 65 (2005) 111. [35] J. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3. [36] C. J. Brinker, G.W. Scherer, Sol-Gel Science: The Chemistry and Physics of Sol-Gel Processing Academic Press, San Diego, CA, 1990. [37] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [38] H. Fang, M. Liu, Z. Zeng, Talanta 68 (2006) 979. [39] W. Liu, Y. Hu, J. Zhao, Y. X u, Y. Guan, J. Chromatogr. A 1102 (2006) 37. [40] M. Liu, Z. Zeng, H. Fa ng, J. Chromatogr. A 1076 (2005) 16. [41] K. Alhooshani, T. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [42] M. Liu, Z. Zeng, C. Wang, Y. Ta n, H. Liu, Chromatographia 58 (2003) 597. [43] X. Li, Z. Zeng, J. Zhou, Anal. Chim. Acta 509 (2004) 27.

PAGE 192

161 [44] D. Wang, J. Xing, J. Peng, C. Wu, J. Chromatogr. A 1005 (2003) 1. [45] R.C. Mehrotra, J. NonCryst. Solids, 121 (1990) 1. [46] S. Kulkarni, A. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50. [47] A. L. Lopes, F. August o, J. Chromatogr. A 1056 (2004) 13. [48] R. Rodriguez, M. Flores, J. Gomez, V.M. Castano, Mater. Lett. 15 (1992) 242. [49] L. Fang, S. Kulkarni, K. Alhoosha ni, A. Malik, Anal. Chem. 79 (2007) 9441.

PAGE 193

162 CHAPTER FOUR SOL-GEL COATINGS WITH COVALEN TLY ATTACHED METHYL-, OCTYL, AND OCTADECYL LIGANDS FOR CAPILLARY MICROEXTRACTION. EFFECTS OF ALKYL CHAIN LEN GTH AND SOL-GEL PRECURSOR CONCENTRATION ON EXTRACTION BEHAVIOR 4.1 Introduction Sol-gel sorbents [1-7] have proven to be quite successful in the extraction of a wide range of analytes by solid-pha se microextraction (SPME). SPME is environmentally friendly, since it completely eliminates the use of hazardous organic solvents in the extraction process [8]. So l-gel extracting phases added a new dimension to the original SPME technique developed by Pawliszyn and co-workers in 1989 [9]. Traditional SPME extracting phases consist of different polysiloxanes [10-13], other polymeric sorbents [14-16], molecularlyimprinted polymers (MIPs) [17,18], and inorganic substances [19]. SPME has traditionally been coupled with gas chromatography (GC) [20], but attempts have also been made to extend its coupling to other separation techniques, including highperformance liquid chromatography (HPLC) [21,22], supercritical fluid chromatography (SFC) [23], and capil lary electrophoresis (CE) [24,25]. Conventional fiber SPME, however, has some inherent shortcomings. The sorbent-coated segment of the fiber is very fr agile and since the coating is situated on the outer surface of the fiber it is susceptible to mechanical damage during operation of the syringe-like SPME device [22]. Metal-base d SPME fibers have been developed to

PAGE 194

163 overcome the problems associated with fibe r breakage [26]. Si nce only a short end segment (~1cm) of the fiber is coated with the extracting phase, fiber SPME is characterized by low sample capacity. Moreove r, fiber SPME is not easily coupled to HPLC, as cumbersome desorbing devices are needed [27]. In-tube SPME was developed to address these problems associ ated with the fiber format of SPME. Initial research on in-tube SPME involved the use of GC capill ary column segments to perform the extraction. Such capillaries were characteri zed by relatively thin (sub-micrometer in thickness) stationary phase co atings which compromised the sample capacity [28]. Also, in general, these conventional in-tube SPME coatings were not chemically bonded to the surface of the capillary [1], which made them unstable under organo-aqueous mobile phase conditions typically used to desorb the extracted analytes into an HPLC system. Poor solvent stability of physically held sorbent coatings limited the widespread hyphenation of in-tube SPME with HPLC. Sol-gel coatings can easily overcome the above mentioned shortcomings of conventional fiber and in-tube SPME coatings. In sol-gel fiber SPME [1], pioneered by our group, an organic-inorganic hybrid extr acting phase is chemically bonded to the external surface of the fused silica fiber. This chemical bonding makes sol-gel extracting phases highly resistant to organi c solvents and provides stability to high temperatures [1]. Sol-gel fiber SPME has been applied to numer ous real-world applic ations, including the extraction of a variety of analytes in envi ronmental [6,29-34], food [35-39], and clinical [5, 40-43,] samples. To materialize the same advantages in the capillary format, our group also introduced sol-gel capillary micr oextraction (CME) [44] – a technique in which a sol-gel coated capillary is used for extraction. Sol-gel tec hnology allows for the

PAGE 195

164 creation of microextraction cap illaries with thicker extrac ting phase coatings (compared to typical coating thicknesses in GC capillary segments used in traditional in-tube SPME) and thereby enhances the extr action capacity of the techni que [44]. Sol-gel coatings provide enhanced thermal and solvent stability in CME since the they are covalently anchored to the capillary wa lls [44,45]. Sol-gel capillary microextraction has been applied to the extract ion of metals in biological samp les [46,47] and the extraction of drugs in human urine samples [48]. Solgel PDMS [44], PEG [49], dendrimer [50], poly(tetrahydrofuran) [51], and cyano-PD MS [52] coatings have demonstrated significantly higher thermal stability in CME-GC analysis. The solvent stability of solgel coatings makes them especially suita ble for on-line hyphenation with HPLC, which uses organo-aqueous mobile phases to desorb the extracted analytes from the sol-gel coated capillary. The successful coupling of sol-gel CME to HPLC is poised to provide a powerful analytical tool for trace analysis. To this end, sol-ge l titania-PDMS [53], cyclodextrin [48], methyl [ 54], poly(dimethyldiphenylsil oxane) (PDMDPS) [55], and titania poly(tetrahydrofuran) [56] coated capillaries have been used in CME-HPLC. Solgel coated microextraction capillaries are quite durable. Sol-gel methyl coated capillaries have been reused in over 300 extractions with little change in extraction performance [54]. Most sol-gel coatings in the literature were prepared using sol-solutions containing sol-gel active polymeric materials together with the sol-gel precursors. During the coating process, these polymeric materials get bonded to the resulting sol-gel network to provide organic-i norganic hybrid extraction phases covalently anchored to the capillary surface. Recently, we reported th e preparation of a sol-gel methyl coated

PAGE 196

165 microextraction capillary hyphe nated on-line with HPLC [54] This approach did not require the inclusion of a solgel active organic ligand or polymer in the sol solution. This coating efficiently extracted nonpolar and moderately polar analytes, and even showed a counterintuitive extr action capability for polar analytes. This study also demonstrated that short alkyl chains covalently attached to the silicon atom in trialkoxysilane sol-gel precursors can serve as the pendant liga nds in the resulting sol-gel coating to provide extraction [ 54]. It should be pointed out that sol-gel SPME coatings with methyl and octyl ligands have been re ported in sol-gel fiber SPME [57]. Recently, Zheng et al. [58] described hyphenation of a sol-gel octyl monolithic microextraction capillary to HPLC. Silicabased coatings containing C18 have also been used for in-tube SPME coupled with HPLC [59,60]. Howeve r, to our knowledge, there have been no systematic studies on the extrac tion behaviors of sol-gel coat ings with alkyl chains of different lengths. In this paper, we report our research on sol-gel alkyl coa tings created for CME using MTMS, C8TMS, and C18TMS as sol-gel precursors pr oviding alkyl chains of different lengths (C1, C8, and C18, respectively) and desire d concentrations in the resulting sol-gel coati ng. We also report our findings on the extraction be haviors of these coatings using aqueous samples containi ng nonpolar, moderately polar, and polar analytes with an aim to understand the effect of alkyl chain length and their concentration in the sol solution on the extraction capabiliti es of the resulting sol-gel coatings.

PAGE 197

166 4.2 Experimental 4.2.1 Equipment Sol-gel CME-HPLC experiments were c onducted using a MicroTech Scientific (Vista, CA, USA) Ultra Plus HPLC system equipped with a Linear UVIS 200 variable wavelength UV detector. A reversed-phase Luna C18 column (15 cm X 4.6 mm i.d.) was used for HPLC analysis. For adequate mixi ng of the sol solutions, a Fisher model G-560 Genie 2 Vortex system (Fisher Scientif ic, Pittsburgh, PA, USA) was employed. Nanopure water (15 M ) was prepared using a Barnstead model 04741 Nanopure deionized water system (Barnstead/Therm odyne, Dubuque, IA, USA). For on-line data collection and processing, Ch rom Perfect version 3.5 (for Windows) computer software (Justice Laboratory Software, Denville, NJ, USA) was used. 4.2.2 Chemicals and materials C18TMS and C8TMS were purchased from Gele st (Morrisville, PA, USA). MTMS, poly(methylhydrosiloxane) (PMHS) PAHs (fluorene, naphthalene, phenanthrene, and acenaphthene), ketones (benzophenone, coumarin, and trans chalcone), phenols (2,3-dichlorophenol, 2-chlorophenol, 3,4-dimethylphenol, and 3,5dimethylphenol), and amines ( o and m -toluidine) were purch ased from Aldrich (Milwaukee, WI, USA). Anthracene and 1-[1,1’ -biphenyl]-4-yl-ethanone were purchased from Eastman Kodak (Rochester, NY, USA) Diphenylamine was purchased from J.T. Baker (Phillipsburg, NJ, USA). Trifluoroace tic acid was purchased from Acros (Morris Planes, NJ, USA). HPLC-grade acetonitrile (ACN), methanol, and methylene chloride were obtained from Fisher Scientific (Pitts burgh, PA, USA). Fused silica capillary (0.25 mm I.D.) was purchased from Polymicr o Technologies (Phoenix, AZ, USA).

PAGE 198

167 4.2.3 Pretreatment of fused silica capillary The inner surface of 0.25 mm I.D. fused sili ca capillary was pretreated prior to the creation of the sol-gel octadecy l-, octyl-, and methyl coatings on it. First, the capillary was sequentially rinsed with 4 mL each of methylene chloride, methanol, and nanopure water. Following this, the capillary was brie fly purged with helium (1 min) leaving only a thin layer of water on the inner walls. Both ends of the capillary were then sealed using an oxy-acetylene torch and the sealed capillary was placed in a GC oven and heated at 350 0C for 2 h. It was subsequently removed from the GC oven and allowed to cool down to room temperature. Finally, the ends of the capillary were cut open using an alumina wafer, and the capillary was installe d in a GC oven and purged with a constant helium flow (1 mL/min). Simultaneously, the oven temperature was programmed from 40 0C to 250 0C, at 5 0C/min. The capillary was held at the final temperature of 250 0C for 2 h, followed by cooling to room temperature. 4.2.4 Preparation of the sol-gel capillaries A total of nine sol-gel methyl, octyl, and octadecyl coated capillaries (three capillaries of each type) were prepared using sol solutions with constant molar concentration of the precursor. To study the effect of precursor concentration on extraction behavior, four additi onal sol-gel octyl coated capi llaries were prepared using sol solutions with varied molar concentration of the precursor. 4.2.4.1 Preparation of sol-gel microextracti on capillaries using constant molar concentration of the precursors Three sol-gel coating solutions were se parately prepared by first weighing 0.22 g of PMHS into each of three clean mi crocentrifuge tubes. Then, 2.47 x 10-4 moles of sol-

PAGE 199

168 gel precursor were added to each microcentrifuge tube, followed by vortexing for 30 s. Next, methylene chloride was added (86.8 L, 126 L, and 153.1 L for the sol-gel octadecyl, octyl, and methyl capillaries, re spectively) to the microcentrifuge tubes, followed by vortexing again for 30 s. Finall y, 75 L of 99% trifluoroacetic acid was added to the mixtures, followe d by vortexing for 1 min. With respect to individual sol-gel precursors, the prepared solutions had c onstant molar concentration (0.514 M) and constant total volume (480.5 L). The compos itions of these sol-gel solutions are given in table 4.1.

PAGE 200

169 Table 4.1 Compositions of the sol solutions used to prepare the sol-gel alkyl coated capillaries with constant molar concentra tion of the sol-gel pr ecursor. The moles of PMHS were calculated based on the aver age molecular weight of 2450 g/mol. ________________________________________________________________________ Sol-gel Coatings Ingredients C1 C8 C18 Sol-Gel MTMS C8TMS C18TMS Precursor (moles) 2.47 x 10-4 2.47 x 10-4 2.47 x 10-4 Deactivating Reagent (PMHS) (moles) ~8.98 x 10-5 ~8.98 x 10-5 ~8.98 x 10-5 Solvent (CH2Cl2) (moles) 2.39 x 10-3 1.97 x 10-3 1.36 x 10-3 Catalyst (99% TFA) (moles) 1.00 x 10-4 1.00 x 10-4 1.00 x 10-4 Total Volume (L) 480.5 480.5 480.5 Sol-Gel Precursor Concentration (M) 0.514 0.514 0.514 ________________________________________________________________________

PAGE 201

170 Pretreated fused silica ca pillary segments (60 cm x 0.25 mm I.D.) were then coated with these sol solutions. To accomplis h this, the sol solution was allowed to flow through the capillary un der nitrogen pressure (10 psi) using a capillary filling/purging device [61]. The capillary was capped with a piece of rubber septum at its exit end and kept under nitrogen pressure (50 psi) for 20 min. The se ptum was then removed, and nitrogen was allowed to flow through the cap illary for 70 min at 50 psi. Following this, the capillary was placed in a GC oven and cons tantly purged with helium (1 mL/min). The temperature of the GC oven was simultaneously programmed from 40 0C to 350 0C, at 5 0C/min. It was held at the final temperatur e for 2 h. The capillary was further cooled to room temperature followed by rinsing with a mixture of 1 mL each of methanol and methylene chloride. Finally, the capillary wa s placed in the GC oven again purging with constant helium flow (1 mL/min) a nd programming the temperature from 40 0C to 350 0C, at 5 0C/min, with a hold time of 30 min at the final temperature. A total of three solgel alkyl capillaries of each type (octadecyl, octyl, and methyl) were prepared under identical conditions. 4.2.4.2 Preparation of sol-gel octyl microextraction capi llaries using varied molar concentrations of sol-gel precursor and constant solution volume To study the effect of sol-gel precursor molar concentration in the sol solution on the extraction behavior, four other octyl coati ngs were prepared using sol solutions with molar concentrations of 0.257 M, 0.514 M, 1.028 M, and 1.542 M with respect to the precursor. For this, 0.22 g of PMHS was adde d to four clean microcentrifuge tubes. Next, C8TMS was added to the micr ocentrifuge tubes (1.24 x 10-4 2.47 x 10-4, 4.94 x 104, and 7.41 x 10-4 moles, respectively), followed by vortexing for 30 s. Following this,

PAGE 202

171 methylene chloride was added (156.4 L, 126 L, 65.2 L, and 4.4 L) to the respective microcentrifuge tubes, again followed by vor texing for 30 s. Finally, 75 L of 99% trifluoroacetic acid was added to each mi xture, followed by vortexing for 1 min. Four pretreated fused silica capillaries were co ated and conditioned following the same exact procedure as described in section 4.2.4.1. The compositions of these sol-gel octyl coating solutions are given in table 4.2.

PAGE 203

172 Table 4.2 Compositions of the sol solutions used to prepare the sol-gel octyl coated capillaries with varied molar concentration of the sol-gel precursor. The moles of PMHS were calculated based on the average molecular weight of 2450 g/mol. ________________________________________________________________________ C8TMS Concentration (M) 0.257 0.514 1.028 1.542 C8TMS (moles) 1.24 x 10-4 2.47 x 10-4 4.94 x 10-4 7.41 x 10-4 Deactivating Reagent (PMHS) (moles) ~8.98 x 10-5 ~8.98 x 10-5 ~8.98 x 10-5 ~8.98 x 10-5 Solvent (CH2Cl2) (moles) 2.44 x 10-3 1.97 x 10-3 1.02 x 10-3 6.87 x 10-5 Catalyst (99% TFA) (moles) 1.00 x 10-4 1.00 x 10-4 1.00 x 10-4 1.00 x 10-4 Total Volume (L) 480.5 480.5 480.5 480.5 ________________________________________________________________________

PAGE 204

173 4.2.5 On-line CME-HPLC analysis using sol-gel C8and C18-bonded microextraction capillaries Figure 4.1 is a schematic of the set-up used for the on-line sol-gel CME-HPLC analysis. In order to perform capillary micr oextraction, 40 cm long pi eces of the sol-gel octadecyl-, octyl-, and methylcoated capillaries were indi vidually installed on the HPLC six-port injection valv e using the capillary as an external samp ling loop. Prior to extraction, the HPLC column was equilibrated with the appropriate mobile phase. One end of a piece of deactivated fused silica capillary serving as a transfer line (60 cm x 0.53 mm I.D.) was connected to the lower end of a vertically placed gravity-fed sample delivery system [44]. A syringe needle attached to the ot her end of the capillary was inserted into the HPLC injection port. Extraction was performed by allowing liquid samples (placed in the sample delivery sy stem) to pass through the 0.53 mm I.D. fused silica capillary tr ansfer line, through the syringe n eedle, into the HPLC injector (maintained in the “load” position), and through the sol-gel coated capillary (flow rate ~ 0.2 mL/min) until extraction equi librium was reached between the sample and the coated sol-gel extracting phase (typi cally 40-60 min, depending on th e sample). The extracted analytes were then desorbed from the so l-gel coating by the HPLC mobile phase and transferred to the HPLC column for separa tion by turning the injector valve to the “inject” position. Isocratic and gradient m odes of elution were used with appropriate mobile phase (H2O/ACN) compositions and program rates.

PAGE 205

174 Figure 4.1 The experimental setup used to perfor m the sol-gel CME-HPLC experiments. During extraction, sample solution flows from the gravity-fed sample dispenser, through the sol-gel alkyl coated capillary, and into waste. During the analysis, the valve is switched to the inject positi on, and the mobile phase flows through the sol-gel alkyl coated capillary, desorbing the analytes a nd eluting them to the HPLC column for separation followed by UV detection.

PAGE 206

175 4.3 Results and discussion A sol-gel coating is formed through hydr olytic polycondensation of one or more sol-gel precursors, which forms a colloidal syst em, often referred to as the sol [62]. As the polycondensation process progresses further, the sol forms a three-dimensional network in a continuous liquid phase known as ge l [62]. For typical silica-based sol-gel microextraction coatings [30,32,63-68], the main sol solution ingredie nts include the solgel precursor (typically al koxysilanes [69]), the sol-ge l active polymer (e.g., hydroxylterminated poly(dimethylsiloxane) (PDMS) [ 1,44], poly(ethylene glyc ols) (PEGs) [49], crown ethers [4], cyclodextrins [5], etc.), the deactiv ating reagent (e.g., poly(methylhydrosiloxane) (PMHS) and/or 1,1,1,3,3,3-hexamethyldixilazane [52,53,62]), water (hydrolysis agent), solvent (e.g., acetone [70], methylene chloride [53], isopropanol [62], and THF [7]), and the solgel catalyst (acid [1,44], base [7 ], or fluoride [71]). In this work, sol-gel alkyl (methyl, octyl, and octa decyl) coated capillaries were prepared to study the dependence of extraction behavior on the alkyl chain length. To study the effects of precursor concen tration on the extraction beha vior, sol-gel octyl coated capillaries were prepared using sol solutions with different molar concentrations of C8TMS (0.257 M, 0.514 M, 1.028 M, and 1.542 M). 4.3.1 Microextraction capillaries prepared us ing sol solution with constant molar concentration of the precursor (Table 4.1) The sol solutions used to create the octad ecyl, octyl, and methyl coated capillaries were prepared using the same number of mole s of each sol-gel alkyl precursor as well as the same amount of 99% trifluoroacetic acid (catalyst) and consta nt amount of PMHS (deactivating reagent). Different amounts of methylene chlo ride (solvent) were added to

PAGE 207

176 equalize the final volumes of the three sol so lutions. This volume adjustment was needed to keep the molar concentration of the solgel precursors constant since equal number of moles of the three precursors had different volumes due to different molecular weights and densities of the sol-gel precursors. W ith equal final volumes and equal numbers of moles of sol-gel precursor, all three sol-gel so lutions had an equal mo lar concentration of 0.514 M with respect to the used individual precursor. Since each sol-gel precursor molecule has one alkyl chain, the number of alkyl chains in each sol solution should theoretically be equal, thus l eaving the length of the alkyl ch ains as the variable between the three different coatings. All three sol-gels had similar appear ances and gelation times (~50 min). A general scheme for the reacti ons leading to the formation of surface-bonded sol-gel octadecyl, octyl, and methyl coatings is shown in scheme 4.1. After hydrolysis and polycondensation, part of the so l-gel network growing in the vicinity of the capillary walls gets chemically bonded to silanol groups on the inner surface of the fused silica capillary through condensation reactions, forming a surface-bonded coating. After expelling the unbonded bulk sol solution from th e capillary under pr essure, the sol-gel coated capillary was purged with nitrogen and subjected to thermal conditioning to accelerate the aging process [44].

PAGE 208

177 R = CH3 for MTMS (CH2)7CH3 for C8TMS (CH2)17CH3 for C18TMS Scheme 4.1 Chemical reactions involved in the form ation of the sol-gel octadecyl, octyl, and methyl coatings.

PAGE 209

178 Extraction profiles were constructed for th e sol-gel octadecyl, octyl, and methyl coated capillaries to determine the time re quired to reach extraction equilibrium between the coating and the sample solution. This was accomplished by extracting representative test analytes: one nonpolar (PAH), one moderate ly polar (ketone), and one polar (phenol) analyte. Three replicate measurements were performed for each test analyte at each of the extraction times set at 10 min, 20 min, 40 min, 60 min, and 80 min. The HPLC peak areas for each extraction time were subse quently averaged and plotted against the corresponding extraction time. The points on the graphs where the average peak areas level off signify the onset of extraction e quilibrium. Figures 4.2, 4.3, and 4.4 depict the extraction profiles for naphthalene (figure 4.2), 1-[1,1’-biphenyl]-4-yl-ethanone (figure 4.3), and 3,4-dimethylphenol (figure 4.4) on the sol-gel octadecyl, octyl, and methyl coated capillaries.

PAGE 210

179 Figure 4.2 Extraction profile of 5 x 104 ng/L naphthalene on sol-gel octadecyl, octyl, and methyl coated capillaries.

PAGE 211

180 Figure 4.3 Extraction profile of 5 x 104 ng/L 1-[1,1’-biphenyl]-4-yl-ethanone on sol-gel octadecyl, octyl, and methyl coated capillaries.

PAGE 212

181 Figure 4.4 Extraction profile of 2 x 105 ng/L 3,4-dimethylphenol on sol-gel octadecyl, octyl, and methyl coated capillaries.

PAGE 213

182 The extraction profiles revealed that naphthalene required 20 min of extraction time on sol-gel octyl coated capillary, 40 mi n of extraction on the sol-gel octadecyl coated capillary, and 60 min of extraction on th e sol-gel methyl coated capillary to reach equilibrium. For all three sol-gel coated capillaries, 1-[1,1’-bi phenyl]-4-yl-ethanone required approximately 60 min of extr action and 3,4-dimethylphenol required approximately 40 min of extraction to reach equilibrium. In subsequent analyses, all analytes were extracted for at least the mi nimum extraction time required for equilibrium to be established between the analytes and al l three sol-gel coatings (40 min for phenols, 60 min for ketones, amines, and PAHs). Since the sol-gel methyl coated capillary has the shortest nonpolar alkyl chain, it was expected to be less efficient in extrac ting nonpolar and modera tely polar analytes than the sol-gel octyl coated capillary, and mu ch less efficient than the sol-gel octadecyl coated capillary, which has the longest nonpol ar alkyl chain. Expe rimental evidence suggests that this pattern does follow when th e capillaries are prepar ed using sol solution composition described in table 4.1. Apparentl y, ketones and PAHs are extracted mostly through dispersion [72] and induc tion interactions [73] between the molecules of these analytes and the alkyl chains within the so l-gel coatings. The chromatograms in figure 4.5 illustrate the differences in the extraction behavior observed for the sol-gel octadecyl (figure 4.5 A), octyl (figure 4.5 B), and methyl (figure 4.5 C). From the three chromatograms, it is clear that the octadecyl coated capillary extracts all of the test compounds most efficiently, followed by the octyl coated capillary, followed by the methyl coated capillary.

PAGE 214

183 Figure 4.5 A Figure 4.5 B Figure 4.5 C Figure 4.5 Chromatograms representing CME-HPLC -UV analysis of amines, phenols, ketones, and PAHs using sol-gel alkyl coat ed capillaries. (Figure 4.5 A) octadecyl (Figure 4.5 B) octyl, (Figure 4.5 C) methyl. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 60 min gravity-fed extraction, at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/ 50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (5 x 104 ng/L), 3 = diphenylamine (1 x 104 ng/L), 4 = trans -chalcone (1 x 104 ng/L), 5 = acenaphthene (1 x 104 ng/L).

PAGE 215

184 In the extraction of amines and phenols it would be expected that the longer nonpolar C18 chains would be less effective in the extraction of these compounds, but the experimental results indicate that th e sorbent with longer alkyl chains (C18) is capable of more efficient extraction of amines and phenol s than the shorter methyl or octyl chaincontaining sol-gel extracting pha ses. This indicates that the extraction of amines and phenols likely occurs through s ynergistic interacti on of both the silanol groups and the alkyl chains in the sol-gel co atings providing molecular level interactions with the polar and nonpolar sites on the amine and phenol mo lecules, as suggested by our previous work [54]. In our previous studies [44,55] on sol-gel coatings with nonpolar sol-gel active polymers in capillary microextraction, the polymers were primarily responsible for extraction of nonpolar analytes. These nonpolar polymers blocked some of the silanol groups in these coatings, rendering them less effe ctive in the extraction of polar analytes. Amines and phenols are presumably extracted by dispersion interactions [72] between the aromatic portions of their molecules and the al kyl chains in the sol-gel coatings, as well as by induction interactions [ 73] between the polar amine and phenol molecules and the nonpolar alkyl chains in the solgel coatings. Silanol groups, pr esent in all th ree types of sol-gel coated capillaries, likely aid in the extraction of these compounds through dispersion, orientation, induction, and other mo lecular level interactions between the silanol groups within the sol-gel coatings and the pol ar functional groups (amine and hydroxyl groups) on the molecules of these analytes [73]. The CME-HPLC-UV detection limit data and run-to-run peak area RSD values for all of the analytes are given in table 4.3. All three sol-gel alkyl coated capillaries demonstrated low and sub ng/L detection lim its for PAHs, low ng/L detection limits for

PAGE 216

185 ketones and diphenylamine, and ng/L detecti on limits for phenols and toluidines (Table 4.3). The detection limits for fluorene and phenanthrene were more than 1000 times lower than those reported for sol-gel octyl monolithic microextraction capillaries on-line coupled to HPLC-UV [58]. The sol-gel alkyl coated capillaries had lower detection limits than other sol-gel microextraction co atings on-line hyphenated with HPLC-UV for ketones [53,55,56], amines [56], a nd phenols [56]. For all of the analytes extracted, the run-to-run HPLC peak area RSD values for th e sol-gel octadecyl, octyl, and methyl coated capillaries ranged from 1.5 to 9.6%, 1.1 to 7.4%, and 3.2 to 8.2%, respectively.

PAGE 217

186 Table 4.3 CME-HPLC-UV peak area repeatabilit y and detection limit data for PAHs, ketones, phenols, and amines for the sol-ge l octadecyl, octyl, and methyl coated microextraction capillaries (0.514 M sol-gel precursor concentration). Compound Octadecyl Octyl Methyl RSD (%) Detection RSD (%) Detection RSD (%) Detection PAHs (n=3) Limit (ng/L) (n=3) Limit (ng/L) (n=3) Limit (ng/L) fluorene 6.9 2.2 7.3 2.6 5.9 2.7 phenanthrene 4.0 1.1 6.0 1.2 7.2 2.1 anthracene 6.1 1.0 4.0 1.5 8.2 1.8 naphthalene 6.5 0.3 2.5 0.4 3.2 0.6 acenaphthene 7.5 0.7 4.3 0.9 3.4 2.2 Ketones coumarin 1.5 1.1 x 101 1.5 1.6 x 101 7.6 1.8 x 101 1-[1,1’-biphenyl]4-yl-ethanone 2.1 2.6 1.9 4.7 7.8 5.5 trans -chalcone 1.6 5.0 1.1 5.4 8.2 5.7 benzophenone 5.1 6.5 3.8 6.8 7.8 7.9 Phenols 2-chlorophenol 4.6 1.9 x 102 7.4 2.1 x 102 5.5 2.1 x 102 3,4-dimethylphenol 7.8 8.6 x 101 4.9 1.2 x 102 7.6 1.2 x 102 3,5-dimethylphenol 3.1 6.9 x 101 3.5 1.0 x 102 4.5 1.2 x 102 2,3-dichlorophenol 3.7 5.1 x 101 2.7 5.3 x 101 3.3 5.4 x 101 Amines m-toluidine 9.6 1.2 x 102 2.6 1.7 x 102 5.3 1.7 x 102 o-toluidine 5.4 1.4 x 102 2.9 1.6 x 102 5.6 1.7 x 102 diphenylamine 5.5 4.1 3.9 5.9 5.6 8.5 Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel octadecyl, octyl, and methyl coated capillaries; extraction time : 60 min. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column; isocratic elution 70/30 ACN/water; 1 mL/min flow rate, UV detection at 260 nm for fluorene, 254 nm for phenanthrene and anthracene, 217 nm for naphthalene and acenaphthene, 200 nm for ketones, phenols, and amines.

PAGE 218

187 Three capillaries of each type (octadecyl, octyl, and methyl) were prepared and tested for capillary-to-cap illary reproducibility. The observed extraction pattern remained the same for all of the compounds test ed on all of the sol-ge l coated capillaries. Also, the sol-gel octadecyl, octyl, and me thyl coated capillaries demonstrated good capillary-to-capillary reproducib ility: the peak area RSD values ranged from 1.3 to 7.9%, 5.3 to 9.4%, and 6.0 to 10.0%, for sol-gel me thyl-, octyl-, and octadecyl-coated capillaries, respectively. The capillary-to-ca pillary reproducibility represented by HPLC peak area RSD values are given in table 4.4. The reproducibility of the sol-gel method used to prepare octadecyl, octy l, and methyl coated sol-gel ca pillaries is illustrated by the chromatograms in figure 4.6, 4.7, and 4.8, respecti vely. The excellent consistency of the chromatograms in figures 4.6, 4.7, and 4.8 confirms the reliability of the sol coating and conditioning procedure used to make these solgel coated capillaries. It also indicates that the surface compositions, including the al kyl and silanol group c oncentrations, of the prepared sol-gel coatings are very similar.

PAGE 219

188 Table 4.4 Capillary-to-capillary CME-HPLC-UV p eak area repeatability for the sol-gel octadecyl, octyl, and methyl co ated microextraction capillaries. Analyte Octadecyl Octyl Methyl RSD (%) RSD (%) RSD (%) (n=3) (n=3) (n=3) m -toluidine 7.7 8.2 8.7 3,4-dimethylphenol 7.9 8.8 9.5 diphenylamine 1.3 9.4 7.9 trans -chalcone 5.2 5.3 6.0 acenaphthene 4.8 6.8 10.0 Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel octadecyl, octyl, and methyl coated capillaries; extraction time: 60 minutes. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C-18 column; gradient elution 45/55 5 min to 60/40 ACN/water in 12 min; 1 mL/min flow rate, UV detection at 200 nm.

PAGE 220

189 Figure 4.6 Capillary-to-capillary CME-HPLC-UV reproducibility of amines, phenols, ketones, and PAHs using three sol-gel oc tadecyl coated capillaries of each type. Extraction conditions: 40 cm x 0.25 mm I.D. cap illary, 60 min gravity-fed extraction, at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dime thylphenol (1 x 105 ng/L), 3 = diphenylamine (4 x 103 ng/L), 4 = trans -chalcone (2 x 104 ng/L), 5 = acenaphthene (5 x 104 ng/L).

PAGE 221

190 Figure 4.7 Capillary-to-capillary CME-HPLC-UV reproducibility of amines, phenols, ketones, and PAHs using three sol-gel octyl coated capillaries of each type. Extraction conditions: 40 cm x 0.25 mm I.D. capillar y, 60 min gravity-fed extraction, at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (1 x 105 ng/L), 3 = diphenylamine (4 x 103 ng/L), 4 = trans -chalcone (2 x 104 ng/L), 5 = acenaphthene (5 x 104 ng/L).

PAGE 222

191 Figure 4.8 Capillary-to-capillary CME-HPLC-UV reproducibility of amines, phenols, ketones, and PAHs using three sol-gel methyl coated capillaries of each type. Extraction conditions: 40 cm x 0.25 mm I.D. capillar y, 60 min gravity-fed extraction, at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (1 x 105 ng/L), 3 = diphenylamine (4 x 103 ng/L), 4 = trans -chalcone (2 x 104 ng/L), 5 = acenaphthene (5 x 104 ng/L).

PAGE 223

192 4.3.2 Microextraction capillaries prepared using sol solution with varied molar concentration of C8TMS Precursor (Table 4.2) Sol-gel octyl coated capillaries we re prepared using different molar concentrations of C8TMS (0.257 M, 0.514 M, 1.028 M, and 1.542 M) to study the dependence of extraction behavior on the so l-gel precursor concentration. The same compounds previously extracted on the octadecy l, octyl, and methyl coated capillaries prepared using sol solutions w ith the same molar concentration of sol-gel precursor were also extracted on the sol-gel octyl coated capillaries prepared with varied precursor concentrations in the sol solution. The chromatograms in figures 4.9, 4.10, 4.11, and 4.12 illustrate the observed differences in extrac tion behavior for the sol-gel octyl coated capillaries with 0.257 M, 0.514 M, 1.028 M, and 1.542 M of C8TMS in the sol solution. A peak area comparison for the four sol-gel octy l coated capillaries prepared with varied sol-gel precursor concentrati on is presented in table 4.5.

PAGE 224

193 Figure 4.9 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, ketones, and PAHs using sol-gel octyl co ated capillaries prepared using 0.257 M concentration of C8TMS in the sol solution. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 60 min grav ity-fed extraction, at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/ 50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (5 x 104 ng/L), 3 = diphenylamine (1 x 104 ng/L), 4 = trans -chalcone (1 x 104 ng/L), 5 = acenaphthene (1 x 104 ng/L).

PAGE 225

194 Figure 4.10 A chromatogram representingCME-HPL C-UV analysis of amines, phenols, ketones, and PAHs using sol-gel octyl co ated capillaries prepared using 0.514 M concentration of C8TMS in the sol solution. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 60 min grav ity-fed extraction, at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/ 50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (5 x 104 ng/L), 3 = diphenylamine (1 x 104 ng/L), 4 = trans -chalcone (1 x 104 ng/L), 5 = acenaphthene (1 x 104 ng/L).

PAGE 226

195 Figure 4.11 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, ketones, and PAHs using sol-gel octyl co ated capillaries prepared using 1.028 M concentration of C8TMS in the sol solution. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 60 min grav ity-fed extraction, at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/ 50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (5 x 104 ng/L), 3 = diphenylamine (1 x 104 ng/L), 4 = trans -chalcone (1 x 104 ng/L), 5 = acenaphthene (1 x 104 ng/L).

PAGE 227

196 Figure 4.12 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, ketones, and PAHs using sol-gel octyl co ated capillaries prepared using 1.542 M concentration of C8TMS in the sol solution. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 60 min grav ity-fed extraction, at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/ 50 to 80/20 ACN/water over 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1 x 105 ng/L), 2 = 3,4-dimethylphenol (5 x 104 ng/L), 3 = diphenylamine (1 x 104 ng/L), 4 = trans -chalcone (1 x 104 ng/L), 5 = acenaphthene (1 x 104 ng/L).

PAGE 228

197 Table 4.5 HPLC peak area comparison of amines, phenols, ketones, and PAHs using solgel octyl coated microextraction capillaries w ith varied sol-gel precursor concentrations. Chemical Chemical Molar Concentration of C8TMS in Sol Solution Class Name 0.257 M 0.514 M 1.028 M 1.542 M Amine m -toluidine 7.4 7.8 6.3 4.6 Phenol 3,4-dimethylphenol 6.5 6.8 6.2 4.2 Amine diphenylamine 10.7 14.4 12.9 2.8 Ketone transchalcone 9.3 9.9 9.7 2.9 PAH acenaphthene 3.5 11.0 6.5 1.2 Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel octyl coated capillaries (0.257 M, 0.514 M, 1.028 M, and 1.542 M). Extraction time : 60 min. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column; gradient elution 50/50 to 80/20 ACN/water in 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambi ent temperature. Peak areas in arbitrary units, average of 3 replicate measurements.

PAGE 229

198 For all of the analytes extracted, the sol-gel octyl coated capillary (prepared with 0.514 M C8TMS in the sol solution) demonstrated the most efficient extraction. The solgel octyl coated capillaries (prepared with 0.257 M, 1.028 M, and 1.542 M C8TMS, in the sol solution) were less efficient in extracting al l of the analytes. For all four capillaries tested, the sol-gel octyl coated ca pillary (prepared with 0.257 M of C8TMS) was second most efficient in the extraction of the polar analytes ( m -toluidine and 3,4dimethylphenol), which indicates that this so l-gel coating may have contained more free silanol groups. The sol-gel octyl coat ed capillary (prepare d with 1.028 M of C8TMS in the sol solution) was second most efficient in the extraction of the moderately polar and nonpolar analytes (diphenylamine, trans -chalcone, and acenaphthene), which indicates that it likely contained fewe r free silanol groups in the resultant sol-gel network. Although diphenylamine is a polar compound, the tw o aromatic rings sterically hinder its polar functional group, allowing it to behave more like a moderately polar compound [74]. The 1.542 M octyl coated capillary was le ast efficient in the extraction of all of the compounds tested. Since the alkyl chains are attached to th e sol-gel precursor, a sol-gel network with one alkyl chain per molecule of sol-ge l precursor is produced. When higher concentrations of sol-gel precursor are used in the sol solution, the created sorbent will not contain a higher concentra tion of alkyl chains. The 0.514 M concentration of sol-gel precursor in the sol solution appears to be op timal for the creation of sol-gel octyl coated microextraction capillaries. When lower concen trations of sol-gel pr ecursor are used, the precursor molecules likely have to diffuse through the sol solution in order to undergo condensation, resulting in fewer condensation re actions and a thinner sol-gel coating.

PAGE 230

199 When higher concentrations of sol-gel precur sor are used, the alkyl chains likely hinder condensation reactions, also pr oducing a thinner sol-gel coati ng. When an intermediate concentration of sol-gel precursor is used, the precursor molecules can diffuse freely and are high enough in concentrati on to readily undergo condensa tion, resulting in a thicker sol-gel coating with better extraction capabi lities. Scanning electron microscope (SEM) images of the 0.247 M (figure 4.13), 0.514 M (figure 4.14), and 1.542 M (figure 4.15) sol-gel octyl coated capillaries were taken. From these images, it is clear that the 0.514 M sol-gel octyl coating is th e thickest, the 0.247 M sol-gel oc tyl coating is thinner, and the 1.542 M sol-gel octyl coating is the thinne st, which is consistent with the observed extraction capabilities of the three capillaries.

PAGE 231

200 Figure 4.13 Scanning electron microscope (SEM) image of the sol-gel octyl coated microextraction capilla ry prepared using 0.247 M concentration of C8TMS in the sol solution.

PAGE 232

201 Figure 4.14 Scanning electron microscope (SEM) image of the sol-gel octyl coated microextraction capilla ry prepared using 0.514 M concentration of C8TMS in the sol solution.

PAGE 233

202 Figure 4.15 Scanning electron microscope (SEM) image of the sol-gel octyl coated microextraction capilla ry prepared using 1.542 M concentration of C8TMS in the sol solution.

PAGE 234

203 4.4 Conclusion Sol-gel octadecyl-, octyl-, and methyl coat ed capillaries are all suitable for use in capillary microextraction online hyphenated with HPLC. Among the three types of solgel capillaries prepared using the same con centration of sol-gel precursor (0.514 M), the sol-gel octadecyl coated ones were the most efficient in extrac ting nonpolar analytes, followed by the sol-gel octyl coated capillaries, followed by the sol-gel methyl coated capillaries. This is expected, since the l onger nonpolar alkyl chains allow for greater molecular interaction with nonpolar compounds. This study also indicates that the alkyl chains in these sol-gel coatings play an active role in the extraction of polar analytes, since the sol-gel octadecyl coat ed capillaries offered the most efficient extraction of polar analytes, followed by the sol-gel octyl coated capillaries, followed by the sol-gel methyl coated capillaries. It is plausible that th e polar compounds are extracted by the sol-gel alkyl coatings through synergis tic molecular level interactio ns of the polar compounds with the silanol groups and alkyl chains present in the sol-gel coatings. All of these solgel coatings provided reasona bly low detection limits (ng/ L) with good run-to-run peak area RSD values and capillary-to-capillary re producibility. When sol-gel octyl coated capillaries were prepared using different mola r concentrations of sol-gel precursor, it was found that the use of a 0.514 M concentration of C8TMS in the sol solution lead to the most efficient extraction capabilities. Cap illaries prepared using both higher and lower concentrations of C8TMS in the sol solution were less efficient in the extraction of all analytes tested. From these stud ies, it is clear that the use of higher concentrations of solgel precursor in the sol soluti on does not translate to a hi gher concentration of alkyl chains in the created sol-gel sorbent. Th is study provides some im portant insight into

PAGE 235

204 solute-sorbent interaction in capillary micr oextraction and may be useful in designing novel sol-gel coatings. 4.5 References for chapter four [1] S. L. Chong, D. Wang, J. D. Hayes, B. W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [2] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [3] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [4] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [5] J. Zhou, Z. Zeng, Anal. Chim. Acta 556 (2006) 400. [6] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [7] C. Basheer, S. Jegadesan, S. Valiyaveett il, H. K. Lee, J. Chromatogr. A 1087 (2005) 252. [8] J. V. Hinshaw, LC-GC N. Am. 21 (2003) 1056. [9] R. P. Belardi, J. B. Pawliszyn, Water Pollut. Res. J. Can. 24 (1989), 179. [10] G. Jiang, M. Huang, Y. Cai, J. Lv, Z. Zongshan, J. Chromatogr. Sci. 44 (2006) 324. [11] M. Yang, Z. R. Zeng, W. L. Qiu, Y. L. Wang, Chroma tographia 56 (2002) 73. [12] Y. Hu, Y. Yang, J. Huang, G. Li, Anal. Chim. Acta 543 (2005) 17. [13] D. Zhang, C. Wu, F. Ai, Chin. J. Chromatogr. 17 (1999) 10. [14] R. Yang, W. Xie, Fore nsic Sci. Int. 139 (2004) 177. [15] T. H. Ding, H. H. Lin, C. W. Whang, J. Chromatogr A 1062 (2005) 49. [16] M. A. Farajzadeh, M. Hata mi, J. Sep. Sci. 26 (2003) 802.

PAGE 236

205 [17] E. H. Koster, C. Crescenzi, W. den Hoedt, K. Ensing, G. J. de Jong, Anal. Chem. 73 (2001) 3140. [18] W. M. Mullett, P. Martin, J. Pawliszyn, Anal. Chem. 73 (2001) 2383. [19] F. Mangani, R. Cenciarini, Chromatographia 41 (1995) 678. [20] R. Eisert, K. Levsen, J. Chromatogr. A 733 (1996) 143. [21] H. L. Lord, J. Ch romatogr. A 1152 (2007) 2. [22] C. G. Zambonin, Anal. Bioanal. Chem. 375 (2003) 73. [23] E. Lesellier, Analusis 27 (1999) 363. [24] B. Santos, B. M. Simonet, A. Rios, M. Valcarcel, Electr ophoresis 28 (2007) 1312. [25] A. Penalver, E. Pocurull, F. Borrull, R. M. Marce, Trends Anal. Chem. 18 (1999) 557. [26] Y. Liu, Y. Shen, M. L. Lee, Anal. Chem. 69 (1997) 190. [27] K. Jinno, M. Taniguchi, M. Hayashid a, J. Pharm. Biomed. Anal. 17 (1998) 1081. [28] H. Kataoka, Anal. Bioanal. Chem. 373 (2002) 31. [29] L. Cai, Y. Zhao, S. Gong, L. Do ng, C. Wu, Chromatographia 58 (2003) 615. [30] X. Li, Z. Zeng, J. Zhou, Anal. Chim. Acta 509 (2004) 27. [31] R. Gomes da Costa Silva, F. Augusto, J. Chromatogr. A 1072 (2005) 7. [32] M. Liu, Z. Zeng, H. Fa ng, J. Chromatogr. A 1076 (2005) 16. [33] Y. Hu, Y. Fu, G. Li, Anal. Chim. Acta 567 (2006) 211. [34] X. Li, J. Gao, Z. Zeng, Anal. Chim. Acta 590 (2007) 26. [35] X. Li, Z. Zeng, J. Zhou, S. Gong, W. Wang, Y. Chen, J. Chromatogr. A 1041 (2004) 1. [36] C. Dong, Z. Zeng, X. Li, Talanta 66 (2005) 721.

PAGE 237

206 [37] V. G. Zuin, A. L. Lopes, J. H. Ya riwake, F. Augosto, J. Chromatogr. A 1056 (2004) 21. [38] M. Liu, Z. Zeng, B. Xiong, J. Chromatogr. A 1065 (2005) 287. [39] M. Liu, Z. Zeng, Y. Tia n, Anal. Chim. Acta 540 (2005) 341. [40] H. Bagheri, A. Es-haghi, M. R. Rouini, J. Chromatogr. B 818 (2005) 147. [41] H. Bagheri, A. Es-haghi, F. Khalilia n, M. R. Rouini, J. Pharm. Biomed. Anal. 43 (2007) 1763. [42] X. Li, Z. Zeng, M. Hv, M. Mao, J. Sep. Sci. 28 (2005) 2489. [43] D. Cha, M. Liu, Z. Ze ng, Anal. Chim. Acta 572 (2006) 47. [44] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [45] D. Wang, S. L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [46] Y. Wu, B. Hu, Z. Jiang, Y. Feng, P. Lu, B. Li, Rapid Commun. Mass Spectrom. 20 (2006) 3527. [47] F. Zeng, B. Hu, Talanta 73 (2007) 372. [48] Y. Fan, Y. Q. Feng, S. L. Da, Z. H. Wang, Talanta 65 (2005) 111. [49] S. Kulkarni, A. M. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50. [50] A. Kabir, C. Hamlet, K. Soo Yoo, G. R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [51] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [52] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [53] T. Y. Kim, K. Alhoosha ni, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165.

PAGE 238

207 [54] S. S. Segro, A. Malik, J. Chromatogr. A 1200 (2008) 62. [55] S. S. Segro, A. Malik, J. Chromatogr. A 1205 (2008) 26. [56] S. S. Segro, Y. Cabezas, A. Malik, J. Chromatogr. A 1216 (2009) 4329. [57] T. P. Gbatu, K. L. Sutton, J. A. Caruso, Anal. Chim. Acta 402 (1999) 67. [58] M. M. Zheng, B. Lin, Y. Q. Feng, J. Chromatogr. A 1164 (2007) 48. [59] Y. Fan, Y. Q. Feng, Z. G. Shi, J. B. Wang, Anal. Chim. Acta 543 (2005) 1. [60] T. Li, J. Xu, J. H. Wu, Y. Q. Feng, J. Chromatogr. A 1216 (2009) 2989. [61] J. D. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3. [62] C. J. Brinker, G.W. Scherer, Sol-Gel Science: The Chemistry and Physics of Sol-Gel Processing Academic Press, San Diego, CA, 1990. [63] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [64] H. Fang, M. Liu, Z. Zeng, Talanta 68 (2006) 979. [65] W. Liu, Y. Hu, J. Zhao, Y. X u, Y. Guan, J. Chromatogr. A 1102 (2006) 37. [66] K. Alhooshani, T. Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [67] M. M. Liu, Z. R. Zeng, C. L. Wang, Y. J. Tan, H. Liu, Chromatographia 58 (2003) 597. [68] D. Wang, J. Xing, J. Peng, C. Wu, J. Chromatogr. A 1005 (2003) 1. [69] R. C. Mehrotra, J. N on-Cryst. Solids, 121 (1990) 1. [70] Z. Zeng, W. Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [71] R. Rodriguez, M. Flores, J. Gom ez, V.M. Castano, Mater. Lett. 15 (1992) 242. [72] H. Kawaki, Chem. Ph arm. Bulletin 56 (2008) 323. [73] A. V. Kiselev, Chro matographia 11 (1978) 117.

PAGE 239

208 [74] K. Grob Jr., G. Grob, K. Grob, J. Chromatogr. 156 (1978) 1.

PAGE 240

209 CHAPTER FIVE SOLVENT-RESISTANT SO L-GEL POLYDIMETHYL DIPHENYLSILOXANE COATING FOR ON-LINE HYPH ENATION OF CAPILLARY MICROEXTRACTION WITH HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY 5.1 Introduction Solid phase microextraction (SPME), a so lvent-free sample enrichment technique, was developed by Pawliszyn and co-workers about two decades ago [1]. It differs from traditional sample preparation techniques because it uses a sorbent coating in place of a solvent to preconcentrate the sample [2]. SP ME has been successfu lly coupled with gas chromatography (GC) [3], high-performance liquid chromatography [4,5], supercritical fluid chromatography [6], and capillary electrop horesis [7,8]. The traditional SPME technique uses a small-diameter (~ 100 m) fused-silica rod (fiber), an end segment (~ 1 cm) of wh ich is provided with a sorbent coating that serves as the extracting phase. Depe nding upon their compositions, extracting phases exhibit distinctive affinities for different analytes [2]. The initial extracting phases used in SPME were organic polymers, which predominantly included organosiloxane materials, especially polysiloxanes with diffe rent functional groups incorporated as side chains [9-12]. Non-silocone materials (Polya mine [13], polyimide [1 4], cellulose acetate [15], and polyvinyl chloride [15]), molecularly imprinted polymer (MIP) coated fibers [16,17], and inorganic coa tings [18] have also been used in SPME.

PAGE 241

210 The major drawbacks of the traditional SPME technique include the fragility of the fiber and needle in the SPME device, the short length of the coated segment, which negatively affects sample capacity and extrac tion loading, and difficu lties in coupling to HPLC, which is evident from the complicat ed coupling interfaces reported in the literature [19]. In-tube SPME, in which the extracting phase coati ng resides on the inner surface of a capillary, thus pr otecting it from mechanical da mage, was developed in order to overcome some of these difficulties. In the in-tube format extr action is performed by passing samples through the coated capillary. Since the length of the coated segment is a few orders of magnitude greater than that used in fiber SPME, in-tube SPME can provide greater extracting phase loading, enhanced sample capacity, and improved extraction sensitivity. However, most in-tube SPME co atings are thin (sub-m), and therefore not conducive to achieving the desired improvement in sample capacity [20]. Stable thick coatings are quite difficult to prepare us ing conventional techni ques. Conventional coatings in both fiberand in-tube SPME ar e not chemically bonded to the substrate, rendering them unstable under organo-aqueous m obile phases typically used to desorb the extracted analytes for HPLC analysis [21]. Thus, the poor solvent stability of conventional coatings serious ly limits the hyphenation of SPME to HPLC, and restricts the widespread application of this hyphenated technique that has the potential to become an extremely powerful analytical tool. Sol-gel coatings were developed to addr ess these problems with solvent tolerance in SPME-HPLC, as well as thermal stability problems encountered in SPME-GC [21]. Sol-gel coatings are attached to the substrat e (fiber or capillary surface) through chemical bonding. This eliminates the problems cau sed by high temperatures (SPME-GC) and

PAGE 242

211 organic solvents (SPME-HPLC) [21]. The so l-gel approach allows for the creation of both thick and thin surface-bonded coatings through manipulation of the sol solution composition and coating time. Thus, the so l-gel technology also provides an effective solution to the problems associated with the cr eation of thick (> 1 m) coatings, which is difficult to solve by conventional techniques [22,23]. The SPME technique using a solgel coated capillary became known as so l-gel capillary microextraction [22]. The covalently-bonded extracting pha se of greater thickness prov ides improved sensitivity. The solvent resistance of the sol-gel coat ed capillary makes it suitable for hyphenation with HPLC. Many successful extracting phase s have been developed for sol-gel SPME in both the fiber and capillary formats [23-28]. In fiber SPME, the reported sol-gel extracting phases include sol-gel PDMS [21], sol-gel PEG [25], sol-gel crown ether [29], sol-gel poly(methylphenylvinylsiloxane) (PMP VS) [10], sol-gel hydroxy fullerene [28], sol-gel calix[4]arene [30], oligome r-based sol-gels [31], and sol-gel -cyclodextrin [27]. In capillary microextraction, sol-gel extracti ng phases include sol-gel PDMS [22], sol-gel dendrimer [32], sol-gel poly(tetrahydrofuran) [23], and sol-gel cyano-PDMS [33]. Solgel titania PDMS [34], sol-gel -cyclodextrin [35], and sol-gel methyl [36] have been used in CME-HPLC. Polydimethyldiphenylsiloxane (PDMDPS) wa s used successfully in both zirconia [37] and germania-based [38] sol-gel coatings for capillary microext raction coupled with GC-FID. However, to our knowledge, there have been no reports on the use of sol-gel PDMDPS coatings in capillary microextraction in hyphenation with HPLC. In this paper, we describe a sol-gel approach to the creation of solvent-re sistant surface-bonded organic-inorganic hybrid sol-gel PDMDPS coated fused silica capilla ries and the use of

PAGE 243

212 such capillaries to provide an effective means to hyphenate capillary microextraction with HPLC. The purpose of this study was to devel op a silica-based sol-gel PDMDPS coating for effective use in capillary microextract ion on-line hyphenated with HPLC. In this coating, PDMDPS offers the advantage of e fficiently extracting nonpolar and moderately polar aromatic analytes, while the chemical anchoring of the PD MDPS into the sol-gel network, which is covalently bonded to the inne r walls of the fused s ilica capillary, offers the advantage of solvent resistance, whic h is necessary for on-line hyphenation with HPLC. The effectiveness of the sol-gel PDMDPS coated capill ary to extract a variety of nonpolar and moderately polar analytes wh en on-line hyphenated to HPLC with UV detection was investigated. The solvent resist ance of this coating was also investigated through solvent stability tests. Furthermore, the high temperat ure solvent stab ility of the sol-gel PDMDPS coating was investigated fo r possible future use in high temperature HPLC. To our knowledge, sol-gel coatings ha ve not been investig ated for potential use in high temperature HPLC in previous studies. 5.2 Experimental 5.2.1 Equipment On-line coupled sol-gel CME-HPLC expe riments were carried out using a MicroTech Scientific (Vista, CA, USA) Ultra Plus HPLC system equipped with a Linear UVIS 200 variable wavelength UV detector. For HP LC separations, a reversed-phase Luna C18 column (15 cm x 4.6 mm I.D.) was used. For thorough mixing of the sol solution ingredients, a Fisher model G-560 Vortex Geni e 2 system (Fisher Scientific, Pittsburgh,

PAGE 244

213 PA, USA) was utilized. A Thermo IEC model Micromax microcentrifuge (Needham Heights, MA, USA) was used for the centrif ugation of the sol solutions. Nanopure water (15 M ) was obtained from a Barnstead model 04741 Nanopure deionized water system (Barnstead/Thermodyne, Dubuque, IA, USA). Chrom-Perfect version 3.5 for Windows computer software (Justice Laboratory Soft ware, Denville, NJ, USA) was used for online collection and processing of the CME-HPLC data. 5.2.2 Chemicals and materials Fused silica capillary (0.25 mm I. D.) was purchased from Polymicro Technologies (Phoenix, AZ, USA). Pol ydimethyldiphenylsiloxane (PDMDPS), methyltrimethoxysilane (MTMS), poly(m ethylhydrosiloxane) (PMHS), polycyclic aromatic hydrocarbons (fluoranthene, fluor ene, naphthalene, phenanthrene, and acenaphthene), ketones (benzophenone, coumarin, and trans-chalcone), aromatic compounds (benzanilide, m -terphenyl, biphenyl, and diphe nyldisulfide) were purchased from Aldrich (Milwaukee, WI, USA). An thracene and 4’phenylacetophenone were purchased from Eastman Kodak (Rochester, NY, USA). Diphenylamine was purchased from J.T. Baker (Phillipsburg, NJ, USA). Tr ifluoroacetic acid was purchased from Acros (Morris Planes, NJ, USA). HPLC-grade acet onitrile, methanol, and methylene chloride were purchased from Fisher Scie ntific (Pittsburgh, PA, USA). 5.2.3 Pretreatment of fused silica capillary The inner surface of the fuse d silica capillary was pretre ated prior to the creation of the sol-gel PDMDPS coating on it. First, it was sequentially rinsed with 4 mL each of methylene chloride, methanol, and deionized water. The capillary was then briefly purged with helium (5 min), leaving a thin co ating of water on the inner surface. Next,

PAGE 245

214 both ends of the capillary were sealed using an oxy-acetylene torch. Following this, it was placed in a GC oven and heated at 350 0C for two hours. It was then removed and allowed to cool. The ends of the capillary were then cut open using an alumina wafer, and it was placed in a GC oven and purged with a continuous flow of helium (1 mL/min). The temperature of the oven was programmed from 40 0C to 250 0C, at 5 0C/min. The fused silica capillary was held at 250 0C for 2 h. 5.2.4 Preparation of the sol-gel poly dimethyldiphenylsiloxane coated microextraction capillary The sol solution was prepared using the following procedure. First, 0.04 g of poly(methylhydrosiloxane) (PMHS), a deactiva ting reagent, was weighed into a clean microcentrifuge tube. Then, 256.2 mg of hydr oxy-terminated PDMDPS, a sol-gel active polymer, was added to it and the mixture was vortexed for 30 s. Next, 300 L of methylene chloride was added to the mixtur e, and the mixture was vortexed again for 30 s. Then, 100 L of methyltrimethoxysilane (M TMS), the sol-gel precursor, was added to the mixture, followed by vortexing for 30 s. Fi nally, 92 L of triflu oroacetic acid, which contained 5% deionized water, was added to the mixture. The so l solution was vortexed once more for 30 s, and then centrifuged for 4 min at 14 000 rpm (15 682 g ) to remove any particulates from the sol solution. After centrifugation, the top cl ear portion of the liquid in the microcentrifuge tube was removed using a micropipette and transfer red to a new clean microcentrifuge tube for further use in coating a hydrothermally pr etreated fused silica cap illary (60 cm x 0.25 mm I.D.). To accomplish this, the sol solution was allowed to flow through the capillary under helium pressure (50 psi) using a capilla ry filling/purging device [39]. The exit end

PAGE 246

215 of the capillary was then sealed using a piece of rubber septum and kept under helium pressure (50 psi) for 40 min. The rubber se ptum was then removed to expel the liquid from the capillary and to allow helium to fl ow through the capillary for 40 min at 50 psi. The capillary was further placed in a GC ove n purging it with continuous helium flow. The temperature of the oven was programmed from 40 oC to 350 oC, at 5 oC/min. It was held at the final temperature for 2 h, follo wed by cooling to room temperature. The capillary was then rinsed with 2 mL of a (1/1, v/v) metha nol/deionized water solution. The capillary was once more conditione d under helium purge by programming the temperature from 40 oC to 350 oC, at 5 oC/min, holding at 350 oC for only 30 min. The capillary was then rea dy for use in CME-HPLC. 5.2.5 On-line CME-HPLC analysis A schematic diagram of the extraction set up is shown in figure 5.1. The ends of a 40-cm long piece of the sol-gel PDMDPS coated capillary were f itted with polyether ether ketone (PEEK) tubing sleeve (0.020” x 0.062”), along with appropriate nuts and ferrules. This capillary was then installed on an HPLC six-port inje ction valve, using it as an external sampling loop. Before st arting extraction on the sol-gel PDMDPS coated capillary, the HPLC column (Luna C18 or Zorbax Phenyl) (15 cm x 4.6 mm I.D.) was equilibrated with the mobile phase compositi on necessary for adequa te separation of the analytes in the sample. A segment of deac tivated fused silica capillary (0.53 mm I.D.) was used to connect a gravity-fed sample di spenser [22] to a syri nge needle, which was inserted into the HPLC injection valve. Extraction was performed by placing liquid samples in the gravity-fed sample delivery system and allowing them to pass through the sol-gel PDMDPS coated capillary. During the extraction, the injector was kept in the

PAGE 247

216 “load” position, and the sample was allowed to drip through the sol-gel coated microextraction capillary (used as an exte rnal sampling loop) to reach an extraction equilibrium (40 min for the samples analy zed) with the sol-gel PDMDPS coating on the capillary’s inner surface. The extracted anal ytes were then transferred to the HPLC column by switching the injection valve to the “inject” position. The flow of the organicrich HPLC mobile phase desorbed the ex tracted analytes from the sol-gel PDMDPS coated capillary and transferred them into th e HPLC column for sepa ration. Isocratic and gradient elution was used with water/ace tonitrile mobile phases of appropriate compositions to achieve separation of the extracted analytes.

PAGE 248

217 Figure 5.1 Sol-gel CME-HPLC experi mental setup. While extr acting, the six-port HPLC injection valve is switched to the “load” pos ition, and the sample solution flows from the gravity-fed sample dispenser, through the so l-gel PDMDPS coated capillary, and into an appropriate waste container. To perform anal ysis, the injection valve is switched to the “inject” position, causing the mobile phase to flow through the sol-gel PDMDPS coated capillary, thus desorbing the ex tracted analytes and eluting th em to the HPLC column for separation, followed by UV detection.

PAGE 249

218 5.3 Results and discussion In the sol-gel process, a colloidal system, the sol, is formed through hydrolytic polycondensation of one or more sol-gel prec ursors (typically al koxides). This is followed by the gelation of the sol to pr oduce a three-dimensional network in a continuous liquid phase, the gel [40]. Sol-gel coatings used in micr oextraction typically involve the use of a number of ingredients in the sol solutio n. These include the sol-gel precursor, the sol-gel active organic polymer, the solvent, the deactiv ating reagent, water, and the catalyst [30,32-34,37,41-47]. Typical sol-gel precursors include alkoxysilanes [48], such as tetramethoxysilane (TMOS), tetraethoxysilane (TEOS), or methyltrimethoxysilane (MTMS). In the so l-gel PDMDPS coated capillary, MTMS was used as the sol-gel precursor. Sol-gel-active organic polymers that have been used in the preparation of sol-gel coatings includ e hydroxy-terminated poly(dimethylsiloxane) (PDMS) [21,22], poly(ethylenegly col) (PEG) [25,49], poly(vinyl alcohol) (PVA) [50], as well as crown ethers [29], cyclodextrins [ 27], and calixarenes [ 30] that contain hydroxy groups. In this study, hydroxy-terminated PDMDPS was utilized as the sol-gel active polymer. Appropriate solvents are used to thoroughly dissolve all of these ingredients into a sol solution and include common solvents such as methylene chloride [34], THF [31], isopropanol [41], and acetone [10]. Methyl ene chloride was used to dissolve all of the ingredients used in the PDMDPS sol solu tion. Deactivating reagents are typically incorporated in the sol-solu tion to derivatize (or block) residual silanol groups on the fused silica capillary surface or in the sol-gel network at the thermal conditioning step that follows. Common deactivating reagents include poly(methylhydrosiloxane) (PMHS) [21,22,24] and 1,1,1,3,3,3-hexamethyldisilazane (HMDS) [33,34,41]. Typical sol-gel

PAGE 250

219 catalysts are acids (e.g., trif luoroacetic acid [21,22]), ba ses (e.g., ammonium hydroxide [31]), or fluorides (e.g., amm onium fluoride [51]). In th e preparation of the sol-gel PDMDPS coated capillary, PMHS and trif luoroacetic acid were employed as the deactivating reagent and catalyst, respectiv ely. This sol-gel PDMDPS coating, along with most sol-gel coatings reported in the literature, is silica based, but titania[34], zirconia[37], alumina[41] and germania[38] based sol-gel coatings have been reported and shown to demonstrat e superior pH stability. In this study, a sol-gel polydimethyldi phenylsiloxane capillary was created using methyltrimethoxysilane (MTMS) as the solgel precursor and di-hydroxy-terminated polydimethyldiphenylsiloxane (PDMDPS) as th e sol-gel active polymer. Schemes 5.1 and 5.2 depict the hydrolytic polyconden sation reactions carried out within a hydrothermally pretreated fused silica capilla ry using trifluoroacetic acid (containing 5% H2O) as the sol-gel catalyst. Through hydr olytic polycondensati on reactions (Scheme 5.1), the PDMDPS and MTMS generated a solgel network with pendant methyl and phenyl groups. This evolving sol-gel network covalently bonded to th e inner walls of the fused silica capillary and led to the formation of a surface -anchored sol-gel coating to serve as the extracting phase in CME (Schem e 5.2). Such a chemical anchorage was achieved through the formation of covalent bonds between the sol-ge l active groups in the network (e.g., silanol or alkoxy groups) a nd silanol groups on th e inner wall of the fused silica capillary. To facilitate this covalent bonding, the fused silica capillary was previously subjected to hydrot hermal pretreatment to promote the formation of silanol groups on its inner surface. Poly(methylhydros iloxane) was added to the sol-gel mixture

PAGE 251

220 to deactivate residual silanol groups in the so l-gel coating. The deactivation process took place primarily in the thermal conditioning steps that followed.

PAGE 252

221 Si CH3OCH3OCH3CH3O + 3 H2O Si CH3OH OH HO +3 CH3OH H y d r o l y s i s of t he so l -ge l p r ecu r so r : catalyst Polycondensation of the hydrolyzed products: Si CH3OH OH HO Si CH3OH OH n HO Si CH3OH HO Si)nCH3OH OH (O n H2O ++ Condensation of hydroxy-terminated polymer to the evolving sol-gel network: + Si Ph Ph HO (Si CH3O)lCH3O (Si Ph Si CH3OH CH3O)m Ph polydimethyldiphenylsiloxane Si CH3OH HO Si)nCH3O OH (O Si Ph Ph (Si CH3O)lCH3O (Si Ph Si CH3OH CH3O)m Ph + H2O Si CH3OH HO Si)nCH3OH OH (O Scheme 5.1 Hydrolysis of sol-gel precursor (M TMS) followed by polycondensation of the hydrolyzed precursor and chemical bonding of the sol-gel active polymer (PDMDPS) to the evolving sol-gel network.

PAGE 253

222 Bonding of sol-gel coati ng to capillary wall: Si CH3OH HO Si)nCH3O OH (O Si Ph Ph (Si CH3O)lCH3O (Si Ph Si CH3OH CH3O)m Ph OH OH OH + OH O OH Si CH3OH Si)nCH3O OH (O Si Ph Ph (SiO)lCH3O (Si Ph Si CH3OH CH3O)m Ph capillary wall surface-bonded sol-gel coating + H2O CH3 Scheme 5.2 Chemical anchoring of the evolving sol-gel PDMDPS network to the inner walls of a fused silica capillary.

PAGE 254

223 Extraction profiles of four different compounds (each representative of a particular chemical class) were investigat ed on the sol-gel PDMDPS capillary. For this, three replicate extraction experiments were performed for each compound for each of the following extraction time pe riods: 10 min, 20 min, 40 mi n, 60 min, and 80 min. The average HPLC peak area for each extraction period was then plotted against the respective extraction time. The point on the graph at which the test compound stops increasing in peak area corresponds to th e time required for the compound to reach equilibrium between the sample solution a nd the sol-gel PDMDPS coating representing the extracting phase (sorbent). All four test analytes (naphthalene, 4’phenylacetophenone, m -tolualdehyde, and biphenyl) reached equilibrium within 40 min of extraction. This is typical of sol-gel coa tings. In all subsequent analyses, the aqueous samples were extracted using the sol-ge l PDMDPS coated capillary for 40 min. The extraction profiles are pr esented in figure 5.2.

PAGE 255

224 Figure 5.2 Extraction profiles of naphthalene (1 x 105 ng/L), biphenyl (2 x 104 ng/L) 4’phenylacetophenone (5 x 104 ng/L), and m -tolualdehyde (2 x 105 ng/L) on the sol-gel PDMDPS coated capillary.

PAGE 256

225 The sol-gel PDMDPS coated capillary demonstrated excellent detection limits in CME-HPLC. The repeatability and detection li mit data is shown in table 5.1. The HPLC peak area RSD values ranged from 2.4 to 9.1%. Depending upon the analyte types and their UV absorption characteristics, the obs erved detection limits ranged from 1.6 to 358.4 ng/L.

PAGE 257

226 Table 5.1 HPLC Peak area repeatability and dete ction limit data for PAHs, ketones, nonpolar aromatic compounds, aromatic amines and aldehydes in CME-HPLC using a sol-gel polydimethyldiphenylsiloxane coated microextraction capillary. Compound Class Compound Peak Area RSD (%, n=3) Detection Limit (ng/L) PAHs fluoranthene 7.6 6.0 anthracene 3.4 2.7 phenanthrene 5.6 2.3 fluorene 6.2 4.5 naphthalene 8.8 1.8 acenaphthene 4.5 1.6 Aromatic m -terphenyl 3.5 7.2 (nonpolar) biphenyl 4.5 1.9 diphenyldisulfide 5.9 4.0 Aromatic benzanilide 9.1 1.4 x 102 (amines) diphenylamine 3.2 1.5 x 101 Ketones benzophenone 4.5 9.2 x 101 coumarin 2.6 1.8 x 101 4’phenylacetophenone 7.7 2.2 x 101 trans-chalcone 2.4 1.5 x 101 Aldehydes m -tolualdehyde 8.4 9.1 x 101 p -nitrobenzaldehyde 8.0 3.6 x 102 Extraction conditions: 40 cm x 0.25 mm I. D. sol-gel PDMDPS coated capillary; extraction time: 40 minutes for all. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column; isocratic elution 70/30 ACN/water; 1 mL/min flow rate, UV detection at 256 nm for fluoranthene and anthracene, 250 nm for phenanthrene, 260 nm for fluorene, 217 nm for acenaphthene and naphthalene, 205 nm for aromatic compounds, 200 nm for ketones and aldehydes.

PAGE 258

227 The sol-gel PDMDPS coated capillary demonstrated remarkably low ng/L detection limits for polycyclic aromatic hydro carbons (PAHs). The analysis of PAHs is important since they are persistant organic pollutants found in the e nvironment [ 52]. All of the PAHs tested had RSD values betw een 3.4% and 8.8%. The detection limits for PAHs ranged from 1.6 ng/L to 6.1 ng/L. A chromatogram illustrating the CME-HPLC analysis of four PAHs is shown in figure 5.3.

PAGE 259

228 Figure 5.3 A chromatogram representing on-line CME-HPLC analysis of PAHs using a sol-gel PDMDPS coated microe xtraction capillary. Extrac tion conditions: 40 cm x 0.25 mm I.D. sol-gel PDMDPS coated capillary, 40 minute gravity fed extraction at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 70/30 to 80/20 ACN/water over 10 min, 1 mL/min flow rate, UV detection at 217 nm, ambient temperature. Peaks: 1 = naphthalene (2 x 103 ng/L), 2 = acenaphthene (4 x 103 ng/L), 3 = anthracene (5 x 104 ng/L), 4 = fluoranthene (1 x 104 ng/L).

PAGE 260

229 Other aromatic compounds were also extr acted using the sol-gel PDMDPS coated capillary in CME-HPLC with UV detecti on. Three of these aromatic compounds, m terphenyl, biphenyl, and diphenyldisulfide, contain no polar functional groups. The analysis of these compounds is important since they are common pollutants. For example, diphenyldisulfide is found in the pesticide phosbutyl [53]. Also, m -terphenyl has been used in planarity recognition st udies on retention in microcolumn liquid chromatography [54]. All of these compounds have good UV absorbance at 205 nm. The observed detection limits for bi phenyl, diphenyldisulfide, and m -terphenyl were 1.9 ng/L, 4.0 ng/L, and 7.2 ng/L, respectively. In CME-HPLC analysis at 205 nm, the two aromatic amines, benzanilide and diphenylam ine, demonstrated detection limits of 136.4 ng/L and 15.4 ng/L, respectively. These higher detection limits can be attributed to the more polar nature of the amines and the nonpol ar nature of methyl and phenyl groups in the sol-gel PDMDPS coating, as well as differe nces in UV absorbance characteristics. Aromatic amines are environmental contam inants, some of which are known to be carcinogenic, even at trace amounts [55]. A ll of the aromatic compounds analyzed had peak area RSD values between 3.2% and 9.1% A chromatogram depicting CME-HPLC analysis of both nonpolar aromatic compounds and aromatic amines is shown in figure 5.4.

PAGE 261

230 Figure 5.4 A chromatogram representing on-lin e CME-HPLC analysis of aromatic compounds using a sol-gel PDMDPS coated microextraction cap illary. Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel PDMDPS coated capillar y, 40 minute gravity fed extraction at room temperature. HP LC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 70/30 to 90/10 ACN/water over 20 min, 1 mL/min flow rate, UV detection at 205 nm, ambient temperatur e. Peaks: 1 = benzanilide (2 x 105 ng/L), 2 = diphenylamine (2.5 x 104 ng/L), 3 = biphenyl (3 x 103 ng/L), 4 = m-terphenyl (1 x 104 ng/L), 5 = diphenyldisulfide (1 x 104 ng/L).

PAGE 262

231 The sol-gel PDMDPS coated capillary was also capable of extracting moderately polar carbonyl compounds. Benzophenone, 4’ phenylacetophenone, coumarin, and transchalcone were tested using the PDMDPS coated capillary in CME-HPLC with UV detection. All of these compounds have good UV absorbance at 200 nm. All of the ketones analyzed had peak area RSD valu es between 2.4% and 8.0%. The detection limits for ketones ranged from 14.8 ng/L to 91.6 ng/L. The higher detection limits observed for ketones can be attributed to the moderately polar nature of these compounds and the nonpolar nature of the PDMDPS coating, as well as differences in the UV absorption characteristics of these analytes. The analysis of ketone s is significant in medical applications, such as the analysis of blood and exhaled breath [56,57], and in environmental applications, such as the analys is of animal waste [58] and exhaust gases [59]. Two aldehydes, m -tolualdehyde and p -nitrobenzaldehyde, were tested using the PDMDPS coated capillary in CME-HPLC us ing UV detection. Aldehydes are a primary component of automobile exha ust odor [60]. These alde hydes have good UV absorbance at 200 nm. The peak area RSD values for p -nitrobenzaldehyde and m -tolualdehyde were 8.0% and 8.4%, respectively. The detection li mits for these aldehydes were 90.8 ng/L for m -tolualdehyde and 358.4 ng/L for p -nitrobenzaldehyde. Alde hydes had overall higher detection limits. This is most likely due to the moderately polar nature of aldehydes tested and the non-polar nature of the solgel PDMDPS coating. Th e highest detection limit observed in this study pertained to p -nitrobenzaldehyde. This can be explained by the presence of a nitro group, which is very polar. Differences in UV absorption characteristics also contribute to differenc es in detection limits. A chromatogram

PAGE 263

232 indicating the extraction of aldehydes and ketones on the sol-gel PDMDPS coated capillary is shown in figure 5.5.

PAGE 264

233 Figure 5.5 A chromatogram representing on-line CME-HPLC analysis of aldehydes and ketones using a sol-gel PDMDPS coated microextraction capillary. Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel PD MDPS coated capillar y, 40 minute gravity fed extraction at room temperature. HP LC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 to 70/30 ACN/water over 20 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. Peaks: 1 = m -tolualdehyde (4 x 105 ng/L), 2 = p -nitrobenzaldehyde (4 x 105 ng/L), 3 = coumarin (1 x 105 ng/L), 4 = 4’phenylacetophenone (2 x 104 ng/L), 5 = trans-chalcone (2 x 104 ng/L).

PAGE 265

234 In order to validate the method for prep aration of the solgel PDMDPS coated capillary, capillary to capillary reproducib ility was determined for 5 analytes, each representative of a different chemical cla ss, on 3 separately prepared sol-gel PDMDPS coated capillaries. The observed capillary to capillary HPLC peak area reproducibility ranged from 3.8 to 10.1%. The capillary to cap illary reproducibility data for the sol-gel PDMDPS coated capillaries is presented in table 5.2.

PAGE 266

235 Table 5.2 Capillary-to-capillary peak area repro ducibility in CME-HPLC for the sol-gel PDMDPS coated capillaries. Chemical Class Chemical Name Capillary to Capillary (n = 3) Peak Area RSD (%) aldehyde m -tolualdehyde 4.3 ketone 4’phenylacetophenone 9.1 aromatic amine diphenylamine 10.1 PAH naphthalene 8.9 nonpolar aromatic biphenyl 3.8 Extraction conditions: 40 cm x 0.25 mm I. D. sol-gel PDMDPS coated capillary; extraction time: 40 minutes for all. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column; isocratic elution 70/30 ACN/water; 1 mL/min flow rate, UV detection at 217 nm for naphthalene, 205 nm for biphenyl and diphenylamine, and 200 nm for m -tolualdehyde and 4’phenylacetophenone.

PAGE 267

236 Aside from being an excellent extraction device, the sol-gel PDMDPS coated capillary provided quite stable and rugged pe rformance. The solvent resistance of the sol-gel PDMDPS coated capillary was proven by its long lifetime and durability when hyphenated on-line with HPLC. One sol-gel PDMDPS coated capillary that was tested survived 84 extraction/desorption cycles over a 2 year one month period with virtually no change in extraction performance (HPLC peak area RSD of 1.3%). This is exemplified by the chromatograms in figure 5.6.

PAGE 268

237 Figure 5.6 A Figure 5.6 B Figure 5.6 On-line CME-HPLC analysis of fluoranthene (1 x 104 ng/L) using a sol-gel PDMDPS coated microextraction capillary. Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel PDMDPS coated capillar y, 40 minute gravity fed extracti on at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Zorbax Phenyl column, isocratic elution 80/20 ACN/water, 1 mL/min flow rate, UV detecti on at 256 nm, ambient temperature. (Figure 5.6 A) = extraction # 2, HPLC peak area = 75626, (Figure 5.6 B) = extraction # 84, HPLC peak area = 77784.

PAGE 269

238 During this time period, the capillary wa s subjected to large amounts of aqueous samples, water and acetonitrile mobile phases, as well as frequent rinsing with organic solvents, such as methylene chloride and methanol. The sol-gel PDMDPS coated capillary is capable of low ng/L detection limits for the non-polar compounds tested in this study. This makes the sol-gel PDMDPS capillary very practical for use in the trace analysis of various non-polar analytes. The sol-gel PDMDPS capillary also has the capability of extracting some more polar compounds, such as aromatic amines, ketones, and aldehydes. Since it reaches equilibrium within 40 minutes for the chemical classes tested, the sol-gel PDMDPS coated capillary offers low detection limits with short analysis time wh en hyphenated on-line with HPLC. The solgel PDMDPS coated capillary had higher de tection limits for PAHs, ketones and aromatic amines than those obtained with a recently developed so l-gel methyl coated capillary [36], but the sol-gel methyl coated capillary requires longe r extraction times, so the sol-gel PDMDPS coated capillary offers the time advantage for the extraction of these compounds. This sol-gel coated capillary woul d be useful in the simultaneous extraction of mixtures of non-polar and moderately po lar analytes. An example of such an extraction is presented in figure 5.7. The sol-gel PDMDPS coated capillary also demonstrat ed potential for use in CME coupled to GC analysis. It remained intact under high-temp erature conditioning right to 350 C with no bleeding problems. The solvent stability of the sol-ge l PDMDPS coated capillaries was further demonstrated by testing a freshly prepared cap illary for its extraction capabilities of a mixture of analytes from different chemical classes both before and after rinsing with 50

PAGE 270

239 mL of 50/50 (v/v) methylene chloride/metha nol. Rinsing with these strong organic solvents actually improved the extraction capabilities of the sol-gel PDMDPS coated capillary. This indicates that the solvents thoroughly cleaned the so l-gel extracting phase, which allowed for the more efficient extracti on of analytes. After rinsing the capillary, an average of 14.8% increase in HP LC peak area was observed. High temperature HPLC is a modern trend in HPLC that aims at providing faster and efficient liquid-phase sepa ration. High temperature HPLC is typically performed at 120 C and has been shown to decrease the visc osity and to increase th e linear velocity of ACN/water mobile phases [61-63]. High temper ature HPLC also offers the advantages of improving peak shape while enabling fast er run times [64]. The use of thermal gradients in high temperature HPLC may actuall y replace solvent gradients [64]. In order for sol-gel capillary microextraction to be effectively hyphenated on-line with high temperature HPLC, it is necessary for the sol-ge l coated capillary to be resistant to high temperature solvents, especially since the m obile phase is typically preheated in high temperature HPLC [64]. A hi gh temperature solvent stability test was performed using a sol-gel PDMDPS coated capillary. For this, a sample containing a mixture of analytes from different chemical classes was first extr acted on a freshly prepared capillary. Next, this capillary was connected, using stainless steel unions on both ends, to two segments (~1 m) of stainless steel HPLC tubing. On e segment of the stainless steel tubing was connected to an HPLC pump, and the other to a waste containe r. The capillary was then placed completely inside a GC oven, so that the two stainless steel unions and approximately one third of both segments of stainless steel HPLC tubing remained completely inside the oven. The GC oven was heated to 150 C as 120 mL of 50/50 (v/v)

PAGE 271

240 acetonitrile/water mobile phase was pumped th rough the capillary at a flow rate of 1 mL/min. After cooling, the same sample was again extracted on the capillary. Little change in performance was noted. The high temp erature test was then repeated, this time heating to 200 C. The capil lary was again tested, with little change in extraction performance. This indicates that the sol-ge l PDMDPS coated capillary is suitable for use in high temperature HPLC up to at leas t 200 C. Chromatograms indicating the extraction of this mixture of analytes befo re heating, after heating to 150 C, and after heating to 200 C are shown in figure 5.7. A peak area comparison is given in table 5.3. Overall, the sol-gel PDMDPS coated capillary is a durable, stable, high-temperature solvent resistant capillary capable of excelle nt detection limits with good reproducibility in CME-HPLC with UV detection.

PAGE 272

241 Figure 5.7 A Figure 5.7 B Figure 5.7 C Figure 5.7 Chromatograms representing on-lin e CME-HPLC analysis of a mixture of polar and moderately polar co mpounds using a sol-gel PDMDPS coated microextraction capillary, (5.7 A) before rinsing and heating, (5.7 B) after rinsing with 120 mL of 50/50 ACN/water (v/v) at 150 C fo r 2 h, and (5.7 C) after rinsing capillary 5.7 B with 120 mL of 50/50 ACN/water (v/v ) at 200 C for 2 h Extraction conditions: 40 cm x 0.25 mm I.D. sol-gel PDMDPS capilla ry, 40 minute gravity fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 60/40 to 65/35 ACN/water over 5 min, to 80/20 ACN/water over 10 min, 1 mL/min flow rate, UV detection at 200 nm ambient temperature. Peaks: 1 = m tolualdehyde (2 x 105 ng/L), 2 = 4’phenylacetophenone (1 x 104 ng/L), 3 = diphenylamine (2.5 x 104 ng/L), 4 = naphthalene (1 x 104 ng/L), 5 = biphenyl (5 x 103 ng/L).

PAGE 273

242 Table 5.3 High temperature solvent stability of the sol-gel PDMDPS coated capillary Analyte Peak Area (arbitrary unit) Before After Rinsing with 120 mL ACN/H20 (50/50, v/v) Rinsing 150 C 200 C m -tolualdehyde 5.2 5.7 5.3 4’phenylacetophenone 5.0 4.9 5.1 diphenylamine 7.8 7.1 7.4 naphthalene 7.2 6.3 7.4 biphenyl 9.7 10.1 11.1 CME-HPLC-UV peak area comparison of a 40 cm x 0.25 mm I.D. sol-gel PDMDPS coated capillary before rinsing and hea ting, after rinsing wi th 120 mL of ACN/H2O (50/50, v/v) at 150 C for 2 h, and after se quentially rinsing with 120 mL of ACN/H2O (50/50, v/v) at 150 C for 2 h and 120 mL of ACN/H2O (50/50, v/v) at 200 C for 2 h.

PAGE 274

243 5.4 Conclusion To the best of our knowledge, this is the first report on th e use of a sol-gel polydimethyldiphenylsiloxane (PDMDPS) coat ing in capillary microextraction in hyphenation with HPLC. This study indicates that the methyl and phenyl groups within the polydimethyldiphenylsiloxane are capable of providing efficient extraction for nonpolar and moderately polar an alytes with good run-to-run repeatability and low ng/L detection limits in CME-HPLC with UV det ection. This sol-gel PDMDPS coating has the potential to become an excellent all-pur pose coating for the extraction of both polar and moderately polar analytes in the same mixture. Thanks to its direct chemical bonding to the capillary wall, the sol-gel PDMDPS coated capillary is highly resistant to solvents, even at high temper atures (up to 200 C). Furt hermore, the sol-gel PDMDPS coated capillary is thermally stable to over 350 C, as indicated by the conditioning process. This demonstrates that the sol-gel approach provides effective immobilization of diverse polymers to produce sol-gel cap illaries for use in CME-HPLC. The high temperature stability of the sol-gel PDMDPS coated capillary expands the capabilities of sol-gel CME to include possi ble applications in hyphena tion with high temperature HPLC. 5.5 References for chapter five [1] C. Arthur, J. Pawlisz yn, Anal. Chem. 62 (1990), 2145. [2] J. Hinshaw, LC-GC N. Am. 21 (2003) 1056. [3] R. Eisert, K. Levsen, J. Chromatogr. A 733 (1996) 143.

PAGE 275

244 [4] H. Lord, J. Chromatogr. A 1152 (2007) 2. [5] C. Zambonin, Anal. Bioa nal. Chem. 375 (2003) 73. [6] E. Lesellier, Analusis 27 (1999) 363. [7] P. Santos, B. Simonet, A. Rios, M. Valcarcel, Electr ophoresis 28 (2007) 1312. [8] A. Penaluer, E. Pocurull, F. Borrull, R. Marce, Trends Anal. Chem. 18 (1999) 557. [9] G. Jiang, M. Huang, Y. Cai, Z. Z ongshan, J. Chromatogr Sci. 44 (2006) 324. [10] M. Yang, Z. Zeng, W. Qiu, Y. Wang, Chromatographia 56 (2002) 73. [11] Y. Hu, Y. Yang, J. Huang, G. Li, Anal. Chim. Acta 543 (2005) 17. [12] D. Zhang, C. Wu, F. Ai, Chin. J. Chromatogr. 17 (1999) 10. [13] R. Yang, W. Xie, Fore nsic Sci. Int. 139 (2004) 177. [14] T. Ding, H. Lin, C. Wha ng, J. Chromatogr. A 1062 (2005) 49. [15] M. Farajzadeh, M. Hatami, J. Sep. Sci. 26 (2003) 802. [16] E. Koster, C. Crescenzi, W. Hoedt, K. Ensing, G. de Jong, Anal. Chem. 73 (2001) 3140. [17] W. Mullett, P. Martin, J. Pawliszyn, Anal. Chem. 73 (2001) 2383. [18] F. Mangani, R. Cenciarini, Chromatographia 41 (1995) 678. [19] K. Jinno, T. Masahiro, M. Hayashid a, J. Pharm. Biomed. Anal. 17 (1998) 1081. [20] H. Kataoka, Anal. Bioanal. Chem. 373 (2002) 31. [21] S. Chong, D. Wang, J. Hayes, B. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [22] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [23] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [24] D. Wang, S. Chong, A. Ma lik, Anal. Chem. 69 (1997) 4566.

PAGE 276

245 [25] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [26] Z. Zeng, W. Qiu, M. Ya ng, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [27] J. Zhou, Z. Zeng, Anal. Chim. Acta 556 (2006) 400. [28] L. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [29] Z. Zeng, W. Qiu, Z. Hu ang, Anal. Chem. 73 (2001) 2429. [30] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [31] C. Basheer, S. Jegadesan, S. Valiyav eettil, H. Lee, J. Chromatogr. A 1087 (2005) 252. [32] A. Kabir, C. Hamlet, K. Yoo, G. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [33] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [34] T. Kim, K. Alhooshani, A. Kabir, A. Malik, J. Chromatogr. A 1047 (2004) 165. [35] Y. Fan, Y. Feng, S. Da, Z. Wang, Talanta 65 (2005) 111. [36] S. Segro, A. Malik, J. Ch romatogr. A (2008) in press. [37] K. Alhooshani, T. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [38] L. Fang, S. Kulkarni, K. Alhoosha ni, A. Malik, Anal. Chem. 79 (2007) 9441. [39] J. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3. [40] C. J. Brinker, G.W. Scherer, Sol-Gel Science: The Chemistry and Physics of Sol-Gel Processing Academic Press, San Diego, CA, 1990. [41] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [42] H. Fang, M. Liu, Z. Zeng, Talanta 68 (2006) 979. [43] W. Liu, Y. Hu, J. Zhao, Y. X u, Y. Guan, J. Chromatogr. A 1102 (2006) 37.

PAGE 277

246 [44] M. Liu, Z. Zeng, H. Fa ng, J. Chromatogr. A 1076 (2005) 16. [45] M. Liu, Z. Zeng, C. Wang, Y. Ta n, H. Liu, Chromatographia 58 (2003) 597. [46] X. Li, Z. Zeng, J. Zhou, Anal. Chim. Acta 509 (2004) 27. [47] D. Wang, J. Xing, J. Peng, C. Wu, J. Chromatogr. A 1005 (2003) 1. [48] R.C. Mehrotra, J. NonCryst. Solids, 121 (1990) 1. [49] S. Kulkarni, A. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50. [50] A. L. Lopes, F. August o, J. Chromatogr. A 1056 (2004) 13. [51] R. Rodriguez, M. Flores, J. Gom ez, V.M. Castano, Mater. Lett. 15 (1992) 242. [52] R. Rodil, M. Schellin, P. Popp, J. Chromatogr. A 1163 (2007) 288. [53] G. Supin, D. Soboleva, E. Khlapova, G. Abramova, Problemy Analiticheskoi Khimii 1 (1970) 172. [54] K. Jinno, C. Okumura, M. Harada, Y. Saito, M. Okamoto, J. Liquid Chromatogr. Related Techn. 19 (1996) 2883. [55] T. Mazzo, A. Saczk, G. Umbuzeiro, M. Zanoni, Anal. Letters 39 (2006) 2671. [56] A. Diskin, P. Spanel, D. Smith, Physiol. Meas. 24 (2003) 107. [57] A. Manolis, Clin. Chem. 29 (1983) 5. [58] D. Smith, P. Cheng, P. Spanel, Rapid Commun. Mass Spectrom. 16 (2002) 1124. [59] D. Smith, P. Spanel, J. Jones, Bioresour. Sci. Tec hnol. 75 (2000) 27. [60] M. Roy, Energy Conversion and Management 49 (2008) 1111. [61] D. Stoll and P. Carr, J. Am. Chem. Soc. 127 (2005) 5034. [62] F. Antia, C. Horvath, J. Chromatogr. A 435 (1988) 1. [63] B. Yan, J. Zhao, J. Brown, J. Bl ackwell, P. Carr, Anal. Chem. 72 (2000) 1253. [64] J. Clark, Pharm. Technol. Europe 16 (2004) 41.

PAGE 278

247 CHAPTER SIX ULTRA-HIGH-STABILITY, pH-R ESISTANT SOL-GEL TITANIA POLY(TETRAHYDROFURAN) COATING FOR CAPILLARY MICROEXTRACTION ON-LINE CO UPLED TO HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY 6.1 Introduction Solid-phase microextraction (SPME) is an important technique for solvent-free sample preparation [1]. In SPME, a sorben t coating created on the end segment (~ 1 cm) of a fused silica fiber is used for extracti on [1]. Various sorbent coatings, including polysiloxanes with a diverse range of side chai ns [2-5], polyamide [6 ], polyvinyl chloride [7], inorganic polymers [8], and molecularly imprinted polymers (MIPs) [9,10] have been reported. To perform extraction, the coated fiber is submerged into a sample under mechanical agitation until analyte sorption-de sorption equilibrium is reached between the fiber coating and the sample [1]. The extracted analytes are most commonly desorbed into the injection port of a gas chromatogr aphy (GC) system [1,11]. SPME has also been coupled with capillary electrophoresis (CE) [12,13], supercritical fluid chromatography (SFC) [14], and high-perform ance liquid chromatography (HPLC) [15,16], but in the latter cases it requires complicated desorp tion devices, often resulting in sample loss, sample dilution, and reduced extraction sensitivity [17]. Shortcomings of fiber SPME include low sample capacity due to small sorbent loading in the short coated se gment of the fiber, bending of the needle on the syringe-like

PAGE 279

248 SPME device, frequently encountered mech anical damage to the coating during operation, and breaking of the delicate fiber. In-tube SPME [18] was developed to overcome these problems inherent in fiber SPME. In this format, SPME uses the sorbent coating on the inner surface of a fused silica capillary. Commonly, a piece of commercial GC column is used for this, and extraction is performed by permitting samples to pass through it until a sorption-de sorption equilibrium of the analytes is reached between the sample and the sorbent coating. In-tube SPME uses a longer coated segment, which should increase the overall ex traction sensitivity. However, conventional in-tube SPME coatings are thin (< 1 m in thickness), and despite using longer coated segments, the extraction sensitivity of this met hod is still low [19]. The creation of stable coatings of greater thickness is extremely difficult using conventi onal coating techniques [19]. Since conventional coatings are not chemically bonded to the capillary surface, such coatings are characterized by moderate thermal stability. This seriously limits the allowable desorption temperature (hence the maximum boiling point) of analytes amenable to in-tube SPME-GC analysis [20]. Finally, the absence of chemical anchorage to the inner surface of the cap illary results in poor solven t stability of conventional coatings that are prone to getting stripped off the capillary by the mobile phase when coupled to HPLC. Sol-gel coatings were developed to address these problems by chemically anchoring the coating to the surface of the fi ber [20] or capillary [21]. Since their introduction in 1997 [20], a variety of co atings have been reported, including poly(dimethylsiloxane) (PDMS)[20], hydr oxy fullerene[22], crown ether[23], oligomer[24], PDMS/poly(vi nyl alcohol)[25], and poly(et hylene glycol) (PEG)[26]

PAGE 280

249 based sol-gels. Introduced in 2002, sol-gel co atings for capillary microextraction (CME) include sol-gel PDMS [21], so l-gel PEG [21,27], sol-gel de ndrimer [28], sol-gel cyanoPDMS [29], electrical ly charged sol-gels [30,31], sol-gel zirconia poly(dimethyldiphenylsiloxane) [32], and solgel germania PDMS [33]. Sol-gel CME was successfully applied to GC [21,2729,32,33], CE [30,31], and inductively coupled plasma mass spectrometry [34,35]. The solvent stability of so l-gel CME coatings allows for an effective on-line hyphenation of CME with HPLC [36-39]. Malik and coworkers [36] introduced the first on-line hyphenation of CME to HPLC in 2004. A silica-based sol-gel -cyclodextrin coated capillary was devel oped and used in the HPLC analysis of non-steroidal anti-inflammatory drugs in urin e samples [37]. Recen tly, we reported solgel methyl [38] and sol-gel poly(dimet hyldiphenylsiloxane) (PDMDPS) [39] coated capillaries for on-line coupled CME-HPLC. Our group introduced a sol-gel t itania PDMS coated capillary [36] and successfully hyphenated CME on-lin e with HPLC. This PDMS coating was effective in extracting non-polar analytes wi th superior pH stability to that of silica-based sol-gel coatings [36]. A sol-gel tit ania-hydroxy-terminated silicone oil fiber was coupled to GC in the analysis of phenols, amines, and pol ycyclic aromatic hydrocarbons (PAHs) [40]. Also, a titania-based sol-gel dimethyl3,7-diaminobenzothiophe ne-5,5-dioxide-3,3,4,4,diphenylsulfone tetracarboxylic dianhydride fiber was developed for the determination of benzene, toluene, ethyl-benzene, xylen e, and halocarbons using SPME-GC [41]. Zirconia[32], alumina[42], and germania[ 33] based sol-gel coatings have also been used in microextraction and they were shown to possess good pH stability.

PAGE 281

250 In a previous study [43], a silica-based sol-gel poly(tetrahydrofuran) coated capillary was found to be very effective in extracting polar and nonpolar analytes when coupled to GC analysis. T itania’s superior pH stability [36] and poly-THF’s excellent extracting capabilities [43] for polar and non-polar analytes inspired us to develop a novel sol-gel coating possessing both of these desirabl e characteristics presented in this paper. To our knowledge, this is the first report on th e creation of a sol-ge l titania-based polyTHF coated capillary and its use in CME on-line coupled to HPLC. 6.2 Experimental 6.2.1 Equipment A Micro-Tech Scientific (Vista, CA, US A) Ultra Plus HPLC system equipped with a Linear UVIS 200 variable wavelengt h UV detector was used for on-line CMEHPLC analysis using a sol-gel titania pol y-THF coated microext raction capillary. A Phenomenex (Torrence, CA, US A) reversed-phase Luna C18 column (15 cm x 4.6 mm I.D.) was used for HPLC separations. A Fisher model G-560 Vortex Genie 2 system (Fisher Scientific, Pittsburg, PA, USA) was employed for thorough mixing of the sol solution constituents. Nanopure water (15 M ) was obtained from a Barnstead model 04741 Nanopure deionized water system (Barnstead/Thermodyne, Dubuque, IA, USA). On-line collection and processing of the CMEHPLC data were carried out using Chrom Perfect (version 3.5 for Windows) computer software (Justice Laboratory Software, Denville, NJ, USA).

PAGE 282

251 6.2.2 Chemicals and materials Polycyclic aromatic hydrocarbons (napht halene, acenaphthene, fluoranthene, and phenanthrene), phenols (2-chl orophenol, 3,4-dimethylpheno l, 2,3-dichlorophenol, 3,5dimethylphenol, 2,5-dichlorophenol, 2,6-di chlorophenol, and 2,4,5-trichlorophenol), amines (N-ethylaniline, N-methylaniline, caffeine, otoluidine and mtoluidine), alcohols (9-anthracenemethanol and cinnamyl alcohol ) and ketones (coumarin, benzophenone, and trans -chalcone) were purchased from Aldrich (Milwaukee, WI, USA). Benzhydrol, polyarginine-tyrosine (4:1) (avg. MW = 43 200), an d poly-glutamic acid-tyrosine (1:1) (avg. MW = 22 000) were obtained from Sigma (St. Louis, MO, USA). Acids (nicotinic and anthranilic), anthracene and 1-[1,1’-biphe nyl]-4-ylethanone were purchased from Eastman Kodak (Rochester, NY, USA). Re sorcinol was procured from Spectrum (Gardena, CA, USA). Diphenylamine and 1-na phthoic acid were purchased from J.T. Baker (Phillipsburg, NJ, USA). ochlorobenzoic acid, otoluic acid, and 2-naphthol were bought from Matheson Coleman & Bell (Cincinn ati, OH, USA). Trifluoroacetic acid (TFA) and tetraethyl orthosilicate (TEOS) were purchased from Acros (Morris Planes, NJ, USA). Titanium(IV) isopropoxide was obtaine d from Gelest (Morrisville, PA, USA). Poly(tetrahydrofuran) (poly-THF, M.W. = 250 g/mol) was a gift from BASF (Parsippany, NJ, USA). From Fisher Scientific, HPLC -grade methanol, methylene chloride, and acetonitrile were procured. Poly(ether et her ketone) (PEEK) tubing (1.59 mm x 0.51 mm x 1.52 m), Rheodyne type ferrules, and nuts (1.59 mm) were purchased from Upchurch (Oak Harbor, WA, USA). Finally, fused si lica capillary (0.25 mm I.D.) was obtained from Polymicro (Phoenix, AZ, USA).

PAGE 283

252 6.2.3 Preparation of the sol-gel titania po ly(tetrahydrofuran) coated capillary 6.2.3.1 Fused silica capillary pretreatment Prior to coating with sol-gel titani a poly-THF, a fused silica capillary was subjected to pretreatment that aimed at clean ing the capillary inner surface and generating silanol groups on the cleaned surface. For th is, a 5-m piece of the capillary (250 m i.d.) was first sequentially rinsed with 4 mL each of methylene chloride, methanol, and deionized water. Then, both ends of the cap illary were sealed utilizing an oxy-acetylene torch. The capillary was further placed inside a GC oven and then heated for 2 h at 350 oC. Subsequently, it was left to cool down to room temperature and the ends were cut open using an alumina wafer. Finally, one e nd of the capillary was connected to the GC injection port, and with continuous helium fl ow (1 mL/min) through the capillary, the GC oven temperature was programmed from 40 C to 250 C, at 5 oC/min, holding at 250 oC for 2 h. 6.2.3.2 Preparation of the sol-gel titania po ly(tetrahydrofuran) sol solution The following procedure was used to in situ create the sol-gel titania poly-THF coating on the inner walls of a fused silica capillary. First, a stock solution containing a 1:1 (v/v) ratio of titanium(IV) isopropoxide and methylene chloride was mixed in a clean microcentrifuge tube and vortexed for approximately 20 s. Next, 230 L of poly-THF (250 MW) was taken in a second microcentrifug e tube and 300 L of methylene chloride was added to it. The content was vortexed for approximately 15 s. Further, 200 L of the previously prepared st ock solution of 1:1 (v/v) ti tanium(IV) isopropoxide and methylene chloride was added to the mixtur e and vortexed for 1 min. A precipitate was formed during the vortexing process. Finally, 40 L of trifluoroacetic acid (TFA), which

PAGE 284

253 contained 73% deionized water, was a dded to the mixture and vortexed for approximately 1 min, during which the precip itate reduced significantly in size. The supernatant was transferred in to a new microcentrifuge tube using a micropipette, and further used to coat the hydrothermally pr etreated fused silica capillary (60 cm x 0.25 mm I.D.). For comparison purposes, a sol-gel titania coated capillary that did not contain any poly-THF was prepared. For this, 100 L of titanium(IV) isopropoxide was mixed with 100 L of methylene chloride, followed by vortexing for 20 s. Further, 60 L of TFA containing 73% water was added and the mixture was vortexed for 1 min. Thermal conditioning of this capillary was performed following exactly the same procedure used in the creation of the sol-gel tit ania poly-THF coated capillary. A sol-gel silica poly-THF coated capill ary was also prepared to further demonstrate the superior pH resistance of the sol-gel titania poly-THF coating. This sol solution was prepared using the exact same pr ocedure used for the sol-gel titania polyTHF coating. However, instead of using titanium(IV) isopropoxide, TEOS was used as the sol-gel precursor. Thermal conditioning of this capillary was also performed following the exact same procedure used in th e preparation of the sol-gel titania polyTHF coated capillary. 6.2.3.3 Sol-gel coating of the fused silica capillary A helium pressure-operated capillary filling/purging device [44] was used to fill the hydrothermally treated fused silica capi llary with the sol-ge l solution under 40 psi pressure. The capillary was completely fill ed with the sol solution and the latter was allowed to drip from the exit end of the capilla ry for 10 s. Following this, the exit end of

PAGE 285

254 the capillary was sealed using a rubber septum and then maintained under helium pressure (40 psi) for 12 min. The rubber se ptum was subsequently removed from the capillary exit, and the liquid content was then expelled fr om the capillary under helium pressure (40 psi). The coating was further dr ied with a helium flow (40 psi) through the capillary for 2 h. 6.2.3.4 Capillary conditioning The coated and dried capillary was in stalled in the GC oven and, under helium purge (1 mL/min), heated by programming the oven temperature from 40 C to 250 C, at 1 oC/min, and held for 2 h at 250 oC. The capillary was further cooled down to room temperature and rinsed with a mixture of 2 mL each of methylene chloride and methanol. Finally, using an HPLC pump, the capillary was rinsed with acetonitrile for 5 min at a flow rate of 1 mL/min. 6.2.4 On-line CME-HPLC analysis The schematic of the system used for on-line sol-gel CME-HPLC analysis is presented in figure 6.1. A water/acetonitrile m obile phase system was used to separate the analytes in both the isocratic and gradient elution modes. A Luna C18 HPLC column (15 cm x 4.6 mm I.D.) was equilib rated with the initial mobile phase composition to be utilized in the HPLC separation. A 38-cm segm ent of the sol-gel titania poly-THF coated capillary was connected to the six-port HPLC injection valve using capillary end fittings consisting of appropriate PEEK tubing sleeves, ferrules, and nuts. In this configuration, the capillary looked like an external sampli ng loop. A liquid sample was added to the gravity-fed sample delivery system [21] and the extraction was carried out by allowing the sample to flow through the sol-gel titania poly-THF coated capillary. To assist in the

PAGE 286

255 flow of the sample through the sol-gel micr oextraction capillary, a 56-cm segment of green PEEK tubing (0.762 mm x 1.575 mm) was connected to the sample delivery system. A stainless steel needle fitted on the exit end of the PEEK tubing was inserted into the HPLC injection port to provide a reliable means to connect the sample reservoir to the HPLC injector. The HPLC injector wa s maintained in the “load” position allowing the sample to pass through the sol-gel coated capillary and achieve extraction equilibrium (approximately 30-50 min extraction time) with the sol-gel titania poly-THF coating. After the equilibrium had been reached, the injection valve was switched to the “inject” position to desorb the extracted analytes and transfer them into the HPLC column with the help of mobile phase flow through the capillary.

PAGE 287

256 Figure 6.1 Experimental setup used to carry out the CME-HPLC experiments using the sol-gel titania poly-THF coated capillary. To perform extraction, aqueous samples containing analytes flow from the gravity-fed sample dispenser through the sol-gel titania poly-THF coated capillary, and then into a wa ste container. To perform analysis, the HPLC injection valve is turned to the inject position. This allows the mobile phase to flow through the coated capillary, which desorb s the analytes and transfers them into the HPLC column for separation follo wed by subsequent UV detection.

PAGE 288

257 6.3 Results and discussion In the course of sol-gel processing, th e sol-gel precursor(s) undergo hydrolytic polycondenation reactions to form a colloidal system known as the sol, which ultimately gets converted into a three dimensional liqui d-filled network structure known as the gel [45]. In this study, titania isopropoxide was used as the sol-gel precursor and poly-THF as the sol-gel active polymer. Methylene ch loride was used to dissolve the coating solution ingredients. TFA, containing 73% deionized water, was used as a chelating agent to decelerate the hydrolys is rate of the precursor [46]. A series of reactions took place in the sol solution within the capillary, resulting in the formation of the sol-gel titania polyTHF coating. Here are the main chemical events. Titanium isopropoxide underwent cont rolled hydrolysis in th e presence of water and the chelating agent [46]. Furt her, the hydrolyzed products underwent polycondensation, producing a three-dimensional sol-gel network. In the course of the hydrolytic polycondensation reactions, the hydroxy-terminated polymer, poly-THF, had the opportunity to get covalent ly bonded to the growing solgel network via condensation reactions. Finally, patches of the sol-gel titania poly-THF network growing in the vicinity of the capillary walls had the chance to condense wi th silanol groups on the inner walls of the pre-treated fuse d silica capillary. This chem ical bonding between the coating and the capillary walls provided the coating wi th solvent resistance – a quality needed for effective on-line hyphenation of CME with HPLC These reactions are illustrated in scheme 6.1.

PAGE 289

258 Scheme 6.1 Chemical reactions involved in the fo rmation of the sol-gel titania poly-THF coated capillary.

PAGE 290

259 To demonstrate that the poly-THF was bonded into the sol-ge l network and is mainly responsible for the extraction, a capi llary was prepared using just titanium(IV) isopropoxide with no poly-THF in the sol so lution. A sample containing six analytes, representing different chemical classes, wa s extracted on both capillaries. It was found that the sol-gel titania (without poly-THF) cap illary is capable of some extraction, but it is not nearly as efficient in the extraction of compounds from all six chemical classes. The chromatograms in figure 6.2 and 6.3 compare extraction on the sol-gel titania coated capillary (figure 6.2) with th at on the sol-gel titania polyTHF coated capillary (figure 6.3).

PAGE 291

260 Figure 6.2 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a so l-gel titania coated ca pillary without polyTHF. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5-dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acen aphthene (1 x 105 ng/L).

PAGE 292

261 Figure 6.3 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a so l-gel titania poly-THF coated capillary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravit y-fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o toluidine (2 x 105 ng/L), 3 = 3,5-dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acen aphthene (1 x 105 ng/L).

PAGE 293

262 In order to determine the time required fo r the analyte extraction equilibrium to be established between the sol-gel coating a nd the sample, an extraction profile was experimentally determined. Six compounds, eac h representing a different chemical class, were extracted on the sol-gel titania poly-THF coated capillary for different lengths of time (10, 20, 30, 40, 50, and 60 min) using three re plicate extractions for each of these periods. The replicate peak areas of each an alyte for each extraction time were averaged and plotted against their corres ponding extraction times (figure 6.4). From this graph, the equilibration time was determined from the point at which the peak area ceased to increase with respect to further increases in extraction time. Thus, otoluic acid, 1-(1,1’biphenyl)-4-ylethanone, 3,4-dimethylphenol, an d benzhydrol each needed approximately 40 min of extraction time, whereas mtoluidine and naphthalene needed 30 min and 50 min of extraction time, respectively.

PAGE 294

263 Figure 6.4 Extraction profiles of m -toluidine, otoluic acid, 1-[1,1’-biphenyl]-4ylethanone, benzhydrol, 3,4-dime thylphenol, and naphthalene for the sol-gel titania polyTHF coated capillary. HPLC cond itions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 50/50 ACN/H2O, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature.

PAGE 295

264 A range of polar analytes from differ ent chemical classes, such as phenols, alcohols, and amines, were extracted utilizing the sol-gel titania poly-THF coated capillary. Peak area RSD values ranging from 1.9 % to 9.7 % were obtained for these polar analytes. The detection limits ra nged from 128.6 ng/L to 451.6 ng/L, 354.1 ng/L to 1011.6 ng/L, and 111.6 ng/L to 660.6 ng/L for phenols, alcohols, and amines, respectively. The peak area RSD values and detection limits for all phenols, alcohols and amines are presented in table 6.1. Since polyTHF is a relatively polar polymer, the solgel titania poly-THF coating was capable of extracting polar analytes, such as phenols, alcohols, and amines. A chromatogram indi cating the extraction of several phenols and alcohols from the same mixture is shown in figure 6.5. A chromatogram indicating the extraction of polar and moderately pola r analytes is presented in figure 6.6.

PAGE 296

265 Table 6.1 HPLC peak area repeatability and dete ction limit data for phenols, alcohols, and amines in CME-HPLC using a sol-gel titania poly-THF-coated microextraction capillary. Chemical Chemical Peak area repeatability Detection limits class name (n=3) (ng/L) Mean peak area RSD (S/N = 3) (arbitrary unit) (%) Phenols 2,3-dichlorophenol 1.7 6.3 1.8 x 102 2-chlorophenol 0.7 2.2 4.5 x 102 3,4-dimethylphenol 0.7 3.4 4.2 x 102 3,5-dimethylphenol 1.3 2.6 2.4 x 102 2,4,5-trichlorophenol 2.4 9.7 1.3 x 102 2,5-dichlorophenol 1.4 3.9 2.2 x 102 2,6-dichlorophenol 1.2 4.9 2.7 x 102 Alcohols benzhydrol 0.9 3.3 3.5 x 102 cinnamyl alcohol 1.3 4.9 5.9 x 102 resorcinol 0.8 5.4 5.1 x 102 2-naphthol 0.7 2.8 4.8 x 102 9-anthracenemethanol 0.4 8.0 1.0 x 103 Amines caffeine 1.0 9.3 6.6 x 102 diphenylamine 1.4 1.9 1.1 x 102 o -toluidine 1.2 8.6 1.3 x 102 m -toluidine 0.8 6.8 4.0 x 102 N-methylaniline 1.5 1.9 2.1 x 102 N-ethylaniline 1.4 5.3 5.5 x 102 Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravit y-fed extraction at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution with ACN/H2O mobile phase, 1 mL/min flow rate, UV detection at 200 nm, at ambient temperature for all.

PAGE 297

266 Figure 6.5 A chromatogram representing CMEHPLC-UV analysis of phenols and alcohols using a sol-gel titania poly-THF coated capillary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O for 5 min to 70/30 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = resorcinol (5 x 105 ng/L), 2 = 2-chlorophenol (4 x 105 ng/L), 3 = 3,4-dimethylphenol (4 x 105 ng/L), 4 = 2-naphthol (4 x 105 ng/L), 5 = benzhydrol (4 x 105 ng/L).

PAGE 298

267 Figure 6.6 A chromatogram representingCME-H PLC-UV analysis of amines and ketones using a sol-gel titania poly-THF coated capillary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 65/35 to 75/25 ACN/H2O in 10 min, 1 ml/min flow rate, UV det ection at 200 nm, ambient temperature. 1 = methylaniline (4 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = diphenylamine (2 x 105 ng/L), 4 = 1-[1,1’-biphe nyl]-4-ylethanone (1 x 105 ng/L), 5 = trans-chalcone (1 x 105 ng/L).

PAGE 299

268 Besides being able to extract polar anal ytes, the sol-gel titania poly-THF coated capillary also provided extraction of modera tely polar analytes (ketones) and even nonpolar analytes (PAHs) with low detection limits. The peak area RSD values and detection limits for all ketones and PAHs are presented in table 6.2. This capability can be attributed to the methylene groups present in the poly-THF structure. The ability of poly-THF to extract nonpolar, moderately polar and polar analytes obs erved in this study is consistent with the results previously reported by us for a s ilica-based sol-gel polyTHF coating [43]. The peak area RSD valu es ranged from 5.4 % to 7.8 % for ketones and from 2.7 % to 9.1 % for PAHs. The de tection limits were between 39.4 ng/L and 263.5 ng/L for ketones and between 11.8 ng/L and 56.9 ng/L for PAHs. The sol-gel titania poly-THF coated capil lary was found to be significa ntly more selective in the extraction of ketones and PAHs when compared to the sol-gel titania PDMS coated capillary, which was also used in CME-HPLC-UV analysis [36]. The sol-gel titania poly-THF coated cap illary was even capable of extracting underivatized organic acids w ithout the need to perform any salting out and/or pH adjustment procedures that are typically us ed in the extraction of acids by SPME using conventional [47-49] or sol-gel coatings [50,51] Eliminating these procedures offers the advantages of time efficiency and a simple r extraction procedure without requiring steps that could potentially lead to sample loss and damage to the coating. For the tested acids, the peak area RSD values ranged from 3.3 to 8.0 % and the detection limits ranged from 447.1 to 2031.2 ng/L, as presented in table 6.2. Differences in the structure and UV absorbing ability of the acids partly contributed to the differences observed in the detection limits of the acids, as well as the other compounds analyzed in this study (table

PAGE 300

269 6.2). A chromatogram illustrating the separati on of two acids and their short retention times can be observed in figure 6.7.

PAGE 301

270 Figure 6.7 A chromatogram representing CME-HPLC -UV analysis of acids using a solgel titania poly-THF coated capillary. Ex traction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 60/40 ACN/H2O, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o toluic acid (2 x 105 ng/L).

PAGE 302

271 Table 6.2 HPLC peak area repeatability and detec tion limit data for acids, ketones, and PAHs in CME-HPLC using a sol-gel titania poly-THF-coated microextraction capillary. Chemical Chemical Peak area repeatability Detection limits class name (n=3) (ng/L) Mean peak area RSD (S/N = 3) (arbitrary unit) (%) Acids o -chlorobenzoic acid 0.4 6.3 7.4 x 102 nicotinic acid 0.1 3.3 2.0 x 103 o -toluic acid 0.4 8.0 6.9 x 102 1-naphthoic acid 1.3 4.4 4.5 x 102 anthranilic acid 0.2 3.6 1.9 x 103 Ketones 1-[1,1’-biphenyl]-4ylethanone 1.1 5.4 6.9 x 101 benzophenone 0.6 6.8 2.6 x 102 coumarin 4.0 6.9 3.9 x 101 trans-chalcone 1.0 7.8 8.3 x 101 PAHs anthracene 1.9 6.9 1.2 x 101 fluoranthene 0.8 6.8 5.7 x 101 naphthalene 1.4 9.1 1.8 x 101 phenanthrene 1.1 2.7 2.0 x 101 acenaphthene 1.3 6.8 3.9 x 101 Extraction conditions: 40 cm x 0.25 mm I.D. cap illary, gravity-fed extraction (40 min for ketones and acids, 50 min for PAHs) at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution with ACN/H2O mobile phase, 1 mL/min flow rate, UV detection at 200 nm for ketones, o -chlorobenzoic acid, and o -toluic acid, 224 nm for 1-naphthoic acid and anthranilic acid, 217 nm for acenaphthene and naphthalene, and 254 nm for anthracene, fluoranthene, and phena nthrene, at ambient temperature for all.

PAGE 303

272 The sol-gel titania poly-THF coated cap illary was capable of extracting polar compounds, such as acids, and non-polar analyt es, such as PAHs, from the same sample. An example of such extraction is shown in figure 6.8.

PAGE 304

273 Figure 6.8 A chromatogram representingCME-HPL C-UV analysis of acids and PAHs using a sol-gel titania poly-THF coated cap illary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 70/30 to 80/20 ACN/H2O in 10 min, 1 mL/min flow rate, UV detection at 220 nm, ambient temperature. 1 = 1-naphthoic acid (1 x 105 ng/L), 2 = naphthalene (1 x 104 ng/L), 3 = acenaphthene (5 x 104 ng/L), 4 = anthracene (4 x 105 ng/L).

PAGE 305

274 To validate the method for preparation of the sol-gel titania poly-THF coated capillary, capillary to capillary reproducibi lity was determined for six analytes, each representative of a different chemical class, on three separately prepared sol-gel titania poly-THF coated capillaries. The method for preparation was found to be very reliable, with capillary to capillary HPLC peak ar ea reproducibility rangi ng from 5.1 to 8.0%. The capillary to capillary re producibility data for the solgel titania poly-THF coated capillaries is pres ented in table 6.3.

PAGE 306

275 Table 6.3 Capillary-to-capillary peak area repro ducibility in CME-HPLC for the sol-gel titania poly-THF coated capillaries. Chemical class Chemical name Capillary to capillary (n = 3) Peak area RSD (%) Acid o -chlorobenzoic acid 5.1 Amine o -toluidine 7.4 Phenol 3,5-dimethylphenol 8.0 Alcohol benzhydrol 6.9 Ketone trans -chalcone 7.3 PAH acenaphthene 5.7 Extraction conditions: 40 cm x 0.25 mm I.D. so l-gel titania poly-THF coated capillaries; extraction time: 50 min. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column; gradient elution 55/45 ACN/water 5 min to 80/20 ACN/water in 15 min; 1 mL/min flow rate, UV detection at 200 nm for all.

PAGE 307

276 The sol-gel titania poly-THF coated capilla ry demonstrated excellent pH stability. Unlike silica-based materials, which are know n to be unstable under alkaline [52] and acidic [53] conditions, the solgel titania-based co ating survived an 18 h exposure to 1 M NaOH (pH 14.0) and 1 M HCl (pH 0.0). Chromatograms demonstrating the extraction of six analytes from six different chemical classes, showing the extraction capability of the sol-gel titania poly-THF co ated capillary before exposure to extreme pHs, after exposure to high pH, and after e xposure to low pH conditions are shown in figure 6.9, 6.10, and 6.11, respectively.

PAGE 308

277 Figure 6.9 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel titania poly-T HF coated capillary before exposure to acidic or basic co nditions. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-f ed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV dete ction at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 309

278 Figure 6.10 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel titania poly-T HF coated capillary after 18 h exposure to 1.0 M NaOH. Ex traction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity -fed extraction at room temperat ure. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV dete ction at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 310

279 Figure 6.11 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel titania poly-T HF coated capillary after 18 h exposure to 1.0 M HCl. Extrac tion conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-fed extraction at room temp erature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detecti on at 200 nm, ambient temperature. 1 = o chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5-dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 311

280 An HPLC peak area comparison for the si x analytes, before and after exposure of the capillary to high and low pH, is depicted in table 6.4. As evident from the table, the exposure of the coating to the base and aci d did not have a significant effect on its extraction performance. The pH stability of th is coated capillary is consistent with the pH stability observed in ot her sol-gel transition metalbased coatings [32,33,36,40,42]. The results of this pH stability test indicate that the sol-ge l titania poly-THF coating is suitable for use under both high and low pH conditions.

PAGE 312

281 Table 6.4 HPLC peak area comparison of acids, amines, phenols, alcohols, ketones, and PAHs before and after exposing the sol-gel titania poly-THF coated capillary to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 18 h. Chemical Chemical Before After NaOH After HCl class name exposure exposure exposure Peak area Peak area % Change Peak area % Change Acid o -chlorobenzoic 5.0 5.1 2.0 4.9 2.0 acid Amine o -toluidine 6.3 6.5 3.2 6.6 4.8 Phenol 3,5-dimethylphenol 13.3 13.7 3.0 13.3 0.2 Alcohol Benzhydrol 8.9 9.0 1.1 9.2 3.4 Ketone transchalcone 7.4 7.3 1.4 7.4 0.4 PAH Acenaphthene 9.3 8.9 4.3 9.6 3.2 Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravit y-fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. Pe ak areas in arbitrary units, average of 3 replicate measurements.

PAGE 313

282 For comparison, a sol-gel silica poly-THF co ated capillary was also prepared and subjected to the same pH treatment as the sol-gel titania poly-THF coated capillary. After subjecting the sol-gel si lica poly-THF coated capillary to 1 M NaOH or 1 M HCl, a significant decrease in extraction efficiency wa s observed. This indicat es that the sol-gel silica poly-THF coating, unlike the sol-gel titani a poly-THF coating, degrades under high and low pH conditions. Although the sol-gel silica poly-THF coating was not completely destroyed after the pH treat ment, like the commercial silica-based PDMS stationary phase tested in previous work [36], the si gnificant decline in performance would make it unsuitable for use under high or low pH conditions. Chromatograms showing the extraction capability of the sol-gel silica poly-THF coated capillary before exposure to extreme pH, after exposure to 1 M NaOH for 18 h, and after exposure to 1 M HCl for 18 h are shown in figures 6.12, 6.13, and 6.14, resp ectively. A peak area comparison is given in table 6.5.

PAGE 314

283 Figure 6.12 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel silica poly-THF coated capillary before exposure to acidic or basic co nditions. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-f ed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV dete ction at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 315

284 Figure 6.13 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel silica poly-THF coated capillary after 18 h exposure to 1.0 M NaOH. Ex traction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity -fed extraction at room temperat ure. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV dete ction at 200 nm, ambient temperature. 1 = o -chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 316

285 Figure 6.14 A chromatogram representing CME-HP LC-UV analysis of acids, amines, phenols, alcohols, ketones and PAHs using a sol-gel silica poly-THF coated capillary after 18 h exposure to 1.0 M HCl. Extrac tion conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravity-fed extraction at room temp erature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detecti on at 200 nm, ambient temperature. 1 = o chlorobenzoic acid (3 x 105 ng/L), 2 = o -toluidine (2 x 105 ng/L), 3 = 3,5-dimethylphenol (4 x 105 ng/L), 4 = benzhydrol (4 x 105 ng/L), 5 = trans -chalcone (1 x 105 ng/L), 6 = acenaphthene (1 x 105 ng/L).

PAGE 317

286 Table 6.5 HPLC peak area comparison of acids, amines, phenols, alcohols, ketones, and PAHs before and after exposing the sol-ge l silica poly-THF coated capillary to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 18 h. Chemical Chemical Before After NaOH After HCl class name exposur e exposure exposure Peak area Peak area % Change Peak area % Change Acid o -chlorobenzoic 4.0 0.8 80.0 0.9 77.5 acid Amine o -toluidine 6.7 3.1 53.7 2.6 61.2 Phenol 3,5-dimethylphenol 15.8 6.4 59.5 5.7 63.9 Alcohol Benzhydrol 13.6 5.2 61.8 4.2 69.1 Ketone transchalcone 20.7 10.6 48.8 6.8 67.1 PAH Acenaphthene 29.5 9.7 67.1 6.9 76.6 Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 50 min gravit y-fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 55/45 ACN/H2O for 5 min to 80/20 ACN/H2O in 15 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. Pe ak areas in arbitrary units, average of 3 replicate measurements.

PAGE 318

287 pH manipulation is often necessary in important environmental and biomedical analyses, including glycomics and proteomics. With its outstanding pH stability, the solgel titania poly-THF coated capillaries may serve as an effective tool for proteomics research, where proteins are often digested in to peptides of various lengths for further study. The extraction and separa tion of these peptides are often conducted under high or low pH conditions [54-58]. Therefore, having a highly pH stable solgel coating is very important for their effective extraction and se paration. To demonstr ate the potential of the sol-gel titania poly-THF coated capill ary for use in proteomics, two polypeptides, poly-arginine-tyrosine (4:1) a nd poly-glutamic acid-tyrosine (1:1), were extracted using the sol-gel titania poly-THF coated capillary at pH values near their isolectric points (12.4 and 4.4, respectively). On-line extr action and HPLC-UV analysis of a 1 x 105 ng/L sample of poly-arginine-tyrosi ne (4:1) is presented in figure 6.15. The average molecular weight of poly-arginine-tyr osine (4:1) was 22 000, corresponding to an average peptide chain length of 107 residues. The three peaks observed in figure 6.15 can be attributed to polypeptides of different chain lengths and molecular weights within the poly-argininetyrosine (4:1) sample, with the shorter, lower molecular weight polypeptides eluting first. This extraction was performed at a pH of 12.4, which indicates that the sol-gel titania poly-THF coated capillary is capable of extracting polypeptides at highly basic pH values. The extraction of a 1 x 105 ng/L sample of poly-glutamic acid-tyrosine (1:1) on the sol-gel titania poly-THF coated capillar y is presented in fi gure 6.16. For the polyglutamic acid-tyrosine (1:1) peptide, the average molecular weight was 43 200, corresponding to a peptide chain length of 263 residues. Again, the multiple peaks observed in the chromatogram (figure 6.16) likely correspond to a few poly-glutamic

PAGE 319

288 acid-tyrosine (1:1) polypeptides with different chain lengths. In this extraction, the pH was 4.4, thus demonstrating that the sol-gel t itania poly-THF coated capillary is capable of extracting polypeptides at ac idic pH values as well.

PAGE 320

289 Figure 6.15 A chromatogram representing CME-HP LC-UV analysis of poly-argininetyrosine (4:1) using a sol-gel titania poly-THF coated capil lary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravity-fe d extraction at room temperature, pH = 12.4. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 70/30 to 90/10 ACN/H2O in 5 min, 1 mL/min flow rate UV detection at 200 nm, ambient temperature. 1, 2, and 3 = poly-arginine -tyrosine (4:1) polypeptides (1 x 105 ng/L) of different lengths and molecular weights.

PAGE 321

290 Figure 6.16 A chromatogram representing CME-HP LC-UV analysis of poly-glutamic acid-tyrosine (1:1) using a sol-gel titani a poly-THF coated capillary. Extraction conditions: 40 cm x 0.25 mm I.D. capillar y, 40 min gravity-fed extraction at room temperature, pH = 4.4. HPLC condi tions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution 80/20 ACN/H2O, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. 1, 2, 3, and 4 = poly-glutamic acid-tyrosine (1:1) polypeptides (1 x 105 ng/L) of different lengths and molecular weights.

PAGE 322

291 6.4 Conclusion The sol-gel titania poly-THF coated cap illaries demonstrated impressive pH stability by surviving exposure to both highl y basic and acidic environments (pH of 14.0 and 0.0, respectively) for extended periods of tim e. These capillaries are also capable of providing simultaneous extraction of nonpolar (e.g., PAHs), moderately polar (e.g., ketones), and polar analytes (e.g., phenols, alcohols, and amines) from the same sample with ng/L level detection limits in CME-HPLC -UV analysis. Especially notable is the fact that these sol-gel cap illaries can provide efficien t extraction of underivatized aromatic carboxylic acids, which are normally difficult to extract without having to perform complicated pretreatment and derivatiz ation procedures. Ex tracting all of these compounds can be important for many environm ental and biomedical applications, such as the detection of toxic chemicals in water samples and the detection of biomarkers in body fluids, respectively. The sol-gel titania poly-THF coated capillary provided higher extraction sensitivity than other sol-gel titaniabased coatings reported in the literature. For all of the analytes extracted on this solgel capillary, the HPLC peak area RSD values ranged from 1.9 to 9.7 %. Finally, the solgel titania poly-THF coated capillary was capable of extracting polypep tides under extremely high and low pH conditions. This indicates that the sol-gel titania poly-THF coated capillary is suitable for use in proteomics applications, especially for the preconcentration of lowabundant proteins and peptides resulting from the digestion of these biomedically important signature molecules.

PAGE 323

292 6.5 References for chapter six [1] C. L. Arthur, J. Pawl iszyn, Anal. Chem. 62 (1990) 2145. [2] G. Jiang, M. Huang, Y. Cai, J. Lv, Z. J. Zhao, Chromatogr. Sci. 44 (2006) 324. [3] M. Yang, Z. R. Zeng, W. L. Qiu, Y. L. Wang, Ch romatographia 56 (2002) 73. [4] Y. Hu, Y. Yang, J. Huang, G. Li, Anal. Chim. Acta 543 (2005) 17. [5] D. Zhang, C. Wu, F. Ai, Chin. J. Chro matogr. 17 (1999) 10. [6] R. Yang, W. Xie, Fore nsic Sci. Int. 139 (2004) 177. [7] M. A. Farajzadeh, M. Hatami, J. Sep. Sci. 26 (2003) 802. [8] F. Mangani, R. Cenciarini Chromatographia 41 (1995) 678. [9] E. H. Koster, C. Crescenzi, W. den Hoed t, K. Ensing, G. J. de Jong, Anal. Chem. 73 (2001) 3140. [10] W. M. Mullett, P. Martin, J. Pawliszyn, Anal. Chem. 73 (2001) 2383. [11] R. Eisert, K. Levsen, J. Chromatogr. A 733 (1996) 143. [12] B. Santos, B. M. Simonet, A. Rios, M. Valcarcel, Electr ophoresis 28 (2007) 1312. [13] A. Penalver, E. Pocurull, F. Borrull, R. M. Marce, Trends Anal. Chem. 18 (1999) 557. [14] E. Lesellier, Analusis 27 (1999) 363. [15] H. L. Lord, J. Ch romatogr. A 1152 (2007) 2. [16] C. G. Zambonin, Anal. Bioanal. Chem. 375 (2003) 73. [17] K. Jinno, M. Taniguchi, M. Hayashid a, J. Pharm. Biomed. Anal. 17 (1998) 1081. [18] R. Eisert, J. Pawlis zyn, Anal. Chem. 69 (1997) 3140. [19] H. Kataoka, Anal. Bioanal. Chem. 373 (2002), 31.

PAGE 324

293 [20] S. L. Chong, D. Wang, J. D. Hayes, B. W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [21] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [22] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [23] Z. Zeng, W. Qiu, Z. Hu ang, Anal. Chem. 73 (2001) 2429. [24] C. Basheer, S. Jegadesan, S. Valiyaveet til, H. K. Lee, J. Chromatogr. A 1087 (2005) 252. [25] A. L. Lopes, F. August o, J. Chromatogr. A 1056 (2004) 13. [26] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [27] S. Kulkarni, A. M. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50. [28] A. Kabir, C. Hamlet, K. Soo Yoo, G. R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [29] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [30] W. Li, D. Fries, A. Malik, J. Sep. Sci. 28 (2005) 2153. [31] W. Li, D. Fries, A. Alli, A. Malik, Anal. Chem. 76 (2004) 218. [32] K. Alhooshani, T. Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [33] L. Fang, S. Kulkarni, K. Alhoosha ni, A. Malik, Anal. Chem. 79 (2007) 9441. [34] F. Zeng, B. Hu, Talanta 73 (2007) 372. [35] Y. Wu, B. Hu, Z. Jiang, Y. Feng, P. Lu, B. Li, Rapid Commun. Mass Spectrom. 20 (2006) 3527. [36] T. Y. Kim, K. Alhoosha ni, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165.

PAGE 325

294 [37] Y. Fan, Y. Q. Feng, S. L. Da, Z. H. Wang, Talanta 65 (2005), 111. [38] S. S. Segro, A. Malik, J. Chromatogr. A 1200 (2008) 62. [39] S. S. Segro, A. Malik, J. Chromatogr. A 1205 (2008) 26. [40] X. Li, J. Gao, Z. Zeng, Anal. Chim. Acta 590 (2007) 26. [41] L. M. Wei, Q. Y. Ou, J. B. Li, Chin. Chem. Letters 15 (2004) 1127. [42] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [43] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1. [44] J. D. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3. [45] C. J. Brinker, G. W. Scherer, Sol-Gel Science: The Chemistry and Physics of Sol-Gel Processing Academic Press, San Diego, CA, 1990. [46] J. Livage, M. Henry, C. Sanchez, Prog. Solid State Chem. 18 (1988) 259. [47] L. Pan, M. Adams, J. Pa wliszyn, Anal. Chem. 67 (1995) 4396. [48] M. Huang, T. Chao, Q. Zhou, G. Jiang, J. Chromatogr. A 1048 (2004) 257. [49] M. Liu, Z. Zeng, Y. Lei, H. Li, J. Sep. Sci. 28 (2005) 2306. [50] M. Liu, Z. Zeng, B. Xiong, J. Chromatogr. A 1065 (2005) 287. [51] M. Liu, Z. Zeng, Y. Tia n, Anal. Chim. Acta 540 (2005) 341. [52] A. Wehrli, J. C. Hildenbrand, H. P. Ke ller, R. Stampfli R. W. Frei, J. Chromatogr. 149 (1978) 199. [53] J. L. Glajch, J. J. Kirkland, J. Koehler, J. Chromatogr. 384 (1987) 81. [54] G. Mitulovic, K. Me chtler, Brief. Funct. Genom. Proteom. 5 (2006) 249. [55] S. Gedela, N. R. Medicher la, Chromatographia 65 (2007) 511. [56] H. Cui, J. Leon, E. Reusaet, A. Bult, J. Chromatogr. A 104 (1995) 27.

PAGE 326

295 [57] L. Zeng, C. E. Nath, P. J. Shaw, J. W. Earl, A. McLachlan, J. Biomed. Chromatogr. 22 (2008) 879. [58] C. K. Larive, S. M. Lunte, M. Zhong, M. D. Perkins, G. S. Wilson, G. Gokulrangan, T. Williams, F. Afroz, C. Schoeneich, T. S. Derrick, C. R. Middaugh, S. BogdanowichKnipp, Anal. Chem. 71 (1999) 389R.

PAGE 327

296 CHAPTER SEVEN SOL-GEL GERMANIA TRI-BLOCK POLYMER COATINGS OF EXCEPTIONAL pH STABILITY IN CAPILLARY MICROEXTRACTION ONLINE COUPLED TO HIGH-PERFORM ANCE LIQUID CHROMATOGRAPHY 7.1 Introduction Since their introduction in solid-phase microextraction (SPME) in 1997 [1], solgel coatings have steadily gained popularity in sampling, sample pr eparation, and analyte preconcentration as they eff ectively overcome the drawbacks of SPME with traditionally coated fibers [2]. The primary shortcom ings of fiber SPME stem from (1) physical location of the sorbent coati ng on the external surface of th e fiber, (2) format of the SPME device where the fiber is moved back a nd forth through the narrow passage within a needle making the sorbent coating locat ed on the external surface of the fiber susceptible to mechanical damage during ope ration, (3) the short length of the coated segment of the fused silica fiber, and (4 ) the lack of chemical bonding between the sorbent coating and the fiber. All this may c ontribute to the vulnerability of the fiber to mechanical damage/failure, and limited therma l and solvent stability [1]. Also, it is difficult to couple fiber SMPE to HPLC since it requires complicated desorption devices, which often result in sample loss, sample dilution, and overall reduced sensitivity [3]. Intube SPME overcomes the disadvantages relate d to external exposur e of the extracting phase by coating it on the inner walls of a capillary [4]. The disadvantages associated with in-tube SPME include thin sub-microm eter coating thickness [5] and lack of

PAGE 328

297 chemical anchoring of the coating to the i nner surface of the capilla ry, seriously limiting the thermal and solvent stability of the sorb ent [1]. Sol-gel coatings overcome this problem since they are chemically anchored to the fiber. An array of silica-based sol-gel coatings have been developed, including so l-gel poly(dimethylsiloxane) (PDMS) [1], poly(ethylene glycol) (PEG) [6], acrylate [7], crown ether [8], hydroxy fullerene [9], and calixarene [10] coatings. Sol-gel capillary microextraction (CME), introduced in 2002 by Malik and coworkers [11], offers improve d extraction capabilities and easy on-line hyphenation with high-performance liquid chromatography [12-17]. Sol-gel CME coatings have also been coupled to GC [11,18-23], CE [24,25], and inductively coupled plasma mass spectrometry [26,27]. The durabil ity and thermal stability of a sol-gel coating stems from the strong covalent bonding of the sol-gel coating to the surface of the fiber [1] or capillary [11]. Most sol-gel coatings are silica-based. However, a significant disadvantage of silica-based sol-gel coatings is the instability of the s iloxane bond under basic [28] and acidic [29] conditions. To overcome this disa dvantage, sol-gel zirc onia[21], alumina[30], and titania[12,16,31,32] based coatings have been developed, and they were shown to possess enhanced pH stability. Howe ver, since these coatings were prepared using transition metal oxides, their surface chemis try differs from that of silica. Recently, the first germania-based sol-gel coatings [33] were reported for use as high and low pH resistant sorbents in sol-gel capillary microext raction coupled to GC analysis and as a GC stationary phase. Germania offers a dis tinct advantage over transition metal oxides because it is an isostructural analog of s ilica and therefore, possesses analogous surface chemistry [33]. In our previous report on sol-gel germania coating [33], poly-

PAGE 329

298 dimethylsiloxane was covalently bonded into the sol-gel germania network via germanium-oxygen-silicon bonds. For the first time, here we report the creation of sol-gel germania tri-block polymer coatings. In addition, this is the fi rst report of a sol-gel germania-based coating used in capillary microextraction on-l ine hyphenated with high-performance liquid chromatography (CME-HPLC). In this work, the sol-gel precursor tetran butoxygermane (TNBG) was used in conjuncti on with a tri-block polymer [34] to develop a novel sol-gel germania hybrid organi c-inorganic coating. In this coating, the hydroxy-terminated tri-block polymer, poly( ethylene oxide)–block-poly(propylene oxide)–block-poly(ethylene oxide ) (PEO-PPO-PEO), was covale ntly bonded into the solgel germania network. Since this polymer is amphiphilic, such a coated capillary can be expected to provide efficient extraction for nonpolar, moderately polar, and polar analytes from aqueous solution. Also, exception al pH stability of ge rmania-based sol-gel coatings, will make such a coating suit able for use under highly acidic and basic conditions often needed in a va riety of separation techniqu es and applications, including ion chromatography [35], hydrophobic interac tion chromatography [35], isoelectric focusing [36], proteomics [37-41], and in HP LC with electrochemical detection, where low or high pH mobile pha ses are required [42-50]. 7.2 Experimental 7.2.1 Equipment A model 04741 Barnstead Nanopure deionized water system (Barnstead/Thermodyne, Dubuque, IA) was used to produce 15M nanopure water for

PAGE 330

299 use in HPLC mobile phases and for the prep aration of aqueous samples for CME. A G560 Fisher Vortex Genie 2 (Fisher Scient ific, Pittsburgh, PA) was used for thorough mixing of the sol solution i ngredients. A Micromax 3590 F microcentrifuge (Thermo IEC, Needham Heights, MA) was used for centrifugation (14 000 rpm, 15 682 x g) to separate the precipita tes from the sol solution. CME-HPLC experiments were conducted using a Micro-Tech Scientific (Vista, CA, US A) Ultra-Plus HPLC system with a Linear UVIS 200 variable wavelength UV detector. An in-house built filling/purging device [51] was used to rinse, fill, and purge th e capillaries under nitrogen pressure. Online data collection and processing was performed usi ng Chrom Perfect versi on 3.5 (for Windows) computer software (Justice Laboratory Software, Denville, NJ). 7.2.2 Chemicals and materials Fused silica capillary (0.250 mm, I.D.) w ith a polyimide external coating was purchased from Polymicro Technologies (Phoe nix, AZ, USA). PEO-PPO-PEO tri-block polymer, benzhydrol, 9-anthracenemethanol, m -toluidine, o -toluidine, N -methylaniline, 2,6-dimethylphenol, 2,4-dichlor ophenol, 2,4,6-trichlorophenol, coumarin, acenaphthene, 1,2-benzanthracene, trans -chalcone, fluorene, phenanth rene, and fluoranthene were obtained from Aldrich (Milwaukee, WI, USA). HPLC-grade solvents, methylene chloride, methanol, and acetonitrile, were bought from Fisher Scientific (Pittsburgh, PA, USA). Naphthalene, trifluor oacetic acid (TFA 99%) and 2chlorophenol were procured from Acros (Morris Planes, NJ, USA). Anthracene and 4’phenylacetophenone were purchased from Eastman Kodak (Rochester, NY, USA). Resorcinol was obtained from Spectrum (Gardena, CA, USA). 2-naphthol was purchased from Matheson, Coleman & Bell (Cincinnati, OH, USA). Diphenylam ine was procured from J.T. Baker

PAGE 331

300 (Phillipsburg, NJ, USA). Tetran -butoxygermane (TNBG) was obtained from Gelest (Morrisville, PA, USA). Poly(e ther ether ketone) (PEEK) tubing (1.59 mm 0.51 mm 1.52 m), Rh eodyne type ferrules, and nuts (1.59 mm) were purchased from Upchurch (Oak Harbor, WA, USA). 7.2.3 Surface cleaning and hydrothermal pretreatment of fused silica capillary A 5-m segment of fused silica capillary was pretreated by sequentially rinsing with 4 mL each of methylen e chloride, methanol, and 15M deionized water using a homemade capillary filling/pur ging device [51] under nitrogen pressure (10 psi). A small amount of water was left in the capillary in the form of a thin surface coating. The ends of the capillary were flame sealed using an oxy-acetylene torch. Ne xt, the capillary was placed in a GC oven and heated at 350 oC for 2 h, then allowed to cool down to room temperature. After this, the cap illary ends were cut open usin g an alumina wafer. It was then installed in a GC oven with one end connected to the inj ection port providing a continuous helium flow through the capillary at a rate of 1 mL/min. The temperature was programmed from 40 oC to 250 oC at a rate of 5 oC/min. The capillary was held at 250 oC for 2 h, then allowed to cool. 7.2.4 Preparation of the sol solution The sol-gel germania triblock polymer coating solution was prepared by first weighing out 0.15 g of hydroxy-terminated PEOPPO-PEO tri-block polymer into a clean microcentrifuge tube. Second, 80 L of methylene chloride was added to the microcentrifuge tube, followed by vortexing for 20 s. Third, 43 L of TNBG was added and the mixture was vortexed for 30 s. La st, 75 L of trifluor oacetic acid, which contained 2% deionized water, was added a nd the solution was vortexed for an additional

PAGE 332

301 30 s. The solution was then centrifuged for 4 min, and the supernatan t was transferred to a clean microcentrifuge tube and subsequently used to coat a pretreated fused silica capillary. 7.2.5 Coating and conditioning of the capillary A 1-m section of the hydrothermally pretr eated fused-silica capillary was inserted into the nitrogen-pressure ope rated capillary filling/purging device [51]. The capillary was then filled with a freshly prepared so l solution under 60 psi of nitrogen pressure. Three drops of the solution were allowed to dr ip down the exit end of the capillary before it was sealed by capping with a rubber septum. The nitrogen pressure was then dropped to 40 psi and maintained at that level for the next 30 min. During this period, a surfacebonded sol-gel germania-tricodk polymer coating was created. After this, while the bulk of the coating solultion still remained in the liquid form, the septum was removed from the exit end of the capillary and the unbonded portion of the sol-gel solution was expelled from the capillary under nitrog en pressure. The capillar y was then purged for an additional 90 min under 40 ps i nitrogen pressure. Next, the sol-gel coated capillary was c onditioned to facilitate the completion of the sol-gel reactions. To accomplish this, the sol-gel germania tri-block polymer coated capillary was placed in a GC oven and purge d with helium flow of 1 mL/min. The temperature of the oven was simultaneously programmed from 40 oC to 200 oC at a rate of 1 oC/min, holding the capillary at the final te mperature for an additional 4 h. After this, the capillary was rinsed with 8 mL of a 1:1 (v/v) mixture of methylene chloride and methanol.

PAGE 333

302 7.2.6 Preparation of aqueous samples for CME-HPLC analysis Concentrated stock solutions (1 mg/mL) of target analytes were prepared by dissolving 10 mg of each analyte into 10 mL of methanol in 20 mL glass vials. Dilute aqueous samples were prepared for CME by transferring the appropr iate amount of the stock solution, using a micropipette, into vol umetric flasks (100 or 250 mL), followed by filling to the calibrated volume mark with nanopure deionized water (15M ). 7.2.7 CME-HPLC analysis of aqueous samples The system used for CME-HPLC analysis is depicted in figure 7.1. A 40 cm section of the sol-gel germania tri-block pol ymer coated capillary was cut and the ends were fitted with 3-cm sleeves of PEEK t ubing (1.59 mm O.D.), nuts, and ferrules to install the capillary as an external sampling loop on a six-port HPLC injection valve. The analyte solution was placed into an in-labor atory designed gravity-f ed sample dispenser [11]. With the injec tion valve in the “load” position, the liquid sample containing the target analyte was allowed to pass through the sol-gel coated capillary via gravity flow until an extraction equilibrium was establis hed between the sample and the sol-gel germania tri-block polymer extracting pha se (typically 30-40 min, depending on the sample). The valve was then switched to th e “inject” position, allo wing the mobile phase (ACN/H20) to desorb the analytes from the sol-ge l coated capillary and carry them to a Luna C18 HPLC column (15 cm x 4.6 mm I.D.) for separation followed by UV detection. Both isocratic and gradient elution with UV detection were used to achieve adequate separation and detection of the analytes in the samples.

PAGE 334

303 Figure 7.1 Experimental setup used to carry out the CME-HPLC experiments using solgel germania tri-block polymer coated capilla ries. To perform gravity-fed extraction, an aqueous sample containing analytes is allo wed to flow from the gravity-fed sample dispenser through the sol-gel germania tri-bl ock polymer coated capillary until extraction equilibrium is established. To perform analysis the HPLC injection valve is turned to the inject position. This allows the mobile phase to flow through the co ated capillary, which desorbs the analytes and transfers them into the HPLC column for separation followed by UV detection.

PAGE 335

304 To investigate the effect of sample fl ow rate through the cap illary on the time required to reach extraction equilibrium, a th ird HPLC pump was c onnected to the waste tube of the six-port injecti on valve. Extraction was performed by pumping the sample solution through the thoroughly cleaned waste tube of the six-por t injection valve, through the sol-gel germania tri-block polymer coated capillary, and out of the injection port. This alternative syst em is depicted in figure 7.2.

PAGE 336

305 Figure 7.2 Experimental setup used to carry out the CME-HPLC analysis using an HPLC pump to pass the sample through sol-gel germ ania tri-block polymer coated capillaries. To perform extraction using enhanced flow rates, aqueous samples containing analytes are pumped using an HPLC pump into the injection valve through the thoroughly cheaned waste line, through the sol-gel germ ania tri-block polyme r coated extraction capillary, and out of the inj ection port to an appropriate waste container. To perform analysis, the HPLC injection valv e is turned to the inject pos ition. This allows the mobile phase to flow through the coat ed capillary, which desorbs the analytes and transfers them into the HPLC column for separation followed by UV detection.

PAGE 337

306 7.2.8 Conversion of peak areas to amounts extracted (ng) In the tables and extraction profiles, th e peak areas obtained using Chrom Perfect 3.0 software were converted to the amounts ex tracted, expressed in ng. To perform this conversion, known sample volumes (20 L) of known concentrations (20 mg/L) were directly injected into the HP LC system, which corresponds to injecting 400 ng of analyte. The peak areas obtained for the direct inject ion of 400 ng of each analyte were used to calculate the mass of each analyte extracted in ng. 7.3 Results and discussion In typical sol-gel reactions, the so l-gel precursor undergoes hydrolytic polycondensation reactions, forming a colloidal system (the sol) The sol subsequently is converted into a three-dimensiona l liquid-filled network (the ge l) [52]. In this work, the germania-based sol-gel precursor, tetran -butoxygermane, was employed. Hydroxyterminated PEO-PPO-PEO served as the sol-ge l active polymer. For the adequate mixing of all of the sol solution constituents, met hylene chloride was used as the solvent. Trifluoroacetic acid (TFA) was utilized as a ch elating agent to slow the rate of hydrolysis of the germanium alkoxide sol-gel precursor A 2% water content was found to be optimum to achieve a gelation time of ~ 60 min. The reaction scheme for the sol-gel germania tri-block polymer coating is illu strated in schemes 7.1 and 7.2. Controlled hydrolysis of tetran -butoxygermane was conducted in the pr esence of trifluoroacetic acid and water [53]. The reac tive hydrolyzed products unde rwent polycondensation to produce an evolving germania-based sol-gel netw ork structure. During this process, the sol-gel active terminal hydroxyl groups of the tri-bloc k polymer had the opportunity to condense into the growing sol-gel matrix. The portions of the sol-ge l germania tri-block

PAGE 338

307 polymer matrix in the vicinity of the inner walls of the pr etreated fused silica capillary condensed with silanol groups to produce a su rface-bonded coating to serve as the sol-gel extracting phase.

PAGE 339

308 Ge O(CH2)3CH3 4 H2O Ge OH OH HO OH 4 CH3(CH2)3OHCo n t r o l l ed h yd r o l ys i s of t h e so l -ge l p r ecu r so r ( t e t r an b u t oxyge r m a n e):Chelating Agent Ge OH OH HO OH Ge OH OH n HO OH n H2OPolycondensation of th e hydrolyzed products:Ge O O HO (O Ge)n O O OH + + + + Condensation of hydroxy-termin ated poly(ethylene oxide)-block-poly(propylene oxide)-blockpoly(ethylene oxide) to the evolving sol-gel network:Ge O O HO (O Ge)n O O OH + n H2O + O(CH2)3CH3O(CH2)3CH3CH3(CH2)3O HO [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ Hy]zGe O O HO (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ y]z Ge O O (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ Hy]z[O (CH2)2 O ]xCH CH3[ CH2 O ](CH2)2[y]z H Scheme 7.1 Chemical reactions involved in the formation of the sol-gel germania triblock polymer coated capillary.

PAGE 340

309 Bo n d i n g t o i n n e r w a l l of fused-s i l i c a c a p i l l a r y: H2O OH OH OHinner capillary wall su r f a ceb o n ded so l g e l g e r m a n i a t r i b l oc k po l y m e r co a t i n g + + Ge O O HO (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ y]z OH O OH Ge O O (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ Hy]z[O (CH2)2 O ]xCH CH3[ CH2 O ](CH2)2[y]z H Ge O O (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ y]z Ge O O (O Ge)n O O O [(CH2)2 O ]xCH CH3[ CH2 O ](CH2)2 O[ Hy]z[O (CH2)2 O ]xCH CH3[ CH2 O ](CH2)2[y]z H Scheme 7.2 Chemical anchoring of the evolving sol-gel germania tri-block polymer network to the inner walls of a fused-silica capillary.

PAGE 341

310 To demonstrate that the tri-block polym er has chemically bonde d into the sol-gel germania network, FTIR spectra were obtai ned for germanium dioxide (figure 7.3), PEOPPO-PEO (figure 7.4), and the sol-gel germania tri-block polymer material (figure 7.5). The appearance of bands at 668 cm-1 and 1027 cm-1 in the spectrum for the sol-gel germania tri-block polymer material (f igure 7.5) indicates that germanium-oxygencarbon bonds have formed [54,55], and provides evidence that the tri-block polymer has chemically bonded into the sol-gel germania network.

PAGE 342

311 Figure 7.3 FTIR spectrum of germanium dioxide.

PAGE 343

312 Figure 7.4 FTIR spectrum of poly(ethylene oxide )-block-poly(propyle ne oxide)-blockpoly(ethylene oxide).

PAGE 344

313 Figure 7.5 FTIR spectrum of the sol-gel germ ania tri-block polymer coating.

PAGE 345

314 The germania-tri-block polymer coated capillary proved to have excellent extraction capabilities for a variety of non-polar (PAHs), mo derately polar (ketones) and polar (amines, alcohols, and phenols) anal ytes, many of which may be carcinogenic, mutagenic, toxic and/or teratogenic [ 11,18,56,57] environmental contaminants. The observed extraction characteristic s can be attributed to the structure of the tri-block polymer, which is polar, making it suitable for the extraction of polar analytes. Since the tri-block polymer also contains methylene a nd methyl groups, it is also suitable for the extraction of moderately polar and nonpolar analytes. The peak area RSD values and detection limit data for the compounds extrac ted using the sol-gel germania tri-block PEO-PPO-PEO polymer coated capillaries is presented in table 7.1. The sol-gel germania tri-block polymer coated capillaries demonstrated detecti on limits ranging from 1.0 x 101 pM to 9.7 x 101 pM for PAHs. For ketones, these sol-gel coated microextraction capillaries had detection limits between 8.8 x 101 and 1.6 x 102 pM. For the polar compounds, the detection limits ranged from 4.3 x 102 to 1.8 x 103 pM for amines, from 8.0 x 102 to 2.8 x 103 pM for alcohols, and from 2.4 x 102 to 1.4 x 103 pM for phenols. The sol-gel germania tri-block polymer coated microextraction capillaries provided efficient extraction of 2,4-dichlo rophenol and 2,4,6-trichlorophenol, both of which are classified by the EPA as major po llutants [57]. The observed differences in detection limits between analytes of the sa me chemical class can be attributed to differences in the chemical structures and UV absorption characteristics of the analytes. The germania-tri-block polymer capilla ry also demonstrated a good run-to-run repeatability, with all HPLC peak ar ea RSD values between 1.1 and 6.8%. A chromatogram depicting the ability of the sol-gel germania tri-block polymer coated

PAGE 346

315 microextraction capillaries to extract analytes from each chemical class is shown in figure 7.6.

PAGE 347

316 Table 7.1 HPLC peak area repeatability and detec tion limit data and distribution constant (Kcs) values for PAHs, ketones, amines, al cohols, and phenols in CME-HPLC using a sol-gel germania tri-block polymer coated microextraction capillary Chemical Class Chemical Name Peak Area Detection Kcs Repeatability Limit (n=3) (pM) RSD (%) (S/N = 3) PAHs fluorene 4.7 1.2 x 101 3.2 x 103 anthracene 1.1 2.1 x 101 1.7 x 104 fluoranthene 5.4 2.9 x 101 1.9 x 104 1,2-benzanthracene 3.6 9.7 x 101 3.7 x 103 phenanthrene 2.1 1.2 x 101 2.0 x 104 naphthalene 5.3 1.0 x 101 5.6 x 103 acenaphthene 4.5 7.8 x 101 2.7 x 103 Ketones coumarin 4.3 1.6 x 102 1.2 x 103 4’phenylacetophenone 6.8 1.4 x 102 1.3 x 103 trans -chalcone 2.7 8.8 x 101 2.4 x 103 Amines diphenylamine 2.7 1.5 x 102 1.5 x 103 o -toluidine 6.0 4.3 x 102 4.2 x 102 m -toluidine 3.9 1.8 x 103 1.5 x 102 N-methylaniline 2.0 1.5 x 103 2.6 x 102 Alcohols benzhydrol 5.0 8.0 x 102 2.0 x 102 resorcinol 2.6 2.8 x 103 8.1 x 101 2-naphthol 4.0 2.6 x 103 1.5 x 102 9-anthracenemethanol 4.4 1.0 x 103 1.0 x 103 Phenols 2-chlorophenol 1.9 1.0 x 103 3.6 x 102 2,4,6-trichlorophenol 2.7 2.4 x 102 8.2 x 102 2,4-dichlorophenol 4.2 3.3 x 102 5.5 x 102 2,6-dimethylphenol 2.9 1.4 x 103 1.1 x 102 Extraction conditions: 40 cm x 0.25 mm I.D. so l-gel germania tri-block polymer coated capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, isocratic elution with ACN/H2O mobile phase, 1 ml/min flow rate, UV detection at 200 nm for ketones, amines, alc ohols, and phenols, 217 nm for naphthalene and acenaphthene, 254 nm for an thracene, phenanthrene, fluoranthene, fluorene, and 1,2-benzanthracene, ambient temperature.

PAGE 348

317 Figure 7.6 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravit y-fed extraction at room temperature. HPLC conditions : 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m -toluidine (1.40 x 103 nM), 2 = 2,4dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 349

318 A scanning electron microscopy (SEM) imag e of the sol-gel germania tri-block polymer coating was also obtained (figure 7.7). From this SEM image, the coating thickness was estimated. The coating thic kness (403.8 nm) was used, along with the length of the extraction capillar y (40 cm) to calculate the volume of the sol-gel germania tri-block polymer extr acting phase coating (Vc = 1.27 x 10-7 L). Using the distribution equation for SPME techniques [58-60 ], the distribution constant (Kcs) of analyte between the sol-gel germania tri-block polymer coati ng and sample was calculated for all analytes extracted. To our knowledge, this is the fi rst report on the determination of solute distribution constants for a sol-gel germ ania extracting phase coating. The SPME distribution equation is: 0C V V K V V K ns c cs s c cs Equation 7.1 n = the amount of extracted analyte (moles) Kcs = the distribution constant of analyte be tween the sol-gel coating and the sample Vc = the volume of the so l-gel extracting phase Vs = the volume of the sample C0 = the original molar concentratio n of analyte in the sample For Vs > > KcsVc, n = KcsVcC Equation 7.2 The amount of extracted analyt e was determined by comparing the peak area obtained for the extracted analyte with the peak area obt ained for a known number of moles of the same analyte (n). The C0 (original molar concentration of analyte in the sample), n, and Vc values were substituted into the equation to calculate the Kcs for each analyte, also

PAGE 350

319 presented in table 7.1. Determining the Kcs is useful in compari ng extraction abilities of the sol-gel germania tri-block polymer coating for different analytes since Kcs values directly reflect extraction capabilities of the coating [58-60], unlike detection limits, which also vary according to UV absorption characteristics of the analytes. Kcs values are also useful for determining original c oncentrations of targ et analytes [58-60].

PAGE 351

320 Figure 7.7 Scanning electron microscopy image of the sol-gel germania tri-block polymer coated microextraction capillary, magnification: 50,000 X.

PAGE 352

321 Experiments were conducted to verify th e pH stability of sol-gel germania-based tri-block polymer (PEO-PPO-PEO) coatings. Fi rst, a sample containing analytes from 5 chemical classes was extracted on a sol-gel ge rmania tri-block polymer coated capillary. To test the acid stability of the sol-gel germania tri-bloc k polymer coating, the capillary was connected to the bottom of a gravity-fed sample dispenser. The sample dispenser was filled with 1.0 M HCl, which was allowed to drip through the sol-gel coated capillary for 20 h. The capillary was then rinsed with deionized water, and the same sample of analytes was extracted again using this acid-tr eated sol-gel coated capillary. The sol-gel germania tri-block polymer coated capillary survived the acid e xposure with a slight increase in the extracted amount, as is refl ected by the increase in HPLC peak area. These results are consistent with thos e observed with germania PDMS coated microextraction capillaries [33] This slight increase in extraction capabilities can be attributed to the acid cleani ng the extracting phase coating. To demonstrate that the solgel germania tri-block polymer coating would remain stable after long-term exposure to low pH conditions, 1.0 M HCl was pumped th rough the capillary using an HPLC pump for an additional 100 h at a flow rate of 10 L /min. The capillary was again rinsed with deionized water, and the same sample of anal ytes was again extracte d using the capillary. The extraction performance remained pr actically unchanged compared to the performance observed after th e initial 20 h exposure to 1.0 M HCl. Chromatograms depict the extraction of this sa mple of analytes by the solgel germania tri-block polymer coated capillary after 20 h of exposure to 1.0 M HCl (figure 7.8), and after 5 days of exposure to 1.0 M HCl (figure 7.9). Extraction comparisons before and after exposure to 1.0 M HCl for 20 h and 5 days are given in table 7.2.

PAGE 353

322 Figure 7.8 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary after 20 h of exposure to 1.0 M HCl. Ex traction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m-toluidine (1.40 x 103 nM), 2 = 2,4-dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans-chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 354

323 Figure 7.9 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary after 5 days of exposure to 1.0 M HCl. Ex traction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m-toluidine (1.40 x 103 nM), 2 = 2,4-dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans-chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 355

324 Table 7.2 Extraction comparison of PAHs, ketones, amines, alc ohols, and phenols before and after exposing the sol-gel germania tri-block PEO-PPO-PEO coated mi croextraction capillar y to 1.0 M HCl (pH 0.0) and 1.0 M NaOH (pH 14.0) for 20 h and 5 Days. Chemical Chemical Before HCl After HCl exposure: Before NaOH After NaOH exposure: class name exposure (20 h) (5 days) exposure (20 h) (5 days) ng ng % ng % ng ng % ng % change change* change change* PAH phenanthrene 59.0 62.6 6.1 61.5 1.6 57.9 61.0 5.3 62.6 2. 5 Ketone trans-chalcone 21.8 23.2 6.5 23.6 3.1 21.4 22.1 3.3 23.2 4.7 Amine m-toluidine 9.6 10.3 6.8 9.9 3.8 9.5 9.9 4.2 10.3 4.0 Alcohol 9-anthracenemethanol 102.3 114.7 12.1 119.4 4.1 103.9 113.2 9.0 108.5 4.1 Phenol 2,4-dichlorophenol 7.7 8.3 8.1 8.1 3.0 7.7 8.4 9.7 8.2 2.9 Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravit y-fed extraction at room temp erature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature. Average of 3 re plicate measurements. *ng comp ared to ng after 20 h exposure to 1.0 M HCl or NaOH.

PAGE 356

325 This process was repeated to test the coating stability under high-pH environments. For this, a 1.0 M solution of NaOH (pH 14) was used. A second piece of the same sol-gel germania tri-block polym er coated capillary demonstrated a slight increase in extraction capabilities after bei ng exposed to base for 20 h. After extended exposure to base (120 h), the extraction perf ormance of the sol-gel germania tri-block polymer coating remained similar to the pe rformance observed afte r initial exposure to base for 20 h. Chromatograms depict the extrac tion of this sample of analytes by the solgel germania tri-block polymer coated cap illary after 20 h of exposure to 1.0 M NaOH (figure 7.10), and after 5 days of exposur e to 1.0 M NaOH (figure 7.11). Extraction comparisons before and after exposure to 1.0 M NaOH for 20 h and 5 days are also given in table 7.2. The pH stability of the sol-gel germania tri-block polymer coated capillaries is comparable or superior to that of previously reported titania [12,16] based sol-gel coated microextraction capillaries. Th erefore, sol-gel germania triblock polymer coatings have potential for applications in areas where pH stability is a pr erequisite (e.g., ion chromatography [35], hydrophobic interaction chromatography [35], proteomics [37-41], and in CME-HPLC with electr ochemical detection [42-50]).

PAGE 357

326 Figure 7.10 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary after 20 h of exposure to 1.0 M NaOH. Extr action conditions: 40 cm x 0.25 mm I.D. capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m-toluidine (1.40 x 103 nM), 2 = 2,4-dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans-chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 358

327 Figure 7.11 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary after 5 days of exposure to 1.0 M NaOH. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min grav ity-fed extraction at room temper ature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m-toluidine (1.40 x 103 nM), 2 = 2,4-dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans-chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 359

328 In addition to pH resistance, the so l-gel germania tri-block PEO-PPO-PEO polymer coated capillaries were also eval uated for stability u nder high temperature solvent conditions, which are us ed in high-temperature HPLC applications. The use of high-temperature HPLC decreases the viscosity while increasing the linear velocity of ACN/H2O mobile phases, which improves peak shape while shortening run time in HPLC [61-64]. In the future, thermal gradie nts may replace solvent gradients in HPLC [64]. In high-temperature HPLC, the mobile phase is typically preheated in an oven to 120 C [61-64]. Therefore, for a sol-gel coa ting to be successfully on-line hyphenated to a high-temperature HPLC system, it must be capable of withstanding high temperature mobile phase conditions [64]. The high-temp erature solvent stability of the sol-gel germania tri-block polymer coating was ev aluated for possible future use in hightemperature HPLC. For this, a sol-ge l germania tri-block polymer coated microextraction capillary was placed inside an oven heated to 200 C as a mobile phase consisting of ACN/H2O (50/50, v/v), was pumped through it for 2 h at a flow rate of 0.1 mL/min. The extraction performance was evalua ted before and after this treatment. The chromatogram in figure 7.12 depicts the extractio n of a mixture of an alytes after exposure to high temperature solvent conditions. Li ke the acid and base exposure, the high temperature solvent exposure also appeared to clean the inner surface of the sol-gel germania tri-block polymer extracting phase, since slightly better extraction performance was noted after the high temperature solvent expo sure. This is consistent with the results obtained in our previous report [15]. An extraction comparision before and after exposing the sol-gel germania tri-block polymer coated microextraction capillary to high temperature solvent conditions is also given in table 7.3.

PAGE 360

329 Table 7.3 Extraction comparison of PAHs, ketones, amines, alcohols, and phenols before and after exposing the sol-gel germania triblock PEO-PPO-PEO coated microextraction capillary to ACN/H20 (50/ 50, v/v) for 2 h at 200C. Chemical Chemical Before After 200C class name e xposure solvent exposure ng ng % Change PAH phenanthrene 59.4 63.0 6.0 Ketone trans-chalcone 22.1 22.8 3.2 Amine m-toluidine 9.9 10.4 5.3 Alcohol 9-anthracenemethanol 102.4 110.1 7.6 Phenol 2,4-dichlorophenol 7.6 8.2 8.2 Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravit y-fed extraction at room temperature. HPLC conditi ons: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 mL/min flow rate, UV detection at 200 nm, ambient temperature, average of 3 replicate measurements.

PAGE 361

330 Figure 7.12 A chromatogram representing CME-HP LC-UV analysis of amines, phenols, alcohols, ketones and PAHs using a sol-gel ge rmania tri-block polymer coated capillary after 2 h exposure to ACN/H20 (50/50, v/v) at 200 C. Extraction conditions: 40 cm x 0.25 mm I.D. capillary, 40 min gravity-fed extraction at room temperature. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column, gradient elution 50/50 ACN/H2O to 80/20 ACN/H2O in 10 min, 1 ml/min flow rate, UV detection at 200 nm, ambient temperature. 1 = m-toluidine (1.40 x 103 nM), 2 = 2,4-dichlorophenol (3.07 x 102 nM), 3 = 9-anthracenemethanol (9.60 x 102 nM), 4 = trans-chalcone (4.80 x 101 nM), 5 = phenanthrene (5.61 x 101 nM).

PAGE 362

331 To evaluate the reproducibility of the sol-gel coating method, capillary-tocapillary reproducibility studi es were also conducted on so l-gel germania tri-block polymer coated microextraction capillaries. For this, five analytes, each representing a different chemical class, were extracted on si x separately prepared sol-gel germania triblock polymer coated capillaries. It was f ound that the preparation method for the sol-gel germania tri-block polymer coated capillaries is quite reliable, with capillary-to-capillary HPLC peak area reproducibility ranging from 5.3 to 6.5 %. The capillary-to-capillary reproducibility data for analytes from each chemical class is presented in table 7.4.

PAGE 363

332 Table 7.4 Capillary to capillary peak area repro ducibility in CME-HPLC for the sol-gel germania tri-block polymer coated capillaries. Chemical class Chemical name Capillary to capillary peak area RSD in CME-HPLC analysis (%) (n = 6) PAH phenanthrene 5.6 Ketone trans-chalcone 5.3 Amine m-toluidine 6.5 Alcohol 9-anthracenemethanol 6.3 Phenol 2,4-dichlorophenol 5.3 Extraction conditions: 40 cm x 0.25 mm I.D. so l-gel germania tri-block polymer coated capillaries; 40 min gravity-fed extraction. HPLC conditions: 15 cm x 4.6 mm I.D. Luna C18 column; gradient elution 50/50 ACN/water to 80/20 ACN/water in 10 min; 1 mL/min flow rate, UV detection at 200 nm for all.

PAGE 364

333 To determine the time required for different analytes to reach equilibrium with the sol-gel germania tri-block polymer coati ng, five analytes, each representative of a particular chemical class (amine, phenol, alc ohol, ketone, and PAH), were extracted three times each using a gravity-fed sample deliver y system (figure 7.1) for 10 min, 20 min, 30 min, 40 min, 50 min, and 60 min. The repl icate amounts, expressed in ng, were subsequently averaged and plotted against the corresponding extrac tion time. The point on the graph in which the amount extracted ceas es to increase with respect to increasing extraction time corresponds to the point at wh ich equilibrium is es tablished between the sample solution and the sol-gel germania triblock polymer coating. From the graph in figure 7.13, it is clear that m-toluidine and 3,5-dimethylphe nol required 30 min of extraction, while 9-anthracenemethanol, trans-chalcone, and phenanthrene required 40 min of extraction to reach equilibrium.

PAGE 365

334 Figure 7.13 Extraction profiles of m-toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans-chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for gravity-fed extraction (~0.2 mL/m in flow rate).

PAGE 366

335 The effect of sample flow rates on extraction equilibrium was also investigated in this study. To accomplish this, an HPLC pu mp was connected to the thoroughly cleaned waste line of the six-port HPLC injection valve (figure 7.1). Two sets of extraction experiments were performed by pumping the sample solution through the sol-gel germania tri-block polymer capillary at two steady flow rates: 1.0 mL/min and 2.5 mL/min. An HPLC pump was employed since it was capable of maintaining a constant, reproducible flow rate. The same representative analytes were extract ed in triplicates at 1.0 mL/min for 5 min, 10 min, 20 min, 30 min, and 40 min and at 2.5 mL/min for 5 min, 10 min, 15 min, and 20 min using an HPLC pum p. The replicate amounts extracted (ng) for each time period were averaged and plotte d against their respect ive extraction times. Using higher flow rates drama tically reduced the time required for the analyte extraction equilibrium to be established between the sa mple solution and the sol-gel germania triblock polymer coating. This is consiste nt with the equation proposed by Eisert and Pawliszyn for in-tube SPME, where it was determined that the time required for equilibrium to be established for the anal ytes between the sample solution and the extracting phase is inversely pr oportional to the flow rate of the sample solution [4]. Equation 7.3 te = extraction time L = length of the capillary holding the extracting phase k = retention factor u = laminar flow rate of the fluid

PAGE 367

336 From the graph in figure 7.14, m-toluidine, 3,5-dimethylphenol, 9anthracenemethanol, and trans-chalcone required approximately 20 min of extraction, while phenanthrene required 30 min of extractio n for equilibrium to be established using a flow rate of 1.0 mL/min. From the graph in figure 7.15, m-toluidine, 3,5dimethylphenol, 9-anthracenemethanol, and phe nanthrene required only 10 min of extraction, while trans-chalcone required between 10 and 15 min of extraction for equilibrium to be established using an extr action flow rate of 2.5 mL/min. The peak areas at equilibrium are very similar for both gravity-fed (figure 7.13) and HPLC pump driven (figure 7.14, figure 7.15) extraction. The use of a higher extrac tion flow rate can dramatically reduce the time required to reac h extraction equilibrium in sol-gel capillary microextraction.

PAGE 368

337 Figure 7.14 Extraction profiles of m-toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans-chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for HPLC pump driv en extraction (1.0 mL/min flow rate).

PAGE 369

338 Figure 7.15 Extraction profiles of m-toluidine, 3,5-dimethylphenol, 9anthracenemethanol, trans-chalcone, and phenanthrene for the sol-gel germania tri-block polymer coated capillary for HPLC pump driv en extraction (2.5 mL/min flow rate).

PAGE 370

339 7.4 Conclusion For the first time, sol-gel germania tr i-block polymer coated capillaries were developed for on-line capillary microextra ction in hyphenation with high-performance liquid chromatography. These capillaries pr ovided simultaneous extraction of polar, moderately polar, and nonpolar an alytes from the same sample. The sol-gel germania triblock polymer coatings achieved detection limits ranging from 1.0 x 101 to 2.8 x 103 pM for environmentally and biomedically sign ificant compounds of varied polarities, including polycyclic aromatic hydrocarbons, ke tones, amines, alcohols, and phenols. The sol-gel germania tri-block polymer coated capillaries also dem onstrated good run-to-run and capillary-to-capillary reproduc ibility. Also for the first time, the analyte distribution constants (Kcs) between the aqueous sample matrix and a sol-gel germania-based coating were determined. Most notabl y, the sol-gel germania tri-bloc k polymer coated capillaries provided excellent stability under extreme pH conditions, surviving long term exposure (5 days) to highly acidic (pH 0.0) and basic (pH 14.0) conditions. These sol-gel germania tri-block polymer coated capillaries are thus suitable for use in applications utilizing mobile phases or samples with ex treme pHs. Sol-gel germania tri-block polymer coated capillaries are also able to withstand exposure to high temperature solvents (200C), making them suitable for possible future use in high-temperature HPLC. The use of higher extraction flow ra tes can reduce the amount of time required for analyte extraction equilibrium to be established between the sample solution and the sol-gel germania tri-block polymer coating.

PAGE 371

340 7.5 References for chapter seven [1] S. L. Chong, D. Wang, J. D. Hayes, B. W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [2] C. L. Arthur, J. Pawl iszyn, Anal. Chem. 62 (1990) 2145. [3] K. Jinno, M. Taniguchi, M. Hayashida, J. Pharm. Biomed. Anal. 17 (1998) 1081. [4] R. Eisert, J. Pawliszyn, Anal. Chem. 69 (1997) 3140. [5] H. Kataoka, Anal. Bioanal. Chem. 373 (2002), 31. [6] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [7] M. Liu, Z. Zeng, B. Xiong, J. Chromatogr. A 1065 (2005) 287. [8] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [9] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [10] X. Li, Z. Zeng, S. Gao, H. Li, J. Chromatogr. A 1023 (2004) 15. [11] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [12] T. Y. Kim, K. Alhoosha ni, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A 1047 (2004) 165. [13] Y. Fan, Y. Q. Feng, S. L. Da, Z. H. Wang, Talanta 65 (2005), 111. [14] S. S. Segro, A. Malik, J. Chromatogr. A 1200 (2008) 62. [15] S. S. Segro, A. Malik, J. Chromatogr. A 1205 (2008) 26. [16] S. S. Segro, Y. Cabezas, A. Malik, J. Chromatogr. A 1216 (2009) 4329. [17] S. Segro, A. Malik, J. Chromatogr. A. 1216 (2009) 7677. [18] S. Kulkarni, A. M. Shearrow, A. Malik, J. Chromatogr. A 1174 (2007) 50.

PAGE 372

341 [19] A. Kabir, C. Hamlet, K. Soo Yoo, G. R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [20] S. Kulkarni, L. Fang, K. Alhooshani, A. Malik, J. Chromatogr. A 1124 (2006) 205. [21] K. Alhooshani, T. Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A 1062 (2005) 1. [22] A. M. Shearrow, G. A. Harris, L. Fang, P. K. Sekhar, L. T. Nguyen, E. B. Turner, S. Bhansali, A. Malik, J. Chromatogr. A 1216 (2009) 5449. [23] A. M. Shearrow, S. Bhansali, A. Malik, J. Chromatogr. A 1216 (2009) 6349. [24] W. Li, D. Fries, A. Malik, J. Sep. Sci. 28 (2005) 2153. [25] W. Li, D. Fries, A. Alli, A. Malik, Anal. Chem. 76 (2004) 218. [26] F. Zeng, B. Hu, Talanta 73 (2007) 372. [27] Y. Wu, B. Hu, Z. Jiang, Y. Feng, P. Lu, B. Li, Rapid Commun. Mass Spectrom. 20 (2006) 3527. [28] A. Wehrli, J. C. Hildenbrand, H. P. Kelle r, R. Stampfli, R. W. Frei, J. Chromatogr. 149 (1978) 199. [29] J. L. Glajch, J. J. Kirkland, J. Koehler, J. Chromatogr. 384 (1987) 81. [30] M. Liu, Y. Liu, Z. Zeng, T. Peng, J. Chromatogr. A 1108 (2006) 149. [31] X. Li, J. Gao, Z. Zeng, Anal. Chim. Acta 590 (2007) 26. [32] L. M. Wei, Q. Y. Ou, J. B. Li, Chin. Chem. Letters 15 (2004) 1127. [33] L. Fang, S. Kulkarni, K. Alhoosha ni, A. Malik, Anal. Chem. 79 (2007) 9441. [34] A. V. Ruzette, Na ture Mat. 1 (2002) 85. [35] Katz, E.; Eksteen, R.; Schoenmakers, P.; Miller, N. Handbook of HPLC, Marcel Dekker, Inc., New York, 1998.

PAGE 373

342 [36] D. B. Wall, M. T. Kachman, S. Gong, R. Hinderer, S. Parus, D. E. Misek, S. M. Hanash, D. M. Lubman, Anal. Chem. 72 (2000) 1099. [37] G. Mitulovic, K. Me chtler, Brief. Funct. Genom. Proteom. 5 (2006) 249. [38] S. Gedela, N. R. Medicher la, Chromatographia 65 (2007) 511. [39] H. Cui, J. Leon, E. Reusaet, A. Bult, J. Chromatogr. A 704 (1995) 27. [40] L. Zeng, C. E. Nath, P. J. Shaw, J. W. Earl, A. J. McLachlan, Biomed. Chromatogr. 22 (2008) 879. [41] C. K. Larive, S. M. Lunte, M. Zhong, M. D. Perkins, G. S. Wilson, G. Gokulrangan, T. Williams, F. Afroz, C. Schoeneich, T. S. Derrick, C. R. Middaugh, S. BogdanowichKnipp, Anal. Chem. 71 (1999) 389R. [42] M. J. Legorburu, R. M. Alonso, R. M. Ji menez, E. Ortiz, J. Chromatogr. Sci. 39 (2001) 425. [43] Y. T. Wu, T. R. Tsai, L. C. Lin, T. H. Tsai, J. Chromatogr. B 853 (2007) 281. [44] A. Mishra, K. P. Joy, Gen. Comp. Endocrin. 145 (2006) 84. [45] S. Moane, S. Park, C. E. Lunte, M. R. Smyth, Analyst 123 (1998) 1931. [46] C. S. Lai, N. K. Nair, A. Muniandy, S. M. Mansor, P. L. Olliaro, V. Navaratnam, J. Chromatogr. B 877 (2009) 558. [47] I. G. Casella, M. Gatta, J. Agric. Food Chem. 50 (2002) 23. [48] H. Iwase, J. Chromatogr. A 881 (2000) 327. [49] X. Cai, M. H. Woo, M. J. Edick, M. V. Relling, J. Chromatogr. B 728 (1999) 241. [50] R. Wintersteige r, M. Macher, Chromatographia 40 (1995) 247. [51] J. D. Hayes, A. Malik, J. Chromatogr. B. 695 (1997) 3.

PAGE 374

343 [52] C. J. Brinker, G. W. Scherer, Sol-Ge l Science: The Chemistry and Physics of SolGel Processing, Academic Press, San Diego, CA, 1990. [53] J. Livage, M. Henry, C. Sanchez, Prog. Solid State Chem. 18 (1988) 259. [54] X. Lu, K. J. Ziegler, A. Ghezelbash, K. P. Johnston, B. A. Korgel, Nano Lett. 4 (2004) 969. [55] N. Inagaki, M. Mitsuuchi, J. Polymer Sci. 21 (1983) 2887. [56] K. Srogi, Environ. Chem. Lett. 5 (2007) 169. [57] National Institute for O ccupational Safety & Health (NIOSH). Pocket Guide to Chemical Hazards. Center for Disease Control. 1997. [58] R. Doong, S. Chang, Anal. Chem. 72 (2000) 3647. [59] Y. Hu, Y. Zheng, F. Zhu, G. Li, J. Chromatogr. A 1148 (2007) 16. [60] G. Jiang, M. Huang, Y. Cai, J. Jv, Z. Zhao, J. Chromatogr. Sci. 44 (2006) 324. [61] D. R. Stoll, P. W. Carr, J. Am. Chem. Soc. 127 (2005) 5034. [62] F. D. Antia, C. Horvat h, J. Chromatogr. 435 (1988) 1. [63] B. Yan, J. Zhao, J. S. Brown, J. Bl ackwell, P. W. Carr, Anal. Chem. 72 (2000) 1253. [64] J. Clark, Pharm. Technol. Europe 16 (2004) 41.

PAGE 375

344 APPENDICES

PAGE 376

345 Appendix A

PAGE 377

346 Appendix A (Continued)

PAGE 378

347 Appendix A (Continued)

PAGE 379

348 Appendix A (Continued)

PAGE 380

349 Appendix A (Continued)

PAGE 381

350 Appendix A (Continued)

PAGE 382

351 Appendix A (Continued)

PAGE 383

352 Appendix A (Continued)

PAGE 384

353 Appendix A (Continued)

PAGE 385

354 Appendix A (Continued)

PAGE 386

355 Appendix B

PAGE 387

356 Appendix B (Continued)

PAGE 388

357 Appendix B (Continued)

PAGE 389

358 Appendix B (Continued)

PAGE 390

359 Appendix B (Continued)

PAGE 391

360 Appendix B (Continued)

PAGE 392

361 Appendix B (Continued)

PAGE 393

362 Appendix B (Continued)

PAGE 394

363 Appendix B (Continued)

PAGE 395

364 Appendix B (Continued)

PAGE 396

365 Appendix C

PAGE 397

366 Appendix C (Continued)

PAGE 398

367 Appendix C (Continued)

PAGE 399

368 Appendix C (Continued)

PAGE 400

369 Appendix C (Continued)

PAGE 401

370 Appendix C (Continued)

PAGE 402

371 Appendix C (Continued)

PAGE 403

372 Appendix C (Continued)

PAGE 404

373 Appendix C (Continued)

PAGE 405

374 Appendix D

PAGE 406

375 Appendix D (Continued)

PAGE 407

376 Appendix D (Continued)

PAGE 408

377 Appendix D (Continued)

PAGE 409

378 Appendix D (Continued)

PAGE 410

379 Appendix D (Continued)

PAGE 411

380 Appendix D (Continued)

PAGE 412

381 Appendix D (Continued)

PAGE 413

382 Appendix D (Continued)

PAGE 414

ABOUT THE AUTHOR Scott Segro was born in Wickliffe, Oh io. He obtained his Asso ciate in Arts degree from Manatee Community College in 2002. He c ontinued his studies at the University of South Florida and received his Bachelor of Arts degree in Chemistry with Biochemistry Emphasis in 2004. In 2005, he entered the chemis try graduate program at the University of South Florida to pursue a Doctorate Degree. The work presented in this dissertation was conducted under the guidance of Dr. A bdul Malik. Scott currently has five publications in international jour nals and one published book chapter.


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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003321
035
(OCoLC)
040
FHM
c FHM
049
FHMM
090
XX9999 (Online)
1 100
Segro, Scott.
0 245
Surface-bonded sol-gel sorbents for on-line hyphenation of capillary microextraction with high-performance liquid chromatography
h [electronic resource] /
by Scott Segro.
260
[Tampa, Fla] :
b University of South Florida,
2010.
500
Title from PDF of title page.
Document formatted into pages; contains X pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2010.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
3 520
ABSTRACT: High-performance liquid chromatography (HPLC) is the most widely used analysis technique. However, its sensitivity is limited. Sample preconcentration methods, such as fiber-based solid-phase microextraction (SPME) and in-tube SPME (capillary microextraction) offer improved detection limits. It is, however, difficult to couple fiber SPME on-line with HPLC due to the need for complicated desorption devices. Such coupling is further complicated due to the limited solvent stability of the extracting phase both in the fiber and in-tube formats of SPME. In this research, surface-bonded sol-gel sorbents were developed to provide the solvent stability required for effective on-line hyphenation of capillary microextraction (CME) with HPLC. These sol-gel sorbents were prepared using (1) silica-based, (2) titania-based, and (3) germania-based sol-gel precursors. Sol-gel reactions were performed within fused silica capillaries to create a number of organic-inorganic hybrid sorbents in the form of surface-bonded coatings: (1) alkyl (methyl, octyl, octadecyl), (2) polydimethyldiphenylsiloxane, (3) titania poly(tetrahydrofuran), and (4) germania tri-block polymer. The sol-gel coated microextraction capillaries were capable of efficiently extracting a wide variety of analytes, including polycyclic aromatic hydrocarbons, ketones, aldehydes, aromatic compounds, amines, alcohols, and phenols with ng/L to pg/L detection limits. The sol-gel methyl coating demonstrated a counterintuitive ability to extract polar analytes. Sol-gel polydimethyldiphenylsiloxane coatings were found to be resistant to high temperature solvent exposure (150C and 200C), making them suitable for use in high-temperature liquid phase separations. To better understand how extraction takes place, effects of alkyl chain length and sol-gel precursor concentration were evaluated in the study on sol-gel alkyl coatings. The sol-gel titania poly(tetrahydrofuran) coating was also capable of extracting underivatized aromatic acids and polypeptides at pHs near their respective isolectric points. The sol-gel titania poly(tetrahydrofuran) coatings and the sol-gel germania tri-block polymer coatings demonstrated impressive resistance to extreme pH conditions, surviving prolonged exposure to 1.0 M HCl (pH ≈ 0.0) and 1.0 M NaOH (pH ≈ 14.0) with virtually no change in extraction behavior. Sol-gel germania tri-block polymer coatings were also stable under high temperature solvent conditions (200C). In addition, for the first time, the analyte distribution constants between a sol-gel germania coating and the aqueous samples (Kcs) were determined.
590
Advisor: Abdul Malik, Ph.D.
653
Sol-Gel
Silica
Titania
Germania
CME
SPME
In-Tube SPME
High-Performance Liquid Chromatography
Solvent Resistance
PH Stability
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.3321