USF Libraries
USF Digital Collections

Polytetrahydrofuran- and dendrimer- based novel sol-gel coatings for capillary microextraction (cme) providing parts per...

MISSING IMAGE

Material Information

Title:
Polytetrahydrofuran- and dendrimer- based novel sol-gel coatings for capillary microextraction (cme) providing parts per trillion (ppt) and parts per quadrillion (ppq) level detection limits in conjunction with gas chromatography and flame ionization detection (fid)
Physical Description:
Book
Language:
English
Creator:
Kabir, Abuzar
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla.
Publication Date:

Subjects

Subjects / Keywords:
Spme
In-tube spme
Pahs
Aldehydes
Ketones
Phenols
Alcohols
Dissertations, Academic -- Chemistry -- Doctoral -- USF   ( lcsh )
Genre:
government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Summary:
ABSTRACT: Sol-gel capillary microextraction (CME) is a new direction in solvent-free extraction and preconcentration of trace analytes. CME presents significant interest in environmental, pharmaceutical, petrochemical, biomedical, agricultural, food, flavor, and a host of other important areas. Sol-gel CME utilizes advanced material properties of organic-inorganic hybrid sol-gel polymers to perform efficient extraction and enrichment of target analytes from a variety of matrices. In this dissertation, two novel sol-gel coatings were developed for CME: (a) sol-gel benzyl-terminated dendrimer coating, and (b) sol-gel polytetrahydrofuran (poly-THF) coating. A detailed investigation was conducted to evaluate the performance of the newly developed sol-gel coatings in solvent-free extraction of a wide range of polar and nonpolar analytes.Sol-gel chemistry was used to chemically immobilize dendrimer- and poly-THF-based hybrid organic-inorganic coatings on fused silica capillary inner surface. The sol-gel coatings were created using a coating solution containing a sol-gel active organic component (dendrimer or poly-THF), a sol-gel precursor (methyltrimethoxysilane, MTMOS), a sol-gel catalyst (trifluoroacetic acid, TFA, 5% water) and a deactivating reagent (hexamethyldisilazane, HMDS). Sol-gel reactions were conducted inside a hydrothermally treated fused silica capillary for 60 min. A wall-bonded sol-gel coating was formed via condensation of silanol groups residing on the capillary inner surface with those on the sol-gel network fragments evolving in close vicinity of the capillary walls. Due to the strong chemical bonding with capillary inner walls, these sol-gel coatings showed excellent thermal and solvent stability in CME in hyphenation with gas chromatography (GC).
Thesis:
Thesis (Ph.D.)--University of South Florida, 2005.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: World Wide Web browser and PDF reader.
System Details:
Mode of access: World Wide Web.
Statement of Responsibility:
by Abuzar Kabir.
General Note:
Includes vita.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 450 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001670331
oclc - 62277684
usfldc doi - E14-SFE0001192
usfldc handle - e14.1192
System ID:
SFS0025513:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001670331
003 fts
005 20051216093254.0
006 m||||e|||d||||||||
007 cr mnu|||uuuuu
008 051115s2005 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001192
035
(OCoLC)62277684
SFE0001192
040
FHM
c FHM
049
FHMM
090
QD31.2 (Online)
1 100
Kabir, Abuzar.
0 245
Polytetrahydrofuran- and dendrimer- based novel sol-gel coatings for capillary microextraction (cme) providing parts per trillion (ppt) and parts per quadrillion (ppq) level detection limits in conjunction with gas chromatography and flame ionization detection (fid)
h [electronic resource] /
by Abuzar Kabir.
260
[Tampa, Fla.] :
b University of South Florida,
2005.
502
Thesis (Ph.D.)--University of South Florida, 2005.
500
Includes vita.
504
Includes bibliographical references.
516
Text (Electronic thesis) in PDF format.
538
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 450 pages.
520
ABSTRACT: Sol-gel capillary microextraction (CME) is a new direction in solvent-free extraction and preconcentration of trace analytes. CME presents significant interest in environmental, pharmaceutical, petrochemical, biomedical, agricultural, food, flavor, and a host of other important areas. Sol-gel CME utilizes advanced material properties of organic-inorganic hybrid sol-gel polymers to perform efficient extraction and enrichment of target analytes from a variety of matrices. In this dissertation, two novel sol-gel coatings were developed for CME: (a) sol-gel benzyl-terminated dendrimer coating, and (b) sol-gel polytetrahydrofuran (poly-THF) coating. A detailed investigation was conducted to evaluate the performance of the newly developed sol-gel coatings in solvent-free extraction of a wide range of polar and nonpolar analytes.Sol-gel chemistry was used to chemically immobilize dendrimer- and poly-THF-based hybrid organic-inorganic coatings on fused silica capillary inner surface. The sol-gel coatings were created using a coating solution containing a sol-gel active organic component (dendrimer or poly-THF), a sol-gel precursor (methyltrimethoxysilane, MTMOS), a sol-gel catalyst (trifluoroacetic acid, TFA, 5% water) and a deactivating reagent (hexamethyldisilazane, HMDS). Sol-gel reactions were conducted inside a hydrothermally treated fused silica capillary for 60 min. A wall-bonded sol-gel coating was formed via condensation of silanol groups residing on the capillary inner surface with those on the sol-gel network fragments evolving in close vicinity of the capillary walls. Due to the strong chemical bonding with capillary inner walls, these sol-gel coatings showed excellent thermal and solvent stability in CME in hyphenation with gas chromatography (GC).
590
Adviser: Abdul Malik, PH.D.
653
Spme.
In-tube spme.
Pahs.
Aldehydes.
Ketones.
Phenols.
Alcohols.
690
Dissertations, Academic
z USF
x Chemistry
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.1192



PAGE 1

Polytetrahydrofuran-and Dendrimer-Based Novel Sol-Gel Coatings for Capillary Microextraction (CME) Providing Parts Per Trillion (ppt) and Parts Per Quadrillion (ppq) Level Detection Limits in C onjunction With Gas Chromatography and Flame Ionization Detection (FID ) by Abuzar Kabir 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. Dean F. Martin, Ph.D. Robert L. Potter, Ph.D. Date of Approval: April 29, 2005 Keywords: SPME, In-tube SPME, PAHs, Al dehydes, Ketones, Phenols, Alcohols Copyright 2005, Abuzar Kabir

PAGE 2

Dedication To my deceased parents, Moqbulur Rahman and Suraiya Rahman, who brought me in this beautiful world and provided their endless inspiration, love and support.

PAGE 3

Acknowledgements I would like to express my sincere acknowledgment a nd earnest appreciation to several people whose contri bution, guidance, and encouragement have made this dissertation possible. First, I would like to th ank my major professor, Dr. Abdul Malik for his able supervision, instruction, patience, s upport and encouragement. I would also like to express my appreciation and gratitude to my Ph.D. committee me mbers: Dr. Milton D. Johnston, Jr., Dr. Dean F. Martin, and Dr. Robert Potter for thei r tireless support and encouragement from the very begi nning of my research endeavor. In addition, I would like to thank Ms. Betty Loraamm of the USF Biology Department for her kind assist ance in acquiring all scanning electron microscopic images throughout this study. I would like to express my sincere appreciation to fellow graduate students C. Shende, T.-Y. Kim, W. Li, K. Alhooshani, Li Fang, Sameer Kulkarni, Anne Marrie Shearrow, and Erica Turner whos e continuous assistance, support, and encouragement helped me stay focused. I wholeheartedly acknowledge USF Department of Chemistry for financial suppor t throughout my graduate career. Last, and of utmost significance, great a ppreciation is extended to the members of my family, in particular, my wife, Salima Kabir, daughter, Nafisa Kabir, and elder brother, Khaled Saifullah. This degree, as well as other accomplishments in my life, would have never been possible without un limited patience, extensive sacrifices and tireless dedication from their side.

PAGE 4

Table of Contents List of Tables vii List of Figures xii List of Schemes xvii List of Abbreviations, Symbols, and Acronyms xviii Abstract xx Chapter One : Solid Phase Microextraction-A Solvent-Free Sample Preparation Technique 1 1.1 Introduction 1 1.2 Evolution of solid phase microextraction: a historical 2 perspective 1.3 Working principles of SPME 8 1.4 Modes of extraction 10 1.4.1 Extraction modes with a coated fiber 10 1.4.2 Extraction modes with in-tube SPME 13 1.5 Preparation of coating on fibers 14 1.6 Experimental parameters affecting extraction efficiency 14 1.6.1 pH adjustment of the matrix 15 1.6.2 Agitation of the matrix 15 1.6.3 Heating the matrix 16 1.6.4 Addition of salt to the matrix 19 1.6.5 Addition of organic solvents 19 1.7 Derivatization 20 1.8 References for Chapter One 23 Chapter Two : An Overview on Stationary Phases Used in Solid Phase Microextraction (SPME) 26 2.1 Introduction 26 2.2 Commercially available sorbents for fiber SPME 32 2.2.1 Homogeneous polymeric sorbents 33 2.2.2 Polymeric composite sorbents 34 2.2.2.1 Polydimethylsiloxane/DVB (PDMS/DVB) 34 2.2.2.2 Carboxen/Polydimethylsiloxane 35 (CAR/PDMS) 2.2.2.3 Carbowax/Divinylbenzene (CW/DVB) 35 2.2.2.4 Carbowax/Templated resin (CW/TPR) 36 i

PAGE 5

ii 2.2.2.5 Divinylbenzene/Carboxen /Polydimethylsiloxane(DVB/ CAR/PDMS) 36 2.3 Commercial GC coatings used in in-tube SPME 37 2.4 Tailor made coatings on SPME fibers 42 2.4.1 Immobilized antibodies 44 2.4.2 Metallic SPME fibers 47 2.4.3 Active carbonaceous sorbents 51 2.4.4 Bonded silica sorbents 54 2.4.5 Flat sheet membranes 57 2.4.6 Miscellaneous sorbents 59 2.5 Tailor made coatings for in-tube SPME 61 2.5.1 Tailor made coatings used predominantly in in-tube SPME 61 2.5.1.1 Molecularly imprinted polymers 63 2.5.1.2 Restricted access materials (RAM) 66 2.5.1.3 Conductive polymers 69 2.5.2 Coatings exclusiv ely used in in-tube SPME 77 2.6 References for Chapter Two 84 Chapter Three : Sol-Gel Technology in Capillary Microextraction 91 3.1 A brief history 91 3.2 Sol-gel technology in SPME 93 3.3 Sol-gel sorbents in SPME: a brief overview 95 3.3.1 Polysiloxane based sol-gel sorbents 97 3.3.1.1 Sol-gel sorbents with homogeneous polysiloxane phases 97 3.3.1.2 Mixed polysiloxane based sol-gel sorbents 106 3.3.1.2.1 Fullerene-polysiloxane mixed sol-gel sorbents 106 3.3.1.2.2 Mixed crown ether-polysiloxane sol-gel sorbents 110 3.3.1.2.3 Mixed calix[4]arene-polysiloxane sol-gel sorbents 132 3.3.1.2.4 Mixed polyvinyl alcohol (PVA) polysiloxane sol-gel sorbents 139 3.3.1.2.5 Mixed polymethylphenylvinylsiloxane (PMPVS) polysiloxane sol-gel sorbents 140 3.3.1.2.6 Mixed divinylbenzene-polysiloxane sol-gel sorbents 141 3.3.1.2.7 Mixed polyphenylmethylsiloxane (PPMS) polysiloxane sol-gel sorbents 144

PAGE 6

iii 3.3.2 Non-polysiloxane based sorbents 149 3.3.2.1 Polyethylene glycol-based sol-gel sorbents 151 3.3.2.2 Non-polysiloxane so l-gel sorbents with alkyl ligands 159 3.3.3 Cyclodextrin-based non-pol ysiloxane sol-gel sorbents 163 3.4 Miscellaneous sorbents 166 3.5 References for Chapter Three 168 Chapter Four : Capillary Microextraction on Sol-Gel Benzyl Terminated Dendrimer Coating 173 4.1 Introduction 173 4.2 Experimental 177 4.2.1 Equipment 177 4.2.2 Chemicals and materials 178 4.2.3 Preparation of sol-gel dendrimer coated extraction capillaries 179 4.2.3.1 Cleaning and hydrothermal treatment of the fused silica capillary 179 4.2.3.2 Preparation of the sol solution 182 4.2.3.3 Coating of the fused si lica capillary with sol solution 183 4.2.3.4 Thermal conditioning of the coated capillary 184 4.2.3.5 Rinsing the capillary with organic solvents to remove unbonded material 184 4.2.4 Preparation of sol-gel PDMS coated capillary GC column 185 4.2.5 Preparation of sol-gel PEG coated capillary GC column 187 4.2.6 Gravity-fed sample dispenser for capillary microextraction 188 4.2.7 Deactivation of glassware 190 4.2.8 Preparation of standard sample solution for sol-gel dendrimer CME 190 4.2.9 Extraction of analytes on sol-gel dendrimer coated capillaries 190 4.2.10 Transferring the extracted analytes to the GC column and Gas chromatographic analysis of the extracted analytes 191 4.2.11 Calculation of the limit of detection (LOD) for individual analyte 194 4.3 Results and discussion 194 4.3.1 Sol-gel dendrimer coati ng and chemical aspects of its preparation 195 4.3.2 Characterization of surface morphology and determination of coati ng thickness using scanning electron microscopy 203

PAGE 7

iv 4.3.3 Determination of ex traction kinetics for both polar and nonpolar analytes 206 4.3.4 Applications of sol-gel dendrimer coated microextraction capillaries 208 4.3.4.1 Polycyclic aromatic hydrocarbons (PAHs) 208 4.3.4.2 Aldehydes and ketones 215 4.3.4.3 Phenols 227 4.3.4.4 CME of butylatedhydroxytoluene (BHT) 233 4.3.4.5 Alcohols 234 4.4 Conclusion 244 4.5 References for Chapter Four 245 Chapter Five : Capillary Microextraction on Sol-Gel Polytetrahydrofuran 250 5.1 Introduction 250 5.2 Experimental 256 5.2.1 Equipments 256 5.2.2 Chemicals and materials 259 5.2.3 Preparation of so l-gel poly-THF coated Microextraction capillaries 259 5.2.3.1 Cleaning and hydrothermal treatment of the fused silica capillary 260 5.2.3.2 Preparation of the sol solution 262 5.2.3.3 Coating fused silica capillary with sol solution 264 5.2.3.4 Thermal conditioning of the coated capillary 264 5.2.3.5 Rinsing the capilla ry with organic solvents to remove unbonded materials 265 5.2.4 Preparation of sol-gel PDMS coated capillary GC columns 265 5.2.5 Preparation of sol-gel PE G coated capillary column 267 5.2.6 Cleaning and deactivation of glassware 268 5.2.7 Preparation of standard solutions for CME on sol-gel Poly-THF coated capillaries 269 5.2.8 Gravity-fed sample dispenser for capillary microextraction 270 5.2.9 Extraction of analytes on sol-gel poly-THF coated capillaries 270 5.2.10 Transferring the extrac ted analytes to the GC column and gas chromat ographic analysis of the extracted analytes 271 5.3 Results and Discussion 273 5.3.1 Sol-gel chemistry of the coating process 273 5.3.2 Characterization of sol-gel poly-THF sorbent 283

PAGE 8

v 5.3.2.1 Fourier-transform infrared spectroscopic characterization of the sol-gel poly-THF coating 283 5.3.2.2 Scanning electron microscopy 285 5.3.2.3 Thermogravimetric analysis 288 5.3.3 Determination of the extraction kinetics 291 5.3.4 Determination of the an alyte enrichment factors of sol-gel poly-THF sorbent 296 5.3.5 Determination of ther mal stability of the sol-gel poly-THF coating 300 5.3.6 Applications of sol-gel poly-THF coating 302 5.3.6.1 Polycyclic aromatic hydrocarbons 302 5.3.6.2 Aldehydes 309 5.3.6.3 Ketones 315 5.3.6.4 Chlorophenols 321 5.3.6.5 Alcohols 327 5.3.6.6 Mixture of polar, moderately polar, and nonpolar compounds 332 5.3.7 Possibility of automation 334 5.4 Conclusion 335 5.5 References for Chapter Five 337 Appendices 342 Appendix A: Capillary El ectrophoresis and Fluorescence Anisotropy for Quantitative Analysis of PeptideProtein Interactions Using JAK2 and SH2-B as a Model System 343 Appendix B: Sol-Gel Approach to In Situ Creation of High pHResistant Surface-Bonded Organic-Inorganic Hybrid Zirconia Coating for Ca pillary Microextraction (In-Tube SPME) 351 Appendix C: High pH-Resistant, Surface-Bonded Sol-Gel Titania Hybrid Organic-Inorganic Coating for Effective On-Line Hyphenation of Capillary Microextraction (In-Tube SPME) with High-Performance Liquid Chromatography 364 Appendix D: Parts Per Quadril lion Level Ultra-Trace Determination of Polar and Nonpolar Compounds via Solvent-free Capillary Microextraction on Surface-bonded Sol-Gel Polytetrahydrofuran Coating and Gas Chromatography-Flame Ionization Detection 375

PAGE 9

vi Appendix E: Capillary Microextraction on Sol-gel Dendrimer Coatings 388 Appendix F: Sol-Gel Poly (ethylene glycol) Stati onary Phase for High-Resolution Capillary Gas Chromatography 398 Appendix G: Sol-Gel Capillary Microextraction 411 Appendix H: Synthesis of Benz yl-Terminated Dendron for Use in High-Resolution Capillary Gas Chromatography 421 About the Author End Page

PAGE 10

vii List of Tables Table 2.1 Commercially available SPME sorbents 32 Table 2.2 Commercial GC colu mns used in in-tube SPME 38 Table 2.3 Tailor-made sorbents on SPM E fibers and their applications 42 Table 2.4 Tailor made sorbents used in in-tube SPME 62 Table 2.5 Coating used in electrochemically controlled SPME 70 Table 3.1 Homogeneous polysiloxane sorbents 98 Table 3.2 Important features of sol-gel hydroxyfullerene sorbent 107 Table 3.3 Sol-gel crown ether sorbents used in SPME 111 Table 3.4 Comparison of partit ion coefficients of fibers with different compositions 113 Table 3.5 Sol-gel calixarene sorbents used in SPME 133 Table 3.6 Salient features of sol-gel PVA, polyphenylmethylsiloxane, divinylbenzene, and polym ethylphenylvinylsiloxane SPME sorbents 148 Table 3.7 Non-polysiloxane based sorbent used in SPME 150 Table 4.1 Name, function and chemical structure of sol-gel dendrimer coating solution ingredients 183 Table 4.2 Chemical structures and pertin ent physical properties of polyaromatic hydrocarbons (PAHs) analyzed using sol-gel dendrimer coating 210 Table 4.3 Run-to-run peak area reproducib ility for PAHs in capillary microextraction using so l-gel dendrimer coatings 212

PAGE 11

viii Table 4.4 Capillary-to-capillary peak area reproducibility for PAHs in capillary mi croextraction using sol-gel dendrimer coatings 213 Table 4.5 Limit of detection (LOD) data for PAHs in CME-GC-FID 214 using sol-gel dendrimer microextraction capillaries Table 4.6 Chemical structures and pertin ent physical properties of 216 aldehydes analyzed using sol-gel dendrimer coating Table 4.7 Run-to-run peak area reproducib ility for aldehydes in 218 capillary microextraction us ing sol-gel dendrimer coatings Table 4.8 Capillary-to-capillary peak area reproducibility for aldehydes 219 in capillary microextraction using sol-gel dendrimer coatings Table 4.9 Limit of detection (LOD) data for aldehydes in CMEGC-FID 220 using sol-gel dendrimer microextraction capillaries Table 4.10 Chemical structures and pertinent ph ysical properties of 222 ketones analyzed using sol-gel dendrimer coating Table 4.11 Run-to-run peak area repr oducibility for ketones 223 in capillary microextraction using sol-gel dendrimer coatings Table 4.12 Capillary-to-capillary peak ar ea reproducibility for ketones 225 in capillary microextraction using sol-gel dendrimer coatings Table 4.13 Limit of detection (LOD) data for ketones in CME-GC-FID 226 sol-gel dendrimer microextraction capillaries Table 4.14 Chemical structures and pertin ent physical properties of 228 phenols analyzed using sol-gel dendrimer coating Table 4.15 Run-to-run peak area reproducibility for phenols in capillary 229 microextraction using sol-gel dendrimer coatings Table 4.16 Capillary-to-capillary peak area reproducibility for phenols in ca pillary microextraction using sol-gel 230 dendrimer coatings Table 4.17 Capillary-to-capillary peak area reproducibility for phenols in capillary microextraction us ing sol-gel dendrimer coatings 231

PAGE 12

ix Table 4.18 Chemical structures and pertinent physical properties of butylated hydroxy toluene (B HT) analyzed using sol-gel dendrimer coating 234 Table 4.19 Run-to-run peak area reproducib ility for BHT in capillary microextraction using sol-gel dendrimer coatings 235 Table 4.20 Capillary-to-capillary peak area reproducibility for BHT in capillary microextraction usi ng sol-gel dendrimer coatings 236 Table 4.21 Limit of detection (LOD) data for BHT in CME-GC-FID using sol-gel dendrimer microextraction capillaries 239 Table 4.22 Chemical structures and pert inent physical properties of alcohols analyzed using sol-gel dendrimer coating 240 Table 4.23 Run-to-run peak area reproducibility for alcohols in capillary microextraction using sol-gel dendrimer coatings 241 Table 4.24 Capillary-to-capillary peak area reproducibility for alcohols in capillary microextraction usi ng sol-gel dendrimer coatings 242 Table 4.25 Limit of detection (LOD) data for alcohols in CME-GC-FID using sol-gel dendrimer microextraction capillaries 263 Table 5.1 Names, functions and chemical structures of sol-gel poly-THF coating solution ingredients 298 Table 5.2 Analyte enrichment factors of commercial PDMS (100m),commercial PA (85 m) and sol-gel poly-THF sorbents for polar and nonpolar compounds 300 Table 5.3 Thermal stability data for sol-gel poly-THF coated microextraction capillaries 303 Table 5.4 Chemical structures and pertin ent physical properties of polyaromatic hydrocarbons (PAHs) analyzed using sol-gel poly-THF coated capillaries 304 Table 5.5 Run-to-run peak area reproducib ility for PAHs in capillary microextraction using sol-gel Poly-THF coating 305 Table 5.6 Capillary-to-capillary peak area for PAHs in capillary microextraction using sol-gel poly-THF coated capillaries 306

PAGE 13

x Table 5.7 Limit of detection (LOD) data for PAHs in CME-GC-FID using sol-gel Poly-THF mi croextraction capillaries 309 Table 5.8 Chemical structures and pert inent physical properties of aldehydes analyzed using sol-gel poly-THF coated capillaries 310 Table 5.9 Run-to-run peak area reproduc ibility for aldehydes in capillary microextraction us ing sol-gel poly-THF coating 311 Table 5.10 Capillary-to-capillary peak area reproducibility for aldehydes in capillary microextraction using sol-gel poly-THF coating 312 Table 5.11 Limit of detection (LOD) data for aldehydes in CME-GC-FID using sol-ge l poly-THF microextraction capillaries 315 Table 5.12 Chemical structures and pertinent physical properties of ketones analyzed using sol-gel poly-THF coating 316 Table 5.13 Run-to-run peak area repr oducibility for ketones in capillary microextraction using sol-gel Poly-THF coating 317 Table 5.14 Capillary-to-capillary peak area reproducibility for ketones in capillary microextraction us ing sol-gel poly-THF coating 318 Table 5.15 Limit of detection (LOD) data for ketones in CME-GC-FID using sol-gel poly-THF mi croextraction capillaries 321 Table 5.16 Chemical structures and pertinent physical properties of chlorophenols (CPs) analyzed using sol-gel poly-THF coating 322 Table 5.17 Run-to-run peak area re producibility for chlorophenols in capilla ry microextraction using so l-gel poly-THF coating 323 Table 5.18 Capillary-to-capillary peak area reproducibility for chlorophenols in capillary microextraction using sol-gel poly-THF coating 324 Table 5.19 Limit of detection (LOD) data for chlorophenols in CME-GC-FID using sol-ge l poly-THF microextraction capillaries 327 Table 5.20 Run-to-run peak area reproducibility for alcohols in capillary microextraction using sol-gel poly-THF coating 328

PAGE 14

xi Table 5.21 Capillary-to-capi llary peak Area (A x 10 -4 ) reproducibility for alcohols in capillary microextraction using sol-gel poly-THF coating 329 Table 5.22 Limit of detection (LOD) data for alcohols in CME-GC-FID using sol-gel poly-THF mi croextraction capillaries 330

PAGE 15

xii List of Figures Figure 1.1 Design of the first commercial SPME device 5 Figure 1.2 (a) Schematic representation of a PDMS coated stir bar 7 (b) Head space stir bar extraction device 7 Figure 1.3 Different modes of SPME operation 11 (a) direct extraction 11 (b) headspace SPME 11 (c) membrane-protected SPME 11 Figure 1.4 Schematic of an internally cooled SPME device 18 Figure 1.5 Derivatization techniques used in solid-phase microextraction 21 Figure 2.1 Graphical representation of numb er of articles published on SPME since its inception in 1989 28 Figure 2.2A Different formats of SPME (a) fiber SPME where extracting phase resides outside of a fiber, (b) in-tube SPME where extracting phase resides inside the capillary 29 Figure 2.2B Comparison of two major formats of SPME 30 Figure 2.3 Wire-in-tube SPME extraction capillary 41 Figure 2.4 Reactions involved in an tibody immobilization 46 (a) silanization of s ilica surface with APTES 46 (b) surface modification with 46 (c) immobilization of antibody 46 Figure 2.5 Scanning electron micrograph of the surface of an anodized aluminum wire 49 Figure 2.6 Scanning electron micrographs of (A) porous layer 55 (B) PDMS coated SPME fibers 55

PAGE 16

xiii Figure 2.7 Drawing of the headspace membrane SPME system 58 Figure 2.8 Different steps involve d in MIP synthesis 64 Figure 2.9 SEM image of ADS partic les immobilized on a fused silica fiber 68 Figure 2.10 Scanning electron micrographs of (a) uncoated metal surface 75 (b) PPY-coated metal surface 75 (c) PPPY-coated metal surface 75 Figure 2.11 Impact of storage on extracted sample using SPME sampler for long-term storage 78 Figure 2.12 Fully automated solid-phase dynamic extraction (SPDE) process 81 Figure 2.13 Schematic representation of a needle trap device 83 Figure 3.1 Classification of sol-gel sorbents used in fiber-SPME/in-tube SPME (CME) 96 Figure 3.2 Scanning electron micros copic image of sol-gel PDMS fiber at 3600-fold magnification 100 Figure 3.3 SPME-GC analysis of aliphatic alcohols (C 10 -, C 12 -, C 14 -, C 16 -, C 18 -)-using sol-gel PDMS fiber and GC/FID system 102 Figure 3.4 Scanning electron micrograph of sol-gel fullerol fiber at 2000fold magnification 108 Figure 3.5 Relative extraction efficiencies of commercial PDMS fiber (100m), commercial CW-DVB (65 m) and OH-DB14C4/OH-TSO (65 m) in amine extraction 117 Figure 3.6 Scanning electron micrograph of the DOH-B15C5 fiber at 800-fold magnification 118 Figure 3.7 Comparison of extraction e fficiencies of commercial PDMS (100 m), commercial CW/DVB (65 m) and DATEG/OH-TSO in extracting BTEX from aqueous solution 131

PAGE 17

xiv Figure 3.8 Comparison of mass absorbed in unit volume of sorbent using five different fibers fo r 10 ng/mL BTEX solution 153 Figure 3.9 Electron scanning microgr aph (600 x magnification of a sol-gel Carbowax 20M ormosil fiber 154 Figure 4.1 Schematic of a homemade capillary filling/purging device for preparation of capillary mi croextraction capillaries and open-tubular sol-gel GC columns 179 Figure 4.2 Schematic of a gravity-fed sample dispensing unit used in sol-gel dendrimer capil lary microextraction 188 Figure 4.3 Schematic representation of the connection of the extraction capillary with the analysis column inside the GC oven using a press-fit quartz capillary 192 Figure 4.4 Scanning electron microscopic image of a 250 m i.d. sol-gel dendrimer coated microextraction capillary illustrating the coating thickness. Magnification: 10,000x 204 Figure 4.5 Scanning electron microscopic image of a 250 m i.d. sol-gel dendrimer coated microextraction capillary illustrating the typical roughened surface obtai ned by sol-gel coating process. Magnification: 10,000x 205 Figure 4.6 Illustration of the extracti on kinetics of a non-polar compound (phenanthrene) and a pola r compound (2,4,6-trichlorophenol) obtained on a 13 cm x 250 m i.d. sol-gel dendrimer coated microextraction capillary using 100 ppb aqueous solution 207 Figure 4.7 CMEGC analysis of PAHs at 10 ppb concentration using sol-gel dendrimer coated microextraction capillary 211 Figure 4.8 Capillary MicroextractionGC analysis of Aldehydes at 100 ppb concentration using so l-gel dendrimer coated microextraction capillary 217 Figure 4.9 Capillary MicroextractionGC analysis of ketones at 100 ppb concentration using sol-gel de ndrimer coated microextraction capillary 224

PAGE 18

xv Figure 4.10 Capillary MicroextractionGC analysis of Phenols at 10 ppb concentration using sol-gel de ndrimer coated microextraction capillary 232 Figure 4.11 Capillary MicroextractionGC analysis of BHT at 10 ppb concentration using sol-gel de ndrimer coated microextraction capillary 337 Figure 4.12 Capillary MicroextractionGC analysis of alcohols at ppb level concentrations using sol-gel dendrimer coated microextraction capillary 243 Figure 5.1 Schematic of a gravity -fed sample dispensing unit in capillary microextrac tion with a sol-gel poly-THF coated capillary 258 Figure 5.2 Schematic of a homemade capillary filling/purging device for preparation of capillary microextraction capillaries and open-tubular sol-gel GC columns 261 Figure 5.3 Schematic representation of the connection of the extraction capillary with the analysis column inside the GC oven using a press-fit quartz capillary 272 Figure 5.4 IR spectra of pure polytet rahydrofuran (left), sol solution having all ingredients except polytetrahydrofuran (middle), sol-gel polytetrahydrofuran coating (right) 284 Figure 5.5a Scanning electron microscopic image of a 320 m i.d. sol-gel poly-THF coated fused sili ca capillary used in CME 286 Figure 5.5b Scanning electron microscopic image of a 320 m i.d. sol-gel poly-THF coated fused sili ca capillary used in CME 287 Figure 5.6a TGA curve of pure poly-THF for programmed heating (10 C/min) under N2 289 Figure 5.6b TGA curve of sol-gel poly-THF for programmed heating (10 C/min) under N2 290 Figure 5.7 Extraction time profiles of PAHs for direct-SPME using a commercial PDMS (30 m) fiber 292

PAGE 19

xvi Figure 5.8 Illustration of the extrac tion kinetics of nonpolar fluoranthene and phenanthrene) and mode rately polar (heptanophenone and dodecanal) compounds extracted on a 12.5 cm x 320 m i.d. sol-gel gel poly-THF coated capillary using 10 ppb aqueous solution of each analyte in a mixture 295 Figure 5.9 Comparison of extraction efficiencies of commercial PDMS (100 m), PA (85 m) and sol-gel poly-THF (0.5 m) coatings 298 Figure 5.10 Capillary microextractionGC analysis of PAHs (20 ppb each using sol-gel poly-THF coated capillary 308 Figure 5.11 Capillary microextractionGC analysis of aldehydes at 20 ppb concentration using poly-THF coated capillary 314 Figure 5.12 Capillary microextractionGC analysis of ketones at (20 ppb) using sol-gel poly-THF coated capillary 320 Figure 5.13 Capillary microextractio nGC analysis of chlorophenols using poly-THF coated capillary 326 Figure 5.14 Capillary microextractionGC analysis of alcohols (100 ppb each) using sol-gel poly-THF coated capillary 331 Figure 5.15 Capillary microextractionGC analysis of a mixture of nonpolar, moderately polar and polar compounds using poly-THF coated 333

PAGE 20

xvii List of Schemes Scheme 4.1 Hydrolysis reaction of (a) methyltrimethoxysilane (MTMOS) and (b) phenyl terminated dendrimer with a trimethoxysilane derivatized root 197 Scheme 4.2 Polycondensation of hydrolyzed methyltrimethoxysilane 198 Scheme 4.3 Condensation of phe nyl-terminated sol-gel-active dendron with a growing sol-gel network formed from MTMOS 199 Scheme 4.4 Chemical anchoring of so l-gel dendrimer stationary phase 200 Scheme 4.5a Deactivation of residual silanol group using HMDS 201 Scheme 4.5b Deactivation of fused s ilica capillary inne r surface with PMHS 202 Scheme 5.1 Cationic ring opening polymerization of tetrahydrofuran 275 Scheme 5.2 Hydrolysis of th e sol-gel precursor, MTMOS 278 Scheme 5.3 Polycondensation of hydrolyzed MTMOS 279 Scheme 5.4 Chemical incorporation of poly-THF into the solgel network 280 Scheme 5.5 Chemical anchoring of the sol-gel hybrid organicinorganic polymer to the silanol groups on the fused silica capillary inner walls 281 Scheme 5.6 Deactivation of residual silanol groups by deri vatization with hexamethyldisilazane (HMDS) 282

PAGE 21

xviii List of Abbreviations, Symbols, and Acronyms AL Alabama AZ Arizona CA California CE Capillary Electrophoresis CEC Capillary Electrochromatography CME Capillary Microextraction CME-GC Capillary MicroextractionGas Chromatography CW/DVB Carbowax/Divinylbenzene CW/TPR Carbowax/Templated Resin CAR/PDMS Carboxen/Polydimethylsiloxane DVB/CAR/PDMS Divinylbenzne/Car boxen/Polydimethylsiloxane FID Flame Ionization Detector GC-FID Gas Chromatography with Flame Ionization Detector GC Gas Chromatography HMDS Hexamethyldisilazane HPLC High-Performance Liquid Chromatography HPLC-MS High-Performance Li quid Chromatography-Mass Spectrometry ICP-MS Inductively Coupled PlasmaMass Spectrometry

PAGE 22

xix INCAT Inside Capillary Adsorption Trap LLE Liquid-Liquid Extraction LOD Limit of Detection MEKC Miceller Electro Kinetic Chromatography MTMOS Methyltrimethoxysilane PA Pennsylvania PAH Polycyclic Aromatic Hydrocarbon PEEK Polyetheretherketone PEG Polyethylene Glycol PMHS Poly(methylhydrosiloxane) PPY Polypyrrole PTV Programmed Temperature Vaporizer RMS Root Mean Square RPM Revolutions per Minute RSD Relative Standard Deviation SEM Scanning Electron Micrograph S/N Signal-to-Noise Ratio SPE Solid Phase Extraction SPME Solid Phase Microextraction TFA Trifluoroacetic Acid THF Tetrahydrofuran USEPA United States Environmental Protection Agency VOC Volatile Organic Compound

PAGE 23

xx Polytetrahydrofuran-and Dendrimer-Based Novel Sol-Gel Coatings for Capillary Microextraction (CME) Providing Parts Per Trillion (ppt) and Parts Per Quadrillion (ppq) Level Detection Limits in Conjunction With Gas Chromatography and Flame Ionization Detection (FID) Abuzar Kabir ABSTRACT Sol-gel capillary microextraction (CME) is a new direction in solvent-free extraction and preconcentration of trace analyt es. CME presents significant interest in environmental, pharmaceutical, petrochemical, biomedical, agricultural, food, flavor, and a host of other important areas. Sol-gel CME utilizes advanced ma terial properties of organic-inorganic hybrid sol-gel polymers to perform efficient extraction and enrichment of target analytes from a variety of matr ices. In this disserta tion, two novel sol-gel coatings were developed for CME: (a) solgel benzyl-terminated dendrimer coating, and (b) sol-gel polytetrahydrofuran (poly-THF) coating. A detailed investigation was conducted to evaluate the pe rformance of the newly deve loped sol-gel coatings in solvent-free extraction of a wide ra nge of polar and nonpolar analytes. Sol-gel chemistry was used to chemically immobilize dendrimerand poly-THFbased hybrid organic-inorganic coatings on fused silica capillary inner surface. The solgel coatings were created using a coating solution containi ng a sol-gel active organic component (dendrimer or poly-THF), a sol-gel precursor (methyltrimethoxysilane, MTMOS), a sol-gel catalyst (trifluoroacetic acid, TFA, 5% water) and a deactivating

PAGE 24

xxi reagent (hexamethyldisilazane, HMDS). So l-gel reactions were conducted inside a hydrothermally treated fused silica capillar y for 60 min. A wall-bonded sol-gel coating was formed via condensation of silanol gr oups residing on the capi llary inner surface with those on the sol-gel network fragments evolving in close vicinity of the capillary walls. Due to the strong chemical bonding w ith capillary inner walls, these sol-gel coatings showed excellent thermal and solv ent stability in CME in hyphenation with gas chromatography (GC). Using a Flame ionizati on detector (FID), low parts per trillion (ppt) and parts per quadrillion (ppq) level de tection limits were achieved in CME-GC for both polar and nonpolar analytes including pol ycyclic aromatic hydrocarbons (PAHs), aldehydes, ketones, phenols, and alcohols. The sol-gel coatings were found to be effective in carrying out simultaneous extr action of both polar and nonpolar analytes from the same sample. To our knowledge, two publications resulting from this research [A. Kabir et al. J. Chromatogr. A 1034 ( 2004) 1-11; A Kabir et al. J. Chromatogr. A 1047 (2004) 1-13] represent the first reports on the developmen t and use of sol-gel dendrimer and sol-gel poly-THF coatings in anal ytical microextraction.

PAGE 25

1 Chapter One Solid Phase Microextraction: A Solvent-Free Sample Preparation Technique 1.1 Introduction An analytical procedure consists of seve ral individual but equally important steps: sampling, sample preparation, separation, quan titation, and statistical interpretation of analytical data, often used in important socio-economical decision making. Since the overall success of an analytical procedure depends on the performance of each individual step, and because the overall speed of the anal ytical procedure is determined by the speed of the slowest step, it is very important to pa y necessary attention to each step. Sampling and sample preparation are considered to be the slowest among all the steps involved in an analytical procedure. Surveys show that more than 80 % of the analysis time is spent on sampling and sample preparation [1]. Therefor e, if we are to expedite the analytical procedure, much attention needs to be paid to sampling and sample preparation steps. Although the last couple of decades have witnes sed significant improvement in analytical instrumentation, inadequate attention has b een paid to improvement of the samplepreparation step. This is paradoxical since th e sample preparation is an unequivocally necessary step to isolate the ta rget analyte(s) from the sample matrix as well as to purify and concentrate the analyte(s) to reach the detection limit of the analytical instrument involved.

PAGE 26

2 Conventional sample preparation proce dures include Soxhlet extraction [2], liquid-liquid extraction (LLE) [3], accelerated solvent extraction (ASE) [4], microwaveassisted solvent extraction (MASE) [5], solid -phase extraction (SPE) [6], supercritical fluid extraction [7], and purge and trap [8]. These conventional sample preparation techniques are time-consuming, labor-inten sive and involve multi-step operation. Moreover, they often employ large volumes of hazardous organic solvents. Most of these techniques are not suitable for field app lication. Therefore, sample collection and transportation to the la boratory is required prior to fu rther processing. As a consequence, incorrect sample handling duri ng collection, transportation, and preservation may lead to significant variability in analysis results putting the validity and authenticity of the method in question. In an attempt to eliminate these sampling and sample preparation related problems, Belardi and Pawliszyn [9] developed solid-phase microextraction in 1989 which not only is a solvent-free technique but also incor porates sample extraction and preconcentration in a single step. Solid-pha se microextraction pr ovides analysts the flexibility in sample preparation both in the laboratory and in the field where the material system under investigation is located. 1.2 Evolution of solid-phase microextraction: a historical perspective The history of solid-phase microextraction dates back to late 1980s when Pawliszyn and co-workers[10] were involve d in laser desorption/gas chromatography experiments to accomplish rapid speed n separation, even for high molecular mass

PAGE 27

3 species. A small piece of optical fiber was used in the experiment. One end of the optical fiber was dipped into the concentrated anal yte solution to generate a coating of the analyte(s) on the surface of it by evaporating th e solvent. Then the fiber was inserted into the GC injection port to transfer the analyt e(s) from the fiber into the injection port through desorption. Desorption was accomplished by transmitting laser light energy through the optical fiber. The desorbed anal ytes then got separated as they moved through the GC analysis column. It was observe d that using laser pulse, analytes could be quantitatively transferred from the optical fi ber probe to GC injection port for further separation, but the sample prepar ation step took hours. As a result, the tota l time for the sample analysis still remained fairly long. This led to a strong recognition of the need for a rapid sample preparation technique that mi ght assist in retaining the time efficiency advantages accomplished by using laser pulse and a high-speed separation instrument. To address the challenge, similar optical fibers having polymeric coating on the outer surface were utilized. The purpose of the polymeric co ating (10-100 m) was to protect the fibers from breakage. In the preliminary work on SPME [9, 11], sections of fused silica optical fibers, both uncoated (obtained by burni ng off the polyimide coating), and coated with liquid and solid polymeric phases were used to verify the conceived idea of SPME. Extraction was carried out by inserting both the coated and uncoated fibe rs into the aqueous solution of the analyte(s) and then placed the fibers into the GC injection port for the thermal desorption of the extracted analyte(s). Prelimin ary data indicated the future potential of the new technique by extracting both polar and nonpolar analyt e(s) rapidly and reproducibly from aqueous solution.

PAGE 28

4 The success in preliminary investigat ion on solid phase microextraction accelerated its rapid development leading to th e invention of a rather simple device by attaching the coated fiber to the plunger of a microsyringe. This, in fact, was the first SPME device [11]. In this first SPME device the extraction fiber was housed inside a Hamilton TM 7000 series microsyringe. The plunger of the syringe was glued to the fiber with epoxy glue. This simple mechanism allowe d facile movement of the coated end of the fiber in and out of the syringe needle by pushing the plunger forward or by retracting it back. A forward movement of the plunger ex posed the fiber coating to the sample, and extraction of the target analyte(s) was acco mplished because of higher affinity of the analyte(s) for the coated sorb ent than the sample matrix. Af ter carrying out the extraction, the extracted analyte(s) were released from the coating into the injection port of a gas chromatograph by thermal desorption. For this, the exposed fiber was first retracted back into the syringe needle to protect it from mechanical damage, and then the needle was used to vertically pierce the septum of the injection port. Followi ng this, the plunger was pushed forward to expose the fibe r to the space in th e glass liner of the hot injection port. As a result, the analyte(s) extracted on the fiber were desorbed and entered into the GC analysis column for separation. Figure 1.1 illustrates the configuration of the first commercial SPME device [12].This configur ation has drawn wide acceptance among the researchers and proved suitabl e particularly for gas phase separation, although several other modifications of the SPME devices were also reported [13, 14].

PAGE 29

Figure 1.1 Design of the first commercial SPME device (Reproduced from Ref. [12] with permission from Elsevier) 5

PAGE 30

6 se td SPME has prompted researchers to develop so called i are d There are several other formats of SPME worth mentioning. Nickerson et al. [15] proposed a sorbent coated vial sealed with a rubber septum. Piercing the septum with a ampling syringe can collect analytes from the headspace inside the sealed vessel. For extracting less volatile compounds, the vessel can be even heated. Another recently popular format is organic polymer coated s tir bar[16] in which sorbent materials are coated on a small stir bar using either sta tic coating technology[17] or sol-gel coating technology[18] originally reported by Malik and co -workers, and subsequently adapted by Liu et al. [19] for stir bar SPME. Figure 1.2 demonstrates a coated stir bar and the device for headspace stir bar extraction [20]. The sorbent coating on the external surf ace of the fiber, however, is not wellsuited for hyphenation with liquid phase sepa ration (e.g., HPLC, CE, CEC, etc.) becau in this case the extracted anal yte(s) needs to be desorbed using an organic sorbent which later would carry the analyte( s) to the liquid mobile phase for separation. The forma related incompatibility of fiber-base n-tube SPME. In this format, the extr action sorbent is located on the inner surface of a fused silica capillary. In most cases, a pi ece of open tubular GC column is used for this purpose. The coating inside the capillar y acts as the sorbent. The aqueous sample containing the analytes is passed through th e capillary and the targ et analyte(s) get extracted by the sorbent. Once the extraction system reaches equilibrium, the analytes washed into the HPLC column using the liquid organo-aqueous mobile phase or any organic solvent which is compatible with th e liquid mobile phase. This format has opene the window for automated sample preparat ion and preconcentration in liquid phase separation (HPLC, CE, CEC, etc.) s

PAGE 31

7 Figure 1.2 (a) Schematic representation of a PDMS coated stir bar and (b) Head space ir bar extraction device (Reproduced from Ref. [20] with permission from American Chemical Society) st

PAGE 32

8n n ting. The time required for establishing the extraction equilibrium depends on various factors: coating on of the the extracting phase and al state of the matrix, etc. The ibed by the following equation: 1.3 Working principle of solid phase microextraction In solid phase microextraction, the extracting phase coated either on the outer surface of a fused silica fiber (fiber-SPME) or inside a fused silica capillary (in-tube SPME, Capillary microextraction) plays a vital role in the extraction process. Unlike solid phase extraction (SPE), solid phase microextraction is an equilibrium extractioprocess. When the extracting phase comes in contact with the analyte(s) to be extracted,the analyte concentration equilibrium is gradually established between the sample matrix and the extracting phase. As soon as the equilibrium is reached, the extracting phase cannot accumulate the analyte(s) any more: the amount of the analyte extracted in a givetime is exactly equal to the amount of analyte desorbed from the coa thickness, partition co-efficient between the analyte and extracting phase, agitati sample matrix, extraction temperature, solute diffusion rates in the sample matrix, viscosity of the sample matrix, physic equilibrium condition can be descr nf = Kfs Vf VsC0 Kfs Vf + Vs Where n Equation 1.1bution constant centration of the analyte(s) in the sample. f = number of moles of the analyte(s) extracted by the extracting phase K fs = fiber coating/ sample matrix distri V f = volume of the extracting phase on the fiber V s = volume of the sample C 0 = Initial con

PAGE 33

9This equation is valid for partitioning equilibrium involving liquid polymeric extracting he the value of the term KfsVf becomes insignificant. Therefore, Equation phases. When solid sorbents are used, this equation is valid only when the concentration of analyte(s) is low and the sorbent is porous. When the volume of the sample is very large compared to the volume of textracting phase, 1.1 can be written as nf =Kfs Vf C0Equa t ion 1.2 As can be seen from the equation 1.2, once the extraction equilibrium is reached, tha direct proportional relationship between the initial concentration of the analyte(s) athe amount (moles) of analyte extracted. This is the theoretical basis for analyte Quantification. This equation clearly expresses the advantage of SPME for field application. The amount of extracted analyte(s) is independ ere is nd ent of the sample volume. There is no need to ollect a defined sample prior to analysis as the extracting sorbent can be exposed irectly to the air, water, production stream where the target analyte is located. The mount of the analytes extracted onto the extracting sorbent will correspond directly to its oncentration in the matrix. Thus SPME incorporates sampling and sample preparation ates the errors associated with sample storage (adsorption n the container wall or decomposition). c d a c into one single step and elimin o

PAGE 34

101.4.1 or en solution, s ches equilibrium between the aqueous phase and the air in the closed container. Extraction is carried out from the gaseous phase. As a result, unwanted matrix interferences (e.g., compounds having high molecular mass, humic materials, proteins, etc.) cannot disturb th e performance of the coating as well as its life-span. Moreover, such an extraction mode allows the modification of the matrix such as pH change, addition of salts etc. frequently used to improve the extraction efficiency of the coating. 1.4 Modes of extraction Extraction modes with a coated fiber When coated fibers are used in solid phase microextraction, there are three different modes of extraction that can be fo llowed: (a) direct extr action, (b) headspace extraction, and (c) membrane protected extraction. Figure 1.3 illustrates the different modes of extraction used in fiber-SPME [12]. In the direct extraction mode, the coated fiber is inserted into the matrix containing the target analyte(s) so that the analyte(s) can be dire ctly transported from matrix to the extracting phase on the fiber (coating). The rate of extraction can be increased in order to achieve faster extr action equilibrium by employing agitation. F gaseous samples, natural conv ection of air is sufficient fo r faster equilibrium. Wh extraction is carried out in a queous solution, different means of agitation (e.g., fast flow of the aqueous solution, rapid fi ber or vial shaking, stirring or sonication of the etc.) may be employed. In the headspace mode, the coated fiber stays above the aqueou phase in a sealed container. The analyte(s) first rea

PAGE 35

11 Figure 1.3 Different modes of SPME operation (a) direct extraction, (b) headspace SPME, (c) membrane-protected SPME (Reproduced from Ref. [12] with permission from Elsevier)

PAGE 36

12 le ion The time needed for accomplishing extrac tion equilibrium greatly depends on the ode of extraction. In headsp ace extraction, analyte(s) to be extracted are first removed om the aqueous phase to the headspace followed by the extraction on the sorbent coating in this case is typically limited by mass transfer rates of the analyte(s) from the aqueous phase to the headspace. As a result, v sfer to r mode compared to direct mode under similar agitation conditions. The third extraction mode is membrane protected extraction. In this case, a membrane is used which surrounds the fiber coating to give prot ection against unwanted sorptive components in the sample matrix. This extracti on mode is partic ularly suitab when the matrix is very dirty. In additi on to protecting the fiber from damage, using appropriate membrane material may provide additional selectivity to the extraction process. One of the major disadvantages of th is mode is its substa ntially longer extract equilibrium. Using thinner membrane and elev ated extraction temperature may results in shorter extraction time [21]. m fr coating on the fiber. Overall mass transfer to the olatiles are extracted fa ster than semi-volatiles due to their faster mass tran the headspace. Temperature has a significant impact on the kinetics of the extraction process because it determines th e vapor pressure of the analyt e(s). As is known, diffusion coefficients in gaseous phase are typically 2-4 orders of magnitude larger than in aqueous phase and since mass transfer from aqueous pha se to vapor phase is faster due to low vapor pressure of the analyte( s), equilibration time for volat iles are significantly shorte in headspace

PAGE 37

13 r static. In activ e or dynamic extraction mode, the matrix contain e terial In dynamic in-tube SPME, it is assumed that an internally coated piece of fused lica capillary (in most of the cases, a piece of open tubular capillary GC column) or a phase di spersed on an inert support (typically a piece of micro-LC capil lary column) is used. In this type of configuration, the concen s a g h the yte 1.4.2 Extraction modes with in-tube SPME In in-tube SPME, there are two different modes of extraction: (a) active or dynamic and (b) passive o ing the analyte(s) is pa ssed through the tube and the an alytes are extracted by th sorbent coating as the sample passes by. In pa ssive or static extraction mode, the capillary is filled with sample matrix and the high affi nity of the analytes fo r the sorbent ma serves as the driving force for their extr action by the coating re siding on the capillary inner walls. si fused silica capillary packed with extracting tration profile along the axis (x) of the tubing contai ning the extracting phase a function of time (t) can be described by a dopting the expression for dispersion of the concentration front [22]. In static in-tube SPME, the extracting phase resides inside a protective tubin (needle), is not exposed direc tly to the matrix sample. The extraction occurs throug static gas phase present in the needle. In this case, the only mechanism of anal transport is diffusion through the gaseous phase contained in the tubing. The static intube SPME is particularly suitable for field sampling.

PAGE 38

14 roaches may be followed. In one approach, the fibers are im mersed into the solution made of organic solvent d lic rods [11]. In preparing commercial coatings, the fiber is first drawn in a tower. Then, the fiber is passe d through a specified diameter rifice (depends on the diameter of the fiber as well as the coating thickness) in a cup that n PA that coats the fibe r. As the fiber is drawn, a coating of the extracting phase is deposited on its outer surface. rgely stirring, 1.5 Preparation of coatings on fibers In order to prepare coatings on fused s ilica fibers, several app and the polymeric material to be deposited for a given period of time. After removing the fibers from the solutions, the solvent is evaporated and the deposite material can be immobilized through cross linking [9]. Coatings can also be made through electrochemical depos ition in which selectiv e coatings are deposited electrochemically on the surf ace of metal o contais the desired phase e.g ., PDMS, 1.6 Experimental parameters affecting extraction efficiency Although the physico-chemical characterist ics of the coating materials la determine their extraction efficiency, ther e are several experimental parameters optimization of which generally increase extr action efficiency and improve the detection limit to a significant extent. Among such experi mental factors is pH adjustment, heating, and salt addition to the matrix.

PAGE 39

15 the commonly used SPME coatings are electrically neutral, they are most suitable for the extraction of neutral com pounds from water (by direct or headspace s which rema in neutral in aqueous solution can be easily extracted by solid phase microextra ction. Changing the pH of the matrix contain1.6.2 n the mode of extraction. In the direct extraction mode, the coated fiber is inserted directly into the ransported direc tly from the sample matrix to the extracting phase. In order to facilitate ra pid extraction, agita tion of the sample matrix is needed to 1.6.1 pH adjustment of the matrix Since extraction). Therefore, compound ing such neutral com pounds does not yield any impact on extraction efficiency. However, compounds like organic acids and base s get dissociated into ionic species when dissolve in water. In order to extract such compounds from aqueous media, pH adjustment is required to convert the ionic species into ne utral molecules. Optimum pH of the matrix, in this case, depends on the pK a or pK b values of organic acids and bases respectively. In order to make sure that 99% of the organic aci d is in neutral form, the pH of the matrix should be at least two units lower than pK a values of the analytes. In case of basic compounds, similarly, pH of the matrix should be at least two uni ts larger than their pK b values. Agitation of the matrix Although, in general, agita tion of the sample matrix significantly reduces the extraction equilibrium time, the impact of agitation mostly depends o sample and the analytes are t

PAGE 40

16 augmen analytes moves rapidly from the headspace to the extract l, if the extraction time is of major concern, the highest temperature which still provides satisfactory sensitivity sh ould be used. When extraction sensitivity, not the extraction time, is the major concern, as in case of trace analysis, extraction t analyte transport from the bulk of the liquid sample to the vicinity of the fiber. Fast sample flow through the capillary (intube SPME), rapid fiber or vial movement, stirring or sonication are most common means employed for agitation. Agitation minimizes the effect caused by so called depletion zone formed close to the fiber as a result of fluid shielding and slow diffusion coefficients of analytes in liquid matrices. In the headspace mode, agitation of the wa ter sample to generate a continuously fresh surface would accelerate the mass transfer of less volatile analytes from the water to headspace. Once in the gaseous phase, ing media due to their large diffusion coefficients in the gas phase. 1.6.3 Heating the matrix Extraction temperature has a profound imp act on overall extraction efficiency. However, the impact is comprised of two opposing forces: (a) diffusion of the analyte increases with the increase in temperature l eading to faster mass transfer from the liquid phase to the headspace and from the headspace to the extracting sorbent. As a result, at higher temperature extraction equilibrium can be reached in relatively quickly which significantly reduces the extraction time. (b) As extraction is an exothermic process, the distribution constant decreases with increase in temperat ure. Therefore, extraction yield would be lower at higher temperature. In genera

PAGE 41

17 should be carried out at as low temperature as possible. o prevent sensitivity loss at elevated ex traction temperature, the extracting phase can be cooled simultaneously with heating the sample matrix. Zhang et al. [23] proposed the design of an internally cooled SMPE devi ce which allows simulta neous cooling of the sorbent as well as heating of the sample matr ix. In this assembly, two concentric fused silica capillaries are used and the sorbent is coated at the outer surface of the larger diameter capillary. The inner capillary is us ed to deliver liquid carbon dioxide to the coated end resulting in a significantly lower coating temperature compared to the matrix temperature. Figure 1.4 presents the design of an internally cooled SPME device [23]. T

PAGE 42

18 Figure 1.4 Schematic of an internally cooled SPME device (Reproduced from Ref. [23] with permission from American Chemical Society)

PAGE 43

19 n of organics from aqueous solution. Although sodium chloride salt is used predom inantly in SPME practice, other salts e.g., 2CO3, K2CO3 may also be used [26]. Although addition of salt usually increases the extracti on efficiency, the opposite behavior has also been re atrix matic y appreciable e ffect on extraction of efficiency in SPME extracti at for pesticide extraction, the presence of ethanol in the extracted so lution is an important parameter for the 1.6.4 Addition of salt to the matrix Addition of salt into the a queous phase containing or ganic analytes may have profound impact on the extracti on efficiency. Several resear ch groups have studied the impact of salt addition in deta il [24, 25]. Addition of salt (a lso called salting out) is a well known phenomenon used for improving extractio CaCl 2 NH 4 Cl, (NH 4 ) 2 SO 4 MgSO4, Na ported [24, 25]. In gene ral, the effect of salt addition increases with the increase in polarity of the compound [25]. 1.6.5 Addition of organic solvents In general, the presence of small amount of organic solvents in the analyte m does not have significant impact on the extract ion efficiency. However, there are several reports on the subject that apparently do not agree with each other. Arthur et al. [27] reported that the presence of up to 1% meth anol in aqueous soluti on containing aro hydrocarbons does not have an on of aromatic hydrocar bons. R. Eisert [28] observed less than 10% loss in extraction yield of selected triazines when methanol content in the aqueous solution was as high as 5%. On the other hand, Urruty et al. [29] reported th

PAGE 44

20 ncy. is therefore apparent that the effect of the presence of or ganic solvent in the extracti to and biological matrices that are also polar or rich in water. To meet these chal lenges, derivatization tec hniques are frequently used. Figure 1.5 summarizes different deriva tization techniques that are compatible with SPME [30]. In direct derivatizat ion, the derivatizing agent is added to the sample vial to convert the analyte(s) into a stable nonpolar compound by chemical reaction with it followed by the extraction on the extraction fiber. extraction method efficie It on matrix is compound and matrix depe ndent and warrants further investigation generalize the impact of the pres ence of solvent in the matrix. 1.7 Derivatization Due to their strong hydrophobi city, nonpolar analytes are easy to extract from aqueous samples using any nonpolar sorbent. Bu t the extraction of polar analytes poses a challenge to the scientist becau se of their affinity for aqueous matrix. As a result such analytes are difficult to extract from environm ental

PAGE 45

21 Figure 1.5 Derivatization techniques used in solid-phase microextraction (Reproduced from Ref. [30] with permission from American Chemical Society)

PAGE 46

22 hen the polarity of the sorbent is su fficient for extracting polar analyte from aqueous phase, derivatization of the analytes may still be required to facilitate better separation in the GC analysis column or to increase detector res ponse by converting the analytes into suitable form. This can be done by employing derivatizat ion in GC injector port. Instead of using derivatization in GC in jector port, another c onvenient way is to use in-coating derivatization following extraction. nother interesting and potentially very useful approach for derivatization is simultaneous derivatization and extraction, performed directly on the coating. This approach yields high efficiency and can be easily used in field applications. In this case, the fiber is doped with a deriva tizing reagent and subsequently is exposed to the sample. The analytes are extracted and simultaneou sly converted to co mpounds having high affinity to the c W A oating.

PAGE 47

23 [3] 7) 93. [5] ev. Anal. Chem. 25 (1995) 43. [7] A. ) 1198. [10] ) 1475. [12] Chromatogr. A 885 (2000) 153. action and tional counterparts 1991. r 1.8 References for Chapter One [1] M. F. Alpendurada, J. Chromatogr. A 889 (2000) 3. [2] V. Lopez-Avila, K. Bauer, J. Milanes, W. F. Beckert, J. AOAC Int. 76 (1993) 864. R.E. Majors, LC-GC Int. 10 (199 [4] B.E. Richter, B.A. Jones, J.L. Ezzel, N.L. Porter, N. Abdalovic, C. Pohl, Anal. Chem. 68 (1996) 1033. A. Zlotorzynski, Crit. R [6] K. Coulibaly, I.J. Jeon, Food Rev. Int. 12 (1996) 131. S.B. Howthorne, Anal. Chem. 62 (1990) 633 [8] M.M. Minnich, J.H. Zimmerman, B. A. Schumacher, J. AOAC Int. 79 (1996 [9] R.P. Belardi, J. Pawliszyn, Water Pollut. Res. J. Can. 24 (1989) 179. J. Pawliszyn, S. Liu, Anal. Chem. 59 (1987 [11] C.L. Arthur, J. Pawliszyn, Anal. Chem. 62 (1990) 2145. H. Lord, J. Pawliszyn, J. [13] J. Pawliszyn, Method and device for solid-phase microextr desorption. PCT, International Pate nt Publication Nu mber WO 91/15745 and na [14] M. McComb, E. Giller, H.D. Gesser, in: 78 th Canadian Society fo

PAGE 48

24 h, Guelph, ON. n within a collection (1999) [17] er-Troyer, M. Crnoja, E. Rosenberg, M. Grasserbauer, [18] J.D. Hayes, B. W. Wilhite, A. Malik, Anal. Food Chem. 50 (2002) 449. [21] Z. Zhang, J. Poerschmann, J. Pawliszyn, Anal. Commun. 33 (1996) 219. [22] J. Crank, in: Mathematics of Diffus ion, Clarendon Press, Oxford, 1989, p.14. [23] Z. Zhang, J. Pawliszyn, Anal. Chem. 67 (1995) 34. [24] A.A. Boyd-Boland, S. Magdic, J. Pawliszyn, Analyst 121 (1996) 929. [25] R. Eisert, K. Levsen, J. Am. Soc. Mass Spectrom. 6 (1995) 1119. [26] T. Kumazawa, X.-P. Lee, K. Sato, O. Suzuki, Anal. Chim. Acta 492 (2003) 49. [27] C.L. Arthur, L.M. Killam, K.D. Buchhol z, J. Pawliszyn, J.R. Berg, Anal. Chem. 64 (1992) 1960. Chemistry Conference and Exhibition, 1995, p. Abs. 528, University of Guelp [15] M.A. Nickerson, Sample screening and preparatio vessel. US Patent number 5,827, 944, 1998. [16] E. Baltussen, P. Sandra, F. David, C. Cramers, J. Microcol. Sep. 11 737. C. Haberhau Fresenius' J. Anal. Chem. 366 (2000) 329. S.L. Chong, D.X. Wang, Chem. 69 (1997) 3889. [19] W. Liu, H. Wang, Y. Guan, J. Chromatogr. A 1045 (2004) 15. [20] C. Bicchi, C. Iori, P. Rubiolo, P. Sandra, J. Agric.

PAGE 49

25 [28] R. Eisert, K. Levsen, G. Wu ensch, Vom Wasser 86 (1996) 1. [29] L. Urruty, M. Montury, J. Agric. Food Chem. 44 (1996) 3871. [30] L. Pan, J. Pawliszyn, Anal. Chem. 69 (1997) 196.

PAGE 50

26 Chapter Two An Overview on Stationary Phases Used in Solid Phase Microextraction (SPME) 2.1 Introduction Solid phase microextraction ( SPME) is an environmentally benign solvent-free sample preparation technique that wa s developed by Pawliszyn and co -workers [1] about sixteen years ago. Because of its simplicity, cost -effectiveness, porta bility, solvent-free operation, environmental friendliness and eas e in automation, SPME has experienced a tremendous growth sinc e its inception in 1989. A recent literature search using SciFinder Scholar 2004 database on March 20, 2005 revealed that ov er 4,379 articles have been published on SPME in last 16 year s of its existence, among which more than 750 papers have been published in 2004 alone. The search results are presented in Figure 2.1 which shows a rapid growth in the num ber of articles published on solid phase microextraction since its inception. With in a very short period of time SPME has already been recognized as a viable alternative to traditional sample prep aration techniques (e.g., liquid-liquid extrac tion and solid phase extraction) that are not only hazardous due to the use of toxic organic solvents in their operation but also tim e consuming, labor-intensive, and cumbersome. Compared to traditional samp le preparation techniques, SPME offers

PAGE 51

27 many advantageous features (e.g., simplic ity, speed, cost-effectiveness, ease in automation, portability, solvent-free operations, environmental friendliness, health safety and so on). Unlike liquid-liquid extraction and solid phase extraction, solid phase microextraction is an equilibrium-based non-exhaustive extraction method. In SPME the extracted amount is practically i ndependent of the sample size. SPME is frequently coupled to gas ch romatography (GC). SPME-GC analysis consists of two major steps: (a) equilibrium extraction of the target analyte(s) from the sample matrix and (b) desorption of the extr acted analyte(s) into the GC for analysis. Success in each of these step s greatly depends on the nature and prope rty of the sorbent coatings used as the extracting phase. Based on the substrate on which the extracting phase is coated, SPME can be classified into two general formats: (a) fiber SPME, and (b) in-tube SPME. In fiber SPME, the sorbent is coated on the oute r surface of a small-diameter (typically100 m o.d.) solid rod called fiber. Fu sed silica fibers are most comm only used for this purpose. Some researchers also reported the use of meta llic fibers (stainless st eel, copper etc.) to facilitate rapid thermal desorption of the ex tracted analytes for GC analysis and to prevent frequent breakage. In in-tube SPME (capillary microextraction), the coating is created on the inner surface of a fused silica capillary. Frequently, a small piece of wallcoated capillary GC column is used for th is purpose. Figure 2. 2A demonstrates two different formats used in SPME. Figure 2. 2B compares the two predominantly used formats in SPME showing the position of the coating.

PAGE 52

28 Articles Published on SPME YearNumber of Articles Published Figure 2.1 Graphical representation of numb er of articles published on SPME since its inception in 1989.

PAGE 53

Figure 2.2A Different formats of SPME (a) fiber SPME where extracting phase resides outside of a fiber, (b) in-tube SPME where extracting phase resides inside the capillary (Reproduced from Ref. [2] with permission) 29

PAGE 54

30 (a) Coated fused silica fiber (b) Polyimide coated fused silica capillary (c) Enlarged cross section of a fused silica capillary Sorbent coating Fused silica fiber Polyimide coated fused silica capillary Polyimide coating on outer surface of the capillary Fused silica glass Sorbent coating inside the capillary Figure 2.2B Two major formats of SPME

PAGE 55

31 uperior ers of n the s provide higher sensitivity. However, thicker usion of analyte(s) within the polymeric extraction phase. Therefore, it is important to use appropr iate coating thickness depend E, (b) c ) tailor-made coatings for fiber SPME, d) ilor-made coatings for in-tube SPME. Although numerous inherent positive attribut es have placed SPME in a s position among currently available sample prep aration techniques, the small numb commercially available fiber coatings are not being able to keep up with the growing demand on new sorbents needed to mainta in the healthy growth of SPME and its penetration into ever increasi ng areas of science and technolo gy. Lack of progress i area of sorbent development may prove to be a serious hurdle for the future growth of SPME. The effectiveness of solid phase microextracion depends on the analyte distribution constant (K) between the SPME coating and the sample matrix. This characteristic parameter describes the affin ity of the SPME coating toward the analyte relative to similar affinity exhibited by the sample matrix. The choice of a particular coating depends on the physiochemical nature of the analytes to be extracted. The principle like dissolves like may serve as a rule of thumb in the selection of SPME coatings. Coating volume is another impor tant factor that determines the method sensitivity and should be given due considera tion in method development, especially for trace analysis. In general, thicker coating coatings require longer extraction times due to the slow diff ing on the required extraction sensit ivity and analysis time. Depending on the format of SPME in which the extracting so rbents are used, SPME sorbents may be grouped into four major classes: (a) commercially availa ble coatings for fiber SPM commercial GC coatings for in-tube SPME, ( ta

PAGE 56

322.2 Commercially available sorbents for fiber SPME SUPELCO Inc. has bly Two different types of coated SPME fibers are available from SU PELCO: homogeneous meric sorbents and composite polymeric sorbents. Table 2.1 prese commercially available SPME coatings [3]. Talble SP sorbent* Polarity Coaing thickness (m) Maximum opg temp. (C) bilization Applications een the on commerc ial source of SPME fibers since 1993. poly nts a list of able 2.1 Commerci ly availa ME s [3] Sorbents t eratin Immo Polydimethy lsiloxane ane/ Carb dime (CAR/PDMS) Carb divinenzen (CW/DVB) Carbowax/tem resin (CW/TPR) Divin Carboxe (DVB olar 3 3 5 crosslinked Partially crosslinked crosslinked lar rganic compounds cs hydrocarbons, air analysis, VOCs VOCs, Polar organics, e.g., alcohols ctants (PDMS) Polyacrylate (PA) Polydimethylsilox Divinylbenzene PDMS/DVB)** ( oxen/Poly Bi-polar 75 320 Partially thylsiloxane crosslinked Hydrocarbons ow ylb ax/ Polar 65 26 e plated ylbenzene/ n/PDMS /CAR/PDMS) Polar Bi-polar 50 50/30 30/30 270 Partially Cross-linked Surfa Odors and flavors Non-p olar P Bi-polar 100 30 7 85 65 60 280 2 80 40 20 270 Nonbonded Nonbonded onded B Partially rosslinked c Partially Nonpolar or moderately po o olar organi P Aromatic Reproduced from Ref. [3] with permission of Wiley InterScience ** In SPME literature, Divinylbenzene (DVB) actually stands fo r poly (divinylbenzene)

PAGE 57

33 4] e PDMS le tion to vent for this ]. bent, immobilized by partial crosslinking, and is available in 85 m thic kness. Being highly polar, it is frequently s fr om different matrices [6]. Unlike other sorbent polyacrylate is a rigid, low-density solid polymer. As a conseq 2.2.1 Homogeneous polymeric sorbents Commercially available homogene ous polymeric sorbents include polydimethylsiloxane (PDMS) and Polyacrylat e (PA). Polydimethylsiloxane (PDMS) [ has been the most frequently used sorbent in SPME because of its inherent versatility, ruggedness, high thermal stability, and wide range of analytes that can be extracted using this phase. SUPELCO markets PDMS coated fi bers in two different forms: bonded and nonbonded. The bonded PDMS coating is crossli nked on the fiber using a cross-linkabl functionality present in the polymeric stru cture. This bonding translates into higher thermal stability (~340 C) as well as solvent stability [3]. The only commercially available bonded PDMS fiber is the one with 7 m coating thickness. The nonbonded coatings are immobilized either by thermalor by UV treatment and are availab in two thicknesses: 100 m and 30 m. Due to the absence of proper immobiliza the substrate, they possess relatively low therma l stability (280 C) as well as low sol resistance. As a nonpolar sorbent, PDMS ex tracts nonpolar analytes very well. PDMS coating can also be used to extract mo re polar compounds by optimizing extraction conditions such as pH, salt con centration, and temperature. One major drawback phase is that the fibers cannot be exposed to a matrix with a pH below 4 or above 10 [4 Polyacrylate (PA) [5] is a highly polar sor recommended for extracting polar analyte s, at room temperature, uence, diffusion of an alytes requires longer time re sulting in longer extraction

PAGE 58

34 Sorbents of this category are made by embedding one or more porous particulate us sorbents, these coating echanical stabilit y but very high selectivity. The following comme DVB is characterized by its high porosity as well as very high surface area (~ 750m2/g) [7]. The pores in the polymer play a k time. Moreover, higher temperatures are re quired for complete desorption of the extracted analytes. Better solvent stability and longer life span have made this coating very popular for the extraction of polar analytes. 2.2.2 Polymeric composite sorbents materials into a crosslinked polymer. Compared with homogeneo s possess lower m rcial composite coatings are cu rrently available from Supelco: Polydimethylsiloxane/Divi nylbenzene (PDMS/DVB), Car boxen/Polydimethylsiloxane (CAR/PDMS), Carbowax/Divinylbenzene (CW/DVB), Carbowax/Templated Resin (CW/TPR), Divinylbenzene/Carboxen/Pol ydimethylsiloxane (DVB/CAR/PDMS). 2.2.2.1 Polydimethylsiloxane/Divi nylbenzene (PDMS/DVB) This composite sorbent was introduced in 1996 [7]. ey role in retaining the analytes. DVB provides PDMS better ability to retain smaller analytes that cannot be retained by PDMS alone. Moreover, the polymer blend demonstrates better affinity for polar an alytes. PDMS/DVB coa ting is suitable for extracting polar compounds e.g., alcohols, nitr ogen containing analytes (e.g., amines), etc. [8].

PAGE 59

352.2.2.2 Carboxen/Polydimethylsiloxane (CAR/PDMS) This fiber was introduced into the market in 1997 [9]. Like DVB, Carboxen is also a highly porous polymeric mate rial having high surface area (~ 715m2/g) and pore volume (~ 0.78mL/g) [7]. Carboxen/PDMS provides high thermal stability (~320 C). Volatile organic compounds (VOCs) [10] and hydrocarbons [11] are among many types of analytes that have been extracted by SPME using CAR/PD MS coating thus far. 2.2.2.3 Carbowax/Divinylbenzene (CW/DVB) Carbowax/Divinylbenzene fiber was in troduced in 1997 [12]. Blending porous divinylbenzene with relatively polar carbow ax provides a composite phase that is frequently used for extracting polar organic compounds (e.g., alcohols) [13] from various matrices. Major shortcomings of this compos ite sorbent include th e swelling tendency of Carbowax in water as well as its high se nsitivity to oxygen at temperatures above 220 C, imposing limitation in its ability to extract fr om aqueous media and its effectiveness for low volatile compounds. There are several appr oaches that may reduce the problem to some extent. Maintaining the maximum inj ection port temperature in the range of 180240 C and using catalytic pur ifiers to obtain oxygen-free carrier gas may improve the lifetime of the fiber [7].

PAGE 60

36 n of polar nalytes. Like other composite sorbents, CW/TPR is made by blending porous templated AS 2), composed of both hydrophobic a nd hydrophilic monomers, is used as in rous polymeric particulate ma terial, contains predominantly esopores with some micropores and macropores. As a result, it effectively extracts 15. Carboxen, on the other erized by its evenly distribu ted micro-, mesoand macropores. It provides ange 2.2.2.4 Carbowax/Templated Resin (CW/TPR) This material [14] has been designed to perform preferential extractio a resin with Carbowax. A patented polym eric material SUPEL COGEL TPR-100 (C RN: 33972-38the porous constituent in the mixture. Due to the presence of both hydrophilic and hydrophobic moieties in the polymeric network, it provides remarkab le selectivity extraction. Extraction of surfactants from aqueous media is one of the many notable applications that have been possible by the virtue of CW/TPR sorbent [15]. 2.2.2.5 Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) This new blended extracting phase was introduced in 1999 by Supelco [16]. Divinylbenzene, a po m organic compounds with carbon atoms in the range of C 6 -C hand, is charact highly effective extractions for C 2 -C 12 organic compounds. Blending particles of these two highly porous polymeric materials w ith liquid polydimethyl siloxane yields a composite phase, which is recommended for extracting C 3 -C 20 analytes from a wide r of polarity [17, 18].

PAGE 61

37 tly. The fiber are ker coating may increase the sample capacity to some extent, long xtract ME not only shortens the tal analysis time but also provides bette r accuracy and precision compared to manual ny ga st ationa suie as in-tube SPME sorbent. In fact, pieces of coated capillary GC co lumns (in most cases, a 60 cm segment) are commonly used for in-tube SPME [20]. In this case, the stationary phase coating on the inner surface of the capill ary serves as the extracting phase. Table 2.2 lists frequently used extr acting phases in in-tube SPME. 2.3 Commercial GC coatings used in in-tube SPME In solid phase microextracti on, the fiber format has been used predominan fiber is installed in a specially design ed SPME syringe. Although the design of the syringe offers good protection to the fiber as well as the coating on its surface, breakage, mechanical damage of the coati ng due to scraping, and needle bending frequently witnessed by analytical chemists. The short length of the coated segment of the fiber provides low stationary phase load ing for extraction. As a consequence, low sample capacity of the fiber imposes limitation on the sensitivity of the SPME process. Although thic eion time may turn out to be the bottleneck in the whole process. Another major drawback of the fiber-SPME is the difficu lty to interface with other analytical instruments e.g., HPLC, CEC, MS, FTIR etc. To overcome these format related shortcomings, in-tube SPME (or capillary mi croextraction, CME) has been introduced [19]. In-tube SPME is suitable for automation. Automated SP to operation. In pri ciple, an s chromatographic ry phase should be table for us

PAGE 62

38Table 2.2 Comcial GCiecs of which are used for tube SPME ame of the endor hase Composition Chemical Structure Refer mer capillary columns short p e inN Phase V P ences B P-1 PB-1 B-Wax upelcowax megawax 50 P-20 Supel-Q-PLOT GE UPELCO ELCO W UPELCO UPELCO GE SUPELCO ne loxane ne ) 5% dimethyl siloxane) oly (ethylene glycol) oly (ethylene glycol) oly (ethylene glycol) oly (ethylene glycol) Porous divinyl benzene polymer DB-1 S PTE-5 SPB-5 D S O 2 B S JW S SUP SUPELCO J S S S 100% dimethyl polysiloxa 100% dimethyl polysi 100% dimethyl polysiloxa Poly (5% diphenyl / 95% dimethyl siloxane Poly (5% diphenyl / 9 P P P P Si CH3 CH 3 O n O Si O Si CH3 CH3 5%95% H OCH2 CH2 O H n CH2 CH2 n [3] [2] [2] [2] [2] [2] [2] [34] [31] [33] [29] 0 [26] [24] [25] 8 4 1 6 5 9

PAGE 63

39 Most of the sorbents used in in-tube SPME are commercially available GC stationary phases (e.g., DB-1, BP-1, SPB-1, SPB-5, PTE-5, Supelcowax, DB-5, Omegawax 250, DB-Wax, BP-20 Wax, Supel-Q-PLOT etc.). The first report of using in-tube SPME was made by Pawliszyn and co-wokers in 1997 [21]. In their experiments, 60 cm indi vidual pieces of different GC capillary columns (Omegawax 250, SPB-1, SPB-5, uncoated fused silica capillary) were used as the extractor and each was coupled to a comm ercial HPLC autosampler. Six phenylureas were extracted as the test analytes using each of the above mentioned capillaries. The Omegawax 250, being the most polar phase, ex tracted the most and the uncoated fused silica capillary extr acted the least. SPB-1 and SPB -5 coatings (nonpolar) also demonstrated poor extraction yield as was expected. Omegawax 250 GC capillary columns use pol y(ethylene glycol) as the stationary phase. It is the most freque ntly used sorbent in in-t ube SPME for extracting polar analytes. Thus far, Omegawax 250 has b een used for extrac ting phenylureas [21], blockers and metabolites in urine and serum samples [ 22], carbamate pesticides [23,24,25], mutagenic heterocyclic amines [26], ranitidine [19], and in drug analysis [27]. Although in both Supelcowax and Omegawax 250 coated capillaries, poly(ethylene glycol) was the common stationary phase, Omegawax 250 demonstr ated higher yield in extracting carbamate from aqueous solution compared to the ot her [25]. In almost all of the above mentioned in-tube SPME publicati ons SPB-1 (100% dimethyl polysiloxane), SPB-5 (poly 5% diphenyl/95% dimethyl silo xane), and uncoated fused silica capillary were used to study the importan ce of sorbent polarity in the extraction of polar analytes. The most polar among the used sorbents, Omegawax 250 was found to be most efficient.

PAGE 64

40 Takino et al. used similar phase DB-Wax (J&W Scientific, Folsom, CA, USA) for the determination of chlorinated phenoxy aci d herbicides in envi ronmental water [28] and successfully coupled SPME-LC with EI/ MS which provided enhanced selectivity and identification capability of the method. Tan et al. [29] have successfully coupled in-tube SPME to GC/FID. One-meter segments of BP-1 (100% methylsiloxane) a nd BP-20 (polyethylene glycol) GC columns were used for extraction of BTEX and phenols respectively from aqueous media. Extraction was carried out by pushing the aqueous medium containing the analytes through the capillary using nitr ogen pressure. Desorption of the extracted analytes was done by using a small plug of organic solven t which carried the analytes from the capillary to the GC injection port. Although SPME has been developed to elimin ate the use of toxic organic solvents in sample preconcentration step, a small amount of solvent is sti ll being used in the desorption process, particularly when coupl ed to LC. To reduce the amount of toxic organic solvent used in an alyte(s) desorption step after the extraction, Saito et al. [30] proposed a wire-in-tube configuration of SPME in which a 20-cm piece of DB-1 (100% polydimethylsiloxane) was used as the extract or. A stainless steel wire (diameter = 200 m) was inserted in to the capillary (dc= 250 m) that significantly reduced the available internal volume of the capillary (9.82 L vs 3.53 L) and thereby reduced the volume of solvent required for desorption. The wire-in-tube SPME was successfully used to analyze antidepressant drugs in a urine sample. Figure 2.3 demonstrates the schematic of a wirein-tube extraction capillary.

PAGE 65

41 Figure 2.3 Wire-in-tube SPME extraction capillary (Reproduced from Ref [30] with permission of Springer-Verlag)

PAGE 66

42 Another GC capillary column frequently used in in-t ube SPME is Supel-Q-Plot. A porous divinylbenzene polymer is used as the sttionary phase in this column. Mester et al. [31] successfully coupled SPME to electrospr ay ionization mass spectrometer. A 60 cm piece of Supel-Q-PLOT column was used to preconcentrate and analyze trimethyland triethyllead species from aqueous media. This system seems to be very promising in lead speciation. The same phase has also been used for the analysis of endocrine disruptors in liquid medicines and intravenous solutions [32] daidzein and genistein in soybean foods [33], bisphenol A, alkylphenols, and phthalate esters in foods contacted ith plastics [34]. .4 Tailor-made coatings on SPME fibers Table 2.3 presents a list of tailor-made sorbents coated on SPME fibers. mmobilized metallic SPME fibers, ac us sorbents, a sorbents, neous sorb a w 2 For discussion purpose, they have been gr ouped into different classes: i antibodies, tive carbonaceo bonded silic PDMS film, and miscella ents.

PAGE 67

43Table 2.3 Tailor-made sorbents on SPME fibers and their applications Sorbents Applications Reference Carbowax 2-Naphthol [1] Octadecyltrichlorosilane 2-Naphthol Liquid Crystalline Poyacrylate (LCPA) 2-Naphthol Polymethylvinylchlo rosilane 2-Naphthol HDEHP modified PDMS Bi (III) [56] PVC Activated Charcoal Fiber Coated on Silver Wire n-Alkanes [42] Graphite Surfactant [48] Adjusted Active Carbon Fiber BTEX [47] Graphitized Carbon Black VOCs [46] Porous-Layer Activated Charcoal BTEX [44] Activated Charcoal PAHs [45] Activated Charcoal-PVC Organophosphorus pesticides [43] Polycrystalline Graphites Nonionic surfactants [15] Pencil Lead Alcohols [49] PDMS PAHs, PCBs [4] Anodized Aluminum Wire Alcohols, BTEX, Petroleum products [37] Modified Copper Wire Primary amines [41] Anodized Zinc Wire VOCs [40] Copper Sulfide Wire Aliphatic alcohols and amines [40] Gold Coating Mercury (II) ions [39] Inorganic/Organic Mesoporous Silica Aromatic hydrocarbons [53] Porous Layer Silica Bonded Phases Aromatic Hydrocarbons [51] C18-bonded Silica Pesticides [52] Silica Particles bonded with Phenyl, C8, C18 PAHs, PCBs [51] Alkyl-Diol-Silica RestrictedAccess Material Benzodiazepines in biological samples Metabolites [69] [70] Immobilized Theophylline Antiserum Theophylline in human serum [35] Poly(pyrrole-sulfated cyclodextrin) Cationic analytes [75] Poly(3-methylthiophene) Arsenate ions [74] Salicylate PVC-based Membrane Organonickel complex [79] Polyimide VOCs [107]

PAGE 68

44 Dibenzo-18-crown-6 (DBC) Hg(II) ions [59] Polysilicone Fullerene (PF) Semi-volatiles [62] Nafion Perfluorinated Resin Polar solvents [57] Aqueous ammonium [58] Cellulose Acetate and PVC Higher alkanes [38] Phenol-based Polymer, BSP3 Ch emical warfare agents [61] Molecularly Imprinted Polymers Clenbuterol and its structural analogues [67] 2.4.1 Immobilzed antibodies The ability to selectively recognize a targ et analyte at the molecular level and extract it from a given matrix is an importa nt aspect in any chemical and biochemical analysis. Stability and commercial availability are two important factors for a molecular recognition compound being considered as a new extraction medium. Keeping these in mind, antibodies are the major candidates that have excellent potenti al to be used as molecular recognition-based ex traction media in SPME. Anti bodies possess high affinity and specific recognition ability toward th eir complementary antigens in biological systems. Antibodies contain several types of r eactive groups that are suitable for covalent bonding with silica supports. Figure 2.4 repres ents a typical reaction scheme involved in antibody immobilization [35]. The credit for the first exploitation of immobilized antibody as SPME sorbent goes to Yuan et al. [35].They reportedly immobilized theophylline antiserum on the surface of a fu sed silica fiber and successfully extracted theophylline from human serum. The new tec hnique, immunoaffinity SPME, has brought new hopes for the determination of drugs in complex biomatrices th at require very high

PAGE 69

45 specificity which, in general, cannot be achieved by existing SPME sorbents.

PAGE 70

46 SiOHOHOHSiOHOHOSiOO(CH3)3NHHSiORORRO(CH2)3NH2HOCHOCSiOHOHOSiOO(CH3)3NCH(CH2)3COHHNHAntibody SiOHOHOSiOO(CH3)3NC(CH2)3CNAntibodySupport Figure 2.4 Reactions involved in antibody immobilization (a) silanization of silica HHSupport+(a)SupportAPTESR= CH2CH3+Glutaraldehyde(b)Support+(c)rface with APTES; (b) Surface modification with glutaraldehyde; (c) immobilization of ted from Ref. [35] with permission of the Royal Society of Chemistry) su antibody (Adap

PAGE 71

47 N.A. Guzman [36] fabricated an improve d SPME device with i mmobilized antibody and coupled it on-line with i mmunoaffinity capilla ry electrophoresis. In the specially designed SPME device, antigen binding fr agments (Fab’) obtained from purified immunoglobulin G (IgG) antibody were immobilized by covalent bonding and were used as the SPME sorbent. Low sample volume, minimal sample preparation, high reproducibility, long lifespan, a nd at least 1000-fold sample preconcentration have been reported to be the major advantageous features of the new device. 2.4.2 Metallic SPME fibers One of the major drawbacks of SPME fiber is its fragility. Almost all SPME fibers are made of fused silica. The sorbent co ated end of the silica fiber which is devoid of protective polyimide coating is very fragile and needs great care in handling. Therefore, developing relative ly strong SPME fibers with long life span is an important issue for the researchers working in this field. Some researchers have proposed miniaturized metallic rods as an alternative to easily breakable silica-based SPME fibers [37, 38]. In some cases, the metal itself and in other cases an oxide layer formed on the meta l surface has served as the SPME sorbent. Guo et al. [39] has successfully combined SPME with electrochemistry. An electrochemically deposited 10 m thick gold coating on a 140 m carbon steel electrode was used as the electrochemical extraction me dium. The gold coating was used to detect inorganic mercury ions in so lutions using ion-trap GC-MS. It was also able to

PAGE 72

48 differentiate between free i norganic mercury and complexed mercury present in aqueous matrix. Moreover, the possibility of direct detection of metallic mercury in gas by gold coated SPME/EC has made the system very prom ising due to the fact that high toxicity of mercury poses a real threat to the environment. In an attempt to develop rela tively strong SPME fiber, Djozan et al.[37] developed anodized aluminum wire. They eval uated different types of aluminum-based wires: (a) polished aluminum wire (to prevent the formation of Al2O3 on the surface), (b) aluminum wire with oxidation product (Al2O3) on the surface, and (c) anodized aluminum wire as the extr action sorbent. Those res earchers found that anodized aluminum wire provided the hi ghest extraction efficiency ( 30 times more sensitive than the oxidized aluminum wire). The dramatic increase in sensitivity was explained by adsorptive nature of the thick (~ 20 m) Al2O3 bed which was formed on the aluminum surface during the anodizing process and also the inherent porous structure of the aluminum oxide bed on the surface. Figure 2. 5 illustrates the porous surface of the anodized aluminum wire [37]. Anodized alum inum wire has been used to extract aliphatic alcohols, BTEX and some petroleum products from gaseous samples. Low cost, high thermal stability (~300 C), mechanical strength, and long life span are among the important advantages of a nodized aluminum fibers.

PAGE 73

49 Figure 2.5 Scanning electron micrograph of the surface of an anodized aluminum wire (Reproduced from Ref. [37] with permission of American Chemical Society)

PAGE 74

50 In developing new sorbents for SPME, it is desirable to have a porous material with large surface area and high sorption cap acity as well as increased selectivity, sensitivity, durability, and reproducibility. Djozan et al. [40] developed a new fiber consisting of copper wire coated w ith a copper sulfide coating (~ 9 m) and efficiently extracted benzyl alcohol, benzyl amine, alip hatic amines and alipha tic alcohols directly from aqueous samples without any derivatiza tion process. The new fiber appears to be highly selective for aliphatic amines and alcohols. Low molecular mass aliphatic amines are im portant intermediates in chemical and pharmaceutical industries. They may form as biodegradation pr oducts of organic materials like proteins, amino acids or other nitrogen-containing compounds and are distributed in environmental water. Low molecular mass aliphatic amines are also important air pollutants due to their unpleasant odor and toxicity. The monitoring of alkylamines is of considerable interest as most of them are toxic, se nsitizers and irritants to the skin, mucous membrane, and respirator y tract. Analysis of aliphatic amines in aqueous media is very difficult because of their high polarity. Preconcentration and derivatization before chromat ographic analysis is a common practice in amine analysis. Djozan et al. [41] developed a modified copper wire (coated with a thin layer of microcrystalline CuCl) as an SPME fiber which showed high selectivity towards aliphatic amines. CuCl is very reactive because it has normally two coordinate bond structures and can accept third and fourth coordination by other ligands such as amino group to form an amino complex according to the reaction: CuCl + n (amine) = Cu(amine)nCl. At high temperatures (during desorption in the GC injection port), the amino complex breaks up to release free amine.

PAGE 75

51 Being easy to produce, considerably ad sorptive, strong enough to endure harsh operational conditions, and cheap, zinc oxide finds its application in SPME as a novel sorbent. Djozan et al [40] prepared this sorbent by anodizing zinc wire. This fiber demonstrated good affinity toward pola r analytes. Among the sulfur compounds extracted, the zinc fiber seems to have the highest sensitivity for thiophenol probably because of the presence of SH functional group in the analyte and is likely to interact selectively with ZnO coating. Farajzadeh et al. [38] reported a metallic SMPE fibe r coated with cellulose acetate and polyvinyl chloride (PVC). The fiber dem onstrated high affinity toward n-alkanes (C10-C20). According to the inventors, this porou s fiber possesses high chemical, thermal, and mechanical stability. 2.4.3 Active carbonaceous sorbents Active carbonaceous sorbents have been proven to be excellent extraction media due to their positive attributes favorable for SPME. They are homogeneous, highly porous, stable at high temperatur es, do not retain water and us ually they do not manifest irreversible adsorption phenomena. Farajzadeh et al. [42] have developed a new sorb ent material by mixing activated charcoal with PVC powder in different proportions and coated on a metal wire. The resultant coating, as the developer clai med, was very firm and porous. A coating composition of 90:10 activated charcoal: PVC de monstrated high effici ency in extracting light alkanes whereas a coating with higher PVC content (25%) showed promise for a

PAGE 76

52 mixture of both heavy and light alka nes. In a more recent article [43], the same group successfully extracted organophosphorus pesticid es from aqueous media but this time a 70:30 activated charcoal: PVC ra tio was found to be optimum. In an attempt to develop a sorbent su itable for extracting volatile organic compounds (VOCs), Djozan et al. [44] reported a porous layer activated charcoal (PLAC) coated (~100 m thickness) fiber. The new material was used to extract BTEX from aqueous media and showed significant se nsitivity enhancement (at least 2 orders of magnitude) in comparison with the reported values for commercial PDMS fiber. Moreover, it provided excelle nt thermal stability (320 C). In an earlier article [45], the same group utilized porous layer activated charco al (PLAC) coated (55 m coating thickness) fiber to extract pol ycyclic aromatic hydrocarbons. Fo r this fiber, the authors had reported a significantly higher temperature (350 C) for thermal desorption of the extracted analytes. A fused silica fiber coated with graphitized carbon black (GCB) was reported by Mangani et al. [46]. The developers of this coating ma terial found it to be suitable for extracting VOCs from different matrices. Jia et al. [47] reported the use of active carbon fiber (ACF) as a sorbent in SPME. An ACF fiber was coated with a unique ex traction phase possessing characteristic pore size distribution, micro pore surface area, surf ace chemical structure, and so on. It was claimed to be an excellent adsorbent for the removal of SO2 and NOx in flue gas from coal combustion. The specific surface area a nd adsorption capacity was dependent on the pre-treatment and activating c onditions. Two different activat ion processes (chemical and water vapor) were used. Water vapor activated fibers provid ed higher affinity to polar

PAGE 77

53 compounds. On the other hand, chemically activ ated fibers showed higher affinity for nonpolar compounds. As the authors predicted, it is possible to fine tune the affinities to different compounds ranging from nonpolar to polar by control ling the activation process and even by combining both the chemical and water activation processes. Kuo et al. [48] coupled SPME to direct elect rospray probe (DEP) and detected trace surfactants from aqueous media. In the SPME-DEP assembly, a 2 cm long 0.3 mm o.d. graphite fiber that acts as the sorbent is inserted into a 1.5 cm long copper coil. A 10-9 M solution of Triton X-100 was used as a test sample which was successfully detected after being extrac ted on the graphite fiber by SPME-DEP-MS even though the same solution did not produced any re sponse when only DEP-MS was used. Glassy carbon, a polycrystalline graphite has found useful applications as a stationary phase in HPLC and SFC because of its high chemical and mechanical stability as well as large surface area. Aranda et al. [15] utilized glassy carbon rods and pencil lead (both are polycrystalline graphite) as sorbents for SPME. The performances of both the sorbents were assessed by extracti ng a nonionic surfactant (Triton X-100). These sorbents seem promising when on-fiber deri vatization is required prior to detection because the derivatizing chem icals are easily adsorbed on the surface of the sorbent. Tong et al. [49] used the same pencil lead as SP ME sorbent and extracted alcohols (e.g., methanol, t-butyl alcohol, is amyl alcohol, n-propane, etc.) from water. A detection limit of 5 ng/mL was achieved for isoamyl alcohol in GC-FID.

PAGE 78

542.4.4 Bonded silica sorbents Slow analyte diffusion into the extraction media is considered to be a bottleneck phenomenon in SPME process. This problem becomes even more prominent in case of using thick coatings. To achie ve high extraction sensitivit y, it is desirable to employ thicker coating, however, use of thicker coatings leads to longer extraction time. In order to achieve higher extraction efficiency as well as faster extraction equilibrium, Lee and co-workers [50] have explored the possibility of using silica bonded phase as an SPME sorbent. Porous silica particles provided hi gh specific surface area which led to high extraction sensitivity. A 30 m thick coating of the bonded silica particles (C8-, C18-) was immobilized on a metal wire surface us ing high-temperature epoxy glue and was compared with untreated silica immobilized in the same way. The surface area of the bonded silica coating was found to be 500 times greater than that of the conventional polymer coating in SPME (0.97 x 10-2 m2 vs. 1.41 x 10-5 m2). As a result, extraction sensitivity was found to increase sharply.

PAGE 79

55 Figure 2.6 Scanning electron micrographs of SPME fibers with: (A) porous layer bonded silic a coating and (B) PDMS coating (Reproduced from Ref. [50] with permission of American Chemical Society)

PAGE 80

56 In contrast with the diffusion controlled mass-transfer process involved in sample enrichment on conventional polymer coatings extraction on bonded si lica coatings was found to be driven by the mass transfer in the bulk soluti on. As a result, extraction primarily depended on the degree of sample agit ation, not on the film thickness. In ideal conditions, adsorption equilibrium could be at tained in few seconds. Desorption should also be faster. The 30 m thick bonded silica coatings de monstrated higher extraction efficiency compared to a 100 m PDMS coating. For instance, C8bonded phase extracted ~40 ng of toluene from a 0.1 ppm aqueous solution of toluene whereas PDMS coating extracted only ~ 5 ng of toluene unde r the same conditions. Due to the use of a metal wire as a substrate, fibers of this kind exhibited great mechanical strength. One of the major drawbacks of this fiber is that at elevated temperatures (300 C) the epoxy glue starts degrading and thus lim iting the applications only up to 250 C. As a continuing effort, the same group reported the application of bonded silica phase (octyl, phenyl, monomeric and polymeric octadecyl) for extracting polyaromatic hydrocarbons from aqueous solution [51]. Among all the bonded phases investigated, C18bonded silica phase showed the highest sensitivity toward PAHs. In order to provide larger sorption ca pacity and higher sorption rate, Xia et al. [52] proposed a C18-bonded silica-coated multifiber that pr ovided 10 times faster sorption rate compared to commercial 100 m PDMS fiber. In terms of analyte preconcentration ability, C18-bonded silica coated fiber demonstrated 40 times more extraction sensitivity compared to PDMS coating. Hou et al. [53] introduced a mesoporous silica material (C16-MCM-41) as an SPME sorbent. The mesoprous material has been characterized by its la rge surface area (1028

PAGE 81

57 m2g-1) and a comparatively huge pore volume (0.94 cm3g-1 ). A 100 m thick coating was immobilized onto a stainless steel wire by e poxy resin glue and was further employed for extracting aromatic hydrocarbons. Results indica ted that the new material had very high extraction efficiency and good se lectivity. Compared to traditi onal silica-based fiber, high mechanical strength, and good chemical and thermal stability were reported to be advantageous features of this material th at could potentially become a popular SPME sorbent. 2.4.5 Flat sheet membranes In order to increase the extraction effici ency and detection limit without sacrificing analysis time, a new format of SPME named thin-film microextraction has been introduced [54]. In this new format, inst ead of using a polymer coated fiber, a thin sheet (1cm x 1 cm x 0.0025 cm) of poly(dim ethylsiloxane) membrane was used as the extraction phase. Having very high surface area (200 mm2) compared to 100 m thick coated fiber (10 mm2), this membrane has extracted 20 times more PAHs compared to 100 m thick PDMS coated fiber. The be nefit of membrane SPME is not only the enhancement in sorbent surface area but also in the sorbent volume (2.55 mm3 vs 0.061 mm3, the volume of 100 m PDMS coating). Both the factors lead to enhanced sensitivity without compromising analysis time. Figure 2.7 illustrates a typical thin-film microextraction process.

PAGE 82

58 Figure 2.7 Drawing of the headspace membrane SPME system. 1. Deactivated stainless steel rod. 2. Flat sheet membrane. 3. Sample solution. 4. Teflon-coated stir bar. 5. Rolled membrane. 6. Injector nut. 7. Rolled membrane. 8. Glass liner. 9. Capillary column (Reproduced from Ref. [54] with permission of American Chemical Society)

PAGE 83

59 rate IR. ion nated resin as a coating on SPME fibe le fibers, Nafion performed better in extracting small alco ost allenge f designed a special SPME device that after doping with crown ether wou Merschman et al. [55] used a rectangular sorben t film (PDMS) to preconcent organic analytes from aqueous media. They successfully interfaced membrane SPME with infrared spectroscopy and determined th e extracted analytes were analyzed by 2.4.6 Miscellaneous sorbents Although SPME has been introduced prim arily for the preconcentration and analysis of organic compounds, it can be easil y applied to the broad field of metal analysis by simply modifying the sorbent. Otu et al. [56] reported the solid phase microextraction of metal ions by modifying PDMS coating. A liqui d ion-exchanger, di(2-ethylhexyl)phosphoric acid (HDEHP) was us ed for the modification. The modified PDMS coated fiber was used to extract bismuth (III) ion from an aqueous sample. Gorecki et al. [57] introduced Nafion perfluori r for the analysis of polar compounds in liquid matrices. Results showed that compared to commercially availab hols. Surprisingly, Nafion demonstrated highest affinity toward methanol, the m polar compounds among the tested materials. Moreover, Nafion coating successfully extracted trace amount of water from organic solvents which might solve a big ch that analytical chemists are still coping these days. Norlin et al. [58] also used Nafion coated fiber for nitrogen isotopic analysis o ammonium in aqueous solutions. Jia et al. [59] ld extract metal ions from aqueous solu tion. He used a piece of fused silica rod,

PAGE 84

60 eccentrically glued to a hydrophobic hollowfiber membrane and then doped the membrane with dibenzo-18-crown-6 (D BC). The SPME device doped with DBC was used for extracting Hg (II) ions that we re subsequently analyzed by HPLC-UV. A detection limit of 500 ppb has been estimated although UV detector is characterized by low sensitivity. Li et al. [60] used plasticized poly(vinylchloride) as an SPME sorbent coated on a primed steel rod housed in a syringe-like SPM E device. The device was used to extract barbiturates from urine and bovine serum samples and was analyzed by CE. For urine and serum samples, the concentration ranges th at could be extracted and analyzed by the method were 0.1-0.3 ppm ~1 ppm, respectively. To detect trace levels of ch emical warfare agents, Harvey et al. [61] developed a novel SPME coating BSP3 on SPME fibe r consisting of hydrogenbond acidic hexafluorobisphenol groups alte rnating with oligo(dimethyl siloxane) segments. The new coating demonstrated remarkable affinity to ward sarin, a nurve ga s. In comparison to commercial PDMS fiber, the BSP3 coated fi ber showed 22-fold higher affinity towards sarin. Fullerene has long been known as a chromat ographic material for its high thermal stability (~360 C), and good selectivity towards aromatic compounds. C. Xiao et al.[62] developed a polysilicone fullerene (PF) coat ing (33 m) for SPME. They compared the efficiency of two coatings, one made of pure PF and the other made of a mixture of PDMS and PF (4:1 ratio). Coating obtained me rely from PF showed better sensitivity toward the test analytes BTEX and PAHs, rather than the mixed one. A detection limit of 0.04g.L-1 was obtained for naphtha lene using GC-FID.

PAGE 85

61.5 Tailor-made coatings for in-tube SPME ade in-tube SPME coatings can be further subdivided into two types inantly in in-tube SPME Several tailor made SPME sorbents have found successful use in both fiberand -tube SPME although their use in the in-tube format predominated. Major classes in is category are: molecularly imprinted polymers (MIPs), restri cted access materials AMs), conductive polymers and mi scellaneous coatings. Table 2.4 presents a list of oatings used in the in-tube SPME format. 2 Tailor-m depending on the exclusivity of their use in in-tube SPME: (1) tailor-made coatings used predominantly in in-tube SPME, and (2) tailor-made SPME coatings developed for exclusive use in in-tube SPME. 2.5.1 Tailor-made coatings used predom in th (R c

PAGE 86

62Table 2.4 Tailor-made sorbents used in in-tube SPME Sorbents Application Analytical Instrument Reference Alkyldiol Silica Restricted Access Material Benzodiazepines from human serum HPLC-UV [70] Molecularly Imprinted Polymer Propranolol HPLC-UV [64] PDMS Amides, alkanes, PAHs, chlorinated solvents BTEX GC-FID GC-FID [98] [101] Polypyrrole Polar pesticides HPLC-ESI-MS [90] N-Nitrosoamines HPLC-UV [92] Verapamil drug LC-MS [91] In Vivo pharmacokinetic studies LC-tandem MS [93] Aromatic compounds HPLC-UV [87] Inorganic anions IC [83] Organarsenic compounds LC-ESI-MS [82] -Blockers in urine and serum samples LC-ESI-MS [85] Catechins, caffeine in tea HPLC-ES-MS [86] Stimulants in human urine and hair LC-ES-MS [89] Tributyltin HPLC-ES-MS [85] Poly-N-phenylpyrrole Aromatics, PAHs, aromatic amines, organoarsenic compounds HPLC-UV [82] Carbon deposited inside capillary VOCs GC [102] Zylon fiber packed capilary Phthalate HPLC-UV [30]

PAGE 87

632.5.1.1 Molecularly imprinted polymers (MIPs) Although natural antibodies ha ve long been used in molecular recognition but their high price, low availability, and short lifetime have led researchers to synthesize antibody mimics such as molecularly imprin ted polymers (MIPs). Molecularly imprinted polymers are crosslinked macromolecules with specific binding sites for a particular analyte. In order to obtain a highly select ive recognition of a certain molecule, this template molecule is incorporated in the mixture of reacting mono mers during synthesis of the molecularly imprinted polymer networ k. After completion of the synthesis, the template molecule is extracted out leaving in the polymer a three dimensional imprint of itself. Figure 2.8 illustrates different steps i nvolved in molecular imprinting process [63]. Mullett et al. [64] reported the application of a mo lecularly imprinted polymer in SPME for selective extraction of propranolol (a -blocker compound) from biological fluids. The propranolol-imprinted polymer particle s were packed in a 80-mm long PEEK tubing (1.59-mm o.d. x 0.76-mm i.d.) and both the ends were capped with a zero volume union fitted with 2-m frit. An autosampler was used to ex tract target analyte and the extracted analyte were analyzed by HPLC -UV. A detection limit of 0.32 g/mL in spiked serum sample with excellent repr oducibility (RSD, <5%) was ach ieved in in-tube SPME using MIP-based sorbent.

PAGE 88

64 Figure 2.8 Different steps involved in MIP synthesis (Reproduced from Ref. [63] with permission of Elsevier)

PAGE 89

65 The same research group [65,66] has also reported online sample preparation of verapamil and its metabolites by an MIP materi al coupled on-line to a restricted access material (RAM) precolumn. The RAM preco lumn helped in removing bulky matrix interferences (e.g., proteins). On the othe r hand, MIP provided se lective extraction of verapamil from the relatively clean matrix obtained after the by RAM precolumn. Verapamil-imprinted polymers were packed in a 40 mm stainless steel column (4 mm i.d.) and capped with a 2 m frit. A detection limit of 5 ng/mL was obtained in LC-MS analysis. Another significan t contribution in MIP-SPME ha s been reported by Koster et al. [67]. A silica fiber was coated with a 75-m thick methacrylate polymer imprinted with clenbuterol. To compare the viability of the imprinting process, another fiber was coated with the same thickness of methacry late but this time no imprinting was used. Experimental results indicated that the fiber with the imprinted polymer coating successfully extracted clenbuterol and its stru ctural analogues with 75 % extraction yield, whereas nonimprinted fiber extracted only with ~5% extraction yield which clearly demonstrates the enhanced selectivity offe red by an imprinted polymer in SPME. The extractability of different stru ctural analogs of cl enbuterol was also in vestigated. It was found that compounds having almost same spat ial arrangement (e.g., bromobuterol) were extracted with higher extraction yield co mpared to compounds having structural differences.

PAGE 90

662.5.1.2 Restricted ac cess materials (RAMs) Restricted Access Materials (RAM) were originally developed for the isolation of low molecular mass drugs from biological fluids with minimum sample treatment. Recently, it has found applicati on in the isolation of herb icides from surface water containing high levels of humic substances [68]. Restricted access materials prevent access of macromolecules (interferences) to the specific sorbent region where analyte retention occurs. Mullett et al. [69] introduced a biocompatible SP ME fiber coated with alkyl-diolsilica (ADS) restricted access material that successfully extracted several benzodiazepines in presence of protein. The aim was to minimize the sample preparation time by eliminating protein precipitation step and therefore, redu ce the probability of sample contamination and analyte loss. The sorbent alkyl-diol-silica (ADS) possesses two different chemical surfaces (diol groups on the outer surface, and alkyl groups on the inner surface) and a pore size that prevents larger molecules (e.g., proteins) from entering into the inner surface. Hydrophilic electroneut ral diol groups on the outer surface of the spherical ADS particles acts like a filter to trap bulky molecules e.g., proteins, whereas hydrophobic alkyl groups extract rela tively smaller target analytes that easily penetrates the outer ADS surface. The ADS-RAM fractiona ted the protein present in the matrix by preventing its access to the adsorption site s, thereby allowing the low molecular mass analytes to be easily extracted and enriched into the interior of the sorbent. Calibration curve constructed for five benz odiazepines over a range 0.550 g/mL demonstrated excellent linearity. Detection limits ranging from 46 to750 ng/mL for different

PAGE 91

67 nd analyzing y HPLC-UV a detection limit of 22-29 ng/mL obtained for different benzodiazepines. Recently, the same group utilized a RAM with ion-exchange capability for onitoring drugs and metabolites in a whole blood sample [71]. A 70-mm piece of steel ire was used as the substrate on the surface of which ADS particles were glued. Figure .8 presents an image of the ADS partic les immobilized on a SPME fiber. benzodiazepines have been achieved in HPLC-UV. Using the same coating, Mullett et al. [70] prepared a highly bio-compatible SPME capillary for the automated and di rect in-tube extr action of several benzodiazepines from human serum. In pr eparing the SPME capillary, a 50 mm PEEK tubing (1.59 mm o.d. X 0.76 mm i.d.) was used and the particles were slurry packed into the capillary. After extracting several benzodi azepines from serum samples a b m w 2

PAGE 92

Figure 2.9 SEM image of ADS particles immobilized on a fused silica fiber (Reproduced from Ref. [71] with permission of Elsevier) 68

PAGE 93

69 The extracted drugs and metabolites were then analyzed by LC-MS. An extensive review on biocompatible coatings (antibodi es, MIPs, RAMs) for SPME by Mullett et al. [72] would provide the readers with further detail s for a better understanding on the topic. 2.5.1.3 Conductive polymers A major setback SPME had experienced in its first decade was the absence of coatings suitable for extraction of ionic an alytes. Because of their neutral charge, conventional commercial coatings are ofte n unsuitable for extracting charged species. Chemical modification of the analytes by de rivatization or by the addition of complexing agents (e.g., crown ethers) ar e the frequently used proce dure to increase extraction efficiency for ionic species [73]. However, these reactions are time consuming and require expensive and toxic reagents. Be sides, completeness of the derivatization reaction in dilute samples can be problematic. In an attempt to solve these problems, conducting polymers have been explored by several research groups. Conducting polymers are versatile materials that possess many positive attributes, and can be efficiently exploited for the advancement of separation science. So far, several conductive polymers have been reported as SPME sorbents among which poly-3methylthiophene, polypyrrole and poly-N -phenylpyrrole ar e noteworthy. The electroactivity and reversible redox properties of these condu ctive polymers have made them suitable extracting phases in electroc hemically controlled delivery devices and separation systems for charged species [74]. Th e extraction of ionic species is realized by

PAGE 94

70 applying a positive/negative potential for a predetermined time. After extraction, the extracted ions are released in a small volume of 0.1 M NaCl solution by applying a negative/positive potential pulse. Table 2.5 lists coatings us ed in electrochemically controlled SPME. Table 2.5 Coating used in electroc hemically controlled SPME Sorbents Application Reference Polypyrrole Aroma in wines [94] Chloride solution [77] Anions, cation [78] Ionic analytes [79] Poly(3-dodecylthiophene) P3DDT Organometallic arsenobetaine [76] Poly(pyrrole-sulfated cyclodextrin) Cationic analytes [75]

PAGE 95

71 T.P. Gbatu et al. [74] reported poly(3-methylthiophe ne) as SPME sorbent to extract ionic arsenic compounds from aque ous media without de rivatization. A ~5m thick coating of poly(3-methylthiophene) was electrochemically deposited onto a platinum fiber. During the extraction, a pos itive charge was maintained on the fiber by applying an electrode potential (+ 1.2 V). At the time of desorption, the polymer was converted back to its neutral state by reve rsing the potential (-0.6 V) and extracted arsenate ions were expelled into a smaller volume of deionized water. Finally the ionic samples were analyzed by ICP-MS. Temsamani et al. [75] investigated the cation uptake and release properties of a poly(pyrrole-sulfated -cyclodextrin) film electrode for prospective application in electrochemically aided solid phase microextract ion. The experimental data revealed that by changing electrode potential, cations like K+, Na+ can be extracted and desorbed efficiently from the electrode. The same group [76] utilized poly(3-dodecylthiophene) for the preconcentration of organometallic arse nobetaine (AsB) from a queous media. Both adsorption and desorption were done elec trochemically. Extraction equilibrium was achieved within a minute. This result shows good prospect in extraction and analysis of other organometallics that still represents challenge for researchers. Liljegren et al. [77] used microband array gold el ectrodes coated with polypyrrole to preconcentrate and detect chloride ions in a capillary based flow system. The results indicate that polypyrrole coated electrodes can be effectively used in miniaturized analytical flow system for analyte preconcen tration. The same group [78] also described a method for the extraction, transfer, a nd desorption of anions and cations under controlled potential using an integrated thre e-electrode device wh ich included a glassy

PAGE 96

72 carbon rod, a silver tube and a stainless steel tube with polyp yrrole coating. The results demonstrated that by changing the dopant i on (perchlorate for an ion extraction and ptoluenesulfonic acid for cation extractio n) during the electro polymerization of polypyrrole on the stainless steel, simultaneous extraction of several anions or cation can be accomplished with high extraction efficiency. In another report, Wu et al. [79] proposed similar el ectrochemically controlled SPME. Exploiting conductive nature of pol ypyrrole film and by applying positive and negative potential, anions and cations can be extracted and ther eafter the extracted cations can be desorbed just by reversing the potential. A salicylate PVC-based membrane (as SPM E sorbent) was coupled to coated graphite membrane electrodes by Ganjali et al. [80] to facilitate selective determination of trace amount of salicylate from biological sample. The new sensor could determine as low as 0.01nM salicylate ion from serum samp le compared to the best reported sensor with a minimum detectable c oncentration limit of 0.1 nM. Another important contribution was the introduction of polypyrrole SPMEcoatings by Pawliszyn a nd co-workers [81]. Among various conducting polymers studied, polypyrrole and its deri vatives have drawn much atte ntion due to its commercial availability, ability to form stable polymeric films by chemical or electrochemical means on various substrate materials ranging from metals and nonmetals. Due to the inherent multifunctional properties of polypyrrole coatings (e.g., interactions, acid-base interactions, hydrogen bonding, di pole-dipole interactions fr om polar functional groups, hydrophobic interactions between the sorbent an d the analytes), they can be used for extracting a wide array of analytes. One impor tant aspect of this polymer is that its

PAGE 97

73 of s ail gure 2.9 presents SEM images of PPY and PPPY coated surface on a metal wire. Wu et al. [83] utilized polypyrrole-coated capillary for speciation of organoarsenic compounds and coupled in-tube SPME with liquid chromagraphy/electrospray ionization mass spectrometry. For comparison, the same extraction was carried out in four commerc ial capillaries (Supel -Q, Omegawax, SPB-1, SPB-5) and an uncoated fused silica capillary. As was expected, the PPY coated capillary provided the best extraction e fficiency. The main driving fo rces for extraction for the above mntioned commercial coatings are hydr ophobic interactions and the interactions from polar functional groups. On the other ha nd, the driving forces for extraction on PPY coating are not limited by only hydrophobic and functional groups interaction, but selectivity can easily be tuned by introducing additional functional groups into the polymer structure. Polypyrrole and its deriva tives can be easily co ated on the surface metal wires by electrochemical deposition as well as the inner walls of fused silica capillaries via chemical polymerization. As a result, polypyrrole and its derivatives have been used as SPME sorbents in both fibe r and in-tube format. Polypyrrole and it derivative poly-N-phenylpy rrole were introduced as SPME sorbents in 1999 by Pawliszyn and his co-workers where polypyrrole (PPY) and poly-Nphenylpyrrole(PPPY) were electrochemically coated on the surface of metal wires and utilized for extracting volatile organi c compounds from aqueous solutions [81]. A det accounts on the preparation of PPY a nd PPPY film on metal fibers (e.g., Pt, Au, stainless steel) as well as the coating insi de the capillary was presented by Wu et al. [82]. This article also highlighted a wide range of applications of th e PPY film and coating which included polar and nonpolar aromatics, polycyclic aromatic hydrocarbons, aromatic amines, organoarsenic compounds. Fi to e

PAGE 98

74 they also include electrostatic interactions that occur between the charged analytes and the positively charged PPY coating.

PAGE 99

75 Figure 2.10 Scanning electron micrographs of (A) Uncoated metal surface, (B) PPY-coated metal surface, (C) PPPY-coated metal surface (Reproduced from Ref [83] with permission of Elsevier)

PAGE 100

76 In a different publication [84], the same group reported the ex traction of various anions (e.g., fluoride, chloride, nitrite, phos phate, sulfate, selenite selenate, arsenate) from aqueous samples and analyzed them by ion chromatography using polypyrrole coated capillaries. Although polypyrrole seem s to be a good extraction medium for ionic compounds, but a closer look at the polymeric structure reveal s that every three or four polypyrrole units contain one pos itive charge that limits th e extraction capacity of the polymer. That problem, howev er, can be easily minimized by incorporating cationic substituents into the polypyrrole sk eleton. In a different report Wu et al. [85] employed polypyrrole coated capillary for speciation and determination of tributyltin from aqueous media. Analysis was done by HPLC-ES-MS. PPY is not only an efficient extraction medium for ionic analytes, it also efficiently extracts neutral analytes by utilizing its diverse modes of molecular le vel interactions. As a result polypyrrole coating has found application in the extraction of a wide range of neutral analytes including -blockers in urine and serum samples [86], catechin and caffeine in tea [87], aromatic compounds in aqueous samples [88], stimulants in human urine and hair samples [89], polar pesticides in water and wine samples [90], verapa mil drug and its metabolites [91], and Nnitrosamines in cell cultures [92]. Recently a polypyrrole coated SPME probe has been introduced by Lord et al. [93] for in vivo pharmacokinetic stud ies in living animals. The same group has also combined SPME with su rface enhanced laser desorption/ionization and ion mobility mass spectrometry (SPM E-SELDI-IMS) using a polypyrrole coated optical fiber [94]. The tip of the fiber acts as the sorben t as well as the transmission medium for the UV laser light. Guadarrama et al. [95] invented an SPME device consisting of a set of 12

PAGE 101

77 polymeric sensors coated with of polypyrro le, poly-3-methylthiophene and polyaniline. The device served as an artifi cial olfactory system to iden tify wine aroma. The device provided a noticeable difference in respons e toward different wines which can be achieved by an automatic headspace sampler. 2.5.2 Coatings exclusively used in in-tube SPME In order to be able to facilitate effect ive sample introduction into an analytical instrument, SPME warrants that extracted analyt es be desorbed as fast as possible. A delay in analysis may result in analyte loss. In an attempt to address this problem, Nardi [96] proposed an SPME sampler for long-term storage. A home-made 15-cm long 0.16 mm i.d. capillary was coated with PS255 (polydimethylsiloxane with ~1% vinyl group) with thickness of 0.5 m. A press-fit cap was used to seal the capillary after the extraction. Even after storing for 30 days, no distinguishable loss was observed for the extracted analytes (BTEX). This extractor seems promising for samples for which immediate analysis after the extrac tion is not possible. Figure 2.10 shows a set of chromatograms illustrating BTEX analysis obtained after storing the extracted analytes in long-term storage assay for different period of time.

PAGE 102

78 Figure 2.11 Impact of storage on extracted samp le using SPME sampler for long-term storage (Reproduced from Ref. [9 6] with permissi on of Elsevier)

PAGE 103

79 One of the major problems of in-tub e SPME hyphenated to GC is ensuring quantitative transfer of the extr acted analyte into the GC injection port. This can be done by using a small solvent plug that transfer the analyte(s) from the extraction capillary to the GC injection port [97]. L. Nardi propos ed a new way to couple in-tube SPME with GC [98] that does not require a heated GC injector. He used a 9 cm long 0.25 mm i.d. capillary with 0.22 m thick immobilized coating of PS255 stationary phase (PDMS gum phase with 1% vinyl groups) for the extracti on. High-quality press-fits were made on both the ends of the extractor using a shar pened tungsten tool and an alcohol flame. Extraction was done using a cap illary extraction tool comprised of a 1 mL sampling syringe, a sampling transfer-lin e ending with a 0.32 mm i.d. fused silica capillary and capillary extractors. One of many advantages of the proposed system is its very fast sampling step, normally a few seconds. After th e extraction, the extr actor was coupled to carrier gas line and the GC column using the press-fits located on both the ends of the extractor. GC elution can be performed keeping the GC oven door open or by temperature programming, if needed. Besides, the extractor can be press-fit capped with the extracted analyte(s) inside and stored for months withou t loosing even volatiles. The system has been proved to be effective for low volatiles (e.g., PAHs) as well as volatiles (e.g., chlorinated solvents, alkanes, amides ). The newly developed in-tube SPME-HRGC is recommended especially for high-throughput an alysis when detecti on sensitivity is not the prime objective. Similar capillary extracto rs ware also used for negligible depletion sampling of BTEX [99], for de termining partition coefficients of BTEX between cross linked polydimethylsiloxane and water [100], analysis of aromatic compounds in water [101].

PAGE 104

80 e demonstrated some prom ise for VOC analysis from air as well as queous media. Musshoff et al. [103] reported a fully automated headspace solid-phase dynamic xtraction (HS-SPDE) followe d by on-coating derivatizatio n and gas chromatographyass spectrometry (GC-MS). SPDE is consider ed to be a new generation solid-phase icroextraction which uses an inside need le capillary absorption trap coated with olydimethylsiloxane. It is th e first commercially availa ble inside-needle device for eadspace analysis using GC-MS. An 8-cm long stainless steel needle coated with a 50 m film of PDMS and 10% act ivated carbon is used as the needle trap. The volume of e extracting phase in SPDE needle (4.40 mm3) was significantly larger than that of 100 m PDMS SPME fiber (0.94 mm3). Apart from its advantage of higher coating volume, its robustness due to its physical construc tion has made SPDE very promising. Higher sample capacity and longer life-span are tw o major attributes of SPDE. The new device was successfully applied for the determination of cannabinoids in hair samples, pesticides in water [104], and synthetic designer drug in hair [105]. Figure 2.11 demonstrates the principle of the SPDE extrac tion process in extracting amphetamines and designer drugs from hair samples. McComb et al. [102] invented an INCAT (inside the needle capillary adsorption trap) device in which a colloidal graphite coa ting inside a steel capilla ry tubing is used as a sorbent. The devic a e m m p h th

PAGE 105

81 2.12 Fully automated solid-phase dynamic extraction (SPDE) process (Reproduced from Ref. [105] with permission of Elsevier) Figure

PAGE 106

82 oziel et al. [106] reported a needle trap device (NTD) that was used for the sampling nd analysis of aerosols and airborne particulate matter fr om an inhaler-administered rug, spray insect repellant, and tailpipe diesel exhaust. The needle trap device was bricated using a 40 mm long pi ece of stainless st eel needle (0.53 mm o.d.) with 5 mm f quartz wool packing secti on near the tip of the need le. The quartz wool packing erforms as the extracting phase. Figure 2.12 represents the schematic of a needle trap evice. K a d fa o p d

PAGE 107

Figure[106] wth pererican Chemical Society) 2.13 Schematic representation of a needle trap device (Reproduced from Ref. imission of Am 83

PAGE 108

84 2.6 [1] R.P. Belardi, J. Pawliszyn, Wate r Pollut. Res. J. Can. 24 (1989) 179. [2] T. Y. Kim, A. Malik, Manuscript in preparation. [3] G. Vas, K. Vekey, J. Mass Spectrom. 39 (2004) 233. [4] C.L. Arthur, L.M. Killam, K.D. Buchhol z, J. Pawliszyn, J. R. Berg, Anal. Chem. 64 (1992) 1960. [5] K.D. Buchholz, J. Pawliszyn, E nviron. Sci. Technol. 27 (1993) 2844. [6] Application Note 17 Supelc o, Bellefonte, PA, USA. [7] V. Mani, in J. Pawliszyn (Ed.), Applications of Solid Phase Microextraction, Royal Society of Chemistry(RSC), Cambridge, UK, 1999, p 63. [8] S.A.S. Wercinski, Ed, Solid Phase Mi croextraction, Mercel Dekker, Inc, New York, 1999. [9] R.E. Shirey, V. Mani, Presentation at Pittcon1997, March 17-20, 1997, Atlanta, GA. [10] S. Tumbiolo, J.-F. Gal, P.-C. Maria, O. Zerbinati, Anal. Bioanal. Chem. 380 (2004) 824. [11] J.A. Lloyd, P.L. Edmiston, J. Forensic Sci. 48 (2003) 130. [12] J. Hall, J.S. Brodbelt, J. Chromatogr. A 777 (1997) 275. References for Chapters Two

PAGE 109

85 [15 atogr. A 888 (2000) 35. togr. 115. [18] rnier, E. Guichard, J. Agric. Food. Chem. 51 (2003) 999) 353Chem. 71 (2000) [24] u, J. Pawliszyn, Anal. Chem. 72 (2000) 2774. 000) 137. wliszyn, J. 2. [13] D. Zuba, A. Parczewski, M. Reiche nbacher, J. Chromatogr. B 773 (2002) 75. [14] A.A. Boyd-Boland, J.B. Pawlisz yn, Anal. Chem. 68 (1996) 1521. ] R. Aranda, P. Kruus, R.C. Burk, J. Chrom [16] D. Poli, E. Bergamaschi, P. Manini, R. Andreoli, A. Mutti, J. Chroma B 732 (1999) [17] O. Ezquerro, M. T. Tena, J. Chromatogr. A 1068 (2005) 201. B. Rega, N. Fou 7092. [19] H. Kataoka, H.L. Lord, J. Pawlis zyn, J. Chromatogr. B 731 (1 359. [20] H.L. Lord, J. Pawlisz yn, LC-GC (1998) 41. [21] R. Eisert, J. Pawliszyn, Anal. Chem. 69 (1997) 3140. [22] H. Kataoka, S. Narimatsu, H.L. Lord, J. Pawliszyn, Anal. (1999) 4237. [23] Y. Gou, C. Tragas, H. Lord, J. Pawl iszyn, J. Microcol. Sep. 12 125. Y. Go [25] Y. Gou, R. Eisert, J. Pawlisz yn, J. Chromatogr A 873 (2 [26] K. Kataoka, J. Pawliszyn, Chromatographia 50 (1999) 532. [27] H. Kataoka, H.L. Lord, S. Yamamoto, S. Narimatsu, J. Pa Microcol. Sep. 12 (2000) 493. [28] M. Takino, S. Daishima, T. Na kahara, Analyst 126 (2001) 60

PAGE 110

86 9) Kiso, K. Jinno, Anal. Biomed. Anal. 9. [34 M. Ise, S. Narimatsu, J. Sep. Sci. 25 (2002) 77. [37] Assadi, S.H. Ha ddadi, Anal.Chem. 73 (2001) 4054. 96) 361. [41] ssadi, G. Karim-Nezhad, Chromatographia 56 (2002) [44] hromatographia 45 (1997) 183. [46 R. Cenciarini, Chromatographia 41 (1995) 678. [29] B.C.D. Tan, P.J. Marriott, P.D. Mo rrison, H.K. Lee, Analyst 124 (199 651. [30] Y. Saito, Y. Nakao, M. Imaizumi, Y. Morishima, Y Bioanal. Chem. 373 (2002) 81. [31] Z. Mester, J. Pawliszyn, Rapi d Commun. Mass Spectrom. 13 (1999) 1999. [32] K. Mitani, S. Narimatsu, F. Izushi, H. Kataoka, J. Pharm. 32 (2003) 469. [33] K. Mitani, S. Narimatsu, H. Kataoka, J. Chromatogr. A 986 (2003) 16 ] H. Kataoka, [35] H. Yuan, W.M. Mullett, J. Paw liszyn, Analyst 126 (2001) 1456. [36] N.A. Guzman, Electrophoresis 24 (2003) 3718. D. Djozan, Y [38] M.A. Farajzadeh, M. Hatami, J. Sep. Sci. 26 (2003) 802. [39] F. Guo, T. Gorecki, D. Irish, J. Pawliszyn, Anal. Commun. 33 (19 [40] D. Djozan, M. Amir-Zehni, Chromatographia 58 (2003) 221. D. Djozan, Y. A 611. [42] M.A. Farajzadeh, A.A. Matin, Anal. Sci. 18 (2002) 77. [43] M.A. Farajzadeh, M. Hatami, Ch romatographia 59 (2004) 259. Dj. Djozan, Y. Assadi, C [45] Dj. Djozan, Y. Assadi, Mi crochem. J. 63 (1999) 276. ] F. Mangani,

PAGE 111

87 ci. [49] nal. Lett. 34(4) (2001) 627. [52] hem. 73 (2001) 2041. [55] .H. Lubbad, D.C. Tilotta, J. Chromatogr. A 829 (1998) [58] Irgum, K.E. A. Ohl sson, Rapid Commun. Mass Spectrom. [59] 1998) 167. [61] 17. ) [47] J. Jia, X. Feng, N. Fang, Y. Wang, H. Chen, W. Dan, J. Environ. S Health A 37 (2002) 489. [48] C.-P. Kuo, J. Shiea, Anal. Chem. 71 (1999) 4413. Z. Tong, L. Guanghan, Y. Xin, A [50] Y. Liu, M. L. Lee, J. K. Hageman, Y. Yang, S.B. Hawthorne, Anal. Chem. 69 (1997) 5001. [51] Y. Liu, Y. Shen, M.L. Lee, Anal. Chem. 69 (1997) 190. X.R. Xia, R.B. Leidy, Anal. C [53] J.-g. Hou, Q. Ma, X.-z. Du, H.-l. Deng, J.-z. Gao, Talanta 62 (2004) 241. [54] I. Bruheim, X. Liu, J. Pawlisz yn, Anal. Chem. 75 (2003) 1002. S.A. Merschman, S 377. [56] E.O. Otu, J. Pawliszyn, Mi crochim. Acta 112 (1993) 41. [57] T. Gorecki, P. Martos, J. Paw liszyn, Anal. Chem. 70 (1998) 19. E. Norlin, K 17 (2003) 936. C. Jia, Y. Luo, J. Pawliszyn, J. Micro. Sep. 10 ( [60] S. Li, S.G. Weber, Anal. Chem. 69 (1997) 1217-1222. S.D. Harvey, D.A. Nelson, B.W. Wri ght, J.W. Grate, J. Chromatogr. A 954 (2002) 2 [62] C. Xiao, S. Han, Z. Wang, J. Xing, C. Wu, J. Chromatogr. A 927 (2001 121.

PAGE 112

88 804 (2004) 3. [65] Walles, K. Levsen, J. Borlak, J. Pawliszyn, J. J. 0 (2002) 307. 963 [71] .M. Mullett, J. Pawlisz yn, J. Chromatogr A 1025 (2004) 85. col. Sep. 10 (1998) o, r. Anal. Commun. 36 (1999) 203. id State Electrochem. 6 (2002) [76] samani, O. Ceyl an, S. Oztemiz, T.P. Gbatu, R.A. LaRue, U. Tamer, Harry B. Mark, Jr., Talanta 58 (2002) 739. [63] J. Nilsson, P. Spegel, S. Nilsson, J. Chromatogr. B [64] W. M. Mullett, P. Martin, J. Pawliszyn, Anal. Chem. 73 (2001) 2383 W.M. Mullett, M Chromatogr. B 801 (2004) 297. [66] M. Walles, W.M. Mullett, K. Levsen, J Borlak, G. Wunsh, J. Pawliszyn Pharm. Biomed. Anal. 3 [67] E.H.M. Koster, C. Crescenzi, W. den Hoedt, K. Ensing, G.J. de Jong Anal. Chem. 73 (2001) 3140. [68] E. Hogendoorn, P. van Zoonen, J. Chromatogr. A 892 (2000) 435. [69] W.M. Mullett, J. Pawliszyn, Anal. Chem. 74 (2002) 1081. [70] W.M. Mullett, K. Levsen, D. Lubda, J. Pawliszyn, J. Chromatogr. A (2002) 325. M. Walles, W [72] W.M. Mullett, J. Pawliszyn, J. Sep. Sci. 26 (2003) 251. [73] J. Chongrong, L. Yuzhong, J. Pawlisz yn, J. Micro 167. [74] T.P. Gbatu, O. Ceylan, K.L. Sutton, J. F. Rubinson, A. Galal, J.A. Carus H.B. Mark, J [75] K.R. Temsamani, O. Ceylan, B. J. Ya tes, S. Oztemiz, T.P. Gbatu, A.M. Stalcup, H.B. Mark Jr., W. Kutner J. Sol 494. B.J. Yates, K.R. Tem

PAGE 113

89 [78] Markides, L. Nyholm, Analyst 127 [79] em. 74 (2002) 4855. ri, 1999, Waterloo, Canada. [83] ster, J. Pawliszyn, Anal. Chim. Acta 424 (2000), 211. nalyst 125 (2000) 391. pectrom.16 (2001) 59. J. Microcol. Sep. 12 (2000) [88] liszyn, Anal. Chem. 73 (2001) 55. [90] Paw liszyn, J. Chromatogr. A 976 (2002) 30 (2002) 307. (2002) 1695. [77] G. Liljegren, L. Nyholm, Analyst 128 (2003) 232. G. Liljegren, J. Pettersson, K.E (2002) 1. J. Wu, W.M. Mullet, J. Pawliszyn, Anal. Ch [80] M.R. Ganjali, R. Kiani-Anbouhi, M.R. Pourjavid, M. Salavati-Niasa Talanta 61(2003) 277. [81] J. Wu, Z. Deng, J. Pawliszyn, Extech., [82] J. Wu, J. Pawliszyn, J. Chromatogr. A 909 (2001) 37. J. Wu, Z. Me [84] J.Wu, X.Yu, H.Lord, J. Pawliszyn, A [85] J.Wu, Z. Mester, J. Pawliszyn, J. Anal. At. S [86] J. Wu, H.L. Lord, J. Pawliszyn, H. Kataoka 255. [87] J. Wu, W. Xie, J. Pawlis zyn, Analyst, 125 (2000) 2216. J. Wu, J. Paw [89] J. Wu, H. Lord, J. Pawlis zyn, Talanta 54 (2001) 655. J. Wu, C. Tragas, H. Lord, J. 357. [91] M.Walles, W.M. Mullett, K. Levsen, J. Borlak, G. Wunsch, J. Pawliszyn, J. Pharm. Biomed. Anal [92] W.M. Mullett, K. Levsen, J. Borlak, J. Wu, J. Pawliszyn, Anal. Chem. 74

PAGE 114

90 [93] H.L. Lord, R.P. Grant, M. Waes, B. Incledon, B. Fahie, J.B. Pawliszyn, Anal. Chem. 75 (2003) 5103. [94] Y. Wang, M. Walles, B. Thom Nacson, J. Pawliszyn, Rapid Commun. Mass Spectrom. 18 (2004) 157. [95] A. G de Saja, Sensors and Actuators, B 77 (2001) 401. a togr. A 985 (2003) 3. [97] 99) 997) [10 r. A (2002) 231. [10 ll son, S. uadarrama, J.A. Fernandez, M. Iniguez, J. Souto, J.A [96] L. Nardi, J. Chrom B.C.D. Tan, P.J. Marriot, H. Kee Lee, P.D. Morrison, Analyst 124 (19 651. [98] L. Nardi, Amer. Lab. 30 (2002) 32. [99] L. Nardi, J. Chroma togr. A 985 (2003) 85. [100] L. Nardi, J. Chroma togr. A 985 (2003) 39. [101] L. Nardi, J. Chroma togr. A 1017 (2003) 1. [102] M.E. Macomb, R.D. Oleschuk, E. G iller, H.D. Gesser, Talanta 44 (1 2137. 3] F. Musshoff, D. W. Lachenmeier, L. K. Kroener, B. Madea, Forensic Sci. Internat. 133 (2003) 32. [104] J. Lipinski, Fresenius' J. Anal. Chem. 369 (2001) 57. [105] F. Musshoff, D.W. Lachenmeier, L. Kroener, B. Madea, J. Chromatog 958 6] J.A. Koziel, M. Odziemkowski, J. Pawliszyn, Anal. Chem. 73 (2001) 47.

PAGE 115

91 apillary Microextraction of S) the silica gel led to diminished th e enthusiasm of the researchers and little technol Chapter Three Sol-Ge Technology in C 3.1 A brief history Sol-gel technology offers a simple but ve rsatile approach to the synthesis inorganic polymers and hybrid inorganic-organic materials. The history of sol-gel technology dates back to mid-1800s with Ebelmens [1] and Grahams [2] studies on silica gel. Both the investigators observed that the hydrolysis of tetr aethoxysilane (TEO under acidic conditions produced SiO 2 in the form of glass-like material. Unfortunately, extremely long drying time (one year or more) required to prevent fracture of ogical development was observed in the next couple of decades. A period between late 1800s and 1920s, however, had witnessed a surge of interest in the sol-gel process. During this time many famous scientists, including Ostwald [3] and Lord Rayleigh [4] investigated the problem of periodic precipitation phenomenon associated with the formation of Liesegang rings [5] and the growth of crystals from gels. During the period 1950s and 1960s Roy and coworkers [6,7] succeeded in using sol-gel methods for synthesizing a large number of novel ceramic oxide compositions involving Al, Si, Ti, Zr, etc ., that was impossible to obtain using ceramic powder

PAGE 116

92 968 -Gel Science and Techno f tructure and physical chemistry f the original-type of li quid solution ( mixture of alc ohol, alkoxide, water, catalyst) table advancement during this time period was the pioneering work of H. chmid methods. Another important br eakthrough during this period wa s Ilers pioneering work on silica chemistry [8] which led to the commercial development of colloidal silica powders known as Du Ponts co lloidal Ludox sp heres. Stober et al. [9] reported in 1 that both the morphology and size of the powders could be controlled using ammonia as a catalyst for the TEOS hy drolysis reaction. The First International Workshop on Glasse s and Glass Ceramics from Gel was held in Padova, Italy, in 1981 and is consider ed to be the beginning of the present era of sol-gel science a nd technology [10]. Since then, there have been nine other such Workshops held biannually at different locations throughout the world. With ever increasing interest in sol-gel science and technology, last two and a half decades alone have witnessed approximately 76,000 publicat ions. The huge surge of sol-gel articles prompted the birth of a new scientific journal Journal of Sol logy in 1993 which successfully brough t all sol-gel scien tists on a single platform. J.D. Mackenzie [11] has classified the achievements in sol-gel chemistry during last two and a half decades (1980-2005) into two broader classes: (1 ) first generation o the sol-gel process and (2) second generation of the sol-gel process. The first generation sol-gel process involved in better understand ing of the s o which resulted in oxide gels. Most no St who successfully incorporated organi c material into inor ganic network by sol gel process [12] which according to J.D. Mackenzie is regarded as the beginning of

PAGE 117

93 brid al hase for n world Sol-gel coating technology for the prepar ation of solid-phase microextraction bers was developed by Malik and co-workers in 1997 [18] The new coating technology second generation. Schmidts invention opened up a new possibility of creating hy organic-inorganic materials using a very si mple sol-gel methodology. The new materi system was named as Ormosils and later renamed as Ormocers by the inventor himself. The credit for using or ganic-inorganic hybrid material as the stationary p electrochromatography goes to Guo and Colon [13]. Malik and co-workers introduced sol-gel column technology for capillary GC [14]. This research effort on sol-gel colum technology was so well accepted am ong the scientific community that the article titled Sol-gel Column Technology for High Resolutio n Gas Chromatography was featured on the front cover of Analytical Chemistr y (Vol. 69, No. 22, pp. 4566-4576, 1997). Since Colon and Maliks pioneering work, a large nu mber of research gr oups all over the have been involved in developing new sol-gel hybrid organic-inorga nic materials system in the area of sample preconcentration and microseparation. Among many scientists working in th e field of chromatography, Tanaka et al. made a significant contribution by developing sol-gel monolithic beds and used them as an HPLC column, [15-17]. The monolithic colu mns with small-sized skeletons and large through-pores can reduce the diffusion path leng th and flow resistance compared to a particle-packed column (most commonly used in HPLC) a nd thereby provide both high permeability and high column efficiency in HPLC. 3.2 Sol-gel technology for SPME fi

PAGE 118

94 ed l into inorganic polymeric networ onding of the created hybrid organic-inorganic sol-gel to the externa h ion materials or reagents. One or more in e polarity of the materia fine as properly designed sol solution. The thi gy. ion of both organic and inorganic phases as well as the intrinsic ability of these organic-inorganic materials to chemically bind to the surface of the substrate have provided the hybrid sol-gel materials with enhanced selectivity, high thermal and solven t stability, and long life span. allow the incorporation of organic polymeric materia k and chemical b l surface of a fused silica fiber. Because of the chemical bonding to the fiber, suc hybrid coatings are character ized by high thermal and chemical stability. Sol-gel approach offers the opportunity to create th ese advanced materials under mild react conditions (most often at room temperature) The technology is very simple, easy-tofollow, single-step and above all does not involve costly organic precursor(s), a sol-gel-active organic component, a sol-gel catalyst, water and solvent(s) are the raw materials involved in the prep aration of sol-gel hybrid organicinorganic composite material. By changing the relative ratio of organic/inorganic materials and by adding one or more surface d eactivating reagents a nd th l can be controlled to some extent provi ding sol-gel chemists the flexibility in tuning the selectivity of such hybrid materials. Furthermor e, the new technology permits the in situ creation of hybrid coatings on the outer surface of a fused silica fiber as well on the inner walls of fused silica capillary. It can also be employed for the creation of a surface-bonded monolithic bed within a capillar y using a ckness of the coating can also be conveniently contro lled using this technolo Positive attributes acquired from the integrat

PAGE 119

953.3 Sol-gel sorbents in SPME: a brief overview An ideal sample preparation technique should be si mple, inexpensive, efficient, easy to use, fast, solvent-free and compatible with a wide range of analytical separation instruments. Since solid phase microextrac tion (SPME) conforms to most of these criteria, it has gained enormous popularity am ong scientists and rese archers shortly after its introduction in 1989 by Pa wliszyn and co-workers. In solid phase microextraction, sorbents/coatings play the most important role in the extraction process. Typically, a sort segment of bare fused silica fiber is coated with a polymeric material. Conventional sorbents, be ing merely physically bonded to the fused silica fiber surface, inherently possess low thermal and solvent stability. In solid phase microextraction, the th ickness of the sorbents is another most important factor that determines the volume of the loaded sorbent, and consequently, the amount of analyte(s) that can be extracted by it. As a re sult, reasonably high extraction sensitivity for an analyte can only be obtaine d when the sorbent has high affinity for the analyte and the sorbent coating thickness is fairly high. Sol-gel sorbents used in SPME/ CME can be groupe d into different types. Scheme 3.1 presents such a classification.

PAGE 120

96 Sol-Gel SPME Sorbents Polysiloxane Sorbents Miscellaneous Sorbents Nonpolysiloxane Sorbents Homogenious Sorbents Mixed Sorbents Low temperature glassy carbon film Sol-gel DHB PEG Dendri mers PolyTHF PDMS, Polymethylphenylsiloxane, PDMS/ Fullarene PDMS/ Calix[4]arene PDMS/ Crown Ether PDMS/PVA PDMS/DVB PDMS/ PMPVS C8Cyclodextrin Classifications of Sol-gel Sorbents Used in SPME/CME PDMS/ PPheMS Figure 3.1 Classification of sol-gel sorbents used in fiber-SPME/in-tube SPME (CME)

PAGE 121

97el sorbents high thermal and oxidative stability, high maxilow miating ss traper low surface energies, easine ss of film formation, high solubilizing power, good selectivity and tunable polar ity have made polysiloxanes the most popular as well as the most commonly used sorbents in all fo rmats of solid phase microextraction [19]. Polysiloxanes with sila nol groups in their structur es are sol-gel active and can participate in sol-gel reactions. In polysiloxane-based SPME sorbents, di fferent functional pe ndant groups are chemically bonded to the polysiloxane backbones. Among the most commonly used pendant groups are m enyl groups. 3.3. orbenogeneous polysiloxane phases ry sorbents, hydroxy terminated PDMS (certain types of silicone oils) have served as organic components in the prepared sol-gel extraction media. precursors (alkoxides of Si, Ti, and Zr) have been used for successful chem tion of these organic moie ties onto the surface of the substrate (outer surface of a fused silica fiber (fiberinner surface of a fused silica capillary (inbe SPME or CME). The sol-gel reactio ns inherently provide molecular level homogeneity in the composition of the prepar ed coating and ensure its chemical bonding to the fused silica glass surface. Table 3. 1 lists different homogeneous polysiloxane 3.3.1 Polysiloxane-based sol-g Chemical inertness, low vapor pressure mum and nimum oper temperat ures, low gla nsition tem atures, ethyl, and ph 1.1 Sol-gel s ts with hom In this catego of SPME Different inorganic ical immobiliza SPME) or tu

PAGE 122

98phases TSolvent ce used as SPME sorbents. able 3.1 Homogeneous polysiloxane sorbents Sorbent Precursor Organic component Thermal/ stability Referen Sol-gel PDMS MTMOS 320 C [18] Sol-gel PDMS MTMOS 300 C [ 21] Sol-gel PDMS MTMOS 350 C [22] Sol-gel Titania Titanium Acetonitrile/ [25] PDMS i-propoxide water Sol-gel Zirconia Zirconia PDMDPS Butoxide 320 C [26] Sol-gel poly phenylmethylvinyl siloxane TEOS, VTEOS Si CH3 O xO Si CH3 CH3 O Si CH CH3 CH2 z y 350 C [50] Sol-gel polyphenylmethyl siloxane/PDMS TEOS, VTEOS CH3 Si H3C CH 3 O Si CH 3 O Si CH3 CH 3 CH3 n [56] 350 C

PAGE 123

99 hong et al. [18] were the first to develop sol-gel coating technology [14] to prepare solid-phase microextraction fibers Hydroxy-terminated poly(dimethylsiloxane)the most widely used polymer as SPME so rbent was reacted with sol-gel precursor methyltrimethoxysilane (MTMOS) in presence of a sol-gel catalyst (trifluoroacetic acid containing 5% water) to produce a hybrid organic-i norganic polymeric network chemically bonded to the fused silica fiber su rface. Due to the strong chemical bonding to the substrate, the hybrid material provid ed exceptionally high thermal and solvent stability. Moreover, the porous structure of the sol-gel coating, as is evident from the scanning electron micrograph (Figure 3.2), facili tated efficient mass transfer of analytes between the sorbent and aqueous media. This in turn, facilitated faster extraction equilibrium compared to conventional thic k PDMS coating which may take hours to reach the equilibrium. Due to the porous stru cture, a sol-gel hybrid organic-inorganic material provides higher surface area, as a re sult of which, a relatively thin coating is sufficient to achieve an analyte detection limit eq uals to or even lower than that achieved using conventionally coated sorbents. Another very impor tant feature of sol-gel poly(dimethylsiloxane) coating is its ability to extract bot h polar and nonpolar analytes from aqueous media. This can be explaine d by the fact that so me of the used hydroxyterminated poly(dimethylsiloxane) molecules may retain one of their terminal hydroxyl groups (e.g., those bonded to th e sol-gel network onl y with one end) and provide affinity towards polar analytes during extraction. C

PAGE 124

Figure: 3.2 Scanning electron microscopic image of sol-gel PDMS fiber at 3600fold magnification (Reproduced fro m Ref. [18] with permission of American Chemical ociety) S 100

PAGE 125

101 Since this feature is absen t in conventional PDMS coa ting, extraction of polar ompounds is impractical to achieve on conventional PDMS coated fibers. Figure 3.3 demon 18] g l S), ong c strates the extraction of highly polar alcohols from aqueous solution using sol-gel PDMS coatings. The porous structure of solgel PDMS coating in conjunction with its high thermal stability ensures faster and comp lete desorption of the extracted analytes into the GC injection port and thus eliminates the anal yte carryover phenomenon which is a frequent problem with the use of conventional thick PDMS coatings. Chong et al. [ also presented a simplified way to tune the polarity of sol-gel PDMS coating by addin deactivating reagents e.g., polym ethylhydroxysilane (PMHS) and/or trimethylmethoxysilane that can react with OHpresent in th e polymer or on the sol-ge network and thus tune the polarity of the sorbent. Although in sol-gel chemistry, tetraalkoxys ilanes [e.g., tetraethoxysilane (TEO tetramethoxysilane (TMOS), etc.] are the most commonly used sol-gel precursors, Chet al. [18] utilized methyltrimethoxysilane (MTMOS) that helped prevent formation of cracks [20] particularly in thick coatings during the drying step, if tetraalkoxysilane precursors are used.

PAGE 126

Figure 3.3 SPME-GC analysis of aliphatic alcohols (C 10 -, C 12 -, C 14 -, C 16 -, C 18 -)-using sol-gel PDMS fiber and GC/FID system (Reproduced from Ref. [18] with permission of American Chemical Society) 102

PAGE 127

103 ated m r he s. d acks of fiber SPME. In conventional SPME fibers, a small piece of fused silica fiber is fiber ing Zhou et al. [21] in fact repeat ed the work done by Chong et al. [18] and valid the original work by Chong et al. [18]. Moreover, new compound classes (e.g., BTEX and organochlorine compounds) have been extr acted using sol-gel PD MS sorbent (40 thickness) and the limits of detection (LOD) were compared with the results obtained on conventional 100 m thick coatings. As was pointed out by Chong et al. [18], being porous, sol-gel organic-inorganic hybrid ma terial provides enhanced surface area for extraction. As a result, a sol-gel coating of moderate thickness can provide comparable o even lower detection limits which is of prim e importance in ultra-trace analysis. As is indicated by the result obtained for BTEX analysis, an order of magnitude lower detection limit has been accomplished by the sol-gel PDMS coating, even though t thickness was two-fifth of the conventiona l PDMS coating. Another comparison was made with highly polar Polyacrylate coating for extracting organochlorine compound The chromatograms presented in the article cl early demonstrated the superiority of solgel PDMS coating over PA coating even in extracting highly polar compounds. Bigham et al. [22] used sol-gel coatings to address several inherent format-relate drawb used as the substrate. A thick layer of coating applied to th e outer surface the serves as the extracting sorbent. Due to the absence of any protective coating (e.g., polyimide), this fiber is very fragile. Mo reover, as the coating stays on the external surface of the fiber, mechanical damage of th e coating due to scraping and syringe needle bending is a frequent phenomenon. Due to the high fragility of fused silica fibers, only a small segment (~ 1 cm) can be used for coati ng resulting in low sta tionary phase load for extraction. As a result, in the fiber fo rmat, SPME suffers from low sample capacity

PAGE 128

104 and consequently poor method sensitivity. Additi onally, this format is difficult to couple with high-performance liquid chromatography. In an attempt to eliminate these format related problems as well as to couple SPME with HPLC, in-tube SPME has been proposed [23]. It uses a coated fused silica capillary (~ 60 cm) (most commonly a piece of open tubular GC column). Although the ne w approach helps interfacing SPME with HPLC, other problems still remain to be solved In the conventional coating process used to make open tubular GC columns, the stationa ry phase inside the fu sed silica capillary is physically immobilized on the surface of th e substrate. Unfortunately, the lack of chemical bonds between the coa ting and the capillary wall results in low thermal and solvent stability to the coating. This probl em is manifested by relatively low thermal stability as well as poor solvent stability of the commercial GC columns. As a result, such capillaries are not suitable for repeated SPME operations usin g organic or organoaqueous HPLC solvents for the desorption of the extracted analytes. All of the underlying format related as well as co ating related problems have been well addressed by Bigham et al. [22]. In the new method, a piece of sol-ge l coated fused silica capillary [14] was used. Due to the strong chemical bonding of sol-gel organic-inorganic hybrid material to the inner walls, such capillaries demonstrated ve ry high thermal stability as well as excellent stability toward conventional organic solv ents. A highly nonpolar phase (PDMS) and a highly polar phase (PEG) [24] were used as new sol-gel SPME sorbents. Both phases demonstrated remarkably superior performan ce in direct extraction of polar and nonpolar analytes from aqueous media compar ed to conventiona l coatings. Although sol-gel tec hnology has been proven to be the most convenient and effective means of chemical immobilization for sorbent coatings, a narrow window of pH

PAGE 129

105 nherent in silica based material s has imposed a serious obstruction to wide pplication of such materials particularly when pH adjustment to very high or very low ) from a given matrix. Cone fact poss bilitynic compared with silica, Kim et al. [25] developed a high pH resistant, surface bonded solrganic-inor ganic coating and dem its application in capillary microextraction by on-line hyphenating with high-performance liquid chromatography. The new sol-gel hybrid material genera ted from titanium isopropoxide (inorganic precursor) and polydimethylsiloxane (organic moiety) provided excellent performance stability even after being exposed to hi ghly basic conditions (pH=13) for prolonged period of time (12 h). Moreover, direct chemical bonding of the TiO -PDMS coating to the capillary inner walls demonstrated excelle nt solvent stability, a prerequisite for coupling SPME with liquid-pha se chromatographic techni ques (e.g., HPLC, CE, CEC, etc.). Pursuing a similar objective, Alhooshani et al. [26] reported a novel organicinorganic hybrid zirconia co ating for capillary microext raction. Although Zirconia has long been known to be better al kali resistant material than other metal oxides, such as alumina, silica, and Titania, this was the fi rst report on sol-gel hybr id organic-inorganic zirconia sorbent used in SPM E. Outstanding pH resistance, ex cellent chemical inertness, and high mechanical strength have made zi rconia an excellent candidate for support material in the field of chromatography and membrane based separa tions created by using zirconium (IV) butoxide as the inorga nic precursor and hydroxy-terminated polydimethyldiphenylsiloxane as the organic moiety, the new sol-gel hybrid inorganicstability i a values are inevitable for the selective extraction of target analyte(s sidering th that Titania esses supe rior pH sta and mecha al strength gel titania hybrid o onstrated 2

PAGE 130

106 organic material provided exceptionally high pH stability (pH 13, 24 h). Moreover, parts per trillion (ppt) level det ection limits were achieved for both polar and non-polar analytes in CME-GC-FID experiments using sol-gel zirconia-PDMDPS as the extraction sorbent. 3.3.1.2 Mixed polysiloxane-base d sol-gel sorbents In polysiloxane based mixed sorbents, one or more organic polymeric entity is used along with PDMS to impart new selectivity to the composite material as well as some other physico-chemical advantages (e.g., higher thermal stability, additional sample capacity, greater surface area, etc.) So far, hydroxyfullerene, calix[4]arene, PVA, crown ether, DVB, PMPVS, PPheMS have been reported in the lit erature [27-40]. 3.3.1.2.1 Fullerene-polysiloxane mixed sol-gel sorbents Fullerenes are closed-cage carbon mo lecules containing pentagonal and hexagonal rings. Because of its unique stru cture and properties, it has drawn wide attention immediately after its discov ery in 1985[27]. The high hydrophobicity of fullerene as well as its high th ermal stability, resistance to oxi dation, and the capability of strong and donor–acceptor interactions with an alytes have made it a very promising SPME sorbent. Table 3.2 presents the importa nt features of solgel hydroxyfullerene sorbent.

PAGE 131

107Table 3.2 Important features of so l-gel hydroxyfullerene sorbent Sorbent Precursor Organic Polymer Thermal/ Solvent Stability Reference Sol-gel hydroxyfullerene MTMOS C60OH n 360 C [27] Yu et al. [27] exploited all these positive attributes of fullerene by developing an SPME coating using hydroxyfullerene with hydroxy terminated sili cone oil (OH-TSO). A solgel approach was used to prepare the coati ng. Sol-gel solution was prepared using fixed volume (100 L) of OH-TSO, variable amount of hydroxyfullerene (0, 10 and 20 mg), 400 L MTMOS, 100 L water and 200 L TFA. Variable amount of fullerene was used to measure the contribution of hydroxyfullerene in extractions. Sol-gel hydroxyfullerene coated SPME fibers were conditioned at as high as 360 C for 5 hour s. Scanning electron micrographs (Figure 3.4) of the fullerol-co ated surface revealed that such coatings possess a porous structure.

PAGE 132

108 Figure Scanning electron micrograph of sol-gel fullerol fiber at 2000-fold magnification (Reproduced from Ref. [27] with permission of Elsevier) 3.4

PAGE 133

109 The high thermal and solvent stability may be attributed to the strong chemical bonding of the sorbent to the substrate, as well as the excellent thermal stability of the participating organic ingredient s (fullerol and OH-TSO). Porous structure of the fullerene based sol-gel coatings facil itated faster mass transfer be tween the sorbent and aqueous media, and helped an extraction equilibrium to establish significantly faster than that on commercial coatings. For example, some PCBs required only 50 min to reach extraction equilibrium on a sol-gel fullere ne coated fiber. Under si milar conditions, a commercial 100 L PDMS coating required even several days to reach equilibrium [28]. The highly porous structure of the coating not only helped in faster mass transfer during extraction but also aided in faster a nd complete desorption, minimizing the memory effect. Sol-gel hydroxyfullerene coatings dem onstrated excellent planar conformation selectivity and molecular recognition. The sele ctivity toward planar PCBs increased with increasing amount of fullerol added in the so l solution. It is belie ved [27] that strong affinity of planar PCBs toward fullerol is due to the charge-transfer between PCBs and the conjugated three-dimensional -electronic system of fullerol. Sol-gel hydroxyfullerene coatings were also used to extract polycyclic aromatic hydrocarbons and aromatic amines. The results revealed th at sol-gel fullerol coatings are not only suitable for nonpolar compounds but also very efficient in extracti ng polar analytes.

PAGE 134

110 ) everal approaches have been suggested to avert the inherent shortcomings of low-mocular-mass crown ethers as GC stationary phases or SPME sorbents. Table 3.3 lists mot common crown ether based polysiloxane sorbents used in SPME. 3.3.1.2.2 Mixed crown ether-polysiloxane sol-gel sorbents Crown ethers are cyclic carbon compounds containing heteroatoms such as oxygen, nitrogen and sulfur. They are char acterized by a cavity structure, medium polarity and strong electronegativ e effect of heteroatoms on the crown ether ring. They have long been used as gas chromatographi c stationary phases because of their unique selectivity resulting from the cavity stru cture and interactio n provided by highly electronegative heteroatoms on the crown ethe r ring. The synthesis of polymeric crown ethers is, perhaps, the easies t among all the crown ether-based phase that have so far been investigated as GC stationary phases. Poor diffusibility, low column efficiency as well as bleeding at high temperatures (due to the generation of lo w molecular mass crown ethers have discouraged further research in the field. So far only one article has been published by Fine et al. in 1985 [29]. S le s

PAGE 135

111Table 3.3 Sol-gel crown ePreOrganic olymer Thermal/ Solvent StReference ther sorbents used in SPME Sorbent cursor P ability Sol-gel ether TEOSa Hydroxycrown O O O O O CHCH2 O(CH2)3 H 2C Si(OCH3)3 OH 350 C [30] So l-gel Hydroxycrown MTMOSb ether O O O O OH O O O O OH 340 C ] [31 n-C15H31 OH O O O O OH t-Bu t-Bu OH Sol-gel benzocrown ether TEOS O O O O O CH2O CH2 3OH CH2OCH2 3OH 350 C [34] Sol-gel Dibenzo-TEOS, 18-crown-6 VTOSc O O O O O O CH2CH CH2 O O O O O CH2 HC H2C OO O O O O CH2OCH2CH2OCH2CH CH2 350 C [38] Sol-gel ether TEOS 380 C [37] Bisbenzocrown a TEOS = tetraethoxysilane; b MTMOS= methyltrimethoxysilane; c VTEOS= vinyltriethoxysilane.

PAGE 136

112 10 he sol solution phase Zeng and coworkers [30] proposed a solgel based methodology to coat SPME fibers with crown ethers having hydroxyl termin i and successfully utilized them in phenol analysis. Sol solution was prepared by mi xing 8 mg of OH-DB14C 4, 90 mg OH-TSO mg PMHS, 100 L TEOS, 50 L 3-(2-C yclooxypropoxyl)propyltr imethoxysilane (KH560) and 100 L methylene ch loride. After thor ough mixing of all the sol solution ingredients, 80 L TFA (5 % water) was added to catal yze the sol-gel reactions (hydrolysis and polycondensat ion) to form a three dimensional sol-gel network chemically bonded to the substrate. Solubi lity of OH-DB14C4 in the solution was a limiting factor that prohibited the use of highe r concentrations of cr own ether in t The bare fused silica fibers were dipped into the sol solution for 30 min each time and freshly prepared sol solution was used to replace the old one after each 30 min period until the coating thickness reached 73 m. Two other fibers having coating thickness 62 m and 85 m were also prepar ed for comparison purpose. To verify the significance of OH-BD14C4 in the sol solution, several other fibers were prepared using different amounts of OH-DB14C 4 (0 mg, 4 mg, and 8 mg) keeping mass of OH-TSO (90 mg) constant. Partition coefficients (K) of different phenols between the aqueous and the sorbent coating containing differe nt proportions of OH-DB14C14 and OH-TSO (0 mg/90 mg, 4 mg/90 mg, 8 mg/90 mg 8 mg/0 mg) were determined.

PAGE 137

113 Table 3.4 Comparison of partition co efficients of fibers with different compositions* OH-DB14C4/OH-TSO (mg/mg) Compound Abbreviation 0/90 (75 m) 4/90 (70 m) 8/90 (73 m) 8/0 (7 6 m) 2-chlorophenol 2CP 179.9 224.7 255.8 6 4.2 2-nitrophenol 2NP 150.3 153.8 202.3 80.9 phenol P 54.3 56.0 59.1 3.3 2,4-dimethylphenol 2,4DMP 157.6 224.8 273.9 84 .8 2,4-dichlorophenol 2,4DCP 215.3 531.1 574.5 211 .5 Reproduced by permission from Anal. Chem. 73 (2001) 2429 As can be seen from the tabulated data (Table 3.4), the partition coefficient ( values increased with the increase of OH-DB14C4 content. When OHDB14C4 was used as the only sol-gel-active orga nic component in the sol solution, K values have been reduced substantially. This was probably due to the fact that the presence of OH-TSO in the sol solution helped not onl y in lengthening the silica network leading to the increased surface area of the composite material but also in spreading the stationary phase uniformly on the fiber surface. Fluorescen ce micrograph of two fibers (one with OHTSO, the other without OH-TSO ) revealed that sol-ge l OH-DB14C4/OH-TSO spreads uniformly on the substrate. On the other hand, sol-gel OH-DB14C4 (no OH-TSO) K)

PAGE 138

114 n the t cation range toward higher boiling compounds. SPME, affinity of an analytes for a given coating depends on their polarity following the principle like di ssolves like. Therefore, polar CW/DVB should be able extract polar phenols in great excess comp ared with the medium polar OH-DB14C4/OHTSO. Bt in reality, this was not the case. The three dimensional network generated by sol-gel coating technology pr ovided DB14C4/OH-TSO coating enhanced surface area as well as higher sample capacity that may have compensated its deficiency due to inferior polarity. Optimization of headspace SPME conditions for phenols revealed that extraction temperature of 40 C, salt-saturate matrix of pH 1, constant stirring during extraction, extraction time of 40 min, desorption for 5 min at 260 C yielded best result with respect to extraction. underwent shrinkage during the gel formation leading to the forma tion of cracks o surface. Life span is one of the most important characteristics of an SPME coating. SPME coatings are damaged (frequent phenomenon for commercial fibers) due to the sudden exposure to the abruptly high temperature during analyte deso rption in GC injection por and/or the presence of solv ent in the matrix. Sol-gel coated OH-DB14C4/OH-TSO SPME fibers were characterized by high thermal a nd solvent stability. No apparent loss in extraction efficiency was obser ved even after being heated at 350 C and also after 150 times repeated extractions. Th e presence of phenyl ring in cr own ether structure, as well as the strong chemical bonding of the sol-ge l network to the fused silica surface, are believed to be responsible for the high th ermal and solvent stability of sol-gel OHDB14C4/OH-TSO coating. Coatings of such hi gh thermal stability coatings are likely to help expanding the SPME appli In u

PAGE 139

115 he same group also prepared three other crown ether-based solid phase microextraction coatings by sol-gel method [31] using hydroxydibenzo-14-crown -4(OH3, 5Cyclooxypropoxyl)-propyltrimethoxysiloxane (KH-560), and 100 L methylene chloride. Then 100 L of TFA (5% water) wa s added and the mixtur e was centrifuged to remove the of particulate materials. In cas e of other two crown ethers, 10 mg of each were used in the sol solution keeping all ot her ingredient unchanged. The thicknesses of sol-gel OH-DB14C4/OH-TSO, DHSU14C 4-OH/TSO, and DBUD14C4/OH-TSO coatings were 65-, 70-, and 70 m, respect ively. In the course of SPME method optimiz und to o erent crown ethers for aromatic amine extraction, 5 differen suggest T DB14C4), dihydroxy-substituted saturated urushi ol crown ether (DHSU14C4), and dibutyl-unsymmetry-dibenzo-14-crown-4-dihydroxy crown ether (DBUD14C4) and investigated their selectivity toward aromatic amines. Sol solution was prepared by thorough mixing of 8 mg OH-DB14C4, 90 mg OH-TSO, 50 L 3-(2ation for aromatic amine extraction, optimum extraction temperature was fo be 55 C. To increase extraction efficiency, th e pH of the matrix wa s adjusted to 13. T compare the selectivity of diff t coatings we re employed (OH-DB14C4/ OH-TSO, DHSU14C4/OH-TSO, DBUD14C4/OH-TSO, OH-TSO, OH-DB14C4). Resu lts obtained by those researchers ed that OH-DB14C4/ OH-TSO (65 m) had the high est extraction efficiency. Among the three sol-gel crown ether coatings, extraction e fficiency decreased with increasing number of alkyl groups (no alkyl group> n-C 15 H 31 > t-Bu) attached to the crown ether ring attribut ed to the decreased polarity of the coatings and increased steric hindrance. Among all the coatings tested, so l-gel OH-DB14C4 had the lowest extraction efficiency. In sol-gel OH-DB14C4 coating, no OH-TSO was added to lengthen the chains

PAGE 140

116 d en ial es on the of the sol-gel network to ach ieve larger pore structure and increased surface area. The presence of OH-TSO in the solgel structure also helps the stationary phase to spread uniformly onto the surface of the substrate. Non-uniform coating, condensed network and low surface area may have been responsible for poor extraction efficien cy of the prepare coating. On the other hand, all sol-gel crown ether co atings demonstrated higher extraction efficiency than sol-gel OH-TSO This clearly demonstrated the higher selectivity of crown ethers for polar compounds. Another comparison was made betwe sol-gel OH-DG14C4/OH-TSO (65 m), comme rcial CW-DVB (65 m) and commerc PDMS coatings. As the extraction process is controlled by the principle like dissolv like, polar aromatic amines should have highe r affinities for the polar coating than that for the non-polar coating. Therefore, it is expected that CW -DVB and sol-gel OHDG14C4/OH-TSO being polar sorbents would extract higher amount of aromatic amines than a non-polar PDMS coating. The extracti on results corroborated with the predicti putting PDMS coating as the least efficient for aromatic amine extraction. Among other two polar coatings, so l-gel OH-DG14C4/OH-TSO and CW-DVB coatings, the sol gel OH-DG14C4/OH-TSO coating demonstrated unambiguous superiority for aromatic amine extraction (Figure 3.5).

PAGE 141

Figure 3.5 Relative extraction efficiencies of commercial PDMS fiber (100 m), commercial CW-DVB (65 m) and OH-DB14C4/OH-TSO (65 m) in amine extraction (A-aniline; MTm-toluidine; NNDEA-N,N-diethylaniline; NEMTN-ethyl-m-toluidine; 3,4DMA3,4-dimehylaniline) (Reproduced from Ref. [31] with permission of Elsevier) 117

PAGE 142

118 The inherently high surface area as well as high sample capacity of sol-gel coatings, originate from their porous structur e, are believed to be responsible for the higher extraction efficiency. All sol-gel crown ether coatings demonstr ated very high ther mal stability (~ 340 C). Their extraction ability showed no signifi cant change even after 150 extractions. Although sol-gel DB14C4/OH-TSO has been pr oved to be a superior coating in many respect compared with commercial coa tings, it represent a relatively low polarity coating [32,33]. Furthermore, its low solubility in so l solution limits the concentration of DB14C leading to low sample capacity as evidenced by the very na rrow linear range of g 15C5). Having bigger ring size, dihydroxy-terminated benzo15-crown-5 should provide higher polarity compared to that of smaller OH-DB14C4 because of the bigger crown ether ring and lower steric hindrance for hydroxy [34]. Moreover, higher solub ility of such crown ether in sol solution should increas e the sample capacity. The presence of both polar and nonpolar functionality in the sol-gel composite material should facilitate extraction of both polar and nonpolar analytes, particularly aromatic com pounds. The synthesis of benzo15-crown-5 has been reported elsewhere [35]. In preparing the sol solution for sol-gel coating 4 phenol [30]. In order to eliminate the inherent sh ortcoming of smaller crown ether, Wanet al. [34] synthesized dihydroxy-terminated benz o-15-crown-5 (DOHB 20 mg DOH-B15C5, 90 mg OHTSO, 10 mg PMHS, 100 L TEOS, 200 L methylene chloride was added and thoroughly mixed. Finally 80 L TFA (5%) was added with ultrasonic agitati on and the mixture was centrif uged to remove particulate materials (if there is any). The fused silica fiber to be coated was dipped into the sol solution for 5 min. For each fiber, the coat ing process was repeated several times each

PAGE 143

119 time for 2 min in the same sol solution un til the desired thickness of the coating was obtained. Using the same sol solution repe atedly in the whole coating process (as opposed to using fresh sol solution every time after finishing the residence time [31] of the fiber in it) minimized the total coating tim e and reduced the material cost. The coating thickness of sol-gel DOH-B15C5/OH-TSO was determined 67 m. Figure 3.5 shows an SEM image of sol-gel DOH-B 15C5 coating illustrating its distinct porous structure.

PAGE 144

Figure 3.6 Scanning electron micrograph of the DOH-B15C5 coated fiber at 800fold magnification (Reproduced from Ref. [31] with permission of Elsevier) 120

PAGE 145

121 Using the same coating protocol, fibers having different coating thickness as well as different concentration of DOH-B15C15 were obtained. In order to assess the role of DOH-B15C15 in coating, different mass (0, 10, and 20 mg of it) was added in the sol solution that yielded the concen tration of crown ether 0, 4.8, and 9.1% respectively. The reproducibility of the coating procedure was verified by coat ing three fibers under same conditions. The coating thicknesses were 65-, 67-, and 69 m, resp ectively, which is clearly indicative of high reproducibility of the coating process. Extraction of different phenols derivatives using these fibers yiel ded low RSD values (i n peak area) ranging from 2.20 to 4.94%. Maximum operating temper atures of different fibers were also evaluated. An SPME fiber with 100 m th ick blank sol-gel coating (no DOH-B15C5) showed cracking at 300 C. The same cracking was observed in 30 m blank sol-gel fiber at 320 C. Such cracking was not apparent in 80 m sol-gel DOH-B15C5/OH-TSO (10 mg) and 70 m sol-gel DOH-B15C5/OH-TSO (20 mg) even after conditioning at 350 C. Such a high operating temperature became po ssible due to the strong adhesion of the coating to the fused silica fiber surfa ce through chemical bonding. Such high thermal stability also helped elimin ating analyte carryover problem and expanding the application range towards high boiling compounds. It became obvious that addition of crown ether in the sol solution significantly increased the ther mal stability of the composite material. In addition to high thermal stability, solgel DOH-B15C5/OH-TSO coating demonstrated excellent solvent stability another indication of strong ch emical bonding of the coating to the substrate. The presence of DOH-B15C5 molecules in the sol-gel network served as a selector for compounds with conjugate electric system. Therefore, such composite sorbents should extract aromatic compounds with high efficiency. BTEX, being one of

PAGE 146

122 ere tion PME ercial s ct differen tion the major environmental contaminants, were extracted using sol-gel DOH-B15C5/OHTSO coated fiber (34 m thickness, 10 mg crown ether). The extraction conditions w also optimized: 2 min extraction at 30 C, 1 min desorption at 280 C, aqueous solu saturated with NaCl and constant stirring were found to be optimum for headspace S of BTEX. In order to compar e the selectivity of different coatings [30 m comm PDMS, 30 m sol-gel PDMS, 34 m solgel DOH-B15C5/OH-TSO (10 mg crown ether)], BTEX solution (10 g/L) was em ployed for headspace SPME. The extraction results demonstrated that 34 m sol-gel DOH-B15C5/OH-TSO (10 mg crown ether) fiber had the highest extraction efficiency and th e 30 m sol-gel PDMS fiber was the least efficient for BTEX extraction under the used co nditions. It is believed that the presence of phenyl in DOH-B15C5 extends interactions with BTEX compounds leading to high extraction efficiency. On the other hand, absence of such functionality in commercial PDMS coating made it in efficient for BTEX extraction. Another class of compounds extracted by sol-gel DOH-B15C5/OH-TSO wa phenols. Most phenols could be forced into the headspace by exploiti ng salting-out effe using NaCl. Acidifying the solution to pH 1 helps phenols maintain their neutral form. Optimum extraction conditions were: 40 min extraction at 40 C, 2 min desorption at 300 C, aqueous solution saturated with NaCl, pH 1 and constant stirring. Selectivity of t coatings (100 m PDMS, 100 m bl ank, 85 m PA, and 67 m crown ether) for phenol extraction was also studied and compared. As was expected, 67 m DOH-B15C5 was found to be the most efficient and 100 m PDMS the least. Although both sol-gel blank and PDMS had the same thickness, solgel blank demonstrated higher extrac efficiency. The excellent extraction effici ency demonstrated by sol-gel DOH-B15C5/OH-

PAGE 147

123 rown n of OCPs using s in 40 ith m coating on capture detector (ECD) has been proved TSO is due to enhanced surface area as well as sample capacity, hydrogen-bonding between the crown ether and phenolic compounds, polarity of the coating, complete desorption due to high thermal stability and so on. Sol-gel DOH-B15C5/OH-TSO coating also efficiently ex tracted a group of carcinogenic aryl amines and showed good selectivity. Application of sol-gel D OH-B15C5/OH-TSO coating has also been extended to trace analysis of organochlorine pe sticides (OCPs) in water [36]. In this case 9.1% c ether was used in the sol so lution and the coating thickness was 80 m. As the extractio efficiency is heavily dependent upon the ex traction conditions and the property of the coating, it is very important to optimize the extraction conditions as well as finding a suitable coating for extracting a particular compound class. In the extraction ol-gel DOH-B15C5/OH-TSO coating, ex traction equilibrium was achieved min under constant agitation, solution saturated with NaCl and temperature maintained at 90C. In a similar fashion, desorption was also very fast and completed in 60 s at 280 C. The selectivity of sol-gel DOH-B15C5/OH-TSO coating for OCPs was compared w commercial PDMS coating. As the results i ndicated, extraction efficiency of the 80 thick crown ether coating wa s significantly higher than th at for 100 L PDMS even though the latter one has higher stationary phas e loading. Thus, solid phase microextraction of OCPs using highly selective sol-gel DOH-B15C5/OH-TSO coating and thereafter detection by highly sensitive el ectr to be a very powerful analytical tool for trace analysis of halogenated organic compounds. Oganophosporuos pesticides (OPs) have l ong been considered as a health and

PAGE 148

124 environmental hazard due to its toxicity and ubiquity in nature. As a result, there is always a great demand for an analytical met hod that is cheap, simple, highly sensitive as well as fast that would ease the monitoring of trace levels of OPs in water, food and other matrices. Yu et al. [37] synthesized allyloxy bisben zo 16-crown-5 trimethoxysilane and used it as a precursor to prepare sol-ge l bisbenzo crown et her/hydroxy-terminated silicone oil coating for SPME. Unlike othe r silicon oil based mixed sorbent systems where the loading of sec ond organic compound is limited by its low solubility and translates into poor to moderate impact on the selectivity of th e composite sorbent, allyloxy bisbenzo 16-crown-5 trimethoxysil ane is highly soluble in sol solution ingredient and may be added to OH-TSO in a greater ratio. Sol solution was prepared using 75 mg of allyloxy bisbenzo 16-cro wn-5 trimethoxysilane, 90 L of OH-TSO, 50 L TEOS, 10 L of PMHS, 300 L of methylene chloride a nd 80 L TFA (5% water). In most polysiloxane-based mixed sorbents, a second organic mo iety is used in conjunction with silicone oil. But in this case, al lyloxy bisbenzo 16-crown-5 trimethoxysilane was used as the sol-gel precursor Due to high solubility of allyloxy bisbenzo 16-crown-5 trimethoxysilane in sol solution ingredients, th e ratio of the mass of benzo-crown ether in this coating was about 33% which is much la rger than crown ethe r concentration (~3%) in sol-gel SPME fiber prepared by common sol-gel method [30]. Due to the strong chemical bonding of the sorbent to the substrate, such sorb ent demonstrated very high thermal stability (350 C). Furthermore, the sol-gel crown ether coating was treated with different solvents (e.g., n-hexane, methylene chloride, acetone, and distilled water) for1 h. No obvious change in extraction sensitivit y was observed from subsequent extractions.

PAGE 149

125 In developing an SPME method, finding th e best sorbent is a challenging task but optimizing other variables (e.g ., extraction time, ionic strengt h of the solution, extraction temperature are) also equally important. In any extraction process, matrix temperature has two fold influences on extraction. Higher temperature increases the diffusion coefficient of analytes in water and thus s hortens the extraction tim e. On the other hand, elevated temperatures decrease the partition co efficient between the coating and analytes because molecular level interactions between th e analytes and the co ating decreases at a higher temperature. It is im portant to find a temperature that provides a reasonable the compromise between two opposing factors. Op timum extraction temperatures were found to be 55 C, 50 C, 32 C, 20 C for honey, orange juice, water and pakchoi matrix, respectively. Extraction equilibria were achie ved for most of the OPs within 120 min. Addition of NaCl helped the extraction of OPs from water, honey and orange juice matrices increasing the peak areas from 10 to 300% depending on the analytes. On the other hand, the addition of NaCl to pakchoi sample decr eased the response of OPs which is because the suspended pieces of pakchoi in heterogeneous system may easily adsorb the OPs although salt can decr ease the solubilities of OPs in water solution. Sample matrix plays a very important role in extraction efficiency. When extractions are carried out from matrices othe r than water, dilution by water may increase extraction efficiency. As was evidenced in th e extraction of OPs from different matrices, the peak area obtained by SPME-GC-FPD for homogenates (honey, orange juice, pakchoi) were less than that from aqueous so lution spiked at the same concentration. Dilution with water slowly increased the peak area of the extracted analyte(s). Interestingly, when diluted to 50-100 times with water, the peak area for most OPs

PAGE 150

126 increased and were higher than the SPME-GC peak areas for same concentration aqueous samples of these OPs prep ared through spiking. The same group also reported sol-gel dibe nzo-18-crown-6 coating for solid phase microextraction of aliphatic amines in lake water and human urin e after treating them with a new derivatizing reag ent, tetraflurobenzoic acid N-hydroxysuccinimide ester (TFBza-suc) [38]. Thermal stability and lifetime of the coating, extraction properties, and optimum extraction conditions were also investigated. Extraction sensitivity and selectivity of the new coating were also compared with two other sol-gel crown ether coatings (3 -allylbenzo-15-crow-5 an d allyloxyethoxymethyl-18-crown-6) as well as commercial PDMS and PA coatings. Extrac tion linearity, reproducibility and method detection limits were also calculated. Aliphatic amines are ubiquitous in nature due to their widespr ead use in industry as well as spontaneous genera tion as biodegradation products of organic compounds like proteins and amino acids or other nitrogenous compounds. Du e to their toxicity and hazardous nature, low-molecular-mass amines are considered to be important air pollutants. Moreover, secondary aliphatic amines are assumed to react with nitrile to form carcinogenic nitrosamines [39]. Therefore, analys is of aliphatic amines at trace levels in biological fluids, air, and water is of gr eat interest. GC is among many analytical approaches routinely used for aliphatic amin e analysis. But because of the high polarity and hydrogen bonding properties of aliphatic am ines, they often produce peak tailing and memory effects. Therefore, derivatization be fore extraction and/or analysis is a common practice. Derivatizing re agents, however, often interferes in separation, cause damage to the column, and in some instances, react w ith hydroxyl, phenol, th iol, and amine groups

PAGE 151

127 leading to poor selectivity for amine detection [38]. Cai et al. [38] utilized a new derivatizing reagent TFBza-suc th at specifically reacted prim ary and secondary aliphatic amines under mild conditions, but not with aromatic amines, hydroxyl, thiol, and phenol groups. The derivatized aliphatic amines are easy to separate by GC and detect by FID or MS. Before solid-phase microextraction, al l aliphatic amines were derivatized by TFBzasuc. To create the sol-gel coating, sol soluti on was prepared using one of the crown ether polymers, an alkoxysilane precurso r (TEOS), a surface deactivation reagent (PMHS), a suitable solvent (methylene chloride ) and a sol-gel catalys t (TFA, 5% water). Besides, VTEOS was added to impart organic character to the silica glass. The vinyl groups in VTEOS and the allyl crown ethers re acted to form cross-li nks in presence of free radical initiator AIBN when exposed to ultraviolet light (125 W, 366 nm). As a result, a surface bonded three dimensional netw ork evolved. The coating thicknesses of three sol-gel coatings DB18C6/OH-TSO, B15C5/OH-TSO, and PSO 18C6/OH-TSO were 80-, 84and 82 m, respectively. The thickn esses of two commercial coatings, PDMS and PA, were 100 and 85 m, respectively. The ex traction efficiencies of all five coatings were compared by extracting derivatized ali phatic amines under iden tical conditions. The results indicated that sol-gel DB18C 6/OH-TSO and B15C5/ OH-TSO had higher extraction efficiencies than sol-gel PSO18C6/OH-TSO. The pr esence of benzyl group in these crown ether is thought to be responsible for higher extr action of aliphatic amines by interactions with TFBza-suc-derivatized am ines. Particularly, the symmetric benzyl groups and greater number of oxygen atoms in the crown ether ring in sol-gel DB18C6/OH-TSO resu lts in stronger interactions between the coating and the

PAGE 152

128 derivatized amines, and therefore, made th is coating most efficient among all five coatings employed in this investigation [38]. Moreover, the extraction efficiencies of two commercial coatings were lower than the cr own ether coatings, PDMS having the lowest. Enhanced surface area and sample capacity provided by sol-gel coating technology is one factor that helped achieve this enhanced extraction sensitivity. Like other sol-gel coatings, sol-gel crown ethers also demons trated high thermal stability. The highest operating temperatures of sol-ge l DB18C6/OH-TSO, B15C5/OH-TSO, and PSO18C6/OH-TSO coatings were reported to be 350, 340, and 320 C, respectively. The presence of two benzene rings on opposite ends of the crown ether ring may have provided relatively higher thermal stability of sol-gel DB18C6/OHTSO. Finally, the solgel DB18C6/OH-TSO coating was used for solid phase microextraction of derivatized aliphatic amines from lake water and urine samples. Excellent re producibility (RSD = 34%), high recovery (92-109%), and low detection limits (0.05-0.005 gL-1) accomplished by the new coating have added new dimensions in amine analysis. Considering the utmost necessity of therma lly stable polar coatings in solid phase microextraction, Yun [40] proposed a -diallyltriethylene glyc ol/hydroxyl terminated silicone oil (DATEG/OH-TSO) coated fi ber prepared by sol-gel technology and investigated its applicability to the extraction of polar compounds (e.g., phenols) as well as nonpolar compounds (e.g., BTEX and phthalate esters from aqueous solution). The sol solution used in preparing the coating was made using 90 mg DATEG, 90 mg OH-TSO, 100 L TEOS, 50 L VTEOS, 9 L benzophenone, 120 L methylene chloride and 120 L TFA (5% water). As is seen from the composition, a remarkably higher amount of DATEG was added to the so l solution which was possible due to its

PAGE 153

129 high soluility in the sol solution. Presen ce of higher mass of DATEG in sol solution ds om aq ce time, fresh sol solution was used to replace the previous solution. The final coating thickness was 55 m. nlike commonly used sol-gel process wh ere only one metal alkoxide (TEOS or MTMO) is used, Yun utilized VTEOS in conjunction with TEOS which reacted with DATEG by radical cross-linking reaction under ul traviolet radiation to produce chemical bonding between crown ether and other coating ingredients. ction S s. As was expected, DMS being nonpolar and the thickest coatings used, should have extracted the nonpolar t sol-gel DATEG/OH-TSO performed best in extracti ng BTEX from aqueous solution (Figure 3.7). Th oth b would eventually translate in to higher polarity of the solgel hybrid sorbent. As a consequence, such sorbent should be able to extract higher am ounts of polar compoun frueous media. The fibers were dippe d into the sol solution for 30 min each time. After the 30-min residen U S Phenols, one of the major class of envir onmental pollutants, were extracted using sol-gel DATEG/OH-TSO and the results were compared with those obtained on a polar CW-DVB (65 m) coatings. Although CW and DATEG have similar polarity, extra results indicated higher ex traction efficiency of solgel DATEG/OH-TSO. Enhanced surface area of the solgel coating is believed to be responsible for the superior performance of the sol-gel coating. Next class of environmental pollutant s extracted by the sol-gel DATEG/OH-TSO fiber was BTEX. For comparison of extracti on efficiency, CW/DVB (65 m) and PDM (100 m) were also employed under the same extraction condition P BTEX with highest extracti on efficiency. But the result s indicated tha is became possible due to the fact that sol-gel DATEG/OHTSO coating has b

PAGE 154

130 polar (DATEG) and nonpolar (OHTSO) structural features as well as enormous surface area, a characteristic feat ure of sol-gel coatings.

PAGE 155

131 Figure 3.7 Comparison of extraction efficiencies of commercial PDMS (100 m), commercial CW/DVB (65 m), and DATEG/OH-TSO coatings in extrac BTEX from aqueous solution (Reproduced from Ref. [40] with p Elsevier) ting ermission of

PAGE 156

132 High thermal stability of sol-gel DATEG/OH-TSO (320 C) allowed for efficient extraction and desorption of high-boiling compounds as opposed to commercial coatings which have relatively low recommended operating temperatures (mostly <300C). Thanks to the ability of sol-gel DATEG/OH-TSO coatings to extract and desorb high boiling compounds, SPME fibers with such co atings were used to extract several phthalate esters, a class of high-boiling contaminants. Dire ct-SPME of phthalate esters were done followed by desorption at 320 C. No noticeable decline in extraction capacity even after 150 cycles of extraction-desorpti on clearly demonstrated superiority of this sol-gel coating compared to availa ble commercial counterparts. 3.3.1.2.3 Mixed calix[4]arene-poly siloxane sol-gel sorbents Calixarenes are the third generation host molecules, the first two being crown ethers and cyclodextrins. They are cyclic oli gomers prepared from th e reaction of phenols and aldehydes. In recent years, calixaren es have received notable attention in supramolecular chemistry. Since they po ssess a molecular cavity of cylindrical architecture similar to that of cyclodextrins, they can form inclusion complexes. The unique characteristics of calixarenes such as small molecular size, good film forming properties, excellent thermal stability, presence of functi onal groups to interact with analytes have made them promising candidates for being used in SPME sorbents. However, the poor solubility of unsubstituted calixarenes in common organic solvents as well as in chromatographic phases (e.g., polys iloxanes) has appeared to be the major obstruction to their widespread use in analytical chemistry. Table 3.5 provides a list of

PAGE 157

133Solvent Stability ce sol-gel calixarene sorbents used in SPME. Table 3.5 Sol-gel calixarene sorbents used in SPME Sorbent Precursor Organic component Thermal/ Referen Sol-gel CaTEOS lixarene 380 C [41] Sol-gel TEOS Calixarene 380 C [79] S t-Bu ol-gel Calixarene TEOS, 3-(2-cyclooxypro poxyl)propyltrimethoxysilane 380 C OEt OEt OH HO t-Bu t B u t-Bu [45]

PAGE 158

134 Zeng and co-workers [41] represent the only rese arch group involved in developing calixarene based sol-gel sorbents for solid phase microextraction. These researchers synthesized 5,11,17,23-tetra-tert-butyl-25,27-diethoxy-26,28dihydroxycalix[4]arene and utilized it as a new SPME sorbent. A sol-gel based approach was followed to prepare the coating a fu sed silica fiber using calix[4]arene and hydroxyterminated silicone oil (OH-TSO ) as organic components, 3-(2Cyclooxypropoxyl)propyltrimethoxysilane (KH-560) and TEOS as sol-gel precursors, poly(methylhydrosiloxane) (PMHS) as deactivat ing reagent, and trifluoroacetic acid (TFA) (5% water) as the sol-gel catalyst. Ca lix[4]arene has limited solubility in OH-TSO. To optimize the amount of calix arene that can be dissolved in OH-TSO, variable amounts of calix[4]arene (0 mg, 20 mg, 30 mg, 40 mg ) ware added in 90 mg OH-TSO along with other sol-gel ingredients a nd the sol-gel coatings ware prepared. OH-TSO was added along with calix[4]arene in the sol solution in order to lengthen the silica network to achieve increased surface area of the coating. Based on SEM data, the prepared coatings had thickness estimated at 65 m (for OH-TSO ), 55 m [for C[4] (20 mg)-OH-TSO], 60 m [for C[4] (30 mg)-OH-TSO], 57.5 m [for C[4] (40 mg)-OH-TSO]. As is evidenced from the thickness data, the sol solution c ontaining 40 mg C[4] generated smaller thickness than the solution containing 30 mg C[4] This is probably due to the fact that C[4] (40 mg) exceeded its solubi lity in the sol solution and eventually precipitated. All calix[4]arene coated fibers were conditioned first at 280 C and then step by step the conditioning temperature was rais ed to as high as 380 C in 20 C increments staying for 1 h at each step. The extraction data revealed no significant loss of extraction efficiency even at 380 C. Of course the credit goes to sol-gel coating technology that made it

PAGE 159

135 possible to create coatings of such high thermal stability. Since calix[4]arene contains phenyl termini as well as a cavity in its structure, it is expected to exhibit high selectivity toward nonpolar aromatic compounds due to interaction, hydrophobic interact ions and cavity-shaped cycl ic molecular structure. Moreover, such coatings s hould efficiently extract polar aromatic amines through hydrogen bonding and dipole-dipol e interactions. As was expected, calix[4]arene coatings demonstrated very high efficien cy in extracting BTEX, PAHs as well as aromatic amines. To compare the extraction capability of sol-gel calix[4]arene-OH-TSO and sol-gel OH-TSO, BTEX was extracted using both the coatings under similar extraction conditions. The amount of BTEX extracted by unit volume of sol-gel calix[4]arene-OH-TSO coating was far greater than that by the sol-gel OH-TSO coating. This was clearly indicative of the positive attribute of calix[4]arene in the coating. The same trend also continued in case of extr acting aromatic amines. Like other sol-gel coatings, sol-gel calix[4]arene-OH-TSO was a ssumed to be highly porous as manifested by its extraordinarily shorter extraction equi librium time (20 min for fluorene in HSSPME). On contrary, Nguyen and co-workers [42] reported 1-2 h ex traction equilibrium time for PAHs on a laboratory-made PDMS coating. Thus, sol-gel coated 5,11,17,23-tetra-tertbutyl-25,27-di-ethoxy-26,28dihydroxycalix[4]arene coating exhibited high se lectivity, sensitivity and faster extraction equilibrium for a wide range of analytes including BTEX, PAHs, and aromatic amines. Very high thermal and solvent stability, excel lent coating and extraction reproducibility, long life span, and low detection limits for ex tracted analytes have made such a coating promising and beneficial in SPME.

PAGE 160

136 In solid phase microextraction, there is a substantial lack in polar coatings. Although commercial phases like Polyacr ylate (PA), Carbow ax/divinylbenzene (CW/DVB), Carbowax/ Templated resin (CW/ TPR) and several other custom made phases have been used as polar sorbents, all of these sorbents possess an inherent shortcoming low thermal stability. In mo st of the cases, the recommended maximum operating temperatures are below 300 C, putti ng a serious limitation to effective thermal desorption of polar analytes. For example, Molder et al. [43] extracted phenols from wastewater using PA fiber. After 5 min of desorption at 280 C, a small carry-over of pentachlorophenol was observed. When desorp tion was done at 300 C for 5 min, carryover problem was gone but after 10 SPME runs the fiber demonstrated a drastic loss of extraction performance. This indicated the poor thermal stability of such phases. In an attempt to contribute to the great demand of polar sorbents, Zeng group [44] utilized previously reported sol-ge l calix[4]arene-OH-TSO [41] for extracting chlorophenols. As was mentioned before, this phase demonstrated excellent ther mal (~ 380 C) as well as solvent stability. For comparison, extracti on of chlorophenols was carried out using commercial PA (85 m), commercial PDMS/DVB (65 m) and in-house prepared sol-gel calixarene C[4]/OH-TSO (85 m) sorbents. As the results indicated, sol-gel C[4]/OHTSO demonstrated highest extraction efficien cy for all PCPs except 2 CP which being extremely polar showed higher affinity toward s more polar PA sorb ent. In case of 2,4 DCP, both C[4]/OH-TSO and PA demonstrated almost sa me extraction efficiency. The high extraction efficiency of C[4]/OHTSO is believed to be due to the interaction, hydrophobic interactions and cavity-shaped cyclic molecular structure. Same C[4]/OH-TSO sorbent was also utilized for the determination of phthalate

PAGE 161

137 ent in young girls, development of breast cancer in tants r o of o water and used for direct SPME. SPME experim mperature for PAEs extracti on. Polar analytes with shor t alkyl chains required 60 min r equilibrium whereas nonpolar analytes with long alkyl chains required longer m if xtraction conditions are maintained [47], all extractions we re carried out for 40 ounts of NaCl were added to verify its impact on extraction sensitivity acid esters plasticizers in polymeric material s by ultrasonic solvent extraction combined with solid phase microextra ction [45]. Phthalic acid es ters (PAEs) are the most commonly used additives in plastic and are be lieved to be the cause of serious health conditions (premature breast developm humans, and so on). Considering the serious health implications of PAEs, the US Environmental Protection Agency (EPA) and analogous regulatory ag encies in several other countries have classifi ed the commonly occurring phthal ates as priority pollu [46]. This implies a great demand of a simple, low cost and rugged method fo determining PAEs in water. The sol-gel coated 5,11,17,23-te tra-tert-butyl-25,27diethoxy-26,28-dihydroxycalix[4 ]arene/hydroxy-terminated silicone oil coated SPME fibers were employed to determine the conten ts of phthalate esterbased plasticizers in blood bags, transfusion tubing, food packaging bag, and mineral water bottle. In order t transfer PAEs from PVC products, they were cut into small pieces and placed in a vial. A small volume of methanol was added to the vi al and was sonicated for 30 min. A part the methanolic extract was then added t ental conditions (e.g., extraction temperature, extraction time, impact of salt addition, desorption temperatur e and time) were optimized to obtain the best possible and reproducible results. Experimental results demonstrated that 30 C was the optimum te fo equilibration time due to slow diffusion. Since it is not necessary to attain equilibriu constant e min. Different am

PAGE 162

138 and a concentration of 180 g/L was found to be optimum. Desorption temperature of 300 C was found adequate for complete release of all of the extracted PAEs in 1 min. Relative affinity of PAEs were inve stigated by employi ng C[4]/OH-TSO and three commercially available fibers (P DMS, PA, and PDMS/DVB). The results unequivocally demonstrated the superiority of C[4)/OH-TSO among a ll four phases. It was assumed believed that the presence of phenyl rings and hydrophobic cavities of C[4)/OH-TSO helped extracti ng aromatic compounds through interactions, hydrophobic interactions, and steric interactions via cavity-shaped cyclic molecular structure. Also, hydrogen bonding and dipole-dipole interacti ons between the analytes and the coating apparently played impor tant roles in the extraction process. Amines are important environmenta l pollutants which may occur as biodegradation products of orga nic matters like proteins, am ino acids and other nitrogen containing organic compounds. Th ey are also used as raw materials or intermediate products for manufacturing a wide range of industrial chemicals. Many amines possess unpleasant odor and harmful to health because of their toxicity. Due to the high polarity and solubility of low-molecular weight amines in water, it is very difficult to extract them without derivatization. Zeng and co-workers [79] utilized previously reported calix[4]arene after slight modification in the structure that allowe d headspace solid phase microextraction of underivatized aliphatic amines from a queous solution. 25,27-dihydroxy-26,28-oxy(2 ,7 dioxo-3 ,6 -diazaoctyl)oxy-p-tert-butylcalix[4]a rene (amide bridged C-[4]) was synthesized in a view that the presence of polar amide bridge in the calix[4]arene structure would increase the affi nity of the sorbent for amines The fibers were prepared

PAGE 163

139 4]arene retained all of butes (e.g., high thermal stability (~380 C), solvent stability, long life lsiloxane d ng kely that the are following a procedure discussed before [41]. The modified calix[ its positive attri span, as well as highly porous surfac e morphology). Owing to the introduction of the polar amide bridge in calixarene molecules, the polarity of the coating increased. As a consequence, such coatings exhibited better sensitivity to most of the investigated aliphatic amines compared to commerc ial polar coatings PDMS/DVB and PA. 3.3.1.2.4 Mixed polyvinyl alcohol (PVA)polysiloxane sol-gel sorbents Lopes et al. [48] prepared a composite sol-gel sorbent using polydimethy (PDMS) and poly (vinyl alcohol ) as organic moieties. Polyvinyl alcohol was incorporate in the growing sol-gel network via polycondens ation and acted as a strong cross-linki agent [49]. The thermal stability of the com posite phase has been found to be superior compared to sol-gel PDMS sorbent. The improved thermal stability of sol-gel PDMS/PVA can be attributed to the additional cross-linki ng provided by the PVA in the reaction mixture. Another important piece of information has been obtained from the thermogravimetric analysis (TGA). Results showed that after heating at 750 C, sol-gel PDMS/PVA retains 36% of its original weight whereas sol-gel PDMS retains 30%. The retained mass represents mostly inorganic silica content of the networ k. It is li comparatively higher inorganic content in sol-gel PDMS/PVA imparts higher thermal stability (350C). It is worth noting that e nhanced thermal stability of SPME sorbents extremely desirable because higher thermal stab ility of the sorbents can extend the array

PAGE 164

140 from herbal infusions of Passiflora L. e sol-gel sorbene of me of analytes amenable to SPME toward hi gher-boiling compounds. Moreover, bleeding from the fiber coating during the thermal deso rption is less likely to happen. The new solgel PDMS/PVA has shown bette r affinity towards polychl orinated biphenyl (PCBs) compared with sol-gel PDMS coating. The same group also expanded the applic ation of sol-gel PD MS/PVA sorbent by successfully employing it in trace determination of organochlorine pesticides (OCP) and organophosphorous pesticides (OPP) 3.3.1.2.5 Mixed polymethylphenylvinylsil oxane (PMPVS)polysiloxan ts Yang et al. [50] described sol-gel poly(methylphe nylvinylsiloxane) coating which is characterized by good thermal stability, long life time, and high extraction efficiency for a wide range of compounds. Due to the inherent multifunctional properties and th features of sol-gel chemistry, sol-gel PMPV S coating was expected to demonstrate good selectivity for both polar and nonpolar analytes. The sol solution was prepared by thoroughly mixing 40 mg of PM PVS, 90 mg of OH-TSO, 100 L of TEOS, 50 L of VTEOS, 10 mg of PMHS a nd 8 mg Benzophenone, 400 L of methylene chloride and 120 L of TFA (5% water). Instead of using single precursor, TEOS, another precursor VTEOS was also used in conjunction with TEOS. Upon e xposure to ultr aviolet light vinyl groups present in VTEOS and PMPVS reacted to form cross-links in presence Benzophenone. Furthermore, ultrasonic dryi ng technique was employed to overco

PAGE 165

141 shrinkage and cracking of the coating which is inevitable in conven tional drying process. Due to the presence of polarizable ph enyl groups in the sol-gel PMPVS/OH-TSO coating, it is expected that such coating should demonstrat e high affinity for aromatic compounds (e.g., BTX, polycyclic aromatic hydr ocarbons, etc.). Two different coatings (sol-gel OH-TSO, 65 m; and sol-gel PM PVS/OH-TSO, 55 m) were employed for headspace SPME of BTX to compare relative ex traction efficiency of both phases. As was expected, sol-gel PMPVS/OH-TSO exhibi ted higher affinity for BTEX due to interaction between the coating and anal ytes. Same trend wa s observed in PAHs extraction from aqueous media. Sol-gel PMPVS/OH-TSO coating (35 m) has been found to be more efficient than commercia l PA (85 m) and PDMS (100 m) coatings. The coating showed high thermal stability ( 350 C) and extended service life. 3.3.1.2.6 Mixed divinylbenzene-polysiloxane sol-gel sorbents Nerve agents are one of the most toxi c compounds. The detection and monitoring of nerve agents have received enormous im portance particularly in changed post 9/11 environment in the USA because of their pote ntial use as weapons of mass destruction. The nerve agents sarin, soman, tabun and VX are representatives of methylphosphonate nerve agents. Due to the absen ce of chromophore (or fluorophore) in their structure, it is difficult to use UV or fluorescence detection which are most commonly used in liquid phase separations. As a viable alternative, SPME-GC or SPME-GC/MS can be explored. One of the major factors that inhibi t the healthy growth of SPME is the availability of suitable coatings for the extr action of a particular class of compounds. In

PAGE 166

142 meric materials are ure selectivity of such sorben gel a le ater matrix Sol solution was pr epared by thoroughly mixing nto the d coating re a second precursor, VTEOS was used to facil and DVB, different ratios of commercial fiber manufacture process, one or more organic poly immobilized on the surface of the silica fibe r by mere physical deposition. When p organic polymers are used as the sorbents, it is difficult to fine-tune ts. Furthermore, lack of chemical bonding between the sorbent and the substrate is considered to be responsible for low thermal and solvent stability of such phases. Solcoating technology has overcome most of the above mentioned problems and provided very simple way to synthesize hybrid organic/inorganic com posite material with tunab selectivity, porous structure providing e nhanced surface area, and chemical bonding to the substrate responsible for hi gh thermal and solvent stability. Liu et al. [51] developed a sol-gel based new sorbent using hydroxy-terminated silicone oil and divinylbenzene polymers for selective extraction of phosphates and methylhosphonates from air and w 180 L of DVB, 60 mg of OH-TSO, 50 L of VTEO S, 10 mg of PMHS, 8 mg benzophenone, 100 L of methylene chloride and 70 L trifluoroacetic acid (5% water) In order to prepare the coati ng, specially treated fused sili ca fibers were dipped i sol solution and kept there for 30 min. Af ter that, new sol solution was employed an continued for another 30 min. When th e coating was completed, the fibers we irradiated under ultraviolet light for 30 min. Finally the fibers were thermally conditioned at 250-380 C under nitrogen flow. The sol solution ingredients were carefully chosen to serve specific purpose. In conjunction with commonly used precursor TEOS, itate free radi cal cross-linking with DVB under ultraviolet light. In order to optimize relative proportions of OH-TSO

PAGE 167

143 both ingredients were used a nd extraction efficiencies of the prepared coatings were evaluated. The following proportions of OH-TSO and DVB were used: 180 mg/0 L; 180 mg/180 L; and 60 mg/ 180 L. The extracti on results suggested that DVB played a major role in the selectivity of the composite material for extracting phosphates and methylphosphonates. The composition 60 mg OH -TSO/ 180 L DVB yielded the highest extraction efficiency. Another comparison was made between so l-gel OH-TSO/DVB and the following commercial coatings: PDMS, PA, and PDMS/DVB. Sol-gel OH-TSO/DVB coating showed the best extr action efficiency among all the phases compared. Like other sol-gel coatings, sol-gel OH-TSO/DVB coating is also characterized by very high thermal stability. After heating up to 380 C, no no ticeable loss in extraction efficiency was observed. It clearl y indicated the operational supe riority of sol-gel OH-TSO/DVB coating. As a result of high thermal st ability, it should provide higher desorption temperature for analytes with high boiling points, and also should eliminate sample carryover still considered to be a co mmon problem for commercial coatings. Other extraction parameters were also investigated and optimized. For headspace SPME of phosphates and methylphosphonates, 5 min extraction at 60 C was found optimum for air samples whereas extraction 15 min extraction at 32 C with constant stirring with 1 gm salt in a 5 mL water sample was found to be optimum for direct extraction.

PAGE 168

1443.3.1.2.7 Mixed polyphenylmethylsiloxane (PPM S)–polysiloxane sol-gel sorbents During the last couple of decades the use of pesticides has increased dramatically to enhance agricultural productivity. The high to xicity and widespread use of pesticides have been a major concern in recent years. Organochlorine pesticides (OCPs) are among the most frequently used pesticides. Their low solubility in water and resistance to metabolism has made them persistent polluta nts in the environment. Some OCPs were banned in 1970 [52]. Consider ing their impact on public h ealth and the environment, several approaches have been proposed to quantify OCPs in different matrices (e.g., water, plants, soils, f oodstuff, etc.) [53-55]. In an effort to simplify these time-consuming methods, Cai et al. [56] developed a microwave-assisted solvent extraction (MASE) and coupled it with SPME-GC/ECD for the determination of organochlorin e pesticides in Chinese teas. Polyphenylmethylsiloxane ( PPMS) and polymethylsiloxane (PMS) coatings were prepared using sol-gel technology and employed them as sorbents. The sol solution was prepared by mixi ng 40 mg PPMS, 100 L OH-TSO, 100 L TEOS, 50 L VTEOS, 10 L PMHS, 10 mg AIBN, 800 L methylene chloride, and 120L TFA (5% water). Coating was prepared by immersing a pretreated fused silica fiber into the sol solution. The coated fibers were then illuminated by ultraviolet light (125 W, 366 nm) for 60 min. followed by th ermal conditioning at 100, -200, -300 and 350 C for 1 h under nitrogen in the GC inj ector. The final coati ng thickness of the prepared sol-gel PPMS fiber was estimated at 70 m. Following the same procedure, another fiber coated with sol-gel PMS was al so prepared. The thickness of sol-gel PMS

PAGE 169

145 coating was estimated at 75 m. Microwave-assisted solvent extraction (MASE) is a relatively new technique which has been applied to the extraction of organic compounds from different matrices [57]. In recent years, MASE followed by SPME has been proven to be a useful combination that may speed up the whole extraction and analysis process [58-61]. In order to find the most suitable sorbent for OCPs extraction, three different coatings were employed: 70 m sol-gel PPM S, 75 m sol-gel PMS and commercial 100 m PDMS. Extractions of OCPs were perfor med in a microwave oven for 10 min. After cooling to room temperature, the extract was transferred to a vial and diluted with water to a predetermined volume for headspace SPME. To verify the effectiveness of MAE, another extraction was carried out using ultrasonic power for an hour followed by headspace SPME. Being nonpolar, OCPs should be extracted well on nonpolar coatings. Therefore, nonpolar sorbents like commercial PDMS as well as sol-gel PPMS and PMS should demonstrate high affinity towards OCPs. Extr action results revealed that both sol-gel PPMVS and PMVS coatings had better extr action efficiency than commercial PDMS although the latter had the highe st sorbent loading. This wa s justified by th e fact that coatings generated by sol-gel process were characterized by porous structure and enhanced surface area compared to commer cial coatings. Furthermore, sol-gel PPMS coating was found to be more efficient th an sol-gel PMS in extracting OCPs. The presence of phenyl groups in sol-gel PPMS is believed to be facilitating better extraction of OCPs through additional interactions. Thermal stability of an SPME sorbent is indicative of its operational superiority.

PAGE 170

146 The higher the maximum operational temperature, the less is the chance of memory effect and appearance of ghost peaks in the chroma togram. Thermal stability of both sol-gel PPMS and sol-gel PMS were also investig ated. No significant loss in extraction efficiency were observed after heating solgel PPMS coating at 350 C and sol-gel PMS coating at 320 C. Such high th ermal stability of sol-gel co atings is due to the strong adhesion of the coating to the substrate through chemical bonding. The better thermal stability of sol-gel PPMS coati ng is believed to be due to the presence of phenyl group in the sorbent structure. Finding the most suitable coating for ex tracting a particular group of compounds is a challenging job for separation scientists. In this regard, the optimization of SPME conditions which include extraction mode, extraction time, extraction temperature, impact of salt addition, etc. are also ve ry important. A comparison was made between headspace SPME and direct SPME by employi ng 100 ng/L concentration of OCPs for equilibrium extraction. Extraction results revealed that almost all OCPs except -HCH and aldrin have higher extraction efficienci es for headspace mode. Extraction equilibria were reached within 20 and 80 min depe nding on the compound. Under similar conditions, the commercial PDMS coating ne eded several hours to reach equilibrium [53]. Optimum extraction temperature was found to be 90 C. Salt addition increased the extraction efficiency; 5 g of NaCl was found optimum in 15 mL aqueous solution of OCPs. Microwave assisted extraction is par ticularly beneficial for extracting organic compounds from solid matrices. Therefore, it has been chosen for extracting OCPs from Chinese tea samples. The optimum MASE extraction condition for OCPs from tea was 80% irradiation power. To evaluate the perf ormances of microwave assisted solvent

PAGE 171

147 xtraction and ultrasonic extract ion, blank green tea spiked with OCPs were employed nd the extraction recovery re sults obtained under identical conditions were compared. he recoveries of OCPs by MASE-SPME-GC /ECD were 39.05-90.13 % for spiked 50 g/Lstandard OCPs and 43.47-101.17% for spik ed 100 ng/L standard OCPs. On the other and, recoveries of OCPs by USE-SPME -GC/ECD were 1.51-27.66% for spiked 50 g/Lstandard OCPs and 2.5-36.29% for spiked 50 ng/L standard OCPs. As is evident om the extraction results, MASE offered much efficient analyte rec overies with shorter xtraction time. Table 3.6 provides a list of miscellaneous polysiloxane-based sol-gel rbents used in SPME e a T n h n fr e so

PAGE 172

148 Table 3.6 Salient features of SPME sorbents based on sol-gel PVA, ethylsiloxane, divinylbenzene, and polymethylphenylvinylsiloxane. Sorbent PrecursorOrganic Component Thermal/ Solvent Stability Reference polyphenylm Sol-gel PVA TEOS CH2CHOHn 350 C [48] Sol-gel Divinyl bTEOS. enzene VTEOS CHHCCH2H2C 380 C [51] Sol-gel polymethylphenyl vinylsiloxane TEOS, VTEOS Si CH3 O xO Si CH 3 CH3 O Si CH CH3 CH 2 z y 350 C [50]

PAGE 173

149 of thumb, all chromatographic stationary phases should be able to serve 3.3.2 Non-polysiloxane sorbents As a rule as SPME sorbents. Among commercially available SPME sorbents, a few are nonpolysiloxane based polymers e.g., PA, CW/D VB, CW/TPR. Surprisingly, only a handful sol-gel coatings are non-polysiloxane based. A recent literature su rvey revealed that polyethylene glycol (superox, Carbowax 20 M), octyl -bas ed silica, dendrimers, polyTHF, cyclodextrins, and fulle rene are the only non-polys iloxane organic components used to prepare sol-gel sorbents.

PAGE 174

150 Table 3.7 Non-polysiloxane sorbents used in SPME Sorbent Precursor Organic Component Thermal/ Solvent Stability Reference Sol-gel PEG MTMOS C H H C H O H OH nH 300 C [62] Sol-gel C8 MTMOS Si OH2CH3C OCH2CH3 CH2(CH2)6CH3 OCH2CH3 Acetonitrile/ water [66] Sol-gel Hydroxyfullerene MTMOS C60OH n 360 C [27] Sol-gel PEG O H3C CH2CH2O nCH2 2N H C N O CH2 H Si OC2H5 OC2H5 OC2H5 3 Si OC2H5 OC2H5 C2H5O CH2 3N H C O N CH2CH2O H CH2 n Si OC2H5 OC2H5 OC2H5 3 300 C [22,24] Sol-gel Dendrimer MTMOS 320 C [79] Sol-gel polyTHF MTOS CH2 4O OH nH 350 C [80]

PAGE 175

151.3.2.1 Polyethylene glycol-based sol-gel sorbents The credit for the first report on utiliz ing Sperox-4, a PEG congener, in solid tive [18]. el en d a sible the 3 phase microextraction goes to Wang et al. [62] who exploited the widely known posi attributes of sol-gel SPM E coating technology develope d by Malik and co-workers Sol solution was prepared by mixing 200 mg Superox-4, 400 L MTMOS, 200 L acetone and 150 L TFA (5% water). The so lution was centrifuged to expel solid particulate material present in the matrix. The clear supernatant was used for sol-gel Superox-4 coating. In order to prepare the coating specially treated fused silica fibers were immersed inside the sol solution and ke pt there for 30 min. A 40-m thick sol-g coating was found to be formed during this residence time. The coated fibers were th conditioned at 300 C for 2 h under continuous nitrogen flow. Scanning electron microscopic investigati on revealed that the coating possesse roughened surface with a porous structure which is a highly desirable characteristic of an SPME sorbent. In order to verify the performance of sol-gel Superox-4 coated fibers, headspace SPME of BTEX solution (10 ng/mL) was carried out. The highly porous structure of solgel Superox-4 coating facilitated faster mass tr ansfer between the gas phase to the sorbent phase. Under constant stirring of the solution, extraction equilibrium was reached within 30 s for benzene and toluene, 40 s for ethy lbenzene and p-xylene, 90 s for o-xylene. Under same conditions, the commercial 100 m PDMS fiber required several minutes to reach equilibrium. The porous structure of solgel Superox-4 coating was also respon for faster and complete desorption of the ex tracted analytes and thereby minimized

PAGE 176

152 possibility of sample carryover or memory effect are frequently encountered by commercial coatings. The complete desorption of extracted BTEX was done within 20 s at 280 C. The porous structure of sol-gel S uperox-4 coating drama tically increased its available surface area as well as sample capacity for extraction. Therefore, even a thinner sol-gel coating was likely to demonstrate a higher extraction capability compared to commercial coatings. In order to compare relative extraction efficiency of sol-gel Superox-4 and four commerci al coatings (85 m PA, 100 m PDMS, 7 m PDMS, and 65 m CW/DVB), BTEX components were extracted under id entical extraction conditions and analyte mass in unit volume of each coating was calculated. The extraction results (Figure 3.8) clearly demonstrated the supe riority of sol-gel Superox-4 coating among all the coatings tested. Mass of BTEX absorbed per unit volume went in the following order: sol-gel Superox-4> CW/DVB> 7 m PDMS> 100 m PDMS> 85 m PA. Due to the presence of benzene ring in CW/DVB coating cap able of providing interactions toward the BTEX analytes, CW/DVB coating demonstrated high affinity towards aromatic compounds. Sol-gel coati ngs demonstrated excellent extraction efficiency, remarkably high thermal stability, and long life span. Even after 150 repeated extraction operations on sol-gel Superox4 coating, no sign of efficiency loss was observed. Sol-gel Superox-4 coated fibers were also used to extract phenols, an environmentally important class of compounds due to their toxicity and ubiquity and widespread use in the indus try. Being highly polar, PA coatings are most commonly employed for extracting phenols. In headspace-SPME using a

PAGE 177

Figure 3.8 Comparison of analyte mass in unit volume of sorbent using five different fibers to extract BTEX from 10 ng/mL aqueous solution (1. benzene, 2. toluene, 3. ethylbezene, 4. p-xylene, o-xylene) (Reproduced from Ref. [62] with permission from Elsevie n r) 153

PAGE 178

154 sol-gel Superox-4 coating ex traction equilibrium was r eached within 40 and 50 min depending on the compound. Since phenols ar e acidic, low pH value of the matrix ensures that analytes are in neutral form and thus enhanced extraction efficiency of the coating. The pH value was adjusted to 1 by adding HCl to the matrix. Salting out is another factor that provides higher extracted amounts. To explo it this factor, 3 g of NaCl was added to 10 mL of aqueous solution. Extraction was carried out on commercial 100 m PDMS, 85 m PA as well as on sol-ge l Superox-4 coatings. Optimum extraction temperature was found to be 30 C. As the affi nity of a particular coating is mostly determined by the principle like dissolves like, PDMS coating being nonpolar is not suitable for extracting polar compounds like ph enols. On the other hand, PA coating being highly polar should ha ve enhanced affinity toward polar compounds like phenols. Actually, both PA and sol-gel Superox-4 coatings demonstrated high affinity for phenols, the sol-gel Superox-4 outperforming the PA coat ing. It is believed that the presence of free hydroxyl group at the free end of the S uperox-4 polymer chain may have enhanced the polarity of the coating to some extent and thus contributed in ex traction efficiency of these polar analytes. Furthermore, highly porous nature of the sol-gel coating increased the surface area and sample capacity of the coa ting. As a result, a 40 m sol-gel Superox4 coating performed even better than a 85 m PA coating. Unlike commercial coatings, sol-gel coatings contain bot h organic and inorganic com ponents that provide unique ine pesticides. Sol-ge l Superox-4 coating demonstrated sufficient selec tivity for all three nonpolar compound classes. It is obvious selectivity to the coating. As a result, su ch coatings are suitable for both polar and nonpolar compounds. Sol-gel Superox-4 coating was also used to extract naphthalene congeners, phthalic diesters and organochlor

PAGE 179

155 that by varying the proportions of the sol solution ingredients or using surface deactivation reagent (to derivatize residual sila nol groups), the net composition as well as polarity of the coating can be changed lead ing to a change in selectivity. er e sol 1 peated five tim r Carbowax 20 M is one of the most p opular and extensiv ely studied polar polyethylene glycol based sta tionary phases used in gas chromatography. Augusto and co-workers [63] recently reported a sol-ge l method for chemical immobilization of Carbowax 20 M onto fused silica fiber and dem onstrated its performa nce superiority ov commercial PDMS and CW/DVB coatings in headspace SPME of BTEX. Physicochemical and morphological characterization of such coatin gs was also presented. Th solution was prepared by thorough mixing of 800 mg Carbowax 20M, 400 L MTMS, 500 L of TFA (5% water). The sol-gel coati ng was made by dipping fused silica fiber ( cm) into the sol solution. The fibers were ke pt inside the solution for 1 h and then fresh sol solution was used to continue the coati ng process. The coating process was re es so that the coating thickness became 8 m. The fibers were then conditioned at 230 C under conti nuous helium flow for up to 60 h. As was revealed from thermogravimetric analysis (TGA), a major lo ss in mass (39.7% of original coating) occurred at 108 C even though the sol-gel material was drie d at 110 C overnight prio to thermogravimetric analysis. It is believe d that conventional dr ying in oven was not adequate to release water and other low molecular mass reaction products entrapped inside the sol-gel polymeric networks. Another diminutive loss in mass was observed at around 230-250 C. This insignifi cant loss in mass may have been due to loss or degradation of unbound molecules of Carbowax 20M. Finally, a major weight loss was observed in the temperature range 375-413 C which accounted for 48.6 % of the initial

PAGE 180

156 of .g., mass. The leftover mass (11.8%) was believed to be pure SiO 2 As the first two events weight loss were likely to be merely due to th e elimination of extraneous substances (e water, low molecular mass of reaction pr oducts, unbonded molecules of Carbowax 20M etc.), the final event of weight loss was cons idered as characteristic thermal degradation of sol-gel Carbowax 20M and can be assigne d to its maximum operating temperature. Scanning electron micrograph of sol-gel Carbowax 20M coating [Figure 3.9] demonstrated a highly porous sponge-like network consis ting of agglomerates of microspheres with up to ~ 2 m in diameter.

PAGE 181

Figure 3.9 Scanning electron micrograph (600 x magnification) of a sol-gel Carbowax 20 M ormosil fiber (Reproduced from Ref. [63] with permission from Elsevier) 157

PAGE 182

158 n ork but also gets chemically in corporated through term inal hydroxyl groups, sol-gel f e nds e to the duced coating thickness (i n this case ~8 m, without compromising extract Several researchers pointed out that the pr esence of PEG in sol-gel matrix helps in controlling pore size distributi on in it, assuming that PEG molecules mostly lie between micropores and mesopores (<20 nm) [64] Perhaps the presence of mesopores and micropores in the sol-gel network are respons ible for massive weight loss at 108 C i thermogravimetric analysis even after heati ng at 110 C overnight in conventional oven which demand more severe condition to expel entrapped water and other unbonded reaction products from the pores. This observation led Augusto and coworkers [63] to adopt a more stringent condition for thermal tr eatment, keeping the fibers at 230 C for 50 h. As was expected, the long period of th ermal conditioning of sol-gel Carbowax 20M increased the precision of analysis, bri nging the RSD to very low value (2.9%). Since Carbowax 20M not only helps cont rolling the pore size distribution of the sol-gel netw Carbowax 20M coated fi bers demonstrated excellent selectivity toward different organic compounds. Extraction e fficiency of sol-gel Carbowax 20M coated fiber was compared to that of commercial PDMS and Carbowax fiber for headspace SPME o BTEX. Extraction equilibrium was reached between 3 min and 10 min depending on th compound in sol-gel Carbowax 20M. Both co mmercial coatings required much longer time to reach extraction equilibrium. Simila rly, complete desorption of all compou required only 20 s. Fast mass transfer to and fr om the fiber coating as observed in sol-gel Carbowax 20M are typical of all sol-gel coat ed sorbents. This became possible du combination of re ion sensitivity) and highly porous coati ng structure provided by sol-gel coating process. Experimental findings indicated that extraction efficiencies were comparable in

PAGE 183

159 M t to C8and C18-coated porous silica particles onto a stainl ess steel support using a highhen ts ently case of benzene and toluene. However, in ca se of o-xylene, the sol-gel Carbowax 20 provided significantly higher ex tracted amount compared to the masses extracted with 100 m PDMS and 65 m Carbowax-DVB fibe rs (230 and 540 % respectively ). Such an efficient extraction was obtaine d on the sol-gel carbowax fiber in spite of the fact tha the sorbent volume of solgel Carbowax 20M (0.017 mm 3 ) was only 2.6 and 4.6% of the coating volume of 100 m PD MS and 65 m Carbowax-DVB coatings, respectively. 3.3.2.2 Non-polysiloxane sol-gel sorbents with alkyl ligands C 8 and C 18 -bonded porous silica particles have long been used as popular chromatographic stationary phases due to their chemical integrity, excellent chromatographic selectivity, and chemical inertness of the silica particles. Tempted by their positive attributes and performance as chromatographic stationary phases as well as highly porous nature of silica particles that may have facilitated faster mass transfer of analytes during extraction, Liu et al. [65] attempted immobilize temperature epoxy resin. Due to the high specific surface area of the immobilized porous silica particles and very thin layer of bonded phases (C 8 or C 18 -), higher extraction sensitivity as well as faster ex traction equilibrium time was observed w compared with commercial coatings. Although worked well with highly volatile compounds, its maximum operating temperature (> 250 C) put a serious limitation on i use in the extraction of hi gher molecular weight compounds. Longer desorption time, sample carryover, and memory effect are among many problems encountered frequ

PAGE 184

160 when such coatings are used to extract se mi-volatile or low-vol atile compounds. Merely physical immobilization of coated silica partic les onto the substrate using epoxy glue is to blame for poor thermal stability of the coati ng. Moreover, such sorb ents may not provide high stability in liquid envir onments involving organic solven ts commonly used in liquidphase separation. In fact, almost all commer cial SPME fibers have been facing thermal and solvent stability restricti ons due to the lack of chem ical bonding to the substrate (fiber or capillary surface). In order to overcome the low thermal and solvent stability of C8-coated silica particles, Gbatu et al. [66] developed a sol-gel chemistry based hydrophobic octyl coating for SPME and evaluated its applicability for extracting organometalic compounds from water followed by HPLC. Other parameters (e.g ., impact of changing the ratio of sol solution ingredients, sol-gel re action time, effect of organic solvents, effect of pH on the coating) were also investigated. The sol solution was prepared by mixing C8-TEOS, MTMOS, methanol, hydrochloric acid and water. As C8-TEOS and MTMOS are the onl y reactants in the sol solution, their relative proporti ons were optimized by using different molar ratios of C8TEOS: MTMOS (0.5:1, 1:1, 2:1). After prep aring the sol solution, one end of the previously treated fused silica fiber (1 cm) was immersed into it and held there for 20 min. During this residence time in the sol solution, a three dimensional porous network gradually evolved. In this pr ocess, a part of the sol-ge l material became chemically bonded to the surface leading to the formation of a coating. The fiber was end-capped by dipping into a solution of trim ethylmethoxysilane/methanol (4:1 v/v). Finally the coated fibers were thermally conditioned at 130 C by placing them in a GC injector under

PAGE 185

161 constant helium flow. Scanni ng electron microscopic (SEM) images were taken for different coatings that were prepared using different mo lar ratio of C8-TEOS/MTMOS (0.5:1; 1:1, 2;1). The SEM images reveal ed that the coating with C8-TEOS/MTMOS molar ratio 2:1 had some crack ing. Molar ratio of 1:1 a nd 0.5: 1 did not produce any apparent crack. All three diffe rent coatings were used for extraction of organometals in order to evaluate their extrac tion efficiencies. Extraction resu lts indicated that the higher the ratio of C8-TEOS/MTMOS in sol solution, the better it extracts nonpolar analytes. But in order to preven t the likelihood of being cracked, a molar ratio of 1:1 was taken as optimum. Optimization of hydrolysis reaction time was done using the sol solution having C8-TEOS/MTMOS in 2:1 molar ratio. Different reaction time was used for generating sol-gel coating and the resultant coatings were evaluated by extracting an aqueous solution of triphenylarsine (Ph3As), diphenylmercury (Ph2Hg) and trimethylphenyltin (TMPhT). It was found that when the hydrolys is reaction was allowed to continue for more than 4 h, extraction efficiency did not increase significantly with time. This signified the completion of hydrol ysis of process and continui ty of polycondensation. The estimated time of hydrolysis was in good ag reement with the data obtained other researchers [13]. In HPLC, organic solvents are used to desorb the extracted analytes from the SPME fibers. However, the stability of the fi ber coating towards the desorbing solvent is a limiting factor. It is worth mentioning that most of the commercial coatings, being physically immobilized on the substrate, do not survive when exposed to the common mobile phases used in HPLC [2 5]. The integrity of sol-gel C8coating in organic solvents

PAGE 186

162 was evaluated by exposing it to xylene and me thylene chloride. In this experiment, solgel coatings were employed for extraction befo re and after exposing to organic solvents for up to 12 h. No apparent loss in extraction efficiency of the coating after such long exposure to organic solvents is indicative of excellent solvent stability of the coating. On the contrary, when commercial SPME fibers are exposed to these solvents, the coating tend swell and slip off the fiber. Another important factor that needs to be checked is the pH stability of the coating because adjusting pH of the sample is a common practice in achieving enhanced extraction efficiency, particularly when the an alytes are organic acids and bases. The solgel coating was found to remain intact even after being exposed to pH 0.3 and pH 13 for 16 h. High coating stability towards organic solvents and under very low and high pH conditions may be attributed to the fact that the coati ng is chemically bonded to the substrate. Sol-gel coatings are characterized by porous structure which remarkably shortens the analyte extraction equilibrium time due to the faster mass transfer. Because of the same reason, sol-gel coatings also provide faster analyte desorption compared to commercial coatings. As was revealed in extraction time profile, sol-gel C8coated fiber required 20 to 30 min to reach extraction equilibrium depending on the compound. By comparison, in case of partially cross linked PDMS/DVB coating, extraction equilibrium was not reached even after an hour. The difference in extraction equilibrium time between the two phases may be a ttributed to the rate of diffu sion within the coating. Solgel C8coating, being porous and thinner, fa vors faster diffusion re sulting in shorter equilibration time.

PAGE 187

163 Jinno and co-workers [67-68] interfaced solid phase mi croextraction with microLC and evaluated the perf ormance of the hyphenated system by employing five SPME fibers: commercial polyacrylate (PA), comm ercial Carbowax/template (CW/TEP), solgel C11 PDMS, commercial PDMS, and methyl-o ctyl poly(dimethylsiloxane) for extracting benzodiazepines from human urine samples. Sol-gel C11 PDMS was found to offer the highest extraction efficiency for benzodiazepines in aqueous media. 3.3.3 Cyclodextrin-based non-polysiloxane sol-gel sorbents Cyclodextrins and their de rivatives have long been used as chromatographic stationary phases especially in chiral separations. Their uni que properties, in particular, presence of a chiral cavity, the shape and si ze selectivity as well as its ability to form inclusion compounds with various analytes have made them very promising chromatographic stationary phases. Multiple retention mechanisms, including interaction of solute with the cyclodextrin cavity are said to be involved in the selectivity of these phases. Considering the positive attributes of cyclodextrins as chromatographic stationary phases, Fan et al.[69] developed a sol-gel met hod for preparing sol-gel -cyclodextrin coating for in-tube SPME coupled to HPLC fo r the determination of non-steroidal antiinflammatory drugs in urine samples. In or der to prepare the sol solution, 0.1 mL TEOS was added to 0.1 mL 0.01 M HCl and s tirred at 60 C until became homogeneous. Another solution was made using 0.05 g 3-glycidoxypropyltrimethoxysilane (KH-560) derivatized -cyclodextrin dissolved in 0.3 mL acet onitrile and 0.5 mL 0.01M HCl. Both

PAGE 188

164 solutions were mixed thoroughly and centrifuge d. The resulting so l solution was then allowed to fill the capillary and kept inside for 20 min. The sol solution was then expelled from the capillary using slow nitrogen flow. To accomplish desired coating thickness, the coating process was repeated three times. Finally the capillary was aged for 48 hrs. A 60 cm piece of sol-gel cylodextrin coated capillary was used for in-tube SPME. The urine samples to be analyzed were centr ifuged to remove any solid particles present in the matrix. Clean and particle-free urine sa mples were then diluted 10 times and spiked with analytes for extraction. A modification in the conf iguration of in-tube SPME-HPLC has been proposed where the coated capillary was directly c onnected to the six-port injection valve minimizing the necessity of additional connecting tubes. The requirement of a costly autosampler could be replaced by using a pum p and a six-port valv e with a PEEK tubing as the sample loop. The sample volume could be controlled accurate ly and precisely by controlling the valve swith-reswitch time interval and the flow rate of the carrier solution. Moreover, as the extraction segment was i ndependent of the analysis segment, both processes could be run simultaneously and t hus minimizing the whole analysis time. The new configuration was believed to minimi ze the analyte accumulation problem in the mobile phase as pointed out by Raghani and Schultz [70]. Extraction profile of ketoprofen (KEP) fenbufen (FEP) and ibuprofen (IBP) on sol-gel -cyclodextrin coating were investigated. In a carrier solution flow rate of 500 L/min, equilibrium was reached for KEP within 69 s which corresponds to 575 L sample volume whereas extraction equilibrium for FEP and IBP was not reached within this time period. It is known that under a given flow rate, the gr eater the sample volume

PAGE 189

165 the higher is the extraction until the extraction equilibrium is reached. In order to make a compromise between the sample volume and method sensitivity for routine analysis, a sample volume of 250 L was selected. The time required to obtain extraction equilibrium depends on various factors, (e.g., analyte distribution c onstant, volume of the coating, sample flow rate et c.). Keeping the sample volume constant, different flow rates were employed to find optimum flow rate A flow rate of 300 L/min was found optimum for the experiment. In general, addi tion of inorganic salts (e.g., NaCl) to the sample matrix increases the extraction efficien cy by reducing the solubi lity of analyte(s) in water. An opposite trend was observed for KEP. It was explaine d assuming that the added salt could compete with the anal ytes to form inclusion compound with cyclodextrin and thereby affect the distributi on equilibrium of the analytes. It is known that -cyclodextrin cavity has a preferential affi nity for the neutral form of acids [71]. As the analytes were weakly aci dic in nature, low pH of the matrix should favor the extraction since it is likely that at low pH they remain neutral in aqueous solution. Contrary to the assumption, extraction effici ency for the drugs was increased with the increase in pH from 3 to 7. Besides inclus ion complexation, there could be other factors that may play important roles in the interaction between the drugs and -cyclodextrin. Sample matrix is such a factor. When the extraction efficiency of a standard aqueous solution and a spiked urine solution was co mpared, a significant loss in extraction efficiency was observed for urine sample. Th e co-existence of inor ganic salt and protein are thought to be responsible for this loss. It is expected that sol-gel -cyclodextrin coatings will prove to be an effective SPM E sorbent not only fo r drugs in biological samples but also for other co mpounds in various matrices.

PAGE 190

166 includ d sol-gel DHB film. with s laser desorption/ionization mass spectrometry (SGALDI-MS) by E fiber. A 6 cm piece of optical fiber (125 m o.d.) was used for this experiment. A clean en dihydr for in optica ction of the target anal yte(s) was carried out by immersing the coated f 1 min. After the extraction, d he ted succes ous solution using so l-gel DBH coated fiber. It is possible to solid(HPLC) [74], (GC) [76]. ared to other forms of carbon. It consists of long sh eets of hexagonally oriented sp2hybridized carbon atoms. 3.4 Miscellaneous sol-gel sorbents In this category, several unique sol-gel based SPME sorbents are discussed that e low temperature glassy carbon film an Teng et al. [72] proposed a method for combining solid phase microextraction ol-gel assisted using an optical fiber coated with a thin sol-gel dihydroxybenzoic acid film as the SPM segmt (2 cm) of this fiber was dipp ed into the sol solution containing 2,5oxyenzoic acid (DBH), tetraethoxysilane (TEOS), water and hydrochloric acid one m. Due to the hydrolysis of the precu rsor and subsequent condensation of its hydrolysis products as well as DBH, a thin film was formed on the outer surface of the l fiber. Extra end othe fiber into the solution containing th e analyte(s) for the fiber was placed on a sample target. Finally the sample target was directly inserte into t mass spectrometer for SGALDI-MS an alysis. Benzo[a]pyrene has been extrac sfully from an aque extend the application of SPME-SGALDI by carefu lly selecting suitable sol-gel material. Glassy carbon is one of the most extensiv ely studied solid sorbent used in phase extraction (SPE) [73], high perfor mance liquid chromatography supercritical fluid chromat ography (SFC) [75], and gas chromatography Glassy carbon has unique chemical and structural features in comp

PAGE 191

167 characteristics, producing greater shape se lectivity than is ob tained on silica bonded lectiv e property of glassy carbon, Giardina et al. ro g the ethylene that leav es a thin coating of at for at least 10 hrs. The stainl ess steel fibers, a sol-gel Kasil No. 1 (SiO2/K2O). The fibers fi into the LTGC e any excess LTGCthickn oroughly investigated d odor contaminants er suppliperimental data revealed that LTGC coating had the highest affinity for ional surface area and polarizability. Another of LTGC processing temperature. The flat surface of the glassy carbon lead s to unique retention and selectivity phases [77] Considering the unique shape se [78] pposed a method of utilizing glassy carbon as an SPME sorbent. In preparin coating, porous silica was coated with LTG C by dissolving it into heptane/m chloride mixture and then slowly drying the so lvent system LTGConto the silica particles. The coated sili ca particles were then thermally treated different temperatures ranging fr om 300 C to as high as 1000 C coating was made on 1.5 cm segments of stainless steel fiber (~ 127 m o.d.). To immobilize the LTGC coated s ilica particles onto the solution was prepared by mixing formamide with were rst soaked with sol-gel solution immediately followed by dipping coated silica particles. The fibers were then gently tapped to remov coated silica particles. The process continued for 10 times to form a coating of desired ess. Selectivity of immobilized LTGC-coated silica has been th by using a wide range of aromatic hydrocar bons and some taste an (e.g., geosmin, 2-methylisoborneol, and 2,4,6trichloroanisole) commonly found in wat es. Ex molecules possessing the greatest cross-sect important finding was that the selectivity of LTGC increased as a function

PAGE 192

168 3.5 References for Chapter Three [1] belmen, Ann. Chim. Phys. 16 (1846) 129. [5] ng, Photogr. Archiv. (1896) 221. iley: New York, 1955. [10] .), J. Non-Cryst. Solids 48 (1982) 1. [12] -Cryst. Solids 73(1985) 681. 4566. nal. Chem. 7 A 7 M. E [2] T. Graham, J. Chem. Soc. 17 (1864) 318. [3] W.Z. Ostwald, Phys. Chem. 27 (1897) 365. [4] L. Rayleigh, Philos. Mag. 38 (1919) 738. R.E. Liesega [6] D.M. Roy, R. Roy, Am Mineral. 39 (1954) 957. [7] R. Roy, J. Am. Ceram. Soc. 52 (1969) 344. [8] R.K. Iler, The Chemistry of Silica, W [9] W. Stober, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62. V. Gottardi (Ed [11] J.D. Mackenzie, J. Sol-Ge l Sci. Technol. 26 (2003) 23. H. Schmidt, J. Non [13] Y. Guo, L. Colon, Anal. Chem. 1995, 67, 2511. [14] D.X. Wang, S.L. Chong, A. Malik, Anal. Chem. 1997, 69, [15] N. Tanaka, H. Kobayashi, K. Nakanishi, H. Minakuchi, N. Ishizuka, A 3 (2001) 420 A. [16] H. Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N. Tanaka, J. Chromatogr. 62 (1997) 135.

PAGE 193

169 a, T. [18] ayes, B. W. Wilhite A. Malik, Anal. Chem. 69 ( msterdam, 1 ase M [22] Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 ( [23] sert, J. Pawliszyn, Anal. Chem. 69(1997) 3140. 518. 004) 165. 1. 978 (2002) 37. [29] .Z. Gearhart, H.A. Mottola, Talanta 32 (1985) 751. [31] Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 5 [17] I. Ishiuka, H. Minakuchi, K. Nakanishi, N. Soga, H. Nagayama, K. Hosoy Tanaka, Anal. Chem. 72 (2000) 1275. S.L. Chong, D.-X. Wang, J. D. H 1997) 3889. [19] H. Rotzsche, Stationary Phases in Ga s Chromatography, Elsevier, A 991. [20] A. Malik, S.L. Chong, in: J. Pawliszyn (Ed.), Applications of Solid Ph icroextraction, Royal Society of Chemistry, Cambridge, UK, 1999, pp. 73. [21] Z.P. Zhou, Z.Y. Wang, C.Y. Wu, W. Zhan, Y. Xu, Anal. Lett. 32 (1999) 1675 S. Bigham, J 2002) 752. R. Ei [24] C. Shende, A. Kabir, E. Townsend, A. Malik, Anal. Chem. 75 (2003) 3 [25] T.-Y. Kim, K. Alhooshani, A. Kabir, D. P. Fries, A. Malik, J. Chromatogr. A, 1047 (2 [26] K. Alhooshani, T.-Y. Kim, A. Kabir, A. Malik, J. Chromatogr. A, 1062 (2004) [27] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A [28] M. Llompart, K. Li, M. Finga s, Anal. Chem. 70 (1998) 2510. D.D. Fine, H [30] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. Z. Zeng, W 1.

PAGE 194

170 2] C.Y. Wu, L.S. Cai, Y.J. Wang, H.M.Han, Z.R. Zeng, Z.Y. Yu, H. Yuan, Chromtographia 37 (1993) 374. 3] J.L. Ge, R.N. Fu, Z.F. Huang, Y.T. Wang, J. Microcol. Sep. 3 (1991) 250. [34] D. Wang, J. Xing, J. Peng, C. Wu, J. Chromatogr A, 1005 (2003)1. [35] C.J. Pedersen, J. Am. Chem. Soc. 89 (1967) 7017. [36] D. Wang, J. Peng, J. Xing, C. Wu, Y.u, J. Chro matogr. Sci. 58 (2003) 57. 7] J. Yu, C.Wu, J. Xing, J. Chromatogr. A 1036 (2004) 101. [38] L. Cai, Y. Zhao, S. Gong, L. Dong, C. Wu, Chromatogra phia 58 (2003) 615. 60 (1988) 1340. [40] Yun, Anal. Chim. Acta 486 (2003) 63. [41] J. Chromatogr. A 1056 (2004) 13. [49] [3 [3 X [3 [39] G. Audunsson, Anal. Chem L X. Li, Z. Zeng, S. Gao, H. Li J. Chromatogr. A 1023 (2004) 15. [42] A.L. Nguyen, J.H.T. Luong, Anal. Chem. 69 (1997) 1726. [43] M. Molder, S. Schrader, U. Franck, P. P opp, Fresenius' J. Anal. Chem. 357 (1997) 326. [44] X. Li, Z. Zeng, J. Zhou, Anal. Chim. Acta 509 (2004) 27. [45] X. Li, Z. Zeng, Y. Chen, Y. Xu, Talanta, 63 (2004) 1013. [46] A. Penalver, E. Pocurull, F. Borrull, R.M. Marce, J. Chromatogr. A 872 (2000) 191. [47] J. Ai, Anal. Chem. 69 (1997) 1230. [48] A.L. Lopes, F. Augusto K. Nakane, T. Jamashita, K. Iwakura, F. Suzuki, J. Appl. Polym. Sci. 74 (1999) 133. [50] M. Yang, Z.R. Zeng, W.L. Qiu, Y.L. Wang, Chromatographia 56 (2002) 73.

PAGE 195

171 matogr. 95 d. C. Silva, F. Augusto, J. Chromatogr. A 1072 (2005) 7. [64] [51] M.M. Liu, Z.R. Zeng, C.L. Wang, Y.J.Ta n, H. Liu, Chroma tographia, 58 (2003) 597. [52] J. Font, A. Marshal, J. Chromatogr. A 811 (1998) 181. [53] S. Magdic, J.B. Pawliszyn, J. Chromatogr. A 723 (1996) 111. [54] C. Aguilar, S. Penalver, E. Pocorull, F. Borrull, R.M. Marce, J. Chro (1998) 105. [55] G.P. Jackson, A.R.J. Andrews, Analyst 123 (1998) 1085. [56] L. Cai, J. Xing, L. Dong, C. Wu, J. Chromatogr. A 1015 (2003) 11. [57] B.W. Renoe, Am. Lab. 8 (1994) 34. [58] Y. Wang, M. Bonilla, H.M. MacNair, M. Khaled, J. High Resolut. Chromatogr. 20 (1997) 213. [59] M. Zhu, F.J. Aviles, E.D. Conte, D.W. Miller, P.W. Perschbacher, J. Chromatogr. A 833 (1999) 223. [60] M.C. Wei, J.F. Jen, Chro matographia 55 (2002) 701. [61] H.W. Hsiung, H.S. Jin, An al. Chim. Acta 428 (2001) 111. [62] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [63] R.G. S. Sato, T. Murakata, T. Suzuki, T. Ohgawara, J. Mater. Sci. 25 (1990) 4880. [65] Y. Liu, Y. Shen, M.L. Lee, Anal. Chem. 69 (1997) 190. [66] T.P. Gbatu, K.L. Sutton, J.A. Caruso, Anal. Chim. Acta, 402 (1999) 67. [67] K. Jinno, M. Taniguchi, H. Sawada, M. Hayashida, in J. Pawliszyn (ed), Applications of Solid Phase Microextra ction, Royal Society of Chemistry, Cambridge, UK, 1999, pp 527.

PAGE 196

172 shida, Analysis 26 (1998) M27. [69] 4) 151. X. Li, Z. Zheng, J. Zhou, S. Gong, W. Wang, Y. Chen, J. Chromatogr. A 1041 [68] K. Jinno, M. Taniguchi, H. Sawada, M. Haya Y. Fan, Y.-Q. Feng, S.-L. Da, Z. -H. Wang, Talanta, 65 (2005) 111. [70] A.R. Raghani, K.N. Schultz, J. Chromatogr. A 995 (2003) 1. [71] Z.I. Manzoori, M. Amjadi, Spectro chim. Acta Part A 59 (2003) 909. [72] C.-H. Teng and Y.-C. Chen, Rapid Co mmun. Mass Spectrom. 17 (2003) 1092. [73] G. Machtalere; V. Pichon; M. C. He nnion, J. High Resolut. Chromatogr. 23 (2000) 437. [74] M.-C. Hennion, V. Coquart, S. Guenu, C. Sella, J. Chromatogr. A 712 (1995) 287. [75] T.M. Engel, S.V. Olesik, J. High Resolut. Chromatogr. 14 (1991) 99. [76] T.M. Engel, S.V. Olesik, M.R. Callstr om, M. Diener, Anal. Chem. 65 (1993) 3691. [77] J. Kriz; E. Adamcova; J.H. Knox; J. Hora; J. Chromat ogr. A 663 (199 [78] M. Giardina, S. V. Olesik Anal. Chem. 73 (2001) 5841. [79] (2004) 1-9. [80] A. Kabir, C. Hamlet, K.S. Yoo, G.R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1. [81] A. Kabir, C. Hamlet, A. Malik J. Chromatogr. A 1047 (2004) 1.

PAGE 197

173 Chapter Four Capillary Microextraction on Sol-Gel Benzyl Terminated Dendrimer Coating 4.1 Introduction PME is e. ntSolidphase microextraction (SPME) [1] is now considered to be a fairly mature sample preparation technique with a wide variety of applications ranging from environmental to biomedical to agricultural, and a host of other samp les of scientific and industrial importance. SPME successfully ove rcomes the inherent shortcomings of conventional sample preparation methods by co mpletely eliminating the use of organic solvents and by integrating a number of samp le handling operations such as extraction, preconcentration, and sample in troduction for instrumental anal ysis. In addition, S a simple, inexpensive, easy-to-automate, porta ble, and fast sample preparation techniqu Due to these positive attributes, SPME has experienced an explosive growth since its inception over a decade ago. SPME is based on the distribution of analyt es between the sample matrix and the extracting phase coated either on the outer surf ace of a solid fiber (f iber SPME) or on the inner surface of a fused sili ca capillary (in-tube SPME or capillary microextraction, CME) [2]. Various SPME coatings have been successfully used to accomplish solve free extraction of analytes from different matrices. Among the used SPME coatings are polydimethylsiloxane (PDMS)[1], Polyacrylat e [3], Carbopack [4], polyimide [5],

PAGE 198

174 polypyrrole [6], molecularly imprinted ma terials [7,8], Carbowax/divinylbenzene (CW/DVB) [9], polydime thylsiloxane/divinylbe nzene (PDMS/DVB) [10], polydimethylsiloxane/Carboxane (PDMS/Carboxane) [11], Carbowax/templated resin (CW/TPR) [11], sol-gel PDMS [12,13], sol-gel PEG [2,14], and sol-gel crown ether [15,16] that have been successfully applied fo r the extraction of anal ytes from different matrices. The extraction affinity is determ ined by various types of intermolecular and steric interactions between the analyte speci es and the coating of the extracting phase. Thus, selective extraction of analytes can be achieved based on their polarity, hydrophobicity, chemical composition, shape/size, etc. Selective extraction by SPME has often been performed based on solute polarity. However, such an approach is not very effective for samples where both polar and nonp olar contaminants are present, and both types need to be analyzed. For such samples, it is very important to have a sorbent that can extract both polar and nonpolar com pounds simultaneously with high extraction sensitivity. Most of the SPME coatings that have been used so far are based on linear organic polymers. Linear polymers have some inhere nt shortcomings for being used as SPME coatings. They possess a wide range of mol ecular weight distribu tion responsible for considerable variations in th eir physical properties [17]. Th e large dispersity of linear polymers makes it difficult to achieve batchto-batch reproducibility. Moreover, linear polymers are highly viscous and poorly solu ble in common organic solvents, putting limitations in their effective use as SPME coatings. Dendrimers [18, 19] are highly branch ed macromolecules that can easily overcome many of the inherent shortcomings of linear polymers. Dendrimers are created

PAGE 199

175 in a step-wise fashion from simple br anched monomeric units, the nature and functionality of which can be easily controll ed and varied. The supramolecular properties of dendrimers can be effectively tailored by creating desired functi onal groups on either the core [20], the peripheral surface [21], the branching unit [22], or at multiple sites within the dendrimer [23]. Dendritic macromol ecules possess physical properties that, in many cases, greatly differ from those of th eir linear analogs. For example, their monodisperse structure is built in generations : layer by layer around a core moiety [24]. In organic solvents, they exhibit high solubility and low vi scosity compared with their linear analogs [25]. These disc repancies in physical properti es are reflections of the fundamental differences in the molecula r architectures of these two types of macromolecules providing drastically differe nt numbers of terminal functional groups [26]. Dendrimers possess open and vacuous structures characterized by channels and pockets, especially in higher generations [27]. Unlike 1st and 2nd generations, the higher generation dendrimers have greater internal surface area compared with the external surface area [28]. Therefore, 3rd and higher generation dendrimers should be well suited for applications where high surface area (both internal and external) is a prerequisite. Because of their tree-like br anched architecture, functi onalized dendrons are potential candidates for novel sorbents in analytical sample enrichment and separations. This opens new possibilities in achieving enhanced selectivity, sensitivity, and performance in chromatographic separations and sample preparations. To date, in the area of analytical separa tions, dendrimers have been used as: (a) pseudo-stationary phases in elect rokinetic chromatography [29-31], (b) bonded stationary

PAGE 200

176 obilization of n ME fibe rs. Low thermal stability of thick coatings ssentially leads to incomplete sample deso rption and sample carryover problems. On the to the hyphenation of in-tube SPME (capillary microextraction) with liqui d-phase separation techniques since organic or organo-aqueous liquids are employed for the desorption of analytes from the SPME coating used for extraction [36,37]. Most of the difficulties associated with the creation and immobilization of thick stationary phase coatings on the fused silica surface can be effectively addressed by using sol-gel coating technology [11,14,38,39] In the context of SPME, major advantageous features of sol-gel technology are as follo ws:(1) it combines the surface treatment, deactivation, coating, and stationary phase immobilization into a single-step procedure fficient, and costhemical bonds be tween the fused silica surface and the sol-gel coating) it provides SPME coatings with high thermal and solvent stabilities, and thereby opens the possibility to expand phases in capillary electrochromatography [32] (c) chiral stationary phases in HPLC [33], and (d) GC sta tionary phases [34]. Effective immobilization of the polymeric coating on fused silica fiber or capillary inner surface is a pr erequisite for the maximum u tilization of its analytical potential. However, it is often difficult to achieve acceptable degree of imm thick SPME stationary phase coatings through conventional approaches [35]. As has bee pointed out by Chong et al. [12], the absence of chemi cal bonds between the polymeric coating and the fused silica fiber surface is responsible for low thermal and solvent stability of conventionally coated SP e other hand, low solvent stability of coati ngs presents a significant obstacle making the whole SPME fiber/capillary manufac turing process fast, e effective; (2) it creates c ; (3 the SPME application range toward higher

PAGE 201

177 boiling compounds on one hand, and thermally labile compounds on the other; (4) it provides an effective pathway to combine or ganic and inorganic ma terial properties to achieve enhanced selectivity; (5) it provide s SPME coatings with a porous structure responsible for increased surf ace area allowing the use of th inner coatings to achieve acceptable extraction phase loading and sample capacity. A recent publication from our group [61] de scribes sol-gel dendrimer coatings in analytical microextraction. In this disser tation, we provide a detail account on sol-gel approach to in situ creation of dendritic sorbent on the inner walls of fused silica capillaries to achieve solven tless extraction of both polar and nonpolar trace analytes from aqueous samples. 4.2 Experimental section 4.2.1 Equipment Sol-gel dendrimer CME-GC experiments were carried out on a Shimadzu model 17A gas chromatograph (Shima dzu Corporation, Kyoto, Ja pan) equipped with a flame ionization detector (FID) and a pr ogrammed temperature vaporizer PTV for sample introduction. An in-house built gravity-fed sample dispenser was used to deliver the aqueous samples through the sol-gel de ndrimer coated capillary during CME experiments. A Fisher Mode l G-560 Ginie 2 vortex (Fisher Scientific, Pittsburgh, PA) was employed for thorough mi xing of different solutions. A Microcentaur model APO 5760 microcentrifuge (Accu rate Chemical and Scientific Corporation, Westbury, NY)

PAGE 202

178 was employed (@ 13000 RPM, 15682 g) to separa te particulates from the sol solutions used for coating the extraction capillaries as well as the GC columns. A homemade, gas pressure-operated filling/purging device [Figure 4.1] was used to fill the fused silica capillary with the sol solution or purge it with helium gas at various stages of coating and extraction procedures. A Barnsted M odel 04741 Nanopure deionized water system (Barnsted/ Thermodyne, Dubuque, IA ) was used to obtain 17.2 M ultra pure water. A JEOL model JSM-35 scanning electron microsco pe was used to obtain SEM images of the sol-gel dendrimer coatings. On-line data collection and proce ssing were done using ChromPerfect software (V ersion 3.5) for Windows (Jus tice Laboratory Software, Denville, NJ). 4.2.2 Chemicals and materials Fused-silica capillary (250 m i.d.) with an external protective polyimide coating was purchased from Polymicro Technologies In c. (Phoenix, AZ). A two-way fused silica press-fit connector (Polymicro Technologies Inc., Phoenix, AZ) was used to interface the extraction capillary with th e front end of the GC capil lary column inside the chromatographic oven. Benzyl-terminated dend ron with triethoysilyl-derivatized stem was synthesized in one of our laboratories following a procedure described elsewhere [34]. Hydroxy-terminated PDMS was purchased from United Chemical Technologies, Inc. (Bristol, PA). Two types of trimethoxys ilyl-derivatized poly( ethylene glycol) (MSIL-5000 and SIL-3400) were obt ained from Shearwater Polymers (Huntsville, AL). Acenaphthene, fluorene, phenanthrene, fluoranthene, pyrene, nonyl aldehyde, m-

PAGE 203

179 tolualdehyde, undecylic aldehyde, buty rophenone, valerophenone, hexanophenone, heptanophenone, benzophenone, 2-chlo rophenol, 3,4-dichlorophenol, 3,5dimethylphenol, and 2,4,6-tric hlorophenol were purchased from Aldrich Chemical Co. (Milwaukee, WI) and n-decyl aldehyde was purchased fr om Sigma Chemical Co. (St. Louis, MO). Methanol (HPLC grade) and all borosilicate glass vials were purchased from Fisher Scientific (Pittsburgh, PA). 4.2.3 Preparation of sol-gel dendrimer coated extraction capillaries Preparation of sol-gel co ated extraction capillaries involves five distinct and sequential operations:(1) clean ing and pretreatment of the fused silica capillary, (2) preparation of the sol solution, (3) coating th e fused silica capillary with the sol solution, (4) thermal conditioning of the coated fused silica capillary, (5) ri nsing of the coating with organic solvents to remove unbonded materials (if there is any). 4.2.3.1 Cleaning and hydrothermal trea tment of the fused silica capillary In order to clean the inner surface of the fused silica capillary from organic contaminants, it was first sequentially rinsed with methylene chloride and methanol. One milliliter of each solvent was passed thr ough the capillary sequentially using the filling/purging device (Figure 4.1). The rinsed fused silica capillary was then purged with

PAGE 204

180 helium for 30 min followed by hydrothermal treatm ent. Hydrothermal treatment is a very important step that helps generating adequa te surface silanol groups required for the formation of strong chemical bonds between the substrate and the growing sol-gel network. To perform hydrothermal treatment, the cleaned fused silica capillary was filled with Nanopure deionized water using the fill ing/purging device and then the water was flushed out of the capillary with the aid of helium gas pr essure. The capillary was then purged with helium gas for 30 min so that on ly a thin layer of water remained on the surface of the capillary. At this point, both the ends of the fused silica capillary were sealed with oxy-acetylene flame. The sealed capillary was then heated at 250 C in a GC oven for 2 hours. Both the ends were then cut open with a ceramic wafer. One end of the fused silica capillary was then connected to the GC injec tion port with the help of a graphite ferrule and the capi llary was heated again in th e GC oven at 250 C for 2 hours using a continuous helium flow through the cap illary. After this, th e capillary was ready for coating.

PAGE 205

181 Figure 4.1 Schematic of a homemade capillary filling/purging device for prep aration of capillary microextraction capillaries and open-tubular sol-gel GC columns. Gas flow Fused silica capillary Pressurization chamber Threaded chamber cap Control valve

PAGE 206

1824.2.3.2 Preparation of the sol solution The used sol solution consis ted of an alkoxide precursor a sol-gel-active (either hydroxy or alkoxy silane terminat ed) organic polymer, one or more surface deactivating reagents, appropriate organic solvent(s), and a sol-gel catalyst. Table 4.1 presents the names, functions, and chemical structures of different ingredients of the sol solution used in preparing sol-gel dendrime r coated microextraction capil laries. The sol solution was prepared by dissolving methyltrimet hoxysilane (sol-gel precursor, 5 L), phenylterminated 3rd generation dendrimer with a triethoxysilyl containing root (sol-gel active organic ligands, 50 mg), hexamethlyldisil azane (surface deactiv ation reagent, 10 L), polymethylhydrosiloxane (surface deactivation reagent, 25 L), and trifluoroacetic acid (containing 5% water) (sol-gel catalyst, 50 L) in methylene ch loride (solvent, 900 L). After adding all the ingredients, the result ant solution was vortexed (5 min) for thorough mixing of the constituents followed by centrif ugation (5 min) to remove any precipitates from the solution before using it for coating.

PAGE 207

183 Table 4.1 Name, function and chemical struct ure of sol-gel dendrimer coating solution ingredients Name Function Structure Methyltrimethoxysilane Sol-gel precursor Phenyl-terminated dendrimer with a trieth oxysilyl root Sol-gel precursor containing a dendritic ligand Presented in Scheme 4.1 Trifluoroacetic acid/Water 95:5 (v/v) Catalyst CF3COOH Methylene Chloride Solvent CH2Cl2 Hexamethylenedisilazane (HMDS) Deactivating reagent Polymethylhydrosiloxane (PMHS) Deactivating reagent 4.2.3.3 Coating of the fused silica capillary with sol solution A hydrothermally treated fused silica capillary (3 m x 250 m i.d.) was installed in the gas pressure operated filling/purging devi ce (Figure 1) to fill the capillary with the specially designed sol solution. After filling, the sol solution was kept inside the capillary for 30 min to facilitate the formation of a surface-bonded sol-gel dendrimer coating. During this in-capillary resi dence time of the sol soluti on, a sol-gel hybrid organicinorganic network was evolving w ithin the sol solution. A part of this sol-gel material CH3O Si CH3OCH3OCH3H3C Si NH Si CH3 CH3CH3 CH3CH3 Si O (Si O )nSi CH3 CH3CH3 CH3CH3 CH3H H3C

PAGE 208

184 growing in the vicinity of the fused silica capillary walls ultimately became bonded to the fused silica capillary walls vi a condensation reaction with s ilanol groups on the surface. The free unbonded portion of the sol solution wa s then expelled from the capillary under helium pressure (50 psi) and the coated cap illary was purged with helium for an hour. The continuous flow of helium facilitate d the evaporation of volatile solvents. 4.2.3.4 Thermal conditioning of the coated capillary The sol-gel coated capillary was th ermally conditioned in a GC oven using temperature-programmed heating from 40 C to 300 C @ 1 C min-1, and holding the capillary at the final temperature for 5 h under helium purge. The purpose of the conditioning was to stabili ze the coated sorbent and re move the non-bonded volatile components of the sol soluti on left in the coating. 4.2.3.5 Rinsing the capillary with organ ic solvents to remove unbonded materials Before using in extraction, the coated cap illary was rinsed sequentially with 2 mL each of methylene chloride and methanol followed by drying in a stream of helium under the same temperature-programme d conditions as in 4.2.3.4, except that the capillary was held at the final temperature for 30 min. The capillary was further cooled down to

PAGE 209

185 ambient temperature and cut into 13-cm long pieces that were further used to perform microextraction. 4.2.4 Preparation of sol-gel PDMS coated capillary columns for GC A 10 m long piece of uncoated fused sili ca capillary (250 m i.d.) was accurately wrapped on a GC column basket in coils. Th e capillary was sequentially rinsed with 1 mL each of methylene chloride and metha nol followed by purging with helium for 30 min. The column was then hydrothermally treated following the procedure described in Section 4.2.3.1. The hydrothermally treate d column was installed on the filling/purging device (Figure 4.1) and the helium flow was adjusted to 40 psi. The coating solution for the sol-gel PDMS GC column was prepared as follows: 50 mg of PDMS was dissolved in 900 L of methylene chloride. A 5 L volume of methyltrimethoxysilane (MTMOS), 25 L of polymethylhydrosiloxane (PMHS), and 50 L of trifluoroacetic acid (TFA) (5% water v/v) were sequentially added into the vial containing the polymer solution. In order to ensure proper mi xing of the ingredients, the sol solution was vortexed after adding each in gredient. As the hydrolysis reaction begins as soon as the sol-gel cataly st is added, a 10-min peri od was allowed for hydrolysis reaction to proceed. At the end of this 10-min period, 20 L NH4F solution (20 mg/mL in methanol) was added to the sol solution to faci litate faster condensation. The resulting sol solution was centrifuged and the clear supernatant was transferred to a clean vial using a micro pipet for further use in column coating.

PAGE 210

186 reated fused silica apillary using helium gas pressure in the filling/purging device (Figure 4.1). Once the capillary was filled with sol-ge l coating solution, the filling/purging device was carefully depressurized and the free exit end of the capillary was sealed with a rubber septum while keeping the other end of the capillary inside filling/purging device. The sol solution was kept inside the capillary for 10 min. During this in-capillary residence time of the sol solution, a surface bonded hybrid organic-inorga nic film gradually evolved. After this, the sol solution was expelled from the fused si lica capillary under helium pressure. The column was then purged with helium for 60 min. The next step was the thermal conditioning of the coated column. For this, one end of the column was connected to the GC injection port using a gr aphite ferrule and a continuous flow of helium was maintained through the column. The other end of the capillary was kept free inside the GC oven. The column was heated from 40 C to 150 C at a rate of 1C/min, maintained this temperature for 5 h, then heated to 330 C at a rate of 5 C/min and holding at this final temperature for 1 h. After the thermal conditioning, the column was rinsed with 2 mL of methylene chloride/methanol mixture (75: 25 v/v), purged with helium for 1 h and then conditioned again from 40 C to 330 C at a rate of 5 C/min, holding it at final temperature for 30 min. The sol solution was then introduced into the hydrothermally t c

PAGE 211

1874.2.5 Preparation of sol-gel PEG c oated capillary column for GC A 10-m long piece of fused silica capillar y was first cleaned and hydrothermally treated as described in section 4.2.4. It was then installed on the filling/purging device (Figure 4.1) to carry out the coating process. A sol-gel coating solution was prepared as follows: 35 mg of methoxypoly(ethylene glycol)-sila ne (PEG 1) and 15 mg of poly (ethylene glycol)-bis silane (PEG2) (sol-gel active organic ligan ds) were dissolved in 600 L of methylene chloride (solvent) contained in a polypropyl ene microcentrifuge vial. A Scientific Products model S8223 Vortex shaker aided the dissolution process (5 min). Then 5 L of MTMOS (precursor), 10 L of bis(trimethoxysily lethyl)benzene (precursor), and 5 L of HMDS (deactivating reagent) were sequentially added to the microcentrifuge vial and thoroughly mixed for 5 min to obtain a homogene ous solution. After this, 50 L of 95% TFA (acid catalyst containing 5% water) was added to the solution and thoroughly mixed. After 10 min, a 20-L volume of NH4F solution (20 mg/mL in methanol) was introduced into the vial. The volume of th e solution was made up to 1000 L by adding the required amount of methylene chloride, and the mixt ure was thoroughly vortexed. The resulting solution was centrifuged at 13 000 rpm (15,682g) for 5 min. The precipitate at the bottom of the vial, if any, was discarde d, and the top clear so l solution was used to fill the hydrothermally treated fused-silica cap illary using a helium pressure of 50 psi. After a set period of in-capillary residence time (10-20 min), the solution was expelled from the capillary under the same helium pre ssure and the capillary was subsequently

PAGE 212

188 purged with helium at 50 psi for an additional 60 min. This was followed by temperatureprogrammed heating in a GC oven from 40 to 150 C at 2 C min-1 with a hold time of 300 min at 150 C and then from 150 to 280 C at 6 C min-1, holding it at 280 C for 120 min. Keeping the temperature programming rate at 6 C min-1, the column was further conditioned in small steps, holdi ng the column for 120 min at each of the following final temperatures: 300, 320, and 340 C. The column was then rinsed with 2 mL of methylene chloride and conditione d again from 40 to 320 C at 6 C min-1. While conditioning, the column was purge d with helium at 1 mL min-1. 4.2.6 Gravity-fed sample dispenser for capillary microextraction A gravity-fed sample dispenser was us ed for capillary microextraction (Figure 4.2). It was constructed by modifying a Chromaflex AQ column (Kontes Glass Co., Vineland, NJ) that consisted of a thick-wall ed Pyrex glass cylinder concentrically placed in an acrylic jacket. Deactivation of th e inner surface of the glass cylinder was accomplished by treating with HMDS solution (5% v/v solution in methylene chloride) followed by heating at 250 C for 1 hour in an inert gas environment. The cylinder was then cooled to ambient temperature, thor oughly rinsed with meth anol and deionized water, and dried in a helium gas flow. The system was then reassembled and was ready for use as a sample delivery device in cap illary microextraction.

PAGE 213

189 Figure 4.2 Schematic of a gravity-fed sample dispensing unit used in sol-gel dendrimer capillary microextraction.

PAGE 214

1904.2.7 Deactivation of glassware All glassware used in this study was fi rst cleaned using Sparkleen detergent, thoroughly rinsed with deioni zed water followed by drying at 150 C for 2 hours. The inner surface of the dr ied glassware was then treated with a 5% v/v solution of HMDS in methylene chloride followed by heating in an oven at 250 C for 8 hours under helium flow. The glassware was then rinsed sequentia lly with methylene ch loride and methanol and dried in the oven at 100 C for 1 hour. Before use, all glassware was thoroughly rinsed with deionized water and dried at room temperatur e in a conti nuous flow of helium. 4.2.8 Preparation of standard sample solutions for sol-gel dendrimer CME All stock solutions were prepared by disso lving 50 mg of each analyte in 5 mL of methanol in a 10 mL deactivated amber glass vial to obtain a concentration of 10 mg/mL. The solution was further diluted to 0.1 mg/mL in methanol. The final solution was prepared by further diluting th is solution in water to achieve g/mL to ng/mL level concentrations depending on the compound class. 4.2.9 Extraction of analytes on so l-gel dendrimer coated capillaries A 13-cm long piece of the sol-ge l dendrimer-coated capillary (250 m i.d.) was conditioned in a GC oven using a temp erature program (from 40 C to 300 C @ 10 C min-1, held at the final temperature for 30 mi n) and simultaneously purging the capillary

PAGE 215

191 with helium. The conditioned capillary was ver tically connected to the lower end of the gravity-fed sample dispenser (Figure 4.2) using a plastic nut a nd a ferrule. A 50-mL volume of the aqueous sample containing trace c oncentrations of the target analytes was added to the inner glass cylinder of the samp le dispenser through the inlet located at the top. A small helium gas pressure (5 psi) was maintained in the system to assist the sample flow. The solution was allowed to pass through the capillary for 30 min. During this time, the analyte molecules were extrac ted by the sol-gel dendrimer coating as the sample passed through the capillary, and th e system moved towards an extraction equilibrium. The capillary was further pur ged with helium gas for 1 min to remove residual water from the capillary walls. 4.2.10 Transferring the extracted analyt es to the GC Column and gas chromatographic analysis of the extracted analytes The extracted analytes were transferred fr om the microextraction capillary to the GC column via thermal desorption. For this, the CME capillary was first installed on the GC injection port and securedly interfaced with the GC capillary column. To facilitate the installation, both the GC injection po rt and the oven were cooled to 30 C, and the quartz wool was removed from the injection port glass liner. Th e CME capillary with the extracted analytes in the coating was then in troduced into the GC injection port from the bottom end of the port so that ~8 cm of the cap illary remained inside the injection port. A graphite ferrule was used to make an airtight connection between the capillary and the injection port. The lower end of the capi llary (residing inside the GC oven) was

PAGE 216

192 connected to the inlet end of GC capillary column using a deactivated press-fit quartz connector. Figure 4.3 demonstr ates the connection of the ex traction capillary with the GC analysis column using a press-fit qua rtz connector. The temperature of the PTV injection port was then rapidly raised to 300 C @ 100 C min-1 to desorb the analytes from the extraction capillary into the carrier gas flow, keeping the GC oven temperature at 30 C during the whole desorption process (5 min). Under these conditions, the desorbed analytes were efficiently carried ove r by the flow of helium to the lower end of the extraction capillary a nd/or front end of the GC column held at 30 C. As soon as the desorbed analytes reached the cooler CME capillary-GC column coupling zone residing inside the GC oven (30 C), the analytes were focused into a narrow band. To facilitate transport of the focused zone through the GC column and its separa tion into individual components, the GC oven temperature wa s then programmed as follows: from 30 C to 300 C @ 15 C min-1 with a 10 min hold at the final temperature.

PAGE 217

193 Figure 4.3 Schematic representation of the connection of the extraction capillary with the analysis column inside the GC oven using a press-fit quartz connector

PAGE 218

194 he limit of detection (LOD) for an analyte is the smallest concentration that can be detected reliably. The LOD is related to both the signal and the noise of the system and usually is defined as the concentration corresponding to a peak whose signal-to-noise (S/N) ratio is 3:1. In order to calculate the limit of de tection (LOD), each compound was extracted individually under same extraction conditions and the peak height of the analyte was calculated in V. The noise was calculated in V from the baseline of the chromatogram using the ChromPerfect for Windows (Version 3.5) computer software (Justice Laboratory Software, M ountain Views, CA). The limit of detection of the compound was calculated through intr apolation using the analyte peak height and the corresponding value of the noise. 4.3 Results and discussion The branched architecture of dendrim ers makes them promising candidates for use as extracting materials with distinct a dvantages over linear polymers used for the same purpose. The main objective of the presen t work was to investigate the possibility of using benzyl terminated dendrimers as a novel extraction medium for solid phase microextraction. This was accomplished by creating immobilized dendrimer coatings on the fused silica capillary i nner surface using principles of sol-gel chemistry. 4.2.11 Calculation of the limit of detection (LOD) for individual analyte T

PAGE 219

1954.3.1 Sol-gel dendrimer coating and chem ical aspects of its preparation Sol-gel column technology [38] provides an elegant single-step procedure for creating organic-inorganic hybrid stationary phase coatings (both thick and thin) inside the capillary that can be further used to pe rform capillary microextraction [2] or highresolution gas chromatographic separations. So l-gel technology also allows the creation of hybrid coatings on the outer surface of a solid fiber [ 12] that can be used in conventional fiber-based SPME analysis. In bo th instances, the coat ing is chemically bonded to the substrate, and provides high thermal stability required for SPME-GC analysis. Thanks to chemical bonding to the substrate, sol-gel coa tings also possess high solvent stability required for hyphenating SPME with liquid-phase separation techniques (e.g., HPLC, MEKC, CEC, etc.) that use organo-aqueous mobile phases. If an organic polymer or ligand is to under go sol-gel reaction, it has to be sol-gelactive. The dendrimer used in this study contained ethoxysilyl groups in its root making the dendrimer molecules sol-gel active. Details of the synthesis of sol-gel active dendrimers can be found elsewhere [34]. The ingredients used in the sol-gel dendrimer coating are presented in Table 4.1. As can be seen in Table 4.1, methyltr imethoxysilane (MTMOS) is the second sol-gel precursor used in the coati ng solution. Under the experi mental conditions used, both MTMOS and the triehoxysilyl moieties in the benzyl-terminated dendron can get hydrolyzed in the presence of th e sol-gel catalyst, trifluoroacetic acid (TFA), as presented in the reaction Scheme 4.1.

PAGE 220

196 The hydrolyzed precursors can then u ndergo polycondensation reactions in a variety of ways to create a sol-gel network. Scheme 4.2 represents a simplified model for the polycondensation reacti on of hydrolyzed MTMOS. The growing chain of the sol-gel pol ymer can also undergo polycondensation reaction with hydrolyzed triet hoxysilyl root of the dendron to form an organic-inorganic hybrid network with th e chemically incorporated dendrim ers as an organic constituent (Scheme 4.3). The condensation reaction can also take place with the participation of silanol groups on the inner surface of the fused silica capillary. The sol-gel dendritic network developed in the vicinity of the fused sili ca capillary inner surface can get chemically anchored to the column walls (Scheme 4. 4) forming a surface-bonded extracting phase film, and remain as such when the sol-gel solution is expelled after 30 min of residence inside the capillary. Both polymethylhydrosiloxane (PMHS) and hexamethyldisilazane (HMDS) used in the sol solution as surface d eactivation reagents lack sol-ge l active sites. Therefore, it can be assumed that they rath er get physically incorporated in the sol-gel network, and subsequently react with the silanol groups (Scheme 4.5) during the post-coating thermal conditioning process. This re sults in a three-dimensiona l deactivation process taking place within the entire thickness of the sol-ge l coating [38] as opposed to traditional twodimensional deactivation process which is c onfined only to the cap illary surface. Thus the sol-gel technology used for the coati ng process elegantly combines column deactivation, coating, and extracting phase f ilm immobilization in a simple and effective anner. m

PAGE 221

197

PAGE 222

198

PAGE 224

200

PAGE 225

201

PAGE 226

202

PAGE 227

2034.3.2 Characterization of surface mo rphology and determination of coating thickness using scanning electron microscopy Figures 4.4 and 4.5 represent two sca nning electron micrographs (SEM) of the inner surface of the so l-gel dendrimer coated capillary. Remarkable uniformity in coating thickness is evident from the SEM image presen ted in Figure 4.4. The coating thickness was estimated at 0.5 m. (Figure 4.4). Moreover, sol-gel dendrimer coating possesses a roughened, porous texture (Figur e 4.5) with enhanced surface area which is favorable for extraction.

PAGE 228

204 Sol-gel dendrimer coating Magnification: 10,000x Coating thickness Figure 4.4 Scanning electron micr oscopic image of a 250 m i.d. sol-gel dendrimer coated microextraction capil lary illustrating the coating thickness. Magnification: 10,000x

PAGE 229

205 Figure 4.5 Scanning electron micr oscopic image of a 250 m i.d. sol-gel dendrimer coated microextraction capilla ry illustrating the typical roughened su rface obtained by sol-gel coating process. Magnification: 10,000x

PAGE 230

206 analytes Figure 4.6 illustrates the CME ki netic profile of a nonpolar analyte (phenanthrene) and a polar analyte (2,4,6-t richlorophenol) extracte d on a sol-gel benzylterminated dendrimer coated capillary. Extractions were carried out using aqueous samples containing 100 ppb concentration of each analyte. Both for the polar and nonpolar analytes, extraction eq uilibria were attained w ithin 30 min (Figure 4.6). Based on these kinetic data, a 30-min extraction time was further used for all samples to ensure that the extraction equilibrium was reached. 4.3.3 Determination of extraction kine tics for both polar and nonpolar

PAGE 231

207 0 200000 400000 600000 800000 1000000 1200000 010203040506070 Time (min)Peak Area (arbitrary units) phenanthrene 2,4,6-trichlorophenol Figure 4.6 Illustration of the extraction kinetics of a nonpolar compound (phenanthrene) and a polar compound (2,4,6-trichlor ophenol) obtained on a 13 cm x 250 m i.d. sol-gel dendrimer coated microextraction capillary using 100 ppb aqueous solutions. Extraction conditions: 13 cm x 0.25 mm i. d. microextraction capillary ; extraction time, 10-50 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C.

PAGE 232

2084.3.4 Applications of sol-gel dendrimer coated microextraction capillaries Sol-gel dendrimer coated capillaries were used to extract a variety of analytes matic hydrocarbons yd, alco and hluene a widely used preservative. The extra tes were further analyzed by GC-FID. .3.4.1 Polycyclic aromatc hydrocarns atic hydrocarbons (PAHs) are among the mt common pollutants found in a, and soil in the USA and other industrialized countries where petroleum products are h eavily used. Toxicity, mutagenicity, and enicity of these compounds in animal s [41prompte Environ rotection Agency (EPA) to place 16 unsubsti tuted PAHs in its list of 129 priority ollutants [42]. Among the 16 EPA promulgate d unsubstituted PAH, 5 were extracted nd analyzed using sol-gel dendrimer coated microextraction capillaries. Table 4.2 rovides a list of 5 selected unsubstituted P AHs, their chemical structures as well as ertinent physico-chemical pr operties. Capillary microext raction results have been resented in Table 4.3 (run-to-run reproducib ility data) and Table 4.4 (capillary-tocapillary reproducibility data). Calculated limit of detection data for the selected PAHs have been presented in Table 4.5. Figure 4.7 represents a gas chromatogram of these unsubstituted polycyclic aromatic hydrocarbons from EPA priority po llutants list. They from a wide polarity range (nonpolar to highl y polar) and of environmental, biomedical and ecological importance. Test analytes included polycyclic aro (PAHs), aldeh es, ketones hols, phenols cted solu d butylate ydroxy to (BHT) 4 i bo Polycyclic arom os environmental ir, water carcinog ] has d US mental P p a p p p

PAGE 233

209 were extracted from an aqueous sample (e ach PAH at 10 ppb) using a sol-gel dendrimer coated microextraction capillary. As can be seen from the data presented in Table 4.2, the detection limits obtained for PAHs in CME-GC -FID range between 2.1 ppt and 3.6 ppt. These values are comparable to or better than the detection limits reported in the literature for conventionally coated SPME fibers. For instance, Doong et al.[ 42] reported a detection limit of 0.25 ng/mL (250 ppt) for fluoranthene obtained by SPME-GC-FID on a commercial 100m PDMS coated fiber, which is mo re than two order of magnitude higher than the value 0.002 ng/mL (2ppt) obt ained on sol-gel dendr imer CME-GC-FID.

PAGE 234

210 Table 4.2 Chemical structures and pertinent physica l properties of polyaromatic hydrocarbons (PAHs) extrac ted using sol-gel dendr imer coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) Acenaphthene 152.20 92.5 150.0 0.8987 Fluorene 166.22 114.8 295.0 1.203 Phenanthrene 178.23 99.2 340 0.9820 Fluoranthene 202.26 107.8 384 1.252 Pyrene 202.26 156 404 1.271

PAGE 235

211 Figure 4.7 CME-GC analysis of P AHs at 10 ppb concentration using sol-gel dendrimer coated microextraction capillary. Extr action conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C, programmed at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) acen aphthene, (2) fluorene, (3) phenanthrene, (4) fluoranthene, and (5) pyrene

PAGE 236

212 Table 4.3 Run-to-run peak area reproducibility for PAHs in capillary microextraction using sol-gel dendrimer coatings* A x 10 -4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Acenaphthene 3.10 3.45 3.42 3.04 3.37 3.28 5.27 Fluorene 4.79 5.25 5.19 4.73 5.13 5.01 4.27 Phenanthrene 5.65 6.07 5.96 5.95 5.89 5.90 2.33 Fluoranthene 6.27 6.46 5.95 5.95 5.89 6.38 1.46 Pyrene 6.24 6.55 6.65 6.63 6.63 6.48 2.56 _______________________________________________ *CMEGC analysis of PAHs at 10 ppb con centration using sol-gel dendrimer coated microextraction capillary. Extr action conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30min. GC anal ysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C, programmed at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) acenaphthene, (2) fluorene, (3) phenanthrene, (4) fluoranthene, and (5) pyrene. A= Actual computer output for peak area in arbitrary unit.

PAGE 237

213 Table 4.4 Capillary-to-capillary peak area repr oducibility for PAHs in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Capillary 1 Capillary 2 Capillary 3 Capillary 4 Capillary 5 Mean RSD (%) Acenaphthene 1.94 2.04 1.98 2.04 2.00 2.00 2.08 Fluorene 4.24 4.21 4.42 4.12 4.40 4.28 2.58 Phenanthrene 4.85 4.93 4.81 4.66 4.81 4.81 2.04 Fluoranthene 6.45 6.46 6.92 6.32 6.54 6.54 3.46 Pyrene 7.88 8.27 9.04 7.88 8.27 8.27 5.72 _________________________________________________ *CMEGC analysis of PAHs at 10 ppb con centration using sol-gel dendrimer coated microextraction capillary. Extr action conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30min. GC anal ysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C, programmed at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) acenaphthene, (2) fluorene, (3) phenanthrene, (4) fluoranthene, and (5) pyrene. A= Actual computer output for peak area in arbitrary unit.

PAGE 238

214 Table 4.5: Limits of detection (LOD) data for PAHs in CME-GC-FID using sol-gel dendrimer microextraction capillaries*. Measured Noise (V) : 0.975 Compounds Concentration (ppb) Measured Peak Height (V) ( H x 10-3) Limit of Detection (S/N 3), ppt Acenaphthene 10 5.79 3.6 Fluorene 10 12.95 2.3 Phenanthrene 10 14.19 2.1 Fluoranthene 10 13.25 22 Pyrene 10 12.72 2.3 _______________________________________________ *CMEGC analysis of PAHs at 10 ppb con centration using sol-gel dendrimer coated microextraction capillary. Extr action conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30min. GC anal ysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C, programmed at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) acenaphthene, (2) fluorene, (3) phenanthrene, (4) fluoranthene, and (5) pyrene. H= Actual computer output for peak height in V.

PAGE 239

2154.3.4.2 Aldehydes and ketones The sol-gel dendrimer-coated CME capillari es were further used to extract trace levels of aldehydes and ketones (carbonyl compounds) in aqueous samples. Carbonyl compounds play an important role in aquatic ox idation processes. In natural waters, these compounds can be produced by the photodegradation of dissolved na tural organic matter [43] as well as a product of microbiological pr ocesses [44]. In recent years, carbonyl compounds have been receiving much attenti on as disinfection and oxidation by-products formed during drinking water tr eatment processes. Many of these by-products have been shown to carcinogens or carcinogen suspects [45-47]. The polar nature and enhanced reactivity of carbonyl compounds in water matrices often im pose the need for their derivatization prior to their extraction and /or detection by chroma tographic techniques [48-49]. However, derivatization of these an alytes, especially when present in trace concentrations, may complicate the analytical process, leading to poor accuracy and reproducibility. Figure 4.8 is a gas chro matogram representing a mixture of 4 underivatized aldehydes. Table 6 provided a list of these underivatized aldehydes. Microextraction results have been presented in Table 4.7, Table 4.8, and Table 4.9 represents the calcu lated limit of detection (LOD) values for these underivatized aldehydes.

PAGE 240

216 Table 4.6 Chemical structures and pertinent physi cal properties of al dehydes extracted using sol-gel dendrimer coating Name of the analyte Chemical structure Molecular Weight (g/mol) Melting Point (C) Boiling Point (C) Density (g/mL) Nonyl aldehyde m-Tolualdehyde O O 142.2406 120.1506 63 93 at 23 mm Hg 199 0.823 1.019 n-Decylaldehyde O156.2674 7 207.209 0.825 n-Undecylaldehyde O 170.2942 109-115 at 5 mm

PAGE 241

217 Figure 4.8 Capillary microextraction-GC analysis of aldehydes at 100 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 mi n. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; sp litless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) nonylaldehyde, (2) m-tolualdehyde, (3) ndecylaldehyde, and (4) n-undecylaldehyde

PAGE 242

218Table 4.7 Run-to-run peak area reproducibility for al dehydes in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Nonylaldehyde 3.14 3.46 3.58 2.99 3.02 3.24 7.36 m-Tolualdehyde 9.69 9.20 9.62 9.86 9.17 9.51 2.90 n-Decylaldehyde 17.43 16.20 16.65 18.15 16.63 17.01 4.09 n-Undecylaldehyde 21.21 20.20 20.36 23.88 21.50 21.36 6.19 __________________________________________________ *Capillary MicroextractionGC analysis of aldehydes at 100 ppb concentration using sol-gel dendrimer coat ed microextraction cap illary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) nonylaldehyde, (2) m-tolualdehyde, (3) ndecylaldehyde, and (4) n-undecylic aldehyde A= Actual computer output for peak area in arbitrary unit.

PAGE 243

219 Table 4.8 Capillary-to-capillary peak area repro ducibility for aldehydes in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Nonylaldehyde 2.99 3.60 3.61 2.91 3.10 3.25 9.20 m-Tolualdehyde 8.91 9.27 9.41 10.84 8.76 9.63 6.79 n-Decyl aldehyde 16.37 15.43 17.88 19.94 17.35 17.41 8.97 n-Undecyl aldehyde 21.18 17.09 19.75 19.41 19.97 19.73 7.60 __________________________________________________ *Capillary Microextraction-GC analysis of aldehydes at 100 ppb concentration using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) nonylaldehyde, (2) m-tolualdehyde, (3) ndecylaldehyde, and (4) n-undecylic aldehyde A= Actual Computer output for peak area in arbitrary units.

PAGE 244

220 Table 4.9 Limits of detection (LOD) for alde hydes in CMEGC-FID using sol-gel dendrimer microextraction capillaries* Measured noise (V) : 0.681 Compounds Concentration (ppb) Measured peak height (V) (H x 10-3) Limit of detection (S/N 3), ppt Nonylaldehyde 100 10.53 19.4 m-Tolualdehyde 100 36.86 5.6 n-Decyl aldehyde 100 63.07 3.3 n-Undecyl aldehyde 100 58.49 3.5 ________________________________________________________ *Capillary Microextraction-GC analysis of aldehydes at 100 ppb concentration using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; Helium carrier gas; FID temperature 350 C. Peak identification: (1) nonylaldehyde, (2) m-tolualdehyde, (3) ndecylaldehyde, and (4) n-undecylaldehyde. H= Actual computer output for peak height in V.

PAGE 245

221 Ketones were extracted from an aqueous so lution containing 100 ppb of each analyte. The data presented in Table 4.9 indicate s that the detection limits obtained for underivatized aldehydes in CME-GC-FID using a sol-gel dendrimer coated microextraction capillary range between 3.5 ppt and 19.4 ppt. These values are fairly comparable to the values reported in th e literature, which were achieved through derivatization proce ss using commercial SPME fibers. For instance, Chancho et al.[ 48] reported a detection limit of 0.02 ng/mL (200 ppt) for decanal obtained by SPME-GCECD on a commercial SPME fiber having 65 m thick DVB-PDMS coating, which is significantly higher than the value 0.003 ng/mL (3 ppt) obtained on sol-gel dendrimer CME-GC-FID. The same trend was also obs erved for other alde hydes. It is worth mentioning that ECD often provides higher sensitivity than FID for oxygenated compounds like aldehydes. Table 4.10 provides a list of 5 ketones wh ich were extracted and analyzed using sol-gel dendrimer coated capillaries. Capillary microextraction results are presented in Table 4.11, Table 4.12 and Tabl e 4.13. Figure 4.9 represents a gas chromatogram of the mixture of 5 underivatized ke tones extracted from an a queous solution containing 100 ppb of each analyte. Excellent reproducibility for the capillary microextraction of ketones (RSD values for run-to-run reproducibility ranges from 1.37 % to 3.39%, while RSD values for capillary-to-capillary reproducibi lity ranges from 2.08 % to 6.45 %) is a good indicator of the superior pe rformance of the sol-gel dendr imer coated microextraction capillaries as well as the method developed in these experiments.

PAGE 246

222 Table 4.10 Chemical structures and pertinent physical pr operties of ketones extracted using sol-gel dendrimer coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) Butyrophenone O 148.2042 11-13 220-222 0.988 Valerophenone O 162.231 -9 105-107 at 5 mm Hg 0.988 Hexanophenone O 176.2578 25-26 265.1 0.958 Heptanophenone O 190.2846 17 155 at 15 mm Hg 0.946 Benzophenone O 18.2214 48.5 305.4 1.5893

PAGE 247

223Table 4.11 Run-to-run peak area reproducibility fo r ketones in capillar y microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Butyrophenone 3.65 3.76 3.64 3.73 3.64 3.68 1.37 Valerophenone 6.38 6.58 6.17 6.30 6.14 6.31 2.52 Hexanophenone 7.75 8.25 8.09 8.13 8.28 8.10 2.33 Heptanophenone 9.15 9.89 9.66 9.68 9.85 9.65 2.31 Benzophenone 6.26 6.75 6.14 6.34 6.09 6.32 3.39 ____________________________________________ *Capillary Microextraction-GC analysis of ketones at 100 ppb concentration using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) bu tyrophenone, (2) valerophenone, (3) hexanophenone, (4) heptanophenone, and (5) benzophenone. A= Actual computer output for peak area in arbitrary unit.

PAGE 248

224 Figure 4.9 Capillary Microextraction-GC analys is of ketones at 100 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 mi n. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; sp litless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) bu tyrophenone, (2) valerophenone, (3) hexanophenone, (4) heptanophenone, and (5) benzophenone.

PAGE 249

225Table 4.12 Capillary-to-capillary peak area repr oducibility for ketones in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Butyrophenone 3.30 30.98 3.18 2.96 3.21 3.15 3.70 Valerophenone 6.28 5.87 6.12 5.88 6.32 60.91 3.12 Hexanophenone 9.71 9.64 10.11 9.19 10.24 9.78 3.81 Heptanophenone 8.29 9.02 9.71 9.20 10.02 9.25 6.45 Benzophenone 6.99 6.60 6.72 6.83 6.93 6.81 2.08 _____________________________________________________ *Capillary MicroextractionGC analysis of ketones at 100 ppb concentration using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) bu tyrophenone, (2) valerophenone, (3) hexanophenone, (4) heptanophenone, and (5) benzophenone A= Actual computer output for peak area in arbitrary unit.

PAGE 250

226Table 4.13 Limits of detection (LOD) for ketones in CME-GC-FID with sol-gel dendrimer microextraction capillaries* Measured Noise (V) : 0.853 Compounds Concentration (ppb) Measured Peak Height (V) (H x 10-3) Limit of Detection (S/N 3), ppt Butyrophenone 200 12.60 44.3 Valerophenone 100 23.11 11.7 Hexanophenone 50 29.95 3.7 Heptanophenone 25 33.10 1.9 Benzophenone 100 16.86 15.2 _____________________________________________________ *Capillary Microextraction-GC analysis of ketones at 100 ppb concentration using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) bu tyrophenone, (2) valerophenone, (3) hexanophenone, (4) heptanophenone, and (5) benzophenone H= Actual computer output for peak height in V.

PAGE 251

2274.3.4.3. Phenols The next class of compounds that were extracted using sol-gel dendrimer coated capillaries was phenols. The presence of phe nolic compounds in the environment is of great concern because of their role in drinki ng water pollution [50], their toxicity [51], and widespread use in the industry [52,53]. Due to their toxicity and persistence in the environment, 11 phenolic compounds have been included in EPA Priori ty Pollutant list [54]. Since phenolic compounds ar e highly polar, it is quite difficult to extract them directly from aqueous media. Derivatization, pH adjustment and/or salting-out are often used to facilitate the extraction [2, 55]. To avoid the analytical complexity due to derivatization, HPLC is a frequent choice for the analysis of phenolic compounds [56, 57]. In the present study, no solute deriva tization, pH adjustment or salting out of the aqueous sample was used to extract phenolic compounds from the aqueous medium. Still, sol-gel dendrimer coat ed microextraction capillaries provided lower detection limits compared to other reports in the literature. Fo r example, we achieved a detection limit of 0.22 ppb for 4-chloro, 3-methyl phenol which is lower than the valu e (1.4 ppb) obtained by Buchholz et al.[2] on a SPME fiber with 95-m thick polyacrylate coating. Same trend was also observed for other pheno lic compounds. A list of phenolic compounds used in this experiment is given in Tabl e 4.14. Run-to-run and capillary-to-capillary microextraction reproducibil ity results are provided in Table 4.15 and Table 4.16, respectively. A gas chromatogram obtained from these experiments is shown in Figure 4.10. Like other analytes extracted and analyzed using sol-gel dendrimer coated microextraction capillaries and GC-FID, phenolic compounds were also characterized by

PAGE 252

228 high run-to-run and capillary-tocapillary microextra ction reproducibility as evidenced by low RSD values in both instances. This has b ecome possible due to the excellent material property of the dendrimer-based sorben t acquired by sol-gel coating technology developed by Malik and co-workers. Table 4.14 Chemical structures and pertinent physical properties of phenols analyzed using sol-gel dendrimer coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) 2-Chlorophenol OH Cl Cl 152.20 92.5 150.0 0.8987 2,5Dimethylphenol OH CH3H3C 166.22 114.8 295.0 1.203 3,4Dichlorophenol OH Cl Cl 178.23 99.2 340 0.9820 2,4,6Trichlorophenol OH Cl Cl Cl 202.26 107.8 384 1.252 4-Cloro-3methylphenol OH Cl CH3 202.26 156 404 1.271

PAGE 253

229Table 4.15: Run-to-run peak area reproducibility for phe nols in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) 2-Chlorophenol 1.58 1.75 1.59 1.64 1.51 1.62 5.29 2,5Dimethylphenol 2.90 2. 82 2.88 2.84 2.90 2.87 1.26 3,4Dichlorophenol 1.85 1.92 1.84 2.04 2.05 1.94 5.06 2,4,6Trichlorophenol 15.46 16. 31 15.16 15.43 15.75 15.57 2.15 4-Chloro-3methylphenol 6.40 6.35 6.35 6.42 6.39 6.38 0.49 _________________________________________________ *Capillary MicroextractionGC analysis of phenols at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C/min, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 2-chlor ophenol, (2) 2,5-dimethylphenol, (3) 3,4dichlorophenol, (4) 2,4,6trichlorophenol, and (5) 4-chloro-3-methylphenol. A= Actual computer output for peak area in arbitrary unit.

PAGE 254

230Table 4.16: Capillary-to-capillary peak area reproducibility for phenols in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) 2-Chlorophenol 1.37 1.27 1.31 1.13 1.36 1.29 7.59 2,5Dimethylphenol 2.69 3. 04 2.72 2.69 2.69 2.76 5.58 3,4Dichlorophenol 1.84 1.92 1.95 1.84 1.89 1.89 2.53 2,4,6Trichlorophenol 14.61 15. 49 14.53 14.05 14.17 14.60 3.87 4-Chloro-3methylphenol 6.99 6.41 6.06 5.99 6.94 6.48 7.37 _____________________________________________ *Capillary Microextraction-GC analysis of phe nols at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C/min, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 2-chlor ophenol, (2) 2,5-dimethylphenol, (3) 3,4dichlorophenol, (4) 2,4,6trichlorophenol, and (5) 4-chloro-3-methylphenol. A= Actual computer output for peak area in arbitrary unit.

PAGE 255

231Table 4.17 Limits of detection (LOD) for phe nols in CME-GC-FID using sol-gel dendrimer microextraction capillaries* Measured noise (V) : 0.681 Compounds Concentration (ppb) Measured peak height (V) ( H x 10-3) Limit of detection (S/N 3), ppt 2-Chlorophenol 100 3.78 840 2,5Dimethylphenol 100 8.36 320 3,4Dichlorophenol 100 6.49 160 2,4,6Trichlorophenol 100 18.26 220 4-Chloro-3mthylphenol 100 11.44 260 _______________________________________________ *Capillary MicroextractionGC analysis of phenols at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C/min, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 2-chlor ophenol, (2) 2,5-dimethylphenol, (3) 3,4dichlorophenol, (4) 2,4,6trichlorophenol, and (5) 4-chloro-3-methylphenol. H= Actual computer output for peak height in V.

PAGE 256

232 Figure 4.10 Capillary Microextraction-GC analys is of phenols at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 mi n. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; sp litless injection; injector temperature, initial 30 C, program rate 100 C/min, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 2-chlo rophenol, (2) 2,5-dimethylphenol, (3) 3,4-dichlorophenol, (4) 2,4,6-trichlor ophenol, and (5) 4-chloro-3-methylphenol.

PAGE 257

2334.3.4.4 CME of butylated hydroxytoluene (BHT) Butylated hydroxytoluene (BHT) was another compound that was extracted on sol-gel dendrimer coated capillaries. Figur e 4.11 represents a gas chromatogram of butylated hydroxytol uene (BHT) obtained in a CME-GC-F ID experiment using a sol-gel dendrimer extraction capilary. Table 4.18 lists the chemical structure and other pertinent properties of BHT. Data obtained from capilla ry microextraction of BHT using sol-gel Dendrimer coated capillaries are presented in Table 4.19 and Table 4.20. Calculated limit of detection data is presented in Table 4.21. BHT is one of the most commonly used a dditives (preservative) used in rubber, petroleum products, organic solvents, plastics, foods, edible fats and oils, cosmetics [58]. Although controversial, BHT is suspected to possess carc inogenic properties [59]. Therefore, the Joint Expert Committee on F ood Additives (JECFA) of the Word Health Organization (WHO) has set th e acceptable daily intake ( ADI) of BHT at 0-0.3 mg/kg body weight. Recently Tombeshi and Freije [60] proposed an SPME-GC/MS technique for the determination of BHT. In SPME-GC-FI D, they achieved a detection limit of 4.2 ng/mL using a 100-m thick PDMS coating on the SPME fiber. By comparison, a detection limit of 3.0 pg/mL was achieved in the present study usi ng CME-GC-FID with a sol-gel dendrimer-coated microe xtraction capillary. This co rresponds to three orders of magnitude improvement in the detection limit and extraction sensitivity.

PAGE 258

234Table 4.18 Chemical structure and pertinent phy sical properties of butylated hydroxy toluene (BHT) extracted usi ng sol-gel dendrimer coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) Butylated hydroxytoluene (BHT) OH220.354 71 265 1.048 Table 4.19 Run-to-run peak area reproducibility fo r BHT in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Butylated hydroxytoluene 7.31 7.75 7.27 7.97 8.14 7.69 5.06 ________________________________________________ *Capillary Microextraction-GC analysis of BHT at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. A= Actual computer output for peak area in arbitrary unit.

PAGE 259

235Table 4.20: Capillary-to-capillary peak area reproducibility for BHT in capillary microextraction using sol-gel dendrimer coatings*. A x 10-4 Compounds name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) Butylated Hydroxytoluene 7.21 6.99 8.00 7.74 7.22 7.43 5.67 __________________________________________ *Capillary Microextraction-GC analysis of BHT at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. A= Actual computer output for peak area in arbitrary unit.

PAGE 260

236Table 4.21: Limit of detection (LOD) data for BHT in CME-GC-FID using sol-gel dendrimer microextraction capillaries* Measured noise (V) : 0.975 Compounds Concentration (ppb) Measured peak height (V) (H x 10-3) Limit of detection (S/N 3), ppt Butylated hydroxytoluene 10 5.85 3.0 ____________________________________________ *Capillary MicroextractionGC analysis of BHT at 10 ppb concentration using sol-gel dendrimer coated microextrac tion capillary. Extraction c onditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. H= Actual computer output for peak height in V.

PAGE 261

237 Figure 11 Capillary Microextractio n-GC analysis of BHT at 10 ppb concentration using sol-gel dendrimer coat ed microextraction cap illary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C.

PAGE 262

2384.3.4.5. Alcohols Figure 4.12 represents a gas chromat ogram of a mixture of alcohols (10 ppb concentration of each). Extraction of these polar compounds was conducted from aqueous samples without any derivatization, pH adjustment or salting-out effects. The presented data shows excellent affinity of the sol-gel dendrimer co ating for these highly polar analytes that are often difficult to extract from aqueous media in the underivatized form using commercial coati ngs. Excellent symmetrical pe ak shapes and high detection sensitivity (Table 4.25) are indicative of outstanding performance of the used sol-gel dendrimer coating in CME and excellent d eactivation characteristics of sol-gel PEG column used for the GC analys is of the extracted alcohols. As is revealed from the data presented in the text, run-to-run and capillary-tocapillary repeatability data for peak area obtained in CME-GC-FID experiments are quite satisfactory. For most solutes, these RSD va lues were under 5%. For the polar analytes, the RSD values were higher than those for nonpolar analytes. Retenti on time repeatability data for PAHs, aldehydes, ketones, phenols, and alcohols were characterized by RSD values of less than 0.14 %. Unique molecular architecture of dendrimer s and the ability of sol-gel dendrimer coatings to provide efficient and reproduc ible extraction of both polar and nonpolar compounds with high detecti on sensitivity makes dendrimer-based materials very promising in analytical extraction technology

PAGE 263

239 Table 4.22 Chemical structures and pertinent phys ical properties of alcohols extracted using sol-gel dendrimer coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) 1-Octanol OH130.2296 -15 195 0.826 1-Nonanol OH144.2564 -8 to -6 125 0.828 1-Decanol OH158.2832 6 230 0.829 1-Undecanol OH172.32 11 146 at 30 mm Hg 0.832 1-Dodecanol OH186.3368 22-26 260.262 0.833

PAGE 264

240Table 4.23 Run-to-run peak area reproducibility for alc ohols in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Run 4 Run 5 Mean RSD (%) 1-Octanol 12.60 13.15 13.74 14.31 14.85 13.73 6.52 1-Nonanol 12.55 13.01 14.13 14.52 13.61 13.52 5.13 1-Decanol 10.91 11.29 11.15 11.38 11.66 11.28 2.46 1-Undecanol 9.35 9.61 10.00 9.94 9.62 9.76 2.88 1-Dodecanol 13.82 15.44 14.41 14.68 14.73 14.62 4.00 __________________________________________________ *Capillary Microextraction-GC analysis of al cohols at ppb level concentrations using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 1-oc tanol (500ppb), (2) 1-nonanol (100 ppb), (3) 1-decanol (30 ppb), (4) 1-undecanol (20 ppb), and (5) 1-dodecanol (50 ppb). A= Actual computer output for peak area in arbitrary unit.

PAGE 265

241Table 4.24 Capillary-to-capillary peak area reproducibility for alcohols in capillary microextraction using sol-gel dendrimer coatings* A x 10-4 Compounds Name Capillary 1 Capillary 2 Capillary 3 Capillary 4 Capillary 5 Mean RSD (%) 1-Octanol 12.35 12.77 13.52 14.21 14.41 13.46 6.64 1-Nonanol 12.73 12.94 13.61 14.42 13. 91 13.46 2.50 1-Decanol 10.97 11.37 11.25 11.47 11.77 11.01 4.27 1-Undecanol 9.25 9.71 10.10 9.96 9.71 9.75 3.30 1-Dodecanol 13.63 15.34 14.71 14.28 14.71 14.53 4.35 _____________________________________________ *Capillary MicroextractionGC analysis of alcohols at ppb level concentrations using sol-gel dendrimer coat ed microextraction cap illary. Extraction conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 1-oc tanol (500ppb), (2) 1-nonanol (100 ppb), (3) 1-decanol (30 ppb), (4) 1-undecanol (20 ppb), and (5) 1-dodecanol (50 ppb). A= Actual computer output for peak area in arbitrary unit.

PAGE 266

242Table 4.25 Limits of detection (LOD) for alco hols in CME-GC-FID using sol-gel dendrimer microextraction capillaries* Measured noise (V) : 0.975 Compounds Concentration (ppb) Measured peak height (V) (H x 10-4) Limit of detection (S/N 3), ppt 1-Octanol 500 3.93 11.2 1-Nonanol 100 5.29 2.3 1-Decanol 30 4.39 1.0 1-Undecanol 20 3.16 1.0 1-Dodecanol 50 2.83 1.8 _________________________________________________________ *Capillary Microextraction-GC analysis of al cohols at ppb level concentrations using solgel dendrimer coated microe xtraction capillary. Extracti on conditions: 13 cm x 0.25 mm i.d. microextraction capillary; extraction tim e, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. sol-gel PDMS column; splitless in jection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperatur e programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 1-oc tanol (500ppb), (2) 1-nonanol (100 ppb), (3) 1-decanol (30 ppb), (4) 1-undecanol (20 ppb), and (5) 1-dodecanol (50 ppb). H= Actual computer output for peak height in V.

PAGE 267

243 Figure 4.12 Capillary Microextraction-GC analysis of alcohols at ppb level concentrations using sol-gel dendrimer co ated microextraction capillary. Extraction conditions: 13 cm x 0.25 mm i. d. microextraction capillary; extraction time, 30 min. GC analysis conditions: 10 m x 0.25 mm i.d. so l-gel PDMS column; splitless injection; injector temperature, initial 30 C, program rate 100 C min-1, final 300 C; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C min-1; helium carrier gas; FID temperature 350 C. Peak identification: (1) 1-octanol (500ppb), (2) 1-nonanol (100 ppb), (3) 1-decanol ( 30 ppb), (4) 1-undecanol (20 ppb), and (5) 1-

PAGE 268

244 dodecanol (50 ppb). 4.4 Conclusion For the first time, sol-gel dendrimer coat ed capillaries were used for solvent-free microextraction and preconcentration in ch emical analysis. Both polar and nonpolar analytes were efficiently extracted from a queous samples on the same sol-gel dendrimer capillary and provided excellent detection sensitivity. Parts per trillion level detection limits were achieved in CME-GC-FID using sol-gel dendrimer-coated extraction capillaries. It should be po ssible to further enhance the extraction sensit ivity by using capillaries with (1) larger -diameters (e.g., 320 m, 520 m ), (2) greater lengths (3) thicker CME coatings, and (4) sol-gel monolith ic extraction beds, or any combination of these factors. Since sol-gel dendrimer ex traction phase shows excellent thermal and solvent stability, sol-gel dendrimer CME capil laries are suitable for coupling with both GC and HPLC.

PAGE 269

245 4.5 References for Chapter Four [1] R.P. Belardi, J. Pawliszyn, Wate r Pollut. Res. J. Can. 24 (1989) 179. [2] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [3] K.D. Buchholz, J. Pawlisz yn, Anal. Chem. 66 (1994) 160. [4] D.W. Potter, J. Pawliszyn, E nviron. Sci. Technol. 28 (1994) 298. [5] C.L. Arthur, J. Pawliszy n, Anal. Chem. 62 (1990) 2145. [6] J. Wu, J. Pawliszyn, J. Chromatogr. A 909 (2001) 37. [7] H. Kataoka, K. Mitani, Jpn. J. Forensic Toxicol. 20 (2002) 251. [8] W.M. Mullett, P. Martin, J. Pa wliszyn, Anal. Chem. 73 (2001) 2383. [9] T.J. Clark, J.E. Bunch, J. Chromatogr. Sci. 35 (1997) 209. [10] O.E. Mills, A. Broome, J. ACS Symp. Ser. 705 (1998) 85. [11] V. Mani, in: J. Pawliszyn (Ed. ), Applications of Solid-phase Microextraction, Royal So ciety of Chemistry (RSC): Cambridge (UK), 1999, 57. [12] S.L. Chong, D. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [13] A. Malik, and S.L. Chong, in J. Pawl iszyn (Ed.) Applicati ons of Solid-phase Microextraction, Royal Society of Chemistry (RSC), Cambridge (UK), 1999, Chap. 6, p. 73. [14] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157.

PAGE 270

246 [15] Z. Zeng, W. Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [16] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [17] M.P. Stevens, Polymer Chemistry, Ox ford University Press: New York, 1999. [18] G.R. Newkome, Z. Yao, G.R. Baker, V.K. Gupta, J. Org. Chem. 50 (1985) 2003. [19] D.A. Tomalia, H. Baker, J. Dewald, M. Hall, G. Kallos, S. Martin, J. Roeck, J. Ryder, P. Smith, Polym. J. 17 (1985) 117. [20] K.W. Pollak, J.W. Leon, J.M.J. Frechet, M. Maskus, H.D. Abruna, Chem. Mater. 10 (1998) 30. [21] M. Albrecht, R.A. Gossage, A.L. Spek G. van Koten, J. Chem. Soc., Chem. Commun. 9 (1998) 1003 [22] J.R. McElhanon, D.V. McGrath, J. Am. Chem. Soc. 120 (1998) 1647. [23] P. W. Wang, Y. J. Liu, C. Devadoss, P. Bharati, J.S. Moore, Adv. Mater. 8 (1996) 237. [24] D.A. Tomalia, A.M. Naylor, W.A. Goddard III, Angew. Chem. Int. Edit. Engl. 29 (1990) 138. [25] M.K. Misra, S. Kobayashi (Eds.), St ar and Hyperbranched Polymers, Mercel Dekker, New York, 1999. [26] K.L. Wooley, J.M.J. Frechet, C. J. Hawker, Polymer 35 (1994) 4489. [27] G.R. Newkome, C.N. Moorefield, F. Vogtle, Dendrimers and Dendron. Concept, Synthesis, Applications. Wiley-VCH, Weinheim, 2001.

PAGE 271

247 [28] G.R. Newkome, J.K. Young, G.R. Baker, R.L. Potter, L. Audoly, D. Cooper, C.D. Weis, K. Morris, C.S. Johnson, Jr. Macromolecules 26 (1993) 2394. [29] S.A. Kuzdzal, C.A. Monning, G.R. Newk ome, C.N. Moorefield, J. Chem. Soc., Chem. Commun. 18 (1994) 2139. [30] P.G.H.M. Muijselaar, H.A. Claessens, C.A. Cramers, J.F.G.A. Jansen, E.W. Meijer, E.M.M. de Brabander-van den Berg, S. van der Wal, J. High. Resol. Chromatogr. 18 (1995) 121. [31] C.P. Palmer, N. Tanaka, J. Chromatogr. A 792 (1997) 105. [32] H.C. Chao, J.E. Hanson, J. Sep. Sci. 25 (2002) 345. [33] B.T. Mathews, A.E. Beezer, M.J. S nowden, M.J. Hardy, J.C. Mitchell, Chromatographia 53 (2001) 147. [34] G.R. Newkome, K.S. Yoo, A. Kabir, A. Malik, Tetrahedron Lett. 42 (2001) 7537. [35] L.G. Blomberg, J. Microcolumn Sep. 2 (1990) 62. [36] A. Malik, in: J. Pawliszyn (Ed.), Compre hensive Analytical Chemistry, Elsevier, Amsterdam, 2002, p.1023. [37] Z. Zhang, M.J. Yang, J. Paw liszyn, Anal. Chem. 66 (1994) 844A. [38] D.X. Wang, S.L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [39] C. Shende, A. Kabir, E. Townsend, A. Malik, Anal. Chem. 75 (2003) 3518. [40] J.D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3. [41] A. Byorseth, T. Ramdahl (Eds.), Handbook of Polycyclic Aromatic Hydrocarbons, Vol. 2, Emission, Sources, and Recent Progress in Analytical Chemistry, Marcel Dekker: New York, Basel, 1985, p.1.

PAGE 272

248 [42] R.A. Doong, S.M. Chang, Y.C. S un, J. Chromatogr. A 879 (2000) 177. [43] R.J. Kieber, K. Mopper, Envir on. Sci. Technol. 24 (1990) 1477. [44] I. Chorus, G. Klein, J. Fastner, W. Rotard, Wat. Sci. Technol. 25 (1992) 251. [45] M.O. Amdur, Air Pollutanta. in: M.O. Amdur, J. Doull, C.D. Klaassen (Eds.), Casarett and Doulls Toxicology: The Basic Science of Poisons, 4th ed., Pergamon Press, New York, 1991, p. 866, Chapter 25. [46] National Research Council. Formal dehyde and other Aldehydes: Board on Toxicology and Environmental Health Hazards, National Academy Press, Washington DC, 1981. [47] G.D. Leikauf, Formaldehyde and other aldehydes. in: M. Lippmann (Ed.) Environmental Toxicants: Human Exposur es and Their Health Effects, Van Nostrand Reinhold: New York, 1992, 299. [48] B. Cancho, F. Ventura, M. T. Galceran, J. Chromatogr. A 943 (2001) 1. [49] M.L. Bao, F. Pantani, O. Griffini, D. Burrini, D. Santiani, K. Barbieri, J. Chromatogr. A 809 (1998) 75. [50] W. Fresinius, K.E. Quentin, W. Schneid er (Eds.) Water Analysis: A Practical Guide to Physico-Chemical, Chemical and Microbiological Water Examination and Quality Assurance, Springer-Verlag, Berlin, Germany, 1988. [51] Material Safety Data Sheet for Phenol Genium Publishing Corp., Schenectady, NY, 1985. [52] J.W. Moore, S. Ramamoorthy, Phenols in Organic Chemicals in Natural Waters. Applied Monitoring and Impact A ssessment, Springer: New York, 1984.

PAGE 273

249 [53] W. Rowe, Evaluation Methods for Envi ronmental Standards, CRC Press, Boca Raton, FL, 1983. [54] Fed. Reg., EPA method 604, Phenols, Part VIII, 40 CFR Part 136, US Environmental Protection Agency, Wa shington, DC, 26 October 1984, p. 58. [55] P. Bartak, L. Cap, J. Chromatogr. A 767 (1997) 171. [56] M. Moder, S. Schrader, U. Frank, P. Popp, J. Fresenius, Anal. Chem. 357 (1997) 326. [57] M. -R. Lee, T. -C. Yeh, W. -S. Hsiang, B. -H. Hwang, J. Chromatogr. A 806 (1998) 317. [58] IRC Monographs on The Evaluation of Carcinogenic Risks to Human and their Supplements, Vol. 40, International Ag ency for Research on Cancer (IARC), IARC Press: Lyon, 1986,161. [59] Clayson, D.B.; Iverson, F.; Nera, E.A.; Lok, E. Toxicol. Ind. Health 9 (1993) 231242. [60] Tombeshi, N.B.; Freije, H. J. Chromatogr. A 963 (2002) 179-183. [61] A. Kabir, C. Hamlet, K.S. Yoo, G.R. Newkome, A. Malik, J. Chromatogr. A 1034 (2004) 1-11.

PAGE 274

250 Chapter Five Capillary Microextraction on Sol-Gel Poly-Tetrahydrofuran Coating 5.1 Introduction Solid-phase microextraction (SPME) [1] is an excellent solventless alternative to the traditional sample preparation technique s like liquid-liquid ex traction (LLE), Soxhlet extraction, solid-phase extraction (SPE), etc. It is a simple, sensitive, time-efficient, costeffective, reliable, easy-to-automate, and portable sample preparation technique. In SPME, analyte enrichment is accomplished by using a sorbent coating in two different formats: (a) conventional fiber format [1] a nd (b) the more recently developed “in-tube” format [2]. In its conventional format, SPM E uses a sorbent coating on the external surface of a fused silica fiber typically 100-200 m in diameter covering a short segment at one of the ends. In the in-tube fo rmat, the sorbent coating is applied to the inner surface of a capillary. SPME completely el iminates of the use of organic solvents in sample preparation, and effectively integrates a number of important analytical steps such as sampling, extraction, preconcentration, and sample introducti on for instrumental analysis. Thanks to these positive attributes SPME has experienced a rapid growth over the last decade. Despite rapid advancements in SPME applications, a number of important

PAGE 275

251 problems in the area of sorbent coatings sti ll remain to be solved. First, existing SPME coatings are designed to extrac t either polar or nonpolar anal ytes from a given matrix. For example, being a nonpolar sorbent, polydi methylsiloxane (PDMS) shows excellent selectivity toward nonpolar anal ytes. The polar polyacrylate coating, on the other hand, demonstrates excellent selectivity toward pol ar compounds. Such an approach is not very convenient for samples where both polar and nonpolar contaminants are present and both need to be analyzed. For such applications it is important to have a sorbent that can extract both polar and nonpolar compounds with high extraction sensitivity needed for trace analysis. Second, in conventional SPME onl y a short length of the fiber is coated with sorbent. The short length of the coated segment on the SPME fiber translates into low sorbent loading which, in turn, lead s to low sample capacity. This imposes a significant limitation on the sensi tivity of the classical fiber SPME. Improving sensitivity is still a major challenge in SPME research. Th is is particularly important for analyzing ultra-trace contaminants that are present in the environment. One possible way of improving extraction sensitivity in SPME is th rough an increase in the coating thickness [3, 4]. However, equilibration time rapidly increases with the increase in coating thickness because of the dynami c diffusion-controlled nature of the extraction process [3]. As a consequence, both extraction a nd subsequent desorption processes become slower, resulting in longer total analysis time. More over, immobilization of thicker coating on fused silica surface is difficult to achieve by conventional approaches [5] indicating to the necessity of an alternative approach to effective immobilization of thick coatings. Third, low thermal and solvent stab ility of SPME coatings represents a major drawback of conventional SPME technology, an d is a direct consequence of the poor

PAGE 276

252 quality of sorbent immobilizat ion. With a very few exceptions, SPME fibers have been traditionally coated by mere physical deposition of the sorbent in th e form of a surface coating. The absence of chemical bonds betw een the sorbent coati ng and the fused silica surface is considered to be the main reason for low thermal and solvent stability of SPME fibers [6]. Low thermal stability of thick coatings forces one to use low desorption temperatures to preserve coating integrity, which in turn, leads to incomplete sample desorption and sample carryover problems. Besides, low solvent stability of the coating poses a significant obstacle to reliable hyphenation of in-tube SPM E with liquid-phase separation techniques (e.g., H PLC) that employ organic or organo-aqueous mobile phases [3]. It is evident that future advancem ents in SPME would greatly depend on new developments in the areas of sorbent chemistry and coati ng technology that will allow preparation of chemically immobilized coatings from advanced material systems providing desired selectivity and performance in SPME. One possible approach to address most of the problems described above is to use sol-gel technology to create sorbent coatings [6-9]. Sol-gel chemistry provides a simple and convenient pathway leading to the synthesis of advanced material systems that can be used to prepare surface coatings [10,11]. In the context of fused silica fiber/capillarybased SPME, major advantages offered by so l-gel technology are as follows: (1) it combines surface treatment, deactivation, coa ting, and stationary phase immobilization into a single-step procedure making the whole SPME fiber/capillary manufacturing process very efficient and cost effective; (2) it creates ch emical bonds between the fused silica surface and the created sorbent coating; (3) it provides surface-coatings with high operational stability ensuring reproducible performance of the sorbent coating under

PAGE 277

253 operation conditions involving high temperature and/ or organic solvents, and thereby it expands the SPME application range toward both higher-boilingas well as thermally labile analytes; (4) it provides the possibili ty to combine organic and inorganic material properties in extracting sorbents providi ng tunable selectivity; (5) it offers the opportunity to create sorbent coatings with a porous st ructure which significantly increases the surface area of the extracti ng phase and provides acceptable stationary phase loading and sample capacity using thinner coatings. A number of shortcomings inherent in conventional fiber SPME originate from the design and physical construction of th e fiber and the syringe-like SPME device. These include susceptibility of the fiber to breakage during coa ting or operation, mechanical damage of the coating due to sc raping, and operational uncertainties due to needle bending. In-tube SPME [2], also term ed capillary microextr action (CME) [7], is practically free from these inherent format -related shortcomings of conventional fiber SPME. It uses a fused silica capillary generally a small piece of GC column with a stationary phase coating on the inner surf ace to perform extraction. The protective polyimide coating outside the capillary rema ins intact and provide s reliable protection against breakage. Moreover, this format provi des a simple, easy, and convenient way to couple SPME to high-performan ce liquid chromatography. Despite numerous advantageous features, in-tube SPME still has drawbacks that originate mainly from the deficiency of the coating technique used to prepare the extraction capillary. Conven tional static coating techni que, commonly employed to prepare GC capillary columns (short segments of which are used for in-tube SPME), is not suitable for generating thic k coatings necessary for enhan ced extraction sensitivity in

PAGE 278

254 SPME. Besides, in general, a conventionally prepared coating is not chemically bonded to the fused silica capillary su rface. As a consequence, such coatings exhibit low thermal and solvent stability. Recently, sol-gel capillary microextraction CME has been proposed [7] to address the above-mentioned problems through in situ creation of surface-bonded coatings via sol-gel technology capable of providing both thick and thin coatings on the capillary inner walls. In both conventional fiber SPME and CME, the sorbent coating plays a critically important role in the extraction process. To date, several types of sorbent coatings have been developed and used for extraction. Thes e coatings can be broa dly divided into two major types: (1) single-phaseand (2) com posite coatings. Single-phase SPME coatings include polydimethylsiloxane (PDMS) [12], Polyacrylate [3], Carbopack [13], polyimide [14], polypyrrole [15], and molecularly imprin ted materials [16,17]. Among the composite coatings are Carbowax /divinylbenzene (CW/DVB) [18], polydimethylsiloxane/divinylbenzene (PDMS/DVB) [19], polydimethylsiloxane/Carboxa ne (PDMS/Carboxane) [20], and Carbowax/templated resin (CW/TPR) [20]. In recent years, sol-gel SPME coatings [6 ,7,21-27] have drawn wide attention due to their inherent advantageous features and performance superiority over traditional coatings (both non-bonded and cross-lin ked types). Sol-gel PDMS [6,7,22,28] coatings possess significantly higher thermal stability (> 360 C) than their conventional counterparts for which the upper temperature limit generally remains within 200-270 C [29]. High thermal and solvent stability have been demonstrated for other sol-gel coatings sol-gel PEG [23] 320 C, sol-gel crown ethers [25] 340 C sol-gel

PAGE 279

255 hydroxyfullerene [27] 360 C), sol-gel polymethylphe nylvinylsiloxane [26]350 C, and sol-gel phenyl-functi onalized coating [69] (350 C). Sol-gel PEG coating [23] has been recommended for polar analytes. Sol-gel crown ether coating [24-25] de monstrated higher extraction efficiencies for aromatic amines compared to CW-DVB fiber. Gbatu et al. [21] described the preparation of sol-gel octyl coatings for SPME-HPLC analysis of organometallic compounds from aqueous solutions. Compared with the commercial SPME coatings, a hydroxyfullerene-based solgel coating [27] showed higher sensitivity, faster mass transfer rate for aromatic compounds and possessed molecular planarity recognition capability for polychlorinated biphenyls (PCBs). Yang et al. [26] prepared sol-gel poly (m ethylphenylvinylsiloxane) (PMPVS) coating using sol-gel technology that provided very high extraction efficiency for aromatic compounds. Poly-THF (also called poly tetramethylene oxide, PTMO) is a hydroxy-terminated polar material that has been used as an organic component to synthesize organicinorganic hybrid materials [30-35]. Sol-gel poly-THF has been used as a bioactive bone repairing material [36], and as a proton conducting solid polymer electrolyte that might allow the operation of high temperature fuel cells [37]. Little work has been devoted to explore the potential of the so l-gel poly-THF material for use as an extraction medium in analytical chemistry. In this chapter, we de scribe a sol-gel chemis try-based approach to in situ creation of poly-THF based hybrid organic-inorganic coa tings on the inner walls of fused silica capillaries and demonstrate the possi bility of using such coatings to achieve parts per trillion (ppt) and parts per quadri llion level detection lim its for both polar and nonpolar analytes after extraction from aqueous sample matrices.

PAGE 280

2565.2 Experimental 5.2.1 Equipments Capillary microextraction-gas chroma tography (CME-GC) experiments with sol-gel poly-THF coating as the extracti ng phase were carried out on a Shimadzu model 17A GC system (Shimadzu Corporat ion, Kyoto, Japan) equipped with a programmed temperature vaporizer (PTV injector) and a flame ionization detector (FID). An in-house designed liquid sample dispenser (Figure 5.1) was used to perform CME via gravity-fed flow of the aqueous samples through the sol-gel poly-THF coated capillary. A Fisher Model G-560 Genie 2 Vort ex (Fisher Scientif ic, Pittsburgh, PA) was used for thorough mixing of sol solution ingredients. A Microcentaur model APO 5760 microcentrifuge (Accu rate Chemical and Scientific Corporation, Westbury, NY) was used for centrifugation (at 13000 rpm, 15682 g) of sol solutions made for coating the microextraction capillaries. An Avatar model 320 FTIR System (Nicolet Analytical Instruments, Madison, WI) was used to obt ain the IR spectra of poly-THF, sol-gel solution, and sol-gel polyTHF sorbent. A JEOL model JSM-35 scanning electron microscope was used for the investigation of the coated capillary surface. A homebuilt, gas pressure-operated fi lling/purging device [38] was used to perform a number of operations: (a) filling the extraction capillary with the sol solution, (b) expelling the solution from the capillary after predetermine d period of in-capillary residence, and (c) purging the microextraction capillary with helium. Ultra pure (17.2 M) water was obtained from a Barnsted Model 04741 Nanopur e deionized water system (Barnsted/

PAGE 281

257 Thermodyne, Dubuque, IA). ChromPerfect (Version 3.5 for Wi ndows) computer software (Justice Laborator y Software, Denville, NJ was used for on-line collection, integration, and processing of the experimental data.

PAGE 282

258 Figure 5.1 Schematic of a gravity-fed sample dispensing unit for capillary microextraction with a sol-gel poly-THF coated capillary

PAGE 283

2595.2.2 Chemicals and materials Fused silica capillary (320 m i.d.) with a protective polyimide coating on the external surface was purchased from Polymicr o Technologies Inc. (Phoenix, AZ). PolyTHF 250 was a gift from BASF Corporation (Parsippany, NJ). Acenaphthene, fluorene, phenanthrene, fluoranthene, pyrene, n-nonanal, n-undecanal, n-dodecanal, n-tridecanal, valerophenone, hexanophenone, heptanophe none, decanophenone, 2,4-dichlorophenol, 2,4,6-trichlorophenol, 4-chloro-3-methyl pheno l, and pentachlorophenol were purchased from Aldrich Chemical Co. (Milwaukee, WI); n-decyl aldehyde, 1-nonanol, 1-decanol, 1-undecanol, and 1-tridecanol were purchase d from Acros Organics (Pittsburgh, PA). Lauryl alcohol was purchased from Sigma Ch emical Co. (St. Louis, MO). HPLC-grade methanol and methylene chloride and all bo rosilicate glass vials were purchased from Fisher Scientific (Pittsburgh, PA). 5.2.3 Preparation of sol-gel poly-THF coated microextraction capillaries Preparation of sol-gel Poly -THF coated microextracti on capillaries involves five distinct and sequential operat ions: (1) cleaning and pretr eatment of the fused silica capillary, (2) preparation of the sol solution, (3) coating the fused si lica capillary with the sol solution, (4) thermal conditi oning of the coated fused sili ca capillary, and (5) rinsing of the coating with organic solvents to remove unbonded ma terials (if there is any).

PAGE 284

2605.2.3.1 Cleaning and hydrothermal trea tment of the fused silica capillary The fused silica capillary inner surface must be cleaned before coating with the sol solution. In order to clean the surface from organic cont aminants, it was rinsed with methylene chloride and methanol. One mL of each solvent star ting with methylene chloride was passed through the capillary sequentially using the filling/purging device (Figure 5.2). The rinsed fuse d silica capillary was then pur ged with helium for 30 min followed by hydrothermal treatment. Hydrotherm al treatment generated adequate surface silanol groups required fo r strong chemical bonding betw een the substrate and the growing sol-gel network. To perform hydrothermal treatme nt, the cleaned fused silica capillary was filled with Nanopure deionize d water using the filling/purging device and then the water was flushed out of the capillary with the aid of helium gas pressure. The capillary was then purged with helium gas fo r 30 min so that only a thin layer of water remained on the surface of the capillary. At this point, both ends of the fused silica capillary were sealed with oxyacetylene flame. The sealed capillary was then heated at 250 C in a GC oven for 2 hours. Both ends we re then cut open w ith a ceramic wafer. One end of the fused silica capillary was furthe r connected to the GC injection port with the help of a graphite ferrule and was heated again in the GC oven at 250 C for another 2 more hours with a continuous helium flow th rough the capillary. Afte r this, the capillary was ready for coating.

PAGE 285

261 Figure 5.2 Schematic of a homemade capillary filling/purging device for the preparation of capillary microextraction capillaries and open-tubular sol-gel GC columns Gas flow Fused silica capillary Pressurization chamber Threaded chamber cap Control valve

PAGE 286

2625.2.3.2 Preparation of the sol solution A sol solution designed to prepare a hybrid sol-gel organi c-inorganic coating consisted of an alkoxide precu rsor, a sol-gel-active (either hydroxy or alkoxy terminated) organic polymer, one or more surface d eactivating reagent, appropriate organic solvent(s), and a sol-gel catalyst. Table 5.1 presents the name, function, and chemical structure of different ingredie nts used to prepare the coating solution to create sol-gel poly-THF coatings for CME. The sol solution was prepared by dissolving methyltrimethoxysilane (s ol-gel precursor, 250 L), hydroxy-terminated poly(tetrahydrofuran), (250 mg), hexamethly ldisilazane (surface deac tivation reagent, 20 L), and trifluoroacetic acid (containi ng 5% water) (sol-gel catalyst, 100 L) in methylene chloride (solvent, 400 L). After adding all the ingredients, the resultant solution was vortexed for 3 min for thorough mixing of the constituents followed by centrifugation for 5 min to remove any precipitates from the sol solution before it was used for coating.

PAGE 287

263 Table 5.1 Names, functions and chemical structur es of sol-gel poly-THF coating solution ingredients Name Function Structure Methyltrimethoxysilane (MTMOS) Polytetrahydrofuran Trifluoroacetic acid / water 95:5 (v/v) Methylene chloride Hexamethyldisilazane Sol-gel precursor Organic ligand Catalyst Solvent Deactivating reagent CH3O Si CH3OCH3OCH3 H O [(CH2)4 O ]nH CF3COOH CH2Cl2 H3C Si NH Si CH3 CH3CH3 CH3CH3

PAGE 288

2645.2.3.3 Coating fused silica capillary with sol solution In order to coat a fused silica capillary w ith the previously prepared sol solution, a 3 -m piece of hydrothermally treated fused silica capillary (320 m i.d.) was installed on the gas pressure-operated filling/purging device (Figure 5.2) that was used to fill the capillary with specially designed sol solution. After filling, the sol solution was kept inside the capillary for 60 min to facilitate the formati on of a surface-bonded sol-gel poly-THF coating. During this residence period of the sol solution inside the capillary, a sol-gel hybrid organic-inorgani c polymeric network evolved w ithin the sol solution, and part of it ultimately became bonded to the fused silica capillary inner surface via condensation reaction with surf ace silanol groups. The bonded part of the sol-gel material served as the CME coating. The free unbonde d portion of the sol solution was then expelled from the capillary under helium pressu re (50 psi) and the coated capillary was purged with helium for an hour. The con tinuous flow of helium facilitated the evaporation of volatile solvents. 5.2.3.4 Thermal conditioning of the coated capillary Thermal conditioning of the sol-gel coating was a very important step in postcoating processing because it (1 ) helped the sol-gel reaction to go to completion, (2) eased removal of entrapped solvents, and unreacted sol-gel ingredients, (3) ensured strong immobilization of the co ating on the surface, and (4) facilitated th e deactivation

PAGE 289

265 reactions on the fused silica su rface and/or the sol-gel coating matrix and thus helped tuning the net polarity as well as the selectivity of the resultant sorbent. The sol-gel coated capillary was thermally conditioned in a GC oven using temperature-programmed heating from 40 C to 300 C @ 1 C min-1, and holding the capil lary at the final temperature for 5 h under helium purge. 5.2.3.5 Rinsing the capillary with organ ic solvents to remove unbonded materials Before using in extraction, the coated cap illary was rinsed sequentially with 2 mL each of methylene chloride and methanol followed by drying in a stream of helium under the same temperature-program med conditions, as in secti on 4.2.3.4 except that the capillary was held at the final temperature for 30 min. The capillary was further cooled down to ambient temperature a nd cut into 13-cm long pieces that were further used to perform microextraction. 5.2.4 Preparation of sol-gel PDMS c oated capillary columns for GC A 10-m long piece of uncoated fused silica capillary (250 m i.d.) was accurately wrapped on a GC column basket in coils. Th e capillary was sequentially rinsed with 1 mL each of methylene chloride and metha nol followed by purging with helium for 30 min. The column was then hydrothermally treated following the procedure described in

PAGE 290

266 Section 5.3.2.1. The hydrothermally treate d column was installed on the filling/purging device (Figure 5.2) and the helium flow was adjusted to 40 psi. The coating solution for the sol-gel PD MS column was prepared as follows: 25 mg PDMS was dissolved in 900 L of methylene chloride. A 5 L volume of methyltrimethoxysilane (MTMOS), 25 L of polymethylhydrosiloxane (PMHS), and 50 L trifluoroacetic acid (TFA ) (5% water v/v) were sequentially added to the vial containing the polymer solution. In order to ensure proper mi xing of the ingredients, the sol solution was vortexed after adding each in gredient. The hydrolysis reaction began as soon as the sol-gel catalyst is added, a nd a 10 min period was allowed for hydrolysis reaction to proceed. At the end of this 10 min period, 20 L NH4F solution (20 mg/mL in methanol) was added to the sol solution to faci litate faster condensation. The resulting sol solution was centrifuged and the clear supernatant from the top was tr ansferred to a clean vial using a micro pipet for fu rther use in column coating. The sol solution was then introduced into the hydrothermally treated fused silica capillary using gas pressure (40 psi) in the filling/purging device (Figure 5.2). Once the capillary was filled with sol-gel solution, the filling /purging device was carefully depressurized and the free exit end of the capillary was sealed with a rubber septum while keeping the other end of the capillary inside filling/purging device. The sol solution was kept inside the capillary for 10 min. During this in-capillary residence time of the sol solution, a surface bonded hybrid organic-inorga nic film gradually evolved. After this, the sol solution was expelled from the fused si lica capillary under helium pressure. The column was then purged with helium for 60 min. The next step was the thermal conditioning of the coated column. For this, one end of the column was connected to the

PAGE 291

267 GC injection port using a gra phite ferrule and a continuous flow of helium (1 mL/min) was maintained through the column. The other end of the capillary was kept free inside the GC oven. The column was heated from 40 C to 150 C at a rate of 1C/min, maintained at 150 C for 5 h, then heated to 330 C at a rate of 5 C/min and maintained at this final temperature for 1 h. After the thermal conditioning, the column was rinsed with 2 mL of methylene chloride/methanol mixture (75:25 v/v), purge with helium for 1 h and then conditioned again from 40 C to 330 C at a rate 5 C/min, holding it at final temperature for 30 min. 5.2.5 Preparation of sol-gel PEG coated capillary column for GC A 10 m long (250 m i.d.) pi ece of fused silica capill ary was first cleaned and hydrothermally treated as described in sect ion 5.3.2.1. It was then installed on the filling/purging device to carry out the coating process. A sol-gel coating solution was prepared as follows: 35 mg of methoxypoly(ethylene glycol)-sila ne (PEG 1) and 15 mg of poly (ethylene glycol)-bis silane (PEG2) (sol-gel active organic ligan ds) were dissolved in 600 L of methylene chloride (solvent) contained in a polypropyl ene microcentrifuge vial. A Scientific Products model S8223 Vortex shaker aided the dissolution process (5 min). Then 5 L of MTMOS (precursor), 10 L of bis(trimethoxysily lethyl)benzene (precursor), and 5 L of HMDS (deactivating reagent) were sequentially added to the microcentrifuge vial and thoroughly mixed for 5 min to obtain a homogene ous solution. After this, 50 L of 95%

PAGE 292

268 TFA (acid catalyst containing 5% water) was added to the solution and thoroughly mixed. After 10 min, a 20-L volume of NH4F solution (20 mg/mL in methanol) was introduced into the vial. The volume of th e solution was made up to 1000 L by adding the required amount of methylene chloride, and the mixt ure was thoroughly vortexed. The resulting solution was centrifuged at 13 000 rpm (15,682g) for 5 min. The precipitate at the bottom of the vial, if any, was discarde d, and the top clear so l solution was used to fill the hydrothermally treated fused-silica cap illary using a helium pressure of 50 psi. After a set period of in-capillary residence time (10-20 min), the solution was expelled from the capillary under the same helium pre ssure and the capillary was subsequently purged with helium at 50 psi for an additional 60 min. This was followed by temperatureprogrammed heating in a GC oven from 40 to 150 C at 2 C min-1 with a hold time of 300 min at 150 C and then from 150 to 280 C at 6 C min-1, holding it at 280 C for 120 min. Keeping the temperature programming rate at 6 C min-1, the column was further conditioned in small steps, holdi ng the column for 120 min at each of the following final temperatures: 300, 320, and 340 C. The column was then rinsed with 2 mL of methylene chloride and conditione d again from 40 to 320 C at 6 C min-1. While conditioning, the column was purge d with helium at 1 mL min-1. 5.2.6 Cleaning and deactivation of glassware To avoid any contamination of the standa rd solutions from the glassware, all glassware used in the current study was t horoughly cleaned with Sparkleen detergent

PAGE 293

269 followed by rinsing with copious amount of de ionized water and drying at 150 C for 2 hours. To silanize the inner surface of the drie d glassware, they were treated with a 5% v/v solution of HMDS in methylene chloride followed by heating in an oven at 250 C for 8 hours under helium purge. The silanized glassware was then rinsed sequentially with methylene chloride and methanol and dr ied in an oven at 100 C for 1 hour. Prior to use, all glassware were rinsed with gener ous amounts of deionized water and dried at room temperature in a flow of helium. 5.2.7 Preparation of standard solutions for CME on sol-gel poly-THF coated capillaries All stock solutions, except those fo r polycyclic aromatic hydrocarbons, were prepared by dissolving 50 mg of each analyte in 5 mL of meth anol in a deactivated amber glass vial (10 mL) to obt ain a solution of 10 mg/mL. The polycyclic aromatic hydrocarbon solutions were prepared by dissolv ing 50 mg of each analyte in 5 mL of tetrahydrofuran. The solution was further dilu ted to 0.1 mg/mL in methanol. The final aqueous solution was prepared by further dilu ting this solution with water to achieve g/mL to ng/mL level concentrations depending on the compound class. Freshly prepared aqueous solutions were used for extraction.

PAGE 294

2705.2.8 Gravity-fed sample dispenser for capillary microextraction A gravity-fed sample dispenser was us ed for capillary micr oextraction (Figure 5.1). It was built by modifying a Chromaflex AQ column (Kontes Glass Co., Vineland, NJ), which consists of a thick-walled Pyre x glass cylinder concentr ically placed in an acrylic jacket. Since glass surfaces tend to ad sorb polar analytes, th e inner surface of the glass cylinder was deactivated by treating with HMDS solution as described before. The cylinder was then cooled down to ambient temperature, thor oughly rinsed with methanol and deionized water, and dried in a helium ga s flow. The system was then reassembled. 5.2.9 Extraction of analytes on solgel poly-THF coated capillaries A 12.5 cm long segment of the solgel poly-THF coated capillary (250 m i.d.) was conditioned under helium purge in a GC oven using a temperature program (from 40 C to 320C @ 10 C/min, held at the fina l temperature for 30 min). The conditioned capillary was then vertically connected to the lower end of the gravity-fed sample dispenser (Figure 1) using a plastic connector. A 50 mL volume of the aqueous sample containing trace concentrations of the target analytes was added to the inner glass cylinder through the sample inle t located at the top of the dispenser. The solution was passed through the capillary for 30 min to f acilitate the extraction equilibrium to be established. The capillary was then detached from the dispenser and purged with helium

PAGE 295

271 for 1 min to remove residual water from the capillary walls. 5.2.10 Transferring the extracted analytes to the GC column and gas chromatographic analysis of the extracted analytes The extracted analytes were transferred from the microextraction capillary to the GC column via thermal desorption. For this the CME capillary was first installed on the GC injection port and securedly interfaced with the GC capillary column. To facilitate the installation, both the GC injection po rt and the oven were cooled to 30 C, and the quartz wool was removed from the injection port glass liner. Th e CME capillary with the extracted analytes in the coating was then in troduced into the GC injection port from the bottom end of the port so that ~8 cm of the cap illary remained inside the injection port. A graphite ferrule was used to make an airtight connection between the capillary and the injection port. The lower end of the capi llary (residing inside the GC oven) was connected to the inlet end of GC capillary column using a deactivated press-fit quartz connector. Figure 5.3 illustrat es the connection of the extr action capillary with the GC analysis column using a pr ess-fit quartz connector.

PAGE 296

272 Figure 5.3 Schematic representation of the interf ace between the extrac tion capillary and the analysis column inside the GC oven using a press-fit quartz connector.

PAGE 297

273 Installation and interfaci ng of the extraction capillary with the GC column were followed by thermal desorption of extracted analytes from the installe d sol-gel poly-THF coated microextraction capillary. For this, the temper ature of the PTV inj ection port was rapidly raised from 30 C to 300 C @ 100 C /min while keeping the GC oven temperature at 30 C (5 min). Under these temp erature program conditions, th e extracted analytes were effectively desorbed from the sol-gel poly-TH F coating and were tran sported b the cooler coupling zone consisting of the lower end se gment of the microext raction capillary and /or to the front end of the GC column both located inside the GC oven and maintained at 30 C. As the desorbed analytes re ached the cooler interface zone (30 C), they were focused into a narrow band. On completion of the 5-min desorption and focusing period, the analytes in this narrow band were an alyzed by GC using temperature-programmed operation as follows: from 30C to 300C @ 20 C /min with a 10 min hold time at the final temperature. 5.3 Results and discussion 5.3.1 Sol-gel chemistry of the coating process Sol-gel chemistry is an elegant synthe tic pathway to advanced materials [8, 3941] that can be effectively utilized to create surface-bonded or ganic-inorganic hybrid coatings on the outer surface of conventional SPME fibers [6] as well as on the inner walls of a capillary for use in CME [7] (intube SPME). Additiona lly, sol-gel technology

PAGE 298

274 can be used for creating both th in [21] and thick [24] coa tings employing a wide variety of sol-gel-active organic ligands [21, 23-27]. Polytetrahydrofuran poly-THF [42] is a medium polarity polymer with terminal hydroxyl groups that can be utilized to bi nd this polymer to a sol-gel network via polycondensation reaction. It consists of tetramethylene oxide repeating units, and is synthesized through cationic ring opening pol ymerization of tetrahydrofuran (Scheme 5.1) using various initiators [43].

PAGE 299

275 H+H2OHO[(CH2) On Tetrahydrofuran Polytetrahydrofuran OH] n4 Scheme 5.1 Cationic ring opening polymerization of tetrahydrofuran

PAGE 300

276 Table 5.1 lists the chemical ingredients used in this work to prepare the sol solution for the fabrication of a so l-gel poly-THF coated capillaries. In order to create the so l-gel poly-THF coating in situ, the sol solution was kept inside the capillary for 60 min. During th is in-capillary residence time, an organicinorganic hybrid sol-gel networ k was evolving in the sol soluti on and a thin layer of this network located in close vicinity of the fuse d silica capillary walls had the opportunity to become chemically bonded to the capillary inner surface as a result of condensation reaction with the silanol groups on the capillary walls. The in situ creation of a highly stable, deactivated sol-gel coating involved the following processes: (1) catalytic hydrolysis of the alkoxide precursors, (2) pol ycondensation of the hydrolyzed precursor with other sol-gel-active co mponents of the sol solution, (3) chemical bonding of polyTHF to the evolving sol-gel network, (4) chemical anc horing of the evolving hybrid organic-inorganic polymer to the inner walls of the capillary, and (5) derivatization of residual silanol groups on the coating by using HMDS. Reacti ons leading to the creation of a surface-bonded sol-gel poly-THF coati ng are presented in schemes 5.2-5.6. Hydrolysis of the sol-gel precursor, methyltrimethoxysilane (MTMOS), in presence of the sol-gel catalyst (TFA) is presented in scheme 5.2. The hydrolysis products of the precurs or can then undergo polycondensation reactions in a variety of ways to create a th ree-dimensional sol-gel network. One possible route for this process is pr esented in scheme 5.3.

PAGE 301

277 Scheme 5.4 represents a simplified mode l for the polycondensa tion reaction of the growing sol-gel network with poly-THF to chemically inco rporate the polymer in the resultant organic-inorganic hybrid network structure. Scheme 5.5 represents the chemical anc horing process of the evolving network to the inner surface of the capillary. Hexamethyldisilazane, used in the coating solution, is capable of deactivating the residual silanol groups on the stationary pha se coating during thermal conditioning of the coated capillary as presented in scheme 5.6.

PAGE 302

278 Si OCH3OCH3CH3O+ 3 H2O CatalystTFA (5%H2OSi OHOHOH+ 3 C3OHMethyltrimethoxysilaneCH3CH3 H Scheme 5.2 Hydrolysis of the sol-gel precursor, MTMOS

PAGE 303

279 Si OHOHOHSi OHOHOHm+ n Si CH3 CH3CH3 (O Si O CH3 )pOHHydrolyzed MTMOSHO Scheme 5.3 Polycondensation of hydrolyzed MTMOS

PAGE 304

280 Si CH3 (O Si O CH3 )pOHy HO[( CH2) O] H Si CH3 (O Si O CH3 O Si OH CH3 )q[( CH2) O] O+Si O Si OHO HOHO 444n Scheme 5.4 Chemical incorporation of poly-THF into the sol-gel network

PAGE 305

281 OH OH OH OH OH O + Si CH3 (O Si O CH3 O Si OH CH3 )q[( CH2) O] O Si O Si OH O Si CH3 (O Si O CH3 O Si OH CH3 )q[( CH2) O] O Si O Si OH O HO OH OH Capillary inner surface Surface bonded sol-gel poly-THF coating Growing sol-gel poly-THF network4 4 n n Scheme 5.5 Chemical anchoring of the sol-gel poly-THF hybrid organicinorganic polymer to the silanol groups on the fused silica capillary inner walls

PAGE 306

282 OHOHOSi CH3 (O Si O CH3 O Si OH CH3 )q[( CH2) O] OSi O Si OHO + H3C Si CH3CH3NH Si CH3CH3CH3 OHOSi CH3 (O Si O CH3 O Si OH CH3 )q[( CH2) O] OSi O Si O O Si CH3H3COH OH CH3CH3O Si CH3 CH3 n4Capillary inner surface Surface bonded sol-gel poly-THF coating HexamethyldisilazaneCapillary inner surface Deactivated Surface bonded sol-gel poly-THF coating4n 5.6 Deactivation of residual silanol groups by derivatization with hexamethyldisilazane

PAGE 307

2835.3.2. Characterization of sol-gel poly-THF sorbent 5.3.2.1 Fourier-transform infrared spectroscopic characterization of the sol-gel poly-THF coating Fourier-transform infrared spectroscopic (FT-IR) measurements were performed in the range of 600-4000 cm-1 using a Avatar Model 320 FTIR system. Instrumental resolution was set at 4 cm-1. Figure 5.4 represents three IR spectr a representing pure poly-THF (left), sol solution having all ingredients except polyTHF (middle), sol-gel poly-THF coating (right). In the bottom spectrum, th e IR data contain the feature Si-O-C identified by 1045 cm-1 that confirms the successful bonding of polytetrahydrofuran to the hydrolyzed product of the alkoxide precursor [35]. The broad band at 3357 cm-1clearly demonstrates the presence of hydroxyl chai n ends in pure poly-THF. In the FT-IR spectrum for sol-gel poly-THF, similar band for OHgroups is seen to be present but in a lesser magnitude. This indicates that some of the OHgroups present in poly-THF have been used up in the process of condensati on reaction that occurre d to incorporate polyTHF moieties into the silica network.

PAGE 308

284 Figure 5.4 IR spectra of pure polytetrahydrofur an (left), sol solution having all ingredients except polytetra hydrofuran (middle), sol-gel polytetrahydrofuran coating (right).

PAGE 309

2855.3.2.2 Scanning electron microscopy (SEM) Surface morphology and the coating thic kness of sol-gel poly-THF coated microextraction capillaries were investig ated using scanning electron microscopy. Figure 5.5 (a, b) represents a scanning elec tron micrographs (SEM) of a sol-gel poly-THF coated capillary. From Figure 5.5a the coating thickness was estimated at 0.5 m. As can be seen from the image, sol-gel poly-THF coati ng is remarkably uniform in thickness.

PAGE 310

286 Figure 5.5a Scanning electron microscopic image of a 320 m i.d. sol-gel poly-THF coated fused silica capillary used in CME.

PAGE 311

287 Figure 5.5b Scanning electron microscopic image illu strating porous stru cture of sol-gel poly-THF coating in a 320 m i.d. sol-gel poly-THF coated fused silica capillary used in CME

PAGE 312

288 Figure 5.5b illustrates porous structure of the sol-gel poly-THF coating. It represents the surface view of the coating obtained at a magnification of 10,000x. It reveals the underlying porous structure of the sol-gel poly-THF coating. Due to the porous nature, the sol-gel poly-THF extracti on medium possesses enhanced surface area, an advantageous feature to achieve enhanced sample capacity. The porous structure also facilitates efficient mass tran sfer through the coating, whic h in turn, translates into reduced equilibrium time during extraction. 5.3.2.3 Thermogravimetric analysis Thermogravimetric analysis is a valuable tool for determining the thermal stability and composition of a material. Thermal stability is a measure of the capability of a material to preserve its properties upon heating. It is an important characteristic, which determines the ma ximum operating temperature of a given sorbent. The most common method of determin ing thermal stability of a sorbent is the non-isothermal TGA, in which the change in weight is determined while the temperature is increased in a linear manne r. The data obtained via this process is called a thermogram, which is a graph of mass change ve rsus temperature. Figure 5.6a represents a thermogram of pure pol y-THF. Degradation begins at ca. 320 C. Figure 5.6b represents a therm ogram of sol-gel poly-THF sorbent. As can be seen, the hybrid material demonstrated at least 100 C higher thermal stability compared to pure poly-THF. This was possible due to th e inherent positive attribute of sol-gel

PAGE 313

289processing technology. Figure 5.6a TGA curve of pure poly-THF for programmed heating (10 C/min) under N2 (Reproduced from Ref. [67] with permission of Elsevier)

PAGE 314

290 Figure 5.6b TGA curve of sol-gel poly-THF for programmed heating (10 C/min) under N2 [TGA 2950, TA Instruments, Inc. DE, USA] Sol-Gel Poly-THF-1012345678902004006008001000TemperatureWeight loss (% ) Series1

PAGE 315

2915.3.3 Determination of the extraction kinetics SPME is a non-exhaustive extraction technique. Quantitation by SPME is based on solute extraction equilibrium established between the sample solution and the coating. Therefore, the time required to reach the equi librium is particularly important. The time required for establishing equilibrium is de pendent on the partition coefficient of the analyte and on the agitation of the sample matrix. For volatile an alytes, the extraction equilibrium time is generally shorter in extract ion from headspace. In order to extract the maximum amount of analyte, the extraction equi librium time has to be reached. But this may be too long and impractical for many compounds. For instance, Doong et al. [63] reported that fluoranthene and pyrene require d 540 min to reach ex traction equilibrium on a commercial PDMS (30 m) fiber in dire ct extraction. The long equilibrium time of high molecular-weight PAHs is due to their low water solubilities and diffusion coefficients. The diffusivities of high mo lecular-weight PAHs in the aqueous solution range from 6.3 x 10-6 to 7.4 x 10-6 cm2/ s which can be even smaller (10-8 to 10-9 cm2/s) in polymeric materials [64, 65]. Figure 5.7 represents the ex traction kinetic profiles of different PAHs in directSPME using commercial PDMS (30 m) fiber [63].

PAGE 316

292 Figure 5.7 Extraction kinetic profiles of PAHs for direct-SPME using a commercial PDMS (30 m) fiber (Reproduced from Ref. [63] by permission from J. Chromatogr. Sci.)

PAGE 317

293 On the contrary, like other sol-gel sorbents sol-gel poly-THF coati ng possesses a porous structure which has been re vealed by the SEM images (Figure 5.4 b). Such porous structure should significantly increase the available surface area of the extracting phase. Due to the inherent porous structure, sol-gel poly-THF provides enhanced extracting phase loading and consequently high sample capacity. Lee and coworkers [68] reported that unlike SPME with polymer coated fiber, which involves a diffu sion-limited process, the kinetics of SPME with a porous layer ar e mainly controlled by mass transfer in the bulk solution. Therefore, the extraction rate depends greatly on the degree of agitation rather than on film thickness. In capillary microextraction, the aqueous solution containing the analytes is under continuous flow due to gravit y, and the extracting phase reaches the extraction equilibria within a shor t period of time. In the present study, solgel poly-THF coated microextraction capilla ry reached extraction equilibrium in a significantly shorter time even for high molecular-weight polycyclic aromatic hydrocarbons. For example, extraction equi libria were reached for two polycyclic aromatic hydrocarbons fluoranthen e and pyrene within 30 min. Figure 5.8 illustrates the CME kineti c profiles of two nonpolar analytes (fluoranthene and pyrene), two moderate ly polar analytes (heptanophenone and dodecanal) and a highly polar analyte (pent achlorophenol) extracted on a sol-gel polyTHF coated capillary. Extractions were carried out using aq ueous solutions of fluoranthene (10 ppb), pyrene (10 ppb), dodecanal (20 ppb), heptanophenone (20 ppb), and pentachlorophenol (50 ppb). As can be seen, both nonpolar, mode rately poloar, and highly polar compounds reached respective equilibr ia within 30 min. This is indicative of

PAGE 318

294 the fast mass transfer in the sol-gel poly -THF coating. Based on these experimental results, further experiments in this work we re carried out using a 30-min extraction time.

PAGE 319

295 Extraction Kinetics 0 20000 40000 60000 80000 100000 120000 140000 160000 1800000102030405060Time (min)Peak Area (arb. units) fluoranthene pyrene heptanophenone dodecanal pentachlorophenol Figure 5.8 Illustration of the extraction kine tics of nonpolar (f luoranthene and phenanthrene) and moderately polar (heptanophenone and dodecanal) compounds extracted on a 12.5 cm x 320 m i.d. sol-gel poly-THF coated capillary using 10 ppb aqueous solution of each analyte in a mixtur e. Extraction kinetics of the highly polar compound, pentachlorophenol, was obtai ned separately on a 12.5 cm x 320 m i.d. solgel poly-THF coated capillary using 50 ppb aqueous solution. Extraction conditions: Extraction time, 1050 min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C.

PAGE 320

2965.3.4 Determination of the analyte enrich ment factors of sol-gel poly-THF sorbent One of the major attributes of solid phase microextraction is its ability to preconcentrate the analyte(s) in the extracti ng phase, and therefore, it is important to measure the analyte(s) enrichment factor fo r a particular extracting phase in conjunction with a particular class of co mpounds. In order to compare rela tive extraction efficiency of different extracting phases, a commercial 100 m PDMS coated fiber, commercial 85 m PA coated fiber, and a sol-gel Poly-THF coated microextraction capillaries were employed. One L of 1 mg/mL solution of n-undecanol (containing 1 ng of the analytes) was injected directly into the GC injection port under splitless mode and the corresponding peak area was recorded. A 20 ppb aqueous solution of fluorene and 100 ppb aqueous solution of n-undecanol were used for direct SPME using SPME fibers and the sol-gel poly-THF microextraction capil lary. The volume of the sol-gel Poly-THF coating was calculated usi ng the equation 1. Vf = 2 rdfL Equation 5.1 Where, Vf = Volume of the sol-gel poly-THF extracting phase r = Radius of the extraction capillary L = Length of the extraction capillary df = Thickness of the coating The volume of the sol-gel Poly-THF ex traction phase (coating thickness ~ 0.5

PAGE 321

297 m) in a 12.5 cm capillary (320 m i.d.) has been calculated as 0.063 L. The sorbent volume in commercial 100 m PDMS (0. 660 L) and 85 m PA (0.500 L) has been found in the literature [66] Direct extraction was carri ed out using the commercial fibers and sol-gel PolyTHF coated microextraction cap illary for 30 min. The area obtained from each extraction (arbitrary unit) corresponded to the amount of analyte(s) present into the extracting phase. The concentration of analyte in the extracting phase was calculated using equation 2. C= Mass of the anal y te extractedintothesorbent(m g ) Volume of the sorbent ( mL) Eq. 5.2 Finally, the analyte enrichment fact or was calculated using equation 3. Enrichment Factor = Concentrationofanalyteintheextracting phase Original concentration of analyte in the sample matrix Eq. 5.3 The results obtained in these experiments ha ve been presented in Figure 5.9 and Table 5.2. As can be seen from the table, sol-ge l Poly-THF coated micr oextraction capillaries demonstrated excellent anal yte enrichment capability.

PAGE 322

298 0200040006000800010000120001400016000 PDMSPAPTHF Undecanol Figure 5.9 Comparison of extraction efficiencies of commercial PDMS (100 m), PA (85 m) and sol-gel poly-THF (0.5 m) coatings [Y-axis represents computer output for peak area in arbitrary unit]

PAGE 323

299Table 5.2 Analyte enrichment factor s of commercial PDMS (100 m), commercial PA (85 m) and sol-gel poly-THF sorbents for n-undecanol* Sorbent Phase volume (L) Concentration of n undecanol in the phase after preconcentration (mg/mL) Analyte enhancement factor for n undecanol PDMS (commercial), 100 m 0.660 0.0353 353 PA (commercial), 85 m 0.500 0.0916 915 Sol-gel poly-THF, 0.5 m 0.063 1.593 16,000 Fiber SPME (using 100 m PDMS and 85 m PA fibers)/ capillary microextraction (using sol-gel poly-THF coated capillary) GC analysis of fluorene (20 ppb) and nundecanol (100 ppb). Extraction time, 30 mi n. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 350 C (280 C for fiber SPME), at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C.

PAGE 324

3005.3.5 Determination of thermal stabilit y of the sol-gel poly-THF coating Thermal stability of an SPME sorbent ch aracterizes resistance to degradation when heated. High thermal stability of a SPM E sorbent is a highly desired attribute, particularly when it is used in combinati on with GC. As sol-gel process allows to incorporation organic moieti es into inorganic polymeri c network through chemical bonding, the resulting hybrid materials genera lly show higher thermal stability compared to pure organic polymers. This is no exception for sol-gel poly-THF. In order to determine the thermal stability of sol-gel poly-THF coating, microextraction capillaries were conditione d at 300 C, 320 C, 340 C, 350 C and 360 C for 2 h at each step. A mixture of anal ytes (comprising of both polar and nonpolar) were extracted for 30 min in three rep licates using sol-gel poly-THF coated microextraction capillary afte r each step of thermal conditi oning. The extraction results are presented in Table 5.3. As can be seen from Table 5.3, there is no noticeable change in extraction efficiency of the sol-gel poly-TH F coated capillary even after heating at 360 C. The high thermal stability of sol-ge l poly-THF coating as evidenced from the experimental data also corr oborates with the data obtai ned from thermogravimetric analysis

PAGE 325

301Table 5.3 Thermal stability of sol-gel poly-THF coated microextraction capillaries through GC peak areas of the solutes ex tracted after conditioning at different temperatures* Thermal conditioning temperature Compounds 300 C 320 C 340 C 350 C 360 C A x 10-5 Nonanal (100 ppb) 2,4,6-Trichlorophenol (400 ppb) Heptanophenone (40ppb) Phenanthrene (20 ppb) Pyrene (40 ppb) 1.26 1.26 3.19 1.74 2.96 1.15 1.29 3.03 1.76 3.04 1.13 1.29 3.13 1.79 2.98 1.15 1.24 3.19 1.81 3.06 1.06 1.21 3.48 1.90 3.18 ___________________________________________________ *Capillary Microextraction-GC analysis of nonanal (100 ppb), 2,4,6-trichlorophenol (400 ppb), heptanophenone (40 ppb), phenanthrene (2 0 ppb), and pyrene (40 ppb) using solgel poly-THF coated capillary. Extraction tim e, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injec tion; injector temperature, initial 30 C, final 300 C, 320 C, 340 C, 350 C and 360 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. A= Actual computer output for peak area in arbitrary unit.

PAGE 326

3025.3.5 Applications of sol-gel poly-THF coating Sol-gel poly-THF coated capillaries were used to extract analytes of environmental, biomedical, and ecological importance, including polycyclic aromatic hydrocarbons (PAHs), aldehydes, ketones, alcohols, and phenols. The extracted compounds were further analyzed by GC. 5.3.5.1 Polycyclic aromatic hydrocarbons PAHs are ubiquitous environmental pollu tants that present potential health hazards because of their toxic, mutagenic, and carcinogenic properties [44, 45]. Because of this, Environmental Protection Agency (EPA) has promulgated 16 unsubstituted PAHs in its list of 129 priority pol lutants [46a]. Table 5.4 list s pertinent physico-chemical properties of 5 EPA promulgated PAHs which were extracted and analyzed using sol-gel poly-THF coated capillaries and GC-FID. Cap illary microextraction results are presented in Table 5.5 and 5.6. Calculated limit of detections for each tested PAHs are presented in Table 5.7. Figure 5.10 shows the gas chroma togram representing CME-GC analysis of the listed unsubstituted polyaromatic hydrocar bons from EPA priority list. They were extracted from an aqueous solution (each at 10 ppb) by capillary microextraction using a sol-gel poly-THF coated capilla ry. As can be seen from the data presented in Table 5.5 and 5.6, run-to-run and capillary -to-capillary repeatability in peak area obtained in CMEGC-FID experiments was quite satisfactory. For all PAHs, the RSD values were under

PAGE 327

303 6%. Moreover, parts per quadrillion (ppq) leve l detection limits were obtained for PAHs in the CME-GC-FID using sol-gel poly-THF microextraction capillaries. These detection limits are significantly lower than those reported by others [46b] via SPME-GC-FID (e.g., 260 ppt for pyrene) using 100 m thick PDMS coated commercial SPME fiber.

PAGE 328

304Table 5.4 Chemical structures and pertinent physica l properties of polyaromatic hydrocarbons (PAHs) extracte d using sol-gel poly-THF coated capillaries. Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) Acenaphthene 152.20 92.5 150.0 0.8987 Fluorene 166.22 114.8 295.0 1.203 Phenanthrene 178.23 99.2 340 0.9820 Fluoranthene 202.26 107.8 384 1.252 Pyrene 202.26 156 404 1.271

PAGE 329

305Table 5.5 Run-to-run peak area reproducibility for PAHs in capillary microextraction using sol-gel poly-THF coating* A x 10-4 Compounds name Run 1 Run 2 Run 3 Mean RSD (%) Acenaphthene 12.47 11.93 13.19 12.53 5.04 Fluorene 11.18 11.35 10.76 11.01 2.75 Phenanthrene 14.46 13.67 13.72 13.95 3.16 Fluoranthene 14.01 13.35 13.48 13.63 2.74 Pyrene 9.57 9.38 9.51 9.49 1.07 _________________________________________________ Capillary Microextraction-GC analysis of PAHs (20 ppb each) using sol-gel poly-THF coated capillary. Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier ga s; FID temperature 350 C. Peaks: (1) Acenaphthene, (2) Fluorene, (3) Phenanth rene, (4) Fluoranthene, and (5) Pyrene. A = Actual computer output for peak area in arbitrary unit.

PAGE 330

306Table 5.6 Capillary-to-capillary peak area for PAHs in capillary microextraction using sol-gel poly-THF coated capillaries* A x 10-4 Compound Name Run 1 Run 2 Run 3 Mean RSD (%) Acenaphthene 13.92 13.84 13.38 13.71 2.13 Fluorene 12.09 12.03 11.50 11.88 2.73 Phenanthrene 14.72 14.07 15.26 14.69 4.02 Fluoranthene 13.92 15.63 13.83 14.46 7.05 Pyrene 9.18 9.50 8.29 8.99 6.96 _________________________________________________ *Capillary MicroextractionGC analysis of PAHs (20 ppb each) using sol-gel poly-THF coated capillary. Extraction time, 30 mi n. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) Acenaphthene, (2) Fluorene, (3) Phenan threne, (4) Fluoranthene, and (5) Pyrene. A=Actual computer output for p eak area in arbitrary unit.

PAGE 331

307Table 5.7 Limits of detection (LOD) for PAHs in CME-GC-FID using sol-gel poly-THF microextraction capillaries* Measured noise (V) : 0.975 Compounds Concentration (ppb) Peak height (V) (H x 10-4 ) Limit of detection (S/N 3), ppq Acenaphthene 20 3.05 625 Fluorene 20 2.93 460 Phenanthrene 20 3.27 400 Fluoranthene 20 3.94 260 Pyrene 20 2.47 750 _____________________________________________ Capillary Microextraction-GC analysis of PAHs (20 ppb each) using sol-gel poly-THF coated capillary. Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier ga s; FID temperature 350 C. Peaks: (1) Acenaphthene, (2) Fluorene, (3) Phenanth rene, (4) Fluoranthene, and (5) Pyrene. H= Actual computer output for peak height in V.

PAGE 332

308 Figure 5.10 Capillary Microextraction-GC analysis of PAHs (20 ppb each) using sol-gel poly-THF coated capillary. Ex traction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier ga s; FID temperature 350 C. Peaks: (1) Acenaphthene, (2) Fluorene, (3 ) Phenanthrene, (4) Fluoranthene, and (5) Pyrene.

PAGE 333

3095.3.5.2 Aldehydes Aldehydes and ketones (car bonyl compounds) are of incr easing concern due to their potential advers e health effects and environmen tal prevalence [47-49]. Aldehydes and ketones can form in water by the photode gradation of dissolved natural organic matter [50]. They may also form as disinfecti on by-products due to chemical reactions of chlorine and/or ozone (frequently used to disinfect water) with natural organic matter present in water [51]. Many of these by-produc ts have been shown to be carcinogens or carcinogen suspects [52]. This is, in part, due to the high polarity and reactivity of carbonyl compounds in wate r matrices [51,53,54]. Table 5.8 lists chemical structure and important physico-chemical properties of 5 e nvironmentally important aldehydes. Figure 10 represents a gas chromatogram of the mi xture of 5 underivatized aldehydes that were extracted from an aqueous solution contai ning 20 ppb of each analyte. Run-to-run and capillary-to-capillary microextraction data ha ve been presented in Tables 5.9 and 5.10, respectively. Low RSD values for both runto-run (ranges from 2.45% to 7.24 %) and capillary-to-capillary (ranges from 3.89 % to 9.31%) microextraction are definitely indicative of high system reproducibility of sol-gel poly-THF coated microextraction capillaries and the extraction method itself.

PAGE 334

310 Table 5.8 Chemical structures and pertinent physi cal properties of al dehydes extracted using sol-gel poly-THF coated capillaries Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) n-Undecanal n-Decanal n-Undecanal n-Dodecanol O O O O142.2406 156.2674 170.2942 184.321 63 7 12 93 at 23 mm Hg 207-209 109-115 at 5 mm Hg 185 at 100 mm Hg 0.823 0.825 0.827 0.829

PAGE 335

311 Table 5.9 Run-to-run peak area reproducibility for al dehydes in capillary microextraction using sol-gel poly-THF coating* A x 10-4 Compound Name Run 1 Run 2 Run 3 Mean RSD (%) n-Nonanal 7.71 7.78 8.08 7.86 2.46 n-Decanal 9.02 10.28 10.24 9.84 7.24 n-Undecanal 7.11 6.67 6.53 6.77 4.46 n-Dodecanal 5.31 4.79 5.38 5.16 6.23 ____________________________________________________ *Capillary microextractionGC analysis of aldehydes at 20 ppb concentration using poly-THF coated capillary. Ex traction time, 30 min. GC an alysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) n-Nonanal, (2) n-Decanal, (3) n-Undecanal, (4) n-Dodecanal. A= Actual computer output for peak area in arbitrary unit.

PAGE 336

312Table 5.10 Capillary-to-capillary peak area reproduc ibility for aldehydes in capillary microextraction using solgel poly-THF coating* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Mean RSD (%) n-Nonanal 8.32 7.71 8.13 8.06 3.89 n-Decanal 9.92 10.76 10.03 10.24 4.47 n-Undecanal 8.01 7.12 7.85 7.66 6.25 n-Dodecanal 6.53 5.53 6.53 6.20 9.31 _____________________________________________ *Capillary microextraction-GC analysis of aldehydes at 20 ppb concentration using polyTHF coated capillary. Extrac tion time, 30 min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) n-Nonanal, (2) n-Decanal, (3) n-Undecanal, (4) n-Dodecanal. A= Actual computer output for p eak area in arbitrary units.

PAGE 337

313Table 5.11 Limits of detection (LOD) for aldeh ydes in CME-GC-FID using sol-gel polyTHF microextraction capillaries* Measured noise (V) : 0.681 Compounds Concentration (ppb) Peak height (V) (H x 10-4) Limit of detection (S/N 3), ppq n-Nonanal 20 1.93 1000 n-Decanal 20 2.43 625 n-Undecanal 20 1.86 750 n-Dodecanal 20 1.42 940 ____________________________________________________ *Capillary microextractionGC analysis of aldehydes at 20 ppb concentration using poly-THF coated capillary. Ex traction time, 30 min. GC an alysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) n-Nonanal, (2) n-Decanal, (3) n-Undecanal, (4) n-Dodecanal. H= Actual computer output fo r peak height in V.

PAGE 338

314 Figure 5.11 Capillary microextraction-GC analysis of aldehydes in water at 20 ppb concentration using poly-THF coated capilla ry. Extraction time, 30 min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) n-Nonanal (2) n-Decanal, (3) nUndecanal and (4) n-Dodecanal.

PAGE 339

315 The data presented in Table 5.11 indicate that a sol-gel poly-TH F coated capillary can extract free aldehydes from aqueous medi a to provide a limit of detection (LOD) which is comparable with, or lower than that achieved through derivatization [53]. For example, LOD for decanal has been reported as 200 ppt [53] (in SPME-GC-ECD) on a 65 m DVB-PDMS coating afte r derivatization with o(2 ,3,4,5,6-pentafluorobenzyl) hydroxylamine hydrochloride (PFBHA) whereas in the present work a significantly lower detection limit (625 ppq) was achieved for th e same analyte using a sol-gel poly-THF coated capillary in hyphenation with GC-F ID, even though ECD often provides higher sensitivity than FID for oxygenated compounds. The same trend has also been observed for other analytes. It should be pointed out th at derivatization of these analytes, especially when they are present in trac e concentration, may complicate the analytical process, thus compromising quantitative accuracy. 5.3.5.3. Ketones Figure 5.12 represents a gas chromatogr am of a mixture of 5 underivatized ketones (20 ppb each) extracted from an aqueous solution. The selected ketones are listed in Table 5.12 along with their chemical structures and pertinent physico-chemical properties. Excellent peak shapes Figure 5.12 and run-to-run (Table 5.13) and capillary-to-capillary ex traction reproducibility Table 5.14 are indicative of preserved separation efficiency in CME-GC analysis and versatility of the sol-gel coating procedure used to prepare the extraction capi llaries and the used GC column.

PAGE 340

316 Table 5.12 Chemical structures and pertinent phys ical properties of ketones extracted using sol-gel poly-THF coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) Butyrophenone O 148.2042 11-13 220-222 0.988 Valerophenone O 162.231 -9 105-107 at 5 mm Hg 0.988 Hexanophenone O 176.2578 25-26 265.1 0.958 Heptanophenone O 190.2846 17 155 at 15 mm Hg 0.946 Decanophenone O 232.37 36-39 168 at 5 mm Hg

PAGE 341

317Table 5.13 Run-to-run peak area reproducibility fo r ketones in capillar y microextraction using sol-gel poly-THF coating* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Mean RSD (%) Butyrophenone 10.81 11.14 11.27 11.07 2.12 Valerophenone 10.25 10.79 10.85 10.63 3.09 Hexanophenone 12.64 12.54 11.00 12.06 7.62 Heptanophenone 11.96 12.51 12.98 12.48 4.12 Decanophenone 7.87 7.51 8.47 7.95 6.06 ___________________________________________________ *Capillary microextraction-GC analysis of ketones at (20 ppb) using sol-gel poly-THF coated capillary. Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, program rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) Butyrophenone, (2) Valerophenone, (3 ) Hexanophenone, (4) Heptanophenone, and (5) Decanophenone. A= Actual computer output for peak area in arbitrary unit.

PAGE 342

318Table 5.14 Capillary-to-capillary peak area repr oducibility for ketones in capillary microextraction using solgel poly-THF coating* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Mean RSD (%) Butyrophenone 11.23 11.74 12.10 11.69 3.73 Valerophenone 11.93 12.03 12.51 12.16 2.55 Hexanophenone 15.01 15.88 14.80 15.23 3.773 Heptanophenone 16.66 15.35 15.49 15.83 4.53 Decanophenone 1.24 10.77 10.93 11.37 8.07 ______________________________________________ *Capillary microextractionGC analysis of ketones at (20 ppb) using sol-gel poly-THF coated capillary. Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, program rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) Butyrophenone, (2) Valerophenone, (3 ) Hexanophenone, (4) Heptanophenone, and (5) Decanophenone. A= Actual computer output for peak area in arbitrary unit.

PAGE 343

319 Table 5.15 Limits of detection (LOD) for ketone s in CME-GC-FID using sol-gel polyTHF microextraction capillaries* Measured noise (V) : 0.853 Compounds Concentration (ppb) Peak height (V) (H x 10-4 ) Limit of detection (S/N 3), ppq Butyrophenone 20 2.91 1000 Valerophenone 20 3.00 460 Hexanophenone 20 3.47 600 Heptanophenone 20 3.05 340 Decanophenone 20 2.03 1000 ___________________________________________ *Capillary microextraction-GC analysis of ketones at (20 ppb) using sol-gel poly-THF coated capillary. Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, program rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) Butyrophenone, (2) Valerophenone, (3 ) Hexanophenone, (4) Heptanophenone, and (5) Decanophenone H= Actual computer output fo r peak height in V.

PAGE 344

320 Figure 5.12 Capillary microextraction-GC analysis of ketones at (20 ppb) using sol-gel poly-THF coated capillary. Ex traction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless inj ection; injector temperature, initial 30 C, final 300 C, program rate of 100 C/min; GC oven temperat ure programmed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) Butyrophenone, (2) Va lerophenone, (3) Hexanophenone, (4) Heptanophenone, and (5) Decanophenone.

PAGE 345

3215.3.5.4. Chlorophenols Chlorophenols (CPs) represent an important class of contaminants in environmental waters and soils due to their widespread use in industry, agriculture, and domestic purposes. Chlorophenols have been widely used as preservatives, pesticides, antiseptics, and di sinfectants [55]. They are also used in pr oducing dyes, plastics and pharmaceuticals. In the environment, chlor ophenols may also form as a result of hydrolysis, oxidation and microbiological degradation of chlorinated pesticides. Chlorine-treated drinking wa ter is another source of ch lorophenols [56]. As a result, chlorophenols are often found in waters [ 57, 58], soils [59], and sediments [59]. Chlorophenols are highly toxic, poorly biodegradable, carcinogenic and recal citrant [60]. Owing to their carcinogenicity a nd considerable persistence, fi ve of the chlorophenols (2chlorophenol; 2,4-dichlorophenol; 2,4,6-trichlorophenol; 4-chloro-3-methylphenol and pentachlorophenol) have been classified as priority pollutants by the US EPA [61]. Since chlorophenols are highly polar, it is quite difficult to extract them directly from polar aqueous media. Derivatization, pH adjustme nt, and/or salting-out are often used to facilitate the extraction [3]. To reduce the an alytical complexity due to derivatization, HPLC-UV is frequently used for the analys is of phenolic compounds [58] but at the expense of detection sensitivity. Figure 13 represents CME-GC analysis of five underivatized chlorophenols extracted from an aqueous medium using a sol-gel polyTHF coated capillary. We did not have to use derivatization, pH adju stment or salting out effect to extract chlorophenols from aqueous medium. Still, we have achieved a lower detection limit e.g., 18 ppt for pentachlorophenol, by CME-GC-FID compared to other reports in the literature 1.4 ppb for the same compound, by SPME-GC-FID [3].

PAGE 346

322 Table 5.16 Chemical structures and pertinent phys ical properties of chlorophenols (CPs) analyzed using sol-gel poly-THF coating Name of the analyte Chemical structure Molecular weight (g/mol) Melting point (C) Boiling point (C) Density (g/mL) 2-Chlorophenol OH Cl 128.5579 7 125.6 1.241 2,4-Dichlorophenol OH Cl Cl 163.003 45 210 1.383 2,4,6Trichlorophenol OH Cl Cl Cl 197.448 69.5 244.5 1.49 4-Chloro-3-methyl phenol OH Cl CH3 142.584 67 235 Pentachlorophenol OH Cl Cl Cl Cl Cl 266.3383 174 310 1.979

PAGE 347

323Table 5.17 Run-to-run peak area reproducibility for chlorophenols in capillary microextraction using sol-gel poly-THF Coating* A x 10-3 Compounds Name Run 1 Run 2 Run 3 Mean RSD (%) 2-Chlorophenol 7.53 6.60 7.71 7.28 7.32 2,4-Dichlorophenol 11.13 12.14 10.92 11.30 5.71 2,4,6Trichlorophenol 14.40 13.71 13.36 13.82 3.83 4-Chloro-3methylphenol 16.51 17.00 17.29 16.93 2.35 Pentachlorophenol 33.01 33.54 31.10 32.55 3.94 _______________________________________________ *Capillary microextraction-GC analysis of chlorophenols using poly-THF coated capillary. Extractions were carried out fr om a solution containing 2-chlorophenol (1 ppm); 2,4-dichlorophenol (50 ppb); 2,4,6-trichlorophenol ( 50 ppb); 4-chloro, 3methylphenol (100 ppb); and pentachloropheno l (50 ppb). Extraction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature pr ogrammed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 2-Chlorophenol, (2) 2,4Dichlorophenol, (3) 2,4,6-Tr ichlorophenol, (4) 4-Chloro -3-methylphenol, and (5) Pentachlorophenol. A= Actual computer output for p eak area in arbitrary unit.

PAGE 348

324Table 5.18 Capillary-to-capillary peak area reproducib ility for chlorophenols in capillary microextraction using sol-gel poly-THF coating* A x 10-3 Compounds name Run 1 Run 2 Run 3 Mean RSD (%) 2-Chlorophenol 4.95 4.22 4.43 4.53 8.74 2,4-Dichlorophenol 8.84 8.74 8.21 8.60 3.96 2,4,6Trichlorophenol 10.10 9.62 11.10 10.27 7.02 4-Chloro-3methylphenol 14.40 13.52 13.27 13.73 4.50 Pentachlorophenol 27.83 28.38 29.56 28.38 3.22 __________________________________________________ *Capillary microextraction-GC analysis of chlorophenols using poly-THF coated capillary. Extractions were carried out fr om a solution containing 2-chlorophenol (1 ppm); 2,4-dichlorophenol (50 ppb); 2,4,6-trichlorophenol ( 50 ppb); 4-chloro-3methylphenol (100 ppb); and pentachloropheno l (50 ppb). Extraction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature pr ogrammed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 2-Chlorophenol, (2) 2,4Dichlorophenol, (3) 2,4,6-Tr ichlorophenol, (4) 4-Chloro -3-methylphenol, and (5) Pentachlorophenol. A= Actual computer output for p eak area in arbitrary unit.

PAGE 349

325Table 5.19 Limits of detection (LOD) for chlor ophenols in CME-GC-FID using sol-gel poly-THF microextra ction capillaries* Measured noise (V) : 0.681 Compounds Concentration (ppb) Measured peak height (V) ( H x 10-3) Limit of detection (S/N 3), ppt 2-Chlorophenol 1 000 3.93 150 2,4Dichlorophenol 50 5.23 85 2,4,6Trichlorophenol 50 5.45 81 4-Chloro-3methylphenol 100 7.35 30 Pentachlorophenol 50 11.34 18 ___________________________________________________ *Capillary microextraction-GC analysis of chlorophenols using poly-THF coated capillary. Extractions were carried out fr om a solution containing 2-chlorophenol (1 ppm); 2,4-dichlorophenol (50 ppb); 2,4,6-trichlorophenol ( 50 ppb); 4-chloro, 3methylphenol (100 ppb); and pentachloropheno l (50 ppb). Extraction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature pr ogrammed from 30 C (hold for 5 min) to 300 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 2-Chlorophenol, (2) 2,4Dichlorophenol, (3) 2,4,6-Tr ichlorophenol, (4) 4-Chloro -3-methylphenol, and (5) Pentachlorophenol. H= Actual computer output fo r peak height in V.

PAGE 350

326 Figure 5.13 Capillary microextraction-GC analysis of chlorophenols using poly-THF coated capillary. Extractions were carried ou t from a solution cont aining 2-chlorophenol (1 ppm); 2,4-dichlorophenol (50 ppb); 2, 4,6-trichlorophenol (50 ppb); 4-chloro, 3methylphenol (100 ppb); and pentachloropheno l (50 ppb). Extraction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PDMS column; splitless injection; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature pr ogrammed from 30 C (hold for 5 min) to 300 C, at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 2-Chlorophenol, (2) 2,4Dichlorophenol, (3) 2,4,6-Tr ichlorophenol, (4) 4-Chloro -3-methylphenol, and (5) Pentachlorophenol.

PAGE 351

3275.3.5.4 Alcohols Figure 5.14 represents a gas chromatogr am for a mixture of alcohols. Being highly polar compounds, alcohol s demonstrate higher affin ity for water and were extracted from aqueous samples using sol-gel poly-THF capillaries without exploiting any derivatization, pH adjustment or salt ing-out effects. Runto-run and capillary-tocapillary microextraction data are presented in Table 5.20 and 5.21 respectively. Limit of detection data are pres ented in Table 5.22.

PAGE 352

328 Table 5.20 Run-to-run peak area reproducibility fo r alcohols in capillar y microextraction using sol-gel poly-THF coating* A x 10-4 Compounds name Run 1 Run 2 Run 3 Mean RSD (%) 1-Heptanol 3.83 4.01 4.33 4.06 6.34 1-Octanol 8.06 8.31 8.06 8.12 2.02 1-Nonanol 9.90 9.87 9.87 9.74 2.57 1-Decanol 13.78 13.81 13.22 13.61 2.44 1-Undecanol 17.33 16.81 16.03 16.73 3.90 1-Dodecanol 14.93 14.21 13.78 14.31 4.08 1-Tridecanol 22.58 21.74 20.75 21.69 4.22 ___________________________________________________ *Capillary microextraction-GC analysis of alcohols (100 ppb each) using sol-gel polyTHF coated capillary. Extrac tion time, 30min. GC analys is conditions: 10 m x 250 m i.d. sol-gel PEG column; splitless injec tion; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 280 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 1-Heptanol, (2) 1-Octanol, (3 ) 1-Nonanol, (4) 1-Decanol, (5) 1-Undecanol, (6) 1-Dodecanol, and (7) 1-Tridecanol. A= Actual computer output for peak area in arbitrary unit.

PAGE 353

329Table 5.21 Capillary-to-capillary peak area reproduc ibility for alcohols in capillary microextraction using solgel poly-THF coating* A x 10-4 Compounds Name Run 1 Run 2 Run 3 Mean RSD (%) Heptanol 2.95 3.42 3.73 3.36 11.77 Octanol 7.01 7.04 6.71 6.92 2.62 Nonanol 8.42 8.31 8.53 8.42 1.30 Decanol 11.29 12.33 12.13 11.92 4.65 Undecanol 14.83 16.05 15.95 15.68 4.71 Dodecanol 13.86 13.26 14.97 14.03 6.17 Tridecanol 18.92 17.54 19.82 18.76 6.12 _______________________________________________ *Capillary microextraction-GC analysis of alcohols (100 ppb each) using sol-gel polyTHF coated capillary. Extrac tion time, 30min. GC analys is conditions: 10 m x 250 m i.d. sol-gel PEG column; splitless injec tion; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 280 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 1-Heptanol, (2) 1-Octanol, (3 ) 1-Nonanol, (4) 1-Decanol, (5) 1-Undecanol, (6) 1-Dodecanol, and (7) 1-Tridecanol. A= Actual computer output for peak area in arbitrary unit.

PAGE 354

330Table 5.22 Limits of detection (LOD) for alcohols in CME-GC-FID using sol-gel polyTHF microextraction capillaries* Measured noise (V) : 0.975 Compounds Concentration (ppb) Measured peak height (V) (H x 10-3) Limit of detection (S/N 3), ppt 1-Heptanol 100 17.14 13 1-Octanol 100 37.16 5 1-Nonanol 100 42.51 0.75 1-Decanol 100 60.63 0.61 1-Undecanol 100 80.32 0.59 1-Dodecanol 100 73.92 1.15 1-Tridecanol 100 76.94 1.15 _____________________________________________________ *Capillary microextraction-GC analysis of alcohols (100 ppb each) using sol-gel polyTHF coated capillary. Extrac tion time, 30min. GC analys is conditions: 10 m x 250 m i.d. sol-gel PEG column; splitless injec tion; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 280 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 1-Heptanol, (2) 1-Octanol, (3 ) 1-Nonanol, (4) 1-Decanol, (5) 1-Undecanol, (6) 1-Dodecanol, and (7) 1-Tridecanol. H= Actual computer output for peak height in V.

PAGE 355

331 Figure 5.14 Capillary microextraction-GC analys is of alcohols (100 ppb each) using solgel poly-THF coated capillary. Extraction time, 30min. GC analysis conditions: 10 m x 250 m i.d. sol-gel PEG column; splitless inject ion; injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 280 C at a rate of 20 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 1-Heptanol, (2) 1-Oc tanol, (3) 1-Nonanol, (4) 1-Decanol, (5) 1-Undecanol, (6) 1-Dodecanol, and (7) 1-Tridecanol.

PAGE 356

332 The presented data indicate excellent affinity of the sol-gel poly-THF coating for these highly polar analytes that are often difficult to extract from aqueous media in underivatized form using commercial coatings Moreover, high detection sensitivity (Table 5.22) and excellent symmetrical p eak shapes also demonstrate outstanding performance of the sol-gel poly-THF coating an d excellent deactivati on characteristics of the sol-gel PEG column used fo r GC analysis, respectively. 5.3.5.6 Mixture of polar and moderately polar and nonpolar compounds Finally, a mixture containing analytes from different chemical classes representing a wide polarity range was extract ed from an aqueous sample using a sol-gel poly-THF coated capillary. As is revealed fr om the chromatogram (Figure 5.15), a sol-gel poly-THF coated capillary can be effectively used to simultaneously extract nonpolar, moderately polar, and highly nonpolar compounds from an aqueous matrix.

PAGE 357

333 Figure 5.15 Capillary microextraction-GC analysis of a mixture of nonpolar, moderately polar and highly polar compounds using poly-THF coated capi llary. Extractions were carried out from an aqueous sample c ontaining 2-chlorophenol (1 ppm); 2,4,6trichlorophenol (50 ppb); pentachlorophenol (50 ppb) ; valerophenone (10 ppb); hexanophenone (10 ppb); nonanal (10 ppb); de canal (10 ppb); fluoranthene (10 ppb); pyrene (10 ppb). Extraction time, 30min GC analysis conditions: 10 m x 250 m i.d. solgel PDMS column; splitless injection (desorpti on of analyte in splitless mode); injector temperature, initial 30 C, final 300 C, programmed at a rate of 100 C/min; GC oven temperature programmed from 30 C (hold for 5 min) to 300 C at a rate of 15 C/min; helium carrier gas; FID temperature 350 C. Peaks: (1) 2-Chlorophenol, (2) Nonanal, (3) Decanal, (4) 2,4,6-Trichlorophenol, (5) Valerophenone, (6) Hexanophenone, (7) Pentachlorophenol, (8) Fluoran thene, and (9) Pyrene.

PAGE 358

334 This may be explained by the existen ce of different polarity domains [62] (organic and inorganic) in the sol-gel poly-THF coating. Run-to-run repeatability and capillary-to-capillary reproducibility are two important characteristics for CME as a micr oextraction technique and for the sol-gel coating technique used for their preparati on. These parameters were evaluated from experimental data involving replicate measurements carried out on the same capillary under the same set of conditions (run-to-run) or on a number of sol-gel poly-THF coated capillaries prepared using the same protoc ol (capillary-to-capillary). The run-to-run repeatability and capillary -to-capillary reproducibili ty for sol-gel capillary microextraction were evaluated through p eak area relative standa rd deviation (RSD) values for the extracted analytes. For nonpol ar and moderately polar analytes (e.g., PAHs, aldehydes, ketones), these parameters had values in the range of 2.19-7.48% and 4.35-10.31, respectively. In the case of polar analytes (phenols and alcohols), these values were less than 7.32% and 11.77 %, respectively. For a sample preparation technique, these peak area RSD values can be regarded as indicat ive of good consistency in CME performance of the microextr action capillaries as well as the good reproducibility in the method us ed for their preparation. 5.3.6. Possibility of automation In the present work, sol-gel CME-GC operation was performed manually which is not convenient from a practical point of view For wide acceptance of the technique, the inconvenience associated with manual installati on of the microextraction capillary in the

PAGE 359

335 GC system needs to be overcome. There are various possibilities to solve this problem, including the use of a robotic arm equipped with devices necessary for performing CME, desorbing the analytes, and tran sferring the desorbed analytes into the separation column. In our opinion, sol-gel capillary microe xtraction technique described in the present dissertation has a great potential fo r automated operation in hyphenation with both gas-phase and liquid-phase separation tech niques. Because of th e tubular format of the extraction device combined with high th ermal and solvent stability of the surfacebonded sol-gel extraction coating, sol-gel capillary microextraction can be expected to offer high degree of versatility in automated operation. 5.4 Conclusion Novel sol-gel poly-THF coating was de veloped for high-performance capillary microextraction to facilitate ultra-trace analysis of polar and nonpolar organic compounds. Parts per quadrillion level detect ion limits were achieved using Poly-THF coated microextraction capillaries in conjunction with GC-FID. To the best of our knowledge, we are the first to report [70] on the use of sol-gel poly-THF sorbent in analytical microextraction. Sol-gel Poly-T HF coatings showed extraordinarily high sorption efficiency for both polar and nonpolar compounds, and proved to be highly effective in providing simultane ous extraction of nonpolar, mode rately polar, and highly polar analytes from aqueous media. Sol-gel poly-THF coated microe xtraction capillaries showed excellent thermal and solvent stability, making them very suitable for hyphenation with both gas-phase and liquid-phase separation techniques, including GC,

PAGE 360

336 HPLC, and CEC. In CME-H PLC and CME-CEC hyphenations, sol-gel poly-THF coated microextraction capillaries ha ve the potential to provide new levels of detection sensitivity in liquid-phase trace analysis, a nd to extend the analytical scope of CME to thermally labile-, high molecu lar weight-, and other type s of compounds that are not amenable to GC. Further sensitivity enhancement should be possibl e through the use of monolithic microextraction capillaries w ith sol-gel poly-THF based hybrid organicinorganic sorbents. This could open up new possi bilities in ultra-trace analysis of organic pollutants in aqueous media.

PAGE 361

337 5.5 References for Chapter Five [1] J. Pawliszyn, Solid-Phase Microextraction. Th eory and Practice, Wiley, New York, 1997. [2] R. Eisert, J. Pawliszyn, Anal. Chem. 69 (1997) 3140. [3] K.D. Buchholz, J. Pawliszyn, Anal. Chem. 66 (1994) 160. [4] D. Louch, S. Motlagh, J. Paw liszyn, Anal. Chem. 64 (1992) 1187. [5] L.G. Blomberg, J. Microcolumn Sep. 2 (1990) 62. [6] S.-L. Chong, D.-X. Wang, J.D. Hayes, B.W. Wilhite, A. Malik, Anal. Chem. 69 (1997) 3889. [7] S. Bigham, J. Medlar, A. Kabir, C. Shende, A. Alli, A. Malik, Anal. Chem. 74 (2002) 752. [8] D.-X. Wang, S.-L. Chong, A. Malik, Anal. Chem. 69 (1997) 4566. [9] A. Malik, S.-L. Chong, In J. Pawliszyn (Ed.), Applications of Solid Phase Microextraction. Royal So ciety of Chemistry (RSC), Cambridge (UK), 1999, 73. [10] C.J. Brinker, G.W. Scherer. Sol-gel Sc ience. The Physics and Chemistry of Solgel Processing. Academic Press, San Diego, CA, 1990. [11] L.C. Klein. Sol-gel Technology for Thin Films, Fibers, Preforms, Electronics, and Specialty Shapes. Noyes Public ations, Park Ridge, NJ, 1988. [12] C.L. Arthur, J. Pawliszy n, Anal. Chem. 62 (1990) 2145.

PAGE 362

338 [13] D.W. Potter, J. Pawliszyn, Environ. Sci. Technol. 28 (1994) 298. [14] C.L. Arthur, L.M. Killam, K.D. Buchholz, J. Pawliszyn, J.R. Berg, Anal. Chem. 64 (1992) 1960. [15] J. Wu, J. Pawliszyn, J. Chromatogr. A 909 (2001) 37. [16] H. Kataoka, K. Mitani, Jan. J. Forensic Toxicol. 20 (2002) 251. [17] H.S. Lee, J. Hong, J. Chromatogr. A 868 (2000) 189. [18] B.J. Hall, J.S. Brodbelt, J. Chromatogr. A 777 (1997) 275. [19] O.E. Mills, A.J. Broome, ACS Symp. Ser. 705 (1998) 85. [20] V. Mani, In J. Pawliszyn (Ed.), Applications of Solid-phase Microextraction. Royal Society of Chemistry (RSC), Cambridge (UK), 1999, p. 57. [21] T.P. Gbatu, K.L. Sutton, J.A. Caruso, Anal. Chim. Acta 402 (1999) 67. [22] Z.P. Zhou, Z.Y. Wang, C.Y. Wu, W. Zhan, Y. Xu, Anal. Lett. 32 (1999) 1675. [23] Z. Wang, C. Xiao, C. Wu, H. Han, J. Chromatogr. A 893 (2000) 157. [24] Z. Zeng, W. Qiu, Z. Huang, Anal. Chem. 73 (2001) 2429. [25] Z. Zeng, W. Qiu, M. Yang, X. Wei, Z. Huang, F. Li, J. Chromatogr. A 934 (2001) 51. [26] M. Yang, Z.R. Zeng, W.L. Qiu, Y.L. Wang, Chromatographia 56 (2002) 73. [27] J. Yu, L. Dong, C. Wu, L. Wu, J. Xing, J. Chromatogr. A 978 (2002) 37. [28] S.L. Chong, M.S. Thesis, University of South Florida, 1997. [29] Manufacturer data sheet. Supelco corp., Bellefonte, PA, 2003, p. 360. [30] H. Goda, C.W. Frank, Chem. Mater. 13 (2001) 2783. [31] A. Fidalgo, L.M. Ilharco, J. Non-Cryst. Solids 283 (2001) 144. [32] C.S. Betrabet, G.L. Wilkes, Chem. Mater. 7 (1995) 535.

PAGE 363

339 [33] T. Higuchi, K. Kurumada, S. Nagami ne, A.W. Lothongkum, M. Tanigaki, J. Materials Science 35 (2000) 3237. [34] A. Fidalgo, T.G. Nunes, L.M. Ilharco, J. Sol-Gel Sci. Technol. 19 (2000) 403. [35] A. Fidalgo, L. Ilharco, J. So l-Gel Sci. Technol. 13 (1998) 433. [36] M. Kamitakahara, M. Kawashita, N. Miyata, T. Kokubo, T. Nakamura, Biomaterials 24 (2003) 1357. [37] I. Honma, O. Nishikawa, T. Sugimoto S. Nomura, H. Nakajima, Fuel Cells 2 (2002) 52. [38] J.D. Hayes, A. Malik, J. Chromatogr. B 695 (1997) 3. [39] C. Shende, A. Kabir, E. Townsend, A. Malik, Anal. Chem. 75 (2003) 3518. [40] B. M. Novak, Adv. Mater. 5 (1993) 422. [41] J. Livage, In J.F. Harrod, R.M. Lain e (Eds.), Applications of Organometallic Chemistry in the Preparation and Pro cessing of Advanced Materials. Kluwer, Dordrecht, The Netherlands, 1995, p. 3. [42] P. Dreyfuss, M.P. Dreyfuss, G. Pruckmayr, Encyclopedia of Polymer Science and Engineering. Wiley, New York, 16 (1989) 649. [43] F. Cataldo, Eur. Po lym. J. 32 (1996) 1297. [44] A. Byorseth, T. Ramdahl, (Eds.), Handbook of Polycyclic Aromatic Hydrocarbons. Vol. 2, Emission, Sources, and Recent Progress in Analytical Chemistry. Marcel Dekker, New York, Basel, 1985, p. 1. [45] K.J. Hageman, L. Mazeas, C.B. Graban ski, D.J. Miller, S.B. Hawthorne, Anal. Chem. 68 (1996) 3892. [46] a. Office of the Federal Registration (OFR ), Appendix A: prior ity pollutants, Fed Reg. 47 (1982) 52309. b. R.A. Doong, S. M. Chang, Y.C. Sun, J. Chromatogr. A 879 (2000) 177. [47] M.O. Amdur, In M.O. Amdur, J. Doull, C.D. Klaassen (Eds.), Air Pollutanta. In Casarett and Doulls Toxicology: The Basic Science of Poisons. 4th ed., Pergamon Press, New York, 1991, Chapter 25, p. 866.

PAGE 364

340 [48] National Research Council. Formal dehyde and other Aldehydes: Board on Toxicology and Environmental Health Hazards. National Academy Press, Washington DC, 1981. [49] U.S. Congress. Compilation of Acts w ithin the jurisdiction of the Committee on Energy and Commerce. U.S. Government Printing Office, Washington DC, 1991. [50] R.J. Kieber, K. Mopper, Envir on. Sci. Technol. 24 (1990) 1477. [51] M.-L. Bao, F. Pantani, O. Griffini, D. Burrini, D. Santianni, K. Barbieri, J. Chromatogr. A 809 (1998) 75. [52] J. Nawrocki, I. Kalkowska, A. Dabr owska, J. Chromatogr. A 749 (1996) 157. [53] B. Cancho, F. Ventura, M.T. Gal ceran, J. Chromatogr. A 943 (2002) 1. [54] E.E. Stashenko, M.A. Puertas, J.R. Martinez, Anal. Bioanal. Chem. 373 (2002) 70. [55] V.H. Kitunen, R.J. Valo, M.S. Salkai noja-Salonen, Envir on. Sci. Technol. 21 (1987) 96. [56] R.C.C. Wegman, A.W.M. Hofst ee, Water Res. 13 (1979) 651. [57] D. Puig, D. Barcelo, Tre nds Anal. Chem. 15 (1996) 362. [58] M. Moder, S. Schrader, U. Frank, P. Popp, Fresenius J. Anal. Chem. 357 (1997) 326. [59] M.-R. Lee, Y.-C. Yeh, W.-S. Hsiang, B.-H. Hwan, J. Chromatogr. A 806 (1998) 317. [60] G.L. Puma, P.L. Yue, Ind. Eng. Chem. Res. 38 (1999) 3238. [61] EPA 822-Z-99-001, U.S. Environmental Protection Agency, Office of Water, Washington, D.C., 1999. [62] A.B. Brennan, G.L. Wilkes, Polymer 32 (1991) 733. [63] R. Doong, S. Chang, Y. Sun, J. Chromatogr. Sci. 38 (2000) 528.

PAGE 365

341 [64] R.P. Schwarzenbach, P.M. Gschwend, D.M. Imboden, Environmental Organic Chemistry, John Wiley & Sons, Inc., New York, NY, 1993. [65] M.L. Chang, S. C. Wu, C.Y. Chen Environ. Sci. T echnol. 31 (1997) 2307. [66] J. Poerschmann, J. Microcol. Sep. 12 (2000) 603. [67] L. Costa, M.P. Luda, G.G. Cameron, M.Y. Qureshi, Polymer Degradation and Stability 67 (2000) 527. [68] Y. Liu, Y. Shen, M. Lee, Anal. Chem. 69 (1997) 190. [69] M. Azenha, C. Malheiro, A. F. Silva, J. Chromatogr. A 1069 (2005) 163. [70] A. Kabir, C. Hamlet, A. Malik, J. Chromatogr. A 1047 (2004) 1.

PAGE 366

342 Appendices

PAGE 367

343Appendix A

PAGE 368

344Appendix A (Continued)

PAGE 369

345Appendix A (Continued)

PAGE 370

346

PAGE 371

347Appendix A (Continued)

PAGE 372

348Appendix A (Continued)

PAGE 373

349Appendix A (Continued)

PAGE 374

350Appendix A (Continued)

PAGE 375

351Appendix B

PAGE 376

352 Appendix B (Continued)

PAGE 377

353 Appendix B (Continued)

PAGE 378

354Appendix B (Continued)

PAGE 379

355Appendix B (Continued)

PAGE 380

356Appendix B (Continued)

PAGE 381

357Appendix B (Continued)

PAGE 382

358Appendix B (Continued)

PAGE 383

359Appendix B (Continued)

PAGE 384

360Appendix B (Continued)

PAGE 385

361Appendix B (Continued)

PAGE 386

362Appendix B (Continued)

PAGE 387

363Appendix B (Continued)

PAGE 388

364Appendix B (Continued)

PAGE 389

365Appendix C

PAGE 390

366

PAGE 391

367Appendix C (Continued)

PAGE 392

368Appendix C (Continued)

PAGE 393

369Appendix C (Continued)

PAGE 394

370Appendix C (Continued)

PAGE 395

371Appendix C (Continued)

PAGE 396

372Appendix C (Continued)

PAGE 397

373Appendix C (Continued)

PAGE 398

374Appendix C (Continued)

PAGE 399

375Appendix D

PAGE 400

376Appendix D (Continued)

PAGE 401

377Appendix D (Continued)

PAGE 402

378Appendix D (Continued)

PAGE 403

379Appendix D (Continued)

PAGE 404

380Appendix D (Continued)

PAGE 405

381Appendix D (Continued)

PAGE 406

382Appendix D (Continued)

PAGE 407

383Appendix D (Continued)

PAGE 408

384Appendix D (Continued)

PAGE 409

385Appendix D (Continued)

PAGE 410

386Appendix D (Continued)

PAGE 411

387Appendix D (Continued)

PAGE 412

388Appendix E

PAGE 413

389Appendix E (Continued)

PAGE 414

390Appendix E (Continued)

PAGE 415

391Appendix E (Continued)

PAGE 416

392Appendix E (Continued)

PAGE 417

393Appendix E (Continued)

PAGE 418

394Appendix E (Continued)

PAGE 419

395Appendix E (Continued)

PAGE 420

396Appendix E (Continued)

PAGE 421

397Appendix E (Continued)

PAGE 422

398Appendix F

PAGE 423

399Appendix F (Continued)

PAGE 424

400Appendix F (Continued)

PAGE 425

401Appendix F (Continued)

PAGE 426

402Appendix F (Continued)

PAGE 427

403Appendix F (Continued)

PAGE 428

404Appendix F (Continued)

PAGE 429

405Appendix F (Continued)

PAGE 430

406Appendix F (Continued)

PAGE 431

407Appendix F (Continued)

PAGE 432

408Appendix F (Continued)

PAGE 433

409Appendix F (Continued)

PAGE 434

410Appendix F (Continued)

PAGE 435

411 Appendix G

PAGE 436

412Appendix G (Continued)

PAGE 437

413Appendix G (Continued)

PAGE 438

414Appendix G (Continued)

PAGE 439

415Appendix G (Continued)

PAGE 440

416Appendix G (Continued)

PAGE 441

417Appendix G (Continued)

PAGE 442

418Appendix G (Continued)

PAGE 443

419Appendix G (Continued)

PAGE 444

420Appendix G (Continued)

PAGE 445

421Appendix H

PAGE 446

422Appendix H (Continued)

PAGE 447

423Appendix H (Continued)

PAGE 448

424Appendix H (Continued)

PAGE 449

425 Appendix H (Continued)

PAGE 450

426ABOUT THE AUTHOR Abuzar Kabir was born in Jessore, a small city in Bangladesh where he received his elementary, secondary and higher secondary education. In 1984, he moved to Dhaka, the country’s capital, to pursue higher studie s. He was admitted into the University of Dhaka where he received his B.Sc. (Honors) and M.Sc. degrees in Applied Chemistry and Chemical Technology in 1990 and 1991, respectively. In 1993, he started his professional career in the Department of Applied Chemistry & Chemical Technology, University of Dhaka, Banglades h as a lecturer and was promot ed to assistant professor in 1996. He came to the United States in 1999 and joined the Department of Chemistry and Biochemistry, University of North Carolina at Greensboro, USA as an M.S. student. On the first day of the new millennium he came to Tampa to join the Department of Chemistry, University of South Florida (USF) where he continued his higher education as a Ph.D. student. He joined Dr. Abdul Malik’s research group and star ted his investigation on developing novel sol-gel organic-inor ganic hybrid material systems for chromatographic separation and analytical samp le preparation. His research at USF has resulted in 8 publications in international journals and 3 US pate nt applications.