Ion Mobility Spectrometry, Second Edition

Página 1 de 1 Cover http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Cover.html 05/05/2007 Pág...

Author: G.A. Eiceman | Z. Karpas

36 downloads 1375 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Página 1 de 1

Cover

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Cover.html 05/05/2007

Página 1 de 2

Page i

ION MOBILITY SPECTROMETRY Second Edition

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_i.ht... 05/05/2007

Página 1 de 2

Page iii

ION MOBILITY SPECTROMETRY Second Edition

G.A.Eiceman New Mexico State University Las Cruces, New Mexico Z.Karpas Nuclear Research Center Beer-Sheva, Israel

Boca Raton London New York Singapore A CRC title, part of the Taylor & Francis imprint, a member of the Taylor & Francis Group, the academic division of T&F Informa plc.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_iii....

05/05/2007

Página 1 de 3

Page iv Published in 2005 by CRC Press Taylor & Francis Group 6000 Broken Sound Parkway NW, Suite 300 Boca Raton, FL 33487–2742 © 2005 by Taylor & Francis Group, LLC CRC Press is an imprint of Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2005. To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk. No claim to original U.S. Government works ISBN 0-203-50475-5 Master e-book ISBN

ISBN 0-203-61617-0 (OEB Format) International Standard Book Number-10: 0-8493-2247-2 (Print Edition) (Hardcover) International Standard Book Number-13: 978-0-8493-2247-1 (Print Edition) (Hardcover) Library of Congress Card Number 2005041956 This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or for the consequences of their use. No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC) 222 Rosewood Drive, Danvers, MA 01923, 978–750–8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy license by the CCC, a separate system of payment has been arranged. Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data Eiceman, Gary Alan. Ion mobility spectrometry/Gary A.Eiceman, Zeev Karpas.—2nd ed. p. cm. Includes bibliographical references and index. ISBN 0-8493-2247-2 (alk. paper) 1. Ion mobility spectroscopy. I. Karpas, Zeev. II. TItle. QD96.P62E33 2005 543′.5–dc22 2005041956

Taylor & Francis Group is the Academic Division of T&F Informa plc. Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_iv....

05/05/2007

Página 1 de 1

Page v

Preface Since the publication of our first monograph, Ion Mobility Spectrometry, over a decade ago, a transformation has occurred in the visibiliiy, utilization, and fundamental understanding of this technique. In this respect, the subject of IMS is virtually unrecognizable from that of a decade ago, necessitating a revised and expanded monograph to cover a combination of multiple advances in technology, computational capabilities, models of the underlying gas-phase ion chemistry, and applications, most notably the large demand for detectors of explosives, toxic chemicals, and pathogenic microorganisms with the rise of international terrorism. A major technical innovation in this area was undoubtedly the introduction of devices in which combined high and low electric fields are used to distinguish between ions with different mobility values. Miniaturization of drift tubes was a direct result of this and other technical developments. Ion mobility sensors based on the deflection of ions in a fixed electric field also appeared on the market. Alternate ionization sources were developed to replace the reliable, standard radioactive source in order to avoid tedious regulatory issues. The combination of mobility spectrometers with chromatographic methods became routine and commercially available, expanding the range of applications and improving the performance of both. During the past decade, advances in computational capabilities dramatically improved the acquisition and treatment of data and the interpretation and presentation of the analytical findings. An improved understanding of the fundamentals of gas-phase ion chemistry underlying the response of mobility spectrometers provided resources to manufacturers and users seeking to avoid some of the problems that plagued earlier instruments and limited the acceptance of IMS. In addition to the traditional areas of contraband (drugs and explosives) detection and monitoring of chemical warfare agents, several completely new and exciting applications for IMS have been explored in medical, biological, environmental, and industrial areas. A striking feature of today’s commercial environment is the number of instrument manufacturers who have released a wide variety of IMSbased instruments from pocket sized to human-scanning portals. The present monograph brings the reader current with all these innovations and reflects the rapidly expanding applications of IMS. In Part I, the history, theory, and basic principles of chemistry are described. The history and evolution of the technology from the late 19th century to the present are described and expanded in Chapter 1. The theory of ion motion in an electric field is given in Chapter 2 with an additional section on the effect

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_v.h... 05/05/2007

Página 1 de 2

Page vi of high electric fields on ion motion. In Chapter 3, the chemistry and physics that govern response in IMS are described. In Part II of the book, the technology of IMS is provided in detail and includes a thorough review of sample introduction methods, drift tubes of various kinds, and modern methods of data analysis and display (Chapter 4). In Chapter 5, hyphenated techniques in which an IMS is combined with a chromatographic inlet or a mass spectrometer detector are discussed. In Part III, the applications of IMS technology are described. In Chapter 6, descriptions are given for the traditional forensic, military, and counterterrorism applications such as the detection of explosives, drugs, chemical agents, and other forensic applications. Whole new areas in which IMS is used in medical and biological applications, as well as developments in the detection of microorganisms and biological agents, are given in Chapter 7. Chapter 8 and Chapter 9 are devoted to industrial, environmental, and other specialized applications of IMS technology. The accompanying CD contains several features or topics that could not be properly included in the book. Among these are a bibliography, tables and mobility data, spectral libraries, some PowerPoint presentations, and detailed treatments of theory. A special feature is the addition of instructions and blueprints that will enable the readers to build their own IMS drift tubes and electronics. The book is meant to serve the needs of specialists in the field who are interested in details of recent developments, as well as researchers, engineers, and students who want a comprehensive yet concise overview of this technology. For those with an interest in IMS and for scientists in related areas, we planned this to be a useful introduction and readable guide to IMS. Gary A.Eiceman Zeev Karpas

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_vi....

05/05/2007

Página 1 de 1

Page vii

The Authors Gary A.Eiceman, Ph.D. is a professor in the Department of Chemistry and Biochemistry at New Mexico State University in Las Cruces. He obtained a B.S. in chemistry from West Chester State College (now West Chester University) in West Chester, Pennsylvania, 1974 and his Ph.D. from the University of Colorado in Boulder in 1978 with Professor H.F.Walton. This was followed by a postdoctoral fellowship with Professor F.W.Karasek at the University of Waterloo in Ontario, Canada, from 1978 to 1980. Since joining the faculty at New Mexico State University in 1980, Dr. Eiceman has been a senior research Fellow at the U.S. Army Chemical Research, Development, and Engineering Center at Aberdeen Proving Ground, Maryland (1987 to 1988), and a senior Fellow with the National Research Council (1992). He has been on sabbatical leaves at the University of Manchester Institute of Science and Technology with Dr. C.L.P.Thomas (1995) and at the Institute of Spectrometry and with Dr. Jorg Baumbach in Dortmund, Germany (2003). Dr. Eiceman has presented over 170 lectures and has authored or co-authored over 150 research articles, chapters, or reviews. His current research interests include the development of ion mobility spectrometry as a field analyzer, the chemistry of ion molecule reactions at ambient pressure, differential mobility spectrometry, and the environmental-analytical chemistry of hazardous organic compounds. He has served on peer review committees for the DOE, EPA, NIOSH, and NASA. He is on the editorial board of Talanta and is founding member of the International Society for Ion Mobility Spectrometry, organizing the first conference in 1992. A consultant for over a dozen agencies or companies, Dr. Eiceman regularly teaches at the undergraduate and graduate levels in quantitative analysis, separation sciences, and chemical instrumentation, and has received the Westhafter Award and University Research Council Award from New Mexico State University in 2004. He lives with his wife Mary and daughter Abigail in Las Cruces. Zeev Karpas, Ph.D. received his B.Sc. and M.Sc. degrees from the Hebrew University, Jerusalem and his Ph.D. from the Weizmann Institute of Science, Rehovot, Israel in 1976. He then spent two years as a post-doctoral Research Fellow at the California Institute of Technology and at the Jet Propulsion Laboratory in Pasadena, California. Upon his return to Israel, Dr. Karpas joined the staff of the Nuclear Research Center, Negev, as a research scientist and eventually became the head of the analytical chemistry department (1989 to 1992). From 1984 to 1985 he spent a sabbatical at the National Bureau of Standards (now the National Institute of Science and Technology) in Gaithersburg, Maryland. He spent another sabbatical (1992 to 1993) at New Mexico State University that was

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_vii.... 05/05/2007

Página 1 de 2

Page viii devoted to studying atmospheric-pressure ionization processes and co-authoring, with Dr. Gary Eiceman, the monograph entitled Ion Mobiliiy Spectrometry, published by CRC Press in 1993. Since 1994, Dr. Karpas has been involved in trace-analysis by ICPMS and radio-toxicological research. He is the co-founder of Q-Scent Ltd., an Israeli company that is concerned with the development of IMS applications for medical diagnostics. Dr. Karpas has been interested in the field of gas-phase ion chemistry since his graduate studies, and particularly in the technology and science of ion mobility spectrometry and its applications.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_viii... 05/05/2007

Página 1 de 1

Page ix

Contents Part I History, Theory, and Basic Principles of Ion Mobility Spectrometry 1 Introduction to Ion Mobility Spectrometry 1.1 Background 1.1.1 Definition of Ion Mobility Spectrometry 1.1.2 Description of Processes in a Mobility Spectrometer 1.1.3 The Formation of Gaseous Ions in Positive Polarity 1.1.4 The Formation of Gaseous Ions in Negative Polarity 1.1.5 The Separation of Ions and the Determination of Mobility 1.2 Studies of Ions in Gases at Atmospheric Pressure 1.2.1 Period of Discovery and Innovation 1.2.2 Period of Foundational Studies (1948 to 1970) 1.2.2.1 Renewed Interest in Ions in Gases at Elevated and Ambient Pressure 1.2.2.2 Drift Tubes for Ion Characterization by Mobility in Weak Electric Fields 1.3 Early Developments of IMS as an Analytical Method (1970 to 1990) 1.3.1 F.W.Karasek and Studies Using IMS for Chemical Analyses 1.3.2 Development of IMS by Military and Security Organizations 1.3.3 Fast-Responding and Pneumatically Sealed Drift Tubes 1.4 Modern Analytical IMS 1.4.1 Chemical Warfare Agents Detection 1.4.2 Explosives Detection 1.4.3 Drug Detection 1.4.4 Ion Mobility Spectrometry—The Book 1.4.5 Field Asymmetric IMS 1.4.6 The IMS Society and International Conferences 1.4.7 The Volatile Organic Analyzer 1.4.8 Biological Applications of Mobility Spectrometry 1.5 Present and Future Trends in IMS 1.5.1 Spectral Libraries and Standardization of Mobility Spectrometers 1.5.2 Instrumentation

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_ix....

3 3 3 3 5 7 7 9 9 12 14 14 19 19 21 25 25 26 26 26 27 27 27 28 29 29 29 30

05/05/2007

Página 1 de 1

Page x 1.5.3 Commercial Scene 1.5.4 Applications References 2 Mobility of Ions in the Gas Phase 2.1 Introduction 2.2 Motion of Slow Ions in Gases 2.2.1 Diffusion of Gaseous Ions 2.2.2 Effect of Electric Field on Ion Motion 2.2.3 Effect of Gas Density 2.3 Models for Ion-Neutral Interactions 2.3.1 Mobility Equations 2.3.2 The Rigid Sphere Model 2.3.3 The Polarization Limit Model 2.3.4 The 12,4 Hard-Core Potential Model 2.4 Models and Experimental Evidence 2.4.1 Introduction 2.4.2 Ion Radii in Homologous Series 2.4.3 Experimental Mobilities in Homologous Series and the Mass-Mobility Correlation 2.4.4 Experimental Mobilities in Homologous Series and the TemperatureMobility Correlation 2.4.4.1 Experimental Results 2.4.4.2 The Effect of Temperature and Drift Gas 2.4.4.3 The Overall Effect of Temperature and Drift Gas on Ko 2.4.5 A Simple Gedanken Experiment 2.4.6 Resolving Peaks by Changing the Composition of the Drift Gas 2.5 Dependence of Mobility on Electric Field References Appendix A Sensitivity of Calculated Mobilities toward the Choice of Parameters A.1 Introduction A.1.1 Choice of a* A.1.2 Choice of r0 and z A.1.3 Choice of n A.1.4 Summary of Sensitivity Analysis 3 Gas-Phase Ion Chemistry in Mobility Spectrometers 3.1 Introduction and General Considerations 3.2 Ion Chemistry at Ambient Pressure 3.2.1 Formation of Reactant Ions 3.2.1.1 Positive-Ion Formation

30 33 33 39 39 40 40 41 42 43 44 45 46 47 48 48 48 50 52 53 53 55 58 59 60 66 69 69 70 73 75 75 79 79 80 80 80

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_x.h... 05/05/2007

Página 1 de 1

Page xi 3.2.1.2 Negative-Ion Formation 3.2.2 Formation of Product Ions 3.2.2.1 Positive-Ion Reactions 3.2.2.2 Negative-Ion Reactions 3.3 Analytical Facets of Gas-Phase Ion Reactions 3.3.1 Quantitative Aspects to Response 3.3.1.1 Effect of Sample Concentration on Response 3.3.1.2 Detection Limits 3.3.1.3 Repeatability, Stability, and Linear Range 3.3.2 Effects of Experimental Parameters on Mobility Spectra 3.3.2.1 Influence of Moisture and Temperature 3.3.2.2 Reagent Gases and Alternate Reactant Ions 3.4 The Interpretation of Mobility Spectra 3.4.1 An Integrated Model 3.4.1.1 Long-Lived Ions 3.4.1.2 Short-Lived Ions 3.4.1.3 Ions with Intermediate Lifetimes 3.4.2 Response to Mixtures 3.4.3 Use of Mobility Spectra for Chemical Identification 3.5 Summary References Part II Technology of Ion Mobility Spectrometry 4 Drift Tubes for Mobility Spectrometers 4.1 Introduction 4.2 Inlets and Introduction of Sample 4.2.1 General Considerations 4.2.2 Reagent Gases 4.2.3 Gases, Vapors, and Ambient Air 4.2.3.1 Membrane-Based Inlets 4.2.3.2 Exponential and Dynamic Dilution 4.2.3.3 Preconcentration of Analytes from Ambient Air Samples 4.2.3.4 Reactive Gases 4.2.4 Liquid Samples 4.2.4.1 Spray and Electrospray Ionization 4.2.4.2 Solid-Phase Micro-Extraction (SPME) 4.2.4.3 Semipermeable Membranes 4.2.5 Solid Samples 4.2.5.1 Thermal Vaporization 4.2.5.2 Vaporization by Laser Heating or Ablation 4.3 Ion Sources 4.3.1 Radioactivity: Nickel, Americium, and Tritium

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_xi....

84 88 88 90 92 92 92 96 98 102 102 104 108 108 109 110 110 110 112 112 113 119 119 121 121 122 125 126 127 129 130 131 131 132 133 133 133 134 136 137

05/05/2007

Página 1 de 1

Page xii 4.3.2 Corona Discharges 4.3.3 Photoionization: Discharge Lamps and Lasers 4.3.4 Surface Ionization Source 4.3.5 Electrospray Ionization (ESI) 4.3.6 MALDI 4.3.7 Flames 4.4 Drift Tubes 4.4.1 Traditional Drift Tubes with Linear Electric Field 4.4.1.1 Design and Structure 4.4.1.2 Gas Flow 4.4.1.3 Electric Field Gradients 4.4.1.4 Ion Shutters 4.4.2 High-Field Asymmetric or Differential Mobility Spectrometers 4.4.2.1 Cylindrical FAIMS 4.4.2.2 Planar DMS 4.4.3 Alternative Designs for Drift Tubes 4.4.3.1 Aspirator Design 4.4.3.2 Planar Traditional Drift Tube 4.4.3.3 Others 4.4.4 Miniaturized Drift Tubes 4.5.1 Detection Devices and Methods 4.5.1.1 Detection Methods 4.5.2 Signal Acquisition and Processing 4.5.3 Analysis of Spectra 4.5.4 Modes of Signal Display 4.6 Selection of Materials 4.6.1 Conducting Materials 4.6.2 Insulating Materials 4.6.3 Miscellaneous Materials 4.7 Summary References 5 Hyphenated Methods with Mobility Spectrometers 5.1 Introduction to Hyphenated Ion Mobility Spectrometry (IMS) Methods 5.2 GC/IMS 5.2.1 Background 5.2.2 Analytical Information 5.2.2.1 Continuous Scanning 5.2.2.2 Reconstructed Ion Chromatograms and Ion Monitoring 5.2.3 Features of Response 5.2.3.1 Control of Selectivity by Reagent Gas or Method of Ionization 5.2.3.2 Quantitative Response

138 140 141 141 142 143 144 144 145 147 149 151 155 157 158 160 160 161 163 163 164 164 165 166 167 169 170 171 172 173 173 185 185 186 186 194 194 197 201 202 205

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_xii.... 05/05/2007

Página 1 de 1

Page xiii 5.2.4 Column Considerations 5.2.4.1 High-Resolution Capillary Columns 5.2.4.2 Multi-Capillary Columns (MCC) 5.2.5 Instruments 5.2.5.1 Environmental Vapor Monitor (EVM) 5.2.5.2 Volatic Organic Analyzer (VOA) 5.2.5.3 GC-IONSCAN 5.2.5.4 Varian Micro Differential Mobility Detector 5.2.5.5 U.S. Army Pyrolysis-GC/IMS 5.3 LC/IMS 5.3.1 A Brief History 5.3.2 Recent Developments 5.4 IMS/MS 5.4.1 Background 5.4.2 Interfaces between IMS Drift Tubes and MS Analyzers 5.4.3 Analytical Information 5.4.3.1 Acquisition of a Mobility Spectrum by IMS/MS 5.4.3.2 Acquisition of a Mass Spectrum by IMS/MS 5.4.3.3 Mass Spectrum for Each Mobility Peak 5.4.3.4 Tuned Ion Mobility Spectrum 5.4.4 IMS/MS Instruments 5.4.4.1 Traditional IMS/MS with a Quadrupole Mass Analyzer 5.4.4.2 Low-Pressure IMS/Quadrupole MS 5.4.4.3 IMS/TOF MS and IMS/Ion Trap/TOF MS 5.4.4.4 Field-Dependent Mobility Analyzers/Mass Spectrometers 5.5 Summary References Part III Applications of Ion Mobility Spectrometry 6 Forensics, Military, Security, and Counterterrorism 6.1 Introduction 6.2 Chemical Weapons 6.3 Detection of Explosives by IMS 6.3.1 General Comments on Detection of Explosives 6.3.2 Measurement with Handheld Devices, Portable Instruments, and Portals 6.3.3 Research and Operational Experience 6.3.3.1 Walk-Through Portals and Systems for Luggage Screening 6.3.3.2 Homemade and Alternate Explosives 6.3.3.3 Database for Explosives

208 208 209 209 210 210 212 214 216 218 218 218 219 219 222 224 224 224 225 226 227 227 228 229 230 232 234 243 243 244 249 249 253 254 257 258 259

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_xiii... 05/05/2007

Página 1 de 1

Page xiv 6.4 Drugs 6.4.1 Introduction and Ion Chemistry 6.4.2 Laboratory and Field Applications 6.4.3 Database for Drugs 6.5 Other Forensic Applications 6.5.1 Lachrymators 6.5.2 Arson 6.5.3 Security of Public Areas 6.6 Conclusions References 7 Biological, Biomolecular Research, and Medical Uses of IMS 7.1 Introduction 7.2 Medical Diagnostics Using IMS 7.2.1 Respired Air as a Measure of Exposure to Anesthetic Gases and of Lung Disease 7.2.2 Diagnosis of Vaginal Infections 7.3 Food Freshness and Odor Detection 7.4 Proteins, Peptides, Amino Acids, and Other Large Biomolecules and Biopolymers 7.4.1 Conformation Studies 7.4.2 Alkali Ions of Biomolecules 7.4.3 Further Studies of Biocompounds 7.5 Detection and Determination of Bacteria 7.5.1 Pyrolysis GC/IMS Methods 7.5.2 Enzyme-Based Immunoassay IMS 7.6 Conclusion References 8 Developed Applications in Industrial and Environmental Monitoring 8.1 Introduction 8.2 Acidic and Corrosive Gases 8.3 Volatile Organic Compounds and Halocarbons 8.4 Ammonia in Water, Air, Clean Rooms, and Process Streams 8.5 Gas Purity and Trap Efficiency 8.6 SF6 Purity in Electrical Switches 8.7 Semiconductor Manufacturing 8.8 Recirculated or Controlled Atmospheres 8.9 VOCs in the Air of the International Space Station 8.10 The Pharmaceutical Industry: Cleaning Verification 8.11 Conclusion References

259 259 261 263 263 263 265 265 266 266 273 273 274 274 276 278 281 282 282 284 284 284 287 288 289 293 293 293 296 299 301 302 303 305 307 309 310 310

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_xiv... 05/05/2007

Página 1 de 2

Page xv 9 Feasible Applications 9.1 Introduction 9.2 Occupational Hygiene and Air Quality 9.2.1 University Stockroom Study 9.2.2 Nicotine Exposure during Production of Skin Patches 9.2.3 Air Quality in a University Research Laboratory 9.3 Fugitive Emissions from Industrial Activity 9.4 Smoke Alarm with Identification of Combustion Sources 9.5 Surface Analysis and Adsorbed Layers 9.5.1 Thermal Desorption of Natural Polymers 9.5.2 Adsorbates and Synthetic Polymers 9.6 Metal and Inorganic Ions 9.7 Aerosols and Electric Mobility Analyzers 9.8 Summary References 10 Present Conditions, Barriers to Advances, and Future Developments in Ion Mobility Spectrometry 10.1 State of the Science and Technology of Ion Mobility Spectrometry (IMS) 10.2 Barriers to Advances in Performance and Uses 10.2.1 Concepts and Practices in IMS 10.2.2 Hardware and Instrumentation 10.3 Future for IMS 10.4 Final Thoughts Index

315 315 315 316 317 318 321 322 324 324 324 326 327 327 329 331 331 333 333 334 335 336 337

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_xv.... 05/05/2007

Página 1 de 2

Page 1

Part I History, Theory, and Basic Principles of Ion Mobility Spectrometry

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_1.h... 05/05/2007

Página 1 de 1

Page 3

1 Introduction to Ion Mobility Spectrometry 1.1 Background 1.1.1 Definition of Ion Mobility Spectrometry The term ion mobility spectrometry (IMS) refers to the principles, methods, and instrumentation for characterizing chemical substances on the basis of velocity of gas-phase ions in an electric field.1 Such principles are simple in concept and, in practice, offer users convenience, high-speed analyses, portability of instrumentation with high reliability, and comparatively low cost of operation. In traditional IMS used for modern analytical measurements, a bundle of ions, known as a swarm, is introduced into a voltage gradient or electric field (E, in units of V/cm). The ion swarm attains a constant velocity through the electric field, called the drift velocity (vd, in units of cm/sec), at ambient pressure in a gas, usually air (Figure 1–1). This velocity is proportional to the electric field strength as in Equation 1–1: vd=KE

(1–1)

The proportionality coefficient, K, is termed the mobility coefficient of the ion in units of cm2V−1sec−1. This relationship is valid only for the ion swarm and not for the speed of individual ions. In air at ambient pressures, swarms of ions between 14 and ~500 amu exhibit velocities of 1 to 10 m/sec in electric fields of 150 to 300 V/cm at temperatures from 25 to 250°C. Calculated mobility coefficients of such drift velocities are 0.8 to 2.4 cm2 V−1sec−1 and are usually normalized to 273 K and 760 torr, yielding a reduced mobility (Ko) as shown in Equation 1–2: Ko=K (273/T)(P/760)

(1–2)

where T is temperature in Kelvin and P is pressure in torr of the gas atmosphere through which the ions move.

1.1.2 Description of Processes in a Mobility Spectrometer Commonly, a measurement begins when a vapor sample is introduced into the portion of a drift tube called the reaction region, as shown in Figure 1–1A.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_3.h... 05/05/2007

Página 1 de 1

Page 4

FIGURE 1–1 Schematic representation of drift tube for IMS. The drift tube is comprised of a reaction region and drift region, both under an electric field gradient (shaded triangle graphic above the drift tube). (A) Two types of neutral sample molecules (small and large circles) are introduced into the ion source region. (B) Sample molecules are ionized (converted to small and large black circles). (C) Ions are injected using an ion shutter into the drift region and separated according to differences in ion mobility. Note that residual sample neutrals are not moved by the electric field.

Molecules in the sample undergo ionization (Figure 1–1B) and are injected using an ion shutter into a drift region where the ion swarms move through the voltage gradient (Figure 1–1C) toward a detector. In Figure 1–1, the ionization and separation of two ions are shown. The time needed for an ion swarm to move the distance (d, in units of cm) between the ion shutter and detector is termed the drift time (td in units of sec or msec) and is referenced to the injection of ions into the drift region (t=0). Differences in the mobility coefficients of the ions lead to different drift velocities, seen as drift times in Equation 1–3: td=d/vd

(1–3)

Drift regions of modern analytical drift tubes are typically 4 to 20 cm long. At ambient pressure and temperature, an ion swarm with Ko=2 cm2 V−1sec−1, moving at 4 m/sec, would have a drift time of 15 msec in a drift tube with E= 200 V/cm and length of 6 cm.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_4.h... 05/05/2007

Página 1 de 1

Page 5

FIGURE 1–2 A positive polarity mobility spectrum for 2-pentanone in air. Sample in IMS is often converted to ions through chemical reactions with a source of charge, the reactant ions (here H+(H2O)n) as described in Section 1.1.3. The reactant-ion peak is apparent at 4.45 msec. The protonated monomer and the proton-bound dimer appear at 5.075 msec and 6.225 msec, respectively.

Ions are neutralized upon collision with the detector, causing a current flow of 10 to 1000 pA, which is amplified and converted into voltage, typically 0 to 10 V. A plot of the detector response vs. drift time is called a mobility spectrum (Figure 1–2) and contains all the information provided by a mobility measurement. This includes the mobility coefficient (which is characteristic of an ion), peak shape (a measure of drift-tube performance), and secondary spectral details including ion fragmentation (a measure of chemical class). The mobility spectrum provides a dimension of selectivity that is not found in simple ionization detectors such as the electron capture detector, owing to the separation of ions on the basis of differences in drift velocities. Reduced mobility values can be used for chemical identification, and analysis of the mobility spectrum can allow classification by chemical family.2–4 Drift times and Ko values can be understood primarily using the size or shape of ions, making IMS a type of ion-size analyzer as described in Chapter 2. The utilization of mobility coefficients and mobility spectra to measure the composition of a sample is predicated upon the successful conversion of sample molecules to gaseous ions.

1.1.3 The Formation of Gaseous Ions in Positive Polarity Ion formation in IMS has been accomplished historically through chemical reactions between molecules of the sample and a reservoir of ions known as reactant ions. In addition to ion separation by mobility coefficients, the chemistry of ionization in air at ambient pressure contributes another layer of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_5.h... 05/05/2007

Página 1 de 1

Page 6 selectivity in the overall analytical response, and this strongly encouraged the development of IMS as a practical measurement technology. In original analytical IMS instruments, the ionization source was 370 MBq (10 mCi) of 63Ni, which emits electrons with a mean energy of 17 keV.5,6 These energetic electrons collide with molecules of the supporting atmosphere, forming ions and producing secondary electrons (see Chapter 3). A sequence of ion-molecule reactions with nitrogen, oxygen, and water vapor in purified air results in the formation of a reservoir of ions of H+(H2O)n in positive polarity and in negative polarity.7–9 These ions are called reactant ions and are essential to the ionization of a sample. The total charge of the reactant ions is governed by the size of the ionization source and establishes the upper limit of the number of molecules that can be ionized. Sample molecules (M) are ionized by collisions with the reactant ions, forming product ions that are stabilized through the displacement of water molecules bound to the cluster ion, as shown in Equation 1–4: (1–4) The cluster ion may be stabilized by collision with another molecule, a thirdbody collision usually from the supporting atmosphere, with a subsequent loss of neutral water adducts. The product ion in Equation 1–4 is called a protonated monomer, and the value of x is commonly understood to be between 1 and 3, though this is dependent upon moisture in the supporting atmosphere. In practice, the introduction of a vapor sample, M, into the reaction region of a drift tube leads to a decline in the peak intensity of the reactant ion and the appearance of a product-ion peak. Differences in mobilities between the reactant and product ions can be resolved in the drift region, where ions appear as distinct peaks with characteristic drift times as seen in Figure 1–2; the drift times can then be converted to Ko values. As the vapor concentration of the sample is increased, a second product ion with a characteristic drift time is often formed with a decline in the intensity of the reactant-ion peaks and the peak of the protonated monomer. This is shown in Equation 1–5 in which another sample neutral attaches to the protonated monomer, forming a proton-bound dimer M2H+(H2O)n−x. (1–5) The proton-bound dimer, resolved from the reactant-ion peak and the protonated monomer (see Figure 1–2), provides additional information in the mobility spectrum. Moreover, the concentration dependence of peak intensities adds useful quantitative information to measurement as described in Chapter 3 (Section 3.3.1.1). The simplicity of Figure 1–2 is generally observed in the mobility spectra of most chemicals and provides a comparatively simple characterization of the ions and, therefore, of the sample.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_6.h... 05/05/2007

Página 1 de 1

Page 7

1.1.4 The Formation of Gaseous Ions in Negative Polarity The formation of reactant ions in negative polarity occurs by means of resonant electron capture, and, therefore, it occurs between low-energy electrons and neutrals such as oxygen. Ion-molecule reactions can take place between reactant ions and a sample molecule (M), leading to the formation of a product ion as shown in Equation 1–6: (1–6) In Equation 1–6, the reaction shown is the formation of an adduct ion between the sample and hydrated O2−. The basis of reactivity with negativeion chemistry is found in the association between the molecule and the anion O2−. This adduct may live long enough to be measured in the IMS drift tube or may undergo further reactions to form M− or (M−1)−. Some chemical groups do not form stable adduct ions and, consequently, show no response with negative polarity. Based on chemical structure, compounds may exhibit a type of preferred response in one polarity over the other. In either polarity and regardless of the method used to ionize a sample, the measurement of ion mobility is a central and distinguishing facet of IMS.

1.1.5 The Separation of Ions and the Determination of Mobility Ion characterization by mobility is simple in practice, providing a rapid and low-cost method to analyze samples. Moreover, the association between the mobility spectrum and sample composition is often direct and straightforward. The central question in mobility measurement is the relationship between the drift velocity of the ion swarm and the chemical identity of the ions in the swarm. Early attempts to relate ion structure or identity to mobility coefficients arose mainly from studies of mono- or diatomic ions in pure gases at subambient pressure.10 Efforts have been made to extend these models to large organic ions with various functional groups in air at ambient pressure.11 Such associations between ion structure and experimental mobility coefficients are complicated by the formation of cluster ions between the ion and neutrals of the supporting atmosphere. Drift velocities are accurate reflections of the mobility of specific ions in a supporting atmosphere, and, therefore, both the identity of the ion and the composition of the neutral gas affect the mobility measurement. Control of instrumental parameters is pivotal to the use of IMS for analytical measurements and the relationship between K and ion identity; this can be attained without difficulty. However, inattention to a few parameters such as moisture level or unwanted neutrals in the drift region can cause confusion between the expected ion identity and the actual ion formed and characterized in the drift tube. These errors plagued early IMS investigations and are discussed in Chapter 3 (Section 3.2).

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_7.h... 05/05/2007

Página 1 de 1

Page 8

FIGURE 1–3 Plots of reduced mobility vs. mass for volatile organic compounds from three chemical families including alcohols (□), esters (Δ), and organophosphates (O).

When all the parameters of instrumentation and chemistry are controlled, the mobility coefficient is governed by the size-to-charge ratio and the reduced mass of the ion in the supporting atmosphere. As shown in Figure 1–3, reduced mobility coefficients are influenced by ion mass in cases in which a linear relationship exists within a homologous series. However, ions of the same mass but different functional groups, or ions of the same functional group but different geometrical arrangements (isomers), often exhibit different Ko values, reflecting the influence of shape and size on mobility. Models were developed to relate the mobility coefficient to properties of the ion, and one formula is shown in Equation 1–712 (1–7) where e is the charge of an electron; N is the number density of neutral-gas molecules at the pressure of measurement; α is the correction factor; μ is the reduced mass of ion and gas of the supporting atmosphere; Teff is the effective temperature of the ion determined by thermal energy and the energy acquired in the electric field, and ΩD is the effective collision cross section of the ion in the supporting atmosphere. Although the mass of an ion strongly affects the mobility coefficient, mobility depends on more than the ion mass and includes information on the ion structure and the collision cross section. This is seen in Equation 1–7 and in Figure 1–3 in which K is inversely proportional to ΩD and ions of the same (or similar) mass exhibit different mobilities. The relationship between K and ΩD, as shown in Equation 1–7, is reasonably well established but incomplete for

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_8.h... 05/05/2007

Página 1 de 1

Page 9 polar organic molecules. For example, the charge distribution within a large organic ion affects observed mobility values and is not comprehensively modeled in current formulas. Thus, limitations exist in the current theory; we are yet to develop a comprehensive model that fully correlates the structure of a molecule with the collision cross section (ΩD) and ΩD to K or Ko. In this regard, analytical IMS remains an empirical measurement technique with incomplete supporting fundamental tools. The preceding discussion was intended to emphasize that IMS should be understood as two sequential processes: (1) the formation of ions that are representative of a sample, and (2) the characterization of these ions for mobilities in an electric field. Both of these can be discussed and treated separately though the final analytical result from a mobility spectrometer is a sum of events from both processes. During the past decade, analytical IMS has come to occupy an unprecedented position of importance in military preparedness and commercial aviation security. These IMS applications were initiated shortly before the first edition of this monograph on IMS was published, and in the ensuing 10 years, have become central in analytical technologies for homeland security. However, the roots of much that is accepted in IMS today were first seen and documented in laboratories of physicists in Europe from the mid-1800s to 1938. As shown in the abbreviated discussions in the following sections, the story of ions in air provides a rich sampling of the history of science and technology; it provides a context for understanding the origins of and developments in IMS today.

1.2 Studies of Ions in Gases at Atmospheric Pressure Analytical IMS arises from the formation and behavior of ions in gases at ambient pressure, and this topic may be classified into periods of discovery and innovation (1850 to 1938) and foundational studies (1948 to 1970). Progress and development were neither continuous nor smooth throughout these periods. Extended periods of inactivity, when interest in ions at ambient pressure waned, were interspersed with periods of intense study, typically in only a few laboratories. Studies of ions using mobility principles never attained the level of interest or development seen with mass spectrometry (MS), though both IMS and MS emerged from the same laboratory and in the same line of scientific inquiry.

1.2.1 Period of Discovery and Innovation The phenomenon of electrical discharges in air and other gases was explored in research centers in England and Germany from the mid to late 19th century. This experience led to the conclusion that gases could be transformed from

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_9.h... 05/05/2007

Página 1 de 1

Page 10 an insulator to a conductor of electricity using various methods. The pioneering investigators understood that changes in the conductivity of air were due to the formation of electrons and gaseous ions. This discovery prompted questions on the identity of the ions and the structure of matter. For example, gases were ionized when exposed to x-rays, and Rutherford measured the mobility of ions formed by x-ray ionization13 within 2 years of Roentgen’s discovery of x-rays in 1895.14 Air was also ionized by radioactivity,15 and soon afterwards, associated ions were characterized using mobility.16 A large body of work on the instrumentation and measurements of ion mobilities in gases occurred from the late 1890s until about 1938 and established an understanding of ion measurements and the meaning of ion mobilities. In Rutherford’s studies of air ionized with a pulsed x-ray source, ions were characterized using a set of parallel plates with an electric field of 13.75 V/cm. Ions at ambient pressure required 0.36 sec to move 8 cm, yielding a velocity of 1.6 cm/sec for an electric field of 1 V/cm.13 Mobility coefficients of 1.8 cm2 V−1sec−1 were measured for negative ions formed by corona discharge in dried air in 189917 and later as 1.37 and 1.80 cm2 V−1sec−1 for positive and negative ions, respectively, in dry air from x-rays in 1908.18 Ions were observed from still other sources including flames19 and ultraviolet light.20 In short, there was an early burst of activity and a broad range and depth of inquiry concerning ions formed in air at ambient pressure. These early discoveries are summarized in a 1928 monograph by Thomson and Rutherford, in which both the instrumentation and related experimental findings are described and discussed in detail.21 By 1938, a wealth of data and experience existed regarding the identity of the ions formed in gases and their mobilities, as well as how these were affected by temperature, pressure, and gas purity.22 That is, the exact composition of an ion in air or other gases at ambient or elevated pressures is governed by the composition of the gas through which the ion moves. The formation of clusters between ions and polar neutral gases was recognized early on in the studies of ion mobility. A specific example of this can be found in two early reports from Lattey who explored the velocity of ions in dried gases and proposed that an envelope of molecules surrounded the positive ions.23 He found that, in the presence of small traces of water vapor, the velocities of negative ions were particularly dependent on moisture levels.24 Naturally, exact mass determinations and the identities of ions in air were not available to these early researchers; however, careful analysis and rationalization provided a general, if not thorough, understanding of the composition and behavior of ions in gases. Three specific developments in these early years were noteworthy for their contribution to subsequent advances in mobility measurements and analytical IMS. The first is the theoretical treatment of mobility and practical experimentation by Langevin who made expansive contributions through two remarkable articles.25,26 Langevin recognized the collisional nature of mobility and the role of attractive forces on effective collision cross sections, and formed an early description of ionmolecule associations and the influence of ion-molecule interactions on mobility. A second development was that of pulsed ion injection

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_10.... 05/05/2007

Página 1 de 1

Page 11 into a drift tube using ion shutters of the type used in drift tubes today. Ion shutters with parallel wires were described in 1929 by both Cravath27 and van de Graaff.28 Bradbury developed this method further with a linear field and two-shutter design with an early boxcar integrator.29 This innovation was employed later in 1970 with the first commercial IMS as described in Section 1.4. A third feature of this period was the reasonably sophisticated understanding of the effect of electric fields and gas pressure on the mobility of ions in gases. Mobility was shown, early in the study of ions in gases, to be independent of the ratio of electric field to pressure (E/N, in units of Td where 1 Td=10−17 V cm2) up to several Td. At values of E/N greater than 2 to 4 Td, mobility became field dependent on E/N, as shown in Figure 1–4.22 Various ions exhibited unique plots of mobility vs. E/N, anticipating the development of high-field IMS as described in Chapter 2. On the whole, methods employed for research into the mobility of gas-phase ions at ambient pressure were gradually supplanted by the new and powerful technique of mass spectrometry, which was free of the problems of secondary

FIGURE 1–4 Plots of mobility vs. E/p for ions in helium. In these plots, the desolvation of an ion with the resultant increase in mobility can be seen in the break in linearity of K vs. E/p. The mobility of strongly hydrated ions such as Li+ was unaffected until high E/p values were reached, whereas weakly hydrated Cs+ showed a break at comparatively low E/p. (Cited in Reference 22, page 68; originally published by Mitchell and Ridler, Proc. Royal Acad. 1934, 146, 911.) E/p is now expressed as E/N; see Huxley, L.G.H., Crompton, R.W., and Elford, M.T., Use of the parameter E/N, Br. J. Appl Phys., 1967, 18, 691.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_11.... 05/05/2007

Página 1 de 1

Page 12 chemical reactions under a vacuum of 10−5 torr. MS emerged from the Cavendish Laboratory where important contributions to the understanding of ionization of gases and ion mobility had been made.30 A small group of researchers used mobility studies for research into specific topics (such as the ionization of air through spark discharges),31 though the study of mobility and gaseous ions became nearly inactive. Later, focus on the basic studies on ion-molecule interactions and the effects of ionization caused by nuclear tests in the atmosphere would lead to a renewed interest in gas-phase ions at ambient pressure. During World War II, a remarkable example of ion detection in ambient air demonstrated the potential for analytical capabilities in air monitoring. Apparently, diesel fumes from submarines were detectable by primitive ionization detectors fitted on some Allied boats. Details and specifications of this analyzer, known as Autolycus, are generally unavailable, apart from a few Internet references.32,33 It “…was intended to be able to pick up the diesel exhaust fumes…although the obvious difficulty was distinguishing submarine fumes from those of any other vessel…”32 Autolycus was a primitive demonstration of the link between air composition and ion measurements 8 years before Lovelock’s discovery that industrial solvents in ambient atmospheres could be detected with ionization detectors, and was the basis for gas analysis by ionization-based technologies such as IMS.

1.2.2 Period of Foundational Studies (1948 to 1970) Advances that laid the foundation of modern analytical IMS can be attributed to Lovelock’s 1948 report that a simple ionization detector responded to ultralow airborne vapor concentrations of industry-related organic vapors released as pollutants into the atmosphere.34 Lovelock had been engaged in medical studies associated with the speculation that breezes could be implicated in human respiratory illness. For these studies, he developed a device in which slight perturbations of wind direction or speed were registered as changes in the ion current at wire collectors. This vapor anemometer (shown in Figure 1–5), was dependent upon the degree to which cross winds deflected slow-moving, large ions from the region between a radioactive source and the wire collectors. The displacement of these ions was related to the direction of the cross-current flow of air and proportional to the strength of the breeze. Lovelock also found that his anemometer responded to certain airborne vapors, especially halocarbons, which reacted with the charge created by the radioactive source and measurably altered the current collected in his device. Versions of this technology were later distributed as the electron capture detector (ECD)35 for gas chromatographs and eventually revolutionized chemical measurement of certain substances, including pesticides in environmental samples. The importance of Lovelock’s studies on air pollution was that a direct link was made between the composition of a vapor sample, such as trace impurities in ambient air, and the ions created in a beta emission source. This can be seen in the direct monitoring of atmospheric tracers such

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_12.... 05/05/2007

Página 1 de 1

Page 13

FIGURE 1–5 (A) Schematic representation of Lovelock’s vapor anemometer, which showed response to atmospheric pollutants. (B) Photograph of analyzer section of the vapor anemometer.

as SF6 with flow-through ECD-based instruments.36 Such associations between gas composition and ionization chemistry provided the missing component that, when combined with advanced ion characterization by mobility, would lead to IMS. There is no suggestion that Lovelock’s studies or other studies with ECD analyzers specifically influenced the thinking of modern pioneers in IMS. However, the ECD and the ion mobiliiy spectrometer share common principles of the chemistry of gas-phase reactions at

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_13.... 05/05/2007

Página 1 de 1

Page 14 atmospheric pressure, and the study of ion chemistry in an ECD was quickly recognized as a parallel to that in a mobility spectrometer. The first generation of IMS researchers used this to rationalize the response of the ion mobility spectrometer to halogenated compounds. 1.2.2.1 Renewed Interest in Ions in Gases at Elevated and Ambient Pressure Interest in gas-phase ion-molecule reactions at elevated pressures arose again in the 1950s as researchers began investigating the radiolysis of air (with the advent of the nuclear age), the reactions in the upper layers of the Earth’s atmosphere (through space exploration and environmental chemistry), and certain specific technologies such as photocopiers. Mass spectrometers were especially valuable in identifying the reaction products and thermochemical parameters from ionization at elevated pressures.37 For example, mass spectrometers were used to identify ions in flames38 and in corona discharges.39–41 Shahin at Xerox Corporation sampled ions in a corona discharge using the pinhole orifice of a mass spectrometer to explore the chemistry of modern photocopy machines and laser printers.39,41 This work was contemporaneous with the study of ionmolecule reactions in chemical ionization mass spectrometry.42 In the 1960s, Kebarle began extensive investigations into the reactions and kinetics of ion-molecule reactions at elevated pressures. These investigations provided a chemical foundation for the interpretation of some, but not all, of the chemical events that occur inside an IMS drift tube.43,44 Through the efforts of these researchers and others not cited here, a base of experience in ion-molecule chemistry existed by the late 1960s to support the next stage of development: ion characterization in air at ambient pressure. 1.2.2.2 Drift Tubes for Ion Characterization by Mobility in Weak Electric Fields The missing component needed to create modern analytical mobility spectrometry was a suitable drift tube for ion characterization. From the late 1950s to late 1960s, foundational studies on the mobility of ions were undertaken in laboratories such as those of Mason at Brown University45 and McDaniel at Georgia Tech.46 Their interests were to probe the interactions between ions and molecules using small symmetrical ions such as Na+ or H+ that were passed through gases such as Ar or CO2 at low pressure. Models of interaction potentials between the ions and neutral molecules, i.e., a buffer gas, were probed and resulted in formulas such as those presented in Equation 1–7. The drift tube used at Georgia Tech was based on a stack of electrically isolated rings, and a linear electric field was established with a voltage divider. An ion source or reaction region could be sampled and ions injected into the drift region.47 The drift tube was equipped with

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_14.... 05/05/2007

Página 1 de 1

Page 15

FIGURE 1–6 Apparatus for determination of mobility of ions in gases from Albritton and McDaniel. (From Albritton, D., The Mobilities of Mass-Identified H3+ and H+ Ions in Hydrogen, Ph.D. thesis, Georgia Institute of Technology, March 1967.)

an aperture to sample ions by MS for mass identification (see Figure 1–6) and was referenced to Crompton et al. in Australia.48 Though the interests of Albritton and McDaniel bore no direct relationship to chemical analyses as found in modern analytical IMS and indeed their drift tube, under partial vacuum, was not a practical device, this design was later adapted for chemical measurements, and it became the core technology for modern analytical IMS. The concept of flooding the drift tube with air and operating the instrument at atmospheric pressure was a next and bold step in instrumentation and chemical measurement science. The concept of characterizing vapors as ions in air using a drift tube at ambient pressure originated in the period from 1964 to 1967 with Martin Cohen and coworkers. Cohen had a background in particle physics and experience in the study of ions in gases. By the mid-1960s, he was employed at Franklin GNO, which received funding through military contracts for chemical monitoring. Some of these are shown in Table 1–1. Developments in this company resulted in a prolific patent record as shown in Table 1–2 and a description of analytical IMS, the so-called Plasma Chromatography™. At a conceptual level, IMS is a blend of Lovelock’s ionization detector and the drift tube at Georgia Tech, though the innovation of Cohen and colleagues was distinctive. The term plasma chromatography was used because a mixture of positive and negative ions in the gas phase was known as a plasma, and the separation of ions was suggestive of the practice of chromatography. The instrumentation (Figure 1–7) that was soon offered commercially came in two configurations: Alpha and Beta, the first with a mass spectrometer behind the ordinary Faraday plate detector, and the second without the spectrometer.49,50 McDaniel’s drift tube (Figure 1–6) and the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_15.... 05/05/2007

Página 1 de 1

Page 16 TABLE 1.1 Partial Listing of Contract Record for Franklin GNO, Predecessor of PCP, Inc. (Now an Italian Company)

Title

Contract No.

Agency

Date

Nonr-4977 (00)

Office of Naval Research

May 1965 March 1967

Investigation of the properties of negative ions produced Nonr-4924 by the interaction of large electronegative gas molecules (00) with free electrons (Studies with SF6.)

Office of Naval Research

June 1965 July 1967

Experimental investigation of electron-attachment characteristics of certain materials in atmospheric air (Studies on dimethylhydrogenphosphite, triethylphosphate, and Sarin with a pulsed D2 lamp at 50–100 torr in negative polarity the drift tube was in a glass chamber.)

Studies of electrophilic gases for plasma quenching

NAS1–5575 NASA—Langley Research Center

October 1965 December 1966

High-temperature studies of electrophilic gases for plasma quenching

NAS1–6884 NASA—Langley Research Center

January 1967 October 1967

Study of material indicated in proposal

DCA100– 68–C–0005

January 1967 October 1967

Defense Communications Agency

FO8635–67– Air Force Armament Performance study of the PCa marking system Laboratory, Eglin Air (Study of personnel detection and chemicals for tracking C–0075 Force Base and marking; this was the first all-metal stainless steel drift tube equipped with 250-μsec shutter pulses; twoshutter coplanar design with continuous UV lamp.)

April 1967 June 1968

PC experimentation with already available PC instrument: design, construct, and laboratory-evaluate modular prototype PC (New drift tube with wide rings, high-temperature operation (200°C), vibration rugged. Operated in truck and helicopter from batteries or generator.)

November 1968 May 1970

DAADO5– 69–C–0139

Land Warfare Laboratory, Aberdeen Proving Ground, U.S. Army

a Plasma chromatography

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_16.... 05/05/2007

Página 1 de 2

Page 17 TABLE 1–2 Partial Listing of Franklin GNO Patent Record on IMS (Plasma Chromatography)

Title

Inventor(s)

U.S.Patent No.a

Dates Filed and Granted

Cohen, Carroll, Wernlund, Kilpatrick

3,699,333

October 23, 1968; October 17, 1972

Time-of-flight mass spectrometer having a flowing gas Cohen, Carroll, stream perpendicular to the ion drift field for increased Wernlund resolution

3,624,389

November 26, 1968; November 30, 1971

Apparatus and methods for separating, concentrating, detecting, and measuring trace gases

Separation and detection of trace substances in gaseous samples containing moisture by diluting with dry air

Cohen, Wernlund, Carroll

3,668,382

November 26, 1968; June 6, 1972

Apparatus and methods for separating, detecting, and measuring trace gases with enhanced resolution

Carroll, Cohen, Wernlund

3,626,180

December 3, 1968; December 7, 1971

Apparatus and methods for separating, concentrating, detecting, and measuring trace gases

Carroll

3,668,383

January 9, 1969; June 6, 1972

Detecting a trace substance in a sample gas through processes such as reacting the sample with different species of reactant ions

Cohen

3,621,239

January 28, 1969; November 16, 1971

Apparatus and methods for separating electrons from ions

Carroll

3,629,574

January 28, 1969; December 21, 1971

Wernlund Gas detecting apparatus with means to record detection signals in superposition for improved signalto-noise ratios

3,526,137

February 11, 1969; December 7, 1971

Apparatus and method for improving the sensitivity of Cohen time-of-flight ion analysis by ion bunching

3,626,182

April 1, 1969; December 7, 1971

Time-of-flight ion analysis with a pulsed ion source employing ion-molecule reactions

Cohen

3,593,018

April 1, 1969; July 13, 1971

Apparatus and methods for detecting and identifying trace gases

Cohen, Carroll, Wernlund, Kilpatrick

3,621,240

May 27, 1969; November 16, 1971

Methods of monitoring the presence or movements of humans

Cohen

4,195,513

June 18, 1969; April 1, 1980

Apparatus and methods employing ion-molecule reactions in batch analysis of volatile materials

Carroll, Wernlund, Cohen

3,639,757

August 4, 1969; February 1, 1972

Plasma chromatograph with internally heated inlet system

Cohen

3,697,748

October 6, 1969; October 10, 1972

Apparatus and methods for measuring ion mass as a

Wernlund,

3,812,355

December 9,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_17.... 05/05/2007

Página 2 de 2

function of mobility

Kilpatrick, Carroll, Cohen

1971; May 21, 1974

a Patents not listed: 3,668,385; 3,596,088; 3,626; 1,783,626; 179; 3,697,749; 3,742,213.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_17.... 05/05/2007

Página 1 de 1

Page 18

FIGURE 1–7 Photograph of the Beta series plasma chromatograph from early 1970s (top) and an Alpha model IMS/MS from the early 1980s (bottom). Manufactured by PCP, Inc, West Palm Beach, FL.

mobility spectrometer (Figure 1–8) among those early commercial instruments show strong similarities in the philosophy of construction and dimensions of the components; McDaniel had become a consultant to this team, now organized as a company, Franklin GNO (for gnostic, Greek for knowledge), later reorganized as PCP, Inc. The relevance of IMS to military preparedness and other security interests would be a recurring theme throughout the next three decades. The beginning of commercialization of IMS technology marked an end to the period of foundational discovery and introduced an era of extensive exploration of IMS as an analytical tool. Thus, the work of F.W.Karasek, who surveyed the response of IMS to organic compounds, was pivotal in stirring interest in plasma chromatography.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_18.... 05/05/2007

Página 1 de 1

Page 19

FIGURE 1–8 Photograph of the drift tube from Beta series plasma chromatograph (or ion mobility spectrometer).

1.3 Early Developments of IMS as an Analytical Method (1970 to 1990) 1.3.1 F.W.Karasek and Studies Using IMS for Chemical Analyses In 1970, F.W.Karasek relocated from Phillips Oil Company to the chemistry department at the University of Waterloo, Ontario, Canada. This coincided with the introduction of plasma chromatography at scientific meetings.49 Karasek showed an immediate interest in the technique50,51 and obtained a Beta-VI plasma chromatograph for his laboratory. He began a research program in IMS, which demonstrated a broad range of response to organic

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_19.... 05/05/2007

Página 1 de 1

Page 20 TABLE 1–3 Selected Publications by F.W.Karasek and Co-workers on IMS, During the Early Years of Exploration (1970–1973)

Title

Reference

Plasma chromatography—a new dimension for gas chromatography and mass spectrometry

50

The plasma chromatograph

51

Qualitative studies of trace constituents by plasma chromatography

52

Plasma chromatography of the polychlorinated biphenyls

53

Trace studies of alcohols in the plasma chromatography mass spectrometer

54

GC-PC interface and its performance in the detection of musk ambrette

55

Plasma chromatography of the monohalogenated benzenes

56

Trace analysis and fundamental studies by plasma chromatography

57

Plasma chromatography of the n-alkyl alcohols

58

Study of electron capture behavior of substituted aromatics by plasma chromatography

59

Scope of plasma chromatography

60

Plasma chromatography of the n-alkyl halides

61

Evaluation of plasma chromatography as a qualitative detector for liquid chromatography

62

compounds and the astonishingly low detection limits for many compounds without sample pretreatment.52–62 The listing of titles in Table 1–3 illustrates the diversity of studies in Karasek’s experiments in the early 1970s. Because the Beta-VI was not equipped with a mass spectrometer, deductive reasoning was needed in many experiments in order to connect gas-phase chemistry with the response in the mobility spectrometer. This was complicated because the association between chemical ionization mass spectrometry and IMS is close but not direct, and the results were sometimes difficult to defend. Another obstacle was the Beta-VI instrument, which was a research instrument without the refinements of later drift tubes. A characteristic of these early Franklin GNO drift tubes was their construction, with drift rings separated by sapphire balls and the entire assembly of rings held together under gentle compression (Figure 1–8). The gap between the rings meant that vapors could diffuse from the inside of the drift tube to the gas volume between the drift tube and a protective shell; moreover, vapors could diffuse from this extra volume back into the drift tube. As a consequence, the flow patterns inside the drift tube were complex and did not prevent inexperienced users from overloading the ion source region with the sample. Excess sample vapor could enter the drift region by diffusion in an uncontrolled manner, forming ion-molecule clusters composed of product ions and neutrals of the sample throughout the length of the drift tube. The size and mobility of these ion clusters were affected by vapor concentrations above certain levels so that the mobility spectra and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_20.... 05/05/2007

Página 1 de 1

Page 21 mobility coefficients appeared to be dependent on concentration.63 This was misunderstood as irreproducible chemical science and affected both the understanding and the acceptance of IMS in the U.S. Also, mistakes arose from the erroneous view of IMS as a type of gas-phase ion chromatography.64,65 This is epitomized in a comment64 that IMS “…was not what we have come to believe a chromatographic method should be.” Indeed, IMS was not then, and is not today, a chromatographic method, and such comparisons, in retrospect, seem ill advised. Another aspect of the Beta-VI was the use of boxcar averaging for collecting a mobility spectrum. In this approach, a second ion shutter was synchronized with the first shutter, and a delay (referenced to the first shutter) in the second shutter swept across a range of time in order to sample the whole of the mobility time scale. Up to 5 min were needed to obtain a mobility spectrum, and fast-changing ion-molecule processes could not be measured or even observed. This limited technology was replaced in the 1980s with inexpensive computers and analog-to-digital converters for digital signal averaging; however, poor opinions strongly held on IMS as an analytical instrument seemed to linger well through the 1980s. Few other academics explored IMS and ion chemistry in air at ambient pressure because of the complicated charge-transfer reactions and the formation of ion clusters. Moreover, concepts of mobility of ions or ion clusters were not appreciated and were sometimes confused with ion mass.66 Horning concluded that ionization at ambient pressure was valuable and that a mass spectrometer was a good detector; however, the mobility spectrometer contributed little to analytical measurements.67 He held that the drift tube in their IMS/MS instruments was an unwelcome complication with no benefit, and the radioactive source was placed directly at the pinhole interface of the mass spectrometer. The ion source was held in a pneumatically tight chamber permitting control of gas composition and sample and a new analytical method was announced—atmospheric pressure ionization mass spectrometry (API-MS).67 By 1975, it appeared that IMS might have become a forlorn technology with no lasting contribution to modern methods of chemical measurement. Exceptions to this gloomy view were successful industrial applications of IMS as a detector of trace impurities on the surfaces of silicon wafers in the electronics industry and as a stack gas monitor for releases of halocarbons.68,69 These were not well known, and from 1975 to 1980 the number of publications and presentations at learned society or professional meetings rapidly declined. Not a single refereed journal article on IMS appeared in 1980.

1.3.2 Development of IMS by Military and Security Organizations Activity in IMS seemed dormant in the late 1970s, apart from the work of Karasek’s team and a few journal articles. However, a remarkable transformation of large laboratory instruments into handheld, rugged analyzers was under way in military establishments in the U.K. and U.S. Parallel to military

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_21.... 05/05/2007

Página 1 de 1

Page 22 efforts to engineer in-field chemical warfare agent detectors, efforts were being made in industrial groups to transform IMS instruments for detection of explosives and drugs. Common to all these initiatives was the understanding that principal objections to IMS related to technology or engineering rather than principles of chemistry and physics. Indeed, IMS was fundamentally attractive to military and security establishments precisely because ion chemistry was favorable for substances used by them. Nerve agents (organophosphorous compounds) exhibit strong binding to gas-phase proton clusters, whereas explosives undergo preferential charge exchange or binding with negative reactant ions. These features provide low detection limits and reasonable selectivity for the target chemicals even in a complex matrix. Finally, the low requirements for power and utilities in IMS-based technologies facilitated the development of small size, lightweight, and low-power analyzers. Consequently, military and security development programs existed from 1965 to 1985 for ion mobility spectrometers and a range of other ionization detectors. These programs were not widely reported, but they catalyzed a renewed interest in IMS as a modern analytical technique beginning in the mid-1980s and going into the early 1990s. Some details of these military development programs are described in the following section. Vapor detectors based on ion-molecule reactions in air were developed during the 1960s by the U.S. Army and were a type of mobility cutoff filter for gas-phase ions. Only ions over a certain size would register response, such as those from nerve agents, whereas the smaller ions commonly formed in clean air would not. These detectors were deployed in the 1970s and included the M-8A1 detector system with the M-43A1 ionization detector cell.70 Two contemporaries of the M-8A1 were DICE71 (detection by ion combination effect), developed at the Chemical Defense Establishment at Porton Down (U.K.) and the mini-ionization detector, built by Honeywell in Germany.72 Both the DICE and the mini-ionization detector were simple devices involving speciation of ions in electric fields. In Finland, a device was developed in which ions traversed a laminar flow of gas in an electric field and were separated according to their path.73,74 The transition from simple analyzers or mobility cutoff filters to authentic mobility spectrometers was associated with a need for the improved specificity and enhanced value that mobility could bring to a measurement. Simultaneous with the investigations done by chemist C.Steve Harden at Edgewood Arsenal, U.S. Army, was the work of physicist David Blyth (at Porton Down). They independently saw advantages in gas-phase ion chemistry and mobility analysis. Eventually, the development of IMS analyzers became a joint program with a division of labor between the U.S. and the U.K. defense establishments. The responsibility of the U.S. military was to produce a continuous air monitor known as ACADA (automatic chemical agent detector and alarm).75 In the U.K., the responsibility was to produce a personal monitor that could operate for short periods in contaminated environments. Eventually, a handheld IMS analyzer, the chemical agent monitor (CAM)76 was developed and produced in large numbers by Graseby

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_22.... 05/05/2007

Página 1 de 1

Page 23

FIGURE 1–9 A photograph of a soldier in a protective suit for chemical warfare with a handheld military-style ion mobility spectrometer (the chemical agent monitor or CAM). The demonstration of IMS in a rugged small field design has invigorated, in part, interest in IMS as an advanced and practical analyzer. Refinements have been made in subsequent years after CAM was introduced.

Dynamics (Figure 1–9). Another application of IMS is the detection of explosives, and this is the second most extensive use of IMS analyzers. Indeed, IMS analyzers have high visibility as explosive detectors and are likely to be noticed by anyone using commercial air transportation today. The gasphase chemistry of ionization of explosives in air in the negative polarity allows part-per-billion (ppb) or subnanogram detection limits for nitroorganic explosives.77–80 A comparatively high selectivity permits explosives to be detected in complex samples without any sample pretreatment or preseparation. The need for detection of explosives coincided with realistic fears of terrorism in commercial aviation, and two configurations of ion mobility spectrometers were proposed to meet the need for fast, noninvasive screening of carry-on luggage and articles. These were the model 400 ION-SCAN from Barringer Research Ltd. (now Smiths Detection)81 and the Itemizer from IonTrack, Inc. (now GE Interlogix).82 The whole story of explosives detection is not broadly published and cannot be told here; nonetheless, the appearance of the IONSCAN and Itemizer (shown in Figure 1–10) assisted in changing perceptions and acceptance of IMS as an analytical measurement technique. As described in regard to CAM, governments and security authorities were willing to entrust the lives of soldiers and the traveling public to the response of ion mobility spectrometers. An unwelcome consequence of these origins of IMS was the limited disclosure on instrumentation and methods in the literature. Understandably,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_23.... 05/05/2007

Página 1 de 1

Page 24

FIGURE 1–10 Commercial IMS-based explosives detectors from two leading manufacturers: Itemizer from GE-Interlogix (top) and Ionscan from Smiths Detection (bottom).

research in support of military preparedness and aviation security yielded sensitive knowledge which remains so today. A consequence of mission-driven developments in IMS was a narrowly focused research program. Thus, the time for engineering and applying IMS arrived and departed without any understanding the basic foundation of IMS instruments, namely, the chemistry of ionization and the behavior of ions at ambient pressure. The contributions from studies on ECD,83–85 chemical ionization MS,86 and high-pressure MS provided a foundation for understanding the kinetics and thermodynamics of gas-phase ion-molecule reactions; however, there was no strong relevance for IMS due to the complications caused by cluster formation, which is common in IMS. These limitations were addressed or clarified by the late 1990s.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_24.... 05/05/2007

Página 1 de 1

Page 25

1.3.3 Fast-Responding and Pneumatically Sealed Drift Tubes Much of the modern era of IMS was initiated by Baim and Hill87 who eliminated several limitations of IMS by the development of a drift tube with unidirectional gas flow, to minimize the residence time of neutrals in the ion source. The goal was to create a fast-responding detector with low peak broadening for use in capillary gas chromatography (GC), in which peak widths at baseline could be 5 to 15 sec. In this scheme, purified gas was introduced into the drift tube at the detector, passed through the drift region and source region, and then vented (Subsection 5.2.1). Effluent from the GC column, containing analyte molecules, was introduced into the reaction region upstream immediately before the ion source such that any unionized sample was rapidly swept from the drift tube. In this design, sample residence in the reaction zone was fixed and limited, and the drift region was kept free of sample. Sample neutrals could no longer enter the drift region unless the analyst grossly overloaded the source region or drift-gas flows were disrupted. This elegant approach to managing gas flow immediately eliminated two of the most complicating and debilitating features of early analytical IMS drift-tube designs: 1. The formation of ion clusters in the drift region between product ions and the sample neutrals was eliminated. This meant that mobility spectra now became truly reproducible and independent of concentration and ended the complications highlighted by Keller et al.63–65 The freedom from cluster reactions allowed spectra to be understood in terms of gas-phase ion-molecule reaction chemistry and ion behavior at ambient pressure. This laid the foundation of a modern analytical method based on physical chemistry models and descriptions. 2. Memory effects seen in the early generation of drift tubes were reduced. Low residence times for vapors meant that IMS drift tubes could be used for comprehensive characterization of complex mixtures using a preseparation inlet such as a gas chromatograph. This has augmented the usefulness of IMS in the same way that GC/ MS enhanced the utility of mass spectrometers. The ramifications of these two improvements helped define the modern period of IMS development with further advances in instrumentation and applications.

1.4 Modern Analytical IMS The era of modern IMS is defined by the sum of advances in instrumentation, improvements in the scientific understanding of IMS, increase in the number of publications and the volume of international communications, and growth

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_25.... 05/05/2007

Página 1 de 1

Page 26 in the volume and number of IMS applications. These were not necessarily related or coordinated but occurred during the 1990s, imparting a net cumulative momentum of change in the perception of IMS and improving the prospects of future development. As might be expected from a dynamic field, the world of IMS is far more diverse than a decade ago.1 A selective and incomplete listing of some of these facets that define modern IMS are listed with annotation in the following text. In each of the categories below, refinements and further activity are on-going in 2005 showing IMS to be a dynamic and maturing analytical method.

1.4.1 Chemical Warfare Agents Detection The CAM underwent testing and deployment throughout the 1980s, and these handheld analyzers were used extensively during combat in the 1991 Persian Gulf War. When under threat in Kuwait and elsewhere, allied troops needed technology for chemical agent monitoring, and CAM provided such capabilities in a rugged and reliable package. This was the first widespread use of IMS, and over 60,000 CAMs can be found in military establishments worldwide today. This represented an unprecedented acceptance and deployment of sophisticated chemical analysis technology and continues with the recent completion of ACADA and a mini-IMS detector that has a nonradioactive source and size smaller than CAM. The U.N. inspectors, looking for chemical weapons production and storage sites in Iraq after the Gulf War, made extensive use of handheld CAMs. The original CAM design has undergone several improvements and has been developed into several derivative instruments as described in Chapter 4 and Chapter 6.

1.4.2 Explosives Detection The use of IMS analyzers in detecting explosives in hand-carried luggage in commercial aviation is the first visible use of IMS outside military venues, though still under supervision of governmental organizations. Today, IMS explosive analyzers are seen at security checkpoints in airports worldwide with more than 15,000 analyzers used for over a million measurements per year. The Smiths Detection IONSCAN 200 and 400 and the GE Interlogix Itemizer can be seen in operation daily at airports. As with the development and use of CAM, IMS gained recognition as a measurement technique within the greater community of analytical scientists from this application. The chemistry of and instrumentation for explosives monitoring are described in Chapter 4 and Chapter 6.

1.4.3 Drug Detection Mobility spectrometers exhibit favorable response toward nitrogen compounds including narcotics such as cocaine and heroin. Consequently, IMS

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_26.... 05/05/2007

Página 1 de 1

Page 27 analyzers are used by customs, police, and drug enforcement agencies in the U.S. and elsewhere. This application is neither as widely used nor as visible as explosives monitoring, though the instruments used are similar in both applications.

1.4.4 Ion Mobility Spectrometry—The Book1 The release of an integrated monograph on IMS by CRC Press in late 1993 (with a 1994 copyright) was helpful as a collection of most facets of analytical IMS. The book was used in courtrooms to support the acceptance of IMS technology in drug analyses and was available as a survey for those interested in learning about IMS along with reading the original literature.

1.4.5 Field Asymmetric IMS When the last edition of this book on IMS was written, little was known about IMS programs in the former Soviet Union apart from a vague appreciation that IMS development had occurred in the former East Germany. In 1991–93, manuscripts from a team in Tashkent and Novosibirsk reported an approach to mobility determination unlike any used in the West.88,89 This method, originally called drift spectrometry and later known as field asymmetric ion mobility spectrometry (FAIMS), is based on the nonlinear dependence of mobility in electric fields (as seen in Figure 1–4). Ions are characterized in an asymmetric electric field in which voltages may reach 20,000 V/cm for nanoseconds, and ion motion is governed by differences in drift velocities and differences in mobility at both low and high electric fields. This method provides a type of mobility filter rather than a dispersion spectrometer and was initially promoted in the U.S. by the Mine Safety Appliance (MSA) Corporation and later by a team at the National Research Council in Ottawa, Canada. Eventually this team formed a company, Ionanalytics.90 Microfabricated analyzers the size of postage stamps were created with comparable principles, demonstrating the innovation and possible new advances in IMS using mobility principles.91,92 The microfabricated analyzer, or differential mobility spectrometer, is now commercially available from Sionex Corp.93

1.4.6 The IMS Society and International Conferences Before 1992, investigators and researchers in mobility spectrometry had never met as a whole, and exchanges of information were made only in a few symposia or small discussion groups. These included the Federation of Analytical Chemistry and Spectroscopy Societies (FACSS) meeting in Ana-heim (1990), CA, a symposium at a joint meeting of the American Chemical Society and the Chemical Society of Canada in Toronto (1988), and a small

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_27.... 05/05/2007

Página 1 de 1

Page 28 TABLE 1–4 Dates and Locations of the Annual Meetings of the International Society for Ion Mobility Spectrometry

Year

Location

No. of Participantsa

1992

Mescalaro, NM

34

1993

Quebec City, Canada

49

1994

Galveston Island, TX

66

1995

Cambridge, U.K.

57

1996

Jackson Hole, WY

54

1997

Bastei National Park, Germany

67

1998

Hilton Head Island, SC

78

1999

Buxton, U.K.

78

2000

Halifax, Nova Scotia, Canada

94

2001

Wernigerode, Harz, Germany

82

2002

San Antonio, TX

90

2003

Umeå, Sweden

92

2004

Gatlinburg, TN

93

2005

Chantilly, France

–

Note: Numbers are taken from the registration list in the meeting booklet and may not include late or on-site registrations.

meeting in Park City, UT (1991). In short, few opportunities existed for investigators in government and academic and industrial laboratories to exchange results and suggestions free of the constraints found in large conferences. In 1992, a four-day workshop organized by the U.S. Army was held in Mescalaro, NM, for discussions among all those known as IMS investigators.94 Though international in intention, the newly reorganized former Soviet states, including the former East Germany, could not be invited in time for the meeting. This was remedied the following year in Quebec City, and the gathering became a self-organized annual conference. The meeting has been successful at forging a forum for users and researchers and served to launch a scholarly society, the International Society for Ion Mobility Spectrometry (IS-IMS). These meetings have facilitated exchange of technology, practices, and experimental discoveries in ways that could not have been possible previously. A listing of the IS-IMS conferences, dates, and locations is given in Table 1–4.

1.4.7 The Volatile Organic Analyzer In 1990, only one team in the world (H.H.Hill and coworkers at Washington State University) had instrumentation for high-resolution capillary GC with an IMS detector. In the early 1990s, NASA selected GC/IMS as a technology for monitoring airborne volatile organic compounds in the International Space Station. The instrument was called the volatile organic analyzer (VOA).95 This investment in GC/IMS via VOA promoted the development of high-temperature drift tubes, stimulated others to couple gas chromatographs to mobility spectrometers, and demonstrated a commitment to this

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_28.... 05/05/2007

Página 1 de 1

Page 29 technology with an unparalleled development program. Today, the VOA is on-board the International Space Station and has demonstrated the high stability of GC/IMS.

1.4.8 Biological Applications of Mobility Spectrometry One innovative application of mobility measurements to supplement mass determinations has occurred with biomolecules in which the shape and size of proteins can be probed through measurements of mobility.96–98 Folded or compact configurations of proteins in the form of ions exhibit small cross sections and high mobilities compared with unfolded or elongated structures of the same substance. In these studies, drift tubes have been operated at reduced pressure, not unlike the drift tubes from proto-IMS analyzers.47,48 These studies have been made possible largely by electrospray ionization methods and may parallel the revolution that occurred in bio-mass spectrometry in the 1990s.

1.5 Present and Future Trends in IMS Certain trends in IMS science and technology may be seen as active lines of investigation currently, whereas others may portend future developments in IMS. One indication of these is the themes that recur in annual presentations at the IS-IMS meetings and in research articles. The following discussion is not comprehensive and lists some of the present developments in IMS.

1.5.1 Spectral Libraries and Standardization of Mobility Spectrometers A challenge for users and investigators of mobility spectrometry through the decades has been the absence of spectral libraries for reference to experimental findings. This was more complicated than may initially be expected because there was no agreement within the IMS community over standard operating conditions or even on the importance of various experimental parameters. There was no agreement (and perhaps there should be none) on common conditions for temperature, moisture in the drift gas, electric field strength, reactant-ion chemistry, and method of ionization. However, the importance of these parameters was clarified during the 1990s, and general agreement exists on the need to document these parameters. An early attempt to create a well-documented spectral library was limited to a single drift-tube temperature and moisture.99 Subsequently, a library was prepared for various temperatures and moistures.2,100 This latter effort also was configured in JCAMP-DX format, which was advocated by

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_29.... 05/05/2007

Página 1 de 1

Page 30 IUPAC101,102 and can be found at their Website.103 Modifications of this format can be anticipated with changes in computational capabilities. In recent years, a chemical standard for calibration of drift tubes has been proposed,104 which may presage the next stage of intercomparison of spectra from mobility measurements. One library is included in the CD-ROM included with this book.

1.5.2 Instrumentation During the past 10 years (since the publication of Ion Mobility Spectrometry), little progress has been made on availability of commercial drift tubes for users or prospective users. Nor is there an affordable traditional drift tube for either specific research studies (GC/IMS or IMS/MS) or for general use (air monitoring or analyses of samples of all phases). The absence of affordable commercial instrumentation is noteworthy and may be a limiting variable in any further broad acceptance of IMS for analytical measurements. Consequently, electronic schematics, engineering specifications for drift tubes, and software for Labview and National Instruments interface and can be found in the CD accompanying this book. One current trend in IMS instrumentation is the miniaturization of conventional drift tube or microfabrication of planar drift tubes for differential mobility spectrometry. This development began in the 1990s, and the instruments have been commercialized by small companies during the past few years. A small conventional ion mobility spectrometer has been developed and marketed by G.A.S. (Gesellschaft für analytische Sensorsysteme, mbH) in Dortmund, Germany,105 and a microfabricated differential mobility spectrometer has been commercialized by Sionex Corporation in Waltham, MA93 (see Table 1–5). Nonradioactive ion sources, long desired as replacements for 63Ni, are now past the conceptual stage and are used in the new miniaturized military analyzers from Smiths Detection, called the lightweight chemical detector (LCD).106 In the LCD, a pulsed corona discharge has been used as a substitute for the radioactive 63Ni ion source. Another emerging feature of IMS drift tubes, which might have short-term and long-term impact on IMS, is a replacement for the Faraday plate. This detector is based on microchannel technology and can multiply ions 104 times at atmospheric pressure.107

1.5.3 Commercial Scene During the past decade, IMS companies have been restructured through consolidation or dissolution, and new small companies have been formed as summarized in Table 1–5. Examples include the fairly rapid consolidation of Graseby Dynamics, Ltd., Barringer Research, and Environmental Technology Group (an early contractor on ACADA) as divisions of one large company (Smiths Detection). Other transfers of ownership and management include the purchase of PCP, Inc. and Molecular Analytics by the SAES Getter Group

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_30.... 05/05/2007

Página 1 de 1

Page 31 TABLE 1–5 Summary of Commercial Activity in IMS from 1990 to 2005

Company Details

Recent History and Comments

Products

Graseby Dynamics

Now Smiths Detection http://www.smithsdetection.com/

GID-3™—chemical warfare agent detection system LCD—lightweight chemical detector CAM—chemical agent monitor MCAD—manportable chemical agent detector

Barringer Research

Now Smiths Detection http://www.smithsdetection.com/

GC-IONSCAN® IONSCAN® 400B IONSCAN® SENTINEL SABRE 2000

Environmental Technologies Group

Now Smiths Detection http://www.smithsdetection.com/

APD2000

Ion Track, Inc.

Acquired by GE. Now GE-Interlogix http://www.geindustrial.com/geinterlogix/iontrack/index.html

Itemiser EntryScan VaporTracer

PCP, Inc.

Acquired by SAES Getters, Italy http://www.saesgetters.com/ Operations in Florida stopped

Molecular Analytics

Now Trace Analytical as part of SAES Getters, Italy http://www.saesgetters.com/ http://www.traceanalytical.com/

Institut für Umwelttechnologien GmbH (IUT)

Acquired by Dräger Safety Ag http://www.draeger.com http://iut-berlin.de/

Bruker-Daltonics

http://www.bruker-daltonik.de/

RAID-M RAID-E RAID-1 RAID-S

G.A.S (Gesellschaft für analytische Sensorsysteme mbH)

http://www.gas-dortmund.de/

μIMS®-ODOR TEIMS-PortablePM UVIMS-MCC-Portable UVIMS Portable

Ta3000 Ta7000 Ta7000F

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_31.... 05/05/2007

Página 1 de 1

Page 32 Company Details

Recent History and Comments

Products

Q-Scent Ltd.

http://www.matiniop.org.il/newrdinf/company/c4858.htm#general Products for medical diagnostics under development

Sionex Corporation

http://www.sionex.com/ Microfabricated drift tubes for differential mobility spectrometry

microDMx™ sensor chip Varian CP-4900 GCDMD, EGIS Defender

Environics Oy

Aspirator style ion characterization http://www.environics.fi/

ChemPro 100 EnviScreen

Ionanalytics

Began with MSA field ion spectrometer and innovated on original design http://www.ionalytics.com

Selectra

Mine Safety Appliance Company

Operations with field ion spectrometer (FIS) stopped Tried to introduce FIS from Russia to North America

Femtoscan Corporation

Ceased business www.femtoscan.com

EVM. A handheld GC/IMS

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_32.... 05/05/2007

Página 1 de 1

Page 33 of Milan, and the purchase of Ion Track Instruments by General Electric. Noteworthy is the growth in IMS-based companies in Europe, including the formation of G.A.S (Dortmund, Germany), Bruker-Saxonia (Leipzig, Germany), and Environics (Finland). Other changes include Femotscan, Inc. (Utah), which stopped sales of a handheld GC/IMS, the Environmental Vapor Monitor, and ceased operations. An Israeli company, Q-Scent, Ltd., is seeking a presence for IMS in medical testing and is in a nascent stage of growth.108

1.5.4 Applications The past decade has seen applications of IMS in environmental, industrial, and biomedical studies in ways that portend the future of IMS. Only one of these, determining SF6 purity in high-voltage switches, has found commercial acceptance (in the hydroelectric industry).109 The fact that few new applications of IMS have seen widespread commercial acceptance during the past decade may be attributed to competing or existing technologies that adequately meet the measurement requirements. However, the inaccessibiliiy of IMS instrumentation may be limiting the exploration and development of new applications. This question has been part of the history of this technology and seems to be as relevant today as 10 or 20 years ago. A hopeful sign is the structure of this monograph, which contains four chapters on applications of IMS, whereas the previous edition contained just a single chapter. A further sign of recognition of the maturity of the technology is that Analytical Chemistry, probably the leading journal in the field, has dedicated an article in the Apages section to the commercialization of IMS.110

References 1. Eiceman, G.A.; Karpas, Z., Ion Mobility Spectrometry, CRC Press, Boca Raton, FL, 1993. 2. Bell, S.E.; Nazarov, E.G.; Wang, Y.F.; Eiceman, G.A., Classification of ion mobility spectra by chemical moiety using neural networks with whole spectra at various concentrations, Anal Chim. Acta 1999, 394, 121–133. 3. Bell, S.E.; Nazarov, E.G.; Wang, Y.F.; Rodriguez, J.E.; Eiceman, G.A., Neural network recognition of chemical class information in mobility spectra obtained at high temperatures, Anal Chem. 2000, 72, 1192– 1198. 4. Eiceman, G.A.; Nazarov, E.G.; Rodriguez, J.E., Chemical class information in ion mobility spectra at low and elevated temperatures, Anal. Chim. Acta 2001, 433, 53–70. 5. Brosi, A.R.; Borkowski, C.J.; Conn, E.E.; Griess, J.C., Jr., Characteristics of Ni59 and Ni63, Phys. Rev. 1951, 81, 391–395. 6. Siu, K.W.M.; Aue, W.A., 63Ni β range and backscattering in confined geometries, Can. J. Chem. 1987, 65 (5), 1012–1024.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_33.... 05/05/2007

Página 1 de 1

Page 34 7. Kebarle, P.; Searles, S.K.; Zolla, A.; Scarborough, J.; Arshadi, M., The solvation of the hydrogen ion by water molecules in the gas phase. Heat and entropies of solvation of individual reaction: J. Am. Chem. Soc. 1967, 89(25), 6393–6399. 8. Arshadi, M.; Kebarle, P., Hydration of OH− and O2− in the gas phase. Comparative solvation of OH− by water and the hydrogen halides. Effects of acidity, J. Phys. Chem. 1970, 74(7), 1483–1485. 9. Kim, S.H.; Betty K.R.; Karasek, F.W., Mobility behavior and composition of hydrated positive reactant ions in plasma chromatography with nitrogen carrier gas, Anal Chem. 1978, 50(14), 2006–2016. 10. McDaniel, E.W.; Mason, E.A., The Mobility and Diffusion of Ions in Gases, Wiley-Interscience, New York, 1973. 11. Lin, S.N.; Griffin, G.W.; Horning E.C.; Wentworth, W.E., Dependence of poly-atomic ion mobility on ionic size, J. Chem. Phys. 1974, 60, 4994–4999. 12. Revercomb, H.E.; Mason, E.A., Theory of plasma chromatography/gaseous electrophoresis: a review, Anal Chem. 1975, 47, 970–983. 13. Rutherford, E., The velocity and rate of recombination of the ions of gases exposed to Röntgen radiation, Phil Mag. 1897, 44, 422–440. 14. Röntgen, W.C., On a new kind of rays, Nature 1895, 53, 274–276 also see, Science 1896, 3, 726. 15. Curie, M.S., Rayons émis par les composes de l’uranium et du thorium, Comptes Rendus 1898, 126, 1101– 1103. 16. Rutherford, E., Uranium radiation and the electrical conduction produced by it, Phil. Mag. 1899, 47, 109– 163. 17. Chattock, A.R, On the velocity and mass of ions in the electric wind in air, Phil. Mag. 1899, 48, 401–420. 18. Franck, J.; Pohl, R., A method for the determination of the ionic mobility in small gas volumes, Ber. Phys. Ges. 1908, 9, 69–74. 19. McClelland, J.A., On the conductivity of hot gases from flames, Phil. Mag. 1898, 46, 29–35. 20. Rutherford, E., The discharge of electrification by ultra-violet light, Proceedings of the Cambridge Philosophical Society 1898, 9, 401–416. 21. Thomson, J.J.; Rutherford, G.P., Conduction of Electricity Through Gases, Cambridge University Press, Cambridge, U.K., 1928. 22. Tyndall, A.M., The Mobility of Positive Ions in Gases, Cambridge Physical Tracts, Oliphant, M.L.E.; Ratcliffe, J.A., Eds., Cambridge University Press, Cambridge, U.K., 1938. 23. Lattey, R.T.; Tizard, H.T., The velocity of ions in dried gases, Proceedings of the Royal Society of London (A) 1913, 86, 349–357. 24. Lattey, R.T., Effect of small traces of water vapor on the velocities of the ions produced by Röntgen rays in air, Proceedings of the Royal Society of London (A) 1911, 84, 173–181. 25. Langevin, P., L’Ionistion des Gaz, Ann. de Chim. Phys. 1903, 28, 289–384. 26. Langevin, P., Une formule fondamentale de théorie cinétique, Ann. de Chim. et de Phys. 1905, 5, 245–288. 27. Cravath, A.M., The rate of formation of negative ions by electron attachment, Phys. Rev. 1929, 33, 605– 613. 28. van de Graaff, R.J., Mobility of ions in gases, Nature 1929, 124, 10–11. 29. Bradbury, N.E.; Nielsen, R.A., Absolute values of the electron mobility in hydrogen, Phys. Rev. 1936, 49, 388–393.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_34.... 05/05/2007

Página 1 de 1

Page 35 30. Thomson, J.J., Rays of Positive Electricity, Green and Co., London, 1933. 31. Loeb, L.L., Statistical factors in spark discharge mechanisms, Rev. Mod. Phys. 1948, 20, 151–160. 32. http://www.home.aone.net.au/shack_one/balkela.html 33. http://www.probertencyclopaedia.com/F1A.HTM 34. Lovelock, J.E., The electron-capture detector—a personal odyssey, in Electron Capture, Zlatkis, A.; Poole, C.F., Eds., Elsevier, New York, 1981, pp. 13–26. 35. Lovelock, J.E.; Wasilewska, E.M., An ionization anemometer, J. Sci. Instr. 1949, 26, 367–370. 36. Benner, R.L.; Lamb, B., A fast response continuous analyzer for halogenated atmospheric tracers, J. Atmos. Oceanic Tech. 1985, 2, 582–589. 37. Ausloos, S.P., Ed., Ion-Molecule Reactions in the Gas Phase, Advances in Chemistry Series No. 58, American Chemical Society, Washington, D.C., 1966. 38. Knewstubb, P.F.; Sugden, T.M., Mass-spectrometric observations of ions in hydrocarbon flames, Nature 1958, 181, 474. 39. Shahin, M.M., Mass-spectrometric studies of corona discharges in air at atmospheric pressures, J. Chem. Phys. 1966, 45(7), 2600–2605. 40. Shahin, M.M., Ion-molecule interaction in the cathode region of a glow discharge, J. Chem. Phys. 1965, 43, 1798–1805. 41. Shahin, M.M., Use of corona discharges for the study of ion-molecule reactions, J. Chem. Phys. 1967, 47 (11), 4392–4398. 42. Munson, M.S.B.; Field, F., Chemical ionization mass spectrometry. I. General introduction, J. Am. Chem. Soc. 1966, 88, 2621–2630. 43. Kebarle, P.; Hogg, A.M., Mass-spectrometric study of ions at near atmospheric pressures. I. The ionic polymerization of ethylene, J. Chem. Phys. 1965, 42(2), 668–674. 44. Good, A.; Durden, D.A.; Kebarle, P, Ion-molecule reactions in pure nitrogen and nitrogen containing traces of water at total pressures 0.5–4 torr. Kinetics of clustering reactions forming H+(H2O)n, J. Chem. Phys. 1970, 52, 212–221. 45. Mason, E.A.; Schamp, H.W., Jr., Mobility of gaseous ions in weak electric fields, Ann. Phys. (N.Y.) 1958, 4, 233–270. 46. McDaniel, E.W., Collisional Phenomena in Ionized Gases, John Wiley & Sons, New York, 1964. 47. Albritton, D.L.; Miller, T.M.; Martin, D.W.; McDaniel, E.W., Mobilities of mass-identified H 3+ and H+ ions in hydrogen, Phys. Rev. 1968, 171, 94–102. 48. Crompton, R.W.; Elford, M.T.; Gascoigne, J., Precision measurements of the Townsend energy ration for electron swarms in highly uniform electric fields, Austr. J. Phys. 1965, 18, 409–436. 49. Cohen, M.J., Plasma chromatography—a new dimension for gas chromatography and mass spectrometry, presented at the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, March 1969. 50. Cohen, M.J.; Karasek, F.W., Plasma chromatography™—a new dimension for gas chromatography and mass spectrometry, J. Chromatogr. Sci. 1970, 8, 330–337. 51. Karasek, F.W., The plasma chromatograph, Res. & Dev. 1970, 21, 34–37. 52. Karasek, F.W.; Kilpatrick, W.D.; Cohen, M.J., Qualitative studies in trace constituents by plasma chromatography, Anal. Chem. 1971, 43, 1441–1447. 53. Karasek, F.W., Plasma chromatography of the polychiorinated biphenyls, Anal. Chem. 1971, 43, 1982– 1986.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_35.... 05/05/2007

Página 1 de 1

Page 36 54. Karasek, F.W.; Cohen, M.J.; Carroll, D.I., Trace studies of alcohols in the plasma chromatograph mass spectrometer, J. Chromatogr. Sci. 1971, 9, 390–392. 55. Karasek, F.W.; Keller, R.A., Gas chromatograph/plasma chromatograph interface and its performance in the detection of musk ambrette, J. Chromatogr. Sci. 1972, 10, 626–628. 56. Karasek, F.W.; Tatone, O.S., Plasma chromatography of the mono-halogenated benzenes, Anal Chem. 1972, 44, 1758–1763. 57. Karasek, F.W., Trace analysis and fundamental studies by plasma chromatography, Int. J. Environ. Anal. Chem. 1972, 44, 157–166. 58. Karasek, F.W.; Kane, D.M., Plasma chromatography of the n-alkyl alcohols, J. Chromatogr. Sci. 1972, 10, 673–677. 59. Karasek, F.W.; Tatone, O.S.; Kane, D.M., Study of electron capture behavior of substituted aromatics by plasma chromatography, Anal. Chem. 1973, 45, 1210–1214. 60. Karasek, F.W., The scope of plasma chromatography, Can. Res. Develop. 1973, 6, 19–23. 61. Karasek, F.W.; Tatone, O.S.; Denney, D.W., Plasma chromatography of the n-alkyl halides, J. Chromatogr. 1973, 87, 137–145. 62. Karasek, F.W.; Denney, D.W., Evaluation of the plasma chromatograph as a qualitative detector for liquid chromatography, Anal. Lett. 1973, 11, 993–1004. 63. Bird, G.M.; Keller, R.A., Vapor concentration dependence of plasmagrams, J. Chromatogr. Sci. 1976, 14, 574–577. 64. Keller, R.A.; Metro, M.M., Evaluation of the plasma chromatograph as a separator-identifier, J. Chromatogr. Sci. 1974, 12(11), 673–677. 65. Metro, M.M.; Keller, R.A., Plasma chromatograph as a separation-identification technique, Separ. Sci. 1974, 9(6), 521–539. 66. Griffin, G.W.; Dzidic, I.; Carroll, D.I.; Stillwell, R.N.; Horning, E.C., Ion mass assignments based on mobility measurements validity of plasma chromatographic mass mobility correlations, Anal. Chem. 1973, 24,1204–1209. 67. Horning, E.C.; Horning, M.G.; Carroll, D.I.; Dzidic, I.; Stillwell, R.N., A new pictogram detection system based a on mass spectrometer with an external ionization source at atmospheric pressure, Anal. Chem. 1973, 45, 936–943. 68. Carr, T.W., Analysis of Polymer Outgassing as Studied by Plasma Chromatography-Mass Spectroscopy, NBS Special Publication (US) 1979, 519, 697–703. 69. Dam, R., Monitoring of toxic vapors, in Plasma Chromatography, Carr, T.W., Ed., Plenum Press, New York, 1984, 177–213. 70. http://www.gulflink.osd.mil/m8alalarms/ 71. Blyth, D.A., A vapour monitor for detection and contamination control, in Proceedings of the 2nd International Symposium on Protection Against Chemical Warfare Agents, Stockholm, Sweden, June 17–19, 1983, pp. 65–69. 72. Boscher, J; von Roedern, C.G., Miniaturized ionization detector system, in Proceedings of the 2nd International Symposium on Protection Against Chemical Warfare Agents, Stockholm, Sweden, June 17–19, 1983, pp. 157–164. 73. Tammet, H., The Aspiration Method for the Determination of Atmospheric-Ion Spectra, Israel Program for Scientific Translations Ltd., Jerusalem, 1970. also see http://ael.physic.ut.ee/tammet/am/ 74. http://www.nbcindustrygroup.com/environics.htm 75. http://www.jpeocbd.osd.mil/acada.htm 76. http://63.89.158.169/products/Default.asp?Product=7

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_36.... 05/05/2007

Página 1 de 1

Page 37 77. Kilpatrick, W.D., Plasma Chromatography and Dynamite Vapor Detection, Final Report FAA-RD-71–7, Contract DOT-FA71WA-2491, Federal Aviation Administration, Washington, D.C., January 1971, AD903108/9. 78. Spangler, G.E.; Lawless, P.A., Measurement of electron capture rates for chlorobenzene with negative ion plasma chromatography, Anal. Chem. 1978, 50(6), 884–892. 79. Wernlund, R.F.; Cohen, M.J.; Kindel, R.C., The ion mobility spectrometer as an explosive or taggant vapor detector, Proceedings of the New Concept Symposium and Workshop on Detection and Identification of Explosives, Reston, VA, October/November 1978, pp. 185–189. 80. Ewing, R.E.; Ewing, G.J.; Atkinson, D.A.; Eiceman, G.A., A critical review of ion mobility spectrometry for the detection of explosives and explosive related compounds, Talanta, 2001, 54, 515–529. 81. http://63.89.158.169/products/Default.asp?Product=16§ion=Transportation 82. http://www.geindustrial.com/ge-interlogix/iontrack/prod_itemiser.html 83. Wentworth, W.E.; Becker, R.S., Potential method for the determination of electron affinities of molecules: application to some aromatic hydrocarbons, J. Am. Chem. Soc. 1964, 84, 4263–4266. 84. Wentworth, W.E.; Chen, E.; Lovelock, J.E., The pulse-sampling technique for the study of electron attachment phenomena, J. Phys. Chem. 1966, 70, 445–458. 85. Wentworth, W.E.; Chen, E.; Steelhammer, J.C., Determination of electron affinities of radicals and bond dissociation energies by electron-attachment studies at thermal energies—electron affinity of acetate radical, J. Phys. Chem. 1968, 72, 2671–2675. 86. Harrison, A., Chemical Ionization Mass Spectrometry, CRC Press, Boca Raton, FL. 87. Baim, M.A.; Hill, H.H., Jr., Tunable selective detection for capillary gas chromatography by ion mobility monitoring, Anal. Chem. 1982, 54(1), 38–43. 88. Buryakov, I.A.; Krylov, E.V.; Nazarov, E.G.; Rasulev, U.Kh., A new method of separation of multi-atomic ions by mobility at atmospheric pressure using a high-frequency amplitude-asymmetric strong electric field, Int. J. Mass Spectrom. Ion Proc. 1993, 128, 143–148. 89. Buryakov, I.A.; Krylov, E.V.; Makas, A.L.; Nazarov, E.G.; Pervukhin, V.V.; Rasulev, U.Kh., Separation of ions according to their mobility in a strong alternating current electric field, Pis’ma υ Zhurnal Tekhnicheskoi Fiziki 1991, 17(12), 60–65 (in Russian). 90. http://www.ionalytics.com/en/technology/faims.shtml 91. Miller, R.A.; Eiceman, G.A.; Nazarov, E.G., A micro-machined high-field asymmetric waveform-ion mobility spectrometer (FA-IMS), Sensor Actuat. B-Chem. 2000, 67, 300–306. 92. Miller, R.A.; Nazarov, E.G.; Eiceman, G.A.; King, T.A., A MEMS radio-frequency ion mobility spectrometer for chemical agent detection, Sensor Actuat. A. Phys. 2001, 91, 301–312. 93. http://www.sionex.com/ 94. The first meeting was funded by the U.S. Army Chemical Research, Development and Engineering Center, Aberdeen Proving Grounds, MD under Contract No. DAAL03–91–C–0034, TCN Number 92–057 (DO. No. 0127), Scientific Services Program. Also, see Proceedings from the 1992 Workshop on Ion Mobility Spectrometry, Eiceman, GA, January 27, 1993.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_37.... 05/05/2007

Página 1 de 2

Page 38 95. Limero, T.; James, J.; Reese, E.; Trowbridge, J.; Hohmann, R., The Volatile Organic Analyzer (VOA) Aboard the International Space Station, SAE Technical Paper Series 2002–01–2407, 32nd International Conference on Environmental Systems, July 2002. 96. Clemmer, D.E.; Jarrold, M.F., Ion mobility measurements and their applications to clusters and biomolecules, J. Mass Spectrom. 1997, 32, 577–592. 97. Stone, E.G.; Gillig, K.J.; Ruotolo, B.T.; Fuhrer, K.; Gonin, M.; Schultz, A.J.; Russell, D.H., Surfaceinduced dissociation on a MALDI-Ion mobility-orthogonal time-of-flight mass spectrometer: sequencing peptides from an “in-solution” protein digest, Anal Chem. 2001, 73, 2233–2238. 98. Hill, H.H.; Chandler, H.; Asbury, G.R.; Wu, C; Matz, L.M.; Ichiye, T., Charge location on gas phase peptides, Int. J. Mass Spectrom. 2002, 219, 23–37. 99. Shumate, C; St. Louis, R.H.; Hill, H.H. Jr., Table of reduced mobility values from ambient pressure ion mobility spectrometry, J. Chromatogr. 1986, 373, 141–173. 100. Wessel, M.D.; Sutter, J.M.; Jurs, P.C., Prediction of reduced ion mobility constants of organic compounds from molecular structure, Anal Chem. 1996, 68, 4237–4243. 101. Baumbach, J.I.; Davies, A.N.; Lampen, P.; Schmidt, H., JCAMP-DX. A standard format for the exchange of ion mobility spectrometry data, Pure Appl. Chem. 2001, 73 1765–1782. 102. http://wwwchem.uwimona.edu.jm:1104/software/jcampdx.html 103. http://www.chemistry.nmsu.edu/eiceman_research/spect_library.html 104. Eiceman, G.A.; Nazarov, E.G.; Stone, J.A., Chemical standards in ion mobility spectrometry, Anal Chim. Acta 2003, 493, 185–194. 105. http://www.gas-dortmund.de/index.htm 106. http://63.89.158.169/products/Default.asp?Product=19 107. Denson, S.; Denton, B.; Sperline, R.; Rodacy, P; Gresham, C, Ion mobility spectrometry utilizing microFaraday finger array detector technology, Int. J. Ion Mobility Spectrom. 2002, 5–3, 100–103. 108. http://www.matimop.org.il/newrdinf/company/c4858.htm 109. Baumbach, J.I.; Pilzecker, P; Trindade, E.; Meinders, J., On-Site and on-line monitoring of GIS using ion mobility spectrometry for sensing of SF6-Decomposition, Entwurf für Transmission & Distribution, August 02,1999. 110. Miller, S., Meeting News, reports from the Mass Spectrometry in Homeland Security Workshop, Knoxville, TN, Sensitive mass detectors for national security applications, Anal Chem. 2003, 75, 493A.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_38.... 05/05/2007

Página 1 de 1

Page 39

2 Mobility of Ions in the Gas Phase 2.1 Introduction The theory underlying ion mobility spectrometry (IMS) describes the motion of slow ions in gases. As the ion moves through a neutral gas (supporting atmosphere) under the influence of an external electric field, different forces act on it. On the one hand, there are forces due to the resistance encountered by the ion from the gas molecules. These are electrostatic forces as well as forces arising from the geometry (size and structure) of the ion and molecules. On the other hand, a diffusive force arising from a concentration gradient of the ions and the influence of the electric field act to enhance ion motion. Therefore, when trying to analyze the phenomenon of ion mobility, must be taken account for the diffusion and nonelectrostatic interaction between the ion and gas molecules, the electrostatic interactions between the ion and polarizable gas molecules, and the effect of the electric field on ion motion. The foundations for IMS were laid close to the turn of the century by Langevin1 and Townsend2 and were refined by several other investigators. Especially noteworthy are the books written by McDaniel and Mason3 and by Mason and McDaniel,4 both of which deal extensively with the phenomena of ion transport in gases from a theoretical point of view as well as from experimental and practical considerations. A more detailed treatment of ion motion in the drift tube of an ion mobility spectrometer was presented by Mason5 and can also be found in other works, especially by Mason, McDaniel, Viehland, and their coworkers.6–9 It is customary to start discussions on the motion of ions in gases with the theory of diffusion, then to introduce the electric field and its effect on ion motion, and finally to observe the combined effect of the two forces. This approach will be shown below and the connection between theory and observations with IMS will be emphasized. Most of this chapter presents an analysis of the more popular and simple theoretical models that deal with the motion of ions in a weak electric field, mainly by making assumptions on the type of potential that represents the interaction between the ion and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_39.... 05/05/2007

Página 1 de 1

Page 40 gas molecules of the supporting atmosphere. The correlation between experimentally measured mobilities and those calculated according to the different models is discussed. The capabilities and limitations of these models to predict the mobility of ions of different masses, in several drift gases at different temperatures are also discussed. Finally, a separate section is dedicated to the treatment of ion motion under high-field conditions10–18 and the theory underlying devices that rely on a combination of DC and AC electric fields for measurement of ion mobility. This will be presented as the differential mobility spectrometry under conditions of asymmetric electric fields.

2.2 Motion of Slow Ions in Gases In this section, the approach of Mason and McDaniel3–6 has been adopted, and emphasis is given to analysis of the motion of a relatively large poly-atomic ion in a buffer gas at atmospheric pressure in a weak electric field.

2.2.1 Diffusion of Gaseous Ions Dispersion will occur through normal diffusion for a bundle or swarm of ions of a single type with a density of n ions per unit volume in a gas of neutral molecules or in the supporting atmosphere. This will be the only process in action in effect if all of the following conditions are met: there is no temperature gradient, no electric or magnetic fields are present, and the density of the ions is low enough that coulombic repulsion may be neglected. The dispersion of ions will create a concentration gradient, and ions will flow from regions of higher concentration toward regions of lower concentration at a rate that is proportional to the magnitude of the concentration gradient, and according to Fick’s law as shown in Equation 2–1: (2–1) where J is the number of ions flowing through a unit area normal to the direction of gas flow in a unit of time, and D is the proportionality constant. The diffusive force depends on the nature of the ions and the gas and is typical of a given combination of ions and neutral molecules. The vector J may also be written as the product of the velocity of the diffusive flow (v) and the number of ions per unit volume (n) as shown in Equation 2–2 and represents the total charge or electric current carried by the ions: J=vn

(2–2)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_40.... 05/05/2007

Página 1 de 1

Page 41 Fick’s law may be rewritten in the form of Equation 2–3 after Equation 2–1 and Equation 2–2 are combined and rearranged: (2–3) This diffusive flow of ions continues until all the ions are uniformly dispersed in the neutral gas and the concentration gradient becomes zero.

2.2.2 Effect of Electric Field on Ion Motion When an electric field is imposed on the ion swarm in the supporting atmosphere, ion motion will be influenced by this field. In contrast, neutral molecules will barely be affected, if at all, by the electric field, and any effect will depend upon the dipole or quadrupole moments of the gases. An additional factor is the electrostatic interaction between the ion and gas molecules; the ion may attract gas molecules that have a permanent dipole, quadrupole, or higher moments. The electrostatic forces would also lead to ion-induced dipole interactions with the gas molecules, the magnitude of which depends on the polarizability of the gas. The interaction potential used to represent these forces will be discussed in Section 2.3. In this section, only the effects of the imposed electric field on ion motion are considered. If the electric field is weak and uniform, the ion swarm will flow along the field lines so that ion motion will be superimposed on the diffusive motion described earlier. The drift velocity of the ions (vd) will be proportional to the magnitude of the electric field (E) as given in Equation 2–4: vd=KE

(2–4)

The term K, called the mobility coefficient, like the diffusion coefficient D, is unique at a fixed temperature for a given combination of an ion and neutral-gas molecules of the supporting atmosphere. The relation between the diffusion coefficient and weak-field ion mobility shown in Equation 2–5 is known as the Einstein equation and is sometimes called the Nernst-Townsend relationship: K=(eD/kT)

(2–5)

where e is the ion charge, k is the Boltzmann constant, and T is the gas temperature. The mobility coefficient is directly proportional to the diffusion coefficient because both express the resistance of ion motion through the gas atmosphere. If the appropriate units (K in cm2 V−1s−1, D in cm2/sec, and T in Kelvin) for a singly charged ion are substituted in Equation 2–5, the mobility coefficient will be directly proportional to the diffusion coefficient and inversely proportional to gas temperature as shown in Equation 2–5.1: K=11605(D/T)

(2–5.1)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_41.... 05/05/2007

Página 1 de 1

Page 42 This expression is valid only when the electric field does not cause heating of the ions; i.e., the ions are no longer thermalized and retain energy acquired from the field. This condition is normally used in IMS drift tubes in which only a few collisions are needed for the ion to reach thermal equilibrium with the neutral molecules. When the electric field intensity is increased at fixed pressure, the ions acquire an average energy well above their thermal energy.10–18 The ions are no longer thermalized, and the mobility coefficient becomes dependent on the ratio E/N (N is the density of the neutral molecules) and represents the condition in which the ion gains energy from the field in excess of the thermal energy. The diffusive forces are no longer spherically symmetrical, and the Einstein equation no longer holds. However, this limitation should not affect the understanding of ion motion in the conventional or linear IMS drift tube in which thermalized conditions apply to all standard applications.

2.2.3 Effect of Gas Density Until now, the effect of the density of the drift-gas molecules (N) on ion motion has not been considered. The motion of ions in an electric field can be regarded as a kind of spasmodic motion in which the ion is accelerated by the field until it collides with a gas molecule and loses, upon collision, part or all of the acquired momentum. This process is repeated throughout the transit of the ion swarm through the electric field (see Section 2.4.6). Therefore, an increase in the electric field strength will increase the drift velocity as per Equation 2–4; an increase in the neutral-gas density will directly diminish this effect with proportional increases in collision frequency and losses in kinetic energy. Thus, the motion of ions in an electric field is governed by E/N in combination rather than as separate terms. The mobility coefficient will be independent of E/N only if the energy acquired by the ion from the electric field is negligible compared with the thermal energy,4 according to Equation 2–6: (m/M+M/m)eEλ<
(2–6)

where m is the mass of the ion, M is the mass of the neutral-gas molecules, and λ is the mean free path between collisions. Thus, eEλ, is the energy gained by an ion with charge e moving in an electric field E over a distance λ, and (m/M+M/m) describes the efficiency of the elastic energy transfer from the ion to the gas molecules. If the gas is assumed to be ideal, then the mean free path is equal to the inverse of the product of the gas density and the collision cross section Q; i.e., λ=1/NQ, so that Equation 2–6 can be rewritten as: (m/M+M/m)eE<
(2–7)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_42.... 05/05/2007

Página 1 de 1

Page 43 which, after rearrangement, is given in Equation 2–8: E/N<
(2–8)

The units of E/N are given in Townsends, defined as 1 Td=10−17 V cm2. A so-called low-field condition in analytical IMS holds as long as E/N is below approximately 2 Td, and the earlier discussion is a valid, though approximate, representation of the effect an external electric field has on the transport of ions in gases. The relevance of this discussion to IMS is that the ion swarm is moved in the direction of the electric field but the drift velocities are comparatively small due to the high frequency of collisions and small mean free path. Furthermore, normal processes of diffusion within the ion swarm will cause broadening of the peak width while the ions are moved through the drift tube. These are opposing phenomena that limit the resolution obtainable with a drift tube and also act as practical boundaries for the dimensions of IMS drift tubes.

2.3 Models for Ion-Neutral Interactions In addition to the diffusive forces mentioned earlier, ion motion in an IMS drift tube is also affected by the electrostatic interactions between the ion and the gas molecules of the supporting atmosphere. The electron cloud surrounding the neutral gas molecule is polarized by the ion, thus inducing a dipole moment in the neutral molecule. This results in an electrostatic interaction between the ion and the neutral molecule—the ion-induced dipole effect. In addition to this, molecules that have permanent dipole or quadrupole moments will be attracted to the ion through ion-dipole or ionquadrupole interactions. Several models have been proposed to account for the overall effect of these three forces on ion motion, and some of the classical models will be discussed in the following text. The usefulness of the models in predicting the mobility of polyatomic ions in different drift gases will be examined. Two simple models are considered first: the rigid sphere model and the polarization limit model. Next, a more refined yet relatively simple-to-use model is described in which a 12,4 hard-core potential represents the ion-neutral interaction. The more complex threetemperature model19–21 will not be discussed because ions in IMS are regarded as thermalized. This is the one-temperature assumption, in which ion temperature is assumed to be equal to the temperature of the drift gas. The critical test of any model is the capability to reproduce experimentally observed correlations between the mass of an ion and its mobiliiy and between temperature and mobility, as well as to accurately predict the mobility coefficients of ions in different gas atmospheres. Normally, the success of a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_43.... 05/05/2007

Página 1 de 1

Page 44 model is examined with relatively simple systems such as a homologous series of ions. For example, one such series is that in which the only changes arise from the addition of methylene groups in the ion with minimal changes in ion size and internal charge distribution. A complication in this discussion is the limited number of such tests with ions and gases representative of analytical IMS conditions. Most of the early published investigations on models of ion mobility were made by physicists in relatively simple systems, mainly those in which monoatomic ions drift through an inert monoatomic or diatomic neutral gas.15–17,22–24 This is evident in Table 1 given in Appendix 1 of Reference 4. As this chapter is concerned with polyatomic ions drifting through noninert gases, especially air, there are not many detailed experimental and theoretical studies that can be cited as relevant.25–37

2.3.1 Mobility Equations One equation for the mobility coefficient of thermalized ions drifting through a gas atmosphere in an electric field was given by Mason and coworkers,3–6 as shown in Equation 2–9: K=(3e/16N)(2π/μkTeff)1/2[(1+α)ΩD(Teff)]

(2–9)

where e, k, and N are as defined earlier, and μ is the reduced mass of the ion-neutral collision pair defined as μ=mM/(m+M). The term Teff is the effective temperature of the ions, and where the onetemperature approximation is regarded as valid, Teff is equal to the temperature of the drift gas. Alpha (α) is a correction factor, generally less than 0.02 for m>M (see Reference 5, p. 50), and ΩD (Teff) is the collision cross section, which depends on the effective temperature. Mason and coworkers3–6 described the collision cross section using Equation 2–10: ΩD=πrm2Ω(1, 1)*(T*)

(2–10)

where Ω (1, 1)*(T*) is a good approximation of the dimensionless collision integral that depends on the ion-neutral interaction potential and is a function of the dimensionless temperature T*=kT/ε0. The term ε0 is the depth of the minimum in the potential surface, and rm is the position of this minimum as shown in Equation 2–11: ε0=e2αp[3rm4(1−a*)4]

(2–11)

where αp is the polarizability of the drift-gas molecules (not to be confused with α from Equation 2– 9). The term a*=a/rm represents the separation between the center of charge and the center of mass of the ion and is not negligible for

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_44.... 05/05/2007

Página 1 de 1

Page 45 polyatomic ions of complex structure. A unit analysis of Equation 2.9 is given in Appendix A.

2.3.2 The Rigid Sphere Model The collision between an ion and a gas molecule is treated as that of rigid spheres in which the ion is equally likely to be scattered in any direction (in the center of mass coordinate system). Mason and McDaniel4 described the mean ion energy in Equation 2–12: (2–12) where, v and V are the velocities of the ion and neutral molecules, respectively, and vd, m, and M are as defined earlier. The underscore represents the average kinetic energy. The term is simply equal to the thermal energy acquired by the ion (3kT/2). The term is the energy gained by the ion from the external electric field, and represents the random motion of the neutral gas molecules gained from the field. In cases in which a heavy ion drifts through a light gas as the supporting atmosphere (as is common in IMS), this last term may be neglected. The mean relative energy may also be taken as the effective temperature and represents the total random energy of the ions. Thus, contributions arise from the thermal motion and the effect of the external field as shown in Equation 2–13: (2–13) An expression for the mobility coefficient in Equation 2–14 can be obtained from Equation 2–4 and Equation 2–9: K=vd/E=(3e/16N)(2π/μkTeff)1/2[1/QD]

(2–14)

where ΩD is the collision cross section. For an ion in a neutral gas, Equation 2–13 and Equation 2– 14 represent the complete momentum transfer theory for mobility. The preceding is an abbreviation of a rigorous treatment4 and was based on considerations of the conservation of momentum and energy, as well as a few approximations such as the one-temperature assumption. The rigid sphere model as shown in Equation 2–14 qualitatively describes certain aspects of observed mobility measurements, including the following: 1. For a given effective temperature, the mobility coefficient is inversely proportional to the neutralgas density. 2. The drift velocity of the ions (vd), the mobility coefficient, and the effective temperature depend on E/N at a fixed temperature. 3. The symmetry (see Reference 4, p. 140) is given correctly.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_45.... 05/05/2007

Página 1 de 1

Page 46 The rigid sphere model also allows a determination of when “low-field” conditions are no longer valid, i.e., when the velocity of drift is no longer linearly proportional to E/N as described in Equation 2–7 and Equation 2–8. This will occur when the random-field energy is no longer smaller than the thermal energy as shown in Equation 2–15 from Mason and McDaniel:4 E/N<<0.78 [m/(m+M)]1/2QD

(2–15)

where E/N is given in units of Td and QD is given in units of Å2 (10−16 cm2). As the masses of the ion and neutral gas are included in Equation 2–15, what may be low-field conditions for a heavy ion may still be high-field conditions for a light ion (or an electron) in a heavy gas. According to the rigid sphere model, the collision cross section given in Equation 2–10 may be written as per Equation 2–16 as: ΩD=πd2

(2–16)

where d is the sum of the radii of the ion and neutral molecule and can be estimated, at least crudely, from the ion size. The mobility coefficient of an ion is approximately given by Equation 2–17, because in cases in which m>M, the correction term α in Equation 2–9 is less than 0.02 (Reference 4, p. 50) and can be neglected: K=(3e/16N)[2π/μkTeff)]1/2[1/(πd2)]

(2–17)

According to Equation 2–17, the mobility coefficient varies inversely as the square root of the temperature and reduced mass. As will be shown in the following text, this is contradictory to experimental findings under conditions used in analytical IMS. In the next simple model, polarization will be added to the interaction between the ion and the molecule, and it can account for such contradictions.

2.3.3 The Polarization Limit Model If the neutral molecule does not have a permanent dipole or quadrupole moment and if there are no ion-neutral repulsive forces, then the interaction between the ion and the neutral molecule is due solely to the ion-induced dipole interaction. This interaction is a function of the polarizability of the neutral molecule αp. The interaction potential, Vpol, varies as a function of the distance, r (this r is not to be confused with rm from Equation 2–10 and Equation 2–11), between the ion and the neutral molecule, according to Equation 2–18: Vpol(r)=ε2αp/(2r4)

(2–18)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_46.... 05/05/2007

Página 1 de 1

Page 47 The collision cross section is proportional to the expression [ε2αp/(kT)]1/2, and all mobility coefficients approach a common limit as the temperature approaches 0 K. This is dependent on the polarizability and is, therefore, called the polarization limit (Kpol) per Equation 2–19: (2–19) The number 13.853 is obtained for Kpol when αp is in units of Å3, m and M are in Daltons, and K has units of cm2 V−1sec−1 at 273 K and 760 torr. When the mass of the ion is much larger than the mass of the neutral molecule, 1/m in the reduced mass term is negligible compared to 1/M, so that the mobility is essentially independent of the ion mass. This contradicts physical intuition as well as experimental observations. In summary, the polarization limit model provides a poor description of several empirical observations in IMS.

2.3.4 The 12,4 Hard-Core Potential Model In the next refinement of mobility theory, repulsive and attractive potentials are combined to describe the ion-molecule interaction. As the ion and neutral molecule approach each other at short ranges, repulsive and attractive interactions develop simultaneously. When a short-range repulsive term is added to the polarization limit potential of Equation 2–18, the interaction potential is modified to an (n,4) potential. Choosing different values for n will change the steepness of the repulsion as well as the width and depth of the potential well. An example of this is given by Viehland36a,b for n=8, 12, and 16 (Reference 4, p. 248). Naturally, attractive forces will lead to increases in ion-neutral interaction and, therefore, to increases in the resistance of the neutral gas to ion motion; correspondingly, mobility will be decreased. Repulsive forces will lead to the opposite result. The type of interaction potential used in most cases is the 12,4 or the so called hard-core potential as given in Equation 2–20: V(r)=(ε0/2) {[(rm−a)/(r–a)]12−3[(rm–a)/(r−a)]4}

(2–20)

where r is the distance between the ion and neutral drift-gas molecule (see Equation 2–18), and all other parameters are as defined previously. As mentioned in the preceding text, in large polyatomic ions, the center of mass and center of charge do not necessarily coincide, and therefore a in Equation 2–20 cannot be neglected. Slightly more sophisticated is the model in which a third interaction potential is found in the 12,6,4 hard-core potential model. This is formulated by adding a further term to the 12,4 potential to include some attractive energy in the form of an r−6 term as shown in Equation 2–21: V(r)={nε0/[n(3+γ)−12(1+γ)]}[(12/n)(1+γ)(rm/r)12 −4γ(rm/r)6−3(1−γ)(rm/r)4]

(2–21)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_47.... 05/05/2007

Página 1 de 1

Page 48 where γ (gamma) is a dimensionless fourth parameter that measures the relative strength of the r−6 and r−4 attraction energies. However, as will be shown in the following text, in most practical cases the 12,4 hard-core potential gives a sufficiently good agreement with experimental data, and these last refinements are often unnecessary.

2.4 Models and Experimental Evidence 2.4.1 Introduction To test the validity of the models described earlier, the experimentally measured reduced mobilities of several ions were compared with coefficients calculated according to the three models. The main features of interest were the correlations of mass with mobility and temperature with mobility; another interesting feature is the effect of the drift gas on mobility coefficients. Six parameters are needed in the modeling: a*, r0, z, polarizability, reduced mass, and temperature. The latter three arise from direct physical measurements. The other parameters (a*, r0, z) are optimized by a fitting procedure to minimize the deviation between calculated and measured mobility constants.37 The values of T* and rm were calculated from a*, r0, and z, and the dimensionless collision cross section, Ω*, was taken from Table 1 in Reference 9. In practice, a discrete value of a* was chosen, and initial values for r0 and z were estimated. The parameters r0 and z were then optimized to obtain a good fit with experimental data points by minimizing the squared sum of deviations, X2, between theory and experiment. Special attention was given to the fact that a broad range of ion masses had to be tested because a reasonable fit could be obtained with most models with a narrow mass range. Amines were chosen as the probe compounds to test the models because of the simple ionization chemistry38 in all the drift gases used in these studies, namely helium, argon, nitrogen, air, carbon dioxide, and sulphur hexafluoride.39

2.4.2 Ion Radii in Homologous Series The radius of the ion strongly affects the distance of approach to the neutral gas molecule, and correspondingly, the interaction parameters. However, ion radii data are not available for most of the large polyatomic ions of interest here, and a homologous series of amines was chosen to get an internally consistent set of data. In such a homologous series of ions, the ion radius is assumed to vary approximately as the cube root of its mass. In the rigid sphere model, this means that the sum of the radii of the ion and the neutral molecule, d, will increase slightly as the chain length and ion mass in the homologous series increase. In the polarization limit model, the ion size is

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_48.... 05/05/2007

Página 1 de 1

Page 49 totally neglected, whereas in the hard-core potential model, rm depends on the ion mass, as shown in Equation 2–22: rm=r0[1+b(m/M)1/3]

(2–22)

where b is a constant representing the relative density of the ion and the neutral molecule, generally taken as unity, and r0 is a constant representing the ion radius for a given homologous series. Because the reduced mass term, μ=(1/m+1/M), is insensitive to small changes in the mass of heavy ions drifting through a light-neutral gas, the ion radius and calculated mobility will not quantitatively follow the experimentally determined values.37 A correction for the effect of changes in ion mass for large ions can be added with a semiempirical correction term, mz. This greatly improves agreement between the calculated and measured values over a broad range of ion masses in a variety of drift gases.37,39–41 The physical significance of this correction term is the underlying assumption that a better representation of the position of the minimum of the interaction potential is obtained by adding a term that depends on the ion mass and accounts for the compressibility of the collision pair. Thus, the dependence that rm has on the ion mass is not just through the cube root of the ratio between the ion and neutral gas masses (see Equation 2–22), but also affected by the z factor mentioned earlier. A modified expression for rm may thus be obtained in Equation 2–23 as: rm=(r0+mz) [1+(m/M)1/3]

(2–23)

Another factor that affects the ion radius and the ion mass arises from the observation that ions drifting in polarizable gases, especially at low temperatures, tend to form clusters with the drift-gas molecules.39–41 Thus, the mass of the ion may be incremented from its original mass, m, by clustering with n neutral molecules, each having a mass of M, such that its effective mass (meff) is altered per Equation 2–24: meff=m+Mn

(2–24)

It should be noted that n here is the average of the number of drift-gas molecules clustered on the core ion drifting through the buffer gas and is not necessarily a natural number. This mass increment will also affect rm obtained in Equation 2–23 when the effective mass is substituted in the equation, giving Equation 2–25: rm=(r0+mz) [1+(meff/M)1/3]

(2–25)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_49.... 05/05/2007

Página 1 de 1

Page 50

2.4.3 Experimental Mobilities in Homologous Series and the Mass-Mobility Correlation Several years ago, Griffin et al.42 tried to demonstrate that the masses of ions in a homologous series may be determined from mobility coefficients (see Figure 1–3 in Chapter 1). In this section, the correlation between mass and mobility will be explored with values obtained experimentally for protonated amines.37–41 Attention will be given to a homologous series of protonated primary and tertiary aliphatic amines and their measured mobilities in air and in helium. The reduced mobility coefficients of these two homologous series of protonated compounds in air and in helium at 250°C are summarized in Table 2–1. A difference can be seen in the behavior of the two homologous series in air and in helium. Whereas in air the mobility of protonated tertiary amines is significantly higher than that of their respective isomeric primary amines, in helium there is little or no difference between isomeric ions, thus confirming some of the assumptions made earlier regarding the ionneutral interaction. In helium, in which this interaction is weak due to its low polarizability, the protonated primary and tertiary ions are affected similarly (and weakly) by the drift-gas molecules. On the other hand, in air, the neutral molecules, especially nitrogen, cluster preferentially with the localized charge on the primary amines, leading to larger electrostatic TABLE 2–1 The Experimentally Measured Reduced Mobilities of Protonated Normal Primary Amines and Tertiary Amines in Air and in Helium at 250°C

No. of Carbon Atoms

Primary Amines

Tertiary Amines

CH3(CH2)n−1NH2

[CH3(CH2)n]3N

Air

He

Air

He

1

2.65

–

–

–

2

2.38

12.1

–

–

3

2.20

10.1

2.36

10.3

4

1.98

9.1

–

–

5

1.85

7.8

–

–

6

1.72

7.2

1.95

7.4

7

1.61

7.1

–

–

8

1.50

6.4

–

–

9

1.42

5.9

1.62

5.7

10

1.35

5.6

–

–

11

1.28

–

–

–

12

1.23

5.1

1.35

4.8

14

–

4.1

1.19

3.9

15

–

–

–

–

16

–

3.8

–

–

21

–

–

0.95

3.0

24

–

–

0.85

2.7

36

–

–

0.64

2.0

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_50.... 05/05/2007

Página 1 de 1

Page 51 interactions, rather than with the delocalized and shielded charge of the tertiary amine ions. Thus, a difference in the mobility in air of isomeric ions should arise, as is, indeed, observed. To study massmobility correlations, Equation 2–9 is commonly rewritten by substitution of the appropriate constants and units, and use the inverse reduced mobility,3–6 as shown in Equation 2–26: Ko−1=1.697×10−4(μeffTeff)1/2rm2[Ω(1, 1)*(T*)]

(2–26)

The factor 1.697×10−4 arises from substituting the appropriate units of Ko cm2 V−1s−1 when the terms are masses in daltons, rm in angstroms, and T in kelvin. In Figure 2–1, experimental inverse mobility coefficients of protonated acetyl compounds in air (circles) are shown.37 The compounds do not belong to a homologous series, as some are aliphatic and some aromatic. The corresponding values were calculated through the fitting procedure described in Appendix A. For the rigid sphere model, the calculation was made with r0=2.60 Å (curve a), curve b was obtained for the polarization limit model, and curve c was obtained for the model employing the 12,4 hard-core potential with parameters r0=2.20 Å, a*=0.2, and z=0.0013 Å/amu. Evidently, the rigid sphere model can reproduce the experimental mobility results only over a very limited mass range even with the appropriate adjustment of r0. The polarization limit dramatically overestimates the reduced mobility over

FIGURE 2–1 The measured inverse mobility of protonated acetyl compounds in air at 200°C as a function of ion mass. Curve a was calculated according to the rigid sphere model with r0=2.60 Å; curve b according to the polarization limit model; curve c according to the hard-core model with a*= 0.2, z=0 Å/amu, and r0=2.40 Å; curve d with a*=0.2, z=0.0013 Å/amu and r0=2.20 Å.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_51.... 05/05/2007

Página 1 de 1

Page 52 the entire range of ion masses and totally fails to reproduce the dependence of mobility on ion mass. On the other hand, the model employing the 12,4 hard-core potential is an acceptable model. By choosing the best-fit parameters for a*, r0, and z, an excellent fit with experimental results was obtained over the entire mass range, in no case deviating by more than 3%. This is even more impressive mindful that the acetyl compounds were not a homologous series.37 In conclusion, the capability, or inability, of each of the three models to account for the correlation between ion mass and reduced mobility has been demonstrated earlier. The capability of each model to account for the dependence of the reduced ion temperature is examined in the next section.

2.4.4 Experimental Mobilities in Homologous Series and the TemperatureMobility Correlation The effect of temperature on the mobility coefficients of ions has been the subject of several experimental studies and much controversy in the literature. The understanding that supports the detailed discussions below is the one-temperature assumption where the effective temperature of an ion is the same as the gas temperature. Thus, heating of the ions by the electric field is negligible. According to the polarization limit model (Section 2.3.3), at the limit T→0, there should be no dependence of Ko on temperature, whereas according to the rigid sphere model (Section 3.2), values for Ko should be proportional to T−1/2. In the model that employs a 12,4 hard-core potential to represent the interaction of the ions with the drift-gas molecules,3–9 the effect of temperature on the Ko of an ion drifting through a buffer gas in an electric field is complex. It depends directly on the temperature through T−1/2, indirectly through ΩD, and even more subtly through meff, which reflects the extent of clustering and is a temperature-dependent phenomenon.40–41 Therefore, analysis of the temperature dependence of the mobility must include the interplay between all these terms. The physical parameters to which Ko is proportional are the collision cross section, the reduced mass, and the temperature. The effect of each of these parameters on mobility will be analyzed using the experimental data. In a simple system of helium, in which clustering and associated effects on mobility are negligible, favorable correlation was obtained between temperature and Ko.39 In contrast, drift gases such as air, CO2, and SF6 exhibited complex behavior. The temperature dependence of the cross section (ΩD), components (rm and Ω*), and effective mass (meff) was examined. This was done for a series of ions with masses varying from 46 to 522 amu. 2.4.4.1 Experimental Results The reduced mobilities of protonated aliphatic and aromatic amines in helium was determined at temperatures of 87, 200, and 224°C.39 These results are compared with reduced mobilities of these or similar ions in air,38 carbon

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_52.... 05/05/2007

Página 1 de 1

Page 53 dioxide40,41 and sulfur hexafluoride39 at temperatures of approximately 90, 150, and 250°C. As the results summarized in Table 2–2 illustrate, the Ko values in helium exhibited a negative temperature dependence, i.e., Ko decreased as the temperature was raised, even for ions with masses as low as 60 amu. In air, the temperature dependence is positive for ions with masses below 90 amu—Ko was increased by about 2% for each 50°C—but negative for ions with masses above 200 amu. Ions of intermediate mass showed almost no temperature dependence in the temperature range of this study.39 In CO2, the temperature dependence is pronounced,40 increasing in some cases, such as pyridine, by 11% between 154 and 250°C, although the biggest change was observed when the temperature was increased from 90 to 154°C. However, very large aliphatic ions, such as protonated trioctylamine (354 amu), show little or no temperature dependence in CO2. In SF6, a positive temperature dependence was also observed, although not as pronounced as in CO2. In Table 2–3, the parameters are shown for the optimizations that gave the best fit between experimental and calculated Ko. From these experimental results, the effect of temperature, drift gas, and clustering on the collision cross section and the reduced mass and therefore on ion mobility can be isolated, and are described in the next two sections. 2.4.4.2 The Effect of Temperature and Drift Gas The reduced mass depends on the masses of the ion and drift gas, and should not depend on the temperature or the polarizability of the drift gas. However, the mass of the ion may be increased through clustering processes, and this is influenced by the gas temperature and polarizability of the drift gas. As shown in Table 2–3, helium does not cluster with ions at ambient pressure at 87°C. In air, the average number of clustered molecules, n, drops from two thirds at 87°C to zero at 215°C In CO2, a steep drop occurs from about 3 molecules at 90°C to zero at 245°C, whereas in SF6 the number of clustered molecules is surprisingly small (n=0.36) at 91°C and drops gradually until reaching zero (by extrapolation) at 335°C Clustering in CO2 is larger than in SF6 below 220°C However, when the effect of clustering on the reduced mass is considered, the molecular weight of the drift gas and the mass of the core ion must be included with values for n. Thus, addition of 0.3 SF6 molecules (MM 146 Da) has the same effect as that of one CO2 molecule (MM 44 Da), or 11 helium atoms (MM 4 Da). Furthermore, the addition of a fixed mass to a small ion will affect it relatively more than the same mass increment to a heavy ion, so that the effect of clustering on the reduced mobility depends on the ion mass. The dimensionless collision cross section, Ω*(T*), for each ion in each drift gas at any given temperature was obtained from calculations based on the experimental data and the Tables in Reference 9 or the Appendix of Reference 4. The ion mass is corrected by substituting the value of n obtained from the fitting procedure into the expression for effective ion mass given earlier (Equation 2–24). The value of rm is obtained by substituting the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_53.... 05/05/2007

Página 1 de 1

Page 54 TABLE 2–2 The Reduced Mobility (cm2V−1s−1) of Several Protonated Aliphatic and Aromatic Amines in Helium, Air, CO2, and SF6 at Different Temperatures

Compound

MW

Helium

Air

CO2

SF6

Temp (°C)

Temp (°C)

Temp (°C)

Temp (°C)

87

200 224

87

150 250

90

154 250

91

150 250

Aliphatic Amines Ethylamine

45 11.0 12.1 11.8 2.16 2.25 2.41 1.06 1.24 1.31 0.62 0.68 0.72

i-Propyl

59 10.1 10.4 10.0 2.02 2.16 2.23

– 1.19 1.26

– 0.64 0.64

Trimethyl

59

– 10.6 10.1

–

– 2.38

– 1.19 1.28

– 0.65 0.65

Diethyl

73

–

– 2.11 2.16

– 1.16 1.25

–

9.1

8.7

–

–

Triethyl

101 8.05 7.58

7.4 1.93 1.94 1.95 0.94 1.07 1.11 0.56 0.61 0.64

Tripropyl

143 6.33

Tributyl

185 5.16 4.78 4.61 1.42 1.40 1.35 0.89 0.93 0.95 0.43 0.48 0.49

Tripentyl

227 4.42

– 3.90 1.25

– 1.19

– 0.83 0.85 0.39 0.42 0.44

Triheptyl

311 3.47

– 3.04 1.02

– 0.95

– 0.69 0.69 0.33 0.34 0.36

Trioctyl

353 3.16 2.83 2.73 0.94 0.89 0.85 0.64 0.64 0.62 0.31 0.32 0.33

Tridodecyl

521

– 5.64 1.65

– 1.62

– 2.08 2.00 0.74 0.69 0.64

– 1.04 1.08 0.49 0.53 0.56

– 0.50 0.48

– 0.25 0.25

Aromatic Amines Pyridine

79 10.3 10.1

9.7 2.08 2.18 2.24 1.07 1.16 1.29 0.61 0.67 0.69

2-Picoline

93 9.35

4-Picoline

93 9.35 8.87 8.54 2.02 2.08 2.08

– 8.54 2.04

– 2.10 1.08 1.18 1.26 0.58 0.62 0.66 – 1.15 1.24 0.58 0.63 0.65

2,4-Lutidine

107 8.42 7.92 7.63 1.93 1.95 1.95 1.07 1.15 1.21 0.56 0.59 0.62

Collidine

121 7.63 7.13 6.89 1.83 1.83 1.81 1.06 1.13 1.17 0.54 0.57 0.59

Quinoline

129 8.05 7.54 7.31 1.78 1.85 1.84 1.02 1.10 1.15 0.52 0.52 0.57

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_54.... 05/05/2007

Página 1 de 1

Page 55 TABLE 2.3 The Parameters of the Fitting Optimization for Helium, Air, CO2, and SF 6

Temp Drift Gas

(°C)

a*

r0

z

n

(Å)

(Å/amu)

(No. of molecules)

Helium

87

0.2

1.33

0.00173 0

Helium

200

0.2

1.27

0.00169 0

Helium

224

0.2

1.32

0.00168 0

Air

87

0.2

2.39

0.00185 0.67

Air

150

0.2

2.37

0.00200 0.17

Air

200

0.2

2.29

0.00219 0

Air

250

0.2

2.23

0.00229 0

CO2

90

0.3

2.17

0.00203 3.4

CO2

154

0.3

2.58

0.00184 0.5

CO2

250

0.3

2.45

0.00204 0.1

SF6

91

0.3

3.57

0.00329 0.36

SF6

150

0.3

3.65

0.00233 0.21

SF6

250

0.3

3.23

0.00271 0.14

optimized fitting parameters r0, z, and meff in Equation 2–25. The collision cross section is then calculated from Equation 2–10. The values thus obtained for rm and Ω*(T*) are summarized in Table 2–4. To normalize the effect of the temperature, the product ΩDT1/2 is shown in the last column of Table 2–4. The value of rm increases with ion mass and with the size of the drift-gas molecules, although for the heavy ions, rm is almost independent of the type of drift gas. On the other hand, rm decreases with temperature, except in helium, in which rm is almost constant for a given ion. The dimensionless collision cross section increases with temperature, except in helium, and decreases with ion mass. For a given ion, Ω*(T*) is much smaller in helium than in the other gases and is almost equal in CO2 and SF6. The value of the product ΩDT1/2 increases with ion mass in all drift gases and also increases with the molecular weight of the drift gas for a given ion. It also increases with the temperature in helium and air but is almost constant in CO2, and actually decreases with temperature in SF6. 2.4.4.3 The Overall Effect of Temperature and Drift Gas on Ko The picture that emerges from these results provides an insight into events occurring with an ion in a supporting atmosphere. For a given ion drifting in helium, rm and Ω*(T*) are almost independent of the temperature, whereas ΩDT1/2 increases with temperature. Clustering does not play a role in helium in these experimental conditions; therefore, values for Ko do not depend on temperature. Thus, the overall effect of raising the temperature

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_55.... 05/05/2007

Página 1 de 2

Page 56 TABLE 2–4 The Calculated Dimensionless Collision Cross Section Ω*(T*), the Distance Parameter rm, and the Product ΩDT1/2 in Helium, Air, CO2, and SF 6 at Different Temperatures for Some Protonated Aliphatic Amines

87°C Ion

rm

(amu)

(Å)

Ω*(T*)

154°C ΩDT1/2

rm

(Å2d1/2)

(Å)

Ω*(T*)

250°C ΩDT1/2

rm

(Å2d1/2)

(Å)

ΩDT1/2 Ω*(T*)

(Å2d1/2)

Helium 46

4.57

0.69

858

4.37

0.68

886

4.51

0.66

939

60

4.96

0.65

952

4.73

0.64

977

4.89

0.63

1054

102

5.93

0.59

1236

5.67

0.58

1273

5.85

0.57

1365

144

6.79

0.55

1510

–

–

–

6.69

0.53

1660

186

7.59

0.52

1784

7.27

0.51

1858

7.47

0.50

1972

228

8.36

0.50

2081

–

–

–

8.22

0.48

2270

312

9.86

0.46

2664

–

–

–

9.69

0.45

2957

354

10.6

0.44

2945

10.2

0.44

3126

10.4

0.43

3261

522

–

–

– 13.0

0.39

4549

13.3

0.39

4929

46

5.81

0.92

1850

5.42

0.97

1840

4.47

1.62

2265

60

6.10

0.87

1928

5.74

0.89

1893

4.76

1.42

2252

102

6.84

0.76

2118

6.55

0.76

2105

5.04

1.08

2303

144

7.50

0.71

2379

–

–

–

6.18

0.92

2459

186

8.10

0.67

2618

7.88

0.66

2646

6.79

0.83

2678

228

8.68

0.64

2872

–

–

–

7.38

0.77

2935

312

9.78

0.60

3419

–

–

–

8.51

0.68

3574

354

10.32

0.58

3680

10.2

0.57

3829

9.07

0.68

3915

522

12.41

0.53

4862

12.4

0.52

5163

11.28

0.62

5486

46

6.77

1.07

2933

5.81

1.30

2847

5.22

1.46

2856

60

–

–

– 6.07

1.20

2868

5.51

1.31

2855

102

7.47

0.92

6.77

1.00

2973

6.26

1.05

2954

144

–

–

– 7.36

0.90

3163

6.88

0.91

3093

186

8.44

0.81

7.92

0.83

3378

7.46

0.84

3356

228

–

–

– 8.45

0.78

3613

8.01

0.79

3639

312

–

–

– 9.46

0.72

4209

9.06

0.72

4243

354

10.3

0.70

9.94

0.70

4487

9.58

0.70

4613

522

–

–

– 11.8

0.64

5866

11.5

0.64

6162

46

7.29

1.19

6.89

1.23

3770

6.03

1.44

3759

60

–

–

– 7.13

1.16

3808

6.27

1.33

3754

Air

CO2

3087 3451

4494

SF6 3783

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_56.... 05/05/2007

Página 2 de 2

102

8.21

0.98

3951

7.77

1.00

3898

6.92

1.11

3816

144

8.85

0.89

4170

8.34

0.91

4087

7.50

0.98

3958

186

9.47

0.83

4453

8.87

0.85

4318

8.04

0.89

4131

228

10.00

0.78

4722

9.38

0.81

4602

8.56

0.84

4419

312

11.2

0.73

5497

10.3

0.75

5178

9.57

0.76

4998

354

11.7

0.71

5904

10.8

0.73

5509

10.1

0.74

5420

522

–

–

– 12.6

0.67

6928

12.0

0.67

6923

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_56.... 05/05/2007

Página 1 de 1

Page 57 for mobility measurements in helium is a decrease in the values for Ko through the T−1/2 term in Equation 2–9. Quantitatively, the decrease is not quite as large as T−1/2, which is indicative of slight temperature dependence on the part of ΩD. In air, rm decreases with temperature more than Ω*(T*) increases, such that the overall effect is a decrease in ΩD. However, due to the T1/2 term, ΩDT1/2 increases with temperature. Clustering also plays a role, with the cluster size decreasing with rising temperature, leading to a small decrease in the reduced mass. Therefore, the overall effect of temperature on Ko of a given ion in air depends on the ion mass. For light ions, the reduction in clustering as the temperature is raised offsets the decrease in mobility due to the increase in ΩDT1/2. On the other hand, for heavy ions drifting through air, Ko is hardly altered by clustering, and the term ΩDT1/2 is dominant, resulting in a decrease in Ko with increasing temperature. In CO2, the decrease in rm and increase in Ω*(T*) combine with the effect of T1/2. Thus, the overall result is that ΩDT1/2 is almost independent of the temperature. Therefore, the observed pronounced increase of Ko with temperature in CO2 is mainly a result of the dramatic decrease in cluster size. In other words, the increase in Ko can be attributed to the change in the effective mass of the ion and the reduced mass μ. In SF6, as in CO2, the overall effect of decreasing rm and increasing Ω*(T*) leads to slight changes in ΩDT1/2, and the observed increase of Ko with temperature is attributed to decreasing cluster size. The dependence of the cluster size on temperature in SF6 is not as strong as in CO2, and therefore the temperature dependence is also weaker. As evident from Table 2–3, the extent of clustering is not directly proportional to the polarizability, and there is likely some dependence also on the shape of the gas molecule. A plausible explanation is that the large, almost spherical SF6 molecules cannot arrange around the ion as compactly as the cylindrical CO2 molecules. In a recent work, the drift time of several compounds in nitrogen was measured over a temperature range of 30 to 250°C.43 The reactant-ion peak served as the reference peak, tR, and a dimensionless “adjusted drift time” (ADT) was defined as (t−tR)/tR, in which t is the drift time of the product ion. The ADT did not change with the electric field in the range 210 to 314 V/cm, but was influenced by the temperature. In summary, Ko values for all ions in a drift gas with low polarizability, in which clustering is negligible, exhibit a negative temperature dependence consistent with the rigid sphere and hard-core models. In a drift gas in which clusters such as CO2 are readily formed, Ko values decrease with decrease in temperature, mainly due to the formation of clusters that increase the effective mass and size of the ion. This effect is pronounced for small ions that exhibit positive dependence of Ko on T. For heavy ions, the small effect from clustering may exhibit an overall increase in Ko with decreasing temperature. For ions of intermediate mass in air, both effects

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_57.... 05/05/2007

Página 1 de 1

Page 58 operate simultaneously on the ion with a near net canceling of any temperature dependence. The polarization limit model does not include a temperature-dependent term, and, therefore, it has no predictive value as far as the effect of temperature on ion mobility is concerned. Drift-gas effects, according to this model, are those that arise from the mass and polarizability of the gas alone. However, as shown, drift gas also plays an important role through its tendency to cluster with the core ion and affect the ion mass. Overall, the polarization limit model overestimates ion mobilities, and as clustering will further lower the measured mobilities, the error would increase for highly polarizable gases. In the rigid sphere model, the temperature dependence of the mobility is through the T−1/2 term, which was shown to hold only for helium. The temperature dependence of the clustering phenomenon is not accounted for by the rigid sphere model. Therefore, this model may be appropriate mainly for predicting the mobility values of ions in helium. The model employing the 12,4 hard-core potential, which represents the electrostatic interaction between the ion and drift-gas molecules, fares much better than the other two models. This is mainly due to the fact that the parameters of the fit in this model are rather flexible and that effects such as clustering, ion shape, etc., may be introduced into the model through its parameters. This brings up the question of the validity and physical significance of the fitting procedure, which is treated in detail in Appendix 1 of this chapter.

2.4.5 A Simple Gedanken Experiment What is the real distance an ion travels from the shutter grid to the detector plate in the drift tube? According to the hard-sphere model, the average velocity of a molecule in air, under conditions of standard temperature and pressure, is about 470 m/sec. At elevated drift-tube temperatures, the velocity would be even larger. On the other hand, for a distance of 5 cm between the shutter grid and the detector plate and a drift time of 5 to 10 msec, the apparent drift velocity of the ion is only 5 to 10 m/sec. Why the difference? The ion does not move in a straight line as it would in a vacuum, but it encounters drift-gas molecules and collides with them. For hard spheres (ion and drift-gas molecules), at standard temperature and pressure conditions, the calculated time between collisions would be 148 psec (148×10−12 sec) and the calculated mean free path between collisions would be 69 nm (69×10−9 m). The resultant path would be like the drunken sailor’s random walk. However, due the influence of the electric field, the net effect is that ion swarms drift along the field lines. Due to the high collision frequency and small free path, the ions are effectively accelerated and decelerated alternately. Based on the ion velocity and drift time, the actual path length is about 50 times larger than the apparent path length (the distance between the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_58.... 05/05/2007

Página 1 de 1

Page 59 shutter and the detector plate). Thus, at atmospheric pressure and room temperature, in the typical low-field conditions of the linear IMS drift tube, the actual distance the ion traverses is roughly 50 times larger than the distance between the shutter grid and the collector.

2.4.6 Resolving Peaks by Changing the Composition of the Drift Gas Ions of different size and shape may exhibit the same drift time or similar drift times in a mobility spectrometer and would appear as an unresolved peak in the mobility spectrum. As the drift time, Ko, reflects the collision cross section or interaction between the ion and molecules of the drift gas, it may be possible to resolve the peaks by changing the drift gas. This approach was examined by Asbury and Hill,44 who measured the drift time of ten different ions, ranging in mass from 92 to 452 Da, in helium, argon, nitrogen, and CO2. The ratio (a) between mobility coefficients for two ions (K01 and K02) was defined by Equation 2–27. α=K01/K02

(2–27)

This should not be confused with definitions employing a described above and is referenced to separation factors used in chromatographic methods. Changing the drift gas may affect the two ions differently, such that their peaks would now be resolved. As an example of the effect the drift gas has on different ions, chloroaniline and iodoaniline, which have quite similar mobility constants in nitrogen, were well resolved in helium and argon (chloroaniline drifts faster) and also in CO2 (iodoaniline drifts faster). If it were only the polarizability of the drift gas or its calculated radius that was responsible for the collision cross section, then nitrogen and argon would show a similar effect, which is clearly not the case. It was concluded that the reduced mass was the primary reason for the mobility spectrum in argon not following the expected trend.44 In general, ions travel faster in drift gases that are less massive and less polarizable. However, these changes may affect different ions in a different manner such that, in principle, unresolved peaks may be resolved by changing the drift gas, as demonstrated earlier. As mentioned earlier and shown in Table 2–1, isomers of primary and tertiary amines that are unresolved in helium due to the similarity in their mobility values (2 to 4% difference) are readily resolved in air (6 to 12% difference). In a way, a similar situation may be encountered in high-field asymmetric waveform ion mobility spectrometers (see discussion in the following section), in which the changes between high-field and low-field conditions may help resolve peaks of ions that have similar mobility constants in a conventional IMS linear-field drift tube. This would be due to the different behavior of the ions in the low-field and high-field parts of the cycle.45

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_59.... 05/05/2007

Página 1 de 1

Page 60

2.5 Dependence of Mobility on Electric Field Most uses of analytical IMS include the understanding or implied condition that the mobility coefficient of an ion is independent of the applied electric field, such that vd=K×E and K≠f(E/N).46 However, Townsend and others recognized in the early 1900s that K is dependent on the energy obtained from the applied electric field by ions between collisions. When E/N is small, energy acquired by the ion from the electric field is considered negligible because collisions with molecules of the supporting gas atmosphere will dissipate any field-acquired energy (i.e., K will not be influenced by the applied electric field at a given pressure). The mobility coefficient will become dependent on the electric field (i.e., K=f(E/N)) with increasing values of E/N as shown in Equation 2–28:15–17,47 K(E/N)=K(0)[1+α2(E/N)2+α4(E/N)4+…α2n(E/N)2n]

(2–28)

where term K(0) is the mobility coefficient under zero field conditions; α2, α4,…α2n are the specific coefficients of even powers of the electric field; and E/N is the electric field normalized for pressure. In Equation 2–28, an even power series in E/N is used due to symmetry considerations (i.e., the absolute value for ion velocity is independent of electric field direction). Ions exhibit patterns of K vs. E/N due to characteristic values of α2n, and Equation 2–28 can be simplified as an alpha function48 to describe the electric field dependence of the coefficient of mobility as per Equation 2–29: K(E/N)=K(0)[1+α(E/N)]

(2–29)

where α(E/N)=α2(E/N)2+α4(E/N)4+…α2n(E/N)2n. This function describes the nonlinear electric field dependence of mobility for an ion. The characterization of mobility dependence has been studied in detail for simple ions and can be found in references such as the Atomic Data of Nuclear Data Tables.15–17,47 However, these measurements were typically made at reduced pressure and often in inert gases. Only a few studies for the measurement of K(E/N) at ambient pressure have been made owing to the technical difficulties associated with creating high fields with conventional time-of-flight drift tubes. For example, an electric field of 21,360 V/cm would be needed to attain values of E/N~80 Td at ambient pressure; this would require a power supply of 106.8 kV for studies with a 5-cm-long drift tube. An alternative approach in IMS technology has enabled studies on field dependence using a field asymmetric ion mobility spectrometer (FAIMS) or differential mobility spectrometer (DMS). The method of high-field asymmetric IMS is a relatively new technique for ion separations based on a nonlinear, high-field dependence of mobility coefficients. The method was proposed in

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_60.... 05/05/2007

Página 1 de 1

Page 61

FIGURE 2–2 Results52 from a differential mobility spectrometer (DMS) show the change in alpha value as a function of the electric field strength (E/N) for a series of ketones with different molecular weights. (A) Protonated monomer shows that alpha increases with the field strength and that the larger the effect, the lower the molecular weight of the ketone; (B) Proton-bound dimers generally show a decrease in alpha with increasing electric field, but low-molecular-weight ketones initially show an increase. The increase in alpha is due to declustering of the ion with increasing field that leads to a reduction in the effective mass of the ion. The effect is more pronounced for protonated monomers due their higher degree of clustering and larger relative change in the effective mass. The numbers to the right of each plot are carbon numbers for each ketone.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_61.... 05/05/2007

Página 1 de 1

Page 62 198249 and experimentally demonstrated in 1993.48 A detailed presentation of this method can be found in Chapter 4 (Section 4.4.2), and the following discussion will be restricted to the dependence of ion mobility on the electric field. Field dependences of K at ambient pressure have been reported for amines,50 chloride,50 positive and negative amino acid ions,51 organophosphates,52 and ketones.53 Results from studies of the field dependence of the ion from ketones are shown in Figure 2–2 as plots of the function alpha (in Equation 2–29) vs. E/N (in Townsends) between 0 to 80 Td. The protonated monomers exhibit a positive dependence of K on E/N (Figure 2–2A) in which the ion at low values for E/N had substantially lower mobility coefficients than that shown by the ion under high E/N values. The core ion was unchanged in these experiments and the trend is contrary to the expected influence of temperature, as in Equation 2–9. Rather, the dependence must be associated with ion size as governed by the formation of clusters between the core ion and small neutrals in the supporting atmosphere. The cross section will be governed by the degree of solvation of the protonated molecule by neutral molecules such as water. At high field, weakly bound molecules will be removed as the effective temperature of the cluster ion increases, while at low field, solvation will increase as the effective temperature is lowered. The same desolvation and solvation processes will occur in each field cycle. Cluster ions such as MH+(H2O)n will, under the low-field portion of the duty cycle, favor clustering and maximum values for n under the experimental conditions of moisture and gas temperature. However, when subjected to the high field portion of the waveform and electric fields of 20,000 V/cm or greater, ion heating should arise and the ion will experience dissociation of hydrates with increase in ion temperature.54 Thus, the basis of α in Equation 2–29 is understood to arise from the declustering and clustering of the core ion as schematically shown in Figure 2–3. In the high-field portion of the waveform, the ion cross section is reduced from declustering. In the low-field portion of the waveform, the ion mobility is decreased due to the formation of clusters. Under this condition, the declustered ion will have a smaller size and lower mass than the clustered ion and, correspondingly, the mobility coefficient will be higher than that of the clustered ion. The processes of cluster formation and declustering are dynamic and continuous throughout ion transport in the drift tube, and the overall effect is a weighted average of the lifetimes and compositions of the clustered and declustered forms of the ion. Thus, the alpha function is gradual and lacks any apparent steps or discontinuous stages. This gradual change in mobility coefficient with E/N has been studied with a large number of organic compounds and observed uniformly regardless of functional group. Returning to the measurement of ketones, as the molecular weight of the ketone increases, effects on the changes in mobility from declustering by ion heating become less pronounced, and the α(E/N) function for ketones with large molar masses nearly flattens (Figure 2–2A). The changes in alpha function with changes in the molecular weight of the ketone are also gradual, with the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_62.... 05/05/2007

Página 1 de 1

Page 63

FIGURE 2–3 A simplified model for positive change in alpha with increasing electric field. During the high-field part of the duty cycle (E 1), ions are declustered, leading to a decrease in Ω (represented by the faint circle around the ion cluster) and an increase in K. At the low-field part of the duty cycle (E 2), ions are reclustered, leading to an increase in Ω and a decrease in K. Consequently, a positive ΔK may be expected with the trends shown in this model.

greatest effect on field dependence observed with the ketone of the lowest mass. The gradual change is also supportive of a general phenomenon rather than a structure- or size-specific process. Nonetheless, the alpha function is small for large ketones and is supported by the model for clusterization-declusterization. An ion of high mass will slightly increase in size and weight with the addition of adducts of water or other small neutrals in the clustered stage, and, therefore, there will be a small difference between the clustered and declustered ions. In contrast, an ion of low mass will undergo significant proportional changes in size and mass with the addition of a few water adducts to the core ion. Between these extremes, intermediate effects should be expected, and were observed in Figure 2–2A. The dependence of mobility coefficients on electric fields for proton-bound dimers is given in Figure 2–2B and shows a negative alpha function that changes with the molar mass of the ketone. As E/N is increased, the mobility of the proton-bound dimer M2H+(H2O)n decreases, indicating that the ion in motion is slower at high electric fields than at low electric fields. This is consistent with decreases in mobility under high-field conditions or increased temperature, according to Equation 2– 17 and Equation 2–26, and presumably occurs through increases in collision frequency with increased temperature. The alpha functions for the proton-bound dimer changed gradually as the mass of the ion was increased, suggesting a weighted average of mobility under conditions of moisture and gas temperature.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_63.... 05/05/2007

Página 1 de 1

Page 64 The results for ketones shown earlier were obtained with air containing 0.1 ppm moisture as the supporting atmosphere. Temperature and pressure were near ambient, and no changes were made in temperature or moisture. However, the cluster model used to explain the positive alpha dependence could be tested with changes in either moisture or temperature. In a second work, the team at New Mexico State University explored the effects of water content on alpha functions with a series of organophosphorous compounds. These compounds were chosen on the basis of high thermal stability and lack of chemical reactivity toward water. Moreover, organophosphorous compounds with the general form of RR2P=O offered a large selection of structures in which R and R were alkyl groups (organophosphates) or in which R was an alkyoxy group (organophosphonates). The results of this study are summarized in Figure 2–4, where alpha values are shown for

FIGURE 2–4

The dependence of the alpha function on moisture for a series of organophosphorous compounds at two different field strengths of 20 Td (A) and 80 Td (B).53 Up to a level of 50 ppm in the drift gas, there is hardly any change in alpha at both field strengths. The changes in mobility between high and low fields for an ion cluster cannot be associated with hydration reactions. Additionally, weakly bound nitrogen molecules are thought to contribute to the effective collision cross section of an ion at low field. This is considered likely owing to the large number of ion-nitrogen collisions in air at ambient pressure.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_64.... 05/05/2007

Página 1 de 1

Page 65 protonated monomers at two E/N of 80 Td (bottom) and 20 Td (top) for moisture levels in air between 0.1 and 10,000 ppm. In these plots, a dependence of alpha on moisture can be divided into two regions with an apparent break in behavior at ~50 ppm moisture in air. Below this value, α(E/N) was not dependent upon moisture though absolute values for alpha were different for each of the protonated phosphates (see discussion on ketones earlier). However, above ~50 ppm, α(E/N) increases monotonically with moisture for ions of each chemical. This same behavior was observed at all values for E/N though only two settings (20 and 80 Td) are shown in Figure 2–4. The earlier-mentioned model for clustering and declustering and the influence of moisture on alpha may be valid only if there is sufficient time during the low-field period for solvent molecules to encounter the ion. At 50 ppm water in air at ambient temperature and pressure there are 1.3×1015 molecules per cm3. If a typical collision rate constant of ~1×10−9 cm3molecule− 1sec−1 is taken for ion-neutral encounters, and knowing that the concentration of ions is far less than that of water, the time between collisions is approximately 1/k[H2O]=0.8 μsec. In the low-field period of ~1.6 μsec, an ion will undergo approximately two collisions with water molecules, a sufficient number if the association reaction is efficient, to change the mass and the cross section of the ion. At elevated water concentrations, the number of collisions in the low-field period will increase, proportionately and reach 400 at 10,000 ppm. With an increase of water concentration, the potential cluster size will increase but the strength of the attachment of each successive water molecule will decrease. The larger clusters will be more susceptible to size diminution in the same high field than are the smaller ones, and hence the value of α will increase with increasing water concentration. The effect of mass of the organophosphorous compounds is also consistent with this model of alpha in which the lower-mass ions show the greatest changes in the values of α as the water concentration is increased. For the ions in this study, lower mass also implies a smaller ion whose cross section will change to a greater extent than will that of a larger ion by the addition of the same number of water molecules. Below ~50 ppm water, an a dependence exists for each ion, and if this change is also due to a change in ion solvation between high and low fields, then the solvating molecule must be either an impurity in the air drift gas or the air drift gas itself. This latter conjecture is consistent with a recent study of the effect of carrier gas type on the high- to low-field mobility of several types of ions. The nitrogen molecule with no dipole moment and a very low polarizability is a very poor solvent for gas-phase ions compared with water. Because an ion will experience a collision with a nitrogen molecule of the drift gas roughly once every 40 psec, an equilibrium solvation can occur in the low-field period (Figure 2–5). Though the associations with nitrogen molecules should be weak at ~19 kJ/mol, the cumulative effect of over 37,000 collisions during the low-field portion of the waveform may account for an increase in effective cross section.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_65.... 05/05/2007

Página 1 de 1

Page 66

FIGURE 2–5 A schematic of the processes that take place in a DMS at high moisture levels.

References 1. Langevin, P., Ann. Chim. Phys. 1905, 5, 245–288. English translation in Collision Phenomena in Ionized Gases (Appendix II), E.W.McDaniel, Ed., John Wiley & Sons, New York, 1984. 2. Townsend, J.S., On the diffusion of ions, Philos. Trans. R. Soc. London A 1899, 193, 129–158. 3. McDaniel, E.W.; Mason, E.A., The Mobility and Diffusion of Ions in Gases, John Wiley & Sons, New York, 1973. 4. Mason, E.A.; McDaniel, E.W., Transport Properties of Ions in Gases, John Wiley & Sons, New York, 1987. 5. Mason, E.A., Ion mobility: its role in plasma chromatography, in Plasma Chromatography, T.W.Carr, Ed., Plenum Press, New York, 1984, 43–93. 6. Revercomb, H.E.; Mason, E.A., Theory of plasma chromatography gaseous electrophoresis—a review, Anal. Chem. 1975, 47, 970–983. 7. Hahn, H.; Mason, E.A., Field dependence of gaseous-ion mobility: theoretical tests of approximate formulas, Phys. Rev. A 1972, 6, 1573–1577. 8. McDaniel, E.W.; Viehland, L.A., The transport of slow ions in gases: experiment, theory and applications, Phys. Rep. 1984, 110, 333–367. 9. Mason, E.A.; O’Hara, H.; Smith, F.J., Mobilities of polyatomic ions in gases: core model, J. Phys. B, At. Molec. Phys. B 1972, 5, 169–176. 10. (a) Wannier, G.H., On the motion of gaseous ions in a strong electric field. I, Phys. Rev. 1951, 83, 281– 289; (b) Wannier, G.H., Motion of gaseous ions in a strong electric field. II, Phys. Rev. 1952, 87, 795–798; (c) Wannier, G.H., Motion of gaseous ions in strong electric fields, Bell Syst. Tech. J. 1953, 32, 170–254. 11. (a) Viehland, L.A.; Kumar, K., Transport coefficient for atomic ions in atomic and diatomic neutral gases, Chem. Phys. 1989, 131, 295–313; (b) Viehland, L.A.;

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_66.... 05/05/2007

Página 1 de 1

Page 67 Robson, R.E., Mean energies of ion swarms drifting and diffusing through neutral gases, Int. J. Mass Spectrom. Ion Proc. 1989, 90, 167–186. 12. Ness, K.F.; Viehland, L.A., Distribution functions and transport coefficients for atomic ions in dilute gases, Chem. Phys. 1990, 148, 255–275. 13. Skullerud, H.R., On the relation between the diffusion and mobility of gaseous ions moving in strong electric fields, J. Phys. B 1976, 9, 535–546. 14. Viehland, L.A.; Mason, E.A., Gaseous ion mobility and diffusion in electric fields of arbitrary strength, Ann. Phys. (N.Y.) 1978, 110, 287–328. 15. Ellis, H.W.; Pai, R.Y.; McDaniel, E.W.; Mason, E.A.; Viehland, L.A., Transport properties of gaseous ions over a wide energy range, At. Nucl. Data Tables 1976, 17, 177–210. 16. Ellis, H.W.; McDaniel, E.W.; Albritton, D.L.; Lin, S.L.; Viehland, L.A.; Mason, E.A., Transport properties of gaseous ions over a wide energy range—Part II, At. Nucl. Data Tables 1978, 22, 179–217. 17. Ellis, E.W.; Thackston, M.G.; McDaniel, E.W.; Mason, E.A., Transport properties of gaseous ions over a wide energy range. Part III, At. Nucl Data Tables 1984, 31, 113–131. 18. Paranjape, B.V., Field dependence of mobility in gases, Phys. Rev. 1980, A21, 405–407. 19. Viehland, L.A.; Lin, S.L., Application of the three-temperature theory of gaseous ion transport, Chem. Phys. 1979, 43, 135–144. 20. Lin, S.L.; Viehland, L.A.; Mason, E.A., Three-temperature theory of gaseous ion transport, Chem. Phys. 1979, 37, 411–424. 21. Waldman, M.; Mason, E.A., Generalized Einstein relations from a three-temperature theory of gaseous ion transport, Chem. Phys. 1981, 58, 121–144. 22. Robson, R.E.; Kumar, K., Mobility and diffusion. II. Dependence on experimental variables and interaction potential for alkali ions in rare gases, Aust. J. Phys. 1973, 26, 187–201. 23. Skullerud, H.R., Mobility, diffusion and interaction potential for potassium ions in argon, J. Phys. B 1973, 6, 918–928. 24. Helm, H., The mobilities of Kr+(2P3/2) and Kr+(2P1/2) in krypton at 295 K, Chem. Phys. Lett. 1975, 36, 97– 99. 25. Viehland, L.A.; Fahey, D.W., The mobilities of NO3−, NO2−, NO+ and Cl− in N2. A measure of inelastic energy loss, J. Chem. Phys. 1983, 78, 435–441. 26. Mason, E.A., Higher approximations for the transport properties of binary gas mixtures, I. General formulas, J. Chem. Phys. 1957, 27, 75–84. 27. Mason, E.A., Higher approximations for the transport properties of binary gas mixtures, II. Applications, J. Chem. Phys. 1957, 27, 782–790. 28. Mason, E.A.; Hahn, H., Ion drift velocities in gaseous mixtures at arbitrary field strengths, Phys. Rev. 1972, A5, 438–441. 29. Mason, E.A.; Schamp, H.W., Jr., Mobility of gaseous ions in weak electric fields, Ann. Phys. (N.Y.) 1958, 4, 233–270. 30. Robson, R.E., Mobility of ions in gas mixtures, Aust. J. Phys. 1973, 26, 203–206. 31. Whealton, J.H.; Mason, E.A.; Robson, R.E., Composition dependence of ion transport coefficients in gas mixtures, Phys. Rev. 1974, A9, 1017–1020. 32. Eisele, F.L.; Perkins, M.D.; McDaniel, E.W.; Mobilities of NO2−, NO3− and CO3− in N2 over the temperature range 217–675 K, J. Chem. Phys. 1980, 73, 2517–2518. 33. Eisele, F.L., Perkins, M.D.; McDaniel, E.W., Measurement of the mobilities of Cl−, NO2·H2O−, NO3·H2O−, CO3·H2O− and CO4·H2O− in N2 as a function of temperature, J. Chem. Phys. 1981, 75, 2473– 2475.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_67.... 05/05/2007

Página 1 de 2

Page 68 34. Holstein, T., Mobilities of positive ions in their parent gases, J. Phys. Chem. 1952, 56, 832–836. 35. Patterson, P.L., Mobilities of negative ions in SF6, J. Chem. Phys. 1970, 53, 696–704. 36. (a) Viehland, L.A.; Mason, E.A., Gaseous ion mobility in electric fields of arbitrary strength, Ann. Phys. N.Y. 1975, 91, 499; (b) Viehland, L.A., Lin, S.L.; Mason, E.A., Kinetic theory of drift-tube experiments with polyatomic species, Chem. Phys. 1981, 54, 341–364. 37. Berant, Z.; Karpas, Z., Mass-mobility correlation of ions in view of new mobility data, J. Am. Chem. Soc. 1989,111, 3819–3824. 38. Karpas, Z., Ion mobility spectrometry of aliphatic and aromatic amines, Anal. Chem. 1989, 61, 684–689. 39. Karpas, Z.; Berant, Z., The effect of the drift gas on the mobility of ions, J. Phys. Chem. 1989, 93, 3021– 3025. 40. Karpas, Z.; Berant, Z.; Shahal, O., The effect of temperature on the mobility of ions, J. Am. Chem. Soc. 1989,111, 6015–6018. 41. Berant, Z.; Karpas, Z.; Shahal, O., The effects of temperature and clustering on mobility of ions in CO2, J. Phys. Chem. 1989, 93, 7529–7532. 42. Griffin, G.W. et al., Ion mass assignments based on mobility measurements, Anal. Chem. 1973, 45, 1204– 1209. 43. Tabrizchi, M., Temperature corrections in the 30–250°C for ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2002, 5, 59–62. 44. Asbury, G.R.; Hill, H.H., Jr., Using different drift gases to changes separation factors (α) in ion mobility spectrometry, Anal Chem. 2000, 72, 580–584. 45. Purves, R.W.; Guevremont, R., Electrospray ionization high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal. Chem. 1999, 71, 2346–2357. 46. Eiceman, G.A.; Karpas, Z., Ion Mobility Spectrometry, CRC Press, Boca Raton, FL, 1993. 47. Viehland, L.A.; Mason, E.A., Transport properties of gaseous ion over a wide energy range, At. Data Nucl. Data Tables 1995, 60, 37–95. 48. Buryakov, I.A.; Krylov, E.V.; Nazarov, E.G.; Rasulev, U.Kh., A new method of separation of multi-atomic ions by mobility at atmospheric pressure using a high-frequency amplitude-asymmetric strong electric field, Int. J. Mass Spectrom. 1993, 128, 143–148. 49. Gorshkov, M.P, Invention Certificate No. 9666583, Russia, G01N27/62, 1983. 50. Viehland, L.A.; Guevremont, R.; Purves, R.W.; Barnett, D.A., Comparison of high-field ion mobility obtained from drift tubes and a FAIMS apparatus, Int. J. Mass Spectrom., 2000, 197, 123–130. 51. Guevremont, R.; Barnett, D.A.; Purves, R.W.; Viehland, L.A., Calculation of ion mobilities from electrospray ionization high-field asymmetric waveform ion mobility spectrometry mass spectrometry, J. Chem. Phys. 2001, 114, 10270–10277. 52. Krylov, E.; Nazarov, E.G.; Miller, R.A.; Tadjikov, B.; Eiceman, G.A., Field dependence on mobilities for gas phase protonated monomers and proton bound dimers of ketones by planar field asymmetric waveform ion mobility spectrometer (PFAIMS), J. Phys. Chem. A 2002,106, 5437–5444. 53. Krylova, N.; Krylov, E.; Eiceman, G.A.; Stone, J.A., Effect of moisture on high field dependence of mobility for gas phase ions at atmospheric pressure: organophosphorus compounds, J. Phys. Chem., 2003, 107, 3648–3648. 54. Kim, S.H.; Betty, K.R.; Karasek, F.W., Mobility behavior and composition of hydrated positive reactant ions in plasma chromatography with nitrogen carrier gas, Anal. Chem. 1978, 50, 2006–2012.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_68.... 05/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_68.... 05/05/2007

Página 1 de 1

Page 69

Appendix A Sensitivity of Calculated Mobilities toward the Choice of Parameters A.1 Introduction Before discussing the physical significance of the fitting procedure described in Chapter 2 and its sensitivity toward the choice of the parameters of the fit, the terms used are briefly reviewed. When an ion of mass m drifts through a bath gas of mass M, there is an electrostatic interaction between them. The interaction potential, V(r), depends on the distance, r, between the interacting species. At the point of closest approach, rm, the depth of the potential curve is ε0 and represents the strongest interaction. The cross section, ΩD, depends on the characteristics of the ion (size, shape, and charge distribution) and neutral molecule (size, shape, and dipole or quadrupole moments). The physical parameters mentioned earlier (V(r), rm, ε0, and ΩD) can be calculated as described in Chapter 2. These calculations are based on the use of other independent parameters, a*, r0, z, and n. These in turn are obtained from the fitting procedure, which involves a comparison of experimentally determined reduced mobility values with that of the calculated values. In the calculations, different choices of these independent parameters are tested until the least deviation between the calculated and measured mobilities is obtained. It should be emphasized that the fitting procedure described in Chapter 2 is a purely mathematical exercise and, to gain insights on the physical correlation between the mobility of ions, their mass, temperature, and drift-gas effects, this procedure has to be evaluated. The effects of the choice of each of the independent parameters (a*, r0, z, and n) on the calculated values of rm, ε0, V(r), and ΩD, and on the reduced mobility, will be discussed in this section. The sensitivity of the fitting procedure was tested on three series of protonated aliphatic amines: primary, tertiary, and cyclic. The structure of the first series is less constrained and well defined than that of the last two. Furthermore, as the proton (the ionic charge) on the tertiary amines is much more shielded from the drift-gas molecules than it is in the cyclic and primary

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_69.... 05/05/2007

Página 1 de 1

Page 70 amines, the interaction of the tertiary ions with the gas molecules should be significantly weaker than that of the other two series.

A.1.1 Choice of a* Values of a* covering the range from 0 to 1 in increments of 0.1 were taken from the tables generated by Mason and co-workers.6 Values of a*>0 indicate separation between the center of charge of the ion and its center of mass, whereas a*=0 indicates that the two centers coincide. The fitting procedure outlined earlier was carried out for each of the three series of compounds mentioned earlier. The results, expressed in terms of X2, which represents the sum of squared deviations between the calculated and experimental reduced mobilities, are summarized in Table A– 1. From these results it appears that the best fit (lowest X2 value) in the series of primary amines was obtained for a choice of a*=0.2. In the series of tertiary amines, all X2 values were relatively large, and choice of a*=0 was only slightly better than a*=0.1. In the series of cyclic amines, the order is reversed, with a*= 0.1 being slightly favored over a value of a*=0. These results are consistent with what would be intuitively expected if the charge is localized mainly on the nitrogen of the amine moiety, which is the preferred site of protonation. Thus, in tertiary amines, a*=0 indicates that the center of charge and center of mass are both on the nitrogen atom and they almost coincide. The value of a*=0.1 in the series of protonated cyclic amines implies that they are slightly separated, whereas in the primary amines a*=0.2 attests to the fact that they are further apart than in the other two series. The choice of a* strongly influences the calculated potential energy surface, V(r), directly (see Equation 2–20) and through its effect on rm(a*=a/rm) and ε0 (Equation 2–11). For example, shown in Figure A–1 is the potential energy surface, calculated by using Equation 2–20, for the interaction of protonated octylamine (130 amu) with the drift-gas molecules in air at 250°C. Curve a was calculated using a*=0.1, rm=7.988 Å, and ε0=0.311 meV, whereas in curve b, a*=0.2, rm=7.142 Å and ε0=0.779 meV. These values of rm and ε0 TABLE A−1 The Sensitivity of the Fitting Procedure to the Choice of a*, as Judged by the Agreement Between Measured and Calculated Reduced Mobility Values

a* 0

0.1

0.2

0.3

0.4

n-primary amines

1.26

0.91

0.34

38

8.8

Tertiary amines

2.74

2.90

5.9

57

123

Cyclic amines

0.15

0.15

0.50

50

4.2

The sum of squared deviations, X2, shown in the table reflects the quality of the fit in each series.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_70.... 05/05/2007

Página 1 de 1

Page 71

FIGURE A–1 The calculated potential energy surface for protonated octylamine ions (130 amu) drifting through air at 250°C. Curve a was calculated taking a*=0.1, rm=7.988 Å, and ε0=3.11 meV; curve b was calculated using a*=0.2, rm=7.143 Å, and ε0=7.79 meV; and curve c with a*= 0.4, rm=7.143 Å and ε0=7.79 meV. (Source: Reference 41, Chapter 2.)

were obtained by calculation according to Equation 2–22 and Equation 2–11, respectively. The stronger interaction calculated with a*=0.2 is reflected in the greater depth and closer position of the minimum in the potential surface. It is interesting to note that using given values of rm and ε0 and changing a* only slightly affects the shape of the potential surface as indicated by curve c in Figure A–1. This curve was calculated using a*=0.4 but with the same rm and ε0 values used in the calculation of curve b. As seen from Figure A–2, a* has only a relatively small effect on the calculated Ko−1 values. Changing a* from 0 (curve a) to 0.2 (curve b) barely affected the fit with the experimental values. However, a larger increase to a*=0.4 noticeably diminishes the quality of the fit as seen in curve c. As pointed out earlier, even though small changes in a* do not strongly affect the quality of the fit with the measured reduced mobilities, the choice of a* greatly influences the values calculated for rm, ε0, and ΩD, which have a significant physical meaning. Therefore, it is necessary to use a fixed value of a* to make comparisons between the calculated physical parameters of the three series of compounds. For the comparison, a value of a*=0.1 was chosen, which is closest to the best value for all three series.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_71.... 05/05/2007

Página 1 de 1

Page 72

FIGURE A–2 The experimentally determined inverse reduced mobility of protonated normal primary amines as a function of ion mass. Curves a, b, and c represent the calculated fit obtained with a*=0, 0.2, and 0.4, respectively. (Source: Reference 41, Chapter 2.)

In each series of ions, rm increases with ion mass, as expected from Equation 2–23. However, the rm value is largest for ions of the primary amines and smallest for the cyclic amines, as can be seen in Figure A–3, traces a and b, respectively. Recall that rm represents the distance between the centers of mass of the ion and neutral. This again corresponds to the intuitive understanding of the shape and size of these ions: the primary amines can be visualized as long chains wagging behind the protonated amino group with the center of mass separated from the charge, whereas the cyclic and tertiary amines have more compact structures, and their centers of mass and charge are closer. The depth of the minimum in the interaction potential curve, ε0, decreases with ion mass in the three series, again as expected from considerations of ion size and proximity of approach for the drift-gas molecules. For a given ion mass, ε0 has smaller values for the primary amines compared to the cyclic amines. This is a manifestation of the trend in the value of rm, as expressed by Equation 2–23. The trend in the value of the collision cross section, ΩD, also reflects the effect of rm, expressed in Equation 2–10. That is, the cross section increases with ion mass (or ion radius) in all series, and for a given ion mass it is largest in the primary amines and smallest in the cyclic amines. As mentioned earlier, increasing a* leads to a marked decrease in rm and therefore in ΩD, whereas ε0 increases. This is demonstrated in Figure A–3 for

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_72.... 05/05/2007

Página 1 de 1

Page 73

FIGURE A–3 The calculated rm values of protonated primary and cyclic amines as a function of ion mass. Curves a and c were calculated with a*=0.1 and 0.2, respectively, for the same experimental data set of primary amines, and curve b was calculated with a*=0.1 for cyclic amines. (Source: Reference 41, Chapter 2.)

primary amines, in which changing a* from 0.1 (curve a) to 0.2 (curve c) strongly affects rm and leads to a fixed decrement of about 0.9 A. The effect of the choice of the independent parameter, a*, on the physical parameter, rm, is clearly seen here; rm for protonated primary amines, calculated with a*=0.2, is smaller than that of the corresponding cyclic amines, calculated with a*=0.1 (curve b). This effect is relatively larger for the smaller ions. It has been shown previously that the value of a* depends not only on the nature of the ion but also on the properties of the drift gas.39 The observed increase of a* with the polarizability of the drift gas was attributed to the dipole moment induced in the polarizable drift-gas molecules by the ion.

A.1.2 Choice of r0 and z It has been previously shown that the addition of a correction factor is especially important when the ion mass range is large and when the drift gas has a low molecular weight.37 This is due to the lack of sensitivity of the reduced mass term to small changes in the ion mass when m>>M. The correction factor, m*z, compensates for this lack of sensitivity by including an additional mass-dependent term in the expression for rm, as shown in Equation 2–23. The choice of z will thus affect the calculated value of rm, and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_73.... 05/05/2007

Página 1 de 1

Page 74

FIGURE A–4 The effect of choice of r0 on the quality of the fit for primary amines in air at 250°C All curves were calculated with a*=0.2. Curve a, r 0=2.28 Å and z=0.00316 Å/amu (fully optimized); curve b, r0=2.48 Å and z=0.00316 Å/amu; curve c, r0=2.08 Å and z=0.00316 Å/amu; curve d, r0=2.48 Å and z=0.00186 Å/amu; and curve e, r0=2.08 Å and z=0.0045 Å/amu. (Source: Reference 41, Chapter 2.)

the latter in turn affects ε0, V(r), and ΩD. The choice of r0, which, for a given homologous series of ions, represents a constant radius typical of that series (Equation 2–22), strongly affects rm, and the latter in turn governs the calculated reduced mobilities. This has been demonstrated previously for ions drifting through air and helium,37 when the correction factor, z, was set to zero. It was also found that for a homologous series of ions spanning a broad range of ion masses, the values of r0 and z must be optimized to obtain a good fit between calculated and measured mobilities. However, it is interesting to test the sensitivity of the calculations when looking at homologous ions covering a limited mass range, such as the primary amines (32 to 186 amu). Curve a of Figure A–4 was calculated for fully optimized values, r0=2.28 Å and z=0.00316 Å/amu, and a good fit with the experimental reduced mobility values was obtained. Optimization of r0 without imposing restrictions on a*, z, and n barely affects the good quality of the fit (curves d and e). On the other hand, changing r0 by less than 10%, while using the same optimized value of z=0.00316 Å/amu, leads to a significant deterioration in the quality of the fit as seen in curve b, in which r0=2.48 Å, and curve c, in which r0=2.08 Å.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_74.... 05/05/2007

Página 1 de 1

Page 75 A similar phenomenon was observed when z and r0 were interchanged. Thus, over a limited range of ion masses, an erroneous choice of r0 (within reason) may be compensated for by z without noticeably affecting the quality of the fit. However, over a broad range of ion masses (up to 500 amu), the quality of the fit is more sensitive to the choice of both parameters.

A.1.3 Choice of n The experimental data given in the preceding text were measured at a relatively high temperature (250°C) and in a drift gas with a relatively low polarizability (helium and air). However, when carrying out mobility measurements at temperatures below 100°C or in gases with a tendency to form clusters, such as CO2 and SF6, the assumption that the effective mass of the ion is that of the core ion is no longer valid.40,41 Thus, although neglecting n is not always justified, this is essentially the case in the calculations made earlier. A different case is shown in Figure A–5, in which the reduced mobilities were measured in CO2 at 154°C and where curves a, b, and c were obtained with n=0, 0.44, and 2, respectively. In Figure A– 5a, the full mass range is shown, and the fit with measured mobilities appears to be fairly linear in all cases. However, a closer look at the lower end of the mass scale (up to 200 amu) in Figure A–5b shows that underestimating the extent of clustering (curve a, n=0) or overestimating it (curve c, n=2) leads to a marked decrease in the quality of the fit. Thus, as was previously demonstrated,40,41 the quality of the fit depends also on the choice of n, and its effect is more pronounced on low-mass ions.

A.1.4 Summary of Sensitivity Analysis In the fitting procedure, which is purely mathematical, four independent parameters are used to fit a set of experimental data points. However, the physical significance of the choice of these parameters is manifested in their values. The most striking example is the value derived for a* from these calculations. For example, the fact that a* increases with the polarizability of the drift gas and with the asymmetry of the ion indicates that it can be used to derive structural and physical information. Furthermore, the choice of a* strongly affects the calculated value of rm, and with it the values of V (r), ε0, and ΩD. Thus, the value of a* must be carefully chosen, bearing in mind the physics underlying the experimental data. For example, due to the small effect of the choice of a* on the agreement between experimental and calculated reduced mobility values, especially in helium,39 choosing a*>0.1 would lead to an unreasonably large calculated attractive interaction between the ion and helium atom. The quality of the agreement between the experimental and calculated reduced mobilities shows a much more pronounced dependence on the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_75.... 05/05/2007

Página 1 de 1

Page 76

FIGURE A–5 The dependence of the quality of the fit between experimental and calculated reduced mobility values on the value of n. Curve a was calculated with n=0, curve b with n=0.44, and curve c with n=2. Figure A–5a shows ions with masses up to 522 amu, while Figure A–5b expands the lower mass end of Figure A–5a. (Source: Reference 41, Chapter 2.)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_76.... 05/05/2007

Página 1 de 2

Page 77 choice of r0 and z. Improperly choosing these parameters leads to a good quantitative fit only over a limited range of ion masses. Changing r0 and z directly affects the calculated value of rm, and through it the values of ε0, V(r), and ΩD. The sensitivity of the fitting procedure on the choice of n is not as dramatic as with the other parameters because its effect is manifested only on the reduced mass and reduced mobility. The effect becomes important only under conditions of low temperature with highly polarizable drift gases and affects light ions more strongly than heavy ions.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_77.... 05/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_77.... 05/05/2007

Página 1 de 1

Page 79

3 Gas-Phase Ion Chemistry in Mobility Spectrometers 3.1 Introduction and General Considerations The chemistry of ionization of constituents in a sample links the observed mobility spectrum to the composition of the sample introduced in the ion mobility spectrometry (IMS) analyzer. This is arguably the most essential step in an IMS measurement because the mobility characterization has little practical analytical value if ions are not formed from a sample or if they are not representative of the sample. In addition to this initial step of ion formation, the subsequent behavior of ions, including the lifetime of the ions and further chemical reactions, in the supporting atmosphere of the drift tube will affect the mobility spectrum. This ionization step in IMS occurs through gas-phase ion-molecule reactions, commonly in air and at ambient pressure. Historically, this has been both the strength of IMS as an analytical method and a practical complication in interpreting the analytical response. The ionization processes occurring in the source region of a mobility spectrometer have been variously referred to as atmospheric pressure ionization (API), atmospheric pressure chemical ionization (APCI), or gas-phase ion-molecule reactions at ambient pressure; the terms are used interchangeably in this chapter. The reactions found in APCI are described by well-established principles of kinetics, thermodynamics, and molecular structure; experimental parameters of temperature, pressure, moisture, and concentration affect response from a mobility spectrometer. Difficulties in the interpretation of mobility spectra have arisen because the effects of these experimental parameters on gas-phase ion chemistry at ambient pressure have not been understood in sufficient detail for the preparation of predictive models. In particular, the formation of cluster ions, which is an inherent feature of analytical IMS, introduces a complexity that has not received thorough treatment by molecular reaction models. Also, ions formed originally in the source region may undergo changes in identity while passing through the drift region. Thus,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_79.... 05/05/2007

Página 1 de 1

Page 80 the mobility spectrum is governed by ion chemistry during the initial step of ion formation and by subsequent chemical reactions with the product ions before arrival at the detector. These ideas were not available as a comprehensive or integrated model until the late 1990s, and are still undergoing refinement. Nonetheless, a few tools have been available to IMS practitioners for the a priori prediction or interpretation of mobility spectra. In nitrogen at 25°C and 1 atm, the density of molecules is 2.5×1019 molecules/ cm3. Molecules with a velocity of 583 m/sec will undergo 4×109 collisions/ sec with ~5×10−7 m distance between collisions. Thus, during a 5-msec residence time in a drift tube, an ion will undergo 2.7×107 collisions with molecules of the supporting atmosphere, providing opportunity for subsequent chemical transformations. There are two consequences to this: (1) ion-molecule reactions will be governed in large part by collision frequency (and by extension the composition of the supporting atmosphere) and (2) the large number of collisions will cause ions to be thermalized, i.e., their additional energy gained from the external field will be lost in collisions when E/N is below 2 Td. A practical, simplified discussion is given in the following sections on the ion chemistry inside an IMS drift tube from the first steps of ionization to the ultimate detection of the ion products. A detailed treatment of the kinetics and thermodynamics of these reactions is provided in Professor J.A.Stone’s article found in the CD accompanying this book.

3.2 Ion Chemistry at Ambient Pressure 3.2.1 Formation of Reactant Ions 3.2.1.1 Positive-Ion Formation Reactant ions are formed from the supporting atmosphere inside the IMS drift tube by high-energy primary electrons emitted by a conventional β source (the standard 10 mCi 63Ni). The energy distribution of the electrons emitted by 63Ni range from 0 to 67 keV with an average value of 17 keV. When a high-energy electron collides with a nitrogen molecule, a positive ion and a secondary electron (perhaps 1 keV) is formed per Equation 3–1. The primary electron will have lost energy in the ionization step but may be considered still a high energy electron capable of subsequent ionizations by collisions with nitrogens.1 In addition, the secondary electron will ionize nitrogen through collisions until its energy is below the ionization potential of air (35 eV needed to produce one thermalized ion pair). (3–1)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_80.... 05/05/2007

Página 1 de 1

Page 81 In addition to N+ ions may also be formed through dissociation, but these and others that may be formed will be neglected to simplify the following discussion. Further ionization of nitrogen molecules by primary and energetic secondary electrons will continue until the energy of the electrons drop below the energy needed to form an ion pair in nitrogen. Because the energy required for the formation of one ion pair is roughly 35 eV, an electron with 17 keV initially could produce ~500 nitrogen ions during the dissipation of energy toward becoming thermalized. A conventional 10 mCi source of 63Ni, with continuous emission of high energy electrons, will produce each second ) form a reservoir of charge ~109 ions/cm3.1 The nitrogen ions and thermalized electrons (or as from which all the subsequent ionization chemistry in the ion source of the IMS is derived. This and related ionization sources such as corona discharge are limited by the magnitude of energy transferred by the primary electrons. An ion of N2+ at ambient pressure will undergo collisions that may lead to a sequence of reactions as shown in Equation 3–2 to Equation 3–6. Such reactions were explored using mass spectrometers by Good et al., with a source at elevated pressure2 and by Shahin3 using a corona discharge ion source at ambient pressure. (3–2) (3–3) (3–4) (3–5) (3–6) Reactions in Equation 3–5 and Equation 3–6 are reversible with the formation rate controlled by the concentration of water vapor. In clean air or nitrogen, the hydrated protons (see Equation 3–5 and Equation 3–6) are the dominant ions present in the ion source and are called the reactant ions, and the term “water-based ionization” is often used to describe this condition. These reactions were identified early in the development of IMS.4 However, other ions including NH4+(H2O)n and NO+ (H2O)2, where n takes values from 1 to 10, can also be observed in mobility spectra in a supporting atmosphere containing a few parts per million of moisture. The formation processes of NH4+(H2O)n and NO+(H2O)2 are not part of the sequence shown in Equation 3–2 to Equation 3–6 and are understood to originate from trace impurities in the ion source or reaction region. Water influences the kinetics of each reaction from Equation 3–3 to Equation 3–6. Remarkably, the effects of moisture in IMS were not reported in systematic

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_81.... 05/05/2007

Página 1 de 1

Page 82

FIGURE 3–1 Mobility spectra for various levels of moisture in the reaction region. Increases in moisture from 1 ppm in air caused displacement of the peak to the right as hydration of the proton was increased.

studies until the late 1990s and are now understood to be essential in the analytical performance of mobility spectrometers. As shown in Figure 3–1, the drift time of the hydrated-proton reactant ion is displaced toward longer drift times proportionally with the increase in the moisture level in the source region and the degree of hydration of the core ion. The effects of temperature on ion composition in IMS were examined in detail by Kim et al. who showed, using IMS/MS, that temperature controlled the extent of clusterization and thus the distribution of hydrated cluster ions.4 Raising the temperature caused a decrease in the number of water molecules in a clustered ion, and this was observed as a corresponding increase in Ko values (Figure 3–2A). Increases in temperature change the reduced mobilities of the peak by gradually diminishing the value of n in the formula H+(H2O)n as shown in Figure 3–2B. Large ion clusters with water molecules are favored at low temperatures, and clusters can be dissociated at elevated temperatures. Although reactant ions are in equilibrium with several hydration levels, only a single peak appears in the mobility spectrum, reflecting the coalescence of multiple ions into a single mobility peak characterized by the weighted sum of all the individual ions of the equilibrium.5 So long as the forward and back reactions of a core ion in dynamic equilibrium with neutrals in the supporting atmosphere are fast compared to drift times, peaks in IMS will be undistorted

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_82.... 05/05/2007

Página 1 de 1

Page 83

FIGURE 3–2 Effects of temperature on the reduced mobility of reactant ions in IMS. (A) The reduced mobility for H +(H2O)n gradually increases as temperature is increased, due to decreasing hydration of the ion. The resultant mobility is the weighted average of all hydrated protons. (B) The distribution of ions according to n is shown. (Adapted from Reference 4.)

Gaussian peaks. Thus the mobility spectrum appears to be simple, although the chemical processes underlying it are complex. Because moisture strongly affects the kinetics of reactions in Equation 3–3 to Equation 3–6, low levels of moisture in the ion source may result in prolonged lifetimes for N2+, H2O+, or N4+. These ions may participate in charge-transfer reactions as shown in Equation 3–7 with molecules in the reaction region including sample or impurities, even alkanes or alkenes6: (3–7)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_83.... 05/05/2007

Página 1 de 1

Page 84 To summarize, the identity of reactant ions is governed by temperature and moisture content in the supporting atmosphere of the reaction region. Since response of an IMS will be affected by the reactant ions (vida infra), response of an IMS to a sample will also be affected by moisture, temperature, and purity of the gas inside the drift tube. 3.2.1.2 Negative-Ion Formation The cross section for formation of negative ions from collisions with electrons is dependent on the energy of the electrons and the electronegativity of the molecule. Generally, cross sections are large for electrons with thermal energies, although some compounds may have a large cross section at a specific energy (resulting in resonant electron capture). At ambient pressure, high energy electrons from 63Ni are rapidly thermalized through the cascade process described earlier. In pure nitrogen, electrons are so effectively thermalized that they can be observed as free electrons in the mobility spectrum.7 These thermalized electrons can be reactive species when pure nitrogen or other inert gases are used as the supporting atmosphere in the source region. In air, thermalized or low-energy electrons can attach to oxygen (with a favorable electron affinity) through three-body collisions8 forming a negative ion and hydrates per Equation 3–8 to Equation 3–10: (3–8) (3–9) (3–10) where M can be O2, H2O, or another neutral molecule. In the presence of traces of CO2, several additional cluster ions are observed in IMS/MS studies, as shown in Figure 3–3.9–12 These ions include clusters containing and with water and CO2. When resolution is low, a single peak is observed in the mobility spectrum (upper frame, Figure 3–3), but this peak comprises several ion species, as shown in Figure 3–3 (bottom frame).9 The peak shape in the mobility spectrum is indicative of a rapidly achieved and dynamic equilibrium of a mixture ions in which reversible reactions are fast relative to the time required for ion transit through the drift region.5 When the supporting atmosphere contains impurities, additional ions may appear as reactant ions through reactions with the impurities. Determining the exact composition of such cluster ions even with IMS/MS could be ambiguous, as evident in the slight discrepancies in Table 3–1, in which two different studies attempted to determine the identity of the ions in similar IMS analyzers. This could be attributed in part to halocarbons, which are common. (3–11) where typically n=1 to 3.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_84.... 05/05/2007

Página 1 de 1

Page 85

FIGURE 3–3 Mobility spectrum (top) and mass spectrum (bottom) for ions formed in air with an ion mobility spectrometer at ambient pressure.10

In some instances, the level of halocarbon contamination in an IMS drift tube is so large that Cl− (H2O)n becomes the de facto reactant ion, which may not be recognized by the user. Just as the hydration for positive ions is affected by moisture and temperature, so is the level of hydration in negative ions. In Figure 3–4, the effects of temperature on reduced mobility from the degree of clusterization can be observed for the chloride ion and may exhibit, in some instances, a reduced mobility coefficient similar to that for O2− (H2O)2, causing confusion over ion identity. The formation of positive and negative ions in the same volume of the ion source will lead to coulombic forces of attraction that can lead to collisions between positive and negative ions (or electrons) and eventually to neutralization of the ionic participants through recombination reactions. These reactions are a sink for ions and can be substantial in the absence of flows and potential gradients in the electric field. In most mobility spectrometers, the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_85.... 05/05/2007

Página 1 de 2

Page 86 TABLE 3–1 Negative Ions Observed in Early Studies of Reactant Ions in Mobility Spectrometers

Carr Ion

Spangler Ko

Mass

CN−

3.37

26

O 2−

3.31

32

O2 −

3.28

Cl−

3.01

35:37

Cl−

2.94

(H2O)OH−

2.94

NO2−

2.70

50

(H2O)O2−

2.59

53

(H2O)OH−

2.94

68

(H2O)2O2−

2.59

76

CO4−

2.41

35 CNO−

2.76

42

NO2−

2.76

46

O(H2O)2−

3.17a

50

CO3−

2.55

60

O 4−

3.11

64

CO4−

−b

Ion

Ko

a Observed in air atmosphere. Mobility value reported only in SF atmosphere. 6 b Not listed.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_86.... 05/05/2007

Página 2 de 2

FIGURE 3–4 The drift times (solid circles) and reduced mobility values (circles) for Cl− (H2O)n at temperatures from 50 to 200°C at constant moisture. Similar trends may be obtained at a constant temperature and decreased level of moisture. (Source: G.A.Eiceman, unpublished findings.) Graphic above the plot illustrates ion declustering.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_86.... 05/05/2007

Página 1 de 1

Page 87

FIGURE 3–5 Plots of ion density vs. ion source radius. The plots were formed from a model provided by Siegel1 and illustrate the losses of ions from ventilation (gas flow) and from recombination rate constants for positive and negative ions of (A) 10−8; (B) 10−7; (C) 10−6 cm3s−1. (Adapted from Reference 1.)

reactions, described above in Equation 3–2 to Equation 3–10, occur in a linear field gradient; so the ions of opposing charges are separated or extracted from the bulk of the ion source. Nonetheless, losses do occur by recombination and diffusion losses to walls of the drift tube. Siegel1 suggested in a steady-state model for ion densities that recombination is more significant than diffusion losses, as shown in Figure 3–5. The uncontrolled or unintentional addition of impurities of organic or inorganic vapors, even at trace levels of 1 to 50 ppb, may alter the identity of the ions formed in the reaction region and affect the observed mobility spectrum, the initial ion chemistry, or the response to the sample. This is discussed in the following sections. The complex nature of multiple reactant ions is no handicap in using IMS analyzers for practical chemical measurements. Moreover, the exact nature or composition of the reactant ions is

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_87.... 05/05/2007

Página 1 de 1

Page 88 noncontroversial provided the supporting atmosphere has been cleaned and kept constant. However, the complex nature of reaction ion chemistry makes detailed physicochemical studies difficult or impossible with analytical grade IMS instruments. In summary, the reactant ions serve as more than just a nondiscriminating reservoir of charge, and they feature in issues of selectivity and sensitivity of IMS analyzers, as described in the Subsection 3.3.2.2.

3.2.2 Formation of Product Ions 3.2.2.1 Positive-Ion Reactions In nearly all analytical mobility spectrometers, reactant ions are produced continuously in the vicinity of the radioactive source and are extracted by the electric field and introduced into the reaction region. In the absence of analyte molecules, the reactant ions pass through the reaction region practically without any chemical reactions and exhibit a distinct spectrum governed by experimental parameters of pressure, temperature, electric field, moisture, and drift length. When analyte molecules (M) from the sample are introduced into the reaction region, reactant ions undergo collisions with them. The collisions yield an energetically excited intermediate adduct ion or transition state ([MH+(H2O)n]*) that may dissociate back to the reactants or may form product ions through other reaction pathways. Such reaction pathways involve a third body (Z) to stabilize the products, as shown in Equation 3–12: (3–12) The reaction shown in Equation 3–12 is the dominant pathway to form product ions in positivepolarity IMS, in which a protonated monomer is formed through association or transfer of a proton to the analyte molecule. This can also be viewed as a displacement reaction in which a water molecule is displaced from the hydrated proton by the analyte molecule. The rate of the forward reaction is established by collision frequency, the concentration of the reactant ions [H+(H2O)n], and the concentration of the analyte molecules [M]. The rate of the reverse reaction is controlled by the energy of association between M and MH+(H2O)n. Any collisions between reactant ions and analyte molecules favor the formation of the intermediate; however, appreciable amounts of product ions are formed only when −ΔG≥5 kcal/mole. The reaction in Equation 3–12 is remarkably simple compared to mass spectra from electronimpact ion sources because proton transfer reactions are considered “soft ionization” processes and are usually not accompanied by fragmentation of the product ion. This is also seen in chemical ionization mass spectrometry.13 As the concentration of analyte molecules is raised, the reservoir of reactant ions is depleted along with an increase in the amount of product ions, MH+(H2O)n−1. In a mobility spectrum, this is seen as a decrease in the intensity of the reactant-ion peak and an increase in the peak for the protonated

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_88.... 05/05/2007

Página 1 de 1

Page 89

FIGURE 3–6 Mobility spectra for a chemical at various gas phase concentrations showing differing intensities of reactant ion peak, protonated monomer and proton bound dimer.

monomer. As the concentration of analyte molecules is increased further, the protonated monomer may be clustered with an additional analyte molecule, forming a proton-bound dimer as shown in Equation 3–13. (3–13) This is seen in a mobility spectrum as a further decrease in the peak intensity of the reactant ion, a decrease in peak height of the protonated monomer, and an appearance of a third peak at a drift time longer than that for the protonated monomer, as shown in Figure 3–6. This third peak is for the proton-bound dimer. An increase in [M] beyond a certain value will lead to total consumption of the reactant ions and elimination of the reactant-ion peak and even the protonated monomer peak from the mobility spectrum. Complete exhaustion of the reservoir of reactant ions is a condition known as saturation of the reaction region, and this condition should be avoided (see discussion in Section 3.3.1.1). Although all substances, in principle, should form protonated monomers, only those that are sufficiently long-lived will be detected in a mobility spectrometer, i.e., observed in the mobility spectrum. The likelihood of detecting an ion is controlled by the association of the molecule and proton, which is governed by the enthalpy and entropy of the ion and by the level

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_89.... 05/05/2007

Página 1 de 1

Page 90 of moisture in the supporting atmosphere. In the instance when the association is weak and water levels are high, the protonated monomer will undergo rapid dissociation to M and H+(H2O)n, which is the reverse reaction from the activated intermediate in Equation 3–12. In this case, there will be no peak observed in the mobility spectrum and no apparent response to the chemical or sample. The central issue is the stabilization of the intermediate or transition state, to form MH+(H2O)n. Reactions in IMS occur between a cluster ion and a neutral molecule; consequently, the free energy of reaction is determined by the nature of reactants and products. The direct application of the proton affinity scale that was developed for free ions in the gas phase at low pressure to the evaluation of the chemistry in the IMS source is quantitatively inaccurate because these scales were derived for unclustered gas-phase ions.14,15 Nonetheless, similarities have been empirically observed between the proton affinity values and the reactivity of ions in IMS. IMS users who wish to employ the documented proton affinity values should be mindful of this limitation and also that the proton affinity of clustered ions may differ from tabular or literature values for proton affinities. However, it is well established that compounds with high proton affinities, as shown Figure 3–7, will generally exhibit similar trends in response with IMS analyzers. In addition to proton transfer, two other reaction pathways may lead to the formation of product ions. Reactant ions in very dry atmospheres may lead to charge transfer reactions such as those shown in Equation 3–7.6 When the ionization potential of the analyte molecule is considerably lower than that of the reactant ions (N2+ or H2O+), charge transfer may be accompanied by fragmentation of the product ions.16,17 Another reaction pathway involves the formation of adduct ions such as MNO+(H2O)n.18,19 3.2.2.2 Negative-Ion Reactions As noted in Section 3.2.1.2, the principal reactant ion in negative polarity with clean air is O2−(H2O) n. Unlike positive-ion chemistry, in which proton transfer is the dominant reaction pathway, the negative-reactant-ion chemistry forms an adduct ion (Equation 3–14) as the first step: (3–14) As the adduct is stabilized by collision with a third body, the ion may survive as a hydrated adduct ion between M and O2−. However, reactions by charge transfer (Equation 3–15) and proton abstraction (Equation 3–16) may also occur: (3–15) (3–16)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_90.... 05/05/2007

Página 1 de 1

Page 91

FIGURE 3–7 Scale of proton affinities for chemicals ranging from low proton affinities (bottom) to high proton affinities (top). Proton affinities were obtained at NIST Chemistry WebBook. Favorable response in an IMS analyzer generally follows the trends seen in proton affinities, which do not account for the formation of cluster ions seen in mobility spectrometers.

The selection between pathways of adduct ion formation and proton abstraction is controlled by the acidity of the protons as demonstrated for a series of fluorinated phenols that allowed a range of acidity on the hydrogen of the phenol group governed by the degree of substitution with fluorine on the phenol ring.20 In instances where the oxygen-hydrogen bond was comparatively strong (low acidity), the FxPh–O–H···O2− adduct was stable on the timescale of an IMS measurement and was observed in the mobility spectra. When the acidity of the proton was increased, thereby increasing the O–H bond length and weakening its strength, proton abstraction and formation of FxPh–O− was observed. Reactions involving the formation of an adduct ion have an important role in applications of IMS, notably the detection of explosives.21 In such fragmentations, NO2− can be formed and can further react forming an adduct ion,

even in the presence of other negative reactant ions.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Page_91.... 05/05/2007

Página 1 de 1

Page 92 Finally, the same dynamics of charge conservation described for positive ions are observed with reactions involving negative ions; that is, the signal intensity of the reactant ion O2–(H2O)n decreases as [M] is increased and the signal intensity of the product ion M O2–(H2O)n increases. Because the formation of MO2–(H2O)n is comparatively rapid, the charge available in O2–(H2O)n may be regarded as finite or stoichiometric on the timescale of an IMS measurement.

3.3 Analytical Facets of Gas-Phase Ion Reactions Ion mobility spectrometers have performed successfully in several applications in which only a threshold response is necessary or important, for example, the detection of explosives and screening of suspect object for drugs of abuse. In these applications, the detection of even trace amounts of the target analyte will trigger an alarm and no further quantitative information is necessary. In other instances, knowledge of both the presence of a substance and the quantitative amounts are required. Examples of such a quantitative use of an IMS instrument include the determination of chemical warfare agents and the screening of air for volatile organic compounds. In both these examples, human mortality or vitality are associated with dose-response relationships, and the level of dose is essential information. In the past decade, improvements in threshold applications of IMS have arisen from improved sampling methods, more sensitive and rugged instrumentation, and better data processing techniques. These are seen with advances in the detection of explosives with handheld as well as with fixed or stationary analyzers. Parallel advances have occurred with instrumentation and procedures requiring quantitative determinations. These have occurred with the increased availability of drift tubes that exhibit fast response, low memory effects, and reproducible delivery of sample to the ion source. Together, these applications have resulted in enhanced precision, sensitivity, and linear range compared with previous generations of analyzers. In the following sections, the principles underlying quantitative behavior for IMS analyzers are discussed.

3.3.1 Quantitative Aspects to Response 3.3.1.1 Effect of Sample Concentration on Response The first step in IMS response, ionization of neutral analyte molecules, fundamentally establishes a starting point for all other events that occur subsequently in the drift tube and establishes the boundaries for quantitative response. Central to all discussions of quantitative aspects of IMS is the limited reservoir of charge available for ion formation (see Section 3.2.1.1). The amount of charge available for formation of product ions is practically

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_92.... 05/05/2007

Página 1 de 1

Page 93 limited by the strength of the ionization source, by the overall sequence of reactions leading to the formation of reactant ions, and, finally, by the reaction rate of formation of product ions. Under typical operating conditions with a 63Ni source, the upper limit for ion density in the source is 109 to 1010 ions per cm−3sec−1, and this establishes the maximum charge available for the formation of product ions.1 In practice, this value may be reduced due to ion losses resulting from poor designs of electric fields and flow patterns. Furthermore, chemical losses in an actual measurement may arise through the presence of impurities, competing reaction pathways, and decomposition or fragmentation of the target product ion. At low concentrations of the analyte, reactant ions are consumed in proportion to the density of the vapor neutrals as per the following equation: Rate=k[M][H+(H2O)n]

(3–17)

The rate constant (k) in Equation 3–17 is a correction on the collision-rate constant and controls the formation of product ions. As shown in Figure 3–8, the density of product ions, for a sample at a constant concentration, is dependent on both the residence time of the sample in the source and its concentration. Depletion of the reaction ions is rooted in the kinetics of Equation 3–17 and parallels inversely the formation of product ions. The relationship is virtually stoichiometric. For a fixed time of residence of sample molecules in the reaction region, the number of product ions and their peak intensity in the mobility spectrum will be quantitatively proportional to the concentration of sample molecules. Also, increases in residence time of M in the source region will lead to an increase in the number of products ions, as shown in Figure 3–8, when the source is not saturated with sample vapor. Once the reservoir of reactant ion charge is totally consumed, no additional increase in product ion intensity will be seen even with additional increase in the concentration of sample molecules. In practice, this limits the upper end of the linear range associated with ion sources at ambient pressure. One of the best-recognized quantitative patterns in mobility spectra is the relationship between protonated monomers and proton-bound dimers, common to several types of compounds including esters, ketones, alcohols, amines, and organophosphorus compounds. As the vapor concentration of the analyte, M, is increased in an ion source (Figure 3–9), a protonated monomer peak first appears with a corresponding decrease in the reactant-ion peak intensity (Equation 3–12). A second peak appears corresponding to the proton-bound dimer with a further increase in analyte concentration; as this occurs, intensity declines for peaks of both the reactant ion and the protonated monomer (Equation 3–13). When the analyte vapors are removed from the ion source and the analyte concentration decreases, these patterns are reversed: intensity declines for the proton-bound dimer and it increases for the protonated monomer peak. Eventually, the protonated monomer peak intensity declines and the reactant-ion peak is restored to the original intensity. The dynamics of these changes are shown in Figure 3–9. Although this pattern is commonly

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_93.... 05/05/2007

Página 1 de 1

Page 94

FIGURE 3–8 Plots of ion number with reaction time or time of mixing for ions and neutrals at ambient temperature and pressure. The curves show response to two analytes of comparable proton affinity. The analytes differ by tenfold in concentration illustrating the kinetic basis of forming product ions.

associated with positive-polarity IMS, similar patterns may be observed with negative ions such as M*C1− and M2*C1−, as observed with volatile halogenated anesthetics22 and other chemicals that form long-lived negative-ion clusters. When the reactant-ion population is completely exhausted, the proportional relationship between product ion intensity and neutral-vapor density no longer holds, and this negates conventional calibrations. Thus, the quantitative value of the response of the analyzer is lost with existing methods and interpretations. Another troubling phenomenon is the diffusion of analyte molecules from the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_94.... 05/05/2007

Página 1 de 1

Page 95

FIGURE 3–9 Plots of ion intensities from vapor sampling of a moving plume of a chemical vapor. The numbers are time in seconds used to generate the plume. These plots show the ion dynamics in an IMS analyzer.

source region into the drift region when concentrations of the sample are increased in the source or reaction region. In the drift region, neutral adducts of the sample may associate with product ions of the same chemical leading to the formation of cluster ions of the type MnH+(H2O)x, in which n can have a value equal to or greater than three. Cluster ions in the ion swarm will rapidly undergo loss, followed by the addition of neutral adducts while the swarm traverses the drift region.5 The observed reduced mobilities for such ions are the weighted averages of the reduced mobilities of all the individual or cluster ions that participate in the equilibrium. A practical implication of this is that drift times of the product ions can vary over a large range, determined by the concentration of sample neutrals in the drift region. Because distribution of sample neutrals in the drift region will be irregular and dependent on sample

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_95.... 05/05/2007

Página 1 de 1

Page 96 concentration, attempts to assign identity to ions using Ko values become difficult or impossible. In summary, depletion of reactant ions should be avoided and a residual level of peak intensity for the reactant ions should be maintained to ensure that the reaction region is not saturated, which would lead to quantitatively unclear behavior of and concentration-dependent drift times for product ions. The most modern IMS drift tubes are designed to avoid this condition, but user carelessness can circumvent good designs. 3.3.1.2 Detection Limits In practice, two definitions for the limit of detection (LOD) or the minimum detectable level (MDL) may be made when application of IMS technology is considered. In the continuous monitoring of ambient air or a controlled atmosphere, MDL is the lowest concentration of the analyte that is observed in the mobility spectrum above a background noise level (signal-to-noise ratio >3). Typical units of such MDLs for volatile compounds are volume-based, in ppm, ppb, mg/m3, or μg/m3. When samples are provided as discrete items, such as filter paper or swipes or as solid-phase microextraction fibers, MDLs are expressed as picograms per sample or nanograms per sample. This is commonly practiced with semivolatile compounds and when IMS drift tubes are used as chromatographic detectors. The excellent detection limits of IMS analyzers may be attributed to the high efficiency of chemical ionization at ambient pressure through high collision rates, long residence times, and efficient conversion of each ion-molecule collision into an activated intermediate (Equation 3–12). In the past decade, quantitative reports on MDLs for IMS measurements have become common. One elementary aspect of the quantitative response of a traditional drift tube, i.e., linear potential gradient with ion shutter, is that only a small portion of the total ion charge in the reaction region is utilized during ion separation and detection. The ion shutter is open to the passage of ions between the reaction and drift regions for only 0.1 to 0.3 msec in a 20-msec duty cycle; thus, only about 1% of all possible ions are sampled, and even fewer are detected. Increases in the strength of the ionization source, the width of the ion shutter, or the duty cycle of the shutter have limited effectiveness because ion-ion repulsions in the drift tube become noticeable at levels above ca. 5×107 ion/cm3 and peak heights decline as the peak widths undergo broadening from coulombic repulsions.23 One of the earliest reports on MDLs in IMS was for dimethylsulfoxide (DMSO), where a peak for 10−10 moles (about 8 ng) of DMSO was seen using an IMS/MS with a GC inlet.24 Shortly thereafter, a GC/IMS/MS was used to detect 10−8 g of musk ambrette,25 and an even better MDL of 10−10 g was reported in an isolated or stand-alone IMS study. However, the method of sample delivery with such IMS configurations involved depositing a solution on a wire and inserting the wire into the heated IMS inlet where the sample vapors were desorbed. This made exact determination of the weight of the sample somewhat problematic. In addition, the studies did not exploit fully

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_96.... 05/05/2007

Página 1 de 1

Page 97 the definition of detection limit26 of the signal-to-noise ratio (S/N). The successful combination of capillary columns and fast-responding drift tubes has allowed reliable determinations of low MDLs. Baim and Hill reported27 an MDL of 6 pg for the methylester of 2,4-D (dichlorophenylacetic acid) and an MDL of 28 pg for di-n-hexyl ether28 using an S/N value of 3. Organophosphorus compounds with strong proton affinities exhibited detection limits of about 10 pg, based on the protonated monomer ion.29 The earliest rigorous evaluation of the ion mobility spectrometer as a stand-alone instrument for MDLs was the detection of sec-butylchloro-diphenyl oxides in biological tissues, where an MDL of about 20 pg was reported.30 Continuous sampling of vapors provided other instances in which IMS was deemed attractive due to its inherent sensitivity to an analyte, as typified by nickel carbonyl in air, with reported detection limits of 0.2 to 0.3 ppb.31 In another report of actual MDLs for continuous monitoring, Karpas et al. found detection limits of 10 ppb for bromine in the ambient air of a chemical plant.32 Other detection limits for continuous monitoring included volatile halogenated anesthetics with MDLs from 0.1 to 1000 ppb22 and nicotine with MDLs of 50 ppb.33 Perhaps the area of the greatest interest in detection limits for ion mobility spectrometers has been that of explosives detection (see also Chapter 6). Karasek first reported the detection of 10 ng of 2,4,6-trinitrotoluene (TNT) with an S/N>1000 from which an MDL in the low-picogram range was predicted by extrapolation.34 Lawrence and Neudorfl showed that ethylene glycol dinitrate (EGDN) undergoes ionization with O2−reactant ions to form EGDN*NO3−—which is an inefficient use of analyte vapor—but that ionization could be enhanced by the use of Cl− as the reactant ion to form EGDN*Cl−.35 Detection limits for EGDN with and without chloride reactant ions were 30 pg and 500 pg, respectively. This was presumably due to the simplification of ionization routes with a single reactant ion, i.e., chloride, rather than the complicated mixtures often seen with air, and perhaps due to the formation of a longlived chloride adduct. The ultimate detection limits for TNT and RDX in air were quoted as 3 ppt (v/v) in air or 0.3 ppt with signal-averaging on a large-volume reactor.36 These low MDLs were attained with a scrupulously clean drift tube in purified nitrogen supporting atmosphere and should not be expected for routine screening instrumentation. Throughout this discussion, the MDLs reported are on an absolute basis for specific analyzers and do not include any sample preparation or preenrichment. For example, the group at Sandia reported 0.03 ppt MDLs for explosives in air when a sampling tube was used to enrich vapors.36 Field instruments, unlike most laboratory instruments, operate at near-ambient temperatures and are equipped with membrane inlets to maintain clean atmospheres for internally recirculated gases. Sample vapors must pass through a membrane that exhibits selectivity governed by the molecular structure of the membrane, the presence of functional groups on the analyte, the temperature of the membrane, and the vapor pressure of the analyte. Membrane efficiency will rarely be 100% and, consequently, MDLs of membrane-equipped instruments will be poorer compared to those in which the sample is deposited

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_97.... 05/05/2007

Página 1 de 2

Page 98 directly into the reaction region. An MDL of ~10 ppb was reported for hydrazine and monomethylhydrazine (rocket propellants) in air using a handheld ion mobility spectrometer and continuous sampling of a vapor stream.37 When such handheld, membrane-equipped analyzers received vapors in short bursts (through a flow injection sample inlet), MDLs were 2 ng for aniline38 and 5 ng for dialkyl phthalates.39 The membrane degrades the detection limits by a factor of 10 to 100, suggesting that the yields of molecules that permeate across the membrane barrier are between 1 and 10%. 3.3.1.3 Repeatability, Stability, and Linear Range Before 1990, the repeatability of signal intensities and drift times had not been considered widely in reports or journal articles on IMS presumably because detection limits were the main concern then. Consequently, there is only a relatively brief record available in the literature on the repeatability of IMS measurements, which is key to any quantitative analytical method. The few examples that are available are concerned with short-term repeatability, and the relative standard deviation (RSD) for peak areas in these is between 5 and 25%.38 In one study with a handheld IMS analyzer, reproducibility was 6 to 27% RSD for 5 to 2,500 ng of dialkylphthalates as shown in Table 3–2.39 Measurements of hydrazine vapors at 10 to 200 ppb using the same instrument showed a precision of ca. 3 to 16% RSD for these highly TABLE 3–2 Quantitative Response of a Mobility Spectrometer for Thermal Desorption of Phthalate Esters from Filter Paper

Peak Area Amount (μg) 0.0050

Mean (N×103)

Std Dev. (% RSD)

2.01

1.49

1.40

(27)

Peak Area Amount (μg) 0.005

1.05 0.05

1.95

1.79

1.71

(7.0)

4.61

(N×103)

Std Dev. (% RSD)

2.36

2.27

2.02

(8.0)

2.42 0.05

1.70 0.25

Mean

2.90

3.15

3.28

(6.0)

3.27 3.90

0.25

3.91

5.92

5.49

5.39

(6.0)

5.16 0.50

10.5

8.42

6.86

(18)

0.50

6.88 1.0

7.55

6.25

(12)

8.22

8.42

13.4

16.2

(26)

1.0

15.5 2.5

8.19

12.5

11.0

9.75

(10)

10.8

23.3

23.6

24.1

(2.0)

2.5

23.2

19.9

20.2

22.5

(9.0)

18.1

Standard deviation includes the sample preparation and thermal desorption steps. 39

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_98.... 05/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_98.... 05/05/2007

Página 1 de 2

Page 99 TABLE 3–3 Precision of Five Replicate Determinations of Phenol by Vapor Desorption-Ion Mobility Spectrometry40

Σ Peak Area (area counts)

Phenol Amount (ng)

% Relative Standard Deviation

0

1235

22.9

10

3934

31.7

20

3395

42.1

30

5313

36.6

40

5526

36.6

50

6486

13.8

60

9481

11.1

70

10426

14.8

80

12800

9.0

90

14252

12.9

100

17495

7.6

200

30389

10.6

300

38872

10.0

400

39904

11.5

500

43049

6.9

600

47225

7.0

700

51181

4.3

800

53888

7.4

900

54753

8.0

1000

66087

3.4

2000

74663

3.8

3000

81964

1.4

4000

84101

1.4

5000

89207

6.1

6000

96092

1.8

7000

98844

1.4

8000

103834

2.2

9000

103831

0.7

10000

100320

11.7

reactive and adsorptive chemicals.37 A fourth example is for the determination of phenol by means of thermal desorption from filter paper. The repeatability was 42 to 0.7% RSD in the range of 10 to 10,000 ng, respectively; experimental results are shown in Table 3–3 and included errors from sample handling.40 All of these studies have been conducted under carefully controlled laboratory conditions. A measure of routine repeatability in IMS can be seen in GC/IMS measurements using the volatile organic analyzer (VOA) (Subsection 5.2.5.1). The repeatability with the VOA for drift time (as 1/Ko) was better than the 0.1% RSD for most chemicals and 0.4% RSD in the worst case. Peak areas were repeatable with an average of 10% RSD for three measurements for a mixture of 18

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_99.... 05/05/2007

Página 2 de 2

volatile organic compounds and a calibrant, as shown in part in Table 3–4.41 This compared favorably with a laboratorybased GC/IMS in which a selected compound in a mixture of organophosphorous compounds was determined at 88 to 1390 pg and repeatability was

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_99.... 05/05/2007

Página 1 de 1

Page 100 TABLE 3–4 Quantitative Response of the Volatile Organic Analyzer for the International Space Station

Concentration, mg/m3 Target

1

2

3

Average Std % RSD

% Error

Methanol

170

200

194

204

199

5

2.5

17.3

2-Propanol

717

833

597

672

701 121

17.2

2.3

1-Butanol

702

442

421

393

419

25

5.9

40.4

Ethanal

202

223

216

205

215

9

4.2

6.3

1793

1625

1943

1839

1802 162

9.0

0.5

o-Xylene

633

587

633

727

649

71

11.0

2.5

Toluene

414

425

561

440

475

75

15.7

14.8

2-Butanone

480

298

269

270

279

16

5.9

41.9

Ethyl acetate

279

199

315

236

250

59

23.7

10.4

Dichloromethane

177

197

191

191

193

3

1.8

9.0

m,p-Xylenes

roughly 6 and 25% RSD for 4 to 6 measurements in each of two reagent gases, water and DMSO.42 The long-term stability of calibration in IMS is rarely documented, although the single opensource example, the VOA, suggests that calibration curves may be stable within repeatability for months. Otherwise, supporting numerical evidence is difficult to obtain. One creative solution to calibration of an IMS analyzer in a changing matrix was that of Dam for monitoring stack emissions.43 In this IMS monitoring system, which had been installed and operated at a DuPont chemical production facility for over a decade without serious shortcomings in stability, an inlet flow system to the IMS was equipped to receive a flow of standard levels of the target chemical while monitoring for the target chemical. This constituted a type of flowing standard addition calibration and was able to compensate for quantitative complications resulting from changes in the matrix. Military applications of IMS require that handheld units be designed to remain in sealed containers with shelf lives of a decade, and when unwrapped, are expected to power up to operating levels within minutes. One of the rare documented instances of stability over a period of weeks is the calibration of a GC/IMS system for hydrazines before and after a handheld unit had flown on flight STS-37 of the U.S. space shuttle. The repeatability of calibration of hydrazines over a 6-week interval was 5 to 10%, which was better than the measured RSD in laboratory exercises.37 Sensitivity is defined as the increase in response per unit concentration or as the slope of the plot of detector response vs. concentration. Of interest here is the quantitative response for different compounds over a range of concentrations. Commonly, linear ranges of 10 to 100 have been reported for IMS, and working ranges can be near or larger than 1000. A typical response curve of a mobility spectrometer is shown for phenol in Figure 3–10.40 Just below

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_100... 05/05/2007

Página 1 de 1

Page 101

FIGURE 3–10 Quantitative response including precision for phenol with IMS determination of phenol on filter paper. Phenol was applied to the filter paper as a solution, dried and thermally desorbed into the IMS drift tube.40

the threshold of response (4 ng), only a very slight increase in the product ion peak area is observed. Between 40 to 100 ng, the peak area increases linearly from 6 to 16 units (Figure 3–10, bottom), i.e., it doubles in intensity when the concentration is doubled. As the amount of phenol is increased to 200 ng, the slope of the response curve changes and the peak area rises from 30 to 65 area units for 200 to 1000 ng of phenol (Figure 3–10, middle), i.e., it doubles in intensity when the concentration is quadrupled. From 2,000 to

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_101... 05/05/2007

Página 1 de 1

Page 102 10,000 ng, the peak area increases from 70 to 105 units (Figure 3–10, top) i.e., a 500% increase in concentration resulted in a 50% increase in the peak area. The amounts of phenol cited here is what was placed on the filter paper; thus, the amounts of phenol in the gas phase should be substantially lower than these amounts owing to inefficiency in the thermal desorption step and efficiency of passage of phenol through the membrane inlet of the handheld analyzer. Nonetheless, the trends in these curves are more significant here than absolute mass calibrations. The plot in Figure 3–10 (top) is what is commonly presented as the calibration curve for IMS and shows a linear range from 40 to 1000 ng, a second linear region from 1000 to 2000 ng, and a third range above 2000 ng where the ion source is nearly saturated (and reactant ions are depleted). Thus, the dynamic or working range is from 40 to 10,000 ng. Repeatability in the plots illustrates that discrimination near the threshold is possible with 10 to 40% RSD, indicating that a fine distinction can be made between 60 ng and 80 ng. In the middle range of response, the RSD is 5 to 10%, and this is reduced to ~2% at the upper range of response.

3.3.2 Effects of Experimental Parameters on Mobility Spectra 3.3.2.1 Influence of Moisture and Temperature Moisture and temperature are central factors in defining the identities of the reactant ions and the product ions, controlling ion lifetimes, and establishing quantitative response. Since 1993 and the publication of the first IMS monograph from CRC Press, Ion Mobility Spectrometry, the role of moisture and temperature have been characterized and found to be two of the most significant variables in an analytical mobility measurement. Remarkably, mobility spectrometers function well across a wide range of moisture and temperature. However, without control of these parameters, reproducibility and comparisons between laboratories may be difficult to achieve with a high level of confidence. This condition would be equivalent to operating mass spectrometers without knowledge or control of vacuum and electron energy. The role of temperature might be understood from careful application of the MS studies of Good et al.,2 Sunner et al.,14,15 and Kim et al.4 The findings showed that at a fixed moisture level, changing the temperature resulted in predictable changes in the ion identity and measured mobility through declustering of reactant ions. However, the role of temperature on the formation of product ions was not examined. A thorough examination on the effect of temperature on mobility spectra was begun by Bell in 199044 and extended by Wang in 1999.45 Temperature had two major effects on the response of a mobility spectrometer: declustering or dissociation of cluster ions and fragmentation of product ions to smaller ions with the high-mobility characteristics of chemical moieties. Declustering of all ions is evident as a decrease in the drift time owing to the reduction in the effective mass of the core ion when water molecules are stripped from the hydration shell. A concomitant change was observed in the reactivity of the reactant ion. Elevated temperatures also lead to the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_102... 05/05/2007

Página 1 de 1

Page 103

FIGURE 3–11 Mobility spectra for six alcohols from butanol to decanol at moisture of 0.1 ppm and temperatures from 100 to 225°C Electric fields were changed to adjust for temperature dependence of mobility. Thus, changes in drift time are due to changes in ion cluster size. Notable is the loss of peak intensity above 8 ms at 225°C, indicating complete fragmentation of alcohol ions.

dissociation of complex ions such as proton-bound dimers. As temperature is increased, the intensity of peaks for protonated monomer increase and the peak abundance of proton-bound dimers decrease. In extreme cases, the core ions can be fragmented in ways reminiscent of electron-impact ionization in MS. An example of fragmentation is shown in Figure 3–11, in which increasing the temperature first caused the disappearance of proton-bound dimers, then the loss of the protonated monomers, and finally the appearance of fragment ions for alcohols at 150°C.46 Other chemical classes exhibited similar behavior, although the temperature at which fragmentation of the ions occurred was class dependent. For example, proton-bound dimers of ketones underwent dissociation only above 175°C, and fragmentation was not observed until the temperature exceeded 225°C45,46 At cryogenic temperatures, ion clusters not usually observed in mobility spectra may be formed with lifetimes that are prolonged sufficiently to enable detection in ion mobility spectometers and appearance in mobility spectra. For example, proton-bound trimers of alcohols were observed when temperatures were decreased to −20°C and dissociated at temperatures from −20°C to +10°C. In comparison, proton-bound dimers of alkyl amines underwent dissociation above −30°C on a 2 to 20 msec timescale. Consequently,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_103... 05/05/2007

Página 1 de 1

Page 104 the dissociation can be observed as a distortion in the peak shape and baseline of a mobility spectrum. This can occur when the ion enters the drift region as a proton-bound dimer and is dissociated before arriving at the detector.48 Thus, under suitable conditions, IMS can be used to measure the kinetics of ion dissociations in gases at ambient pressure. The dependence of IMS response on moisture begins with the reactant ions available in the reaction region of the drift tube and the possible reactions of sample vapors. The assumption that hydrated protons will be the prime reactant ion throughout all regimes of temperature does not hold when moisture levels are reduced below 10 ppm. Two pronounced effects arise from decreasing the moisture level in the supporting atmosphere of the drift tube. As mentioned in the preceding text, the kinetics of the reactions in Equation 3–3 to Equation 3–6 are slowed down with low levels of moisture, and ions such as H2O+, N4+, and N2+ are increasingly available for reactions with sample vapors. Consequently, the normally dominant proton-transfer reactions will be replaced with chargetransfer and dissociative charge-transfer reactions. The analytical consequences of reduced moisture then include changes in the relative sensitivities of response to analytes and the formation of new types of ions that are not commonly observed. For example, product ions from alkanes are not observed in an ion mobility spectrometer in which moisture level exceeds 10 ppm; however, ionized alkanes (M.+) formed in charge-exchange reactions are observed when moisture levels are below 1 ppm.6 Fragment ions have been found for nearly all chemical families even at low temperatures when moisture levels are below 0.5 ppm.16,17 3.3.2.2 Reagent Gases and Alternate Reactant Ions Control of the composition of the reactant ion permits a user to introduce a degree of selectivity into an IMS measurement by modifying the gas-phase ion chemistry in the source region. In addition, a reagent or dopant gas can suppress background interferences, concentrate the charge reservoir into one or a few preferred ions, and simplify spectral interpretation. In addition, the use of reagent gases may resolve overlapping peaks through improvements in ion separation and increase sensitivity in certain applications. When a reagent gas is added to the reaction region, the hydrated protons in positive polarity are converted into new or alternate reactant ions as per Equation 3–18: (3–18) where R is the reagent gas and m is the number of reagent neutrals on the ion cluster. For a molecule to show response with a reactant ion of RmH+, the displacement of R by the sample (e.g., A) will be needed as per Equation 3–19: (3–19)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_104... 05/05/2007

Página 1 de 1

Page 105 In the case in which m=1, a proton-transfer reaction occurs. However, more complex reactions may occur with the displacement of water or of the reagent molecule. When the reagent gas concentration is ~1 ppm in the reaction region, all hydrated protons will be converted to the alternate reactant ion when the proton affinity of the reagent gas is substantially higher than that of water. If the enthalpy of association of A is much lower than that of R, there may not be any product ions formed even for high concentrations of A. Thus, compounds that can be regarded as matrix interferences do not participate in gas-phase ionization, and detrimental matrix effects are reduced or eliminated. When a molecule A has an enthalpy of association equal to or higher than that of the reagent gas, the reactions shown in Equation 3–19 will occur, forming product ions and decreasing the intensity of the peak of the alternate reagent ion. A careful selection of the reagent gas can thus eliminate matrix interferences while preserving response to the target analyte. This simplifies both ionization chemistry and mobility spectra. As shown schematically in Figure 3–12, a mixture of seven chemicals (A to G) with various proton affinities give a complex response with hydrated protons. This response would be difficult to interpret, and it would be impossible to obtain quantitative determinations from the mobility spectrum.49 If an alternate reagent gas such as ammonia is used, only components B, D, and G of the mixture will appear in a simplified mobility spectrum (Figure 3–12, middle frame). The selection of a reagent gas with high proton affinity such as an amine will preclude the ionization of all chemicals except D and the mobility spectrum will contain a single product ion peak. A caveat in the use of reagent gases is that analytes that have proton affinities below that of the reagent gas will provide no response or a greatly diminished one. In this case, another ion source or another reagent gas should be considered. In mobility spectrometers with acetone at an elevated concentration (~1 ppm) as the alternate reagent gas, the major reactant ion is a proton-bound dimer of acetone (Ac). In the presence of small amounts of a compound with high proton affinity such as an organophosphate, the product can be described as a mixed proton-bound dimer as shown in Equation 3–20: (3–20) Reagent gases may also be used to selectively alter the mobility of a certain ion, leading to improved separation of overlapping peaks in a mobility spectrum. In the examples discussed here, the reagent gas is distributed throughout the drift tube to avoid any ion-decomposition reactions in the drift region of cluster ions. Such reactions would readily occur in a purified air or nitrogen gas atmosphere. In the monitoring of HF in ambient air by negative-mode IMS, an overlap between the O2− peak and the HF peak prevents the determination of HF. The addition of a reagent gas, here methyl salicylate, forms a cluster ion with O2−, shifting this peak to a longer drift time. Thus, the reactantion peak

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_105... 05/05/2007

Página 1 de 1

Page 106

FIGURE 3–12 Hypothetical example of the concept of selectivity with the choice of reactant ions (RIP). (A) RIP with high proton affinity (PA) gives a response only for compound D; (B) RIP with intermediate PA gives a response only for D, B, and G; (C) RIP with low PA gives a response for all compounds A to F that have higher PAs.

is displaced from the HF peak and a complete resolution between productand reactant-ion peaks is obtained. The drift time and intensity of the HF peak are unaffected by the reagent gas, and this enables the development of a commercial HF monitor based upon IMS (Chapter 8).50 Another example in which a reagent gas is used to resolve overlapping peaks is the separation of ammonia, hydrazine (Hz), monomethyl hydrazine (MMH), and unsymmetrical dimethylhydrazine (UDMH). Product ions for these chemicals appear as a single broad peak in a water-based IMS analyzer that is also vulnerable to interferences in ambient air. The addition of acetone as a reagent gas eliminate most of the background interferences and provide partial but insufficient separation of the analyte peaks.51 Large ketones (K) such as 5-nonanone form long-lived cluster ions with the analytes, and the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_106... 05/05/2007

Página 1 de 1

Page 107

FIGURE 3–13 Drift times for hydrazines and ammonia with chemical modification of the internal atmosphere of a chemical agent monitor. A ketone, 5-nonanone, was introduced into the drift gas at concentrations from 10 ppb to 1.5 ppm. Clustering between the ketone and the protonated bases caused shifts in drift times and resolution between ammonia and the hydrazines. The effect of clustering was pronounced between 10 and 1000 ppb of 5-nonanone in the drift gas.37

number of ketone molecules in the cluster depends on the number of sites available for proton binding. Thus, the cluster ion structures were: NH4+K4, HzH+K3, MMH*H+K2, and UDMH*H+K. These were formed at high vapor levels of ketone in the drift tube, and the order of drift times followed the degree of clusterization and not the molecular weight of the analyte, as shown in Figure 3–13. Another advantage of the use of reagent gases has been mentioned in Section 3.3.1.2, where the detection limit for EGDN was improved more than tenfold by the use of Cl− as a reactant ion.35 The use of halide ions for enhanced sensitivity has been particularly effective in the detection of aliphatic and aromatic explosives. Indiscriminate operation of an IMS analyzer may lead to unexpected responses due to the inadvertent presence of impurities that act as a de facto reagent gas and produce unexpected and often unwelcome alternate reactant ion chemistry. These may arise from impurities in the drift gas, contaminants in the drift tube, or trace impurities in the reagent gas. In such cases, the response obtained may not meet the expectations either in magnitude or in quality of response. Because the use of a reagent gas affects the quantitative response in IMS, the calibration of the analyzer should be made under the same conditions of measurement, i.e., with the same reagent gas at the same concentration.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_107... 05/05/2007

Página 1 de 1

Page 108

3.4 The Interpretation of Mobility Spectra 3.4.1 An Integrated Model In the preceding sections, the ionization chemistry for the formation of reactant and product ions in the reaction region of an IMS has been described. In most cases, the mobility spectrum can be interpreted using the reactions in the preceding discussion. Thus, mobility spectra have often been directly associated with the ion chemistry of the source region, and this may seem reasonable because product-ion mobility spectra vary as sample levels are changed in the reaction region. In a conventional IMS drift tube with unidirectional flow, ions pass from the reaction region containing the sample into the drift region containing a gas of uniform composition, typically purified air, as shown in Figure 3–14 (top). Sample neutrals are prevented from entering the drift region and rapidly purged from the reaction region. In a well-designed drift tube with bidirectional flow (Figure 3–14 [bottom]), a comparable condition exists, in which the sample is released through a vent from the drift tube near the

FIGURE 3–14 Two flow patterns for mobility spectrometers. In the top graphic, sample is introduced into the drift tube and drift gas sweeps the sample through the ion source and out of the drift tube. In the bottom graphic, sample is carried into the ion source and flows are drawn from the drift tube near the ion shutter.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_108... 05/05/2007

Página 1 de 1

Page 109 ion shutter. In either design, the drift region should be free of sample neutrals. The transit of product ions through a clean gas atmosphere may produce reactions that are not immediately evident to a user or analyst. Indeed, the failure to observe a response for a chemical in IMS may be associated with a low stability of the product ion which dissociates in the purified gas atmosphere of the drift region. Such reactions will depend on the kinetics of dissociation of product ions (positive or negative) when removed from the vapor rich environment of the reaction region; in a clean gas environment, product ions can only undergo back reactions to those shown in Equations 3–12 to 3– 16. These concepts are shown graphically in Figure 3–15. 3.4.1.1 Long-Lived Ions Ions that have lifetimes greater than that of the drift time will pass through the drift region essentially intact and unchanged, yielding a distinct spectrum

FIGURE 3–15 Ion stability and observed response in mobility spectra. In the top graphic, the product ion is stable and traverses the drift region without any dissociation or decomposition reactions. A product ion peak is observed in the mobility spectra at a characteristic drift time. In the middle graphic, a product ion is injected into the drift region and undergoes dissociation before arriving at the detector. A distortion in the baseline is usually evident of such reactions. In the bottom graphic, the product ion is sufficiently unstable so dissociation or decomposition is complete in the reaction region. The mobility spectrum also shows a single peak at reproducible drift time.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_109... 05/05/2007

Página 1 de 1

Page 110 (Figure 3–15 [top]) characterized by a symmetrical peak and an undistorted baseline. This is commonly found in the mobility spectra for robust ions including amines, organophosphates, and ketones. 3.4.1.2 Short-Lived Ions Ion with lifetimes that are much shorter than the time needed for drift to the detector will undergo decomposition or fragmentation reactions before reaching the ion shutter or shortly after entering the drift region. These ions also appear in mobility spectra as symmetrical peaks without baseline distortion (Figure 3–15 [bottom]). However, drift times are comparatively short and are disproportionate to the parent ion and the molecular weight of the compound from which it was formed. A well-known example is the dissociative electron capture of haloalkanes where only peaks for halide ions are observed in negative-ion mobility spectra. 3.4.1.3 Ions with Intermediate Lifetimes If an ion has a lifetime between 1 and 10 msec in a conventional drift tube, the mobility spectrum may reflect fairly complex events occurring with the product ion. Some fragment ions may be formed in the source and arrive at the detector without further change due to short drift times that are typical of fragment ions. Other ions will enter the drift region intact and undergo fragmentation as the ion swarm traverses the drift region. In this case, drift times will be a weighted average of that for the parent ion and the fragment ions. Because ion decomposition could occur throughout the length of the drift tube, a broad range of weighted drift times may be seen and would be evident as a distorted baseline. Such baseline distortions may be used to determine the kinetics of decomposition of ions in air at ambient pressure.43 A portion of the ions formed in the ion source may reach the detector intact, showing a symmetrical peak. The combination of these three possibilities is shown in Figure 3–15 (middle). In practice, the lifetime of an ion is influenced by the temperature and the moisture level so that an ion that fragments at an elevated temperature may remain intact at a lower temperature, as shown in Figure 3–11.

3.4.2 Response to Mixtures The response of an ion mobility spectrometer to a chemical mixture can be complex owing to the distribution of charge in an ionization source on a competitive basis, the number of reaction pathways, and the coexistence of multiple reactant ions. This can be partially controlled by the discriminate use of a reagent gas, as described in Section 3.3.2.2. In MS, simultaneous introduction of more than a few compounds also yields a complex mass spectrum that is difficult to interpret even when no ion chemistry between components occurs. In an ion mobility spectrometer, in which the resolution

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_110... 05/05/2007

Página 1 de 1

Page 111 of ions is low and collisions are frequent among constituents, the mobility spectrum is too complex to extract quantitative chemical information about all the components of a multicomponent sample. In some instances, competitive ionization may mask the presence of some compounds. This is both a weakness and a strength of IMS because preferential ionization is the basis of the most successful applications of IMS in which the target molecule is preferentially ionized in the presence of other compounds. In this discussion, a matrix is defined as background composition of little analytical interest, whereas in a mixture, several components of the sample have analytical value. In an IMS analyzer, some components of a mixture may affect the quantitative determination of other components. In a mixture with two or three constituents of comparable proton affinities,49 quantitative response was observed for the components of the mixture. However, more than three constituents resulted in complex interdependent response and mobility spectra that eluded quantitative determination of the components. This occurred because some peaks could not be detected as the product ion was not formed (i.e., it was not favored in competitive ionization). Such responses in IMS may be generally observed provided the components of the mixture are at trace level and have comparable ionization properties. Examples of this are rare. Just as chromatographic preseparation has become integral to the use of mass spectrometric measurements with mixtures, so also chromatographic combinations with mobility spectrometers increase the analytical value for the comprehensive analysis of complex mixtures. In hyphenated GC/IMS or LC/IMS, the preseparation simplifies the chemistry in the ion source by minimizing the number of constituents present at a given moment in the reaction region (such methods are described in Chapter 5). Thus, product ions are formed without interference from other constituents. In some cases in which coelution occurs, the additional dimension of characterization provides analytical value for identification and quantitative determination. In a sample containing an analyte in a matrix including a number of other constituents of lesser or no importance, the matrix may affect the response of the analyte. This constitutes one of the most difficult considerations when an ion mobility spectrometer is assessed as a prospective chemical sensor. When potential interferences have ionization properties that favor charges residing on the target molecule, matrix effects are minor and may not affect quantitative response. In cases in which matrix components interfere with the ionization processes of the analyte, IMS may become useless. The use of other ionization sources or the attachment of a gas chromatograph to the inlet can possibly alleviate the situation. When the matrix attenuates the quantitative response of IMS, the technique of standard addition may be adapted for IMS, as demonstrated by Dam.43 In this approach, a blend of the sample and the calibration compound were introduced into the IMS in a pulsed fashion. The ratio of the total response of the ion mobility spectrometer to the increase in response with the known addition was automatically determined, just as in traditional standard addition methods. The advantages of this standard addition approach

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_111... 05/05/2007

Página 1 de 1

Page 112 to overcome the presence of matrix interferences were also demonstrated with flow injection IMS.38 The results showed that standard addition was suitable for some extreme cases of matrix effects.

3.4.3 Use of Mobility Spectra for Chemical Identification In practical applications of IMS, the drift time of a peak is converted to a reduced mobility that is characteristic of an ion. Under controlled operating parameters, the ion is distinctive of the compound from which the ion was derived. This may be compared directly to a list of reference values of mobility coefficients obtained under identical conditions for authentic standards. Methods of handling mobility spectra parallel techniques employed elsewhere in analytical chemistry, including separation methods in which two-dimensional information is obtained and the empirical determination is dependent on the conditions of the measurement. For a given compound, mobility coefficients are highly reproducible and will give a unique value. However, several compounds may yield similar mobility coefficients, a situation that is reminiscent of coeluting peaks in chromatographic methods. The range of mobility coefficients in most applications of IMS is from 1.2 to 2.2 cm2 V−1sec−1, and the accuracy is typically no better than 0.01 cm2 V−1sec−1. Consequently, there may be several compounds for each mobility coefficient, and identification based upon mobility coefficient alone, without additional information on the sample, may lead to misinterpretation. As in the preceding discussions, temperature and moisture strongly affect ion identities and thus mobility coefficients. During the past decade, awareness of this has grown, and standards for calibration of the mobility scale have been developed.52 This also resulted in the development of spectral libraries where reliable data regarding specific temperatures and moistures in the drift gas are referenced.53 A special library is available on the CD-ROM accompanying this book.

3.5 Summary Gas-phase ion chemistry in air at ambient pressure is the strength of IMS as an analytical method that provides useful detection limits. Enhanced selectivity may be achieved by controlling the ion chemistry with the addition of reagent gases. Alternatively, IMS may serve as a universal analyzer when gas atmospheres permit the use of charge-exchange reactions in combination with protonbased chemistry. All chemicals from alkanes to organophosphorus compounds can be determined under such conditions. At present, the link between the chemistry in the ion source and the observed mobility spectrum is clear and also involves understanding of the processes the ions undergo as they move through a cleaned-gas atmosphere in the drift region.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_112... 05/05/2007

Página 1 de 1

Page 113 Moving long-lived, stable ions through the clean-gas atmosphere provides sharp and distinctive peaks, as these ions travel as well-characterized, homogeneous ion swarms with no changes in structure that would otherwise arise if the ions moved through sample vapors. However, the cleangas atmosphere creates a type of ion filter so that only ions with lifetimes greater than the drift time will arrive at the detector and create a response. This is the basis for the simplicity of mobility spectra (no cluster ions with more than two sample neutrals attached to the core ion) and also the limitation (poor response to ions with short lifetimes) of such spectrometers. Authors’ Note: This chapter reflects the contributions, involvement, and influence of Professor John Stone in the authors’ laboratory and research program. The comparison of this chapter with the treatment in the previous edition will illustrate the extent of Prof. Stone’s role in guiding the thinking and development in bridging the worlds of gas-phase ion-molecule chemistry and analytical chemistry. He has been a willing guide in our development, and without his help this discussion would be weaker in depth and breadth.

References 1. Siegel, M.W., Atmospheric pressure ionization, in Plasma Chromatography, Carr, T.W., Ed., Plenum Press, New York, 1984, chap. 3, pp. 95–113. 2. Good, A.; Durden, D.A.; Kebarle, P, Ion-molecule reactions in pure nitrogen and nitrogen containing traces of water at total pressures 0.5–4 torr. Kinetics of clustering reactions forming H+(H2O)n, J. Chem. Phys. 1970, 52, 212–221. 3. Shahin, M.M., Mass-spectrometric studies of corona discharges in air at atmospheric pressures, J. Chem. Phys. 1966, 45(7), 2600–2605. 4. Kim, S.H.; Betty K.R.; Karasek, F.W., Mobility behavior and composition of hydrated positive reactant ions in plasma chromatography with nitrogen carrier gas, Anal. Chem. 1978, 50(14), 2006–2016. 5. Preston J.M.; Rajadhyax, L., Effect of ion-molecule reactions on ion mobilities, Anal. Chem. 1988, 60(1), 31–34. 6. Bell, S.B.; Ewing, R.G.; Eiceman, G.A.; Karpas, Z., Characterization of alkanes by atmospheric pressure chemical ionization mass spectrometry and ion mobility spectrometry, J. Am. Soc. Mass Spectrom. 1994, 5, 177–185. 7. Tou, J.C.; Ramstad, T.; Nestrick, T.J., Electron mobility in a plasma chromatograph, Anal. Chem. 1979, 51, 780–782. 8. Stockdale, J.A.; Christophorou, L.G.; Hurst, G.S., Capture of thermal electrons by oxygen, J. Chem. Phys. 1967, 47(9), 3267–3269. 9. Spangler, G.E.; Carrico, J.P., Membrane inlet for ion mobility spectrometry (plasma chromatography), Intl. J. Mass Spectrom. Ion Phys. 1983, 52, 267–287. 10. Spangler, G.E.; Collins, C.I., Reactant ions in negative ion plasma chromatography, Anal Chem. 1975, 47, 393–402. 11. Carr, T.W., Comparison of the negative reactant ions formed in the plasma chromatograph by nitrogen, air, and sulfur hexafluoride as the drift gas with air as the carrier gas, Anal Chem. 1979, 51, 705–711.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_113... 05/05/2007

Página 1 de 1

Page 114 12. Carr, T.W., Negative ions in plasma chromatography-mass spectrometry, Anal. Chem. 1977, 49, 828–831. 13. Harrison, A.G., Chemical Ionization Mass Spectrometry, CRC Press, Boca Raton, FL, 1986, p. 87. 14. Sunner, J.; Nicol, G.; Kebarle, P., Factors determining relative sensitivity of analytes in positive mode atmospheric pressure ionization mass spectrometry, Anal. Chem. 1988, 60, 1300–1307. 15. Sunner, J.; Ikonomou, M.G.; Kebarle, P., Sensitivity enhancements obtained at high temperatures in atmospheric pressure ionization mass spectrometry, Anal. Chem. 1988, 60, 1308–1313. 16. Bell, S.E.; Nazarov, E.G.; Wang, Y.F.; Eiceman, G.A., Classification of ion mobility spectra by chemical moiety using neural networks with whole spectra at various concentrations, Anal. Chim. Acta 1999, 394, 121–133. 17. Bell, S.E.; Nazarov, E.G.; Wang, Y.F.; Rodriguez, J.E.; Eiceman, G.A., Neural network recognition of chemical class information in mobility spectra obtained at high temperatures, Anal. Chem. 2000, 72, 1192– 1198. 18. Karasek, F.W.; Denney, D.W., Role of nitric oxide in positive reactant ions in plasma chromatography, Anal. Chem. 1974, 46(6), 633–637. 19. Eiceman, G.A.; Kelly, K.; Nazarov, E.G., Nitric oxide as a reagent gas in ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2002, 5, 22–30. 20. Eiceman, G.A.; Bergloff, J.A.; Karpas, K., Proton abstraction and association reactions in atmospheric pressure negative chemical ionization (APNCI) of fluorinated phenols, J. Am. Mass Spectrom. 1999, 10, 1157–1165. 21. Ewing, R.E.; Ewing, G.J.; Atkinson, D.A.; Eiceman, G.A., A critical review of ion mobility spectrometry for the detection of explosives and explosive related compounds, Talanta 2001, 54, 515–529. 22. Eiceman, G.A.; Shoff, D.B.; Harden, C.S.; Snyder, A.R; Martinez, P.M.; Fleischer M.E.; Watkins, M.L., Ion mobility spectrometry of halothane, enflurane, and isoflurane anesthetics in air and respired gases, Anal. Chem. 1989, 61, 1093–1099. 23. Spangler, G.E.; Collins, C.I., Peak shape analysis and plate theory for plasma chromatography, Anal. Chem. 1975, 47, 403–407. 24. Cohen, M.J.; Karasek, F.W., Plasma Chromatography™—a new dimension for gas chromatography and mass spectrometry, J. Chromatogr. Sci. 1970, 8, 330–337. 25. Karasek, F.W.; Keller, R.A., Gas chromatograph/plasma chromatograph interface and its performance in the detection of musk ambrette, J. Chromatogr. Sci. 1972, 10, 626–628. 26. Long, G.L.; Winefordner, J.D., Limits of detection: a closer look at IUPAC definitions, Anal. Chem. 1983, 55, 712A–714A. 27. Baim, M.A.; Hill, H.H., Jr., Determination of 2,4-dichlorohenoxyacetic acid in soils by capillary gas chromatography with ion mobility detection, J. Chromatogr. 1983, 279, 631–642. 28. St. Louis, R.H.; Siems, W.F.; Hill, H.H., Jr., Evaluation of direct axial sample introduction for ion mobility detection after capillary gas chromatography, J. Chromatogr. 1989, 479, 221–231. 29. Eiceman, G.A.; Harden, C.S.; Wang, Y.-F.; Garcia-Gonzalez, L.; Schoff, D.B., Enhanced selectivity in ion mobility spectrometry analysis of complex mixtures by alternate reagent gas chemistry, Anal. Chim. Acta 1995, 306, 21–33. 30. Tou, J.C.; Boggs, G.U., Determination of sub parts-per-million levels of secbutyl chlorodiphenyl oxides in biological tissues by plasma chromatography, Anal. Chem. 1976, 48, 1351–1357.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_114... 05/05/2007

Página 1 de 1

Page 115 31. Watson, W.M.; Kohler, C.F., Continuous environmental monitoring of nickel carbonyl by fourier transform infrared spectrometry and plasma chromatography, Environ. Sci. Technol. 1979, 13, 1241–1243. 32. Karpas, Z.; Pollevoy, Y; Melloul, S., Determination of bromine in air by ion mobility spectrometry, Anal. Chim. Acta 1991, 249, 503–507. 33. Eiceman, G.A.; Sowa, S.; Lin, S.; Bell, S.E., Ion mobility spectrometry for continuous on-site monitoring of nicotine vapors in air during the manufacture of transdermal systems, J. Hazard. Mater. 1995, 43, 13–30. 34. Karasek, F.W.; Denney, D.W., Detection of 2,4,6-trinitrotoluene vapours in air by plasma chromatography, J. Chromatogr. 1974, 93, 141–147. 35. Lawrence, A.H.; Neudorfl, P., Detection of ethylene glycol dinitrate vapors by ion mobility spectrometry using chloride reagent ions, Anal. Chem. 1988, 60, 104–109. 36. Schellenbaum, R.L.; Hannum, D.W., Laboratory Evaluation of the PCP Large Volume Ion Mobility Spectrometer (LRVIMS), Sandia Report SAND89–0461*UC515, March 1990. 37. Eiceman, G.A.; Salazar, M.R.; Rodriguez, M.R.; Limero, T.F.; Beck, S.W.; Cross, J.H.; Young, R.; James, J.T., Ion mobility spectrometry of hydrazine, monomethylhydrazine, and ammonia in air with 5-nonanone reagent gas, Anal. Chem. 1993, 65, 1696–1702. 38. Eiceman, G.A.; Garcia-Gonzalez, L.; Wang, Y.-F.; Pittman, B.; Burroughs, G.E., Ion mobility spectrometry as flow-injection detector and continuous flow monitor for aniline in hexane and water, Talanta 1992, 39, 459–467. 39. Poziomek, E.J.; Eiceman, G.A., Solid-phase enrichment, thermal desorption, and ion mobility spectrometry for field screening of organic pollutants in water, Environ. Sci. Technol. 1992, 26, 1313–1318. 40. Smith, G.B.; Eiceman, G.A.; Walsh, M.K.; Critz, S.A.; Andazola, E.; Ortega, E.; Cadena, F., Detection of Salmonella typhimurium by hand-held ion mobility spectrometer: a quantitative assessment of response characteristics, Field Anal. Chem. Technol. 1997, 4, 213–226. 41. Limero, T.; Reese, E.; Trowbridge, J.; Hohman R.; James, J.T., The Volatile Organic Analyzer (VOA) Aboard the International Space Station, Paper Offer Number 02ICES-317, 2002, Society of Automotive Engineers. 42. Eiceman, G.A.; Wang, Y.-F.; Garcia-Gonzalez, L.; Harden, C.S.; Shoff, D.B., Enhanced selectivity in ion mobility spectrometry analysis of complex mixtures by alternate reagent gas chemistry, Anal. Chim. Acta 1995, 306, 21–33. 43. Dam, R., Analysis of toxic vapors by plasma chromatography, in Plasma Chromatography, Carr, T.W., Ed., Plenum Press, New York, 1984, pp. 177–214. 44. Bell, S.E., Ion Mobility Spectrometry of Selected Organic Compound Classes, Ph.D. dissertation, New Mexico State University, Las Cruces, NM, May 1991. 45. Wang, Y-F., Effects of Moisture and Temperature on Mobility Spectra of Organic Chemicals, MS thesis, New Mexico State University, June 1999. 46. Zhou, Q., Fragmentation of Gas Phase Ions at One Atmosphere in Ion Mobility Spectrometry, M.S. thesis, New Mexico State University, May 2001. 47. Ewing, R.E., Stone, J.A.; Eiceman, G.A., Proton bound cluster ions in ion mobility spectrometry, Int. J. Mass Spectrom. 1999, 193, 57–68. 48. Ewing, R.E., Kinetic Decomposition of Proton Bound Dimer Ions with Substituted Amines in Ion Mobility Spectrometry, Ph.D. dissertation, New Mexico State University, December 1996.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_115... 05/05/2007

Página 1 de 2

Page 116 49. Eiceman, G.A.; Blyth, D.A.; Shoff, D.B.; Snyder, A.R, Screening of solid commercial pharmaceuticals using ion mobility spectrometry, Anal Chem. 1990, 62, 1374–1379. 50. Spangler G.E.; Epstein, J., Detection of HF using atmospheric pressure ionization (API) and ion mobility spectrometry (IMS), Paper, The 38th ASMS Conference on Mass Spectrometry and Allied Topics, Tucson, Arizona, June 1990. 51. Bollan, H.R., The Detection of Hydrazine and Related Materials by Ion Mobility Spectrometry, Ph.D. dissertation, Sheffield Hallam University, England, March 1998. 52. Eiceman, G.A.; Nazarov, E.G.; Stone, J.A., Chemical standards in ion mobility spectrometry, Anal Chim. Acta 2003, 493, 185–194. 53. http://www.chemistry.nmsu.edu/eiceman_research/spec_library.html

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_116... 05/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_116... 05/05/2007

Página 1 de 2

Page 117

Part II Technology of Ion Mobility Spectrometry

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_117... 05/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_117... 05/05/2007

Página 1 de 1

Page 119

4 Drift Tubes for Mobility Spectrometers 4.1 Introduction An ion mobility spectrometer is composed of a central component, the drift tube, where ion formation and characterization occur, and other components that support the measurement made with the drift tube. These other components may be categorized as the utilities, including power supplies, heaters, and gas supply, and the electronics, including detector amplifiers, signal processor, and ion-shutter controller. These are shown in a block diagram for a generalized arrangement in Figure 4–1. The drift tube can be further divided into an ion source, a reaction region, a drift region, and a detector (as shown in Figure 4–2). Although every ion mobility spectrometry (IMS) analyzer will share some or all of the generic design in Figure 4–1, identifying a standard or average IMS instrument is not possible due to the variety of components and designs of drift tubes, the possible configurations for inlets, materials of components, and electronics. Although IMS has been under development for over 30 years as a modern analytical technique, there is little uniformity in analyzer designs, and distinctiveness has actually increased during the past decade with new materials and components, innovations in ion formation and separation, and new commercial analyzers. An exception to this generalization can be found in the instruments produced in large numbers for chemical weapons detection and explosives determinations. In such analyzers, uniformity in construction and performance within a model line is desired, and analyzers are held to defined specifications. Another special feature of commercial analyzers is that the components seen in Figure 4–1 and Figure 4–2 are all integrated into a single package and may not be recognizable as discrete items. In contrast, research-grade instruments can be built from individual components that are available commercially as separate packages, except the ion-shutter controller. This is a specialized electronic unit and is unavailable as a standard item from vendors. With either commercial or user-built instrumentation, choices are made regarding the components of the drift tube and the design and manufacture

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_119... 05/05/2007

Página 1 de 1

Page 120

FIGURE 4–1 A block diagram of the main components of a generic mobility spectrometer with electric field gradient.

of components. The properties and specifications of drift-tube components will affect the overall performance of an analyzer and may significantly influence the suitability of a configuration for a specific application. This chapter was written to describe broadly the variety of possibilities with

FIGURE 4–2 Diagram of the drift tube of an ion mobility spectrometer. The sample vapors enter the reaction region, where ionization and ion formation take place. Ions are injected through the ion shutter into the drift region and arrive at the detector plate with drift times that depend on drift velocity.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_120... 05/05/2007

Página 1 de 1

Page 121 designs of IMS analyzers, and emphasis is given to the components. Accordingly, the chapter is divided into five sections: 1. The sample delivery, or inlet, systems for gaseous, liquid, and solid samples (Section 4.2) 2. The methods and sources used to produce ions at ambient pressure (Section 4.3) 3. The different designs of drift tubes for ion separation (Section 4.4) 4. The methods for ion detection and acquisition, analysis, and display of mobility spectra (Section 4.5) 5. The properties and limitations of materials used to fabricate components of a drift tube (Section 4.6) Included in the CD accompanying this book are schematics for the design of a research-grade drift tube, ion-shutter electronics, circuits for a highvoltage supply, and an amplifier designed for bandwidths of analytical mobility spectrometers. This chapter is organized around the components of IMS analyzers and both research and commercial instruments can be understood as assemblies of components. Thus, an evaluation of capabilities or performance of any instrument should be undertaken mindful of the properties and interdependence of the components. Discussions of drift tubes with preseparation methods, including gas chromatography (GC), liquid chromatography (LC), and supercritical fluid chromatography, or those with advanced detectors such as mass spectrometers can be found in Chapter 5. Finally, supplementary discussions of IMS instrumentation can be found in two monographs1,2 and numerous reviews published in English,3–9 Russian,10 German,11–13 Polish,14 and Chinese.15

4.2 Inlets and Introduction of Sample 4.2.1 General Considerations Ion mobility spectrometers are designed to separate gas-phase ions that are formed from molecules in a sample delivered to the ion source or reaction region. Inlet systems or methods to deliver sample to the drift tube must satisfy a requirement that gas-phase ions be produced from the sample at ambient pressure, often in air. A second requirement is that the IMS measurement provides an analytically reliable measure of the sample; that is, the inlet or interface between the sample and analyzer should not appreciably skew or distort the chemical information sought from the sample. The earliest modern IMS analyzers were used for characterizing vapors or gases in controlled atmospheres or laboratory settings. In these measurements, the gaseous sample was drawn or pushed directly into the reaction region.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_121... 05/05/2007

Página 1 de 1

Page 122 This approach was so simple and effective that few variations could be made on the inlet design. Infield vapor monitoring could be enhanced through online calibration of the analyzer, and reagent gases could be added to the drift tube to reduce chemical interference from the sample matrix. In the early use of IMS for determining semivolatile or nonvolatile compounds in liquids or solids, the sample was deposited from solution onto a platinum wire attached to a glass rod. After the solvent was evaporated in air, the wire was inserted into a heated gas flow near the ion source. Desorbed vapors were swept by gas flow into the ion source and reaction region. Such methods were valuable only when samples were comparatively simple and measurements made in laboratories. Increases in the scope of IMS applications during the past 10 to 20 years and in the complexity of samples analyzed by IMS instruments have necessitated the development of specialized inlet systems or methods to deliver the sample to the drift tube. Examples include membrane inlets to exclude moisture, dust, and interfering chemicals from the reaction region; preconcentration devices for enriching trace amounts of substances in samples; and thermal desorbers for compounds with low vapor pressures as particulate on filters. Other advances were made with the delivery of liquid samples directly to a drift tube through electrospray ionization and the direct introduction of solid samples through laser-based techniques such as laser desorption and ionization. In 2005, a variety of inlets or interfaces are found on commercial instruments and on mobility spectrometers in research laboratories. The choice of an inlet system will affect the analytical performance of the drift tube and should be informed by some understanding of the matrix, the analyte, the anticipated range of amount or concentration, and practical constraints of cost, size, weight, and power consumption. The inlet is not a passive element in a mobility spectrometer and is often used in conjunction with reagent gases to improve the selectivity of response. Both the inlet and reagent gas can aid user-tailored response, which is analytically advantageous when an analyzer is used without an interface to a chromatograph. Inlet configurations may govern subsequent decisions on a reagent gas, and the topics are often interdependent. Consequently, a discussion of reagent gases has been included in this section, and additional examples of the use of reagent gases in IMS are described in Chapter 6 to Chapter 9. Although the chemical basis for reagent gases is discussed in Chapter 3, examples and the mechanics for reagent gases are emphasized in the following text.

4.2.2 Reagent Gases The number of possible reagent gases for IMS is vast in principle, although comparatively few are actually used for routine analyzers or research instruments. Reagent gases in positive polarity include acetone, ammonia, dimethylsulfoxide (DMSO), nicotinamide, and nonylamine. In negative polarity, small chlorocarbons primarily are used to produce chloride ions. The strategy underlying the use of reagent gases may be summarized in the form of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_122... 05/05/2007

Página 1 de 1

Page 123 two simple generalizations: control of proton transfer in positive polarity and control of anionattachment reactions in negative polarity. In positive-ion chemistry, a reagent gas may enhance the specificity of response in proton transfer reactions by discriminating against molecules with low proton affinity. In negative polarity, the major pathways for negative-ion formation in the IMS are charge-transfer reactions, ion-attachment or clustering reactions, and electron attachment, followed by stabilization of the product ion or dissociation into fragment ions. Although the chemical basis for ionization with reagent gases can be understood generally (Chapter 3), the selection of an optimum reagent gas is still in an empirical stage of development.16 Several flow configurations may be used to introduce reagent gases into a drift tube. Vapors can be passed directly into the reaction region, the carrier gas stream, or gas flows throughout the entire drift tube, as shown in Figure 4–3A to Figure 4–3C, respectively. The choice of flow pattern is governed by design requirements for the instrument and by some subtle details. For example, the pattern in Figure 4–3C is used in handheld IMS analyzers, where product ions are formed by reactions between sample (M) and a reactant ion that is a protonbound dimer of acetone (Ac) as shown in Equation 4.1: (4.1) In order to preserve the mixed proton-bound dimer (MAcH+) through the drift region, vapors of acetone must be present throughout the drift region.18 Other reagent gases such as methanol, acetic acid, and DMSO were used in comparable flow patterns without complete success, presumably due to excessive clustering of reagent gas with ions in the drift region.18 In other cases, the objective of using a reagent gas is to increase the sensitivity and specificity of gas-phase chemistry in the reaction region only, and flow patterns in Figure 4–3A and Figure 4–3B could be used. In such an arrangement, the reagent gas is used to control ionization chemistry in the reaction region, and product ions are characterized in the drift region containing purified air or nitrogen. Here, DMSO exhibited increased selectivity in ion formation.19 Although ammonia has been shown effective as a reagent gas20 and is used in some military-grade analyzers, ammonia can form cluster ions, thus mitigating selectivity through proton affinities and was discounted in a study of reagent gases.19 Reagent gases may be used with nonradioactive ion sources (see Subsection 4.3.2) and may enhance ionization effectiveness, as demonstrated with naphthalene and photoionization with a microchip laser. Vapors of naphthalene were ionized by multiphoton ionization at 266 nm, and this was followed by secondary charge-transfer reactions with an analyte.21 This was described for measurements using two ionization sources with two different reagent gases. Selectivity of response was governed by principles of ionization energy. In the same instrument, a second source of common 63Ni was operated with ammonia as the reagent gas. In the second source, selectivity of ionization occurred through proton affinities with proton transfer reactions.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_123... 05/05/2007

Página 1 de 1

Page 124 Thus, response to a sample was obtained through two independent principles of ion formation, enhancing the analytical capabilities of the measurement. In negative polarity, a chloride ion has been the preferred reactant ion, which may be formed from reagent gases such as chloroform, methylene chloride, and other chlorocarbons.18,22 When the reactant ions in negative polarity are simplified into a single-ion species such as Cl−(H2O)n or Br− (H2O)n, product ions from a compound can be coalesced into one or a few peaks, improving the detection limits. This is seen with molecules that tend to fragment and was best demonstrated with ethylene glycol dinitrate and other explosives.22,23 Reagent gases may also be used to improve the separation between product ions and reactant ions by forming cluster ions and selectively increasing the effective ion mass of either the reactant or product ions. In the IMS determination of hydrogen fluoride (HF), the mobility of the product ions was close to that of the reactant ions, and the mobility of the reactant ions was altered by adding a reagent gas that clustered selectively with the reactant ions.24 Methylsalicylate was added to the gas atmosphere of the drift tube, and the reactant ion O2–(H2O)n was converted to an adduct ion with methylsalicylate. The product ion for HF was unaffected by the addition of the reagent gas. Because this adduct ion had a greater drift time than that of the O2–(H2O)n and the drift time of the product ion for HF was largely unaffected; the peak for HF was then well separated from that for the reactant ion.24 The main method used to generate a continuous and stable supply of a reagent gas in the supporting atmosphere or gas flow of the drift tube is the permeation source as shown in Figure 4–3 (inset). A permeation source consists of a short length of plastic tubing filled with the reagent, which is often a liquid or solid at ambient pressure and temperature; the tube is sealed or plugged at the ends.25 Vapors from the reagent permeate through the tubing wall and are transferred from the outer surface of the tube into the gas flow around the permeation source. The rate of permeation is controlled by the tubing material, wall thickness, tube length, temperature, and vapor pressure of the reagent in predicable ways.25,26 The concentration of the reagent gas in the source region is governed by the permeation rate and by the rate of gas flow used to sweep vapors from the permeation tubes. These variables are arranged to reach a concentration of 0.1 to 10 ppm for the reagent gas in the drift or reactant regions, and exact values can be calculated from gravimetric analysis of the permeation tube.27 Permeation sources emit vapors continuously, and their operating life expires when the reservoir of chemical is depleted. Thus, an IMS analyzer equipped with a permeation source will have a shelf life limited by depletion of the reagent, and maintenance may be needed when instruments are left unused for prolonged periods. Also, when a mobility spectrometer with a permeation source is unused for an extended time, reagent gas will accumulate in the flow system, and high vapor levels of the gas will flood the supporting atmosphere when the instrument is powered on. This excess will cause distortions in the mobility spectrum and will persist until steady-state conditions are reached with a flow from a fresh supply of gas or as the vapors

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_124... 05/05/2007

Página 1 de 1

Page 125

FIGURE 4–3 Flow configurations for drift tubes. Reagent gas can be introduced: (A) directly into the ionization region, (B) into the carrier gas stream, (C) throughout the entire drift tube. Gas is removed (not shown) from the drift tubes at the side near the ion shutter or at the source end of the tube.

are scrubbed in a recycled flow system. In practice, 5 to 10 min may be needed for the instrument to reach a stable and functional condition.

4.2.3 Gases, Vapors, and Ambient Air The direct introduction of sample into a drift tube is rarely used in IMS except when purified gases are being assayed for impurities.28–30 In such cases, the sample is brought to the reaction region with as little contact as possible with materials or surfaces where impurities in the gas might be absorbed or that might introduce impurities into the gas sample. In other IMS analyzers, the inlet system is the interface between the ion source or reaction region and the ambient atmosphere or sample. Suitable inlets are vital in the operation of field instruments and serve to isolate, in the case of membranes, the purified internal atmosphere of the analyzer from the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_125... 05/05/2007

Página 1 de 1

Page 126 ambient air being sampled.11 Membranes are thin synthetic polymer films through which organic compounds diffuse preferentially over moisture. Additionally, membranes prevent dust or other particulate matter from entering the drift tube. Other inlets include devices to control quantitative levels of sample before analysis or to prefractionate a sample by chromatography Regardless of the type or principle of operation, inlets may be characterized by efficiency and enrichment as defined for membranes with gas chromatograph/mass spectrometer.31 The yield is the percent of sample successfully passed through the inlet or interface to the drift tube. Enrichment is the yield multiplied by the ratio of gas flows into and from the inlet. In a handheld ion mobility spectrometer, the flow to the membrane is 500 ml/min, and the flow against the inside surface of the membrane is ~100 ml/min. Thus, this ratio of enrichment through flows would be ~5, and a yield of 15% for a compound providing 75% improvement in response. All inlets have a possible deleterious effect on the sample in contact with the surfaces of an inlet. Even with a seemingly passive apparatus such as an exponential dilution flask, losses through adsorption on surfaces can lead to poor yields. Consequently, the design and materials for inlets affect the overall analytical performance and practical value of the IMS apparatus; materials are discussed in Section 4.6. A last aspect of sampling gases or atmospheres is the definition of an IMS analyzer as a point detector. A single IMS analyzer can provide a continuous record of the presence of a chemical at the location of the analyzer (this can be converted to a concentration vs. time profile when the analyzer is calibrated).32 However, modifications to the inlet can extend the capabilities of a single analyzer to a multipoint sampling system where tubing is used to bring sample into a single analyzer, in near real time, from several locations.33 Alternatively, IMS analyzers may be made mobile or portable to obtain a map or plot of concentrations in a given region or terrain.34 An extension of this concept is to transform a point detector to full mobility as an airborne analyzer. This has been done using an unmanned aerial vehicle that can sample a large volume of the atmosphere.35 In these and other applications of mobility spectrometers for measurements of chemical complex samples, samples receive some type of preprocessing through a membrane interface or a sorbent trap to extract and preconcentrate analyte for improved detection limits. 4.2.3.1 Membrane-Based Inlets The advantages of using a membrane interface between the drift tube and ambient atmosphere samples have been demonstrated, a direct benefit being improved gas purity inside the drift tube.36 The membrane excludes water and ammonia efficiently and limits the entry of polar molecules into the drift tube. Failure to control moisture will be evident in the drift time for the reactant-ion peak (RIP), which will be dependent upon ambient conditions. Cluster ions cause losses in resolution in the mobility spectrum and may

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_126... 05/05/2007

Página 1 de 1

Page 127 lead to erratic behavior of the ion signals. However, the challenges of making such membranes for handheld analyzers have also been described, and membrane manufacture is complicated by mechanical issues of thickness (~12 μm), support on a holder, and leak-free bonding between the membrane and holder.37 Finally, the yield for membranes is influenced by the molecular mass of the analyte and membrane temperature.31 Consequently, some control of membrane temperature is needed to ensure stable analytical performance, and a heating element placed against the membrane can be used to thermostat the membrane to temperatures from near-ambient to 150°C Membrane-based interfaces are especially important for IMS analyzers with the drift tube operated at or near ambient temperature. Though the membrane preferentially passes organic vapors and provides a degree of isolation from the high levels of moisture in ambient air, membranes do not exclude water with 100% efficiency, and some water vapor will enter the internal atmosphere of the analyzer under normal operating conditions. Nonethless, the partial exclusion of water will extend the life of the molecular sieves or adsorptive materials used to scrub moisture from the recycled atmosphere of the analyzer. The result is that drift tubes may be reliably operated at ambient temperature, conserving power consumption, size and weight. Otherwise, heating the drift tube to suppress the effects of moisture on analyzer response would be costly in power, size, and weight. Eventually, sieves need replacement when capacity is exceeded. The disadvantage of a membrane inlet can be diminished sensitivity because yield is governed by solubility of the analyte in the membrane; solubility is controlled by membrane temperature, vapor pressure of the substance, and polarity of the membrane and substance. Thus, large molecules and highly polar compounds will be poorly detected by unmodified, membrane-based handheld analyzers. Furthermore, such instruments may exhibit slight degradation in response times and significant hysteresis from 30 sec to several minutes for full restoration of the analyzer to clean response after an exposure to high concentrations of a chemical. The poor yield with membranes has a beneficial feature because the dynamic response range of the measurement may be extended beyond ranges considered normal for mobility spectrometers (see Subsection 3.3.1.3).37 4.2.3.2 Exponential and Dynamic Dilution Exponential dilution of a gas sample or vapor is a convenient method to provide a range of concentrations quantitatively to an analyzer or detector.38,39 Dilution methods may be used if the sample concentration saturates the reaction region of an IMS analyzer because an overloaded reaction region will result in incorrect measurements. If the drift tube is greatly overloaded with sample, the instrument may be rendered unusable for an extended period. In exponential dilution, a gaseous sample with volume Vs (ml) is introduced into a flask with volume Vf (ml), creating an initial concentration Co of sample usually in purified air or nitrogen. The flask receives a flow

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_127... 05/05/2007

Página 1 de 1

Page 128 F (ml/min) of fresh gas, and the concentration of gas flow from the apparatus Ct at a given time t (min) is described by the exponential relationship given in Equation 4–2: (4–2) The best methodology for exponential dilution includes stirred gas inside the flask, a period without gas flow for the vapor to evenly distribute throughout the flask after introduction (and often volatilization if the sample is a liquid), control of the temperature (surfaces warmed to avoid sample condensation) and mass flow to ensure constant gas flow with time. Exponential dilution may be applied to analyzers with a membrane inlet or can be used for direct introduction of sample into a drift tube. The method may also be used with liquids that can be volatilized at temperatures below that of the flask and transfer lines. The batch nature of an exponential dilution method is limiting when a high sampling frequency is needed. High-speed preparation or dilution of sample can be made with inlets that control vapor entry into the reaction region. In these designs, sampling or dilution is based on a servo-control mechanism in which peak intensities are continuously measured and feedback is given to the inlet for control of sampling.40,41 In one design, the rate of change of RIP intensity is determined continuously. When the rate of change exceeds a criterion, a flow of clean air is used to dilute the sample intake stream. This protects the analyzer from overloading of the ion source and extends the dynamic range of response.40 A variation of this servo-control approach is based on a sheath-flow inlet design in which sample is delivered to the ion source through a fused-silica tube, such as used in gas chromatography.42,43 As shown in Figure 4–4, high flows of sheath gas allow vapors eluting from the fused-silica column to be dispersed into the reaction region with high efficiency for ionization of sample to product ions because sample and reactant ions are mixed well. When the sheath flow is reduced, effluent from the fused-silica column is swept along the walls of tubing lower, and the mixing of reactant ions and sample is reduced. Consequently, response is variable, depending upon sheath flows. The absolute value for the RIP intensity is monitored, and sheath gas flow is controlled by automatic feedback to flow controllers.43 If the RIP intensity falls below threshold with increased levels of sample, the sheath flow is reduced until the threshold is restored; this can be dynamic and fast, preventing saturation of the source and providing an enhanced dynamic range. The range of linear response with this dynamic servo was extended by an order of magnitude without sacrificing the sensitivity of the instrument. The original intention was for this design to be used with GC/IMS (Chapter 5), although the approach could be useful with any direct inlet so long as fused-silica tubing was used to carry the flow of sample.42

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_128... 05/05/2007

Página 1 de 1

Page 129

FIGURE 4–4 Dynamic sheath flow—a flow of gas between the capillary column and a concentric tube carries vapors eluting from the column into the ion source. Decreasing this flow causes the effluent to flow back along the external tube without reaching the reactant ions. (From Young, D., Eiceman, G.A., Breach, J., Brittain, A.H., Thomas, CL.P., Automated control and optimization of ion mobility spectrometry responses using a sheath-flow inlet, Anal. Chim. Acta 2002, 463, 143–154.)

4.2.3.3 Preconcentration of Analytes from Ambient Air Samples Until the 1990s, mobility spectrometers had achieved the prerequisite detection limits for the accepted applications of chemical agent and explosives detection. The development of new applications in which target vapors were potentially below the amounts needed for response necessitated the adaptation of preconcentration methods pioneered earlier for GC/MS.44,45 In preconcentration methods, a large volume of sample containing the analyte is drawn through a trap where analyte molecules are retained, whereas nitrogen, oxygen, and other gases are not retained. The principle of retention can be through adsorption, partition, or condensation as described in the following text. The trapped substances are released with heat into a low flow rate of gas, so the analyte can be provided to the analyzers as a concentrated vapor stream. The degree of concentration or the preconcentration factor,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_129... 05/05/2007

Página 1 de 1

Page 130 CF, is defined according to Equation 4–3 by the volume of air sampled, Vs; the trapping efficiency, Y; and the gas volume or carrier stream, Vc, into which the analyte is released: CF=Vs(Y/Vc)

(4–3)

The principles of these methods and the materials are well developed.44 Laboratory or fixed instruments may be equipped with cryotraps, which are cooled vessels or surfaces cold enough to condense the analyte molecule from the air sample. Compounds condensed into a thin film on the surface of the trap can be released by adding heat to the trap. This method can be complicated by condensation of water that can foul subsequent measurements. A practical alternative to cold traps is a tube containing an adsorbent material that can selectively retain volatile organic compounds. Commercially available synthetic polymers are hydrophobic and thermally stable, and are attractive because of their high breakthrough volumes at ambient temperature and little or no water retention.45 Trapped substances may be recovered using solvents or through heating the adsorbent tube while a stream of purified air or nitrogen is passed through the trap. Rapid heating with a resistive element wrapped around the trap will release the substances as a concentrated effluent into a gas chromatograph or directly to the IMS analyzer. The polymers are reusable and reliable; however, partial or irreversible adsorption with incomplete recovery of trapped substances can occur with polar or reactive compounds. Sorbent-based methods of preenrichment were refined during the 1980s and 1990s and are now standard analytical methods for trace determinations of organic compounds.46 The sole application of this approach with mobility spectrometers was air monitoring on spacecraft using a gas chromatograph/mobility spectrometer, the Volatile Organic Analyzer or VOA (Subsection 5.2.5.2).47 Traps were made of two adsorbents in a single trap to collect a range of compounds. During desorption, gas flow is reversed for best possible recovery of compounds. 4.2.3.4 Reactive Gases Reactive gases such as hydrogen chloride, hydrogen fluoride, and other inorganic acids can be characterized using mobility spectrometers when vapors are introduced directly into the drift tube. When reactive gases make contact with surfaces of the inlet, unacceptable losses may occur through adsorption or chemical reactions. However, such gases will react corrosively with nearly all metal surfaces inside a drift tube including the 63Ni foil, a popular ion source for mobility spectrometers. The formation of radioactive nickel salts from reactions with acid gases can lead to an unacceptable formation and spread of salt particulate, making drift tubes and effluent from the drift tubes radioactive. Corrosive acid gases can be determined in mobility spectrometers when the 63Ni foil is bathed in nitrogen or air and pneumatically isolated from

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_130... 05/05/2007

Página 1 de 1

Page 131 reactive gases.48 This is done by isolating the ion source, in a bath of purified gas, from the reaction region where the sample is introduced. Thus, the reactive gas is restricted to a small volume in the drift tube and contact with the nickel source is prevented, ensuring no corrosion of the radioactive foil. Reactant ions are withdrawn from the ion source using an electric field and are passed into a reaction region. In a reaction region, reactive gases are introduced and mixed with the reactant ions; product ions are extracted into a drift region. The reactive gases enter and exit the drift tube (perpendicular to ion flow) from the sides of the tube, clean gas flows from the ion source, and the drift region is vented with the sample. Consequently, only a comparatively small volume of the drift tube, and none of the radioactive source, is exposed to the acid gases. This design philosophy can also be used with noncorrosive vapors that might foul a reaction region or source through overload or adsorption. This configuration has been called the side flow design and is used for detecting explosives in complex mixtures of thermally desorbed vapors.49

4.2.4 Liquid Samples 4.2.4.1 Spray and Electrospray Ionization Liquid samples may be introduced directly into a mobility spectrometer so long as the dissolved analyte can be converted to a gas-phase ion. Alternatively, the entire liquid sample must be volatilized and the analyte ionized in the solvent-rich atmosphere. This is simple in concept but unpromising in practice. At ambient pressure, liquids are more than three orders of magnitude higher in density than gases so that 1 ml of a liquid, when volatilized, will produce about 11 of gas. Thus, the reaction region can be rapidly overloaded with solvent for a sample containing low levels of an analyte. Additionally, direct injection of a liquid into a drift tube is technically dangerous through the possible accumulation of condensate and arcing or sparking between drift-tube components under high voltage. Thus, liquid samples should be introduced into a drift tube only with certain restrictions: the volume and flow rate of the liquid sample must be small compared with drift-gas flow, the liquid should produce a gas with negligible formation of aerosols or with aerosol sizes below certain values, and the drift tube should be heated. During the 1990s, a revolution occurred with mass spectrometry (MS) through the development of electrospray ionization for the determination of soluble proteins in liquid or aqueous samples.50,51 Similarly, liquid samples can be analyzed through electrospray ionization with mobility spectrometers.52,53 In electrospray ionization, a small liquid flow of ~50 μl/min is nebulized through a syringe needle or a capillary tube that is placed at high potential (as shown in Figure 4–5) and provided a stream of an inert gas or air at the needle tip. Ions formed from analyte molecules in electrospray (see Subsection 4.3.5) may be multiply charged ions, particularly when the molecule contains several protonation sites such as peptides and other bioorganic molecules. In electrospray IMS, droplets

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_131... 05/05/2007

Página 1 de 1

Page 132

FIGURE 4–5 A configuration for electrospray ionization with an IMS drift tube. The syringe needle is positioned about 1 cm from a mesh screen at the entrance to the drift tube.

formed in the electrospray source encounter hot gas when entering the drift tube. The hot gas assists in the desolvation of the droplets, forming gas-phase ions that can then be characterized for mobility in the drift region. The use of electrospray ionization as a sample introduction inlet is particularly useful for nonvolatile compounds or for compounds with low volatility that cannot be easily vaporized and introduced in the gas phase. Some biological and environmental applications of ESI/IMS are described in Chapter 7 to Chapter 9. Further details on ESI sources can be found in Subsection 4.3.5. 4.2.4.2 Solid-Phase Micro-Extraction (SPME) SPME is a technique for preconcentrating a target compound from an aqueous sample or from a gas flow through principles of solubility and enrichment with a stationary phase.54 The stationary phase is typically a polymer bonded to a silica substrate and functions similar to the bonded-phase capillary columns that are popular in GC. The retention or extraction of chemicals from a sample is governed by partition of molecules between the bonded phase and the sample matrix—liquid or gas. In this, SPME resembles the sorbent traps described earlier with an additional advantage of flexibility of design. The mechanical form of SPME may be a fiber that is retractable into a syringe-like device, a glass fiber filter, a tube containing a packed bed, or a short length of capillary column. Phases of various chemical polarities allow some selectivity toward analytes, and the effectiveness of SPME depends on the matrix and the compatibility in polarity between the analyte and bonded phase. After the sample is extracted with

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_132... 05/05/2007

Página 1 de 1

Page 133 the SPME material, compounds can be desorbed by heating the SPME device in a stream of purified gas. Desorbed vapors can be passed to a gas chromatograph or directly to the IMS analyzer. The main advantages of SPME include solvent-free extraction, elimination of glassware for extraction, and ease of use with existing instruments. Some use of SPME has been made for IMS and includes the detection of chemical warfare agents in water.55 The SPME fiber of a copolymer of DMS and divinylbenzene (DVB) was placed directly in the heated anvil of a commercial IMS analyzer, the IONSCAN 400. Also, filters based upon SPME principles were used with a chemical agent monitor (CAM) for the determination of phthalate esters in water.56 Water containing analyte was filtered, and a portion of the filter paper was thermally desorbed in a heated chamber with a small flow of nitrogen. Desorbed vapors were directed to the inlet of a CAM, providing quantitative determinations at trace levels of phthalate esters in water. 4.2.4.3 Semipermeable Membranes Membrane techniques can be seen as flexible methods for preparing or extracting the sample in preparation for measurement rather than simply a barrier to isolate flows between an analyzer and the surrounding environment. Although materials for membrane extractors may be comparable to those for membrane barriers, the dimensions and methods are distinctive and were developed earlier for MS.57 In one configuration, a section of semi-permeable tubing is immersed in water or an aqueous solution. Organic molecules that cross the membrane wall and enter the gas flow inside the tubing are swept by a stream of purified gas into the drift tube for ionization and characterization.58 Thus, the IMS instrument can be used as an on-line monitor for dissolved organic matter in solution as well as in the gas phase. Another material, Nafion, is permeable to water and can be used for dehydrating gas flows, although this method has not been adapted for IMS. In brief, membrane techniques are available for advanced designs of IMS monitors although little has been described, unlike membrane MS, which has seen considerable development.57

4.2.5 Solid Samples 4.2.5.1 Thermal Vaporization Although measurements are made in IMS instruments with ions in the gas phase, there are no intrinsic limitations with solid samples if analyte molecules can be vaporized and introduced as neutrals into the reaction region. This principle has found wide application in detecting compounds with low vapor pressures, especially illicit drugs and explosives.8 Particles of the material are trapped on a filter (made of metal, cloth, or a synthetic polymer) either by swiping the suspect object or by drawing air through the filter with suction. The filter is then placed in an assembly that is part of the IMS

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_133... 05/05/2007

Página 1 de 1

Page 134

FIGURE 4–6 The heated anvil and inlet design for solid samples (Barringer IONSCAN 4000). The filter with the sample is placed near the heated anvil and vapors are transported by the carrier gas flow into the drift tube.

inlet. In this assembly, called a heated anvil (shown in Figure 4–6), the sample is rapidly heated to 150 to 250°C, and target compounds are vaporized into a gas flow that is passed into the drift tube for analysis. Although this method is characteristic for explosives and drug determinations with the IONSCAN family of instruments, thermal desorption can be used for any sample, as demonstrated with direct heating of bacteria.59 4.2.5.2 Vaporization by Laser Heating or Ablation Solid samples may be heated and vaporized using laser energy, with the additional benefit that compounds may also be ionized with the same laser.60–65 Alternatively, vapors desorbed by exposure to a laser beam may be ionized with a conventional ion source in a familiar reaction region. In this approach, solid samples may be placed directly in the drift tube and ionized with the laser or may be placed in a separate sample chamber. Vapors formed from the sample by laser desorption are swept with a flow of carrier gas into the reaction region.66 Additional discussion on some of the applications of this approach is presented in Chapter 9. Photons from the laser may be used for both desorption and ionization of molecules, as demonstrated for polycyclic aromatic hydrocarbons, explosives, and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_134... 05/05/2007

Página 1 de 1

Page 135

FIGURE 4–7 Two designs for laser desorption and ionization of solids or adsorbed films in combination with mobility spectrometers. Drift tubes may contain an ion shutter (b)62 or can be operated without an ion shutter (a).63

others.60–64 This has been accomplished by several teams using differing drift-tube designs, although a common feature was drift-tube measurements referenced to the laser pulse. Two designs are shown in Figure 4–7. The pulsed laser provided both the reference on the time base and the source of ions; band broadening has been associated with laser desorption and ionization, presumably through prolonged reactions in the gas phase over the desorbed plume or through coulombic repulsion in the confined volume of the ionizing beam. Lasers can be useful ion sources for large, heat-sensitive biological molecules that may be converted to gas-phase product ions through a technique

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_135... 05/05/2007

Página 1 de 1

Page 136 known as MALDI (matrix-assisted laser desorption/ionization.67 A solid sample is mixed with a solvent or matrix, usually a semivolatile carboxylic aromatic compound, and placed on solid support (glass or metal). The support is inserted into the drift tube, and laser photons are used to vaporize the sample, causing the formation of a vapor plume of analyte and matrix. In this plume, proton transfers occur between the acidic matrix and the analyte, making product ions suitable for mobility analysis.68,69 A more detailed discussion of MALDI methods is presented in Subsection 4.3.1.6.

4.3 Ion Sources The formation of gas-phase ions precedes the processes of ion separation and detection of product ions by mobility measurements, although ionization can occur in the same step as sample introduction as is evident from the discussion in the previous section. Ionization in analytical IMS commonly occurs in air at ambient pressure; consequently, the methods or reactions used to produce ions must operate with the levels of moisture and oxygen found in ambient air. Ion sources have included radioactive sources, photo-discharge lamps, lasers, electrospray ion sources, flames, corona discharges, and surface ionization sources. During the past decade, a few trends were evident with increased interest in nonradioactive sources and ion sources suited for liquid and solid samples. This latter trend is especially evident with interest in semivolatile and nonvolatile molecules, particularly for biological and environmental applications. Some of the features of ionization sources that are in use in commercial and research instruments are summarized in Table 4–1. TABLE 4–1 Summary of Ionization Techniques Used in Ion Mobility Spectrometry

Source

Type of Chemicals

Maintenance

Cost

Comments

Radioactive

Universal

Low

Medium/low

Licensing required

Corona discharge

Universal

High

Medium

Maintenance required

Photoionization

Selective (IP)

Medium

Medium

Low efficiency

Surface ionization

Nitrogen bases

High

Medium

Complex

Electrospray

Liquids

Medium

Medium

Long clearance

MALDI

Solids

High

High

Laboratory use

Flame

Selective

Medium

Low

Molecular structure lost

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_136... 05/05/2007

Página 1 de 1

Page 137

4.3.1 Radioactivity: Nickel, Americium, and Tritium Radioactive sources are favored for use in IMS analyzers because these provide stable and reliable operation with ionization chemistry that is well suited for current applications of IMS. Furthermore, radioactive foils do not require an external power supply and have no moving parts or maintenance requirements. In 2005, the most widely used and best understood of all ion sources for IMS is the long-favored radioactive 63Ni source that is also widely used in electron capture detectors for GC. The radioactive source most favored is 10 mCi of 63Ni coated as a thin layer on a metal strip, generally nickel or gold.70 The maximum energy of the electrons emitted from the 63Ni source is 67 keV, with an average energy near 17 keV. Almost all the energy of this source is dissipated in air at an ambient pressure within 10 to 15 mm, from the surface of the metal, establishing guidelines for the optimum diameter when the source is used as a cylinder, a common geometry in IMS analyzers.71 The electrons emitted from the 63Ni produce ions and secondary electrons (see Chapter 3). This process is repeated until the secondary electrons are no longer energetic enough to ionize the gas molecules of the supporting atmosphere. The formation of an ion pair requires ~35 eV, so each beta particle emitted from the source could ideally produce about 500 ion pairs on average.72,73 Negative ions may be produced also through electron-attachment processes that, in most cases, proceed efficiently when the electrons are at thermal energy, thus further increasing the ion yield. Radioactive isotopes other than 63Ni have been used with IMS drift tubes, including a betaemitting tritium source and an alpha-emitting isotope, 241Am, which is similar to the source used in household smoke detectors.74 Alpha particles emitted from 241Am are highly energetic, with energies above 5.4 MeV, and have a short effective range in air so that ionization is efficient in small-volume sources.75 Tritium poses less radiation hazards than 63Ni sources and has been used as an ionization source in a study of environmental monitoring of ppb levels of toxic compounds in ambient air.76 Despite the attractions described in the preceding text of radioactive ion sources in IMS, use of Ni, Am, or T sources are discouraged today due to several developments that are financial, organizational, and technical. Radioactive sources require special permits and licensing procedures, and general licenses for alpha-emitting sources with appropriate activity are particularly difficult to obtain. Once in service, radioactive-based drift tubes require semi-annual checks for leakage of radioactivity and general laboratory hygiene. The burdens of cost of assays and keeping records, and the legal implication of workplace hygiene may discourage prospective and current users. Another concern with radioactive sources is the care needed to prevent chemically reactive conditions in the IMS drift tube; for example, elevated drift-tube temperatures in the air atmosphere may eventually cause oxidation of the metal foil and the formation of nickel oxides or salts. These salts or oxides are mechanically unstable and may be released into the ambient environment if the drift tube is vented without a particulate filter. A paper

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_137... 05/05/2007

Página 1 de 1

Page 138 filter fitted on the drift-tube effluent will trap radioactive particulate. Finally proper disposal of drift tubes equipped with radioactive sources requires permits and can be expensive.

4.3.2 Corona Discharges The ion chemistry and electrical stability of a point-to-plane corona discharge (CD) ion source was described using a tandem mass spectrometer equipped with an atmospheric pressure source.77 This source was a continuous current-regulated discharge with a DC power supply and demonstrated the regions of stability with potential and distance between the needle and plane. CDs with various designs and control of electronics have also been developed or explored with IMS,78–81 particularly during the 1990s, and CDs are available in some production instruments.82 In order to form a CD, a sharp needle or thin wire is placed 2 to 8 mm from a metal plate or discharge electrode with a voltage difference of 1 to 3 kV between the needle and plate. An electric discharge develops in the gap between the needle or wire and the opposing conductor, and the ions formed in the gap can resemble closely those found in the widely used 63Ni ion source. These ions are then available for subsequent ion-molecule reactions with the sample. A CD has been used in pure nitrogen with negative polarity79–80 in which a large number of electrons were produced and used for ionization. However, electronegative substances can quench the discharge, and a drift tube was designed and evaluated for detection of halogenated methane and some nitrocompounds. Positive- and negative-ion formation was shown to be consistent with the Townsend formula (I/V is a linear function of V), and the total ion current obtained from the corona ionization source was ten times greater than that of the 63Ni source. Although DC corona discharges are possible with IMS,79,80 a pulsed corona may be attractive due to the reduced power needed to operate the source. However, a pulsed source introduces an element of time or time-resolved chemistry that begins with the start of the discharge. Changes in ion composition occur as the ions move away from the center of the discharge.81 A complication of this chemistry can be found in the negative polarity where nitrogen oxides and ozone formed in the corona may interfere with gasphase ion-molecule chemistry and degrade the response of an IMS analyzer to certain chemicals. Two solutions to this problem have been proposed: increasing the distance between the electrodes77 and reversing the air flow past the corona needle.78 CDs do require maintenance because electrodes undergo erosion and require replacement. Nonetheless, several applications of CD-based drift tubes have been described, including the determination of chlorobenzene in water samples83 and the monitoring of acetone in air.84 A pulsed radio-frequency (RF) discharge was proposed as a means of extending the life of the needle, improving the reliability of the source, and reducing the interference from unwanted oxides as shown in Figure 4–8.85

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_138... 05/05/2007

Página 1 de 1

Page 139

FIGURE 4–8 Pulsed corona discharge ionization source—(a) high voltage is applied between two gold wires to form the discharge; (b) the voltage pulse sequence for positive-and negative-ion formation.

In summary, a CD has been deemed valuable because there is no radioactivity, the ion currents are relatively high, design and assembly are simple, and certain applications such as the direct analysis of liquid samples may be best made using a CD source. The disadvantages include the need for an external power supply, corrosion of components, maintenance of the discharge, and the formation of corrosive chemical vapors such as NOx and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_139... 05/05/2007

Página 1 de 1

Page 140 ozone. Stability may be degraded by corrosion of the needle; this is cumulative and is governed by use.

4.3.3 Photoionization: Discharge Lamps and Lasers Photo-discharge Iamps86–93 and Iasers60–69 may be used as a means of ionizing neutral molecules in air at ambient pressure. The lasers may also serve as a means of vaporizing solid samples as adsorbed films and solids before ionization.62,66 Photo-discharge lamps emit photons from the electrical excitation of gases filled in the lamp,94 and lamps that are commercially available can provide energies of 9.5, 10.2, 10.6, and 11.7 eV.95 The formation of positive ions with photons has been described as direct ionization through the reaction given in Equation 4–4: (4–4) where hυ is the photon energy and M the neutral molecule. The energy needed for this, the ionization potential, with organic compounds is generally between 7 to 10 eV, and M+ ions are routinely observed with aromatic hydrocarbons.62 The exact mechanism of ion formation in air at ambient pressure should be considered as incompletely understood for compounds of high proton affinity or compounds for which protons can be released. For example, product ions for ketones with a discharge lamp include MH+ species rather than M+. This may suggest that intermediate reactive species are active in ambient pressure ionization, although the reactions are not fully described. Negative ions are not formed directly through photoionization processes but arise through chemical reactions with the electron shown in Equation 4–4. The electron may attach directly to a molecule or may undergo dissociative ionization; also, the electron may react with oxygen and proceed through association reactions (see Chapter 3). A 10.6-eV low-pressure gas discharge lamp has been used for the continuous detection of alcohols in the concentration range between 1 and 100 ppmv,92,93 and in other studies aliphatic and aromatic hydrocarbons were measured by UV-IMS. The main advantage of photo-discharge sources is that some selectivity in response may be ensured by the choice of an appropriate ionization energy or wavelength. The disadvantages of photo-discharge lamps are the requirement for an external power supply, the cost of the lamps, and the need to periodically replace them due to the finite lifetimes of such lamps. A drift tube without an ion shutter was equipped with a pulsed xenon lamp photoionization light source and characterized for parameters and performance with negative-ion detection.97 The performance of this design was found to be comparable to that of conventional IMS drift tubes. Lasers could be used to provide, in principle, any photon energy in the ultraviolet-to-infrared range of wavelengths with dye lasers. However, the common Nd-YAG laser offers conveniently four wavelengths: a fundamental

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_140... 05/05/2007

Página 1 de 1

Page 141 at 1064 nm and harmonics at 532, 355, and 266 nm. Laser-based, single-photon ionization is possible only for photons in the UV part of the spectrum, and lasers that operate in this energy range are mainly excimer lasers. Little may be said about gas-phase ionization by lasers with IMS because only a handful of studies have been described.60,61,96

4.3.4 Surface Ionization Source An ion source based on surface ionization principles was developed primarily by the group of Rasulev in Uzbekistan98,99 and was demonstrated on a mobility spectrometer.100–101 As shown in Figure 4–9, the source consists of an emitter made from a single crystal of molybdenum doped with iridium (or another platinum group metal) and heated to 300 to 500°C Certain types of molecules, mainly nitrogen bases, will undergo electron transfer upon collision or contact with the heated reactive surface, resulting in the formation of a positive ion. Ionization of tertiary amines was shown to be favored over that of secondary amines that were more efficiently ionized than primary amines. Compared with a conventional 63Ni source, the surface ionization source was reported to have a comparatively large dynamic range, a selective response to certain types of compounds, and none of the regulatory problems associated with radioactive ionization sources. Amines, tobacco alkaloids, and triazine herbicides were all shown to exhibit picogram limits of detection with a dynamic range of five orders of magnitude.98 The success of a surface ionization source is dependent upon the preparation of stable emitters with a simple design.99 One of the problems with this source is the poisoning of the surface after exposure to certain compounds. The surface can be regenerated but only under special conditions in a controlled atmosphere. Another complication in the use of this source for several substances is that the response is highly dependent upon the structure of the analyte.

4.3.5 Electrospray Ionization (ESI) ESI may occur when an aerosol is formed between a needle tip under a potential of several thousand volts relative to a grid or plate. The mechanism of ion formation was described in detail by Kebarle and Tang.51 ESI has become attractive as a method for IMS determination of bio-molecules52,53,102– 105 and for the analysis of samples of environmental interest106–108 An example of the value of an ESI source can be seen in the determination of chemical warfare agents (CWA) and their degradation products in water through analysis by ESI/IMS/MS.106 Water samples were injected through an ESI system into a high-resolution IMS drift tube, in which separation according to mobility occurred. A mass spectrum of each peak in the mobility spectrum was obtained using a time-of-flight mass spectrometer (TOF-MS). Traces of CWAs and their degradation products

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_141... 05/05/2007

Página 1 de 1

Page 142

FIGURE 4–9 Surface ionization: A schematic of surface ionization shows that a molecule strikes the surface and loses an electron. The photograph shows the crystal in an SI assembly for IMS.

were rapidly detected and identified using this approach. ESI may be used also for detection of inorganic ions in water, such as uranyl acetate.108 The advantages of ESI are that liquid samples may be introduced directly into the IMS, molecular information is retained due to the soft ionization processes, and in some cases multiply charged ions may be formed. The main limitation with ESI for practical analytical IMS is that relatively long rinsing times are needed between samples because memory effects in the fluid delivery system can be quite significant.

4.3.6 MALDI A variation of laser-based ionization that merits separate discussion is MALDI. In this, a solid sample may be directly desorbed, vaporized, and ionized inside an IMS drift tube when an intense laser pulse is directed at the sample. This method has gained popularity with MS for the direct ionization

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_142... 05/05/2007

Página 1 de 1

Page 143 of solid samples, particularly for large, nonvolatile biochemicals and also as a way of introducing metal ions into the gas phase. The rapid vaporization of biological samples prevents their dissociation and fragmentation and thus provides a way to obtain molecular and structural information. Also, MALDI provides a way of producing different types of cationized molecular ions with Li+, Na+, Cu+, and Ag+, in addition to the protonated species usually formed in atmospheric pressure ionization processes.109 The most serious drawbacks of using MALDI are the price of a suitable laser and the complexity and limited applicability to certain types of solid samples. Although early studies were made with MALDI MS with the ion source region under vacuum, the demonstration of atmospheric pressure MALDI with MS encouraged MALDI IMS. Some examples of studies with MALDI and IMS include the use of MALDI to generate sodiated parent ions of a number of oligosaccharides without fragmentation and cationized forms of bradykinin. Cross sections of the ions were obtained and compared with predictions made by molecular mechanics or molecular dynamics calculations.110 The gas-phase conformations of poly (styrene) oligomers cationized by Li+, Na+, Cu+, and Ag+ were examined using MALDI, and ion collision cross sections in He were measured by ion mobility methods.109 Chemically modified DNA oligonucleotides up to eight bases in length and noncovalent complex formation between peptides were characterized by MALDI followed by IMS separation and time-of-flight mass analysis (MALDI/IM/TOFMS).111 Comparison of MALDI/IM/TOFMS with MALDI/IM/MS analysis yielded results similar to that by nanoelectrospray mass spectrometry.112

4.3.7 Flames Flames were among the earliest sources of ions at ambient pressure, and the mobility of ions in flames was measured as recently as 1978 in order to explore the properties of a flame ionization detector widely used in GC. Studies with IMS demonstrated that ions in the flame as residual current were hydrated protons, and ions such as protonated monomers of a compound were not observed when a chemical was introduced into the flame.113 The possibility of a flame source, however, was suggested in the work of Atar et al., who described the use of ions and electrons produced in a hydrogen or hydrocarbon flame to ionize sample molecules, in a fashion similar to a flame ionization detector.114 Several possibilities for utilizing this idea were proposed. Among these was a pulsed flame with considerably reduced fuel consumption. This ionization method may increase sensitivity to certain types of compounds and extend the range of IMS applications to include chemicals used in the microelectronics industry such as hydrides and fluorides. The hydrides, which have relatively low proton affinities,115 do not readily form stable positive ions in the IMS drift tube but may be converted in the flame into the corresponding oxides, readily forming negative

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_143... 05/05/2007

Página 1 de 1

Page 144 ions. The advantages of flame ionization sources are the possibility of producing high ion currents, capability to detect additional types of compounds, and elimination of a radioactive or photoionization source. The main disadvantages are adding considerable complexity to the drift tube and losing some of the specificity of the detector because molecular identity is lost when oxides are formed. Although promising, this source has not yet been incorporated into an IMS drift tube.

4.4 Drift Tubes The drift tube is the most crucial component of an IMS analyzer, and design flaws or choice of wrong materials may result in low sensitivity, poor resolution, and severe memory effects. In this section, different approaches to the design and construction of drift tubes are described. There are two basic designs that are well recognized: the linear electric field (DC) drift tube and the field asymmetric design (AC or RF) drift tube. A number of other designs have been demonstrated, and some have been commercialized, such as the aspirator design. The features that are common to all drift tubes are that they contain an ion source and reaction region, where ions are formed; a separation or drift region, where ions are differentiated according to mobility; and a detector, where the ion signals are registered. In well-designed drift tubes, all the ion chemistry should be completed in the reaction region so that once the ions are formed, they retain their identity in the drift region until the detector is reached. In commercial instruments, the drift tube is usually integrated into the inlet, controllers, and data system as shown in Figure 4–1 and Figure 4–2. In contrast, research-grade instruments are usually modular and can be adapted to suit the application.

4.4.1 Traditional Drift Tubes with Linear Electric Field Drift tubes with linear electric fields are the most basic and best-understood drift tubes for IMS, and the principles of this design can be described in detail. The tubular reaction and drift regions are separated by an ion shutter, which is used to inject packets of ions produced in the reaction region into the drift region. The ion packet or swarm traverses the distance between the ion shutter and the detector, thus establishing the drift time or the time of flight for ions of a given mobility coefficient in a constant electric field. These basic features are common to all linear-field drift tubes; however, designs of drift tubes differ in the method of ionization (see the preceding text), flow regime, technique used to create the electric field gradient, position, type, and number of ion shutters, operating temperature, and detection method.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_144... 05/05/2007

Página 1 de 1

Page 145 4.4.1.1 Design and Structure Descriptions of a drift tube for analytical use first appeared in refereed journals in 1970 with articles by Karasek,116 and Cohen and Karasek.117 The metal-conducting rings were held together under compression with a spring and were insulated by sapphire balls of about 1 mm diameter, allowing a gap between rings. The entire drift tube including the supporting structure was placed in a sealed vacuum-tight metal chamber. A photograph of this drift tube is shown in Figure 1–10 of Chapter 1. The drift tube was made of cylindrical elements in a stack that were biased at different voltages to maintain an electric field of 200 to 300 V/cm along the central axis of the drift tube. Neighboring conducting rings were insulated from each other using sapphire balls, and the gap between conducting rings (up to a few millimeters) was greater than the width of the ring by a factor of ten. Ionization was by a polished beta-emitting radioactive 63Ni source inserted into the first drift ring. The drift tube contained two ion shutters of the Bradbury-Nielson type,118 where a potential difference was applied between two coplanar sets of thin wires positioned in parallel and at close distances. Neighboring wires were electrically isolated, and a strong electric field (~600 V/cm) was generated between neighboring wires. This field was perpendicular to the axis of the field gradient across the drift tube, blocking ion passage to the drift region as ions were annihilated on the wires. The gate was opened by bringing the two sets of wires to common potential, typically for ~0.2 msec every 20 msec, so that only about 1% of the ions formed in the source were injected into the drift region. A flat, circular collector plate or Faraday plate was used to measure the ion current. A biased aperture grid was placed close to the detector in order to minimize current flow induced by an approaching ion swarm. Thus, the spectrum was freed of artifacts and distortion of the peak shape. In the early drift-tube designs, a second ion shutter was located near the detector end of the drift tube. This second ion shutter was open at a time that was delayed relative to the closing of the first shutter. Only ions with drift times equal to this delay time would pass the second shutter and generate a signal at the detector. A mobility spectrum was obtained by sweeping the delay time, creating a plot of signal intensity vs. delay (or drift) time. In one of the early designs, the lengths of the reaction and drift regions were 6 and 8 cm, respectively. A flow of carrier gas, typically set to 100 to 200 ml/min, entered the reaction region at the front end of the drift tube, and the drift gas, set at 400 to 700 ml/min, entered the drift region from the detector end of the drift tube. These opposing flows were exhausted, in principle, through an exit vent, just in front of the shutter grid. The carrier and drift gases in these early studies were nitrogen or air, purified or scrubbed using molecule sieves. The carrier gas served to transport the sample into the ion source or reaction region, where ion-molecule reactions occurred. However, if ions continued to react in the drift region, drift times would be a composite between that of the original ion and that of the final product

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_145... 05/05/2007

Página 1 de 1

Page 146 ion, making the mobility spectrum difficult to interpret. In order to inhibit these reactions, the drift gas should not contain any sample neutrals with which the ions could cluster or react when in the drift region; hence the emphasis rightly on purified gases for carrier and drift flows. The metal chamber containing the drift tube could be heated to temperatures above 200°C and could be pumped to moderate vacuum with a rotary pump for a bake out to enhance cleaning and reduction of memory effects. This early drift tube suffered from a few essential design limitations. The space between the drift rings permitted diffusion of molecules of sample into the gas volume of the chamber and throughout the chamber. These molecules were not easily removed from the chamber and reentered, in a somewhat uncontrolled manner, the drift tube in the reaction and drift regions. Consequently, ionmolecule clustering reactions could occur when sample amounts or vapors from previous samples were persistent and at elevated levels. This led to large memory effects, long clearance times for samples, and drift times and mobility spectra that were sometimes dependent upon sample concentration. Furthermore, the boxcar integrator was slow, and several minutes were needed for acquisition of each mobility spectrum. Consequently, use of this spectrometer for analytical measurements was cumbersome, particularly if quantitative results were needed. This basic design was considerably improved within a decade with digital signal processing and by using a closed ring structure in which the sapphire balls were replaced by Macor® insulating rings as shown in Figure 4–10.

FIGURE 4–10 A schematic view of the improved drift-tube design from PCP, Inc. The conducting guard rings were separated by Macor rings that, when compressed by a set of springs, provided a closed drift-tube design.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Page_14...

06/05/2007

Página 1 de 1

Page 147 The second ion shutter was removed, and the time to acquire a mobility spectrum was reduced from several minutes to a few seconds with computer-based signal processing and digital signal averaging. 4.4.1.2 Gas Flow The carrier gas in the drift tube (described in the preceding subsection) was used to transport sample molecules from the inlet into the reaction region and provide the supporting atmosphere in which ionization occurred. The drift gas was intended to maintain a clean gas environment in the drift region, free of unwanted impurities. The counterflow design and open ring structure made these goals difficult to attain. In 1982, Baim and Hill simplified this design with a single gas flow that was passed from the detector end, through the drift and reaction regions, and vented as shown in Figure 4–11.119 This was called a unidirectional drift tube and was used with a closed drift-tube structure of alternating conducting and insulating rings. The sample was introduced through a GC capillary column near the ion source and was swept by the drift gas (~300 ml/min) through the source and rapidly out of the drift tube. The ions derived from the sample vapors were drawn through the length of

FIGURE 4–11 A segmented drift tube with unidirectional flow. The conducting guard rings were separated from each other by glass or ceramic insulating rings, providing an airtight drift tube.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_147... 06/05/2007

Página 1 de 1

Page 148 the tube by the electric field, whereas the neutral sample molecules that were not ionized spent only a short time in the reaction region before being swept out. This enabled the mobility spectrometer to serve as a detector for a GC capillary column in which elution profiles could be as narrow as 15 sec. The performance of this design was validated by Eiceman et al., who further simplified the drift-ring design.120 In laboratory instruments, or portable instruments intended for brief intervals of use, carrier and drift gases can be supplied from compressed gas cylinders. However, portable—particularly handheld—instruments cannot be supplied with gas cylinders and instead rely on onboard purification of ambient air through molecular sieves or adsorbent traps. Such traps are part of a closed-loop or recycled flow system with an internal pump that supplies scrubbed carrier and drift flows to the drift tube and also draws ambient air into the inlet of an analyzer. The flow system is referenced to ambient pressure, so changes in barometric pressure cannot cause the air pressure inside the drift tube to exceed that of the surrounding atmosphere. In some laboratory studies, the standard carrier and drift gases of nitrogen or air were replaced by inert gases (mainly helium or argon) or heavier compounds, such as CO2 or SF6. Helium has several advantages as a drift gas with high ionization potential, low polarizability, and no dipole moment. Thus, helium is good for cooling ions and consequently is the most compatible with theoretical predictions. Helium has been the drift gas of choice for studies of large biological molecules. When helium or other noble gases have been used as drift gases, the tendency of arcing between components is increased, so that voltages must be decreased in comparison with analyzers operated with air. Changing the drift gas may be advantageous in cases where ions with similar drift times in one gas may be resolved in a different drift gas.121 In laboratory studies, regardless of the source of gases, knowledge of gas moisture and temperature are essential. Moisture levels can be low when nitrogen is drawn from a liquid-nitrogen reservoir. Otherwise, scrubbing towers containing molecular sieves are helpful in reducing moisture levels below 1 ppm or even up to 50 ppb. Moisture can be monitored using one of several technologies. A convenient commercial method is based on an aluminum oxide sensor and provides response from 10 ppb to 10,000 ppm. The temperature of an IMS experiment should be determined by using gas temperature measured with a low-mass thermocouple placed in the gas flow as exhausted from the drift tube. The community of IMS investigators is becoming aware of these topics of moisture and temperature, and instrument drift tubes with affordable sensors that provide meaningful or accurate results are under development. A variation of recycled internal gas flows can be found in a miniaturized palm-sized mobility spectrometer in which the drift tube has an open structure and is surrounded by a molecular sieve.122 The internal atmosphere of the drift tube is in direct contact with the molecular sieve, which provides passive or diffusion pumping of impurities, sample, and moisture in the internal atmosphere. Although this design is capable of receiving only

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_148... 06/05/2007

Página 1 de 1

Page 149 microliter amounts of sample through a pinhole orifice, the simplicity of this design is attractive for a field analyzer. This drift tube is now available as the lightweight chemical detector (Smiths Detection). 4.4.1.3 Electric Field Gradients Electric fields between conducting components within a drift tube have been formed using a highvoltage supply and a voltage divider with individual resistors. So that power demand on the voltage supply can be kept low and current flow maintained throughout components, resistors are typically ~1 MΩ. Resistors with good precision are recommended and are often mounted externally to the drift tube with heated drift tubes. However, high-temperature resistors are commercially available and can be attached directly to the tube components. Similarly, unheated drift tubes can be equipped with resistors internal to the drift tube. Discrete resistors can be replaced with resistors printed on ceramic, although each conducting element must be connected by wire to nodes in the resistor series. The electric field applied to the drift region is generally linear and uniform between 200 to 300 V/cm in IMS instruments; however, the designs and fields used in the reaction region and particularly near the ionization source, the shutter grid, and the detector plate can be specific to individual drift tubes. Calculations based on ion paths, using the finite elements method, showed that ion transmission times could be better controlled by using a slightly focusing electric field gradient.123 This would lead to a narrower spread of the arrival times at the collector of ions of the same type leaving from different spots in the shutter grid, thus improving resolution of the drift tube. This better control was achieved by using a ratio of about 1.5:1 between the widths of the insulating and conducting rings, as employed in the Rotem Prototype IMS.123 A study of the density profile of the ions arriving at the detector after being produced by a cylindrical 63Ni source showed that this profile reflects the electron density in the ionization source.124 The electric fields inside drift tubes have been explored in recent years by Soppart and Baumbach,125 who modeled drift-tube designs and their effects on electric fields. They emphasized the importance of including the housing or chamber in modeling field lines. This was also considered by Eiceman et al.,126 who described quantitatively the effects of other components on overall instrument performance. The apparent charging of surfaces and the difficulties with thermal expansion were identified as features of a drift tube containing Teflon insulating rings (see Section 4.6). In all the designs referenced in the previous discussion, discrete conducting rings, insulating rings, and resistors were used to create the electric field for ion characterization. This approach is demanding, requiring numerous components and experience in the assembly of these components. An alternative approach was developed by Spangler and coworkers at Bendix (later, Environmental Technologies Group, ETG).127,128 Here, the inside surface of a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_149... 06/05/2007

Página 1 de 1

Page 150 ceramic tube is coated with a thick-film resistor ink layer, which when fired provided a resistive surface. The ink, when applied with even thickness throughout the tube, could provide a total resistivity comparable to a drift tube with external resistors of 10 to 20 MΩ. When a potential drop is applied across the tube, a field gradient is produced and ions pass through this tube with speed and band broadening indistinguishable from conventional discrete component designs. In principle, this design could provide a simplified, economic, and compact drift tube as shown in Figure 4–12. However, the application of the film evenly throughout the length of the cylinder is nontrivial except for small tubes. This design was commercialized in the 1990s and is today the design of one manufacturer. In most drift-tube designs, the high-voltage supply is applied to the ion source or reaction region, and the voltage gradient extends throughout the drift tube with ground potential at the detector end. The detector (see the following text) is placed at virtual ground on the input of an operational amplifier. However, in an alternative design, the source is grounded and the detector is placed at high potential. A battery-operated preamplifier can float with the detector, and a signal can be drawn from the preamplifier for further amplification through optoisolators. This strategy was employed with a high-resolution drift tube129 and a twin drift tube that was designed and

FIGURE 4–12 Coated ceramic tube that was inside the miniature IMS. The electric field was produced by a voltage gradient over the resistive coating along the entire length of the drift tube.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_150... 06/05/2007

Página 1 de 1

Page 151 built by Graseby Dynamics but has not been described in the open literature. The twin drift-tube design had some distinct advantages not found in other IMS drift tubes, namely, the simultaneous detection of ions of each polarity When the ion source is at ground potential, positive and negative ions are extracted into each drift tube with corresponding electric field gradients. No further development of this concept has occurred. The high-resolution drift tube described by Brokenshire129 with values up to 200 for ratios of drift time to peak width (compared with 25 to 50 for most drift tubes) was based on a large-bore glass tube, (50 cm long×10 cm diameter) with discrete copper bands as drift rings applied on the outside of the glass tube. The ion-shutter grid and detector assemblies were adapted from the CAM analyzer, providing sampling of ions only in the center of the drift tube. The voltage applied to this drift tube was from 15 to 20 kV, making the instrument a laboratory-based research-grade tool. Two final aspects of drift tubes should be noted here. Electric fields in drift tubes need not be limited to linear or static gradients and can be non-linear, or even reversible or dynamic. Blanchard proposed an idea for improving performance of the IMS that relied on a reversible field gradient to selectively collect ions of a limited range of reduced mobilities.130 Finally, two drift tubes may be operated in parallel with opposite polarity in order to obtain mobility spectra in positive and negative polarities for a sample. Although this is similar to the twin drift-tube design described above, it features a conventional electrical arrangement. A photograph and schematic of a miniature IMS with two parallel drift tubes are shown in Figure 4–13. 4.4.1.4 Ion Shutters The commonly employed method to inject ions into the drift region from the reaction region is the ion shutter, often termed the Bradbury and Nielson design.118 However, the invention should be credited to Cravath131 and van de Graaff,132 who independently described this style of ion shutter in 1929. In this ion shutter, thin wires are placed parallel and coplanar at close separations and under tension. Alternate wires are mechanically and electrically isolated and are usually held on a nonconducting support. For example, wires may be placed on a ceramic ring with a pattern of holes to allow proper alignment of wires; however, initial fitting of wires with proper tension is tedious because wires are prone to breaking during assembly. Also, wires may sag or break from thermal expansion or contraction during heating and cooling cycles. This can be remedied with two ceramic halves to hold the wires and a thermally stable metal spring placed between the halves in order to maintain tension even with thermal cycles. In any design, performance will be degraded when wires are not parallel or contain bends or other defects. A functionally comparable ion shutter was described by Tyndall and consists of parallel wires as found in the Bradbury-Nielson shutter except that the wires are parallel in planes offset by a small distance.133 The advantage

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_151... 06/05/2007

Página 1 de 1

Page 152

FIGURE 4–13 Miniature ion mobility spectrometer called the lightweight chemical detector (LCD) made by Smiths Detection: (a) photograph of the device; (b) schematic showing the two parallel drift tubes (positive and negative polarity) and the loudspeaker membrane that serves to introduce the ambient air sample through a pinhole into each miniature drift tube.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_152... 06/05/2007

Página 1 de 1

Page 153 of the Tyndall design is that shutters may be crafted in three parts: two geometrically correct wire sets and an insulator. These can be pressed together to make a convenient, economic, and functional ion shutter. A description of an ion shutter of this design is excerpted here,126 in which the term grid refers to one assembly of wires. The grids were the same as those found in the CAM and were pressed together with a mica insulator between the grids. The excerpt follows: “The grids have parallel wires with diameters of 0.05 mm and wire-to-wire distances of 0.5 mm center-to-center. The grid wires span a hole of 13 mm diameter on a disk of 30 mm diameter. The two grids were separated by 0.3 mm with a thin mica sheet and the juxtaposition of the wires made an interdigitated arrangement with line-of-sight wire-towire distances of 0.25 mm, center-to-center. The grid nearest the source, grid 1, was always at a potential appropriately suited to the electric field gradient as established by the voltage divider. The second grid, grid 2, was placed at 90–100 V with respect to grid 1 when the ion shutter was closed, preventing the passage of ions into the drift region, i.e., 960 V for grid 1 and 1060 V for grid 2. Ion injection into the drift region occurred when the potential for grid 2 was made common with that for grid 1. In almost all experiments, the ion shutter was open for 208 msec and each scan was initiated on the rising edge of the rectangular shutter pulse. The timescale was set at the center of the rectangular wave form.” Although the Tyndall and Bradbury-Nielson ion shutters differ slightly in structure, electrical control as described in the preceding text can be the same. A slight variation in electrical control is when wire pairs are both changed in potential equidistance from a common potential used for ion injection. Regardless of electrical or mechanical arrangements, these grids both constitute limitations of the analyzer seen directly in the duty cycle. Typically, ions are sampled from the reaction region for only 300 μsec every 20 to 30 msec; thus only 1% of all ions are utilized for a measurement. Additionally, solutions to the delicate assembly of the Tyndall or Bradbury-Nielson shutters have been sought with two metal mesh grids that are separated by an insulating gasket and operated electrically as in the traditional ion shutters. Descriptions of this shutter are not available, and the design presumably is not used because ion losses from collisions on the grids, which would fill much of the cross section of the drift tube, would be large. The low duty cycle of the ion shutter has motivated the development of Fourier transform ion mobility spectrometry (FT-IMS). In FT-IMS, the boxcar waveform on the ion shutter can be replaced with a sawtooth or sinusoidal waveform of variable frequency. Thus, ions penetrate the shutter constantly in amounts governed by the waveform amplitude, and so the duty cycle for ion sampling is increased in comparison with the ordinary boxcar waveform. Alternatively, a rectangular waveform with a range of frequencies may be employed. In either instance, the ion signal depends on the frequency of the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_153... 06/05/2007

Página 1 de 1

Page 154 waveform. An inferogram profile is transformed to the time domain to extract the mobility spectrum.134,135 The sinusoidal waveforms failed to demonstrate anticipated improvements in detection limits, presumably because ion-ion repulsions occurred in the drift region when the volume was comparatively rich in ion density.136 Tarver has employed a rectangular waveform with a range of frequencies and has reported a sevenfold increase in sensitivity with a 50% increase in duty cycle.135 Although the concept of boxcar integration using two ion shutters in a drift tube was once a common feature of early drift tubes,116,117 there may seem to be little value for this design with computer acquisition of the signal (see the following text). However, a second ion shutter can be helpful with interfaces of IMS drift tubes with mass spectrometers. In such combinations, a second ion shutter is useful in isolating a region of the mobility spectrum to pass into a mass spectrometer so that a mass spectrum of only the selected peak is obtained. This has not been widely practiced owing to poor signalto-noise ratios in such experiments but should not be discounted in future designs. The ion shutter, or rather the shutter pulse width, is an important variable in establishing peak shape in a mobility spectrum.137 Because the minimum peak width will be established by the time needed for ions to move through the wire structure from the reaction region to the drift region, the minimum peak width is practically between 10 to 100 μsec. Moreover, this figure is dynamic and will depend upon the fields around the shutter wires, established by potential differences on the wires, relative to the field provided by the drift-tube rings.138 Ion motion near the ion shutters appears to be governed by a balance between the electric fields on the wires and the drift-tube field. When ion-shutter fields are low, ion penetration through the wire occurs continuously, adding a DC component to the mobility spectrum. As fields in the ion shutter become too large, ion passage does not occur, presumably because ions after entering the drift tube during a shutter pulse are captured by the ion shutters when fields are restored on the trailing edge of the shutter pulse. An optimum field exists and is seen as an optimum voltage for an ion shutter.138 This appears to be referenced to the drift-tube fields. When drift-tube fields in the vicinity of the ion shutter are increased, with constant fields in the shutter, ion penetration of the shutter occurs even when it is closed.62 In summary, the traditional Bradbury-Nielson ion shutter can provide a convenient means of introducing ions into a drift tube to obtain a mobility spectrum and has been effective in both laboratory and portable IMS analyzers. However, this ion-shutter design suffers from intrinsic limitations pertaining to duty cycle, minimum pulse width, and complexity or cost for assembly with a drift tube. There are no alternatives to the ion shutters described in the preceding text and unfortunately no methods apparently exist to circumvent practical and theoretical limitations of these ion shutters for injecting ions in a drift region. Such limitations are associated with the requirements to move ions using mobilities of ion swarms at ambient pressure.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_154... 06/05/2007

Página 1 de 1

Page 155

4.4.2 High-Field Asymmetric or Differential Mobility Spectrometers In traditional IMS, the mobility coefficient is considered constant (Ko) for an ion at ambient pressure in a comparatively low electric field (E) of 100 to 300 V/cm. Changes in E will alter drift velocities for ions, so the ratio of vd to E is a constant, i.e., Ko. However, when E is increased beyond that commonly used in linear-field drift tubes, the mobility coefficient becomes dependent on the strength of the electric field (at constant pressure or gas neutral density, N) as described in Chapter 2. Although this has been known for over a century, the concept was exploited only a decade ago to provide ion characterization through what has become known as differential mobility spectrometry (DMS). Different names have been given to methods and devices based on this principle, among them field ion spectrometry (FIS), field asymmetric ion mobility spectrometry (FAIMS), ion drift spectrometer, high-field asymmetric waveform ion mobility spectrometry, and radio-frequency ion mobility spectrometry (RFIMS); choices of investigators will be respected as much as possible, although the term differential mobility spectrometry (DMS) will be used to describe the principle. The technique of DMS for analytical measurements was developed in Russia and reported in journal articles in the early 1990s,139,140 and a public presentation was given in 1995 at the 4th International Workshop on Ion Mobility Spectrometry in Cambridge, England.141 In a drift tube based upon field-dependent mobilities, ions are carried with gas flow from an ion source and through an analyzer region that is comprised of two parallel plates or a pair of concentric tubes. This is shown in Figure 4–14A for a planar design in which plates are separated by 0.5 mm. One electrode is held at ground potential, whereas an RF electric field, E(t), is applied to the other plate. The waveform for the electric field is typically a 1 MHz asymmetric rectangular shape in which Emax≥ 20,000 V/cm and Emin~1000 V/cm. This field is applied perpendicular to the gas flow and causes the ions to oscillate between the plates in a direction transverse to that of gas flow. Ions are displaced toward the top plate (or electrode) during the high-field portion of the waveform with a velocity described in Equation 4–5: (4–5) where terms are ion velocity perpendicular to gas flow direction; K(E), field-dependent mobility; and E(t), electric field. During the low-field portion, the ion movement is reversed, and the ions drift toward the lower plate with a velocity also described by Equation 4–5; however, the value of K(E) will be different from that during the high-field portion of the waveform. For each RF period, the ion is displaced toward the top plate by a net amount Δh. The average value of ±Δh for an ion species is determined by the duty cycle of the RF field and the field dependence of the mobility K(E). Thus, the total displacement of the ion as it moves along the x-axis in the direction of the y-axis

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_155... 06/05/2007

Página 1 de 1

Page 156

FIGURE 4–14 The differential mobility spectrometer (DMS) made by Sionex: (A) schematic of operation—only ions of a given mobility (α=0) traverse the space between the parallel plates, whereas ions with α>0 or α<0 collide with one of the plates; (B) the ion current as a function of the compensating voltage (the mobility spectrum), showing that whereas protonated monomers reach the detector when the compensation voltage is negative, proton-bound dimers reach the detector when the compensating voltage is positive.

is nΔh as governed by the number of waveform periods (n) experienced by the ion. Ion displacement in the y-axis can be understood as a net displacement caused by the differential mobility ΔK as shown in Equation 4–6: ΔK–K(E)highfield–K(E)lowfield

(4–6)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_156... 06/05/2007

Página 1 de 1

Page 157 Only ions with a total transverse displacement less than the distance to the upper plate will reach the detector, whereas all other ions will collide with drift-tube wall (or plates) and be neutralized or annihilated. Ions of a given ΔK can be kept from striking the upper plate (allowing a stable passage to the detector) by applying a low DC voltage (called the compensation voltage) to the lower plate to draw the ion away from a collision with the upper plate. An ion that successfully passes through the analyzer reaches a detector (Faraday plate) that is located down flow from the plates, and a signal is generated. As shown in Figure 4–14B, the position of ions in the spectrum depends on the polarity and amplitude needed for the compensation voltage and is characteristic of an ion, reflecting the field dependence for that ion. A sweep of the compensation voltage will provide a differential mobility spectrum, i.e., a measurement of all the ions in the analyzer. Unlike a traditional mobility spectrometer, the DMS has no ion shutters and ions are continuously introduced into the analyzer. In this way, the operation of a DMS analyzer is analogous to a quadrupole mass spectrometer (or mass filter), whereas a traditional drift tube resembles a TOF-MS in operation, although detailed principles of ion characterization by IMS and DMS cannot be compared directly. 4.4.2.1 Cylindrical FAIMS A configuration of field-dependent mobility with a cylindrical design, as shown in Figure 4–15, was introduced into North America and was called field ion spectrometry or FIS.142 In early configurations,139,140 sample molecules were ionized and passed with the gas flow between two concentric tubes (the inner and outer electrodes). The asymmetric electric field, or separation

FIGURE 4–15 Schematic of the cylindrical field ion spectrometer (manufactured by MSA). Ions with suitable mobility traverse the space between the two concentric tubes and reach the collector; other ions strike one of the tube walls.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_157... 06/05/2007

Página 1 de 1

Page 158 field, was applied between these inner and outer electrodes. In later designs, the ions were separated from the flow of sample into a stream of purified air. The cylindrical design was first introduced into the U.S. through Mine Safety Appliances (Pittsburgh, PA) in the mid-1990s,142 but the program of development and marketing was stopped after a few years. However, the technology was embraced by a team in Canada, and they continued the exploration and development of a cylindrical design.143 An early discovery was that the cylindrical design had an ion-focusing effect.144 Due to the differences in curvature of the inner and outer tubes, electric field lines are not homogeneous in the cross section, and this leads to ion focusing in the space between the surfaces. This focusing effect enhances the ion peak height or ion yield with increases in separation voltage. Another improvement made on this design was that formation of ions in an external ion source and the injection of ions from the sample matrix or supporting atmosphere into the drift tube. Thus, ions from an electrospray ion source could be introduced into the analyzer, practically without residual sample neutrals.145 The cylindrical design has been renamed field asymmetric ion mobility spectrometry (FAIMS) and is being actively developed and commercialized by lonalytics Corporation (Ottawa, Ontario).146 The limitations with the cylindrical design are the cost and attention needed in manufacture to ensure even gap throughout the drift tube and maintain this geometry in portable instruments. An advantage in the FAIMS and all other analyzers based on field-dependent mobilities is the exclusion of ion shutters in the drift tube. 4.4.2.2 Planar DMS An innovative approach to building an FAIMS instrument using microelectronics micromachined (MEMS) technology was developed by Miller et al.147 In this approach, metal electrodes were bonded onto glass or ceramic plates, using methods of mass production found in microfabrication technology Two plates of mirror image were separated by insulating rails and pressed together (Figure 4–16A) to form the drift tube. The size of the gap between the plates, a gastight seal, and the flow channel were provided by the gasket. A photograph of the metal elements on ceramic plates is given in Figure 4–16B. Dimensions of the separating electrodes were 5 mm×13 mm. A 5 mm×5 mm detector is located on each plate with a 5-mm gap between the detector and metal electrodes. A detector on one plate is floated to +5 V and the other detector is placed at −5 V so negative and positive ions, respectively can be simultaneously detected in one analyzer. Several ion sources have been combined with this device, including a photo-discharge lamp147,148 and a radioactive 63Ni source.149 As with the large cylindrical design, the duty cycle for sampling ions from the source is 100%, and unlike other designs, drift tubes can be manufactured using methods of mass production developed for microelectronics. The term DMS is used for this technology that is now commercially available as the microDMx™ Sensor System through Sionex Corporation (Waltham, MA).150 The number of investigations and applications of DMS with microfabricated

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_158... 06/05/2007

Página 1 de 1

Page 159

FIGURE 4–16 Planar microfabricated differential mobility spectrometer: (A) photograph showing that the device is the size of a quarter (Sionex Corp.); (B) a photograph of the plates (ceramic) shows the size of the drift region (5 mm×13 mm) and detectors (5 mm×5 mm).

analyzers has grown in the past few years through exploration by the original group151–156 and others.157 A microDMx™ Sensor System is now available on a commercial gas chromatograph, as described in Subsection 5.2.5.4. One advantage of the small size of this analyzer is that residence times for ions and vapors are low, and combinations with capillary gas chromatographs are simple and effective.151,152 Indeed, a planar DMS analyzer showed negligible or minor increases in variance in chromatographic peaks compared with a flame ionization detector, and these could be remedied with increased detector temperature.152 The planar small analyzer was easily interfaced to a mass spectrometer.158

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_159... 06/05/2007

Página 1 de 1

Page 160

4.4.3 Alternative Designs for Drift Tubes The characterization of ion velocity or ion mobility in air at ambient pressure can be accomplished using methods other than those described in the preceding text. Indeed, ion mobilities were not determined using an ion shutter and drift rings until 1929, and all previous designs involved creative combinations of mechanical and electrical methods. The only common feature in drift tubes between 1898 and 1929 was the motion of ions in an electric field, AC or DC. Some of these alternative designs are described in the following text and include one commercialized analyzer, the Environics aspirator design. 4.4.3.1 Aspirator Design One commercially available mobility-based analyzer, not described in the first edition of this book, is the aspirator-type analyzer.159 This drift tube is based on ion-counting technology dating from as early as 1901,160 in which air was drawn through a tube and atmospheric ions counted at the collecting electrode. Thorough modern treatments of aspiration counters were given in 1970 by Tammet161 and in 1986 by Misakian et al.162 The transition from an ion counter to a mobility analyzer was made as early as 1975 with a cylindrical design.163 Current commercial designs used for chemical measurements are planar, and only this design will be described. In an aspirator design for IMS, as shown in Figure 4–17, a flow of gas is passed between two plates in which an electric field is placed perpendicular to gas flow. The flow of gas is laminar, and ions formed external to the chamber are introduced through an orifice or slit into a stream of purified air. Ions are drawn by an electric field across a flow of gas based upon ion mobility. Ions eventually arrive at the plate at ground potential with some distance displaced by flow, as governed by mobility between the plates. Ions of high mobility will traverse the flowing gas quickly and arrive at the collecting plate after only a short displacement (Figure 4–17). Ions with low mobility require an extended time to move between the plates and are thus displaced further along the x-axis, striking the plate at a location different from the lighter ions. A set of detectors located on the plate and isolated electrically will register ions striking that location. Because there is no ion shutter, and ion characterization is occurring constantly, the duty cycle of ion characterization should be high. Thus, a record of all mobilities is available constantly, giving this analyzer an advantage over all other IMS analyzers. Nonetheless, this design is limited by diffusion broadening of ions, inlet geometries, and the number and width of detectors as shown in Figure 4–17. All these sum to provide an analyzer with comparatively low resolution of chemical information. The analytical result is a set of patterns corresponding to current developed at each detector. This provides a type of mobility histogram with concomitant low resolution. Still, these drift tubes can provide analytical information and are available commercially. Application areas of the aspiration ion mobility spectrometer are chemical weapons and toxic vapors,164,165 odors from fish,166 and ethanol for on-line

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_160... 06/05/2007

Página 1 de 1

Page 161

FIGURE 4–17 A schematic of the unique design of the IMCell (Environics, Finland). Ambient air is sampled and ionized, and the ions are deflected by an electric field into a clean gas flow moving right to left in the lower channel. Ion separation based upon mobility occurs in the bottom channel with ions reaching detectors with independent amplifiers.

analysis of ethanol beer.167 In this latter example, a polypropylene membrane was employed to stabilize moisture inside the drift tube. The aspirator design is inviting because of the inherent simplicity of the design, and one improvement was attempted through the scanning of the electric field, which is swept in discrete nonuniform steps.168 A plot of current vs. voltage (field) at a single detector was treated with a discrete transform based on the equations of Tammet to form a mobility spectrum.168 Another variation of this design was described with 13 collecting plates of differing lengths.169 This unit was used to measure ions in air during periods of dry and wet atmospheric conditions. 4.4.3.2 Planar Traditional Drift Tube Constructing a familiar cylindrical drift tube, as described in the preceding text, requires insulating and conducting rings (except the coated ceramic-tube design127,128), a machine shop to produce the components, and experience in correctly assembling the many parts of drift tubes. The advantages obtained from the planar microfabricated DMS described earlier has been the motivation for the design of a traditional planar drift tube.170 This is shown in Figure 4–18A, in which the drift rings and associated electronics were made using microfabrication methods on two printed circuit boards. The printed circuit boards were pressed to a 3-mm thick Teflon gasket that defined the drift-tube volume (10 mm wide×3 mm high×10 cm long) and were held mechanically in a metal frame. The planar electrodes were used to establish a voltage gradient in the drift tube as done with a voltage divider in a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_161... 06/05/2007

Página 1 de 1

Page 162

FIGURE 4–18 Microfabricated planar linear-field IMS (NMSU design): (A) board schematic with the planar conducting elements made of gold stripes; (B) schematic of the assembled drift tube (two boards placed in parallel separated by an insulator); (C) typical mobility spectrum obtained with the device.

traditional cylindrical drift tube. Other components included an ion source, ion shutter, aperture grid, and detector, and these were added to the drift tube as exchangeable components. In Figure 4–18B, a mobility spectrum for 2-pentanone is shown, in which the field strength was 300 V/cm. Further development of this design is underway, and the implications for drift-tube design are yet unknown.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_162... 06/05/2007

Página 1 de 1

Page 163 4.4.3.3 Others Two other drift tubes merit mention here for the unconventional drift region, and have implications for drift designs. The drift tube of Irie et al.30 was formed with a drift region composed of no drift rings. Rather, ions were able to move from the ion shutter to the detector, a distance of 1 to 4 cm with a voltage of 1 to 5 kV, without additional drift rings. Peak width was ~45 μsec with a drift time of ~700 μsec, yielding a ratio for td:width of ~15. Although this ratio is low, the simple drift region demonstrates that mobility can be determined without the addition of drift rings. At the other extreme of ring structures is the 1970 design of Stevenson et al., who employed grids in place of rings.171 Potentials on the grids were established in forward and backward directions through the entire drift length, so the electric field has a triangular pattern with drift length. The grids were spaced at 1 cm, and two sawtooth waveforms of 0 to 1000 V and 150 Hz to 10 kHz were placed on alternating rings. A mobility constant can be related to the frequency of the sawtooth waveform, and only an ion of a given mobility is transmitted through the drift region. This drift tube was operated without an ion shutter and constituted a type of linear ion filter. Peak widths at half height were ~2 cm2 V−1sec−1 for a peak at 20 cm2 V−1sec−1 or a ratio of drift time:peak width of 10. Although this is low in comparison with analytical grade drift tubes, the ion density was large. Thus, this drift tube could be a type of preparative-scale mobility spectrometer.

4.4.4 Miniaturized Drift Tubes The first commercial mobility spectrometers had a footprint of an executive-size desk and had a height of ~1.5 m. Within 15 years they had been made handheld size, in great measure due to advances in the miniaturization of electronics. However, this reduction in size was accompanied by the first significant decrease in drift-tube size of the handheld CAM analyzer, in which the size was reduced to 25 mm OD (outside diameter)×85 mm length from 90 mm OD×200 mm length for the Beta-VI in 1970.116 Interest in the miniaturization of the drift tube was seen as early as when the small ceramic tubes were prepared at Bendix (then Environmental Technology Group in Baltimore).127,128 Baumbach et al. advocated a small analyzer—called an ion mobility sensor—that lacked the resolution of a conventional spectrometer.172 Baumbach’s team has continued to miniaturize drift tubes, and current drift tubes are now the size of a cigarette.173 This drift tube is commercially available as the μIMS®-ODOR from G.A.S. (Gesellschaft für analytische Sensorsysteme mbH). Another miniaturized and simplified drift tube was incorporated into the palm-sized unit now known as the lightweight chemical detector (LCD).122,174 In this instrument, the drift rings are attached to a circuit board that contains other parts of the electronics. The structure is the same as other traditional drift tubes although the drift-tube components are made up of metal squares 15 mm on a side and with a length of 50 mm. The instrument is commercially available from Smiths Detection.175 A last commercial effort at miniaturization

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_163... 06/05/2007

Página 1 de 1

Page 164 is the microfabricated DMS analyzer now available commercially from Sionex and described earlier.147–158 As shown in Figure 4–16, the active region in which ions are separated is only 5 mm×13 mm rectangles separated by 0.5 mm. Another example of miniaturization is that of Oak Ridge National Laboratory, in which a drift tube was developed with conduction rings of 1.7-mm diameter, and the drift tube was 35 mm in length.176 Twenty-five rings were insulated by spacers and attached to miniresistors. The drift tube was placed in a metal support that was heated between 22°C and 250°C with resistive cartridge heaters. The ion source was an ultraviolet laser with a pulse energy of 0.1 mJ/pulse. As with other laser sources with no ion shutters, bandwidth was broad and particularly so for the small dimensions of the drift tube. Miniaturization may continue to be of interest in the development of IMS analyzers, and the ion shutter will continue to be a limiting constraint with traditional designs.

4.5.1 Detection Devices and Methods 4.5.1.1 Detection Methods The most common and simplest detector used in a traditional drift tube is a circular metal plate that serves as a Faraday plate. Collision and annihilation of ions on the detector, which is attached to the inverting input of an operational amplifier that is placed at virtual ground, are accompanied by a current flow. Currents between ~10−10 and 10−11 A are amplified and converted to voltages of 1 to 10 V. A very important feature present in most designs is a metal mesh, called the aperture grid or screen, which is placed at a distance of 0.5 to 2 mm from the collector and biased to a voltage so the field between the aperture grid and detector is from 300 to 600 V/cm in air at ambient pressure. The purpose of the aperture grid is to prevent induced current flow in the detector with the approach of the ion swarm through the drift region. Distortion of the mobility spectrum can be observed when the aperture grid is removed, although this will be dependent upon the length of the drift tube and ion speed. With the aperture grid, the detector is electronically blind to the approaching swarm until the ions actually pass through the aperture grid. Between the aperture grid and the detector, the swarm induces a current flow before the ions reach the detector; however, this distortion is negligible on the timescale of the mobility spectrum.138 The potential between the aperture grid and detector affects both peak shape and ion intensity, and these are influenced by the fields immediately in front of the aperture grid. Thus, the electric field between the aperture and detector should be adjusted or optimized whenever changes are made in the field strength in the drift region.126 The size of the aperture grid and collector, relative to each other and to the diameter of the driftregion guard rings, affect the sensitivity and resolution of the analyzer. To put it simply, a collector and aperture that have the same guard-ring diameter would provide maximum sensitivity because all ions that traverse the drift region could strike the collector plate. On the other hand, the signal would be broadened and resolution reduced if all ions

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_164... 06/05/2007

Página 1 de 1

Page 165 of the same type that follow different paths would reach the collector plate because the spread in time would be large.124 Resolution may be improved if either the collector plate or the aperture grid has a smaller diameter than the guard rings. This would reduce the ion current and sensitivity but would allow only ions that travel close to the center of the drift region to strike the collector plate, decreasing the spread in arrival times of ions. In a recent paper, high resolution in a high-flow drift tube was achieved by restricting the opening of the aperture grid.177 One of the most exciting developments in technology of IMS drift tubes may be found in the development of a detector that is capable of electron multiplication for ions in air at ambient pressure.178 Although technical details of this detector are under development, reported gains suggest that detection limits may be decreased to a few tens or hundreds of ions.

4.5.2 Signal Acquisition and Processing In an IMS drift tube with a linear electric field, the ion mobility spectrum is a plot of the signal intensity (ion current) as a function of the drift time, which is established by the ion shutter. However, the definition of the time base requires attention to detail and several possibilities exist: Zero can be set to the leading edge, the falling edge, or the center of the ion shutter pulse and some correction may be needed to the axis for drift time. The output from a mobility spectrometer consists of analog signals, which today are digitized and commonly averaged before storage. Signal averaging is generally necessary because the spectrum from a single sweep is too noisy to be of practical use. Averaging n similar spectra will improve the signal-to-noise ratio by a factor equal to the square root of n. Considering that a single mobility spectrum is typically acquired in 5 to 20 msec, averaging 400 spectra should take about from 2 to 8 sec (depending on the duty cycle of the instrument) and will improve the signal-to-noise ratio by a factor of 20. Doubling the number of averaged spectra will result in a further improvement of only 40% of the signal-to-noise ratio. The actual amount of averaging required would depend on the signal and noise levels and the minimum detectable level needed. Some time must be allowed for computer “overhead,” depending on the hardware and software used, for displaying and processing the information. Thus, a reasonably good averaged signal may be obtained in under 10 sec, which is a short enough response time for most applications. The availability of suitable software gives users a choice between readymade programs, generally supplied with the IMS by its manufacturer, and a software package that allows the user to custom-design the software according to specific needs. Digital signal averaging was once a topic of active development in IMS179–181 but is not being advanced further today. Needs are apparently met by commercial interface cards and software. Although signal averaging is a useful technique, this method presupposes the subtlety of user knowledge of spectral content; furthermore, there is an expectation that chemical information will not be lost by averaging. Once

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_165... 06/05/2007

Página 1 de 1

Page 166 the spectra are averaged and stored, there is no longer any opportunity to subject the spectra to further processing. Harrington has argued that a preferable alternative to averaging is the storage of all spectra in an experiment or measurement. Because spectra are obtained often at 30 Hz for times from a few to 40 min and each spectrum may be ~2 KB, the amount of data handled can easily exceed 3.6 Mb/min. This necessitates computational processing capabilities that have been until recently considered unreachable for ordinary applications. However, favorable trends in computation with improving speed, storage, and data handling may make this suggestion by Harrington possible. Nonetheless, Harrington has described simple-to-use interactive self-modeling mixture analysis (SIMPLISMA),182 which can allow a user to interrogate spectra in real time and after the measurement. The original SIMPLISMA algorithm was modified and used to extract information from the mobility spectra so that the concentration of each component in a mixture could be determined.182 To facilitate large data storage, two-dimensional wavelet compression (WC2) was developed.183 The WC2 method was used to compress data and reduce the noise level while retaining the chemical information.

4.5.3 Analysis of Spectra One of the most promising areas of spectra analysis in IMS is the use of neutral networks (NN) for interpreting response from an IMS and the incorporation of NN tools into IMS software.184–187 Boger used an experimental data set of nine IMS spectra from measurement of bromine concentrations185 to train the neural network. From the 330 channels used for teaching, only ten were retained after five network teaching and reduction cycles, with four hidden neurodes (49 connections). The prediction of one bromine concentration that was not used in the teaching gave an error of 2.7%. Boger reported that the training process done off-line took less than an hour (on an old IBM/PS2 computer), but the actual calculation took only seconds once the network was trained. Other examples of utilization of NN methods to qualitatively and quantitatively interpret IMS spectra were recently made.186 In the former case, mobility spectra of several mixtures of up to six aliphatic amines in air were analyzed with the aid of a trained NN and presented in the form of a truth table. The results were quite satisfactory with less than 20% errors (false positive or false negative). Different concentrations in air of an analyte (hydrogen fluoride, bromine, or dimethyl formamide) were measured and the results interpreted by NN methods with deviations of less than 10% from the known concentrations. Mobility spectra have been analyzed with neural networks in order to learn what parts of mobility spectra are significant to identifying a chemical or to assigning a spectrum to a chemical family.188– 190 It was possible to categorize spectra from IMS characterization of chemicals at temperatures as low as 40°C in drift gas containing low moisture (~0.1 ppm) into chemical families with high success when only the earliest part of the mobility spectrum was used for network training. Indeed, ions of low mobility were not helpful in

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_166... 06/05/2007

Página 1 de 1

Page 167 classifying spectra. Ions of high mobility or low drift time were found in each mobility spectrum in low abundance, previously overlooked, and the patterns of these ions were suggestive of fragmentation of ions. The fragmentation was understood to form ions common to a chemical family; thus, similar fragment ions were formed in all aldehydes, unique ions were produced in alcohols, and so forth, for other functional groups of compounds. These fragment ions were evident when the intensity scale was changed from normal to logarithmic format, and spectra were processed for noise reduction and deconvoluted. Elevated temperatures were seen to increase fragmentation.

4.5.4 Modes of Signal Display The standard mode for display of mobility data is the regular mobility spectrum (ion current vs. drift time), such as the spectrum shown in Figure 1–2 of Chapter 1. Other common modes include a display of the signal intensity (or analyte concentration) as a function of time, which is particularly useful for applications in which continuous monitoring is required. A watershed or pseudo 3-D plot combines these two modes, as shown in Figure 4–19.

FIGURE 4–19 Watershed mobility spectra of toluene di-isocyanate (TDI) obtained with a handheld CAM (Graseby Dynamics). Each new mobility spectrum is displayed in the front, whereas spectra obtained earlier are gradually moved to the back. Thus, a change in intensity of each ion may be monitored as a function of elapsed time. (Source: Brokenshire, J.L., Dharmarajan, V., Coyne, L.B., and Keller, J., Near real time monitoring of TDI vapor using ion mobility spectrometry (IMS), Proc. SPI. Annu. Tech/Marketing Conf., 1989, 459–469.)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_167... 06/05/2007

Página 1 de 1

Page 168 Each mobility spectrum is displayed in the standard mode and is moved one step back every time a new spectrum is acquired, so that the overall effect is similar to a watershed. The change of concentration with time may be observed from the variation in the peak height of a signal with the appropriate drift time (such as the TDI peak in Figure 4–19). A different type of display, particularly useful for hyphenated techniques, such as GC/IMS, is the pseudo 3-D (or topographic) display shown in Figure 4–20. A topographic plot is shown from continuous monitoring of ambient air over 9 days using a DMS analyzer equipped with a photodischarge ion

FIGURE 4–20 Three dimensional (3-D) display of continuous measurement of the mobility spectrum of aromatic hydrocarbons (BTEX) in ambient air: (A) topographic plot of the differential mobility spectra vs. time (in days); (B) a plot of the ion intensity at a compensation voltage of −2 V as a function of time.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_168... 06/05/2007

Página 1 de 1

Page 169 source.148 In Figure 4–20B, a plot of ion intensity against time is shown for the product ions appearing at −23V on the compensation voltage axis. In Figure 4–20A, the topographic plot shows the differential mobility spectra across the entire sampling episode. The advantage of the topographic plots is apparent as they provide a summary view of all features of the data set. In cases in which the IMS is used as a monitor, the concentration of the target chemical or compound of interest can be displayed numerically. The IMS is generally connected to some alarm system that is triggered once a preset threshold is passed. As similar setups are used in most other monitoring systems and are not really a part of IMS signal processing, they will not be described here.

4.6 Selection of Materials In the previous discussions, a large number of drift-tube designs and components for drift tubes have been described. Little has been described heretofore, or in other reviews of IMS technology, regarding the materials used to construct drift tubes. The choices made regarding materials will govern both the range of operating temperatures possible for a drift tube and the quality of the drifttube response, in ways that are somewhat independent of the design or style of the drift tube. In particular, two complications may arise from improper choice of materials or the use of materials under unsuitable operating conditions: chemicals or impurities will be emitted or desorbed from the materials, causing unwanted contamination or changes in the gas-phase ionization chemistry, or compounds in a sample may be adsorbed or lost to surfaces of materials through chemisorption or decomposition. The following discussion is of importance to both instrument builders and users of IMS analyzers. Users should be mindful of the issue because commercial instruments have been designed for particular applications and materials included with best compatibility for the application. Thus, a CAM contains plastic components that are wholly acceptable for a drift tube operated at ambient temperature. However, the CAM drift tube cannot be operated without unwelcome consequences at temperatures above ~70°C. In contrast, drift tubes designed for explosives and drug detection at elevated temperature will contain materials that do not off-gas and ideally contain surfaces on the inlet that are not too adsorptive or reactive for hot vapors of polar molecules such as drugs, or thermally labile compounds such as explosives. Instrument builders should appreciate that RIPs in carefully designed and built instruments could be protonated monomers or proton-bound dimers of compounds released from even minor components in a drift tube and that there is no substitute for careful use of proper materials.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_169... 06/05/2007

Página 1 de 1

Page 170

4.6.1 Conducting Materials Conducting materials in a drift tube are used in traditional IMS designs for the drift rings, ion shutters, aperture grid, detector plate, gas lines, protective housing, and connecting wires for the drift-tube components. Metals might also be found in the ion source assembly (electrospray needle, electrodes in a CD source, and the holder for 63Ni foil) and in the inlet as a sample transfer line. Stainless steel is the metal preferred for all drift tubes, owing to durability, resistance to chemical attack such as oxidation, and thermal properties of low or no off-gassing. Gold plating has been used with some drift-tube designs but is believed to be unnecessary, only incurring extra cost and complications. Stainless steel is a poor conductor of heat and will require extended time to warm and cool. Nonetheless, the cost and availability of stainless steel favors its use as a detector plate, aperture grid, and drift rings. The disadvantages of stainless steel are the cost and difficulty of machining components from bar, rod, or plate stock. Thus, stamped or formed pieces will limit the costs of manufacture. For example, drift rings can be made quickly and easily from stainless steel washers available from commercial sources. The washers are faced and trimmed on OD and ID (internal diameter) on a lathe with savings in time and cost vs. complete manufacture from rod stock. In one drift-tube design from Baumbach’s team in Germany, the drift tube was defined by a Teflon tube, and the drift rings were on the outside of the Teflon tube.125 In this design, requirements for chemical reactivity could be relaxed and drift rings could be copper, brass, or even aluminum. Similarly, a high-resolution drift tube described by Brokenshire and manufactured by Graseby Dynamics in the early 1990s consisted of a glass tube with thin copper bands as drift rings attached on the external surface of the tube.129 Metals share common electrical properties, so considerations will largely be chemical reactivity, cost, and physical durability of the metal. One special feature is thermal expansion, which can become a consideration when drift tubes are heated; aluminum has the worst coefficient of thermal expansion of metals suitable for IMS drift tubes, and this should factor into designs involving aluminum housings or flanges. Metal can also be part of an inlet and encounter or make contact with sample. Such metal should be limited to stainless steel or nickel, whereas other common metals such as copper can be excessively reactive or adsorptive toward organic compounds and corrosive gases and should be avoided. Teflon is highly advisable for surfaces of the inlet. Alternatively, metals could be passivated with a thin layer of gold, treated through a proprietary deactivation method, or replaced with silica-lined steel tubing. Inexpensive metals such as copper may be used for prototyping drift tubes, although use should be restricted to low temperature and nonabsorptive chemicals.170 Ion shutters may be made from stainless steel wire, and considerations of thermal expansion may arise within a shutter assembly. A sophisticated shutter design was based upon wires stretched between two ceramic halves.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_170... 06/05/2007

Página 1 de 1

Page 171 The ceramic pieces were separated by thermally stable springs so that expansion or contractions could occur with variations in temperature without sagging or snapping of wire from changes in tension. Original literature sources should be consulted for guidance on metals and materials for these specialized components.

4.6.2 Insulating Materials Electric fields are established in a drift tube through voltage differences on conducting pieces that are electrically insulated and separated by distances that are ideally known and fixed. In the earliest drift-tube designs of the Beta VI and subsequent generations from the same company, rings were separated by three sapphire balls ~2 mm in diameter symmetrically arranged around the drift ring. In the CAM, the drift rings are held by three stainless steel rods with Teflon sleeves. The rings are insulated with Teflon, and the spacing between the rings is established by three 4-mm long ceramic tubes with internal diameters large enough to fit over the Teflon-covered rods. In both of these designs, the insulators are Teflon, ceramic, or an inorganic crystalline solid. The choice of insulator is as important as the conductor for the same analytical performance and practical reasons: chemical reactivity, physical integrity, cost, and convenience of manufacture. Additionally, plastics and other materials are prone to off-gassing with increases in temperature through impurities or small oligomers retained from the production process or decomposition of the polymer. Another property of the insulator is high resistance to electrical conduction with DC volume resistivities of up to 1016 ohm-cm. Insulators used in drift tubes include glass, machinable glass, ceramic, machinable ceramic, mica, and plastic materials such as polyethylene, polyimide, and Teflon. An insulator of high resistivity, high thermal tolerance, and good chemical inertness is Macor™ (55% fluorophlogopite mica and 45% borosilicate glass), which has the added benefit of being machinable.191 Although Macor can be worked on a lathe and milling machine, precautions are needed to avoid inhalation of dust, and the use of liquid streams may produce a waste liquid that is corrosive to the metal of the lathe and milling machine. Machinable, almost soft, ceramic is available that can be fired to temperatures over 600°C to become hardened ceramic.192 Changes in size after firing can be over 10%, and heating must be uniform and held contant (±2°C), making the application of this ceramic for delicate or fitted parts somewhat difficult. Polymers such as polypropylene and fluorocarbon resins (the Teflon family of polymers) offer ease of machining and chemical inertness, and drift tubes may contain nylon or polypropylene insulating rings so long as temperatures are low. The best of the fluoropolymers regarding temperature stability is PTFE (polytetrafluoroethylene) with an upper temperature of operation >260°C.193

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_171... 06/05/2007

Página 1 de 1

Page 172 These materials have excellent electrical properties for the high impedances of IMS drift tubes and do not exhibit electrical breakdown under high voltage. Sadly, the thermal expansion coefficients of PTFE make deformation a concern, and drift tubes made with PTFE and operated at high temperature must be kept in a durable external shell. Finally, PTFE is a poor conductor of heat, so the application of heat to a drift tube made from PTFE must be made gradually, or localized overheating may occur with melting or decomposition of the plastic. However, the cost and convenience of machining PTFE makes this material attractive for drift tubes even when operated at temperatures as high as 250°C.

4.6.3 Miscellaneous Materials The same considerations used for insulators and conductors described in the preceding text are applicable to miscellaneous items such as tubing, gaskets, seals, and specialized materials such as adhesives and shells. In low-temperature drift tubes with handheld analyzers, unions for tubing and pump seals and other connections are made using plastics. In research-grade analyzers, tube connections are usually made with compression unions or are silver-soldered. However, material for other parts of a drift tube should not be ignored because even small sources of contamination can render ultrasensitive analyzers such as IMS drift tubes unworkable. Tubing can be used for preheating the drift gas because gas temperature in the drift and reaction regions will affect ion clustering and fragmentation of ions. Thus, the appearance of a mobility spectrum will be influenced by gas temperature. No systematic description of preheating a gas has been given in which the entire gas stream has been monitored by sensors capable of accurately measuring the gas temperature, so that the subject remains unsettled even after 30 years of modern analytical IMS. In practice, thermocouples have often been attached to the metal housing of a drift tube, providing false measures of the temperature of the gas. In a design common to NMSU, several meters of 0.25-in.-OD aluminum tubing (a good conductor of heat) are wrapped around the body of the drift tube. Drift gas is passed through the tubing before the gas enters the drift tube near the detector. Preliminary studies made on meter lengths of identical tubing (independent of the drift tube) show that the gas temperature is not identical to the wall temperature of the tubing and that preheating of gas had occurred, as elevated temperatures were seen with thermocouples placed in the gas effluent. Although the gas was warmed during passage through the pre-heat tubing, measurements of temperature were not deemed reliable owing to thermocouple mass. Cooper tubing should be avoided completely, if preheating is needed, owing to the formation of copper oxides. Under prolonged and elevated temperatures in oxidizing gases such as air, copper oxides may flake and send particulate into the gas streams and places downflow. Drift tubes under a slight pressure drop must be engineered with seals and gaskets to prevent inflow of surrounding atmosphere. This can be

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_172... 06/05/2007

Página 1 de 1

Page 173 accomplished with thermoplastics or metal-to-metal seals when elevated temperatures are needed or durable elastomers for ambient-temperature instruments. Housings and flanges have been universally stainless steel or aluminum. Membranes have been made from polypropylene and methylpolysilicone and have been made in Teflon holders, in simple modifications of a Swaglock union,194 and in stainless steel flanges.195

4.7 Summary A mobility spectrometer may be seen as an assembly of components, and the discussion in this chapter was intended to illustrate the opportunities and complexities of modern IMS designs. There are few analytical challenges that can be envisioned for which an IMS instrument could not be proposed, even bearing in mind that some substances or ions may not be stable in air at ambient temperature. Nonetheless, the trend in drift-tube technology is toward increased diversity and design, and this is understood as healthy for a dynamic and growing technology. IMS should no longer be considered an emerging technology, although there are still technological advances that would be welcome in nearly all aspects of the drift tube. The number of instruments has increased, security capabilities have improved, and so has preparedness for the familiar threats that IMS is deployed against. Today, handheld analyzers are available for explosives detection, and palm-size instruments are available for monitoring chemical weapons. The companies that addressed these markets continue to do so, and neither general instruments nor components for drift tubes are available from these manufacturers. Nothing remotely close to the developments seen with GC in the 1950s and 1960s, or mass spectrometers in the 1960s and 1970s exists with IMS today. However, several small companies have begun to provide instruments that may be considered general purpose. Those requiring traditional high-temperature drift tubes must undertake in-house development with an array of options. The preceding discussion may be helpful in choosing components for an intended application. On the CD with this book, circuits and other drawings are available to further assist such endeavors.

References 1. Carr, T.W. (Ed.), Plasma Chromatography. Plenum Press, New York, 1984, 259 pp. 2. Eiceman, G.A.; Karpas, Z., Ion Mobility Spectrometry, CRC Press, Boca Raton, FL, 1994. 3. St. Louis, R.H.; Hill, H.H., Jr., Ion mobility spectrometry in analytical chemistry, CRC Crit. Rev. Anal Chem. 1990, 21, 321–355.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_173... 06/05/2007

Página 1 de 1

Page 174 4. Karpas, Z., Ion mobility spectrometry in forensic science, Forens. Sci. Rev. 1989, 1, 104–119. 5. Eiceman, G.A., Ion mobility spectrometry as a fast monitor of chemical composition, Trends Anal Chem. 2002, 21, 259–275. 6. Ewing, R.G.; Atkinson, D.A.; Eiceman, G.A.; Ewing, G.J., A critical review of ion mobility spectrometry for the detection of explosives and explosive related compounds, Talanta 2001, 54, 515–529. 7. Hill, H.H.; Simpson, G., Capabilities and limitations of ion mobility spectrometry for field screening applications, Field Anal. Chem. Technol. 1997, 1, 119–134. 8. Baumbach, J.I.; Eiceman, G.A., Ion mobility spectrometry: arriving on site and moving beyond a low profile, Appl. Spectrosc. 1999, 53, 338A–355A. 9. Turner, R.B.; Brokenshire, J.L., Hand-held ion-mobility spectrometers, Trends Anal. Chem. 1994, 13, 275– 280. 10. Karachevtsev, G.V., Methods for ion mobility measurements and their application, Uspekhi Khimii 1996, 65(8), 712–717. 11. Stach, J., Ion mobility spectrometry—principles and applications, Analytiker-Taschenbuch 1997, 16, 119– 154. 12. Schweitzer, W.; Latscha, H.P.; Keller, T., Ion mobility spectrometry (IMS). A short introduction. GIT Labor-Fachzeitschrift 2003, 47(11), 1154–1157. 13. Adler, J.; Arnold, G.; Baumbach, J.I.; Doering, H.-R., Principle and possible application (to trace gas analysis) of ion mobility spectrometry. ZfI-Mitt. 1990, 154, 5–11. Also see Arnold, G.; Doering, H.-R., Ion mobility spectrometry, ZfI-Mitt. 1990, 154, 13–34 and related articles in Volume 154. 14. Witkiewicz, Z.; Stryszak, E., Plasma chromatography. I. Physicochemical fundamentals of plasma chromatography and the apparatus involved, Chemia Analityczna 1985, 30, 491–518. Also see Witkiewicz, Z.; Stryszak, E., Plasma chromatography. II. Physicochemical fundamentals of plasma chromatography and the apparatus involved, Chemia Analityczna 1985, 30, 633–697. 15. Long, Y.T.; Guo, Y., Ion mobility spectrometry and its recent advances, Fenxi Huaxue 1997, 25, 97–103. 16. Li, E; Xie, Z-.; Schmidt, H.; Sielemann, S.; Baumbach, J.I., Ion mobility spectrometer (IMS): a new online monitor of trace volatile organic compounds, Guangpuxue Yu Guangpu Fenxi 2002, 22(6), 1025–1029; also see: Ting, L.Y.; Yan, G.; Qin, L.M., Alternate reagent ion method in ion mobility spectrometry, Rev. Anal. Chem. 2000, 19, 179–189. 17. Preston, J.M.; Rajadhyax, L., Effect of ion-molecule reactions on ion mobilities, Anal. Chem. 1988, 60, 31– 34. 18. Proctor, C.J.; Todd, J.F.J., Alternative reagent ions for plasma chromatography, Anal. Chem. 1984, 56, 1794–1797. 19. Eiceman, G.A.; Wang, Y.F.; Garcia-Gonzalez, L.; Harden, C.S.; Shoff, D.B., Enhanced selectivity in ion mobility spectrometry analysis of complex mixtures by alternate reagent gas chemistry, Anal. Chim. Acta 1995, 306, 21–33. 20. Munro, W.A.; Thomas, C.L.P.; Langford, M.L., Characterization of the molecular ions produced by a dinitroalkane in positive mode ion mobility spectrometry with water, dichloromethane and ammonia reactant ion chemistries, Anal. Chim. Acta 1998, 374, 253–267. 21. Kunz, R.; Dinatale, W.F.; Becotte-Haigh, P, Comparison of detection selectivity in ion mobility spectrometry: proton-attachment versus electron exchange ionization, Int. J. Mass Spectrom. 2003, 226, 379–395.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_174... 06/05/2007

Página 1 de 1

Page 175 22. Lawrence, A.H.; Neudorfl, P., Detection of ethylene glycol (ethanediol) dinitrate vapours by ion mobility spectrometry using chloride reagent ions, Anal. Chem. 1988, 60, 104–109. 23. Spangler, G.E.; Lawless, P.A., Ionization of nitrotoluene compounds in negative ion plasma chromatography, Anal Chem. 1978, 50, 884–892. 24. Spangler, G.E.; Epstein, J., Detection of HF using atmospheric pressure ionization (API) and ion mobility spectrometry (IMS), 38th ASMS Conference on Mass Spectrometry and Allied Topics, Tucson, AZ, 1990. 25. McKinley, J.; Majors, R.E., The preparation of calibration standards for volatile organic compounds, LCGC 2000, 18, 1024–1033. 26. Saltzman, B.E.; Burg, W.R.; Ramaswamy, G., Performance of permeation tubes as standard gas sources, Environ. Sci. Technol 1971, 5, 1121–1128. 27. Maria, P.-C; Gal, J.-F.; Balza, M.; Pere-Trepat, E.; Tumbiolo, S.; Couret, J.-M., Using thermogravimetry for weight loss monitoring of permeation tubes used for generation of trace concentration gas standards, Anal Chem. 2002, 74, 305–307. 28. Ketkar, S.N.; Dheandhanoo, S., Use of ion mobility spectrometry to determine low levels of impurities in gases, Anal Chem. 2001, 73, 2554–2557. 29. Dheandhanoo, S.; Ketkar, S.N., Improvement in analysis of O2 in N2 by using air drift gas in an ion mobility spectrometer, Anal. Chem. 2003, 75, 698–700. 30. Irie, T.; Mitsui, Y.; Hasumi, K., A drift tube for monitoring ppb trace water, Jpn. J. Appl Phys., Part 1 (Regular Papers, Short Notes & Review Papers), 1992, 31, 2610–2615. 31. Grayson, M.A.; Wolf, C.J., Two-stage molecular separator for connecting a gas chromatograph to a mass spectrometer, Anal. Chem. 1970, 42, 426–430. 32. Karpas, Z., Operational experience with IMS for process control and environmental monitoring, Fifth International Workshop on Ion Mobility Spectrometry, Jackson, WY, USA, 1996. 33. Dam, R.J., Analysis of toxic vapors by plasma chromatography, in Plasma Chromatography, T.W.Carr, Ed., Plenum Press, New York, 1984, chap. 6. 34. Eiceman, G.A.; Snyder, A.P.; Blyth, D.A., Monitoring of airborne organic vapors using ion mobility spectrometry, Int. J. Environ. Anal. Chem. 1990, 38, 415–425. 35. Cao, L.; Harrington, P.B.; Harden, C.S.; McHugh, V.M.; Thomas, M.A., Nonlinear wavelet compression of ion mobility spectra from ion mobility spectrometers mounted in an unmanned aerial vehicle, Anal. Chem. 2004, 76, 1069–1077. 36. Spangler, G.E.; Carrico, J.P., Membrane inlet for ion mobility spectrometry (plasma chromatography), Int. J. Mass Spectrom. Ion Phys. 1983, 52, 267–287. 37. Bocos-Bintintan, V.; Brittain, A.; Thomas, C.L.P., The response of a membrane inlet ion mobility spectrometer to chlorine and the effect of water contamination of the drying media on ion mobility spectrometric responses to chlorine, Analyst 2002, 126, 1539–1544. 38. Ritter, J.J.; Adams, N.K., Exponential dilution as a calibration technique, Anal. Chem. 1976, 48, 612–619. 39. Lorenz, L.J.; Culp, R.A.; Dixon, R.T., Improved exponential dilution flask for gas chromatography, Anal. Chem. 1970, 42, 1119–1120. 40. Dworzanski, J.P.; Kim, M.-G.; Snyder, A.P.; Arnold, N.S.; Meuzelaar, H.L.C., Performance advances in ion mobility spectrometry through combination with high speed vapour sampling, preconcentration and separation techniques, Anal. Chim. Acta 1994, 293, 219–235. 41. Dworzanski, J.P.; Kim, M.G.; Snyder, A.P.; Arnold, N.S.; Meuzelaar, H.L.C., Performance advances in ion mobility spectrometry through combination with

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_175... 06/05/2007

Página 1 de 1

Page 176 high speed vapour sampling, preconcentration and separation techniques, Anal Chim. Acta 1994, 293, 219– 235. 42. Young, D.; Thomas, C.L.P.; Breach, J.; Brittain, A.H.; Eiceman, G.A., Extending the concentration and linear dynamic range of ion mobility spectrometry with a sheath flow inlet, Anal Chim. Acta 1999, 381(1), 69–83. 43. Young, D.; Eiceman, G.A.; Breach, J.; Brittain, A.H.; Thomas, C.L.R, Automated control and optimization of ion mobility spectrometry responses using a sheath-flow inlet, Anal. Chim. Acta 2002, 463, 143–154. 44. Harper, M, Sorbent trapping of volatile organic compounds from air, J. Chromatogr. A 2000, 885, 129–51. 45. Dettmer, K.; Engewald, W., Adsorbent materials commonly used in air analysis for adsorptive enrichment and thermal desorption of volatile organic compounds, Anal. Bioanal. Chem. 2002, 373(6), 490–500. 46. Poole, C.F.; Schuette, S.A., Isolation and concentration techniques for capillary column gas chromatographic analysis, HRC & CC, J. High Resolut. Chromatogr. Chromatogr. Commun. 1983, 6(10), 526–49. 47. Limero, T.; Reese, E., First operational use of the ISS-VOA in a potential contingency event, Intl. J. Ion Mobility Spectrom. 2002, 5(3), 27–30. 48. Bacon, A.T.; Reategui, J.; Getz, R.C., Acid Gas Monitor Based on Ion Mobility Spectroscopy, Environmental Technologies Group, Patent Numbers: US 5032721 A (07/16/91) & US 534701 A (06/01/90). 49. Schellenbaum, R.L.; Hannum, D.W., Laboratory Evaluation of the PCP Large Reactor Volume Ion Mobility Spectrometer (LRVIMS), Technical Report No. SAND89–0461*UC-515, Sandia National Laboratories, Albuquerque, NM, March 1990. 50. Cech, N.B.; Enke, C.G., Practical implications of some recent studies in electro-spray ionization fundamentals, Mass Spectrom. Rev. 2001, 20, 362–387. 51. Kebarle, P.; Tang, L., From ions in solution to ions in the gas phase, Anal. Chem. 1993, 65, 972A–986A. 52. Shumate, C, Electrospray ion mobility spectrometry, Trends Anal Chem. 1994, 13, 104–109. 53. Wittmer, D.; Luckenbill, B.K.; Hill, H.H.; Chen, Y.H., Electrospray-ionization ion mobility spectrometry, Anal Chem. 1994, 66, 2348–2355. 54. Penalver, A.; Pocurull, E.; Borrull, F.; Marce, R.M., Trends in solid-phase microextraction for determining organic pollutants in environmental samples, TrAC, Trends Anal Chem. 1999,18(8), 557–568. 55. Fallman, A.; Rittfeldt, L., Detection of chemical warfare agents in water by high temperature solid phase microextraction-ion mobility spectrometry (HT-SPME-IMS), Int. J. Ion Mobility Spectrom. 2001, 4, 85–87. 56. Poziomek E.J.; Eiceman, G.A., Solid-phase enrichment, thermal desorption, and ion mobility spectrometry for field screening of organic pollutants in water, Environ. Sci. Technol. 1992, 26, 1313–1318. 57. Johnson, R.C.; Cooks, R.G.; Allen, T.M.; Cisper, M.E.; Hemberger, P.H., Membrane introduction mass spectrometry: trends and applications, Mass Spectrom. Rev. 2000, 19(1), 1–37. 58. Garcia-Gonzalez, L., Extraction of Volatile Organic Compounds from Aqueous Solution using a Membrane with Detection by Ion Mobility Spectrometry, M.S. thesis, New Mexico State University, Las Cruces, NM December 1992. 59. Vinopal, R.T.; Jadamec, J.R.; deFur, P.; Demars, A.L.; Jakubielski, S.; Green, C; Anderson, C.P.; Dugas, J.E.; DeBono, R.F., Fingerprinting bacterial strains using ion mobility spectrometry, Anal Chim. Acta 2002, 457, 83–95.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_176... 06/05/2007

Página 1 de 1

Page 177 60. Lubman, D.M.; Kronick, M.N., Plasma chromatography with laser-produced ions, Anal Chem. 1982, 54, 1546–1551. 61. Lubman, D.M.; Kronick, M.N., Resonance-enhanced two-photon ionization spectroscopy in plasma chromatography, Anal Chem. 1983, 55, 1486–1492. 62. Young, D.; Douglas, K.M.; Eiceman, G.A.; Lake, D.A.; Johnston, M.V., Laser desorption-ionization of polycyclic aromatic hydrocarbons from glass surfaces with ion mobility spectrometry analysis, Anal. Chim. Acta 2002, 453(2), 231–243. 63. Illenseer, C; Lohmannsroben, H.-G., Investigation of ion-molecule collisions with laser-based ion mobility spectrometry, Phys. Chem. Chem. Phys. 2001, 3(12), 2388–2393. 64. Gormally, J.; Phillips, J., The performance of an ion mobility spectrometer for use with laser ionization, Int. J. Mass Spectrom. Ion Proc. 1991, 107(3), 441–451. 65. Eiceman, G.A.; Anderson, G.K.; Danen, W.C.; Ferris, M.J.; Tiee, J.J., Laser desorption and ionization of solid polycyclic aromatic hydrocarbons in air with analysis by ion mobility spectrometry, Anal Lett. 1988, 21(4), 539–52. 66. Huang, S.D.; Kolaitis, L.; Lubman, D.M., Detection of explosives using laser desorption in ion mobility spectrometry/mass spectrometry, Appl. Spectrosc. 1987, 41(8), 1371–1376. 67. Laiko, V.V.; Baldwin, M.A.; Burlingame, A.L., Atmospheric pressure matrix-assisted laser desorption/ionization mass spectrometry, Anal. Chem. 2000, 72(4), 652–657. 68. Bramwell, C.J.; Creaser, C.S.; Reynolds, J.C.; Dennis, R., Atmospheric pressure matrix-assisted laser desorption/ionization combined with ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2002, 5(3), 87–90. 69. Steiner, W.E; Clowers, B.H; English, W.A; Hill, H.H., Jr., Atmospheric pressure matrix-assisted laser desorption/ionization with analysis by ion mobility time-of-flight mass spectrometry, Rapid Commun. Mass Spectrom. 2004,18(8), 882–888. 70. Simmonds, P.G.; Fenimore, D.C.; Pettitt, B.C.; Lovelock, J.E.; Zlatkis, A., Design of a nickel-63 electron absorption detector and analytical significance of high-temperature operation, Anal Chem. 1967, 39, 1428– 33. 71. Siegel, M.W., Rate equations for prediction and optimization of chemical ionizer sensitivity, Int. J. Mass Spectrom. Ion Phys. 1983, 46, 325–328. 72. Thomson, J., Ionizing efficiency of electronic impacts in air, Proceedings of the Royal Society of Edinburgh 1931, 51, 127–141. 73. Jesse, W.P., Absolute energy to produce an ion pair in various gases by β-particles from sulfur-35, Phys. Rev. 1958, 109, 2002–2004. 74. Yun, C.-M.; Otani, Y; Emi, H., Development of unipolar ion generator—separation of ions in axial direction of flow, Aerosol Sci. Technol. 1997, 26, 389–397. 75. Paakanen, H., About the applications of IMCELLTM MGD-1 detector, Int. J. Ion Mobility Spectrom. 2001, 4, 136–139. 76. Leonhardt, J.W., New detectors in environmental monitoring using tritium sources, J. Radioanal. Nucl. Chem. 1996, 206 (2, International Conference on Isotopes, Proceedings, 1995, Pt. 4), 333–339. 77. Eiceman, G.A.; Kremer, J.H.; Snyder, A.P.; Tofferi, J.K., Quantitative assessment of a corona discharge ion source in atmospheric pressure ionization-mass spectrometry for ambient air monitoring, Int. J. Environ. Anal. Chem. 1988, 33, 161–183. 78. Bell, A.J.; Ross, S.K., Reverse flow continuous corona discharge ionization, Int. J. Ion Mobility Spectrom. 2002, 5(3), 95–99.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_177... 06/05/2007

Página 1 de 1

Page 178 79. Tabrizchi, M.; Khayamian, T.; Taj, N., Design and optimization of a corona discharge ionization source for ion mobility spectrometry, Rev. Sci. Instrum. 2000, 7, 2321–2328. 80. Tabrizchi, M.; Abedi, A., A novel electron source for negative-ion mobility spectrometry, Int. J. Mass Spectrom. 2002, 218, 75–85. 81. Hill, C.A; Thomas, C.L.P., A pulsed corona discharge switchable high resolution ion mobility spectrometer-mass spectrometer, Analyst 2003, 128(1), 55–60. 82. Taylor, S.J.; Turner, R.B.; Arnold, P.D., Corona-discharge ionization source for ion mobility spectrometer, PCT Int. Appl. 1993, 32 pp. CODEN: PIXXD2 WO 9311554 A1 19930610. 83. Borsdorf, H.; Rammler, A.; Schulze, D.; Boadu, K.O.; Feist, B.; Weiss, H., Rapid on-site determination of chlorobenzene in water samples using ion mobility spectrometry, Anal Chim. Acta 2001, 440(1), 63–70. 84. Khayamian, T.; Tabrizchi, M.; Taj, N., Direct determination of ultra-trace amounts of acetone by coronadischarge ion mobility spectrometry, Fresenius’ J. Anal Chem. 2001, 370(8), 1114–1116. 85. Marr, A.J.; Cairns, S.N.; Groves, D.M.; Langford, M.L., Development and preliminary evaluation of a radio-frequency discharge ionization source for use in ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2001, 4,126–128. 86. Baim, M.A.; Eatherton, R.L.; Hill, H.H., Jr., Ion mobility detector for gas chromatography with a direct photoionization source, Anal. Chem. 1983, 55, 1761–1766. 87. Leasure, C.S.; Fleischer, M.E.; Anderson, G.K.; Eiceman, G.A., Photoionization in air with ion mobility spectrometry using a hydrogen discharge lamp, Anal. Chem. 1986, 58, 2142–2147. 88. Sielemann, S.; Baumbach, J.I.; Schmidt, H., IMS with non radioactive ionization sources suitable to detect chemical warfare agent simulation substances, Int. J. Ion Mobility Spectrom. 2002, 5(3), 143–148. 89. Borsdorf, H.; Nazarov, E.G.; Eiceman, G.A., Atmospheric pressure chemical ionization studies of nonpolar isomeric hydrocarbons using ion mobility spectrometry and mass spectrometry with different ionization techniques, J. Am. Soc. Mass Spectrom. 2002, 13(9), 1078–1087. 90. Xie, Z.; Sielemann, S.; Schmidt, H.; Li, F.; Baumbach, J.I., Determination of acetone, 2-butanone, diethyl ketone and BTX using HSCC-UV-IMS, Anal. Bioanal. Chem. 2002, 372(5–6), 606–610. 91. Miller, R.A.; Nazarov, E.G.; Eiceman, G.A.; King, A.T., A MEMS radio-frequency ion mobility spectrometer for chemical vapor detection, Sensor. Actuat. A: Phys. 2001, A91, 301–312. 92. Sielemann, S.; Baumbach, J.I.; Schmidt, H.; Pilzecker, P, Detection of alcohols using UV-ion mobility spectrometers, Anal Chim. Acta 2001, 43, 293–301. 93. Sielemann, S.; Baumbach, J.I.; Schmidt, H.; Pilzecker, P, Quantitative-analysis of benzene, toluene, and mxylene with the use of a UV-ion mobility spectrometer, Field Anal Chem. Tech. 2000, 4, 157–169. 94. Driscoll, J.N., Evaluation of a new photoionization detector for organic compounds, J. Chromatogr. 1977, 134(1), 49–55. 95. See http://www.hnu.com/ and links. 96. Eiceman, G.A.; Vandiver, V.J.; Leasure, C.S.; Anderson, G.K.; Tiee, J.J.; Danen, W.C., Effects of laser beam parameters in laser-ion mobility spectrometry, Anal. Chem. 1986, 58(8), 1690–1695.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_178... 06/05/2007

Página 1 de 1

Page 179 97. Begley, P.; Corbin, R.; Foulger, B.E.; Simmonds, P.G., Photo-emissive ionization source for ion mobility detectors, J. Chromatogr. 1991, 588, 239–249. 98. Rasulev, U.K.; Khasanov, U.; Palitcin, V.V., Surface-ionization methods and devices of indication and identification of nitrogen-containing base molecules, J. Chromatogr. A 2000, 896, 3–18. 99. Rasulev, U.K.; Iskhakova, S.S.; Khasanov, U.; Mikhailin, A.V., Atmospheric pressure surface ionization indicator of narcotic, Int. J. Ion Mobility Spectrom. 2001, 4, 121–125. 100. Rasulev, U.Kh.; Nazarov, E.G.; Palitsin, V.V., Surface ionization gas-analysis devices with separation of ions by mobility, Fourth International Workshop on Ion Mobility Spectrometry, Cambridge, U.K., 1995. 101. Wu, C; Hill, H.H.; Rasulev, U.K.; Nazarov, E.G., Surface-ionization ion mobility spectrometry, Anal. Chem. 1999, 71, 273–278. 102. Srebalus, C.A.; Li, J.W.; Marshall, W.S.; Clemmer, D.E., Gas-phase separations of electrosprayed peptide libraries, Anal. Chem. 1999, 71, 3918–3927. 103. Henderson, S.C.; Valentine, S.J.; Counterman, A.E.; Clemmer, D.E., ESI/ion trap/ion mobility/time-offlight mass-spectrometry for rapid and sensitive analysis of biomolecular mixtures, Anal. Chem. 1999, 71, 291–301. 104. Hoaglund, C.S.; Valentine, S.J.; Sporleder, C.R.; Reilly, J.P.; Clemmer, D.E., 3-Dimensional ion mobility TOFMS analysis of electrosprayed biomolecules, Anal. Chem. 1998, 70, 2236–2242. 105. Hudgins, R.R.; Jarrold, M.F., Conformations of unsolvated glycine-based peptides, J. Phys. Chem. B 2000,104, 2154–2158. 106. Steiner, W.E.; Clowers, B.H.; Matz, L.M.; Siems, W.F.; Hill, H.H., Jr., Rapid screening of aqueous chemical warfare agent degradation products: ambient pressure ion mobility mass spectrometry, Anal. Chem. 2002, 74, 4343–4352. 107. Wu, C; Siems, W.F.; Asbury, G.R.; Hill, H.H., Electrospray-ionization high-resolution ion mobility spectrometry—mass-spectrometry, Anal. Chem. 1998, 70, 4929–4938. 108. Dion, H.M.; Ackerman, L.K.; Hill, H.H., Jr., Detection of inorganic ions from water by electrospray ionization-ion mobility spectrometry, Talanta 2002, 57, 1161–1171. 109. Gidden, J.; Bowers, M.T.; Jackson, A.T.; Scrivens, J.H., Gas-phase conformations of cationized poly (styrene) oligomers, J. Am. Soc. Mass Spectrom. 2002, 13, 499–505. 110. Lee, S.; Wyttenbach, T.; Bowers, M.T., Gas-phase structures of sodiated oligosaccharides by ion mobility ion chromatography methods, Int. J. Mass Spectrom. 1997, 167, 605–614. 111. Koomen, J.M.; Ruotolo, B.T.; Gillig, K.J.; Mclean, J.A.; Russell, D.H.; Kang, M.J.; Dunbar, K.R.; Fuhrer, K.; Gonin, M.; Schultz, J.A., Oligonucleotide analysis with MALDI-Ion-Mobility-TOFMS, Anal. Bioanal. Chem. 2002, 373, 612–617. 112. Woods, A.S.; Koomen, J.M.; Ruotolo, B.T.; Gillig, K.J.; Russel, D.H.; Fuhrer, K.; Gonin, M; Egan, T.F.; Schultz, J.A., A study of peptide-peptide using MALDI-Ion mobility o-TOF and ESI mass-spectrometry, J. Am. Soc. Mass Spectrom. 2002, 13, 166–169. 113. Bolton, H.C.; Grant, J.; McWilliam, I.G.; Nicholson, A.J.C.; Swingler, D.L., Ionization in flames. II. Mass-spectrometric and mobility analyses for the flame ionization detector, Proceedings of the Royal Society of London, Series A: Mathematical, Physical, and Engineering Sciences 1978, 360, 265–277. 114. Atar, E.; Cheskis, S.; Amirav, A., Pulsed flame—a novel concept for molecular detection, Anal. Chem. 1991, 63, 2061–2064.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_179... 06/05/2007

Página 1 de 1

Page 180 115. Lias, S.G.; Liebman, J.F.; Levin, R.D., Evaluated gas phase basicities and proton affinities of molecules; Heats of formation of protonated molecules, J. Phys. Chem. Ref. Data 1984, 13, 695–808. 116. Karasek, F.W., The plasma chromatograph, Res. & Dev. 1970, 21, 34–37. 117. Cohen, M.J.; Karasek, F.W., Plasma chromatography—a new dimension for gas chromatography and mass spectrometry, J. Chromatogr. Sci. 1970, 8, 330–337. 118. Bradbury, N.E.; Nielson, R.A., Absolute values of the electron mobility in hydrogen, Phys. Rev. 1936, 49, 388–393. 119. Baim, M.A.; Hill, H.H., Jr., Tunable selective detection for capillary gas chromatography by ion mobility spectrometry, Anal. Chem. 1982, 54, 38–43. 120. Eiceman, G.A.; Leasure, C.S.; Vandiver, V.J.; Rico, G., Flow characteristics in a segmented closed-tube design for ion mobility spectrometry, Anal. Chim. Acta 1985, 175, 135–145. 121. Asbury, G.R.; Hill, H.H., Using different drift cases to change separation factors-(alpha) in ion mobility spectrometry, Anal Chem. 2000, 72, 580–584. 122. Snyder, A.R; Harden, C.S.; Shoff, D.B.; Ewing, R.G.; Katzoff, L.; Bradshaw, R.; Turner, R.B.; Adams, J.N.; Taylor, S.J.; FitzGerald, J., Miniature ion mobility spectrometer monitor. Proceedings of the ERDEC Scientific Conference on Chemical and Biological Defense Research, Aberdeen Proving Ground, MD, November 15–18, 1994, 145–151. 123. Avida, A.; Friedman, M., The Design of an Ion Drift Tube with a Uniform Electric Field, NRCN(TN)099, 1986 (in Hebrew). 124. Karpas, Z.; Eiceman, G.A.; Ewing, R.G.; Algom, A.; Avida, R.; Friedman, M.; Matmor, A.; Shahal, O., Ion distribution profiles in the drift region of an ion mobility spectrometer, Int. J. Mass Spectrom. Ion Proc. 1993, 127, 95–104. 125. Soppart, O.; Baumbach, J.I., Comparison of electric fields within drift tubes for ion mobility spectrometry, Meas. Sci. Technol. 200,11, 1473–1479. 126. Eiceman, G.A.; Nazarov, E.G.; Stone, J.A.; Rodriguez, J.E., Analysis of a drift tube at ambient pressure: models and precise measurements in ion mobility spectrometry, Rev. Sci. Instrum. 2001, 72, 3610–3621. 127. Carrico, J.P.; Sickenberger, D.W.; Spangler, G.E.; Vora, K.N., Simple electrode design for ion mobility spectrometry, J. Phys. E: Sci. Instrum. 1983, 16, 1058–1062. 128. Spangler, G.E.; Vora, K.N.; Carrico, J.P., Miniature ion mobility spectrometry cell, J. Phys. E: Sci. Instrum. 1986, 19, 191–198. 129. Brokenshire, J.L., High resolution ion mobility spectrometry, 1991 Joint meeting FACSS/Pacific Conference and 27th Western Regional ACS Meeting, Anaheim, CA, October, 6–11, 1991. 130. Blanchard, W.C., Using non-linear fields in high pressure spectrometry, Int. J. Mass Spectrom. Ion Proc. 1989, 95, 199–210. 131. Cravath, A.M., The rate of formation of negative ions by electron attachment, Phys. Rev. 1929, 33, 605– 613. 132. van de Graaff, R.J., Mobility of ions in gases, Nature 1929, 124,10–11. 133. Tyndall, A.M., The Mobility of Positive Ions in Gases, Cambridge University Press, Cambridge, U.K., 1938. 134. Knorr, F.J.; Eatherton, R.L.; Siems, W.F.; Hill, H.H., Jr., Fourier transform ion mobility spectrometry, Anal Chem. 1985, 57, 402–406. 135. Tarver, E.E., External second gate, Fourier transform ion mobility spectrometry: parametric optimization for detection of weapons of mass destruction, Sensors 2004, 4(1–3), 1–13.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_180... 06/05/2007

Página 1 de 1

Page 181 136. Spangler, G.E., Space charge effects in ion mobility spectrometry, Anal. Chem. 1992, 64, 1312. 137. Spangler, G.E.; Collins, C.I., Peak shape analysis and plate theory for plasma chromatography, Anal Chem. 1975, 47(3), 403–407. 138. Eiceman, G.A.; Vandiver, V.J.; Chen, T.; Rico-Martinez, G., Electrical parameters in drift tubes for ion mobility spectrometry, Anal. Instrum. 1989, 18(3–4), 227–242. 139. Buryakov, I.A.; Krylov, E.V.; Makas, A.L.; Nazarov, E.G.; Pervukhin, V.V.; Rasulev, U.Kh., Separation of ions according to their mobility in a strong alternating current electric field, Pis’ma υ Zhurnal Tekhnicheskoi Fiziki 1991, 17(12), 60–65. 140. Buryakov, I.A.; Krylov, E.V.; Nazarov, E.G.; Rasulev, U.Kh., A new method of separation of multiatomic ions by mobility at atmospheric pressure using a high-frequency amplitude-asymmetric strong electric field, Int. J. Mass Spectrom. Ion Proc. 1993, 128, 143–148. 141. Carnahan, B.; Day, S.; Kouznetsov, V; Tarrasov, A., Development and applications of a traverse field compensation ion mobility spectrometer, Fourth International Workshop on Ion Mobility Spectrometry, A. Brittain, Ed., Cambridge, U.K., 1995. 142. Carnahan, B.; Day, S.; Kouznetsov, V.; Matyjaszczyk, M.; Tarassov, A., Field ion spectrometry. A new analytical technology for trace gas analysis, Adv. Instrum. Control 1996, 51, 87–96. 143. Purves, R.W.; Guevremont, R.; Day, S.; Pipich, C.W.; Matyjaszczyk, M.S., Mass spectrometric characterization of a high-field asymmetric waveform ion mobility spectrometer, Rev. Sci. Instrum. 1998, 69 (12), 4094–4105. 144. Guevremont, R.; Purves, R.W., Atmospheric pressure ion focusing in a high-field asymmetric waveform ion mobility spectrometer, Rev. Sci. Instrum. 1999, 70, 1370–1383. 145. Purves, R.W.; Guevremont, R., Electrospray ionization high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal Chem. 1999, 71, 2346–2357. 146. http://www.faims.com/ and http://www.ionalytics.com/en/index.shtml 147. Miller, R.A.; Eiceman, G.A.; Nazarov, E.G. King, A.T., A novel micro-machined high field asymmetric waveform ion mobility spectrometer, Sens. Actuat. B 2000, 67, 300–306. 148. Eiceman, G.A.; Nazarov, E.G.; Tadjikov, B.; Miller, R.A., Monitoring volatile organic compounds in ambient air inside and outside buildings with the use of a radio-frequency-based ion-mobility analyzer with a micromachined drift tube, Field Anal Chem. Technol 2000, 4(6), 297–308. 149. Eiceman, G.A.; Tadjikov, B.; Krylov, E.; Nazarov, E.G.; Miller, R.A.; Westbrook, J.; Funk, P, Miniature radio-frequency mobility analyzer as a gas chromatographic detector for oxygen-containing volatile organic compounds, pheromones and other insect attractants, J. Chromatogr. A 2001, 917, 205–217. 150. http://www.sionex.com/ 151. Eiceman, G.A.; Tarassov, A.; Funk, P; Miller, R.A.; Nazarov, E.G.; Hughes, E., GC-PFAIMS as smart smoke alarm. Identification of combustion sources by patterns of retention time and compensation voltage, Int. J. Ion Mobility Spectrom. 2002, 5, 71–75.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_181... 06/05/2007

Página 1 de 1

Page 182 152. Eiceman, G.A.; Nazarov, E.G.; Miller, R.A.; Krylov, E.V.; Zapata, A.M., Micro-machined planar field asymmetric ion mobility spectrometer as a gas chromatographic detector, Analyst 2002, 127(4), 466–471. 153. Krylov, E; Nazarov, E.G.; Miller, R.A.; Tadjikov, B.; Eiceman, G.A., Field dependence of mobilities for gas-phase-protonated monomers and proton-bound dimers of ketones by planar field asymmetric waveform ion mobility spectrometer (PFAIMS), J. Phys. Chem. A, 2002, 106, 5437–5444. 154. Eiceman, G.A.; Krylov, E.V.; Tadjikov, B.; Ewing, R.G.; Nazarov, E.G.; Miller, R.A., Differential mobility spectrometry of chlorocarbons with a micro-fabricated drift tube, Analyst 2004, 129, 297–304. 155. Krylova, N.; Krylov, E.; Eiceman, G.A.; Stone, J.A., Effect of moisture on the field dependence of mobility for gas-phase ions of organophosphorus compounds at atmospheric pressure with field asymmetric ion mobility spectrometry, J. Phys. Chem. A 2003, 107, 3648–3654. 156. Miller, R.A.; Nazarov, E.G.; Eiceman, G.A.; King, A.T., A MEMS radio-frequency ion mobility spectrometer for chemical vapor detection, Sens. Actuat. A-Phys. 2001, A91(3), 301–312. 157. Veasey, C.A.; Thomas, C.L.P., Fast quantitative characterisation of differential mobility responses, Analyst 2004, 129, 198–204. 158. Eiceman, G.A.; Nazarov, E.G.; Miller, R.A., A micro-machined ion mobility spectrometer-mass spectrometer, Int. J. Ion Mobility Spectrom. 2000, 3(1), 15–27. 159. http://www.environics.fi/ 160. Ebert, H., Aspirations apparat zur bestimmung des ionengehaltes der atmosphare, Phys. Z. 1901, 2, 662– 666. 161. Tammet, H.F., The Aspiration Method for the Determination of Atmospheric-Ion Spectra, available at http://ael.physic.ut.ee/KF.public/sci/publs/AM/default.htm 162. Misakian, M.; McKnight, R.H.; Fenimore, C, Calibration of Aspirator-Type Ion Counters and Measurement of Unipolar Charge Densities, National Bureau of Standards Technical Note 1223, Issued May 1986, 77 pp. 163. Knudsen, E.; Israelsson, S., Mobilities of small ions in the atmospheric surface layer, Pure Appl. Geophys. 1975, 113(4), 525–533. 164. Tuovinen, K.; Paakkanen, H.; Hanninen, O., Determination of soman and VX degradation products by an aspiration ion mobility spectrometry, Anal Chim. Acta 2001, 440(2), 151–159. 165. Utriainen, M.; Karpanoja, E.; Paakkanen, H., Combining miniaturized ion mobility spectrometer and metal oxide gas sensor for the fast detection of toxic chemical vapors, Sens. Actuat. B-Chem. 2003, B93(1– 3), 17–24. 166. Raatikainen, O.; Pursiainen, J.; Hyvonen, P; Von Wright, A.; Reinikainen, S.-P; Muje, P, Fish quality assessment with ion mobility based gas detector, Mededelingen—Faculteit Landbouwkundige en Toegepaste Biologische Wetenschappen (Universiteit Gent) 2001, 66(3b), 475–480. 167. Kotiaho, T.; Lauritsen, F.R.; Degn, H.; Paakkanen, H., Membrane inlet ion mobility spectrometry for online measurement of ethanol in beer and in yeast fermentation, Anal Chim. Acta 1995, 309, 317–325. 168. Sacristan, E.; Solis, A.A., A swept-field aspiration condenser as an ion-mobility spectrometer, IEEE Transactions on Instrumentation and Measurement 1998, 47(3), 769–775. 169. Cockbaine, D.R., Ion Mobility scanning ion chamber, I am seeking the reference. 170. Eiceman et al., in preparation.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_182... 06/05/2007

Página 1 de 1

Page 183 171. Stevenson, P.C.; Thomas, R.A.; Lane, S., Ion-mobility spectrometer for radio-chemical applications, Nucl Instrum. Methods 1970, 89, 177–187. 172. Baumbach, J.I.; Berger, D.; Leonhardt, J.W.; Klockow, D., Ion mobility sensor in environmental analytical chemistry—concept and first results, Int. J. Environ. Anal Chem. 1993, 52(1–4), 189–193. 173. See http://www.gas-dortmund.de/pdf_docs/uIMS-ODOR_0310_en.pdf for the commercial product and http://www.gas-dortmund.de/pdf_docs/Mini_Flyer_020312.pdf for a description of the drift tube. 174. http://www.smithsdetection.com/product.asp?product=LCD+%2D+Lightweight+Chemical+Detector&prodgroup-Continuous+Monitoring&prodcat=Chemical+Agent+Detection+% 26+TICs&prodarea=Trace+detection 175. Xu, J.; Whitten, W.B.; Ramsey, J. Space charge effects on resolution in a miniature ion mobility spectrometer, Anal Chem. 2000, 72(23), 5787–5791. 176. Wu, C; Steiner, W.E.; Tornatore, P.S.; Matz, L.M.; Siems, W.F.; Atkinson, D.A.; Hill, H.H., Jr., Construction and characterization of a high-flow, high-resolution ion mobility spectrometer for detection of explosives after personnel portal sampling, Talanta 2002, 57, 123–134. 177. Denson, S.; Denton, B.; Sperline, R.; Rodacy, P.; Gresham, C, Ion mobility spectrometry utilizing microfaraday finger array detector technology, Int. J. Ion Mobility Spectrom. 2002, 5(3), 100–103. 178. Prini, A.; Lawrence, A.H.; Laframboise, S., Compact digital signal averager for ion mobility spectrometry, J. Phys. E: Sci. Instrum. 1987, 20, 1422–1424. 179. Goubran, R.A.; Lawrence, A.H., Experimental signal analysis for ion mobility spectrometry, Int. J. Mass Spectrom. Ion Proc. 1991, 104, 163–178. 180. Davis, D.M.; Kroutil, R.T., Application of digital filters to process data for ion mobility spectrometry, Anal. Chim. Acta 1990, 232, 261–266. 181. Rauch, P.J.; Harrington, P.B.; Davis, D.M., Near real-time self-modeling mixture analysis, Chemometrics and Intelligent Laboratory Systems 1997, 39, 175–185. 182. Chen, G.X.; Harrington, P.B., Real-time two-dimensional wavelet compression and its application to real time modeling of ion mobility data, Anal. Chim. Acta 2003, 490, 59–69. 183. Boger, Z., Possible Roles of Neural Networks in Developing Expert Systems for the Nuclear Industry, International Atomic Energy Agency Technical Experts Meeting/workshop to Demonstrate and Reviews Expert System Proto-types, Springfield, U.K., 30/9–4/10/91. 184. Karpas, Z.; Pollevoy, Y.; Melloul, S., Determination of bromine in air by ion mobility spectrometry, Anal Chim. Acta 1991, 249, 503–507. 185. Boger, Z.; Karpas, Z., Use of neural networks for quantitative measurements in ion mobility spectrometry (IMS), J. Chem. Info. Comput. Sci. 1994, 34, 576–580. 186. Boger, Z.; Karpas, Z., Application of neural networks for interpretation of ion mobility and X-ray fluorescence spectra, Anal Chim. Acta 1994. 292, 243–251. 187. Bell, S.E.; Nazarov, E.G.; Wang, Y.F.; Eiceman, G.A., Classification of ion mobility spectra by chemical moiety using neural networks with whole spectra at various concentrations, Anal Chim. Acta 1999, 394, 121–133. 188. Bell, S.E.; Nazarov, E.; Wang, Y.F.; Rodriguez, J.E.; Eiceman, G.A., neural network recognition of chemical class information in mobility spectra obtained at high temperatures, Anal Chem. 2000, 72, 1192– 1198.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_183... 06/05/2007

Página 1 de 2

Page 184 189. Eiceman, G.A.; Nazarov, E.; Rodriguez, J.E., Chemical class information in ion mobility spectra at low and elevated temperatures, Anal. Chim. Acta, 2001, 433, 53–70. 190. http://www.corning.com/lightingmaterials/products/macor.html 191. http://www.aremco.com/machinable-fired-ceramics.html 192. http://jenseninert.com/content/techinf.htm, also see http://www.polymer-plastics.com/fluoro_tefb.shtml 193. Scott, R.B.; Brown, P., A compact inexpensive gas chromatograph/mass spectrometer silicone rubber membrane separator, J. Chem. Educ. 1977, 54, 40. 194. Gough, T.A.; Webb, K.S., Use of a molecular separator in the determination of trace constituents by combined gas chromatography and mass spectrometry, J. Chromatogr. 1972, 64, 201–210. 195. Flagan, R.C., History of electrical aerosol measurements. Aerosol Sci. Technol. 1998, 28(4), 301–380.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_184... 06/05/2007

Página 1 de 1

Page 185

5 Hyphenated Methods with Mobility Spectrometers 5.1 Introduction to Hyphenated Ion Mobility Spectrometry (IMS) Methods In previous chapters, mobility spectrometers have been described as analytical instruments that are used independent of other chemical equipment to measure the chemical composition of a sample. Those descriptions emphasize the attractions of IMS analyzers as portable, fast, convenient, rugged, and economical. However, IMS drift tubes may be combined with other chemical instruments, and such combinations result in synergistic benefits. When a mobility spectrometer is combined with a gas chromatograph, the chemical information obtained and the characteristics of the measurements are improved by more than the individual results put together. Mobility spectra can be obtained for column effluent throughout a chromatographic separation, providing, from the measurement, additional analytical information, which is somewhat orthogonal to the retention timescale. Thus, mobility spectra enrich a chromatographic separation through a second dimension of mobility. However, improvements in the quality and value of the mobility spectrometer response result from the prefractionation of a complex sample and, ideally, the delivery of components as individual constituents to the IMS analyzer. When prefractionation is done, the ion chemistry in the reaction region is simplified and demands on ion separations in the drift tube are reduced. In this way, the chromatograph enhances the analytical value of the mobility spectrometer. Thus, the combination of a chromatograph and a mobility spectrometer is mutually beneficial and provides a data set that is suitable for advanced data handling. A final attraction for the use of both a chromatograph and an IMS analyzer is that the interface is inexpensive and uncomplicated because the IMS drift tube is operated at ambient pressure. This alleviates the demands of interfacing vacuum and ambient pressure as found in gas chromatography/ mass spectrometry (GC/MS) or liquid chromatography/mass spectrometry (LC/MS) instruments.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_185... 06/05/2007

Página 1 de 1

Page 186 Although practical benefits result from operating an IMS drift tube at ambient pressure, a consequence of moving ions through gases at ambient pressure is that ion identities will be altered by the chemical composition of the supporting atmosphere. In general, the core ions in a drift tube are robust and are not altered by the gas environment; however, a mobility coefficient is determined for the complete set of ions in the clustered form, including loosely held neutrals. Ions form clusters with neutrals such as water, making ion identity dependent upon moisture and temperature. Also, the presence in the supporting atmosphere of reagent gases and impurities which subtly affect ion identities with respect to adducts or clustered neutrals. All these are manifest in the Ko value which can reflect small changes in ion composition; but reduced mobility coefficients, unlike ion mass, cannot be regarded as intrinsic to a chemical. Accordingly, the drift time or mobility coefficient may not be used for ion identifications in the same way as mass is used normally in mass spectrometry. Although this is inconsequential for most applications, ion identification in IMS can be made with certainty only when a mass spectrometer is used with the drift tube and ions from the drift tube are mass-identified. Alternatively, ion identities can be discussed when conditions inside the drift tube are known and can be referenced to prior studies where ions were mass-identified. Historically, the mass spectrometer has been thought to enhance the value of a mobility measurement with identifications of ions in the mobility spectrum. Indeed, mass analysis of ions provides a level of understanding of the response of the IMS that is not available otherwise. Mobility spectrometers are seen increasingly as beneficial when used as inlets for mass spectrometers. Mobility analyzers at ambient or elevated pressure can pre-filter an ion mixture from an electrospray ion source and provide a first level of selectivity for subsequent MS measurements. This concept for IMS/MS, where the mobility spectrometer reduces chemical noise in complex ion mixtures, may anticipate rapid clinical screening of biological or medical samples. Once again, a hyphenated technique benefits the practices and principles of both components of a measurement. The advantages of combining drift tubes with chromatographic methods and mass spectrometry were understood early in the development of modern analytical IMS, as is suggested by the title of a historic publication in IMS: “Plasma Chromatography™—A New Dimension for Gas Chromatography and Mass Spectrometry”1

5.2 GC/IMS 5.2.1 Background In GC, a sample is passed as a vapor through a column with a flow of gas, the carrier gas, and the components emerge or elute from the column at characteristic times as shown in Figure 5–1. The carrier gas is commonly an inert gas, usually nitrogen or helium, and is routinely provided to the column

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_186... 06/05/2007

Página 1 de 1

Page 187

FIGURE 5–1 (A) Diagram of the main components of a gas chromatograph that include the injection port through which the sample is introduced, the chromatographic column where components of a mixture are separated, and the detector that records the presence of these components, (B) A chromatogram that provides qualitative and quantitative information about a sample may be obtained from the analysis.

under pressure from 100 to 400 kPa. The sample is introduced into the inlet of the chromatograph, where liquid or solid samples are volatilized and swept by the carrier gas into and through the column. In a capillary column, the predominant type of column in use today, a liquid or stationary phase is cross-linked and chemically bonded on the inner wall of a fused-silica column. The constituents of the sample can be dissolved in the stationary phase through interactions between a component and stationary phase, so passage of the component through the column will be retarded. Components that exhibit little interaction or solubility with the stationary phase will be retarded slightly in passage through the column and will elute quickly. As the strength of interactions increases, the time for a component to elute from the column, or the retention time, also increases. All vapors emerge from the column with retention times that depend on the carrier gas flow rate, column

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_187... 06/05/2007

Página 1 de 1

Page 188 temperature, and stationary phase and molecular properties of the constituents, principally vapor pressure. Secondary influences on retention include the polarity or molecular structure of the compounds, and these provide refinements in separation that make GC a valued analytical technique. Vapors eluting from the column enter a detector where a signal is generated through a physical or chemical process, and this signal is amplified and displayed as a plot of detector response vs. retention time. This plot is called a chromatogram (Figure 5–1B), and provides the only direct information from a gas chromatographic analysis when simple detectors are employed. In the chromatogram, quantitative information is contained in peak heights or areas, and qualitative information is derived from the retention time. Consequently, identification of a peak is generally regarded as incomplete unless additional information is obtained on the eluting peak. Although a range of detectors has been developed that provide specificity for chemical substituents or certain atoms, an identification has been considered supportable only with a mass spectrum, or perhaps an infrared spectrum. Historically, combinations of gas chromatographs and mass spectrometers underwent rapid development in the 1970s and coincided with the growth of environmental and medical applications of GC and the technical advances of MS. Today, the use of GC/MS instruments has become routine although most GC/MS instruments are considered laboratory devices, due to the requirements of pure gas supply, high vacuum, and power. In the combination of a gas chromatograph with an ion mobility spectrometer, effluent from the column is introduced directly into the reaction region of the drift tube, as shown in Figure 5–2, in ways that are analogous to GC/ MS instruments. The first experimental results with a GC/IMS analyzer were reported in 1972 for the characterization of musk ambrette,2 and other early applications of GC/IMS from 1970 to 1980 are shown in Table 5–1.3–10 These references have historical interest only and the column technology of the time severely limited the analytical capabilities of GC/IMS. Columns were glass or metal tubes with 1 to 2 mm ID and contained a packing made from a solid support material such as diatomaceous earth coated with a thin film of stationary phase. Although the stationary phases were high-molecular-weight polymers, often with good thermal stability, low levels of impurities from polymer synthesis were released at elevated temperatures. In some instances, prolonged exposure of stationary phases to elevated temperatures caused decomposition of the polymer, which was only coated on the support. This could lead to loss of the stationary phase and changed column properties. Release of impurities or decomposition products of the polymer, even at very low levels, into the column effluent would alter the ion chemistry in the reaction region of the drift tube, affecting the response and reliability of the mobility spectrometer.8 Additionally, pack columns exhibited poor efficiency of separation and limited the value of a GC/IMS measurement. Finally, clearance of vapors from the reaction region was slow due to the designs of these early-generation analyzers, and this further eroded the utility of GC/IMS as

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_188... 06/05/2007

Página 1 de 1

Page 189

FIGURE 5–2 Block diagram of a gas chromatograph/ion mobility spectrometer. The effluent from the GC column flows through a heated transfer line and enters the reaction region of the IMS drift tube. Signals from the detector are amplified and processed using a computer where individual spectra can be stored for later analysis. TABLE 5–1 Examples of Publications in GC/IMS from 1970–1980

Year

Title

Reference

1972 Gas chromatograph/plasma chromatograph interface and its performance in the detection of musk ambrette

2

1973 Coupling of high speed plasma chromatography with gas chromatography

3

1973 Fast scan ion mobility spectra of diethyl, dipropyl, and dibutyl ethers as determined by the plasma chromatography

4

1974 The plasma chromatograph as a qualitative detector for gas chromatography

5

1974 Evaluation of the plasma chromatograph as a separator-identifier

6

1977 Gas chromatographic detection modes for the plasma chromatograph

7

1978 Studies of the effects of volatile components from gas chromatographic liquid phases on plasma chromatographic performance

8

1979 Characterization of isomeric compounds by gas and plasma chromatography

9

1980 Plasmagram spectra of some barbiturates

10

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_189... 06/05/2007

Página 1 de 1

Page 190 an analytical method; similar compromises in chromatography were also made with GC/MS measurements of that time. In the late 1970s, a revolution in chromatographic separations occurred, with high-efficiency capillary columns where the stationary phase was cross-linked and bonded to the surface of a fusedsilica tube. These columns became available commercially, and the bonded phases virtually eliminated contamination of the IMS analyzer from the stationary phase. These capillary columns were small in diameter, contained a 50-micron thick film of stationary phase, and were operated with carrier gas flows of only 1 to 5 ml/min. Bond-phase capillary columns became an essential development without which the chemistry of ionization could not be properly managed in an IMS, but this alone could not advance GC/IMS. The second significant and pivotal advance in modern GC/IMS was made by Baim and Hill in 1982 to solve the problem of residence time and band broadening in the drift tube.11 Unlike the pack columns where flows were 30 to 50 ml/min, capillary column flows are small and details on the interface between the column and the drift tube become important. Poor plumbing can lead to band broadening by nonchromatographic processes, and any extracolumn broadening is unacceptable for capillary column. They designed a drift tube so that column effluent could be introduced directly from a fused-silica column into the reaction region. The drift gas was introduced into the drift tube at the detector and was passed through the drift and reaction regions, forming what is known as unidirectional flow. In a unidirectional flow design, the volume of the reaction region is rapidly purged of excess or unused sample that is swept from the drift tube. Thus, effluent from the gas chromatograph has an established and brief residence time in the reaction region and is prevented from entering the drift region by virtue of the direction of the drift gas flow. With unidirectional flow, high-speed response is possible for fast-eluting GC peaks, and ion mobility spectrometers do not contribute appreciably to peak broadening compared to other GC detectors. However, there are several secondary and essential consequences of this design: clustering reactions between ions and sample neutrals in the drift region are prevented, and the time of residence for sample in the reaction region is repeatable and known. These make mobility spectra reproducible and provide a foundation for refined GC/IMS measurements. The connection of a gas chromatographic column to an ion mobility spectrometer might be thought of as uncomplicated and convenient due to the compatibility of pressures, and this has been largely correct. Two possible methods to connect a capillary column to a mobility spectrometer include the attachment of the column to the drift tube on axis to the ion source with a concentric inner tube (Figure 5–3A) or through the side of the drift tube (Figure 5–3B). The side design is effective, although complicated by the requirements of a gas-tight connection between the column and the drift tube. There are the practical inconveniences of adding insulation and a protective shell on the drift tube with a side tube, and orientating or locating the drift tube at the oven of the gas chromatograph. This is not a severe limitation, as shown in the photograph of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_190... 06/05/2007

Página 1 de 1

Page 191

FIGURE 5–3 Two designs for connecting a capillary column to a mobility spectrometer with unidirectional flow: (A) On-axis to the ion source with a concentric inner tube, (B) Through the side of the drift tube.

a research-grade drift tube (Figure 5–4), but it is also not particularly convenient for drift-tube assembly or repair. In the second design, the capillary column is extended on axis into the ion source, from the end of the drift tube. Care is taken to keep the capillary column in the center of the ion source, and the column is positioned on the upstream side of the source as shown in Figure 5– 3A. The advantages of the axial design include the convenience of interchanging columns, which can now be accomplished without working around the heating elements, insulation, and housing of the drift tube, and the convenience of locating the drift tube in places provided for detectors on commercial gas chromatographs. A difficulty with the axial design is that response is sensitive to the position of the capillary column in the ion source.12 Differences from 10-fold to 50-fold have been reported for displacements of the column end by ~10 mm, and complete loss in response can occur with a poor column position. This loss in response can be attributed to the streaming of effluent vapors (upper trace of Figure 5–5) along the outer walls of the capillary column.13 The streaming results in poor mixing between effluent and reagent ions and low efficiency of product ion formation. A modification of the axial design includes a sleeve with an

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_191... 06/05/2007

Página 1 de 1

Page 192

FIGURE 5–4 Photograph of a home-made GC/IMS with side connection of the GC column through the insulation and a protective shell of the IMS drift tube.

additional gas flow (sheath flow) to create turbulence and mixing of vapors, as shown in the lower trace of Figure 5–5.13 This design is the basis for the Volatile Organic Analyzer (VOA) described in Subsection 5.2.5.2. When this sleeve flow is used, the interface of the GC to the IMS is uncomplicated and less sensitive to the location of the capillary outlet, although the response is affected by the flow of gas in the sleeve. The flow of the sleeve gas was employed to extend the linear range of response for a GC/IMS with an automatic servo system referenced to the reactant-ion peak intensity.14 Two essential details must be considered when a fused-silica-or aluminum-clad capillary column is connected to a mobility spectrometer. The protective coating on fused-silica columns is a thermally stable polyimide, which, when placed in the reaction region of a drift tube, will exude trace levels of vapors from the polymer. These are released in quantities sufficient to alter the ionization chemistry of the IMS analyzer. Although the exact nature of the vapor has not been identified, the compound is an imide or amine-like molecule with a high proton affinity from the nitrogen atom in the compound. Thus, the direct insertion of a polyimide-coated fused-silica capillary column into a drift tube will lead to the de facto formation of an alternate reagent gas. Removal of the polyimide, which protects the capillary column surface, is difficult. One solution is to use a short section of an aluminum-clad column coupled with a small dead-volume fitting to the fused-silica column. However, this may lead to electrical arcs between the column and the drift rings in the drift tube; naturally, this is undesirable. The aluminum coating can be easily removed

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_192... 06/05/2007

Página 1 de 1

Page 193

FIGURE 5–5 Mobility spectra showing reactant ion peak (RIP) O2–(H2O)n at 4 ms and

a product ion peak at 3.5 ms (CI–(H2O)n from a chlorocarbon) with a capillary column inlet as shown in Figure 5–3A.

by stripping in a solution of sodium hydroxide while the inside of the column is protected with a wax plug in the column end. When the aluminum cladding is removed and the length of base-treated column washed, a small part of the column containing the plug can be broken and discarded. The column can now be inserted into the drift tube without danger of electrical arcs or chemical contamination. In the period from 1970 to 2000, the only configurations of GC/IMS were based on the traditional drift tube with ion shutters and voltage gradient.1–17 The development of drift tubes employing fielddependent mobilities in the past several years has led to interfaces between these analyzers and gas chromatographs.18–23 These instruments, whether a field ion spectrometer or a differential mobility spectrometer, are similar to a quadrupole mass spectrometer with ions continuously removed from the ion source and passed to the analyzer section. Also, a scan of ions is obtained by sweeping a DC potential, and instead of drift times, compensation voltage values are used. The importance of this development is that the cost and footprint of this

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_193... 06/05/2007

Página 1 de 1

Page 194 analyzer have motivated rapid commercialization of a GC with a differential mobility spectrometer.24 In the following discussions, traditional configurations will be treated for each topic first, and then the performance or features of field-dependent configurations will be described.

5.2.2 Analytical Information Although mobility spectrometers cannot match the current standards of resolution and spectral identification of mass spectrometers, comparisons between GC/IMS and GC/MS are inevitable, and the two may be seen as functionally analogous instruments. There are strong parallels in both the philosophy of use and the utilization of chemical information obtained from a measurement. Differences in size, cost, weight, power, simplicity, and maintenance requirements naturally favor the mobility spectrometer, whereas history, sophisticated commercial instruments, software, availability of spectral libraries, and resolution favor the mass spectrometer. In a GC/IMS measurement, a selection of methods is available to acquire data or to analyze and display data and these include continuous acquisition of mobility spectra, selected ion monitoring, and reconstructed ion chromatograms. In continuous acquisition mode, complete mobility spectra are obtained and stored throughout a chromatographic separation. When the analysis is completed, this data set can be used to extract mobility spectra for any peak or any location in the chromatogram. Also, the data matrix comprising drift time, retention time, and ion current or detector response can be processed using graphics software. Alternatively plots can be made of ion current at specific drift times or Ko values with retention time. These constitute a type of reconstructed ion chromatogram commonly used in GC/MS though differences between Ko and m/z make precise comparisons between such plots from GC/IMS and GC/MS impossible. In traditional drift-tube designs, singleion monitoring is possible when the drift tube is fitted with dual ion shutters, and the second shutter is operated at a constant delay with respect to the first shutter. Examples and discussions of each of these is given in the following subsections. 5.2.2.1 Continuous Scanning When a GC/IMS technique is employed for separation and identification of components, mobility spectra are generally collected throughout an entire chromatographic analysis. The total time for such analyses may span several 100 msec in high-speed GC to nearly 60 min for temperatureprogrammed analysis with 30-m capillary columns and complex mixtures. Regardless of the duration of the GC separation, mobility spectra are acquired in intervals of 0.1 to 2 sec, governed by drift-tube parameters and the choice of 5 to 25 individual IMS scans for digital signal averaging to a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_194... 06/05/2007

Página 1 de 1

Page 195 mobility spectrum. Ideally, all mobility spectra obtained during an analysis are stored and may be utilized after the analysis for identification of constituents. The data set will be a matrix in three dimensions, including retention time, drift time, and ion intensities at each drift time when the measurement is completed. The presentation or manipulation of this data set to extract information parallels the methods developed for GC/MS during the last three decades. The matrix of data from a GC/IMS analysis can be illustrated graphically in a topographic plot (Figure 5–6). In this graph, data are plotted using the three axes of chromatographic retention time (y-axis), drift time from IMS (x-axis), and IMS signal intensity (color coded). The merit of this representation is that many facets of the analysis can be inspected or observed in a single plot. A disadvantage is that spectral details may not be evident or may be obscured by the other patterns of data. A second method of analyzing data is with a reconstructed gas chromatogram that can be generated in a 2-D graph by plotting ion intensity vs. retention time as shown in Figure 5–7. Unlike mass spectrometry where all the ions in the spectrum are summed and used as signal intensity, the reactant-ion peak intensity in the mobility

FIGURE 5–6 Topographic plot of GC/IMS analysis of a mixture of chemicals where the IMS drift time is on the x-axis, the chromatographic retention time on the yaxis, and IMS signal intensity is color-coded.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_195... 06/05/2007

Página 1 de 1

Page 196

FIGURE 5–7 Presentations of GC/IMS data. (A) Plot of reactant-ion intensity from topographic plot in Figure 5–6. A decrease in the reactant-ion peak indicates the formation of product ions and the elution of an analyte from the GC column, (B) The mobility spectrum for the chemical eluting with a retention time of 10 min, (C) The mobility spectrum for the chemical eluting with a retention time of 15 min.

spectra can be plotted against retention time. In this approach, the presumption is that reactant-ion peak intensity is a reliable measure of product ions formed from constituents. Decreases in the reactant-ion peak indicate the appearance of components in the GC effluent from which product ions are formed, and plots are seen as the inverse of the reactant-ion peak (RIP) intensity. This approach becomes complicated only when peaks for the product ions or fragments of product ions overlap with the reactant-ion peak. In such

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_196... 06/05/2007

Página 1 de 1

Page 197 a case, the intensity of the RIP is not completely reliable as a measure of sample composition, and the sum of product ions intensity at drift times other than that for the RIP may be an alternative way to display results. When continuous scanning methods are used, a mobility spectrum can be extracted from the data set at any retention time and used to characterize a constituent or sample composition. This is shown Figure 5–7B and Figure 5–7C for two of the components in the sample, and these illustrate the value of obtaining mobility information that is orthogonal to chromatographic retention. These spectra can be compared with spectral libraries and referenced to retention times of authentic standards for the identification of the components by using the mobility spectra. In a gas chromatography/differential mobility spectrometry (GC/ DMS) measurement, scanning of the compensation voltage can be done continuously throughout a chromatographic separation.19–21 The data set will be analogous to that from GC/IMS except that the principle of mobility characterization will be unlike that in a traditional mobility characterization. However, an advantage with GC/DMS for analytical measurements is that positive ion and negative ion spectra may be obtained simultaneously. As above, 3-D or 2-D graphs of results may be made from the full set of data. 5.2.2.2 Reconstructed Ion Chromatograms and Ion Monitoring 5.2.2.2.1 Reconstructed Ion Chromatograms A complete set of data from a GC/IMS measurement may be processed, after the analysis in order to extract information about components with certain retention times and particular drift times, or mobility coefficients. Generally, retention times are known for a particular analyte, and a plot of signal intensity at a certain drift time can be used to determine rapidly the possible presence and amount of a chemical in a simple or complex chemical mixture. An advantage of reconstructed ion chromatograms is that intensities for ions at several drift times may be graphed, and these ion signals can be isolated from the entire data set (Figure 5–6). This is shown in Figure 5–8 where intensities for particular drift times are extracted and plotted against retention time. In Figure 5–8A only a single component of the mixture produces an analyte ion at the drift time, whereas in Figure 5–8B there are several compounds that produce analyte ions at comparable drift times. This method provides no enhanced selectivity in the actual measurement but does provide advantages in examining the data. In a GC/DMS experiment, data are obtained by scanning the compensation voltage (i.e., Δk) as described earlier in Subsection 4.4.2. Results are shown in Figure 5–9 for the separation of a halocarbon mixture and appear to resemble those from a GC/IMS measurement: the plot is a pseudo 3-D representation of three dimensions of data. Differential mobility spectra are available for each step in chromatographic time, and reconstructed ion chromatograms can be extracted by plotting ion intensity at a given

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Page_19...

06/05/2007

Página 1 de 1

Page 198

FIGURE 5–8 Selected ion plots from the data set in Figure 5–6 corresponding to analyte ions with two drift times: (A) A single compound gives rise to an ion with drift time of 9.0 msec, (B) several compounds give rise to ions at 7.0 msec.

compensation voltage against GC retention time as shown in Figure 5–10. However, the intrinsic differences in the characterization of ions by DMS make a GC/DMS experiment resemble the principles underlying familiar GC/MS measurements with a quadrupole mass analyzer. The sampling of ions in DMS is continuous with complete transfer, excepting ion losses, of ions from the source region to the drift tube. Thus, the duty cycle (time of ion sampling or total time of analysis) for DMS is 100% rather than the 1% found in the conventional IMS drift tube with a pulsed ion shutter. An immediate consequence for data collection and handling is that specificity and selectivity in ion monitoring can be controlled with dwell times on one or several ions and by adjustment of the width of the compensation voltage window. These possibilities were suggested in calculations but have not been confirmed experimentally. 5.2.2.2.2 Ion Monitoring Before advanced data processing became generally available with inexpensive computers and signal processing interface cards, the large amounts of data described earlier for continuous scanning could not be

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_198... 06/05/2007

Página 1 de 1

Page 199

FIGURE 5–9 Topographic plot from analysis of mixture of halocarbons by GC/DMS: (A) Topographic plots of the compensation voltages for negative ions, (B) Topographic plots of the compensation voltages for positive ions, (C) Chromatogram of the GC retention time only.

acquired with the speed and capacity needed for later graphic analysis. Thus, reconstructed chromatograms were not possible with ordinary IMS analyzers of that time. Instead, ion-specific information with chromatographic retention was obtained using a second ion shutter placed in the drift tube near the detector. This second ion shutter could be opened at a time delayed with respect to the first ion shutter by a predetermined amount and controlled by a boxcar integrator or similar device. Ions were passed to the detector only during the time interval when the second ion shutter was opened, providing a type of notch filter for ions at a certain drift time. Unlike postanalysis processing, with complete data sets from scanning GC/IMS or GC/DMS, the dual-shutter approach excludes all other ions in the mobility spectrum, and these are removed by collisions and neutralizations on the wires when the second ion shutter is closed to ion passage. All chemical information in the measurement except that for a single drift-time window is lost, making the measurement highly directed and somewhat wasteful. This loss of chemical information is compensated by user control of signal-to-noise ratio through control of the width of the second ion shutter. However, increases in pulse width of the second or the first ion shutter would lead to decreases in specificity. Another advantage of this method is that requirements for data handling are minimal, and only the information sought is collected and recorded.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_199... 06/05/2007

Página 1 de 1

Page 200

FIGURE 5–10 Plots from a GC/DMS analysis of a halocarbon mixture (Figure 5–9) showing traces for positive ions at two compensation voltages (A and B) and for the positive reactant ion (C).

Finally, foreknowledge of the sample and the spectral properties of the analyte must be available with high accuracy. These and other limitations noted earlier have discouraged the use of ion monitoring as an analytical technique, although the utility and advantages are clear for some yet undeveloped applications. This approach to GC/IMS measurements was first demonstrated by Karasek and Kim5 and Karasek et al.,7 and expanded by Hill and coworkers.11,12,16,17 As suggested in the preceding text, any ion can be monitored, and this includes the reactant ion in the case when a completely general measurement is desired. An example of ion monitoring is shown in Figure 5–11 for the reactant ion and for a single ion. The extension of this concept to the monitoring of several ions has not been considered realistic for traditional drift tubes due

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_200... 06/05/2007

Página 1 de 1

Page 201

FIGURE 5–11 Schematics for a GC/IMS measurement where a second ion shutter is used to pass ions at a certain drift time to the detector: (A) Diagram of the dual-shutter drift tube, (B) Diagram of pulsing sequence of both shutters, (C) Mobility spectrum obtained by changing the delay between the two shutters.

to complications in timing of ion injection, ion drift, and changes in the second ion shutter. Jumps between various delays in drift time while ions are repeatedly injected into the drift tube do create difficulties in synchronization of events. However, this does not limit DMS or field-dependent analyzers, and selected ion monitoring as practiced with quadrupole mass spectrometers may be anticipated as the next development in DMS data handling.25

5.2.3 Features of Response Most of the considerations regarding the response in a GC/IMS measurement, such as the selection of a reagent gas chemistry, are identical to those that guide choices in the use of mobility spectrometers alone. The best

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_201... 06/05/2007

Página 1 de 1

Page 202 selectivity of response is possible when the ionization properties of the reagent gas are near but slightly below those for the analyte and the ionization properties for all other constituents are above that of the reagent gas. If a response that is as comprehensive as possible is desired, a charge exchange reagent with very dry supporting atmosphere26 or hydrated protons should be used. 5.2.3.1 Control of Selectivity by Reagent Gas or Method of Ionization Although the GC preseparation may be considered all the selectivity needed for a GC/IMS measurement, unresolved peaks and complex patterns may be obtained still for some environmental or biological samples. A GC/IMS measurement with such samples may be improved through the use of reagent gases with elevated proton affinities in order to remove response from chemicals with low proton affinities.27 This parallels in philosophy and practice the methods used in GC/MS with chemical ionization methods using methane, methanol, and ammonia. In the determination by GC/IMS of organophosphorous compounds in a mixture of volatile organic compounds, reagent gases of acetone and dimethylsulfoxide (DMSO) were employed to simplify the product ion chromatograms. Response with hydrated protons gave a complex and partially unresolved chromatographic profile with little possibility of recognizing the organophosphorous compounds. The retention times obtained for individual compounds with hydrate proton reactant ions are listed in the column entitled “Water” in Table 5–2. When acetone was introduced as a reagent gas at ~1 ppm in the reaction region, some chemicals were no longer ionized, as seen in the list of compounds detected in Table 5–2 in the column entitled “Acetone.” Compounds not detected with acetone as the reagent gas are marked as empty cells. Acetone provided enhanced selectivity for the complete GC/IMS measurement through the first stage of detection, i.e., ionization of the chemicals through ion-molecule reactions. This results in a simplified chromatogram compared with ionization chemistry using hydrated protons. When DMSO was used as the reagent gas, response was limited to amines, organophosphorous compounds, and a few chemicals from the other families. Although not completely selective to organophosphorous compounds, DMSO provided enough selectivity through ionization to simplify the chromatogram so that individual organophorphorous compounds could be recognized at anticipated retention times (Table 5–2). Mobility spectra for the organophosphorous compounds with DMSO reagent gas chemistry were distinctive and characteristic of each chemical as shown in Figure 5–12.27 In cases when the proton affinities of the analytes of interest are near or lower than those of interfering components, another method for ionization may be useful or necessary to add selectivity. The most widely used of these is a photo discharge lamp described first for GC/IMS in 1983.28 Ionization by photo discharge lamps is governed by rules that are based upon ionization energy, rather than proton transfer. A particular advantage of a photo ionization

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_202... 06/05/2007

Página 1 de 1

Page 203

FIGURE 5–12 Chromatogram (right) and mobility spectra (left) from GC/IMS analysis of a mixture of organophosphorus compounds (see Table 5–2 for peak identifications). Mobility spectra corresponding to peaks in the chromatogram. Reagent gas is DMSO.

device (PID) ion source is the unrestricted transportation or movement of such instruments, in contrast with regulatory restrictions for relocation and deployment of instruments with radioactive sources. Another advantage of a PID source is the possibility of selecting lamps of a particular energy level from four energy levels (11.1,10.6, 9.5, and 8.3 eV) that are conveniently available from commercial sources. Studies with photo discharge lamps have employed both traditional drift tubes28 and differential mobility spectrometers.29 The group of Baumbach et al. in Germany are strong proponents of PID as sources for GC/IMS instruments and have also employed multicapillary columns (see Subsection 5.2.4.2) for high-speed GC separations.30–35 This team has described the advantage of separations in reducing competitive charge-exchange reactions and has also extensively developed the use of graphic representation of the results and data analysis. Apart from the different rules of selectivity available with a photo discharge lamp, the mobility spectrum does not contain a reactant-ion peak, making spectral profiles simpler to analyze when product ions have mobility coefficients near that of the reactant ion, as found in an 63Ni-based drift tube. In some cases, a reagent gas with a large cross section for photons, such as aromatic compounds, is used in order

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_203... 06/05/2007

Página 1 de 2

Page 204 TABLE 5–2 Retention Times for Chemicals Detected by GC/IMS with Three Reagent Gases

Retention Time (min) Compound

Water

Acetone

DMSO

Alcohols Ethanol

3.27

–

–

n-Propanol

3.76

–

–

n-Butanol

5.02

5.02

5.01

n-Pentanol

7.60

–

–

n-Hexanol

11.32

11.31

11.32

n-Heptanol

15.66

15.67

–

n-Octanol

19.28

–

–

n-Nonanol

21.19

21.20

21.21

n-Decanol

22.68

22.61

21.62

Ketones Acetone

3.39

–

–

2-Butanone

4.05

–

–

2-Penatnone

5.48

5.48

5.47

2-Hexanone

8.30

–

–

2-Heptanone

12.16

12.16

–

2-Octanone

16.09

–

–

2-Nonanone

19.40

19.40

–

2-Decanone

23.38

23.40

–

Aldehydes Propanol

4.61

–

–

Butanal

5.48

5.48

5.47

Pentanal

8.71

8.73

–

Hexanal

11.78

11.79

11.78

Heptanal

12.44

12.45

–

Octanal

15.92

15.92

–

Nonanal

16.68

–

–

Decanal

20.97

20.98

–

Ethylacetate

4.28

–

–

Methylpropanoate

4.61

–

–

Propylacetate

6.06

–

–

Methyl-t-valerate

6.33

–

–

Methylisovalerate

7.90

–

–

Ethylbutanoate

8.71

8.73

–

Methylpentanoate

9.57

–

–

12.16

12.16

–

Esters

Propylbutanoate

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_204... 06/05/2007

Página 2 de 2

Methylhexanote

12.44

12.45

–

Methylisocaproate

13.49

Ethylhexanoate

16.40

16.41

–

Methylheptanoate

17.35

17.29

–

Ethyloctanoate

23.38

23.40

–

–

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_204... 06/05/2007

Página 1 de 1

Page 205 Retention Time (min) Compound

Water

Acetone

DMSO

Amines Diethylamine

4.48

–

–

Isobutylamine

5.02

5.02

5.01

Diisopropylamine

5.48

5.48

5.47

n-Amylamine

11.32

11.31

11.32

Diisobutylamine

13.69

13.70

13.80

Benzylamine

16.87

16.88

16.88

DMHP

8.78

8.84

–

DMMPa

11.78

11.79

11.78

TMPa

13.69

13.70

13.80

DEMpa

16.87

16.88

16.88

DIMP

18.99

19.00

19.01

DEEP

19.85

19.86

19.86

TEPa

20.97

20.98

20.99

DElpa

21.19

21.20

21.21

OSDEMpa

22.60

22.61

22.62

TIPPa

23.38

23.40

23.40

DEMSMP

27.63

27.63

27.64

DMMPa

28.19

28.19

28.20

TPP

29.25

29.25

29.27

DEESMP

29.82

29.82

29.82

DPMP

32.70

32.68

32.67

TIBP

33.00

33.00

33.00

DBBP

35.70

35.68

35.67

TSBP

36.71

36.67

36.64

TBP

36.71

36.67

36.64

Organophosphorous Compounds

a OPC with a coeluting VOC possible.

to efficiently produce reactant ions and thus enhance the sensitivity of the PID source. 5.2.3.2 Quantitative Response The quantitative characteristics of response with IMS analyzers were described in Chapter 3, and these should be largely unaffected by the use of a gas chromatograph as an inlet for the drift tube. However, the delivery of each particular analyte from the sample in a narrow band of 10 to 20 sec width in a Gaussian profile will change the units for reporting quantitative

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_205... 06/05/2007

Página 1 de 1

Page 206 results. Quantitative units for a GC or GC/IMS measurement are typically reported in amounts of analyte (such as picograms, nanograms, or femtomol) or amount per unit of time (such as nanograms per second or picomols per second) rather than the units concentration (such as parts per billion or milligrams per cubic meters) found in continuous sampling with IMS or for a complete measurement including sample collection and preconcentration. This difference in units hinders direct comparisons of the quantitative performance of a drift tube although the attractions of low detection limits in IMS analyzers are retained with GC/IMS instruments. The entire measurement will reflect the contributions of each component, such as inlet, column, transfer lines, and the drift tube, in both losses of mass and addition of variance to overall peak shape. These can be minimized in a welldesigned instrument when attention is given to cold spots and active sites for adsorption on surfaces of the interface between the chromatographic column and the drift tube. In this regard, the same principles that apply in all gas chromatographs apply also to GC/IMS instruments. The minimum detectable levels (MDL) for capillary GC/IMS were reported as <1 fmol/sec for tributylamine, phenyl sulfide, and tributyl phosphate, making GC/IMS at least 100 times better in detection limits compared with the well-known flame ionization detector (FID).36 Absolute MDL values for the determination of ketones by GC/DMS were about 0.05 to 0.1 ng, which was equal to or slightly better than a gas chromatograph equipped with a flame ionization detector but worse than in GC/IMS studies, likely because the ion source was only 1 mCi vs. 10 mCi for the large drift tube from GC/IMS results.20 Detection limits for GC/DMS were also determined for oxygen-containing compounds as 16 to 550 pg, with an average detection limit for all compounds of 94 pg.19 Vinyl chloride in air was determined by GC/IMS at 2 ppb by volume (ppbv) level.37 When a 1 1 sample was preconcentrated on an adsorbent trap, the MDL was reported as 0.02 ppbv, illustrating the difference between an MDL for an instrument vs. an MDL for a complete method. A gas chromatograph coupled to a field ion spectrometer, which was developed by Mine Safety Appliance Company from a Novosibirsk design and is no longer available, was reported to have MDLs from a few nanograms to >100 mg/m3 for actual chemical warfare agents of GB, VX, and HD.18 Recently, the MDLs for a GC/ IMS method were determined for methyl tertbutyl ether (MTBE) as 2 μg/l in nitrogen with a photo discharge lamp as the ion source and 30 pg/l for 63Ni ion source.33 The corresponding values for water samples were 20 mg/l and 1μg/l, respectively. Reproducibility was deemed good with relative standard deviations (RSD) of between 2.9 and 9%. Similar methods gave MDLs of 2.7 μg/l for acetone and 2-butanone and 3.0 μg/l for diethyl ketone in nitrogen.34 The assay linear dynamic range was from 4 to 320 μg/l. Concerns about the reproducibility of a GC/IMS measurement have not been broadly documented, apart from the VOA, for which a limited but impressive set of data on quantitative performance is available.38 Repeated

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_206... 06/05/2007

Página 1 de 1

Page 207 determinations of volatile organic compounds in the air of an enclosed atmosphere showed that the precisions of retention times ranged from 0 to 0.7% RSD. Concentrations derived from calibration curves exhibited precisions, including sample collection, of from 1 to 10% RSD except for two outliers. These numbers were obtained from an instrument operating automatically without expert attention. Consequently, the numbers represent values that might be considered realistic for this technology without optimization or recalibration. Recently, a calibration curve was prepared from a model of a chromatographic profile and from on-column mass.29 As shown in Figure 5–13, individual points along an elution profile could be transferred to a calibration curve, providing quantitative information comparatively rapidly and conveniently. This method is dependent upon a Gaussian peak shape, which is not difficult to obtain with a high degree of accuracy with modern capillary columns. Finally, the conditions of ion source or reaction region will affect both MDLs and response curves or sensitivity (Δresponse/Δconcentration). Reagent gases with high proton affinities will suppress sensitivity even to compounds with high proton affinities. In addition, moisture and temperature will affect sensitivity, as seen with IMS response to alkanes from methane to butane.

FIGURE 5–13 Approach of creating a calibration curve from the elution profile of a chromatographic peak in a GC/IMS or GC/DMS experiment.29

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_207... 06/05/2007

Página 1 de 1

Page 208 Although an ultradry atmosphere permitted mobility spectra with high spectral quality to be obtained for small hydrocarbons, the quantitative response was deemed poor. However, at moistures of 1 ppm and larger, virtually no response is obtained for hydrocarbons. Also, high temperature of the drift tube can lead to fragmentation of ions, when the charge is distributed among several ions. This leads to a loss in MDL values, which was explored early in the development of refined GC/IMS and found not to be excessive compared with other sensitive detectors for GC.39 In summary, the quantitative behavior of a GC/IMS instrument is distinguished by low MDLs and high sensitivities. These can be affected by the purity of the supporting atmosphere in the drift tube and by differences in free energies of proton transfers from reactant ion to analyte molecule. A benefit of pre-separation is the added reliability to a quantitative determination through lessened demands on the chemistry of ionization. This occurs as the effects of competitive ionization, which may occur when mixtures are introduced directly into an IMS analyzer, are eliminated or minimized.

5.2.4 Column Considerations Choices for the column in GC/IMS are fairly limited due to the predominance of high-resolution capillary columns with bonded phase, the maturity of development of these columns, and the selection of columns that are commercially available. Only eight to ten stationary phases are currently available with capillary columns and exhibit a range of polarities from non-polar to polar. All these are bonded and cross-linked, making their physical properties compatible with IMS analyzers as detectors. Hence, choices can be made regarding the chromatographic properties of the sample. Other requirements of time and resolution further guide selection. 5.2.4.1 High-Resolution Capillary Columns Contemporary capillary columns with a cross-linked bonded phase on fused silica are the preferred choice for laboratory-based GC/IMS, and advantages of such columns include easy availability, high efficiency, reliability, and nearly undetectable column bleed even at elevated temperatures. The disadvantages of such columns are the comparatively long times for a measurement (20 to 60 min), and the mechanical requirements of size for 30-m-long columns. Cost could be a consideration with most column prices at $200 or $400 for 15- and 30-m long columns, respectively. Historically, with IMS methods, the delay in time has been paramount and motivated the use of high-speed columns. High speed in GC separations can be obtained either with increased flow of carrier gas or with decreased lengths of column, as done with some instruments such as the Environmental Vapor Monitor (EVM; see Subsection 5.2.5.1). The most serious disadvantage of the short length of a conventional narrow-bore bonded-phase capillary column is that ~50 ng of chemicals will

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_208... 06/05/2007

Página 1 de 1

Page 209 cause the measurements to occur on a nonlinear portion of the partition isotherm for a 0.1-mm ID capillary column. The isotherm is a plot of CL vs. CG with the partition coefficient (K) equal to CL/CG, where CL is the analyte concentration in the liquid phase and CG the analyte concentration in the gas phase. The immediate consequence of this is that retention and peak shape for the separation are dependent upon the mass introduced into the column; this makes the time of retention unreliable. One solution to this is a bundle of identical capillary columns of short lengths, preserving high speed of analysis while providing a high capacity so that retention time is independent of sample concentrations. Although this possibility had been considered by chromatographers for decades, practical or working versions of such a column became known in the mid-1990s as a multicapillary column (MCC).40 5.2.4.2 Multi-Capillary Columns (MCC) The technology of MCCs was developed in research centers of the former USSR and is a remarkable technical accomplishment. As many as 900 capillary columns with a diameter of 40 μm are bundled into a single column with a diameter of ~2 mm. The columns are available as straight or coiled versions and have been used with GC/IMS by Baumbach et al.30–35 and by Buryakov et al. with field-dependent mobility analyzers.22,23,41 The main use of MCCs in Baumbach’s team has been for the separation of benzene, substituted benzenes, and other low-molecular-weight volatile organic compounds. This team has employed drift tubes that in recent years have been miniaturized versions of the traditional field-gradient design (Chapter 4). In contrast, Buryakov has employed MCCs for the separation and determination of drugs and explosives, such as 2,4-dinitrotoluene (DNT), 2,4,6trinitrotoluene (TNT), and pentaerythritol tetranitrate (PETN), and heroin, cocaine hydrochloride, and crack. The mobility analyzer used by this team was a field-dependent mobility design with the ion source at 190°C. The attractions of MCCs include high speed, high capacity, flow rates of over 50 ml/min and convenient dimensions. A disadvantage is that a conventional capillary column of 0.53 mm diameter (large-bore column) has comparable sample capacity as an MCC, negating any advantage of MCCs with respect to sample loading. Perhaps the most difficult facet of MCCs today is cost and poor commercial availability.42

5.2.5 Instruments In the first version of this book in 1994,43 the commercial availability of instruments was identified as a barrier to the development of IMS. There is today still no supplier of instruments or components for the assembly of IMS analyzers that meet specific needs or interests of a customer. This is also true for GC/IMS instruments. However, a number of GC/IMS instruments have been described or have been produced for targeted applications and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_209... 06/05/2007

Página 1 de 1

Page 210 may be regarded as models or successful demonstrations of GC/IMS analyzers. Such examples may be useful guides to those wishing to build and use GC/IMS instruments and some prominent examples of these instruments are described in the following subsections. The list is not comprehensive and emphasizes commercial or prototype instruments and excludes instruments from research laboratories. 5.2.5.1 Environmental Vapor Monitor (EVM) EVM was a handheld GC/IMS instrument manufactured in the early 1990s through the combination of the successful military analyzer, the Chemical Agent Monitor (CAM) made by Graseby Dynamics, Ltd. (Watford Herts, U.K.), and a capillary column with an all-glass inlet developed for valveless sampling vapors at ambient pressure (Figure 5–14).44–47 The gas pumps used originally in the CAM lowered the pressure (0.5 atm) in the drift tube. The pressure drop across the column allowed air to be pulled as the carrier gas passed through the column. Purified air was provided continuously at 8 ml/min to the inlet of the capillary column (see schematic in Figure 5–14) with an air purification system. A sample was taken for analysis only when this clean gas flow was throttled so that the capillary column made contact with sample gas entering the inlet at 60 ml/min. After a predetermined sampling period, clean air flow was re-established to the capillary column, completing an injection of sample. In this design, the sample encounters only silica surfaces removing or reducing the possibilities for contamination or loss of sample. The GC column was operated isothermally at ~70°C Mobility spectra were collected repeatedly at 0.1 sec intervals during a chromatographic analysis of about 8 sec. The efficiency of the column in this example is not high, with 2000 to 5000 theoretical plates; yet it was sufficient to deliver the pesticides individually to the ion source, as shown in Figure 5–15. The attraction of this GC/ IMS is handheld portability and speed of response, whereas disadvantages include inflexibility in sampling and limited chromatographic performance. The EVM was produced in limited numbers in the early 1990s and introduced into the commercial environmental monitoring markets simultaneously with the decreasing demand for on-site analytical instruments. The company created to market the EVM was dissolved in the mid-1990s, and the EVM is no longer available. The significance of the work was that a GC/ IMS analyzer for the determination of volatile organic chemicals in ambient air could be made handheld, rugged, and reliable. 5.2.5.2 Volatile Organic Analyzer (VOA) The development of a high-speed GC/IMS, called the VOA, was part of an initiative to screen the ambient air on board the U.S. space station for volatile organic compounds.38,48,49 This instrument was developed with the objectives of low power, low weight, medium-resolution GC separations, and air sampling with preenrichment of vapors. A block schematic of the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_210... 06/05/2007

Página 1 de 1

Page 211

FIGURE 5–14 Photograph (top) and schematic diagram (bottom) of the Environmental Vapor Monitor (EVM) made by Graseby, U.K. A watershed mobility spectrum is shown next to the photograph.

VOA and a couple of photographs of the flight version of this instrument are shown in Figure 5–16. The unit consists of two parallel GC/IMS instruments with two stationary phases, and operation of the drift tube in positive and negative polarities as shown in the schematic of Figure 5–16. This provides a broad assay consisting of two measurements on each column and in each polarity for the IMS analyzer. The VOA was designed for use during contingencies when unexpected and potentially hazardous events occur on-board the space station and for routine screening of air before the crew enters an unoccupied station. The first operational use of the VOA occurred in February 2002 after high levels of contaminants from a metal

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_211... 06/05/2007

Página 1 de 1

Page 212

FIGURE 5–15 Topographic plot from GC/IMS determination of a mixture of several pesticides using the handheld Environmental Vapor Monitor (EVM). The reactant ion peak is seen at 3.2 ms. Product ions from pesticides are evident from 3.5 to 7.5 ms.

oxide canister used in extravehicular activities were released into the air.49 The instrument provided on-site and rapid analysis of the air, demonstrating the value of onboard instrumentation for the protection of crew health and safety. Data from screening a synthetic mixture of volatile organic compounds are shown in Figure 5–17, where the column efficiency was 80,000 theoretical plates. In the VOA, the drift tubes were heated, and the unit was engineered at the standard for flight hardware. Only a few units were built, and neither the VOAs nor derivatives of VOAs are commercially available. 5.2.5.3 GC-IONSCAN The highly successful IONSCAN instrument for detecting explosives and drugs was modified with a GC inlet in the early 2000s while retaining most of the features that made it a popular instrument.50– 53 Features retained in this modification included the heated anvil for analysis of filter paper or swabs, all of the performance criteria of the drift tube, high sensitivity, and ease of use. The analysis with the GC-IONSCAN was fast and did not appreciably increase the size of the analyzer because the gas chromatograph was fitted onto the front panel of the instrument with little increase in footprint of the instrument. A particular feature of this instrument (Figure 5–18) is the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_212... 06/05/2007

Página 1 de 1

Page 213

FIGURE 5–16 Two photographic views (top) of the volatile organic analyzer (VOA) in flight on board the space station and a schematic diagram of the instrument (bottom).

integration of the GC column with the drift tube; both must be operated at elevated temperatures without cold spots in the transfer lines. The modified IONSCAN has been used in screening cargo containers in combination with highvolume air sampling of the atmosphere inside containers.51 Samples were screened for cocaine and ecgonidine methyl ester, and the addition of a gas chromatograph improved analytical performance but was unable to correct for complications due to chemicals sensitive to temperature or surfacebased reactions. This was observed with the desorption of RDX and HMX explosives with a heated anvil desorber used in another IONSCAN technology.51 Nonetheless, chromatographic separation

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_213... 06/05/2007

Página 1 de 1

Page 214

FIGURE 5–17 Two gas chromatograms obtained from screening synthetic mixtures of volatile organic compounds with the volatile organic analyzer (VOA).

provided valuable details about sample composition in ways not possible with an IMS analyzer alone. The significance of the GC-IONSCAN was that a heated IMS drift tube that is commercially available was modified to accept comparatively low vapor-pressure compounds and enhance the features that made the IONSCAN a valuable analyzer. This may well be the first GC/IMS that can be used by nonspecialists and might foreshadow the development of GC/IMS as a commercially available technology50 The current limitations may be seen as affordability and software. 5.2.5.4 Varian Micro Differential Mobility Detector In the spring of 2004, a milestone in the history of hyphenated mobility analyzers was reached with the formal launch of the first affordable GC instrument with a detector based on principles of mobility, in this instance, differential mobility.54 The gas chromatograph, a Varian model CP-4900, was equipped with an option for a microfabricated differential mobility

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_214... 06/05/2007

Página 1 de 1

Page 215

FIGURE 5–18 Photograph of the GC-IONSCAN (a) made by Barringer Research (now Smiths Detection) and schematic of the IMS drift tube (b) with the GC transfer line inlet on the left.

detector (μ-DMD), and the intended application was the determination of odors in natural gas.55 The differential mobility detector was used to determine a series of mercaptans in pure air and in a mixture of methane in air. Negative-ion differential mobility spectra were distinctive for mercaptans and provided the determination of odorants in high levels of hydrocarbons. However, the utility of the GCμDMD has been rapidly extended to

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_215... 06/05/2007

Página 1 de 1

Page 216

FIGURE 5–19 The microfabricated differential mobility detector (μ-DMD) with a CP4900 gas chromatograph, made by Varian Corporation. Left panel: View of instrument with computer and chromatographic separation (top) and of the DMD drift tube; Right panel: Side view of the inside of the instrument showing modular components.

chlorofluorocarbons and volatile halogenated anesthetics that are known to have favorable response in traditional IMS drift tubes.56 The introduction of a mobility-based analyzer by a large established instrument manufacturer such as Varian may mark the beginning of the general availability of IMS analyzers in GC/IMS combinations. The current model gas chromatograph is still somewhat targeted in applications and is provided without the full features of a temperature-programmed oven. A photograph of this unit is shown in Figure 5–19. 5.2.5.5 U.S. Army Pyrolysis-GC/IMS A GC/IMS instrument that merits discussion in this chapter was developed by Snyder et al. and has been operated as a field GC/IMS analyzer for pyrolysis (py) analysis of bacteria in aerosols.57–60 This instrument arose from the same efforts that led to the production of the EVM (see earlier discussion), and a schematic is shown in Figure 5–20. The instrument comprises a CAM drift tube as the IMS detector, and a 2-m capillary column of low polarity is connected to the ion source of the drift tube. As with the EVM, the drift tube, and column and inlet are all pumped to a slightly subambient pressure, and scrubbed air is used as the carrier gas. A pyrolysis inlet is fitted to the gas chromatograph so that airborne bacterial aerosols could be deposited in the tube, and rapidly heated to 350°C in ~5 sec so that the pyrolyzate is transferred to the GC column. Analysis by GC/IMS is made in several minutes with a rapid temperature program on the GC column. This GC/IMS has been extensively tested and validated in outdoor studies that were carried

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_216... 06/05/2007

Página 1 de 1

Page 217

FIGURE 5–20 Photograph of a pyrolysis-gas chromatograph-ion mobility spectrometer. The instrument contains a CAM drift tube as the IMS detector and a lowpolarity capillary column. Airborne bacterial aerosols, deposited in the tube, are rapidly heated to 350°C in ~5 sec and the pyrolyzate is transferred to the GC column. (From Snyder, A.R; Thornton, S.N.; Dworzanski, J.P.; Meuzelaar, H.L.C., Detection of the picolinic acid biomarker in Bacillus spores using a potentially field-portable pyrolysis-gas chromatography-ion mobility spectrometry system, Field Anal Chem. Technol. 1996, 1, 49–59; Dworzanski, J.P.; McClennen, W.H.; Cole, P.A.; Thornton, S.N.; Meuzelaar, H.L.C.; Arnold, N.S.; Snyder, A.R, A field-portable, automated pyrolysis-GC/IMS system for rapid biomarker detection in aerosols: a feasibility study, Field Anal. Chem. Technol 1997, 1, 295–305; Snyder, A.R; Maswadeh, W.M.; Tripathi, A.; Dworzanski, J.P., Detection of Gram-negative Erwinia herbicola outdoor aerosols with pyrolysis-gas chromatography/ion mobility spectrometry, Field Anal Chem. Technol. 2000, 4, 11–126; Snyder, A.R; Tripathi, A.; Maswadeh, W.M.; Ho, J.; Spence, M, Field detection and identification of a bioaerosol suite by pyrolysis-gas chromatography-ion mobility spectrometry, Field Anal Chem. Technol 2001, 5, 190–204.)

out in desert areas of western U.S. and Canada. In early work, five aerosol particles per liter of air for the Gram-negative bacterium Erwinia herbicola could be detected in approximately 2.5 min (after sample collection). However, refinements have been made on this instrument, and reproducible detection limits are now below 0.5 bacteria particles per liter of air from 2000 liters of air, concentrated using a high-volume sampler. The methods developed for this Py-GC/IMS analyzer were able to discriminate between aerosols of Gram-positive spores, a Gram-negative bacterium, and a protein (ovalbumin) through the detection of certain biomarkers with correct retention times and drift times. Although not commercially available, this instrument has been under sustained development during a 10-year period, with a linage traceable to the EVM. As with the EVM, the work demonstrates that the advantages that made IMS analyzers attractive to users such as the U.S.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_217... 06/05/2007

Página 1 de 1

Page 218 Army, namely portability and reliability, may be realizable also with GC/ IMS instruments. The advantages of the GC column in clarifying the mixture generated as pyrolysate are implicit in the measurements. Further refinements may be expected in this instrument with improvements to the drift tube and temperature control. In summary, GC/IMS has been recognized during the past decade as a next step in one path of refinement of IMS analyzers, and this is evident in the developments described in the preceding text. The development of integrated GC/IMS instruments by Smiths Detection and by Varian Corporation. suggests that the transition of the technology from research laboratories to commercial ventures has occurred. As with regular IMS instruments, the availability of technology to meet the needs of a broad range of users, as might be found in ordinary GC, has been elusive with gas chromatograph/ mobility spectrometers. Compelling applications and a market requirement will be needed for this to change.

5.3 LC/IMS 5.3.1 A Brief History Early in the development of IMS, the possibility of interfaces with liquid chromatographs was investigated, but attempts to operate a liquid chromatograph/ ion mobility spectrometer were complicated by the excessive volumes of solvent of the effluent.61,62 Although the results of these early-1970s developments were mixed, an IMS drift tube should be attractive as a detector for condensed-phase methods due to the ambient-pressure operations of the drift tube. Indeed, advances in liquid chromatography during the 1980s resulted in some explorations of LC/IMS in the early 1990s.63–65 Later, supercritical fluid chromatography66–72 and electrophoresis73 instruments were interfaced to a mobility spectrometer for the same reasons as any hyphenated chromatograph/drift tube combination. In each of these cases, the chromatograph produced comparatively large volumes of gas from the liquid or supercritical effluent (a 1000-fold increase in volume occurs when a liquid is converted to a gas).

5.3.2 Recent Developments In the previous edition of this book,43 the prospects of LC/IMS methods were assessed as unpromising, and “…any future for IMS as a detector for these other chromatographic techniques is not guaranteed.” Although this judgment may seem harsh, few developments or advances in the technology, practice, or application of LC/IMS have occurred as judged by the number

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_218... 06/05/2007

Página 1 de 1

Page 219 of publications in the refereed literature. Twelve years on from that prediction, this lack of development is slightly perplexing in view of the developments with liquid-based analysis by mass spectrometry. Mass spectrometry is, as a topic, now dominated by biomedical and biological applications that have been made possible by the popularization of electrospray ionization. Electrospray ionization is an effective method to form gas-phase ions from large, thermally labile molecules in air at ambient pressure and directly from the liquid or aqueous state. Most of the advantages of electrospray ionization could be transferred to mobility spectrometers. Indeed, one of the earliest applications of electrospray ionization was with mobility spectrometers in 1984.74 Because a mobility spectrometer can be an economical alternative to a mass spectrometer, development of LC/IMS methods might have been expected during the later 1990s. Instead, despite the promise of a convenient and effective interface between liquid chromatographs and mobility spectrometers, only three journal articles appeared in the refereed literature, including the determination of carbohydrates75 and the determination of benzodiazepines and triazine herbicides.76 The question then again is: What is the balance of advantages and disadvantages of IMS as a detector for LC? Unlike the fusion between electrospray ion sources and mass spectrometers, in which MS instruments were commercially available and able to be quickly adapted to electrospray ionization, no such condition exists for mobility spectrometry. What instruments do exist cannot be easily adapted to electrospray ionization. Consequently, even if a need exists in the industrial or medical worlds, there is no hardware to support development. Although the technical prospects for LC/IMS instruments still appear potentially promising, practical conditions in 2005 do not seem favorable for even a small measure of acceptance in the analytical measurement community.

5.4 IMS/MS 5.4.1 Background One of the first published works on IMS, which was then still known as plasma chromatography, described a sophisticated GC/IMS/MS instrument (Figure 5–21).1 The function of the gas chromatograph was to preseparate the components of the sample; the IMS could be used either as a detector or as an additional separation technique, and the mass spectrometer was used to identify and quantify the ions emerging from the mobility spectrometer. Subsequently, the gas chromatograph was removed from the front end of the instrument, and IMS/MS remained a powerful analytical tool, used in the measurement of ions formed in atmospheric ionization processes. A stand-alone IMS instrument cannot provide definitive identification of ions

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_219... 06/05/2007

Página 1 de 1

Page 220

FIGURE 5–21 Photograph (top) and schematic diagram (bottom) of Alpha series GC/IMS/MS instrument (made by PCP, West Palm Beach, FL) from the early 1970s. The size of this innovative instrument reflects the state-of-the-art electronics and mechanics of the period.

in a mobility spectrum, although prior knowledge of the sample composition and in-depth understanding of the gas-phase ion chemistry can be combined to make an informed estimate of the probable identity of an ion. The record shows this to be a serious risk with ions in air at atmospheric pressure, even when experimental conditions are thought to be well controlled. Only the mass identification of ions by IMS/MS can definitively allow a peak in a mobility spectrum to be identified; even so, supersonic expansion of ions between ambient pressure and high vacuum can alter the exact identity of an ion by removal of clustered molecules while preserving the core of the ion. A word of caution may be added here: analytical or monitoring applications

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_220... 06/05/2007

Página 1 de 1

Page 221 based on changes in the response of the IMS instrument to different compounds or different concentrations may reflect the effects of contaminants or additives to the sample rather than the effects due to the target compound; in such circumstances, MS is needed to verify ion identities. The early IMS/ MS studies provided the foundations for understanding the intricate ion processes that occur inside a drift tube operated at ambient pressure and on the important role played by moisture, temperature, and gaseous impurities. Before 1990, only a few IMS/MS instruments were employed in research supporting the development of IMS as an analytical method. A listing of many of the early IMS/MS investigations is shown in Table 5–3; the average is about one publication every 2 years. The subjects of these investigations were varied and ranged from the identification of reactant ions80,83 to studies in support of IMS measurements.77,79 The limited number of investigations may be attributed to the low number of instruments in research laboratories. During the 1990s, improvements occurred in the availability and variety of IMS/MS instruments and publications became routine and nearly mandatory when describing ionization chemistry. The intrinsic difference between the operating pressure (650 to 760 torr) of the mobility spectrometer and that of the mass spectrometer (10−4 to 10−6 torr) means that the design of the interface so that ions can be transported without loss or significant identity changes presents theoretical and practical challenges. TABLE 5–3 Early Reports on IMS/MS Investigations

Year

Title

Reference

1971 Trace studies of alcohols in the plasma chromatograph-mass spectrometer

77

1976 Mass-identified mobility spectra of p-nitrophenol and reactant ions in plasma chromatography

78

1976 Plasma chromatography of heroin and cocaine with mass-identified mobility spectra

79

1977 Negative ions in plasma chromatography-mass spectrometry

80

1977 Analysis of surface contaminants by plasma chromatography-mass spectroscopy

81

1978 Plasma chromatography of benzene with mass-identified mobility spectra

82

1978 Mobility behavior and composition of hydrated positive reactant ions in plasma chromatography with nitrogen carrier gas

83

1984 Mass-analysed ion mobility studies of nitrobenzene

84

1985 Ion mobility spectrometry/mass spectrometry (IMS/MS) of two structurally different ions having identical ion mass

85

1988 Differentiating between large isomers and derivation of structural information by ion mobility spectrometry mass spectrometry techniques

86

1990 An ion mobility spectrometry-mass spectrometry (IMS-MS) study of the site of protonation in anilines

87

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_221... 06/05/2007

Página 1 de 1

Page 222

5.4.2 Interfaces between IMS Drift Tubes and MS Analyzers As a gas-phase ion moves through the pinhole or orifice between ambient pressure and high vacuum, ions and molecules of the supporting atmosphere undergo supersonic expansion.88 This results in unregulated and unpredictable changes in the internal energy of the ion and in the composition of the cluster between the core ion and neutral molecules. Examples of this uncertainty can be found in the degree of clusterization for ions of acetone (MH+, 59 amu and M2H+, 117 amu) with water at different potentials on the orifice and first lens in the high vacuum. As shown in Figure 5–22, small changes in these potentials can cause quantitative changes in both absolute ion transmission and relative abundances. These phenomena are essential to MS applications with IMS. Only MS can provide an answer with regard to the exact composition of an ion from a mobility spectrum and determine the identity of the protonated species observed in the IMS: MH+, MH+H2O, MH+(H2O)2, or higher levels of hydration. In other studies, nitrogen or carbon dioxide adducts to a molecule have been identified, i.e., MO2CO2+, or MO2CO2N2+. Were these ions present in the ion swarm or were the clusters formed during later stages of expansion when ions are cooled? Such changes occur not just with small lightly held molecules but also with proton-bound dimers, even of ketones which are strongly held ions. Consequently, the mass analysis of an ion by IMS/MS will be accurate for the core ion only and will not be a reliable measure of true ion identity in the drift tube. This is inherent to all atmospheric pressure source-based mass spectrometers as well as IMS/ MS instruments. This precludes IMS/MS as a tool to explore exact cluster sizes and transformation of clusters in drift tubes. Rather, the main goal of most IMS/MS measurements is to identify the core ion. In most IMS/MS investigations and virtually all studies in support of analytical IMS, identification of the core ion is sufficient, driven by limits of technology and principles. Two designs of interfaces between drift tubes and mass spectrometers are known and include either a pinhole orifice with a diameter of 20 to 50 μm or a skimmer cone with 100 μm, with direct transition of ions from ambient pressure to high vacuum and a differentially pumped design in which ions pass through two or more pressure differentials. The choice of interface is governed by the speed of pumps, ion efficiency, and user preference. For example, a cryogenic pumped-mass spectrometer is able to hold vacuum with a single pinhole or orifice but has limited pumping time requiring recycling of cryosurfaces after 8 h of constant operation. Differentially pumped instruments with turbomolecular or diffusion pumps typically are operated continuously without the recycling limitation of the cryogenic pumped-vacuum systems. Differentially pumped interfaces are often designed around two skimmer cones or a skimmer cone and large pinhole membrane. A main consideration in the interface region is the configuration of the electric fields so that ions can pass from the drift tube, through the interface, and into the mass analyzer without barriers in potentials. In this, the orifice or skimmer cones are electrically isolated from the vacuum housing and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_222... 06/05/2007

Página 1 de 1

Page 223

FIGURE 5–22 Mass spectra of ions created from acetone in ambient air with a corona discharge ion source showing the effect of various potentials on the orifice of the interface between ambient pressure and high vacuum. Potentials on the oriface were (A) 60, (B) 50, and (C) 40 with a potential of 40 V on the next lens element, a quadruple lens.

raised in potential above the mass analyzer, a common practice with quadrupole-based mass spectrometers. Then, the entire drift tube is floated at a potential above the orifice so that the last electrical ring or grid in the drift tube is some tens or hundreds of volts above the orifice with fields of ~200 to 400 V/cm in this region. Care must be given to the proximity of housing and other components that may be grounded and cause distortions in field lines, drawing ions away from the orifice or pinhole. In interfaces with time-of-flight mass spectrometers (TOF-MSs), other configurations may be used in which the orifice is at ground potential or below ground, and the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_223... 06/05/2007

Página 1 de 1

Page 224 entire TOF-MS is lowered and isolated at several thousands of volts below the orifice. The control of field and geometry is still under active discussion, and each IMS/MS instrument in use today is built by individual research laboratories and is unique. Regardless of pumping system or interface design, the goal in an IMS/MS experiment is to garner information about the mobility spectrum, the sample, or the ionization chemistry associated with the sample, ion source, and supporting atmosphere.

5.4.3 Analytical Information The drift tube in the original ion mobility spectrometer/mass spectrometer (an Alpha model, from Franklin GNO, then PCP, Inc, West Palm Beach, FL) contained two ion shutters, separated by about 5 cm (see Figure 5–21). The first gate was a regular pulsed ion shutter allowing a swarm of ions to enter the drift region of the IMS at a frequency of about 30 Hz. The second ion shutter may be opened continuously to allow all ions to enter the interface and the mass spectrometer. Alternatively, the second gate could be operated with a boxcar integrator to obtain a mobility spectrum. Finally, the mass spectrometer could be set to a specific ion and used as an ion specific detector for the drift tube. A fourth possible operation of the ion shutters, namely, the selection of ions in a certain peak in a mobility spectrum for characterization by MS was not used, presumably because of low signal level due to large ion losses in the interface. The emphasis in the following discussion is on examples of data acquisition by IMS/MS instruments. 5.4.3.1 Acquisition of a Mobility Spectrum by IMS/MS Drift tubes in early and contemporary IMS/MS instruments are equipped with a grid or plate with a hole at the detector end of the drift tube, and this is floated electrically to a proper position in the overall field gradient and used as a detector of ions with normal Faraday plate designs found in standalone IMS analyzers. The output of the IMS detector is displayed, just as in regular IMS operation, and no use is made of the mass spectrometer. Alternatively, ions may be passed from the drift tube to the detector of the mass spectrometer without any mass filtering. In this way, the electron multiplier serves as the detector for the mobility spectrometer. Although the distance the ions traverse from the gate to the mass spectrometer detector is more than twice the distance to the IMS detector, the drift times differ only by about 1 to 2%. This is because the ions encounter very little resistance to movement in the low pressure of the mass spectrometer. Regardless of the method of detection, the complete mobility spectrum is valuable in demonstrating the proper performance of the IMS/MS instrument and in providing a mobility spectrum for the measurement or experiment. 5.4.3.2 Acquisition of a Mass Spectrum by IMS/MS By leaving the two shutter grid gates in the drift tube open continuously, ions produced in the reaction region of the IMS drift tube are passed

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_224... 06/05/2007

Página 1 de 1

Page 225 without any mobility analysis to the mass spectrometers. In this mode, the mass spectrum obtained is a measure of the total number of ions; caveats regarding ion identities, ion losses, and ion transformations in the interface still apply here, and the relationship between ions in the mass spectrum and those in the mobility spectrum (see preceding text) must be treated with caution. Ions that are long-lived or stable will be seen in the mass spectrum, modified only by the addition or exclusion of neutral adducts such as water, carbon dioxide, etc. Unstable ions may undergo decomposition in the interface, and the mass spectrum may not represent the chemical identity of peaks in the mobility spectrum. In extreme instances, decomposition of an ion may be observed in the mobility spectrum even before an ion reaches the interface. Thus, a mass spectrum must be interpreted in light of the known, calculated, or expected stability of the ions, which can sometimes be predicted. The mass spectrum is useful in clarifying which ions may be found in the mobility spectrum and was routinely obtained with the early commercial IMS/MS instrument in preparation for obtaining tuned ion mobility spectra (see Subsection 5.4.3.4). Although no mobility information is available in this mode of data acquisition, the best signal-to-noise values for any IMS/MS experiment are provided when a mass spectrum of all ions is collected. An example of a mobility spectrum from an IMS/MS experiment is shown in Figure 5–23 (top frame) for a butyl acetate.89 Details of ion behavior in the drift tube can be obtained from either mass spectra for individual peaks or from tuned ion mobility spectra, as described in the following subsections. 5.4.3.3 Mass Spectrum for Each Mobility Peak In this mode of operation, a second ion shutter is pulsed with a delay corresponding to the drift time of a peak in the mobility spectrum. The mass spectrometer is used to mass-analyze all the ions that constitute this peak, providing a complete mass spectrum for only this component of the mobility spectrum. The immediate usefulness of this method lies in the identification of the core ion and, hence, the chemical associated with the peak. This method of acquiring MS data is not commonly used in IMS/MS measurements because signal levels are reduced in three parts or steps: the first ion shutter, the second ion shutter, and the interface. Consequently, signal averaging of as many as 500 to 2000 mass spectra may be needed to obtain a mass spectrum with an acceptable signal-to-noise ratio. Interpretation of a mass spectrum may also be complicated because a peak in a mobility spectrum arises from an ion swarm, and several ions may exist in a swarm. Thus, ions of different masses may appear as a single peak in the mobility spectrum but will be resolved and determined at each m/z in the mass spectrometer. Well-known examples of these are cluster ions, in which the core ion is surrounded by several neutral molecules, especially by water and nitrogen molecules. An example of this was shown in Figure 3–3 in Chapter 3, in which the several different negative ions

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_225... 06/05/2007

Página 1 de 1

Page 226

FIGURE 5–23 Results from IMS/MS characterization of butyl acetate are shown as: the whole mobility spectrum (lower trace) and the tuned ion mobility spectra for the reactant ions at m/z 37 Da, a fragment ion at m/z 57 Da and the protonated monomer at m/z 117 Da (upper trace).89

appeared as a single peak in the mobility spectrum. The facility of isolating a peak for MS analysis was not a feature on early instruments and is not commonly used today. 5.4.3.4 Tuned Ion Mobility Spectrum Once the ions are determined from a mass spectrum (see Subsection 5.4.3.2), the mobility behavior and relationships between ions can be gleaned through what was termed a tuned ion mobility spectrum. In this mode of operation,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_226... 06/05/2007

Página 1 de 1

Page 227 the mass spectrometer is preset to measure only ions with specific m/z values in ways that parallel single-ion monitoring with GC/MS measurements. A plot of ion intensity vs. time is obtained for each ion, in which the mobility spectrometer is operated with a single ion shutter and the mass spectrometer serves to filter and detect ions. This is shown in Figure 5–23 for ions of butyl acetate, which may be compared on the basis of mobility and intensity, and is particularly useful when mobility spectra may arise from ion-molecule reactions in the drift region, as shown in Figure 5–23. The appearance of tuned ion mobility spectra can be used to define relationships and reactions in the IMS drift tube. This was illustrated for the decomposition of butyl acetate in the 2 to 15 msec timescale, as shown in Figure 5–23. Comparisons of tuned ion mobility spectra for protonated monomer, and proton-bound dimer and fragment ions were used to suggest a mechanism for decomposition.89 In another study, the site of protonation of aniline derivatives was investigated.87 According to theoretical calculations, the energy difference between the nitrogen-protonated and ring-protonated configurations is very small, and the two isomeric species cannot be resolved by MS alone. However, two distinct peaks were observed in the IMS/MS spectrum, in which the ringprotonated species had a higher mobility than the nitrogenprotonated species due to delocalization of the ion charge that resulted in a smaller cross section.87 The tuned ion mobility spectrum can also be used to clarify which ions in the mass spectrum of all ions are associated by clustering. In an early work with IMS/MS instrumentation, the distribution of ion clusters in reactant ions for positive polarity were determined as a function of the temperature of the drift tube.83 The determinations should not be regarded as absolute, due to limitations from ion declustering in the interface region.

5.4.4 IMS/MS Instruments 5.4.4.1 Traditional IMS/MS with a Quadrupole Mass Analyzer Although IMS/MS instruments all still mainly a research tool for laboratory studies, several different designs have been proposed. The traditional, or classic, IMS/MS (MMS-160 made by PCP, Inc.) is shown in Figure 5–21. The instrument included a complete ion mobility drift tube with a 63Ni ion source at ambient pressure, an interface region that was pumped with a turbomolecular pump to 10−4 torr, a standard electron impact (EI) ionization source, and a quadrupole mass spectrometer that operated at 10−5 to 10−6 torr and was evacuated with a diffusion pump. Gas or vapor samples were introduced into the IMS, where ionization took place and ions were separated in the drift region. In addition to the four operation modes described earlier, the EI ionization source could be used to ionize neutral molecules that entered the interface region. This made it possible to study the components of samples that were not ionized, or did not form stable ions, under the conditions of the IMS. However, judging from the published literature, not

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_227... 06/05/2007

Página 1 de 1

Page 228

FIGURE 5–24 An IMS/MS instrument with differential pumping and two cones (skimmer and sampler) in the interface and a quadrupole mass spectrometer. (From Hill, C.A.; Thomas, C.L.R, A pulsed corona discharge switchable high resolution ion mobility spectrometer-mass spectrometer, Analyst 2003, 128, 55–60.)

much use was made of this feature. The MMS-160 and similar instruments, provided ion identification, showed that a single peak in the mobility spectrum could actually be due to several ions and clusters of different masses, and also gave valuable structural information that could be obtained from the tuned ion mode of operation. An instrument with related engineering was built by Graseby Dynamics and was rebuilt in the laboratory of C.L.R Thomas in Manchester and is shown in Figure 5–24.90 Variations of this design include the tandem mass spectrometer at New Mexico State University, where an existing MS/MS with an atmospheric pressure ion source was fitted with a mobility spectrometer. Electrospray sources have been fitted on IMS/quadrupole analyzers for a variety of applications.91–95 5.4.4.2 Low-Pressure IMS/Quadrupole MS Variations on the traditional IMS/MS design were made with an elongated drift tube to provide improved resolution of mobility peaks. This was used for studies of structural isomers and to obtain information on the collision cross section of large polyatomic ions. For example, a 63-cm-long drift tube was operated using helium as the drift gas at 500 torr, and the detector was

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_228... 06/05/2007

Página 1 de 1

Page 229 a quadrupole analyzer.96 In this particular study, a pulsed laser was used to desorb and vaporize the sample; stable ionized clusters of 60 and 70 carbon atoms were observed in positive-mode operation and a whole series of silicon-atom clusters (n ranging from 6 to 46) were seen in negative-polarity mode. In addition, clusters of sodium chloride molecules attached to a chloride ion of negative polarity and attached to a sodium ion of positive polarity were characterized. 5.4.4.3 IMS/TOF MS and IMS/Ion Trap/TOF MS The major change in recent years has come with the combination of TOF mass spectrometers with IMS drift tubes. When this combination is equipped with an electrospray ion source, a powerful tool is available for studying the configuration of large biomolecules. This design is shown schematically in Figure 5–25.97 The liquid sample is introduced and ionized by the ESI inlet and enters the desolvation region that is 5-cm long and maintained at a pressure of 1 to 10 torr. Some of the ions pass through an orifice and enter the drift tube that is 42.8-cm long and contains helium at a low pressure (~3 torr), which serves as the buffer gas. In the drift region, the ions are rapidly thermalized by collisions and separated according to their mobility coefficient (corresponding to their collision cross section). The ions are then focused into the TOF mass spectrometer where they are massanalyzed. This system has been used to study the conformation of several large molecules of biological interest, such as ubiquitin.

FIGURE 5–25 Electrospray ionization/low-pressure ion mobility spectrometry/time-offlight mass spectrometry instrument used for studying biological molecules. (From Hoaglund, C.S.; Valentine, S.J.; Sporleder, C.R.; Reilly, J.P.; Clemmer, D.E., Three-dimensional ion mobility/TOFMS analysis of electrosprayed biomolecules, Anal Chem. 1998, 70, 2236–2242.)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_229... 06/05/2007

Página 1 de 1

Page 230

FIGURE 5–26 Electrospray ionization/ion trap/low-pressure ion mobility spectrometry/time-of-flight mass spectrometry (ESI/IT/IMS/TOFMS) instrument that includes an ion trap between the ionization source and drift tube. (From Hoaglund, C.S.; Valentine, S.J.; Clemmer, D.E., An ion-trap interface for ESI-ion mobility experiments, Anal Chem. 1997, 69, 4156–4161.)

A somewhat different approach was to use an ion-trap interface between the ESI-source drift tube and the TOF-MS, as shown in Figure 5–26.98,99 Here too, liquid samples are introduced and ionized by ESI, and the ions are trapped after desolvation in order to accumulate an ion bunch before injecting them into the drift tube. The application of this device for studies of the conformation of heme molecules yielded unique structural information, as depicted in Figure 5–27, where the folded conformer is shown with a significantly smaller cross section that the unfolded conformer. Tryptic digest products of large biological molecules have been studied by a combination of a high-pressure MALDI (matrix-assisted laser desorption/ionization) source for desorption and ionization of the sample, a mobility drift tube, and a TOF-MS positioned orthogonal to the drift tube (MALDI/IM-o-TOF), shown schematically in Figure 5–28.100 The size of the drift tube (12.5 cm) and TOFMS (20 cm) made this apparatus much more compact than those described earlier, but with a corresponding degradation of the analytical performance (resolution in the drift tube and TOF). 5.4.4.4 Field-Dependent Mobility Analyzers/Mass Spectrometers Although IMS/TOFMS instruments have afforded insights into the shapes and folding of biomolecules, the analytical utility is uncertain for the current measurement requirements in the burgeoning biological

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_230... 06/05/2007

Página 1 de 1

Page 231

FIGURE 5–27 Mobility spectra of heme molecules, obtained with ESI/IT/IMS/TOFMS instrument clearly show difference between the cross section of folded conformers and unfolded conformers. The former have smaller cross sections and higher mobilities than the latter. (From Hoaglund, C.S.; Valentine, S.J.; Clemmer, D.E., An ion trap interface for ESI-ion mobility experiments, Anal. Chem. 1997, 69, 4156–4161.)

MS efforts. Perhaps a significant impact near term may come from the union of various fielddependent mobility analyzers such as the field asymmetric IMS (FAIMS) instrument or the differential mobility spectrometer. In 1999, a Canadian team interfaced an FAIMS analyzer to a mass spectrometer and began a series of investigations establishing an FAIMS drift tube as a valuable inlet tool to pre-filter ions before MS measurements.101–107 In the FAIMS analyzer, ions are passed continuously through the analyzer, unlike the pulsed operation of traditional drift tubes. This allows the analyzer to serve as an effective ion filter before the mass spectrometer, in which ions of only a certain compensation voltage are passed to the mass spectrometer. The reduction of chemical noise with a FAIMS or DMS inlet for a mass spectrometer has been demonstrated with biological samples and has been commercialized as an enhancement for ESI-MS measurements. A micro-fabricated drift tube operated with field-dependent mobilities has also been coupled to a mass spectrometer in order to verify ion identities108 and is shown in Figure 5–29.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_231... 06/05/2007

Página 1 de 1

Page 232

FIGURE 5–28 Schematic of the (MALDI/IM-o-TOF) instrument. A nitrogen laser is used for desorption and ionization of the sample, and ions are separated in the drift tube and mass-analyzed by a TOFMS instrument positioned orthogonally. (From Gillig, K.J.; Ruotolo, B.; Stone, E.G.; Russell, D.H.; Fuhrer, K.; Gonin, M.; Schultz, A.J., Coupling high-pressure MALDI with ion mobility/orthogonal time of flight mass spectrometry, Anal. Chem. 2000, 72, 3965–3871.)

5.5 Summary The combination of ion mobility analyzers with gas chromatographs has experienced rapid growth during the past decade and recently has been noteworthy for the first affordable GC/mobility analyzer from a leading manufacturer of gas chromatographs, the Varian μ-DMD instrument. The improvements in performance in each of the components nearly guarantee that further development and refinement may be expected. The capabilities of GC/IMS instruments underwent large changes in the 1990s, moving from the laboratory to handheld analyzers, demonstrating that the concept is valuable and realizable without enormous expense or effort. In contrast, there seems to be only little interest in the use of IMS analyzers as detectors for LC and, rather, some have suggested that a mobility analyzer should replace liquid chromatographs in LC/MS instruments. While a mobility spectrometer can provide selectivity before a mass spectrometer, the chemistry of ionization with complex samples may permit matrix effects or other interferences. Such complications in the formation of ions may preclude a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_232... 06/05/2007

Página 1 de 1

Page 233

FIGURE 5–29 DMS/MS schematic (A) and photograph (B) of the microfabricated field-dependent drift tube coupled to a quadrupole mass spectrometer. (From Eiceman, G.A.; Nazarov, E.G.; Miller, R.A., A micro-machined ion mobility spectrometer-mass spectrometer, Int. J. Ion Mobility Spectrom. 2001, 3, 15– 27.)

simple exchange of LC/MS by IMS/MS. As a research tool, mobility spectrometers have been accepted as valued tools largely through the efforts of teams working with conformation and structure determinations for large biological molecules (see Chapter 7). Although these developments seem promising, IMS/MS instruments are still research instruments and the only commercial manufacturer (PCP, Inc.) has ceased operations. Thus, IMS/MS

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_233... 06/05/2007

Página 1 de 1

Page 234 instruments will be built only by investigators in the foreseeable future. A recent review may provide a measured overview of IMS/MS.109

References 1. Cohen, M.J.; Karasek, F.W., Plasma chromatography™—a new dimension for gas chromatography and mass spectrometry, J. Chromatogr. Sci. 1970, 8, 330–337. 2. Karasek, F.W.; Keller, R.A., Gas chromatograph/plasma chromatograph interface and its performance in the detection of musk ambrette, J. Chromatogr. Sci. 1972, 10, 626–628. 3. Cram, S.P.; Chesler, S.N., Coupling of high speed plasma chromatography with gas chromatography, J. Chromatogr. Sci. 1973, 11, 391–401. 4. Metro, M.M.; Keller, R.A., Fast scan ion mobility spectra of diethyl, dipropyl, and dibutyl ethers as determined by the plasma chromatography, J. Chromatogr. Sci. 1973, 11, 520–524. 5. Karasek, F.W.; Kim, S.H., The plasma chromatograph as a qualitative detector for gas chromatography, J. Chromatogr. 1974, 99, 257–266. 6. Keller, R.A.; Metro, M.M., Evaluation of the plasma chromatograph as a separator-identifier, J. Chromatogr. Sci. 1974, 12, 673–677. 7. Karasek, F.W.; Hill, H.H., Jr.; Kim, S.H.; Rokushika, S., Gas chromatographic detection modes for the plasma chromatograph, J. Chromatogr. 1977, 135, 329–339. 8. Ramstad, T.; Nestrick, T.J.; Tou, J.C., Studies of the effects of volatile components from gas chromatographic liquid phases on plasma chromatographic performance, J. Chromatogr. Sci. 1978, 16, 240– 245. 9. Hagen, D.F., Characterization of isomeric compounds by gas and plasma chromatography, Anal Chem. 1979, 51, 872–874. 10. Ithakissios, D.S., Plasmagram spectra of some barbiturates, J. Chromatogr. Sci. 1980, 18, 88–92. 11. Baim, M.A.; Hill, H.H., Jr., Tunable selective detection for capillary gas chromatography by ion mobility monitoring, Anal. Chem. 1982, 54, 38–43. 12. St. Louis, R.G.; Siems, W.G.; Hill, H.H., Jr., Evaluation of direct axial sample introduction for ion mobility detection after capillary gas chromatography, J. Chromatogr. 1989, 479, 221–231. 13. Young, D.; Thomas, C.L.P.; Breach, J.; Brittain, A.H.; Eiceman, G.A., Extending the concentration and linear dynamic range of ion mobility spectrometry with a sheath flow inlet, Anal Chim. Acta 1999, 381, 69– 83. 14. Young, D.; Eiceman, G.A.; Breach, J.; Brittain, A.H.; Thomas, C.L.P., Automated control and optimization of ion mobility spectrometry responses using a sheath-flow inlet, Anal Chim. Acta 2002, 463, 143–154. 15. Baim, M.A.; Eatherton, R.L.; Hill, H.H., Jr., Ion mobility detector for gas chromatography with a direct photoionization source, Anal Chem. 1983, 55, 1761–1766. 16. Baim, M.A.; Hill, H.H., Jr., Determination of 2,4-dichlorophenoxyacetic acid in soils by capillary gas chromatography with ion-mobility detection, J. Chromatogr. 1983, 279, 631–642.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_234... 06/05/2007

Página 1 de 1

Page 235 17. Baim, M.A., Schuetze, F.J.; Frame, J.M.; Hill, H.H., Jr., A microprocessor-controlled ion mobility spectrometer for selective and nonselective detection following gas chromatography, Am. Lab. 1982,14, 59– 70. 18. Schneider, A.A.; Carnahan, B.L.; Matyjaszczyk, M.S.; Zuckerman, L.; Fertig, G.; Williams, M., High sensitivity GC-FIS for simultaneous detection of chemical warfare agents, AT-PROCESS 2000, 5, 124–136. 19. Eiceman, G.A.; Tadjikov, B.; Krylov, E.; Nazarov, E.G.; Miller, R.A.; Westbrook, J.; Funk, P, Miniature RF-ion mobility analyzer as a gas chromatographic detector: oxygen-containing volatile organic compounds, pheromones and other insect attractants, J. Chromatogr. 2001, 917, 205–217. 20. Eiceman, G.A.; Nazarov, E.G.; Miller, R.A.; Krylov, E.; Zapata, A., Micromachined planar field asymmetric ion mobility spectrometer as gas chromatographic detectors, Analyst 2002, 127, 4, 466–471. 21. Eiceman, G.A.; Tadjikov, B.; Ewing, R.G.; Nazarov, E.G.; Krylov, E.; Miller, R., Differential mobility spectrometer of chlorocarbons with micro-fabricated drift tube, The Analyst 2004, 129, 297–304. 22. Buryakov, I.A.; Kolomiets, Y.N., Rapid determination of explosives and narcotics using a multicapillarycolumn gas chromatograph and an ion-mobility spectrometer, J. Anal. Chem. (Translation of Zhurnal Analiticheskoi Khimii) 2003, 58, 944–950. 23. Buryakov, I.A.; Kolomiets, Y.N.; Louppou, V.B., Ion non-linear drift spectrometer (INLDS)—a selective detector for high-speed gas chromatography, Int. J. Ion Mobility Spectrom. 2001, 4, 13–15. 24. CP-4900 DMD based Analytical Solution, CFC’s Gas Analyzer available at http://www.varianinc.com/image/vimage/docs/products/chrom/gc/microgc/shared/DS-CP501682–03– 041.pdf and Sulfur Odorant in Natural Gas Analyzer available at http://www.varianinc.com/image/vimage/docs/products/chrom/gc/microgc/shared/DS-CP501679–02– 042.pdf 25. Eiceman, G.A.; Tarassov, A.; Miller, R.A.; Nazarov, E.G.; Hughes, E.; Funk, P, Discrimination of combustion fuel sources using gas chromatography-planar field asymmetry ion mobility spectrometry, J. Sep. Sci. 2003, 26, 585–593. 26. Kojiro, D.R.; Cohen, M.J.; Stimac, R.M.; Wernlund, R.F.; Humphry, D.E.; Takeuchi, N., Determination of C1–C4 alkanes by ion mobility spectrometry, Anal. Chem. 1991, 63, 2295–300. 27. Eiceman, G.A.; Wang, Y.-F.; Garcia-Gonzalez, L.; Harden, C.S.; Shoff, D.B., Enhanced selectivity in ion mobility spectrometry analysis of complex mixtures by alternate reagent gas chemistry, Anal Chim. Acta 1995, 306, 21–33. 28. Baim, M.A.; Eatherton R.L.; Hill, H.H., Jr., Ion mobility detector for gas chromatography with a direct photoionization source, Anal. Chem. 1983, 55, 1761–1766. 29. Veasey, C.A.; Thomas, C.L.P., Fast quantitative characterisation of differential mobility responses, Analyst, 2004,129, 198–204. 30. Sielemann, S.; Baumbach, J.I.; Schmidt, H.; Pilzecker, P, Quantitative analysis of benzene, toluene, and mxylene with the use of a UV—ion mobility spectrometer, Field Anal Chem. Technol. 2000, 4, 157–169. 31. Sielemann, S.; Baumbach, J.I.; Schmidt, H., IMS with nonradioactive ionization sources suitable to detect chemical warfare agent simulation substances, Int. J. Ion Mobility Spectrom. 2002, 5, 143–148. 32. Vautz, W.; Sielemann, S.; Baumbach, J.I., Determination of terpenes in humid ambient air using ultraviolet ion mobility spectrometry, Anal. Chim. Acta 2004, 513, 393–399.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_235... 06/05/2007

Página 1 de 1

Page 236 33. Baumbach, J.I.; Sielemann, S.; Xie, Z.; Schmidt, H., Detection of the gasoline components methyl tertbutyl ether, benzene, toluene, and m-xylene using ion mobility spectrometers with a radioactive and UV ionization source, Anal. Chem. 2003, 75, 1483–1490. 34. Xie, Z.; Sielemann, S.; Schmidt, H.; Li, F.; Baumbach, J.I., Determination of acetone, 2-butanone, diethyl ketone and BTX using HSCC-UV-IMS, Anal Bio-anal. Chem. 2002, 372, 606–610. 35. Ruzsanyi, V.; Sielemann, S.; Baumbach, J.I., Determination of microbial volatile organic compounds (MVOC) using IMS with different ionization sources, Int. J. Ion Mobility Spectrom., 2002, 5, 138–142. 36. St. Louis, R.H.; Siems, W.F.; Hill, H.H., Jr., Detection limits of an ion mobility detector after capillary gas chromatography, J. Microcolumn Sep. 1990, 2(3), 138–145. 37. Simpson, G.; Klasmeier, M.; Hill, H.; Atkinson, D.; Radolovich, G.; Lopez-Avila, V.; Jones, T.L., Evaluation of gas chromatography coupled with ion mobility spectrometry for monitoring vinyl chloride and other chlorinated and aromatic compounds in air samples, J. High Resolut. Chromatogr. 1996,19, 301–312. 38. Limero, T; Reese, R.; Trowbridge, J.; Hohmann, R., Validation of the Volatile Organic Analyzer (VOA) Aboard the International Space Station, SAE Technical Paper Series 20030–01–2646. 33rd International Conference on Environmental Systems, San Antonio, Texas, July 2002. 39. Baim, M.A.; Hill, H.H., Jr., Effects of contamination on ion mobility detection after gas chromatography, J. Chromatogr. 1984, 299, 309–319. 40. Baumbach, J.I.; Eiceman, G.A.; Klockow, D.; Sielemann, S.; v. Irmer, A., Exploration of a multicapillary column for use in elevated speed gas chromatography, Int. J. Environ. Anal Chem. 1997, 66, 225–239. 41. Buryakov, I.A., Express analysis of explosives, chemical warfare agents and drugs with multicapillary column gas chromatography and ion mobility increment spectrometry, J. Chromatogr. B 2004, 800, 75–82. 42. http://www.uiggm.nsc.ru/uiggm/sibertex/column.html 43. Eiceman, G.A.; Karpas, Z., Ion Mobility Spectrometry, CRC Press, Boca Raton, FL, 1994. 44. Dworzanski, J.P.; Kim, M.-G.; Snyder, A.P.; Arnold, N.S.; Meuzelaar, H.L.C., Performance advances in ion mobility spectrometry through combination with high speed vapor sampling, pre-concentration and separation techniques, Anal Chim. Acta 1994, 293, 219–235. 45. Arnold, N.S.; Dworzanski, J.P; Sheya, S.A.; McClennen, W.H.; Meuzelaar, H.L.C., Design considerations in field-portable GC-based hyphenated instrumentation, Field Anal Chem. Technol. 2000, 4, 219–238. 46. Snyder, A.P.; Harden, C.S.; Brittain, A.H.; Kim, M-G.; Arnold, N.S.; Meuzelaar, H.L.C., Portable handheld gas chromatography/ion mobility spectrometry device, Anal Chem. 1993, 65, 299–306. 47. McClennen, W.H.; Arnold, N.S.; Meuzelaar, H.L.C., Field-portable hyphenated instrumentation: the birth of the tricorder?, TrAC: Trends Anal Chem. 1994, 13, 286–293. 48. http://194.105.117.18/products/Default.asp?Product=2 49. Limero, T.; Reese, E., First operational use of the ISS-VOA in a potential contingency event, Int. J. Ion Mobility Spectrom. 2002, 5, 27–30. 50. http://194.105.117.18/products/Default.asp?Product=12

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_236... 06/05/2007

Página 1 de 1

Page 237 51. DeBono, R.; Grigoriev, A.; Jackson, R.; James, R.; Kuja, F.; Loveless, A.; Nacson, S.; Rudolph, A.; Sun, Y, Separation of mixtures using gas chromatography coupled to ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2002, 5, 194–201. 52. Lafontaine, P.; Pilon, P.; Morrison, R.; Neudorfl, P., The use of GC-IMS to analyze high volume vapour samples from cargo containers, Int. J. Ion Mobility Spectrom. 2001, 4, 34–36. 53. Su, C.-W.; Babcock, K., The use of solid phase desorption/gas chromatography/ ion mobility spectrometry (SPD/GC/IMS) for explosive detection, Int. J. Ion Mobility Spectrom. 2002, 5, 67–70. 54. http://www.varianinc.com/cgi-bin/nav?products/chrom/gc/microgc/dmd&cid-LHQIIJKFP 55. Nazarov, E.G.; Miller, R.A.; Eiceman, G.A.; Krylov, E., Monitoring of trace amounts of sulfur compounds in air/hydrocarbon gas streams using a differential mobility spectrometer, Int. J. Ion Mobility Spectrom. 2002, 5, 76–81. 56. Eiceman, G.A.; Shoff, D.B.; Harden, C.S.; Snyder, A.R; Martinez, P.M.; Fleischer, M.E.; Watkins, M.L., Ion mobility spectrometry of Halothane, Enflurane, and Isoflurane Anesthetics in air and respired gases, Anal. Chem. 1989, 61, 1093–1099. 57. Snyder, A.R; Thornton, S.N.; Dworzanski, J.P.; Meuzelaar, H.L.C., Detection of the picolinic acid biomarker in Bacillus spores using a potentially field-portable pyrolysis-gas chromatography-ion mobility spectrometry system, Field Anal. Chem. Technol. 1996,1, 49–59. 58. Dworzanski, J.P.; McClennen, W.H.; Cole, P.A.; Thornton, S.N.; Meuzelaar, H.L.C.; Arnold, N.S.; Snyder, A.R, A field-portable, automated pyrolysis-GC/ IMS system for rapid biomarker detection in aerosols: a feasibility study, Field Anal. Chem. Technol. 1997, 1, 295–305. 59. Snyder, A.R; Maswadeh, W.M.; Tripathi, A.; Dworzanski, J.P., Detection of Gram-negative Erwinia herbicola outdoor aerosols with pyrolysis-gas chromatography/ ion mobility spectrometry, Field Anal. Chem. Technol. 2000, 4, 11–126. 60. Snyder, A.R; Tripathi, A.; Maswadeh, W.M.; Ho, J.; Spence, M, Field detection and identification of a bioaerosol suite by pyrolysis-gas chromatography-ion mobility spectrometry, Field Anal Chem. Technol. 2001, 5, 190–204. 61. Issenberg P.; Essigmann, J., The application of the plasma chromatograph as a detector in liquid chromatography, presented at the Pittsburgh Conference of Analytical Chemistry and Applied Spectroscopy, Cleveland, OH, March 1972. 62. Karasek, F.W.; Denney, D.W., Evaluation of the plasma chromatograph as a qualitative detector for liquid chromatography, Anal. Lett. 1973, 6, 993. 63. Shumate C B; Hill H.H., Jr., Coronaspray nebulization and ionization of liquid samples for ion mobility spectrometry, Anal. Chem. 1989, 61, 601–6. 64. Hill, H.H., Jr.; Siems, W.F.; Eatherton, R.L.; St. Louis, R.H.; Morrissey, M.A.; Shumate, C.B.; McMinn, D.G., Gas, supercritical fluid, and liquid chromatographic detection of trace organics by ion mobility spectrometry, Instrum. Trace Org. Monit. 1992, 49–64. 65. McMinn, D.G.; Kinzer, J.A.; Shumate, C.B.; Siems, W.F.; Hill, H.H., Jr., Ion mobility detection following liquid chromatographic separation, J. Microcolumn Sep. 1990, 2, 188–192. 66. Eatherton, R.L.; Morrissey, M.A.; Hill, H.H., Jr., Comparison of ion mobility constants of selected drugs after capillary gas chromatography and capillary supercritical fluid chromatography, Anal Chem. 1988, 60, 2240–2243.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_237... 06/05/2007

Página 1 de 1

Page 238 67. Huang, M.X.; Markides, K.E.; Lee, M.L., Evaluation of an ion mobility detector for supercritical fluid chromatography with solvent-modified carbon dioxide mobile phases, Chromatographia 1991, 31, 163–167. 68. Morrissey, M.A.; Widmer, H.M., Ion-mobility spectrometry as a detection method for packed-column supercritical fluid chromatography, J. Chromatogr. 1991, 552, 551–561. 69. Eatherton, R.L.; Morrissey, M.A.; Hill, H.H., Jr., Ion mobility spectrometry after supercritical fluid chromatography, J. High Resolut. Chromatogr. Chromatogr. Commun. 1986, 9, 154–160. 70. Rokushika, S.; Hatano, H.; Hill, H.H., Jr., Ion mobility spectrometry after supercritical fluid chromatography, Anal Chem. 1987, 59, 8–12. 71. Morrissey, M.A.; Siems, W.F.; Hill, H.H., Jr., Ion mobility detection of polydimethylsilicone oligomers following supercritical fluid chromatographic separation, J. Chromatogr. 1990, 505, 215–225. 72. Hannan, R.M.; Hill, H.H., Jr., Analysis of lipids in aging seed using capillary supercritical fluid chromatography, J. Chromatogr. 1991, 547, 393–401. 73. Hallen, R.W.; Shumate, C.B.; Siems, W.F.; Tsuda, T.; Hill, H.H., Jr., Preliminary investigation of ion mobility spectrometry after capillary electrophoretic introduction, J. Chromatogr. 1989, 480, 233–245. 74. Gieniec, J.; Mack, L.L.; Nakamae, K.; Gupta, C; Kumar, V.; Dole, M., Electrospray mass spectroscopy of macromolecules: application of an ion-drift spectrometer, Biomed. Mass Spectrom. 1984,11, 259–268. 75. Lee, D.S.; Wu, C; Hill, H.H., Detection of carbohydrates by electrospray ionization ion mobility spectrometry following microbore high-performance liquid chromatography, J. Chromatogr. A 1998, 822, 1–9. 76. Collins, D.C.; Xiang, Y.; Lee, M.L., Comprehensive ultra-high pressure capillary liquid chromatography/ion mobility spectrometry, Chromatographia 2002, 55, 123–128. 77. Karasek, F.W.; Cohen, M.J.; Carroll, D.I., Trace studies of alcohols in the plasma chromatograph-mass spectrometer, J. Chromatogr. Sci. 1971, 9, 390–392. 78. Karasek, F.W.; Kim, S.H.; Hill, H.H., Jr., Mass-identified mobility spectra of p-nitrophenol and reactant ions in plasma chromatography, Anal. Chem. 1976, 48, 1133–1137. 79. Karasek, F.W.; Hill, H.H. Jr.; Kim, S.H., Plasma chromatography of heroin and cocaine with massidentified mobility spectra, J. Chromatogr. 1976, 117, 327–336. 80. Carr, T.W., Negative ions in plasma chromatography-mass spectrometry, Anal. Chem. 1977, 49, 828–831. 81. Carr, T.W., Analysis of surface contaminants by plasma chromatography-mass spectroscopy, Thin Solid Films 1977, 45, 115–122. 82. Kim, S.H.; Betty, K.R.; Karasek, F.W., Plasma chromatography of benzene with mass-identified mobility spectra, Anal Chem. 1978, 50, 1784–1788. 83. Kim, S.H.; Betty, K.R.; Karasek, F.W., Mobility behavior and composition of hydrated positive reactant ions in plasma chromatography with nitrogen carrier gas, Anal Chem. 1978, 50, 2006–2016. 84. Proctor, C.J.; Todd, J.F.J.; Turner, R.B., Mass-analysed ion mobility studies of Nitrobenzene, Int. J. Mass Spectrom. Ion Proc. 1984, 60, 137–145. 85. Kim, S.H.; Spangler, G.E., Ion mobility spectrometry/mass spectrometry (IMS/ MS) of two structurally different ions having identical ion mass, Anal Chem. 1985, 57, 567–569.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_238... 06/05/2007

Página 1 de 1

Page 239 86. Karpas, Z.; Stimac, R.M.; Rappoport, Z., Differentiating between large isomers and derivation of structural information by ion mobility spectrometry mass spectrometry techniques, Int. J. Mass Spectrom. Ion Proc. 1988, 83, 163–175. 87. Karpas, Z.; Berant, Z.; Stimac, R.M., An Ion Mobility Spectrometry-Mass Spectrometry (IMS-MS) study of the site of protonation in anilines, Struct. Chem. 1990, 1, 201–204. 88. Spangler, G.E., Characterization of the ion-sampling pinhole interface for an ion mobility spectrometer/mass spectrometer system, Int. J. Mass Spectrom. 2001, 208, 169–191. 89. Eiceman, G.A.; Shoff, D.B.; Harden, C.S.; Snyder, A.P., Fragmentation of butyl acetate isomers in the drift region of an ion mobility spectrometer, Int. J. Mass Spectrom. Ion Proc. 1988, 85, 265–275. 90. Hill, C.A.; Thomas, C.L.P., A pulsed corona discharge switchable high resolution ion mobility spectrometer-mass spectrometer, Analyst 2003, 128, 55–60. 91. Wu C; Klasmeier, J.; Hill, H.H., Jr., Atmospheric pressure ion mobility spectrometry of protonated and sodiated peptides, Rapid Commun. Mass Spectrom. 1999, 13, 1138–1142. 92. Wu, C; Siems, W.F.; Asbury, G.R.; Hill, H.H., Jr., Electrospray ionization high-resolution ion mobility spectrometry-mass spectrometry, Anal. Chem. 1998, 70, 4929–4938. 93. Matz, L.M.; Hill, H.H., Separation of benzodiazepines by electrospray ionization ion mobility spectrometry-mass spectrometry, Anal Chim. Acta 2002, 457, 235–245. 94. Asbury, G.R.; Wu, C; Siems, W.F.; Hill, H.H., Separation and identification of some chemical warfare degradation products using electro-spray high resolution ion mobility spectrometry with mass-selected detection, Anal. Chim. Acta 2000, 404, 273–283. 95. Wu, C; Siems, W.F.; Hill, H.H., Jr., Secondary electrospray ionization ion mobility spectrometry/mass spectrometry of illicit drugs, Anal. Chem. 2000, 72, 396–403. 96. Dugourd, P.; Hudgins, R.R.; Clemmer, D.E.; Jarrold, M.F., High resolution ion mobility measurements, Rev. Sci. Instrum. 1997, 68, 1122–1129. 97. Hoaglund, C.S.; Valentine, S.J.; Sporleder, C.R.; Reilly, J.P.; Clemmer, D.E., Three-dimensional ion mobility/TOFMS analysis of electrosprayed biomolecules, Anal Chem. 1998, 70, 2236–2242. 98. Hoaglund, C.S.; Valentine, S.J.; Clemmer, D.E., An ion trap interface for ESIion mobility experiments, Anal. Chem. 1997, 69, 4156–4161. 99. Henderson, S.C.; Valentine, S.J.; Counterman, A.E.; Clemmer, D.E., ESI/ion trap/ion mobility/time-offlight mass spectrometry for rapid and sensitive analysis of biomolecular mixtures, Anal Chem. 1999, 71, 291–301. 100. Gillig, K.J.; Ruotolo, B.; Stone, E.G.; Russell, D.H.; Fuhrer, K.; Gonin, M.; Schultz, A.J., Coupling highpressure MALDI with ion mobility/orthogonal time of flight mass spectrometry, Anal Chem. 2000, 72, 3965–3871. 101. Guevremont, R.; Purves, R.W., High field asymmetric waveform ion mobility spectrometry-mass spectrometry: an investigation of leucine enkephalin ions produced by electrospray ionization, J. Am. Soc. Mass Spectrom. 1999,10(6), 492–501. 102. Purves, R.W.; Barnett, D.A.; Guevremont, R., Separation of protein conformers using electrospray-high field asymmetric waveform ion mobility spectrometry-mass spectrometry. Int. J. Mass Spectrom. 2000, 197, 163–177.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_239... 06/05/2007

Página 1 de 2

Page 240 103. McCooeye, M.A.; Mester, Z.; Ells, B.; Barnett, D.A.; Purves, R.W.; Guevremont, R., Quantitation of amphetamine, methamphetamine, and their methylenedioxy derivatives in urine by solid-phase microextraction coupled with electrospray ionization-high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal Chem. 2002, 74, 3071–3075. 104. Gabryelski, W.; Wu, F.; Froese, K.L., Comparison of high-field asymmetric waveform ion mobility spectrometry with GC methods in analysis of haloacetic acids in drinking water, Anal Chem. 2003, 75, 2478–2486. 105. Gabryelski, W.; Froese, K.L., Characterization of naphthenic acids by electro-spray ionization high-field asymmetric waveform ion mobility spectrometry mass spectrometry, Anal Chem. 2003, 75, 4612–4623. 106. McCooeye, M.; Ding, L.; Gardner, G.J.; Fraser, C.A.; Lam, J.; Sturgeon, R.E.; Mester, Z., Separation and quantitation of the stereoisomers of ephedra alkaloids in natural health products using flow injectionelectrospray ionization-high field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal Chem. 2003, 75, 2538–2542. 107. Guevremont, R.; Barnett, D.A.; Purves, R.W.; Vandermey, J., Analysis of a tryptic digest of pig hemoglobin using ESI-FAIMS-MS, Anal Chem. 2000, 72(19), 4577–4584. 108. Eiceman, G.A.; Nazarov, E.G.; Miller, R.A., A micro-machined ion mobility spectrometer-mass spectrometer, Int. J. Ion Mobility Spectrom. 2001, 3, 15–27. 109. Collins, D.C.; Lee, M.L., Developments in ion mobility spectrometry-mass spectrometry, Anal Bioanal. Chem. 2002, 372, 66–73.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_240... 06/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_240... 06/05/2007

Página 1 de 2

Page 241

Part III Applications of Ion Mobility Spectrometry

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_241... 06/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_241... 06/05/2007

Página 1 de 1

Page 243

6 Forensics, Military, Security, and Counterterrorism 6.1 Introduction The high sensitivity of ion mobility spectrometers (IMS) toward compounds that have high electronegativity (EN) or proton affinity (PA), fast response times, and man portability make IMS analyzers ideally suited for detecting trace amounts of certain chemicals. Among those are most of the common explosives that are highly electronegative due to the presence of nitro groups, several illicit drugs that contain amide groups and have a high proton affinity and most of the chemical warfare agents that are either organophosphorus derivatives (high PA) or contain halogen atoms (high EN). This potential for forensic applications was recognized early in tests of IMS on behalf of several U.S. government agencies for detection of personnel,1 explosives,2–4 and narcotics.4 Chemistry is central to these applications and discussions of the principles and technology may be found in Chapter 3 and Chapter 4. Specialized discussions or review articles have appeared regarding forensic applications of, IMS,5–8 and the large number of publications in forensic journals illustrate the prominent role of IMS and possible future uses in this area. The common features of these applications should be highlighted prior to any detailed discussion on the use of IMS technology in detecting chemical warfare agents (CWAs), explosives, narcotics, and other forensic-related substances. In all cases, the first stage of analysis is sampling with a need to introduce suitable amounts of analyte into the ionization region (just how much is enough will be discussed later). The second stage is identification or the need to ascertain that the target compound was correctly identified and to avoid false-positive and false-negative responses. A last stage that is particularly important for continuous monitoring applications is how to ensure that the IMS analyzer is still fully operational and not disabled through saturation of the reaction region or other malfunction. An assumption here is that those interested in detecting CWAs, explosives, or illicit drugs will be familiar with the chemical and physical properties of these

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_243... 06/05/2007

Página 1 de 1

Page 244 substances and do not need to see a schematic of molecular formulas, structures, popular names, and scientific nomenclature. Thus, only certain technical details that are relevant to detection by IMS are mentioned (i.e., vapor pressure, electron affinity, and so forth) in the following discussion. An exact number of IMS instruments deployed worldwide for these applications is difficult to obtain or verify. An estimate of the numbers of IMS analyzers, based upon open-source literature and discussions with researchers and manufacturers is 60,000 to 100,000 analyzers for the detection of CWAs, mainly for field use by NATO members, and 15,000 to 20,000 units for the detection of explosives (principally) and illicit drugs at major airports and other controlled areas. The sum total of all other applications combined is about two orders of magnitude lower than these, illustrating the importance of IMS for these measurements and the importance of these applications for the community of IMS researchers and users.

6.2 Chemical Weapons The favorable response of mobility spectrometers in the positive ion mode to organophosphorus compounds was reported from the 1970 until the 1990s.9–12 Since then, several groups have studied the response of specific instrument configurations to CWAs or their degradation products in the environment and to simulants of CWAs.13–25 In some of the studies, the IMS was combined with a chromatographic column to separate the complex mixture of compounds before they enter the drift tube.13,24 The studies of CWA degradation products in the environment15–21 involved preconcentration from liquid samples with a solid-phase microextraction or SPME fiber,21 electrospray ionization with a high-resolution IMS drift tube,19–20 or even direct aspiration of the liquid sample.16–18 Nerve agents designed for use in chemical warfare are mostly derivatives of organophosphorus compounds with high proton affinities. Thus, the sensitivity and specificity of response and the high speed of measurement make mobility spectrometers superb instruments for infield or on-site determinations. Additionally, blister agents and related breakdown products can be detected in the negative polarity. Positive gas-phase ion chemistry governs the reactions of nerve agents or simulants such as DMMP (dimethylmethyl phosphonate) in the IMS ionization source (see Chapter 3). The main reaction pathway between the reactant ion, R2H+(H2O)n, and an organophosphorous molecule, G, involves displacement of an R with G to form a heterogeneous proton bound dimer, GRH+(H2O)n−1, as shown in Equation 6–1. This could be followed by the formation of a homogeneous proton bound dimer, G2H+(H2O)n−1, as shown in Equation 6–2 if the concentration of G is increased. Proton transfer: (6–1)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_244... 06/05/2007

Página 1 de 1

Page 245 Dimerization: (6–2) In order to control the ionization processes and reduce interfering reactions, a reagent gas that forms well-defined reactant ions may be added. Thus, R might be one of several chemicals including ammonia or acetone.26 In principle, the ion G2H+(H2O)n−1 will be formed if the association between G and H+ is strong.27 Furthermore, proton-bound trimers, G3H+(H2O)n, and higher clustered ions can form in the reaction region where ions and sample neutrals mix. When such ions are extracted from the reaction region into drift region, largely free of sample neutrals, decomposition of trimer ions is rapid. Such ions exhibit lifetimes too short to be measured in the drift tube at ambient temperature. Blister agents that generally contain halogen atoms, such as mustard gas (HD), have a low proton affinity and, therefore, do not form stable positive ions at ambient pressure. However, these chemicals (termed H agents in military codes but called B here) can form an adduct ion, i.e., B*O2− (H2O)n−1 in the negative polarity as shown in Equation 6–3: (6–3) Elevated moisture levels may interfere with this reaction and enhance the reverse reaction of Equation 6–3, namely, the dissociation of the adduct ion. Programs to develop IMS-based detectors and monitors for CWAs were undertaken with the support of defense establishments in several governments, notably the U.K. and U.S. Research toward the development of an automatic chemical agent detector alarm (ACADA) was begun by Bendix Corp., later Environmental Technology Group (ETG) in the U.S.,28 and the result has been placed into service through Smiths Detection as the GID-3™ (Figure 6–1). The intention for this analyzer is to continuously monitor the presence of CWAs in ambient air and sound the alarm when threshold values are exceeded. The GID-3 and all other IMS-based analyzers are regarded as point sensors and cannot determine chemical compositions at distances greater than a few meters in the absence of breezes. A handheld chemical agent monitor (CAM) was successfully produced by Graseby Dynamics, Ltd.29 in the U.K. for use in postattack scenarios and was intended for an individual to screen surfaces, equipment, and personnel before further contact. The response of the C AM would then guide decisions regarding handling and decontamination of the object or person. The CAM analyzer has been regularly upgraded, and an improved model is now deployed as the i-CAM (Figure 6–2). Handheld IMS analyzers for use in battlefield environments were developed also in Germany by Bruker-Saxonia as the RAID series and the IMS200030 as shown in Figure 6–3. A general resemblance exists between these handheld analyzers though technical

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_245... 06/05/2007

Página 1 de 1

Page 246

FIGURE 6–1 Photographs of the GID-3 (Smiths Detection, U.K.) mobility spectrometer for continuous monitoring of airborne vapors for specific detection of chemical warfare agents.

specifications differ slightly. In Finland, Environics Oy produced a series of mobility-based analyzers with the aspirator designs for CWA monitoring, including the IMCELL™ MGD-1.17 Finally, a pocket-sized CWA monitor for individual soldiers (Figure 6–4) has been developed by Smith’s Detection and deployed with the British armed forces. All of these instruments were designed for users who have limited training and experience with analytical measurement. Moreover, the conditions of use could be hostile and demanding. Thus, the analyzers had to be rugged, simple to operate, and practically maintenance free for long periods. In practice, a soldier challenges the analyzer with a simultant (a confidence tester) before screening for CWAs in order to validate the proper operation and performance of the instrument. Over 20 years ago it was found that the sensitivity and specificity of the earlier generation of instruments toward nerve agents could be enhanced by the use of acetone-based reagent ion chemistry26 In the positive ion mode, the acetone molecules readily protonate and associate to produce protonated

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_246... 06/05/2007

Página 1 de 1

Page 247

FIGURE 6–2 Photograph of the improved handheld chemical agent monitor or iCAM. These instruments were developed by Graseby Dynamics Ltd., now Smiths Detection, in Watford, Herts, U.K., and were the first rugged handheld mobility spectrometers. Over 50,000 CAMs have been produced for the armed forces of several nations, including the U.S.

FIGURE 6–3 Photograph of the handheld IMS2000 RAID produced by Bruker-Saxonia in Leipzig, Germany

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_247... 06/05/2007

Página 1 de 1

Page 248

FIGURE 6–4 Two views of miniaturized mobility spectrometers. A model of a lightweight chemical agent detector or LCAD is shown and was developed for the U.S. Army. A variant, the LCD, was developed for use by the U.K. armed forces. Both are from Smiths Detection.

acetone monomers and dimers, thus effectively eliminating many interferents, while not losing response to organophosphorus nerve agents that have proton affinities above that of acetone31 (196 kcal/mol). Acetone does not significantly affect negative ion chemistry and does not prevent blister agent detection with a drift tube containing acetone as a reagent gas. The exact data on the sensitivity of the military units for nerve and blister agents are not available from research articles or objective studies, apart from compliance with established NATO requirements. However, in some brochures of the manufacturers of IMS equipment the performance is listed (Table 6–1). For

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_248... 06/05/2007

Página 1 de 1

Page 249 TABLE 6–1 Chemical Warfare Agents

Name

MDL

Ko(−)

Ko(+)

ng/l

cm2/Vs

cm2/Vs

Reference

GA

5 1.52, 1.76, 2.44

1.51, 1.06

Leipzig Report, 1990*

GB

5

1.62, 1.22

Leipzig Report, 1990*

GD

5

1.51, 1.06

Leipzig Report, 1990*

VX

5

1.23

Leipzig Report, 1990*

HD

20 1.55, 2.40, 2.73

HN

20 1.33

Leipzig Report, 1990* 1.47

Leipzig Report, 1990*

1.84

Leipzig Report, 1990*

CG

2000 1.76, 1.90, 2.58

AC

2.44, 2.98

Leipzig Report, 1990*

CK

2.12, 2.44, 2.58

Leipzig Report, 1990*

G, V

5

RAID-16

H, L

20

RAID-16

AC

200

RAID-16

Note: Data obtained from *Ionenbeweglichkeitsspektrometrie zur Gasspurenanalyse, ZFI-Mitteilungen Nr. 154, Berichte des Zentralinstituts fuer Isotopen- und Strahlenforschung, Akademie der Wissenschaften der DDR, Leipzig, April 1990; and from brochure of Bruker RAID series.

example, in the brochure of Bruker’s handheld IMS2000, the range for the nerve agents GB and GA is given as 20 to 600 μg/m3. In the listing of the minimal detectable limits (MDL) of the civilian GPIMS and FP-IMS instruments made formerly by ETG, Inc., 5 and 10 ppb are given as the MDLs for nerve and blister agents, respectively. Information about field tests of these instruments is also not available to the general public. The high sensitivity of IMS devices toward CWAs is true also for the nontraditional mobilitybased analyzers such as the field-dependent mobility analyzers or differential mobility spectrometers (see Chapter 4). For example, MDLs for GB and phosgene were reported as 8 and 4 parts-per-trillion (by volume), respectively, by the Field Ion Spectrometer formerly marketed by Mine Safety Appliance, Inc.32 Deployment of IMS analyzers on unmanned airborne vehicles (UAV) for detection of clouds of CWAs has also been demonstrated with simulants.33 When a point sensor is made mobile, the capabilities of the analyzer are augmented by providing chemical determinations at points within a large area.

6.3 Detection of Explosives by IMS 6.3.1 General Comments on Detection of Explosives The past decade may be characterized by an intense research effort to improve the sensitivity and reliability of IMS instruments and to expand the list of detectable explosive substances. This has occurred due to the rising need for rapid, efficient, and reliable detectors of explosives that are

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_249... 06/05/2007

Página 1 de 1

Page 250 used in acts of terrorism worldwide. Several commercial IMS-based analyzers are competing for a share in this rapidly growing market segment. Thanks to the high sensitivity of IMS technology toward several types of explosives, applications have been made in detecting hidden explosives and identifying suspect materials for postdetonation investigations and even for detecting explosives in water and underwater. Common explosives consist of nitro compounds, which are highly electronegative and form stable negative ions under conditions of ambient pressure. In fact, the general public is likely to encounter an IMS device, often unawares, as airline passengers being screened for explosives. Several manufacturers are very active in this field, particularly after the attack on the Twin Towers at the World Trade Center in New York City on September 11, 2001. After this event, awareness of global terrorism increased, although that incident had no direct relationship with nitro-organic explosives and no explosive detector would have prevented the attack. In the development of IMS analyzers for explosives detection, stages of technology can be identified. A burst in activity occurred with government contracts after the initial discovery of the capabilities of IMS, and the emphasis was on detecting volatile compounds such as EGDN, nitroglycerine, and TNT. These are the most volatile explosive compounds or related chemicals in the family of nitrated organic compounds as listed in Table 6–2. Gradually, the threat to commercial aviation safety from explosives with lower vapor pressures such as RDX, PETN, HMX, and their composites was understood and techniques for sampling such compounds with heated inlet systems to transfer sample to the analyzer were developed towards the end of the 1980s (see Chapter 4). However, these analyzers were suited only for screening inanimate items such as hand luggage and cargo. In recent years, portals have been developed for rapid screening of humans at controlled entry points such as airport boarding gates, secure buildings, or restricted zones. Naturally, the throughput on analysis must be high and the technology TABLE 6–2 Molecular Weight and Vapor Pressure of Common Explosives

Name

Class

Molecular Weight

Vapor Pressure (torr)

EGDN—Ethyleneglycol dinitrate

Aliphatic

152

4.8×10−2

NG—Nitroglycerin

Aliphatic

227

2.3×10−4

DNT—Dinitrotoluene

Aromatic

182

1.1×10−4

TNT—Trinitrotoluene

Aromatic

227

4.5×10−6

RDX—Cyclonite

Cycloaliphatic

222

1.1×10−9

PETN—Pentaerythritol tetranitrate

Aliphatic

316

3.8×10−10

Source: Conrad, F.J., Explosives detection—The problem and prospects, Nucl. Mater. Manage. 13, 212,1984.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_250... 06/05/2007

Página 1 de 1

Page 251

FIGURE 6–5 Summary of negative polarity ionization chemistry for explosives: product ions can arise though one of four reaction channels that arise from the initial product ion, which is a complex cluster ion (center). Product ions may include a molecular ion (M−, top left); an adduct ion with O2− (top right); a proton-abstracted ion (bottom right) or a fragment such as the nitrate ion (bottom left).

must be seen as noninvasive with a balance between public safety and individual privacy. The ionization chemistry for explosives is governed by reactions in the negative polarity, and ions formed in the reaction region for molecules of explosives are shown in Figure 6–5. The reactions between a reactant ion, such as O2−, and electronegative molecules may involve charge transfer from the reactant ion (Equation 6–4) to the neutral molecule, which may then dissociate and form a more stable fragment ion, F−: Charge transfer (6–4) Other reaction channels involve proton abstraction from the analyte molecule, which is essentially proton transfer to the reactant ion (Equation 6–5), forming an (M–H)− ion. Proton abstraction (6–5)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Page_25...

06/05/2007

Página 1 de 1

Page 252

FIGURE 6–6 The Vapor Tracer, a handheld explosive analyzer (made by GE Interlogix).

More common are association or ion attachment reactions (similar to Equation 6–3) that can be followed by dimerization or clustering (Equation 6–6) with increased concentrations of M and collisional stabilization: Dimerization (6.6) Charge transfer and proton abstraction reactions have been reported only in a few cases, such as the reaction of O2− with TNT in air.34–37 Often several association and attachment reaction channels occur simultaneously and compete with one another so that the explosive molecules attach to the different ions that are present in the ionization chamber, forming several different negative product ions. Commonly, the ionization process is controlled by adding a reagent gas, generally chloride or bromide ions, which then forms well-defined product ions, as proposed by several investigators,38–41 although in some cases improved results were obtained without the chloride reagent.42 In other pathways, fragment ions such as NO2− or NO3− are formed through dissociative charge transfer and may then associate with another parent molecule. Though any distribution of charge among several ions damages the limits of detection, a variety of product ions can be helpful in confirming the presence of explosives and reducing the rate of false-positive responses. An important consideration with the detection of explosives is elimination of false negatives. Although a false-negative response for detecting drugs may not directly affect the lives of those involved, failure to detect contraband explosives may actually endanger the safety or lives of hundreds of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_252... 06/05/2007

Página 1 de 1

Page 253 people, such as passengers in an airplane or train. Therefore, the performance criteria placed upon explosive detection systems are demanding. Explosive detection methods with high reliability must have a very low false-positive rate (below 5%) and a practically zero false-negative rate. In addition, in order to be certified by the Federal Aviation Administration (FAA), the system should be sensitive and fast enough to detect 2 lb or less of explosives in or under 6 sec. Such requirements for an explosive detector were summarized by Lucero.43 The use of volatile compounds as taggants to enable detection of explosives by IMS was investigated in the late 1970s44 and more recently.45 Although the advantages of tagged explosives for counterterrorism are self evident, the proposal to tag all explosives has not been accepted by all manufacturers of explosives (and probably will not be accepted in the future). In any case, there are available stocks of untagged explosives, and reliance solely upon taggants is impractical. Although luggage may be examined for explosives by radiation techniques based on neutron activation or bombardment of the object with high-energy x-rays, people must be screened by methods that are harmless and nonintrusive. These are mainly techniques based on sniffing or gentle sampling of particles attached to the subjects’ clothing or body, e.g., in walk-through portals and by handheld instruments. The detection of relatively volatile explosives, such as EGDN, NG, or even TNT, the last having a vapor pressure above 10−6 torr at 25°C (see Table 6–2), may be done by sampling vapors or particulate matter from the subject or suspect piece of luggage. However, detection of plastic explosives such as RDX or PETN, which have room temperature vapor pressures below 10−8 torr, or of well-sealed volatile explosives, requires a different sampling strategy. This could be preconcentration of the vapors or collection of microparticles containing the adsorbed explosive vapors before transfer into the detector for heating and analysis. Thus, the first barrier for detecting explosives is that of sampling—bringing enough of the compound into the IMS. One approach for screening people suspected of handling explosives or suspect objects was recently presented.46 A gas jet entrains particles and carries them to a Teflon filter, from which vapors are aspirated into an IMS for analysis. In addition to detection of explosives, handheld IMS devices may be applied to postdetonation identification of the type of explosive used. This can be done on-site with the IMS serving as a qualitative analytical instrument, or using IMS to screen the debris and select the pieces with traces of the explosive for more detailed and precise analysis at the forensic laboratory.47–49 The detection of traces of explosives as pollutants in the environment are described in Chapter 8.

6.3.2 Measurement with Handheld Devices, Portable Instruments, and Portals Handheld IMS devices for detecting explosives can be used much like handheld metal detectors to screen people entering a controlled area by passing

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_253... 06/05/2007

Página 1 de 1

Page 254 the device close to the subjects’ bodies, clothes, and belongings without actually touching the subject. These devices may also be used by security patrols to examine suspect objects and verify “sterile zones” in public places. The efficiency of this approach is limited to detection of unwrapped volatile explosives or to traces of these on the exterior of the suspect object, but has the advantage of deterring terrorists and instilling a feeling of security among the public. In addition, such devices may be readily rushed to the scene to enhance security measures. Some commercial devices of this kind include the GE-Interlogix VaporTracer (shown in Figure 6–6) and the Smiths Detection Sabre 2000. Desktop instruments are usually stationed at control points such as airport boarding gates and rely on human operators who use swipes or air suction techniques to trap particles on a filter that is then inserted into the inlet of the analyzer and placed on the anvil heater unit. Sampling of this type is only marginally intrusive as it requires gentle physical contact with the suspect subject or object. These instruments are often combined with, or positioned adjacent to, an x-ray machine so that the operator may be able to immediately test any object that appears as suspicious in the imaging device. Though larger than the handheld analyzer, these explosive detectors are portable and may be moved or relocated conveniently to a controlled area for additional capability in screening. Two of these portable analyzers are shown in Figure 6–7. Portals for explosive detection should be like those used for metal detection: screen people rapidly and with minimal intrusiveness as they pass through a portal.50–53 Portals have the capability to draw an air sample from a person’s whole body, preconcentrate the vapors, and perform an analysis by IMS within a few seconds. Portals may also employ jets of air and gentle physical contact to detach particles, trap them on a filter, and thus enhance the sensitivity of the detector. In principle, a security checkpoint with such a portal could be designed to lock and trap a would-be bomber and limit the damage that may be inflicted by a suicide bomber. An automatic portal for luggage, which can possibly be adapted for inspection of people, is described in the next section. Interestingly, two opposing approaches to sampling subjects in walkthrough portals have been developed. In the portal of GE-Interlogix,53 streams of air directed against a body are drawn through ports at the top of the portal into a preconcentrator. This design was based mainly on the work of Gowadia and Settles54 and their mapping of the natural aerodynamic portal sampling. In a second design by Sandia National Laoratory now licensed by Smiths Detection,50 streams of air are drawn downward to the bottom of the portal into a preconcentrator and IMS detector (Figure 6–8). A more detailed technical description is given in the following text.

6.3.3 Research and Operational Experience In this section, some of the operational experiences and recent developments for explosives detection will be discussed. Most of the data in this section

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_254... 06/05/2007

Página 1 de 1

Page 255

FIGURE 6–7 Portable explosive analyzers: the Itemizer made by GE Interlogix (top) and the IONSCAN made by Smiths Detection (bottom).

was published by law enforcement agencies or supplied by manufacturers of commercial IMS devices. Thus, there is an emphasis on practical aspects and field tests. The descriptions are based mainly on the manufacturers’ claims and brochures, except where objective reports on comparative evaluations and field tests are available. The common or conventional explosives that contain nitro functional groups can be divided into two categories: relatively volatile substances such as nitroglycerine and EGDN, and low vapor pressure materials such as RDX, PETN, and HMX. There are several commercial and military composites that are mixtures of two or more of the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_255... 06/05/2007

Página 1 de 1

Page 256

FIGURE 6–8 Sentinel II portal for screening humans for explosives residues: streams of air are used to sample the subject’s body and the air is drawn through ports at the bottom of the portal into a pre-concentrator and IMS detector (Smiths Detection).

above, such as Composite B and C, and Semtex. Recently, new types of alternative explosives, including homemade compounds that can be easily prepared from available materials, have been involved in acts of terrorism. Among those are ammonium nitrate mixed with fuel oil (ANFO) that was used in the Oklahoma City bombing, triacetone triperoxide (TATP) and urea nitrate that have been used principally by Palestinians and others. The common explosives mentioned in the preceding text can be readily detected by IMS operated in the negative mode, with subnanogram detection limits in laboratory testing reported by several investigators and manufacturers. Most commercial instruments for detection of these substances are operated at elevated temperatures (from ~100 to 200°C), make use of a doping agent to control the reactant-ion chemistry, and can be used for detection of vapors (“sniffing”) or trapped particles with a filter insertion inlet. Under these conditions, the major mechanisms for product ion formation are clusterization and dimerization (Equation 6–2 and Equation 6–6). As identification of the explosives is based on the reduced mobility of the ions in the spectrum, the dopant affects the product ions and is therefore a consideration in detection algorithms. Several of the older studies on the suitability of IMS for the detection and identification of explosives arising early in the development of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_256... 06/05/2007

Página 1 de 1

Page 257 modern analytical IMS, have been summarized by Karpas.5 Some of these studies were concerned with the ion chemistry of explosive vapors, although attempts to determine minimum detectable limits were made in others. A word of caution is needed when discussing or conducting quantitative measurements with explosives, particularly in the vapor phase. Most explosive substances are highly polar molecules that tend to absorb on surfaces or undergo thermal decomposition.54 Thus, there may be a significant difference between the amount of vapor released from the explosives source,54–56 say a heated permeation tube, and the amount actually reaching the analyzer. Consequently, the actual concentrations of explosives detected may be lower than that calculated on the basis of weight loss from a vapor generator. Later, the missing material or decomposition products may reappear when displaced or desorbed from the surfaces of the generator or the analyzer. Quantitative calibration of the IMS response may be determined best by depositing a known amount of the substance (usually from a calibrated solution of the explosive in a volatile organic solvent) on a suitable filter and desorbing the sample directly into the IMS drift tube from the heated inlet. 6.3.3.1 Walk-Through Portals and Systems for Luggage Screening Two of the major manufacturers of IMS equipment, Smiths Detection (formerly Barringer Research) and GE-Interlogix (formerly Ion Track Instruments), are actively developing walk-through portals for humans or for automatic screening of luggage. Barringer Research had produced a series of portable IMS-based instruments for detection of drugs and explosives, known as the IONSCAN, and a handheld analyzer. Similarly, Ion Track Instrument had built an instrument family based on IMS to meet the same requirements for aviation safety. The core technology for these instruments are now integrated into luggage screening systems57 and portals for human screening.58 A few articles and technical reports on screening uses for IMS analyzers have been made available and disclose some of the features of analysis. Only some of the more significant facets will be discussed here. In routine operation, a powerful vacuum cleaner collects microscopic particles that are then trapped on a filter. The filter is inserted into a heater connected directly to the IMS and desorbed vapors are transported by a carrier gas into the drift tube (see Figure 4–6 in Chapter 4). Detection limits of 50 to 350 pg were claimed for most common explosives, including plastic explosives. With the support of the Federal Aviation Administration, an automatic trace detection system for the detection of explosives’ vapors and particles in luggage had been developed.57 The system consists of a conveyor belt that carries the luggage into a tunnel where particles are picked up by an array of 225 flexible sampling tubes and carried to a preconcentrator unit. A coated mesh vapor concentrator further increases the concentration of the vapors that are then analyzed by an IONSCAN. Meanwhile, the vapors given off by the luggage are collected by another preconcentrator unit and released into a second set of coated mesh concentrators and an IMS instrument. The drift tube and inlet temperatures

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_257... 06/05/2007

Página 1 de 1

Page 258 for vapor detection were set at 60 and 110°C, respectively, and for particle detection at 115 and 245° C, respectively. The overall time for each piece of luggage was 20 sec: 7 sec for sampling, 3 sec for desorption and rotation of the table, and 10 sec for particle desorption and analysis. However, the actual throughput is higher as the next bag enters the tunnel while the first one is being sampled, and then two more bags follow while the analysis of the first two is being carried out. Thus, the throughput of the system was reported as 720 bags an hour, with the IMS performing six analyses per minute.57 A Large Volume (LV) IMS analyzer was produced specifically for detecting explosives and tested at Sandia Laboratories. The results are summarized in a technical report.50 The system tested was comprised of a walk-through portal equipped with a purge gas flowing at rates of up to 2300 1/min and a large-diameter (with a cross section of 100 cm2) IMS drift tube capable of taking a carrier gas flow of 16.61/min (two orders of magnitude higher than regular IMS drift tubes). The main findings in this study were that 0.5 g of TNT placed in a plastic bag inside the portal and mixtures of TNTRDX (composition B) could be detected. However, pure RDX or PETN were not detected. Limits of detection were found to be close to those of a conventional IMS with preconcentration. In a recent study, 17 of the most common suspected interferents for detection of TNT by IMS technology in airport scenarios were investigated.59 The reactant ion was chloride, the most common reactant ion in explosives detectors. Ten of the interferents showed no IMS response, and of the other seven, only two presented a problem: 4,6-dinitro-o-cresol had a close reduced mobility value and 2,4-dinitrophenol competed with ionization of TNT. 6.3.3.2 Homemade and Alternate Explosives As mentioned earlier, new types of homemade or alternates to common nitrated explosives have appeared in several crime scenes and pose a threat to public safety. In order to counteract these threats, instrumentation has to be calibrated, data systems have to be revised, and some modifications in operational parameters may be necessary. GE-Interlogix has characterized operational conditions for the detection of several alternative explosives without severe deterioration of performance for the standard explosives. The temperature of the drift tube was lowered to 169°C and the desorber was set to 220°C. In addition, the instrument was operated in dual mode, i.e., measurement of the negative and positive mobility spectra almost simultaneously. With these modifications, operational detection limits (ODL) defined by the authors as the mass of target substance that was required to produce an alarm at a given detection threshold of 50 to 100 ng were established for TATP, ammonium nitrate, and black gunpowder. Under these conditions, the response and detection limit for TNT is degraded by a factor of two, but those for RDX and NG are barely affected; the detection limits for PETN was improved.60 Recently, TATP was investigated by IMS and tandem mass spectrometry, which was used to identify the peaks in the mobility spectrum. A peak at mass 223 Da was identified as arising from

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_258... 06/05/2007

Página 1 de 1

Page 259 TATR61 In the mobility spectrum, a cluster of three peaks was observed, and a large increase in peak intensity was observed after dissolution of TATP in toluene. The reduced mobility of the main peak was reported as 2.71 cm2/V sec. This high mobility value must correspond to an ion with low molecular weight and not to the molecular ion. Two other studies concerned with the detection of improvised explosives in positive and negative polarity were presented at ISIMS meetings in 2001 and 2003.62,63 In the former, response of the IMS to TATP was found in both polarities, whereas in the latter study ions arising from TATP were found only in positive mode operation. The humidity level, that had to kept below 100 ppmv, affected the detection of TATP and HMTD, the peroxide compounds.63 6.3.3.3 Database for Explosives In Table 6–3, the reduced mobility values of most of the common explosives are shown. There are several different reduced mobility values for some of the compounds associated with unique product ions, which were not always identified by mass spectrometry. Clustering of target molecules with ions, as shown in the reaction in Equation 6–6, plays an important role in explosive detection. A comprehensive study showed that the prominent ion was an adduct or cluster between the molecule and reactant ions.38 These association processes were strongly temperature dependent and were abundant at a relatively low temperature (50°C).

6.4 Drugs 6.4.1 Introduction and Ion Chemistry The potential for detecting contraband drugs, especially heroin and cocaine, by IMS was recognized in the early 1970s.4 The actual application of IMS technology for detection and identification of illicit drugs can be divided into two parts: one is the detection of hidden contraband material, such as heroin and cocaine, usually in luggage, containers, and illicit laboratories; and the other is screening of suspect substances and identification of illicit materials as opposed to legitimate substances. This section will focus on the properties and ionization chemistry of such substances in air at ambient pressure, which is important for correct identification, on the sampling methods used to transport enough molecules from the suspected object into the IMS, and on the procedures that were developed for drug bioassay (mainly skin, hair, and urine). The common illicit drugs can be divided into three main groups: opiates or narcotic substances such as heroin and cocaine, amphetamines (benzodiazopines) and their derivatives, and all other substances. Several of the common substances used as illicit drugs consist of nitrogen compounds,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_259... 06/05/2007

Página 1 de 2

Page 260 TABLE 6–3 The Reduced Mobility Values of Common Explosives

Ion Identity Name

(assumed)

Ko cm2/Vs

MDL

Parameters

Nitro-aromatics MNT

1.74

air, 166°C

1.81 2.40 2,4 DNT

1.68, 2.10

air, 200°C

3,4 DNT

1.54

air, 50°C

2,6 DNT

1.67

air, 250°C

200 pg 1.45

air, 166°C

TNT

1.49 1.54 1.59 Nitro-aliphatics Dynamite NG

NG·Cl−, NG·NO3−

2.10, 2.48

air, 200°C

50 pg 1.32, 1.34

air, 150°C

200 pg 1.28 EGMN

NO3−

2.46

air, 150°C

EGDN

NO3−

2.46

air, 150°C

1.30, 1.25

air, 250°C

RDX

200 pg 1.48 1.39

air, 250°C

RDX·Cl−, RDX·NO3−

800 pg 1.31

Low vapor pressure HMX RDX

(RDX)2C1− PETN

1000 pg 0.95 1.48 1.21

PETN PETN(-H) PETN·Cl− PETN·NO3−

air, 166°C

80 ng 1.145 200 pg 1.10 1000 pg

Tetryl

1.45 1.62

air, 250°C

Comp B

1.57, 1.70, 1.81

air, 200°C

Improvised TATP

TATP·NH4+

HMTD

HMTD·H+

1.50

100–130°C

Black gunpowder

N(CH2O)3H·C1−

1.88

150°C

~μg 1.36

100°C (Marr)

Note: Air—the drift gas, ionization source flow, and the sample vapors were carried by air. Cl−—chloride reactant ion present.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_260... 06/05/2007

Página 2 de 2

Source: Fetterolf, D.D.; Clark, T.D., Detection of trace explosive evidence by ion mobility spectrometry, J. Forens. Sci. 1993, 38, 28–39 (Barringer IONSCAN 200); Marr, A.J.; Groves, D.M., Ion mobility spectrometry of peroxide explosives TATP and HMTD, ISIMS 2003, 6.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_260... 06/05/2007

Página 1 de 1

Page 261 mainly amides, that have a high proton affinity. Such compounds form stable positive ions by proton transfer reactions as shown in Equation 6–7: (6–7) Karasek et al. reported IMS/MS measurements of heroin and cocaine64 and they found that the major ions formed from heroin were the molecular ion (M+), a quasimolecular ion (M-H2)+, and a fragment ion (M–CH3CO2)+. In cocaine, the molecular ion (M)+ and two fragment ions (M– C6H5CO2)+ and (M–C6H5CO2–CH3CO2)+ were the dominant ions in the positive ion mobility spectra.64 The response in positive mode and negative mode to several types of barbiturates has been reported,65 and later a Fourier transform IMS with capillary column GC was employed for the determination of barbiturates.66

6.4.2 Laboratory and Field Applications Lawrence and coworkers at the Canadian National Research Council (NRC) made significant advances in drug detection by IMS in the mid-1980s by developing new sampling methods and applications.67–73 Simulated field tests using a detachable sampling cartridge involved searching mail, luggage, and personnel for drugs.67 It was reported that some potential interferents, including headspace volatiles of coffee and tea, did not affect detection of target drugs. In another test, letters spiked with narcotics were distinguished from negative controls. Sampling close to the zipper of a spiked suitcase, the hands and pockets of an individual who had handled drugs gave distinct signals. However, in these tests, sampling the suitcase externally and away from the zipper did not give a positive response. The most significant results of this test were that false alarms were not received from innocent items and that near real-time performance for simultaneous detection of cocaine and heroin was achieved. A similar test to search for heroin and cocaine in various customs scenarios, such as letter mail and cargo containers, was described by Chauhan et al. of the Customs and Excise Authority of Canada.74 Four out of the 339 letters examined contained drugs, and all were correctly identified by IMS. Two false-positive results were reported. Both false-positives were for heroin, one of which was found, after GC/MS study, to contain a morphine-based tar. Of 18 containers tested for cocaine, the IMS gave correct responses in 13 instances and showed 5 false-negatives. In addition to customs and mail scenarios, a method for detecting and identifying drug residues on the skin of subjects was developed. Hand swabs of subjects were obtained and tested by directly inserting the sampling needle into the IMS inlet.69,70 Several prescription and illicit drugs were successfully detected and identified by this method. The use of hand (fingers and palm) and nostril swabbing for noninvasive preliminary screening of patients arriving with drug overdoses at hospital emergency rooms was described.70 The reported rate of correctly identifying the illicit or

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_261... 06/05/2007

Página 1 de 1

Page 262 prescription drug (as confirmed later by laboratory tests) was as high as 53% for patients being tested by IMS within less than 30 min of arrival at the emergency room. This is quite an accomplishment for such a simple testing procedure, considering the variety of substances measured and that several of the prescription drugs were actually coated tablets or capsules, thus leaving little or no traces of the active substance on the subject’s hands. Eiceman et al. demonstrated that IMS can be used to verify the content of some common over-the-counter pharmaceuticals.75 The tablet or capsule was slightly warmed and headspace vapors were sampled by a handheld IMS; mobility spectra were compared with known spectra to confirm the prescription. The use of cartridges or filters with high volumetric sampling rates appears to be the most effective method for drug detection. Due to the low vapor pressure of most contraband drugs, detection must be based on the entrapment of drug microparticles and their evaporation at an elevated temperature in the analyzer inlet. This approach has been adopted by Barringer Research in the IONSCAN analyzer. Preliminary results on drug detection in field tests were described earlier and presented by Fetterolf et al. of the FBI.76 Detection limits of nanograms were reported for several drugs and reduced mobility values were measured. Cocaine residues on hands reportedly could be detected 1.5 h after a brief contact with the substance, and that simply washing the hands failed to completely remove the traces of the drug. Residues of narcotics were detected in a quick screening of bookkeeping records of drug dealers.77 Field experiences described by personnel of the U.S. Coast Guard77,78 and agents of the Drug Enforcement Administration (DEA)79 have illustrated the difficulty in collecting good swipe samples in unfavorable conditions. Complications arise from surfaces covered with moisture, oil stains, or paint residues. Tests were made to select the swipe material that gave the best results, and methods were developed for gradual desorption of the suspect material. A handheld explosives monitor was evaluated for extended at-sea deployment.78 The Sabre 2000 was compared to an IONSCAN B for false-negative and false-positive rates for illicit drugs including heroin and cocaine. The limit of detection for heroin was determined from response curves as 20 ng with close agreements with three analyzers. In contrast, the IONSCAN showed 2-ng limits of detection. The false-positive alarm rate was ~1%, whereas the false-negative rate was ~50% for quantities below 250 ng, ~25% for 250 to 500 ng, and ~10% for more than 500 ng.78 Keller et al. extended the subject of bioassays for drug testing of skin as described in the preceding text and developed special procedures for treating hair samples collected from suspects in order to test them by IMS.80–85 After rinsing the hair sample to remove external contamination, the sample was digested by a methanolic alkaline solution. A 0.05-ml portion of the digest was applied to a membrane filter and dried before detection of designer drugs82 and metamphetamines.80,81 Exotic substances that cause hallucinations, such as psychedelic fungi that contain psilocybin and psilocin, were identified using mobility spectra.83 Quantitative analysis was

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_262... 06/05/2007

Página 1 de 1

Page 263 performed by GC/MS after a simple one-step extraction involving homo-genization of the dried fruit bodies of fungi in chloroform followed by derivatization.83 Solid-phase microextraction (SPME) has been used to preconcentrate the analyte (narcotic drugs) from headspace vapors.86 This approach was novel and combined sample preparation by SPME with sample ionization electrospray. Field-dependent mobility analyzers and mass spectrometric identification were used to detect amphetamine and derivatives87 morphine and codeine88 in human urine. Matz and Hill used an ESI/IMS/MS instrument to study charge competition within the ESI during analysis of amphetamines89 and to test for benzodiazepines.90 In real-life situations, cigarette smoke and the nicotine derivatives it contains may interfere with the detection of illicit drugs, particularly amphetamines. Advanced signal processing methods that use the dynamic data generated by the IMS were developed by Harrington’s group to overcome this problem.91,92

6.4.3 Database for Drugs A database for the reduced mobilities of prescription and illicit drugs is presented in Table 6–4. Differences in the reported reduced mobility values of the same compound by different workers arise mainly from the use of different experimental conditions in the measurement. While agreement is generally good in comparisons of Ko values, differences may arise from experimental conditions where the product ion is slightly changed. These changes are noticable in some instances (Table 6– 4), though good agreement between measurements made far apart in terms of time and space and by different instruments was also seen.

6.5 Other Forensic Applications Ion mobility spectrometry has matured to a high level of technology and methodology for the chemical determination of explosives, chemical weapons, and drugs as demonstrated in the discussion of the preceding sections. However, other somewhat narrow and limited uses of IMS analyzers exist in forensics and security. These are derivative interests of those mentioned earlier but merit individual description.

6.5.1 Lachrymators Lachrymators are chemicals that cause irritation of mucosal membranes and result in streaming tears and a running nose, as well as difficulties in functioning normally. Therefore, such chemicals have gained popularity among law enforcement forces trying to control unruly crowds or disperse demonstrators with less than lethal means. People wishing to defend themselves

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_263... 06/05/2007

Página 1 de 2

Page 264 TABLE 6–4 A Database for the Reduced Mobilities of Prescription and Illicit Drugs and Pharmaceuticals

Ko cm2/Vs

Name

Parameters

Acetaminophen

1.70, 1.76, 1.97

Air, 220°C, skin/cart

N-Acetylamphetamine

1.53

Air, 220°C, sol/wire

Acetylcodeine

1.09, 1.21

Air, 220°C, sol/wire

Alprazolan

1.15

Air, 220°C, sol/wire

Amitryptiline

1.19

Air, 220°C, skin/cart

Amobarbital

1.36, 1.53

N2, 230°C, sol/GC

Amphetamine

1.66

Air, 220°C, sol/wire

Aprobarbital

1.39, 1.56, 1.75

N2, 230°C, sol/GC

Barbital

0.99, 1.50

N2, 230°C, sol/GC

Bromazepam

1.24

Air, 220°C, sol

Butabarbital

1.28, 1.35

N2, 230°C, sol/GC

Cannabinol

1.06

Air, 220°C, sol/wire

CDA

1.18

Air, 220°C, sol/wire

Cocaine

1.16

Air, 220°C, sol/wire

Codeine

1.18, 1.21

Air, 220°C, sol/wire

Diazepam

1.21

Air, 220°C, sol/wire

Flurazepam

1.03

Air, 220°C, sol/wire

Heroin

1.04, 1.14

Air, 220°C, sol/wire

LSD

1.085

Lorazepam

1.19, 1.22

Air, 220°C, sol/cart

Methamphetamine

1.63

Air, 220°C, sol/wire

MDA

1.49

Air, 220°C, sol/wire

MDMA

1.47

Mephobarbital

1.63, 1.81

N2, 230°C, sol/GC

Methyprylan

1.52

Air, 220°C, sol/wire

Morphine

1.22, 1.26

Air, 220°C, sol/wire

Nicotine

1.57

Nitrazepam

1.22

Air, 220°C, sol/wire

OMAM

1.13, 1.26

Air, 220°C, sol/wire

Opium

1.55

Air, 250°C, wire

Oxazepam

1.23, 1.28

Air, 220°C, sol/cart

PCP

1.27

Pentobarbital

1.38

N2, 230°C, sol/GC

Phenobarbital

1.44

N2, 230°C, sol/GC

Phenylcyclidine

1.27, 1.63, 2.01, 2.23

Air, 220°C, sol/wire

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_264... 06/05/2007

Página 2 de 2

Procaine

1.31

Secobarbital

1.31, 1.48

N2, 230°C, sol/GC

THC

1.05

Air, 220°C, sol/wire

Thelaine

1.14

Air, 220°C, sol/wire

Triazolam

1.13

Air, 220°C, sol/wire

Air—the drift gas, ionization source flow, and the sample vapors were carried by air; wire—insertion of a metal wire with absorbed vapors or solution. Source: DeTulleo-Smith, A.M.; Methamphetamine vs. nicotine detection on the Barringer ion mobility spectrometer, IMS Meeting, Jackson Hole, WY, 1996; Lawrence, A.H.; Detection of drug residues on the hands of subjects by surface sampling and ion mobility spectrometry, Forens. Sci. Int. 1987, 34, 73–83.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_264... 06/05/2007

Página 1 de 1

Page 265 from attack by muggers may use spray cans with agents that have similar effects to repel attackers. However, criminals have also discovered the potential of these agents for disabling their victim, and residues are occasionally found at crime scenes.93–96 The current methods for the detection of lachrymator residues require extensive sample preparation and spectral measurements in the laboratory. Allinson and McLeod have studied the gas-phase ion chemistry of some popular lachrymators: 2-chlorobenzylidenemalono-nitrile (CS), alpha-chloroacetophenone (CN) and pepper sprays (capsaicin and crude pepper extracts). They also explored the possibility of using a handheld IMS for detection of lachrymator residues. By comparing positive-and negative-ion mode spectra, it was possible to differentiate not only between CN and CS but also between their isomers and breakdown products. The MDLs varied, depending on the chemical properties of the substance (vapor pressure, proton affinity, and electronegativity) and ambient conditions (temperature and humidity). Generally, MDLs were in the milligram range for CS and pepper sprays and submicrogram range for CN.

6.5.2 Arson Accelerants at the scene of arson crimes or suspected arson cases may be detected by IMS that can also be used to locate the site of ignition.97 Though laboratory-based gas chromatograph-mass spectrometers are accepted methods for analysis of samples for accelerants, the portability of a handheld analyzer is attractive. Such an analyzer can be brought to a suspected crime scene and debris after the fire can be screened immediately on-site. The complexity of samples probably necessitates a GC/IMS analyzer such as the GC/IONSCAN and the Varian GC/DMD.

6.5.3 Security of Public Areas With the increasing threat of acts of terrorism against public areas, facilities, and buildings where large crowds may be present, the need for an alarm system that will alert the public to the presence of chemical, biological, and nuclear threats has increased. Sensitive potential targets like nuclear installations need protection and access control, and such a system based on IMS has been proposed in Russia.98 Systems that monitor the ambient air in such places and provide early alarms were developed by Smiths Detection (the Centurion) and by Environics as described in Chapter 8, Section 8.8. An example of such an application was the monitoring of diborane (B2H6) vapor by a Centurion device that can be equipped with multiple detector heads in the air system of a building.99 The effects of the humidity level on the response of the instrument at different concentrations of diborane were determined. It was reported that the device could serve to detect the presence of diborane with an MDL of 0.17 ppm, well below the required IDLH (Immediate Danger to Life and Health) level of 15 ppm.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_265... 06/05/2007

Página 1 de 1

Page 266

6.6 Conclusions At this writing, there is no other measurement technique or technology as widely used in such critical venues for human safety and security as IMS technology. The chemical properties of explosives, drugs, and chemical weapons, with the formation of stable gas-phase ions preferentially over many interferences found in relevant samples, made the principles of ion mobility spectrometry attractive to measurement science. However, the technology and the cumulative experience in support of those principles was developed over a 20-year span; fairly rapid application followed this induction period. The literature contains a significant amount of redundancy, and this provides measures of security or confidence with the principles and methods. The future may include improvements in detection limits and selectivity as well as miniaturized analyzers and improved affordability for instruments. Regardless of these future hopes, present demands for rugged on-site analyses suggest that IMS analyzers will be familiar tools in public venues and battlefields in the near future.

References 1. Carroll, D.I., Personnel Marking and Detection Based upon Plasma Chromatography Concept, Final Report 510-F; AFATL-TR-68–79, Franklin GNO Corporation, West Palm Beach, FL, July 1968. 2. Kilpatrick, W.D., Plasma Chromatography and Dynamite Vapor Detection, Final Report FAA-RD-71–7, Contract DOT-FA17WA-2491, Federal Aviation Administration, Washington, D.C., January 1971. 3. Cline, J.E.; Hobbs, J.R.; Barrington, A.E., Laboratory Evaluation of Detectors of Explosives’ Effluents, Final Report DOT/TSC, Cambridge, MA, November 1972. 4. Hall, W.A.; Gage, H.M., Determination of the Sensitivity and Specificity of Vapor Detection Systems for Explosives, Narcotics, and Related Compounds, Technical Report, USALWL, Aberdeen Proving Ground, MD, 1973. 5. Karpas, Z., Forensic applications of ion mobility spectrometry, Forens. Sci. Rev. 1989, 1, 103–119. 6. Keller, T.; Binz, R.; Regenscheit, P; Bernhard, W.; Dirnhofer, R.; Chlewinski, A., lonenmobilitatsspektrometrie in der forensischen analytik und kriminalistik, Krim. Schweiz. 1996, 67–70, 137–141. 7. Ewing, R.G.; Atkinson, D.A.; Eiceman, G.A.; Ewing, G.J., A critical review of ion mobility spectrometry for the detection of explosives and explosive related compounds, Talanta 2001, 54, 515–529. 8. Karpas, Z., Ion mobility spectrometry in forensic science, in Encyclopedia of Analytical Chemistry, John Wiley & Sons, London, 2000. 9. Moye, H.A., Plasma chromatography of pesticides, J. Chromatogr. Sci. 1975, 13, 285–290.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_266... 06/05/2007

Página 1 de 1

Page 267 10. Preston, J.M.; Karasek, F.W.; Kim, S.H., Plasma chromatography of phosphorus esters, Anal Chem. 1977, 49, 1746–1750. 11. Kim, S.H.; Spangler, G.E., Ion-mobility spectrometry-mass spectrometry of two structurally different ions having identical ion mass, Anal. Chem. 1985, 57, 567–569. 12. Karpas Z.; Pollevoy, Y., Ion mobility spectrometric studies of organophosphorous compounds, Anal Chim. Acta 1992, 259, 333–338. 13. Dworzanski, J.P.; Kim, M.-G.; Snyder, A.P.; Arnold, N.S.; Meuzelaar, H.L.C., Performance advances in ion mobility spectrometry through combination with high-speed vapor sampling, preconcentration and separation techniques, Anal. Chim. Acta 1994, 293, 219–235. 14. Turner, R.B.; Brokenshire, J.L., Hand-held ion mobility spectrometers, Trends Anal Chem. 1994, 13, 275– 280. 15. Leonhardt, J.W., New detectors in environmental monitoring using tritium sources, J. Radioanal Nucl Chem. 1996, 206, 333–339. 16. Tuovinen, K., Paakkanen, H.; Hanninen, O., Determination of soman and Vx degradation products by an aspiration ion mobility spectrometry, Anal. Chim. Acta 2001, 440, 151–159. 17. Paakanen, H., About the applications of IMCELL™ MGD-1 detector, Int. J. Ion Mobility Spectrom. 2001, 4, 136–139. 18. Kättö, T.; Paakkanen, H.; Karhapää, T., Detection of CWA by means of aspiration condenser type IMS, in Proceedings of the 4th International Symposium on Protection Against Chemical Warfare Agents, Stockholm, Sweden, 1992, pp. 103–108. 19. Asbury, G.R.; Wu, C.; Siems, W.F.; Hill, H.H., Separation and identification of some chemical warfare degradation products using electrospray high-resolution ion mobility spectrometry with mass-selected detection, Anal. Chim. Acta 2000, 404, 273–283. 20. Steiner, W.E.; Clowers, B.H.; Matz, L.M.; Siems, W.F.; Hill, H.H., Jr., Rapid screening of aqueous chemical warfare agent degradation products: ambient pressure ion mobility mass spectrometry, Anal. Chem. 2002, 74, 4343–4352. 21. Fällman, A.; Rittfeldt, L., Detection of chemical warfare agents in water by high temperature solid phase microextraction-ion mobility spectrometry (HT-SPME-IMS), Int. J. Ion Mobility Spectrom. 2001, 4, 85–87. 22. Harden, C.S.; Blethen, G.E.; Davis, D.M.; Harper, S.; McHugh, V.M.; Shoff, D.B., Detection and analysis of explosively disseminated CW agents in 400 m3 chamber, Int. J. Ion Mobility Spectrom. 2001, 4, 17–21. 23. Sielemann, S.; Li, E; Schmidt, H.; Baumbach, J.I., Ion mobility spectrometer with UV-ionization source for determination of chemical warfare agents, Int. J. Ion Mobility Spectrom. 2001, 4, 81–84. 24. Sielemann, S.; Baumbach, J.I.; Schmidt, H., IMS with non radioactive ionization sources suitable to detect chemical warfare agent simulation substances, Int. J. Ion Mobility Spectrom. 2002, 5, 143–148. 25. Ringer, J.; Ross, S.K.; West, D.J., An IMS/MS investigation of lewisite and lewisite/mustard mixtures, Int. J. Ion Mobility Spectrom. 2002, 5, 107–111. 26. Spangler, G.E.; Campbell, D.N.; Carrico, J.P., Acetone reactant ions for ion mobility spectrometry, Pittsburg Conference on Analytical Chemistry and Applied Spectroscopy, Atlantic City, NJ, 1983, Paper No. 641.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_267... 06/05/2007

Página 1 de 1

Page 268 27. Preston, J.M.; Rajadhyax, L., Effect of ion-molecule reactions on ion mobilities, Anal Chem. 1988, 60, 31– 34. 28. Carrico, J.P.; Davis, A.W.; Campbell, D.N.; Roehl, J.E.; Sima, G.R.; Spangler, G.E.; Vora, K.N.; White, R.J., Chemical detection and alarm for hazardous chemicals, Am. Lab. 1986, 18, 152, 155–7, 159–163. 29. Brochure of the Chemical Agent Monitor (CAM), Graseby, Watford, U.K. http://www.smithsdetection.com/ 30. Brochure of Bruker RAID series and IMS2000 hand-held instruments, Bruker-Saxonia, Germany. http://www.army-technology.com/contractors/nbc/bruker/ 31. Lias, S.G.; Liebman, J.F.; Levin, R.D., Evaluated gas phase basicities and proton affinities of molecules: heats of formation of protonated molecules, J. Phys. Chem. Ref. Data 1984, 13, 695–808. 32. Carnahan, B.; Day, S.; Kouznetsov, V.; Tarrasov, A., Development and applications of a traverse field compensation ion mobility spectrometer, Fourth International Workshop on Ion Mobility Spectrometry, Ed. A.Brittain, Cambridge, U.K., 1995. 33. Cao, L.; de B. Harrington, P.; Harden, C.S.; McHugh, V.M.; Thomas, M.A., Nonlinear wavelet compression of ion mobility spectra from ion mobility spectrometers mounted in an unmanned aerial vehicle, Anal. Chem. 2004, 76, 1069–1077. 34. Karasek, F.W.; Tatone, O.S.; Kane, D.M., Study of electron capture behavior of substituted aromatics by plasma chromatography, Anal. Chem. 1973, 45, 1210–1214. 35. Karasek, F.W.; Denney, D.W., Detection of 2,4,6-trinitrotoluene vapors in air by plasma chromatography, J. Chromatogr. 1974, 93, 141–147. 36. Karasek, F.W., Detection of TNT in air, Research/Development 1974, 25, 32–34. 37. Spangler, G.E.; Lawless, P.A., Ionization of nitrotoluene compounds in negative ion plasma chromatography, Anal. Chem. 1978, 50, 884–892. 38. Danylewych-May, L.L., Modifications to the ionization process to enhance the detection of explosives by IMS, Paper C-10, Proceedings of the 1st International Symposium Explosion Detection Technology, Atlantic City, NJ, November 1991. 39. Proctor, C.J.; Todd, J.F.J., Alternative reagent ions for plasma chromatography, Anal Chem. 1984, 56, 1794–1797. 40. Spangler, G.E.; Carrico, J.P.; Campbell, D.N., Recent advances in ion mobility spectrometry for explosives vapor detection, J. Test. Eval. 1985, 13, 234–240. 41. Lawrence, A.H.; Neudorfl, P, Detection of ethylene glycol dinitrate vapors by ion mobility spectrometry using chloride reagent ions, Anal. Chem. 1988, 60, 104–109. 42. Daum, K.A.; Atkinson, D.A.; Ewing, R.G.; Knighton, W.B.; Grimsrud, E.P., Resolving interferences in negative mode ion mobility spectrometry using selective reactant ion chemistry, Talanta 2001, 54, 299–306. 43. Lucero, D.P., User requirements and performance specifications for explosive vapor detection systems, J. Test. Eval. 1985, 13, 222–233. 44. Wernlund, R.F.; Cohen, M.J.; Kindell, R.C., The ion mobility spectrometer as an explosive taggant detector, Proceedings of the New Concepts Symposium and Workshop on Detection Identification of Explosives, Reston, VA, 1978, p. 185.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_268... 06/05/2007

Página 1 de 1

Page 269 45. Ewing, R.G.; Miller, C.J., Detection of volatile vapors emitted from explosives with a handheld ion mobility spectrometer, Field Anal. Chem. Technol. 2001, 5, 215–221. 46. Phares, D.J.; Holt, J.K.; Smedley, G.T.; Flagan, R.C., Method for characterization of adhesion properties of trace explosives in fingerprints and fingerprint simulations, J. Forens. Sci. 2000, 45, 774–784. 47. Spangler, G.E.; Carrico, J.P.; Kim, S.H., Analysis of explosives and explosive residues with ion mobility spectrometry (IMS), Proceedings of the International Symposium on Detection of Explosives, Quantico, VA, 1983, p. 267. 48. Fetterolf, D.D.; Clark, T.D., Detection of trace explosive evidence by ion mobility spectrometry, J. Forens. Sci. 1993, 38, 28–39. 49. Garofolo, F.; Migliozzi, V.; Roio, B., Application of ion mobility spectrometry to the identification of trace levels of explosives in the presence of complex matrixes, Rapid Commun. Mass Spectrom. 1994, 8, 527– 532. 50. Schellenbaum, R.L.; Hannum, D.W., Laboratory Evaluation of the PCP Large Reaction Volume Ion Mobility Spectrometer (LRVIMS), Report SAND-89–0461, 1990, 37 pp. 51. Elias, L.; Neudorfl, P, Laboratory Evaluation of Portable and Walk-Through Explosives Vapor Detectors, Report NAE-LTR-UA-104, CTN–91–60015, 1990, 26 pp. 52. Wu, C; Steiner, W.E.; Tornatore, P.S.; Matz, L.M.; Siems, W.F.; Atkinson, D.A.; Hill, H.H., Construction and characterization of a high-flow, high-resolution ion mobility spectrometer for detection of explosives after personnel portal sampling, Talanta 2002, 57, 123–134. 53. Gowadia, H.A.; Settles, G.S., The natural sampling of airborne trace signals from explosives concealed upon the human body, J. Forens. Sci. 2001, 46, 1324–1331. 54. Eiceman, G.A.; Preston, D.; Tiano, G.; Rodriguez, J.; Parmeter, J.E., Quantitative calibration of vapor levels of TNT, RDX, and PETN using a diffusion generator with gravimetry and ion mobility spectrometry, Talanta 1997, 45, 57–74. 55. Cohen, M.J.; Wernlund, R.F.; Kindel, R.C., An adjustable vapor generator for known standard concentrations in the fractional parts per billion range, Proceedings of the New Concepts Symposium and Workshop on Detection and Identification of Explosives, Reston, VA., 1978, p. 41. 56. Davies, J.P.; Blackwood, L.G.; Davis, S.G.; Goodrich, L.D.; Larson, R.A., Design and calibration of pulsed vapor generators for 2,4,6-trinitrotoluene, cyclo-1,3, 5-trimethylene-2,4,6-trinitramine, and pentaerythritol tetranitrate, Idaho National Engineering Laboratory, Anal. Chem. 1993, 65, 3004–3009. 57. Fricano, L.; Goledzinowski, M.; Jackson, R.; Kuja, E; May, L.; Nacson, S., An automatic trace detection system for the detection of explosives’ vapors and particles in luggage, Int. J. Ion Mobility Spectrom. 2001, 4, 22–26. 58. http://www.geindustrial.com/ge-interlogix/iontrack/prod_entryscan.html. 59. Matz, L.M.; Tornatore, P.S.; Hill, H.H., Evaluation of suspected interferents for TNT detection by ion mobility spectrometry, Talanta 2001, 54, 171–179. 60. McGann, W.J.; Haigh, P; Neves, J.L., Expanding the capability of IMS explosive trace detection, Int. J. Ion Mobility Spectrom. 2002, 5, 119–122. 61. Buttigieg, G.A.; Knight, A.K.; Denson, S.; Pommier, C; Denton, M.B., Characterization of the explosive triacetone triperoxide and detection by ion mobility spectrometry, Forens. Sci. Int. 2003, 135, 53–59.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_269... 06/05/2007

Página 1 de 1

Page 270 62. McGann, W.J.; Goedecke, K.; Becotte-Haigh, P.; Neves, J.; Jenkins, A., Simultaneous dual-mode IMS detection system for contraband detection and identification, Int. J. Ion Mobility Spectrom. 2001, 4, 144– 147. 63. Marr, A.J.; Groves, D.M., Ion mobility spectrometry of peroxide explosives TATP and HMTD, Int. J. Ion Mobility Spectrom. 2003, 6. 64. Karasek, F.W.; Hill, H.H., Jr.; Kim, S.H., Plasma chromatography of heroin and cocaine with massidentified mobility spectra, J. Chromatogr. 1976, 117, 327–336. 65. Ithakissios, D.S., Plasmagram spectra of some barbiturates, J. Chromatogr. Sci. 1980, 18, 88–92. 66. Eatherton, R.L.; Siems, W.F.; Hill, H.H., Jr., Fourier transform ion mobility spectrometry of barbiturates after capillary gas chromatography, J. High Resolut. Chromatogr. Chromatogr. Commun. (HRC & CC) 1986, 9, 44–48. 67. Lawrence, A.H.; Elias, L., Application of air sampling and ion-mobility spectrometry to narcotics detection: a feasibility study, Bull. Narc. 1985, 37, 3–16. 68. Lawrence, A.H., Ion mobility spectrometry/mass spectrometry of some prescription and illicit drugs, Anal. Chem. 1986, 58, 1269–1272. 69. Lawrence, A.H., Detection of drug residues on the hands of subjects by surface sampling and ion mobility spectrometry, Forens. Sci. Int. 1987, 34, 73–83. 70. Nanji, A.N.; Lawrence, A.H.; Mikhael, N.Z., Use of skin sampling and ion mobility spectrometry as a preliminary screening method for drug detection in an emergency room, J. Toxicol. Clin. Toxicol. 1987, 25, 501–515. 71. Lawrence, A.H.; Nanji, A.A., Ion mobility spectrometry and ion mobility spectrometry/mass spectrometric characterization of dimenhydrinate, Biomed. Environ. Mass Spectrom. 1988,16, 345–347. 72. Lawrence, A.H.; Nanji, A.A.; Taverner, J., Skin-sniffing ion mobility spectrometric analysis: a potential screening method in clinical toxicology, J. Clinical Lab. Anal. 1988, 2, 101–107. 73. Lawrence, A.H., Characterization of benzodiazepine drugs by ion mobility spectrometry, Anal. Chem. 1989, 61, 343–349. 74. Chauhan, M.; Harnois, J.; Kovar, J.; Pilon, P., Trace analysis of cocaine and heroin in different customs scenarios using a custom-built ion mobility spectrometer, J. Can. Soc. Forens. Sci. 1991, 24, 43–49. 75. Eiceman, G.A.; Blyth, D.A.; Shoff, D.B.; Snyder, A., Screening of solid commercial pharmaceuticals using ion mobility spectrometry, Anal. Chem. 1990, 62, 1374–1379. 76. Fetterolf, D.D.; Donnelly, B.; Lasswell, L.D., III, Detection of heroin and cocaine residues by ion mobility spectrometry, 39th Conference American Society of Mass Spectrometry, Nashville, TN, 1991. 77. Su, C.-W.; Babcock, K.; Rigdon, S., The detection of cocaine on petroleum contaminated samples utilizing ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 1998,1, 15–27. 78. Su, C.-W.; Babcock, K.; deFur, P.; Noble, T.; Rigdon, S., Column-less GC/IMS (II)—a novel on-line separation technique for ionscan analysis, Int. J. Ion Mobility Spectrom. 2002, 5, 160–174. 79. DeTulleo, A.M.; Galat, P.B.; Gay, M.E., Detecting heroin in the presence of cocaine using ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2000, 3, 38–42. 80. Miki, A.; Keller, T.; Regenscheit, P; Bernhard, W.; Tatsuno, M.; Katagi, M.; Nishikawa, M.; Kim, L.; Hatano, S.; Tsuchihashi, H., Determination of internal and external methylamphetamine in human hair by ion mobility spectrometry, Jpn. J. Toxicol Environ. Health 1997, 43, 15–24.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_270... 06/05/2007

Página 1 de 1

Page 271 81. Miki, A.; Keller, T.A.; Regenscheit, P.; Dirnhofer, R.; Tatsuno, M.; Katagi, M.; Nishikawa, M; Tsuchihashi, H., Application of ion mobility spectrometry to the rapid screening of methylamphetamine incorporated in hair, J. Chromatogr. B: Biomed. Appl.1997, 692, 319–328. 82. Keller, T.; Miki, A.; Regenscheit, P.; Dirnhofer, R.; Schneider, A.; Tsuchihashi, H., Detection of designer drugs in human hair by ion mobility spectrometry (IMS), Forens. Sci. Int. 1998, 94, 55–63. 83. Keller, T.; Schneider, A.; Regenscheit, P.; Dirnhofer, R.; Rucker, T.; Jaspers, J.; Kisser,W., Analysis of psilocybin and psilocin in Psilocybe subcubensis Guzman by ion mobility spectrometry and gas chromatography-mass spectrometry, Forens. Sci. Int. 1999, 99, 93–105. 84. Keller, T.; Schneider, A.; Tutsch-Bauer, E.; Jaspers, J.; Aderjan, R.; Skopp, G., Ion mobility spectrometry for the detection of drugs in cases of forensic and criminalistic relevance, Int. J. Ion Mobility Spectrom. 1999, 2, 22–34. 85. Keller, T.; Miki, A.; Tatsuno, M.; Katagi, M.; Nishikawa, M.; Tsuchihashi, H.; Regenscheit, P.; Dirnhofer, R., Detection of methamphetamine, MDMA and MDEA in human hair by means of ion mobility spectrometry (IMS), Int. J. Ion Mobility Spectrom. 1998,1, 38–42. 86. Orzechowska, G.E.; Poziomek, E.J.; Tersol, V., Use of solid-phase microextraction (SPME) with ion mobility spectrometry, Anal. Lett. 1997, 30, 1437–1444. 87. McCooeye, M.A.; Mester, Z.; Ells, B.; Barnett, D.A.; Purves, R.W.; Guevremont, R., Quantitation of amphetamine, methamphetamine, and their methylenedioxy derivatives in urine by solid-phase microextraction coupled with electrospray ionization high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal. Chem. 2002, 74, 3071–3075. 88. McCooeye, M.A.; Ells, B.; Barnett, D.A.; Purves, R.W.; Guevremont, R., Quantitation of morphine and codeine in human urine using high-field asymmetric waveform ion mobility spectrometry (FAIMS) with mass spectrometric detection, J. Anal. Toxicol. 2001, 25, 81–87. 89. Matz, L.M.; Hill, H.H., Jr., Evaluating the separation of amphetamines by electrospray ionization ion mobility spectrometry/MS and charge competition within the ESI process, Anal. Chem. 2002, 74, 420–427. 90. Matz, L.M.; Hill, H.H., Separation of benzodiazepines by electrospray ionization ion mobility spectrometry-mass spectrometry, Anal. Chim. Acta 2002, 457, 235–245. 91. Reese, E.S.; Harrington, P.B., The analysis of methamphetamine hydrochloride by thermal desorption ion mobility spectrometry and SIMPLISMA, J. Forens. Sci. 1999, 44, 68–76. 92. Shaw, L.A.; Harrington, P.D., Seeing through the smoke with dynamic data-analysis: Detection of methamphetamine in forensic samples contaminated with nicotine, Spectroscopy 2000, 15, 40–45. 93. Allinson, G.; Mcleod, C.W., Characterization of lachrymators by ambient-temperature ion mobility spectrometry, J. Forens. Sci. 1997, 42, 312–315. 94. Allinson, G.; Mcleod, C.W., Characterization of tear gas residues by ion mobility spectrometry, Appl. Spectrosc. 1997, 51, 1880–1889. 95. Allinson, G.; Saul, C; McLeod, C.W.; Gilbert, J., Identification of tear gases in suspect spray cans and cloth samples by ion mobility spectrometry, J. Forens. Sci. 1998, 43, 845–849.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_271... 06/05/2007

Página 1 de 2

Page 272 96. Kataoka, M; Seto, Y; Tsuge, K.; Noami, M, Stability and detectability of lachrymators and their degradation products in evidence samples, J. Forens. Sci. 2002, 47, 44–51. 97. Stotz, J.; Harrington, P., Accelerant detection for arson investigations using ion mobility spectrometry, #884, PittCon, New Orleans, LA, 1998. 98. Matsaev, V.T.; Gumerov, M.F.; Chilipenko, L.L.; Kozlov, N.N., Use of IMS for access control application at Russia’s nuclear power facilities, Int. J. Ion Mobility Spectrom. 2002, 5, 115–118. 99. Sun, Y; DeBono, R.; Burton, J., Monitoring diborane vapor with Smiths Detection’s Centurion, 12th International Conference on Ion Mobility Spectrometry, Umeå, Sweden, July 27–32, 2003.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_272... 06/05/2007

Página 2 de 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_272... 06/05/2007

Página 1 de 1

Page 273

7 Biological, Biomolecular Research, and Medical Uses of IMS 7.1 Introduction Advances in methods to introduce samples into IMS drift tubes as described in Chapter 4 have enabled the range of applications of IMS to be expanded beyond the measurements of volatile and semivolatile compounds normally associated with IMS. Molecules and samples, wholly new to IMS measurements, such as peptides, proteins, and carbohydrates, have been explored during the past decade and may have importance in medical research or clinical use. Such studies have been made possible in large part through the adaptation of methods of sample delivery or ionization that have been successfully pioneered in mass spectrometry during the past 20 years. These include electrospray ionization (ESI) for aqueous samples or matrix-assisted laser desorption and ionization (MALDI) for solid samples as discussed in Chapter 4. The importance of these methods in chemical sciences was recognized with the award of the 2002 Nobel Prize in Chemistry to John B. Fenn1 and Koichi Tanaka2 for developing them. The rationales for exchanging mass spectrometers with mobility spectrometers in biological measurements parallel those used previously with other measurements described in Chapter 6: mobility spectrometers are inexpensive, portable, and sensitive analyzers that use low power. What may be compromised in density of information in comparison to a mass spectrometer is rewarded with savings in cost and convenience. In other instances, IMS drift tubes can be combined with mass spectrometers as described in Chapter 5 for enhanced analytical capabilities in measuring biomolecules. Apart from the applications that are likely to remain as laboratory or research tools, IMS analyzers have been examined or developed for clinical uses such as the diagnosis of diseases or for measurements of exposure to anesthetic gases. Bacteria can be determined using IMS by a method of fast sample preparation that produces volatile or semivolatile compounds. In another approach, bacteria are thermally decomposed to produce volatile compounds and these may be used to determine bacteria in air. A few examples of these

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_273... 06/05/2007

Página 1 de 1

Page 274 types of IMS studies are described in this chapter to illustrate certain applications that may alter both the perceptions and value of IMS during the next decades. This is not an exhaustive review but an effort was made to offer a balanced presentation of the wide range of methods or applications. In general, practical applications with a commercial potential, based on IMS drift tubes that are operated at ambient pressure, can be distinguished from laboratory or research drift tubes that are operated at reduced pressure with an inert gas and usually as an IMS/MS instrument.

7.2 Medical Diagnostics Using IMS Since ancient times, odors or vapors from breath or urine have been understood to reflect illnesses in humans. The use of advanced analytical instruments, including gas chromatograph/mass spectrometers, has been explored since the 1960s to replace human senses with refined chemical measurements. Extensions of this concept to gas chromatograph/mobility spectrometers or to mobility spectrometers alone have been made in the past decade and are not restricted only to physiological vapors but also included vapors from molds, bacteria, and certain medically useful gases such as anesthetics. Except for the determination of anesthetics by GC/IMS or GC/DMS instruments, these methods have not yet been commercialized.

7.2.1 Respired Air as a Measure of Exposure to Anesthetic Gases and of Lung Disease The potential for using IMS to detect, identify, and monitor volatile compounds such as halothane, enflurane and isoflurane, which are used as anesthetic gases in operating theaters, was investigated by Eiceman et al.3 The presence and concentration of these compounds following inhalation of a small dosage was monitored in the respired air exhaled directly into a handheld IMS analyzer. In the negative-ion mode, the dominant ions were the chloride adduct ion and chloride-bound dimer with increases in anesthetic gas concentrations. In general, residual low levels of these anesthetic gases were detectable even more than 1 h after the end of treatment. Realtime monitoring of breath for these substances could make possible an objective measure of the depth of anesthesia to complement existing clinical observations. In addition, minutes after the onset of treatment, vapors sampled from the skin surface provided evidence of an exposure event as shown in Figure 7–1.4 In principle, other volatile chemicals inhaled by a subject could be observed in respired air or from skin emissions and could provide a noninvasive assessment of the extent of exposure. Lung diseases may be recognized through changes in the chemical composition of volatile organic compounds in the air exhaled by an ill subject.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_274... 06/05/2007

Página 1 de 1

Page 275

FIGURE 7–1 Plots of peak intensity for the product ion of isoflurane in IMS determination of vapors containing isoflurane and inhaled by a rabbit, and vapor concentrations of enflurane obtained from gases respired from skin. The source of isoflurane was removed from the inhalation chamber at a time marked by the dashed line. Both the inhaled vapors and the emissions from the skin were monitored using military grade chemical agent monitors in negative ion polarity.

Preliminary studies by Ruzsanyi et al.5 indicated that the mobility spectra from GC/IMS analysis of breath of those with lung damage of the type that may be incurred by cancer differs from that of healthy subjects. The underlying assumption was that the respired volatile organic compounds (VOCs) reflect their concentration in blood through exchange of gases at the lung interface. In that study, 11 substances (mainly ketones, alkanes, and diones) with room-temperature-reduced mobility values of 1.08 to 1.97 cm2 V−1sec−1 served as markers of illness. A photo-discharge lamp (10.6 eV) afforded some degree of selectivity, although some kind of preseparation technique is preferred to resolve mutual interferences between the components. In another study by the same group, a highspeed capillary column was coupled to an IMS drift tube with a UV ionization source to determine volatile organic compounds that may arise through diseases or certain occupational exposures. These compounds included some common ketones, benzene, and some substituted benzenes.6 Another investigation examined the presence of acetone in breath using a membrane extraction module, a sorbent trap, and a gas chromatograph with dual detectors: a flame ionization detector and a mobility spectrometer. The last 0.25–1 portion of the stream of exhaled breath was analyzed. The membrane removed much of the respired moisture, blocking interference with the analyses.7

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_275... 06/05/2007

Página 1 de 1

Page 276 These few reports on the determination by GC/IMS of organic compounds in breath or respired air are consistent with, and supported by, centuries of experience in traditional medicine. Examples of diagnostic vapors include acetone from diabetes, isoprene from lung damage, and styrene, 2methylheptane, propylbenzene, decane, and undecane from lung cancer.8 Given the sensitivity and specificity of IMS analyzers to such compounds, commercial development of volatile organic compounds for clinical diagnosis of illness through analysis of exhaled air seems inevitable. However, this is dependent upon a suitable instrument configuration, probably a fast GC/IMS analyzer.

7.2.2 Diagnosis of Vaginal Infections The degradation of proteins, peptides, and amino acids leads to the formation of several lowmolecular-weight compounds including biogenic amines. Biogenic amines are a family of compounds that includes monoamines, diamines, triamines, and tetramines such as trimethylamine (TMA), putrescine and cadaverine, and spermidine and spermine, respectively. These and several other compounds such as histamine, tyrosine, skatole, and other amines are particularly interesting for IMS determinations due to the high proton affinities of amines. The vapors may arise through several degradation pathways including chemical reactions, enzymatic reactions, and microbial processes. For example, decarboxylation of lysine produces cadaverine, whereas loss of CO2 from histidine produces histamine, as shown in Equation 7–1 and Equation 7–2: (7–1) (7–2) An example of these amines being indicative of illness is bacterial vaginosis (BV), which is the most common form of vaginal infection that afflicts females of almost all ages, races, and societies.9 Bacterial vaginosis may occur when the delicate balance is disturbed in the vaginal discharge fluid between lactobacilli (microorganisms that excrete lactic acid and peroxides and maintain a low pH from 3.8 to 4.2) and pathogenic microorganisms. This may be due to several external and internal effects, such as use of medication (antibiotics, in particular, reduce the lactobacilli population), habits of personal hygiene (rinsing too frequently or not enough), allergy, and more. One of the manifestations of BV is the enhanced production of biogenic amines, TMA and putrescine, particularly. This has been extensively investigated in Israel and the U.S. by Q-Scent.10–12 In practice, a swab of vaginal discharge fluid is collected by the gynecologist and placed in a vial and capped. The addition of an alkaline solution to the sample enhances the volatilization of amines in solution even at room

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_276... 06/05/2007

Página 1 de 1

Page 277 temperature, and headspace vapors containing volatile amines, if present, may be transferred to an IMS analyzer for quantitative determination. After about 10 sec the vial can be heated rapidly, treated with a few drops of a dilute acid solution, and headspace can be sampled for other volatile amines. In order to minimize interferences, n-nonylamine is used as the reagent gas. In a sample from a BV-infected patient, TMA will appear initially and putrescine will be observed later when the sample is heated. During the measurement, the dynamic trends for a sample collected from a patient with BV will show first an increase in the TMA peak area and concomitant decrease in the reactantion peak area; this is followed by a decrease in the TMA peak area and increase in the putrescine peak area (Figure 7–2). Samples from a healthy subject will exhibit only the reactant-ion peak without peaks for biogenic amines. Other common vaginal infections such as candidiasis (yeast infection) and trichomoniasis are also detectable as elevated levels of other biogenic amines, cadaverine, and putrescine, with little or no TMA. This analysis is sensitive, specific (overall above 95% accuracy with low false negative and false positive), rapid (about 90 sec), simple in methodology and inexpensive compared to other screening methods for BV. Thus, an IMS-based test could replace the conventional diagnostic procedure (i.e., the Amsel test) and diagnostic methods for other infections, as suggested by the favorable classifications shown in Figure 7–3.13 This method is not yet commercially available but could become one of the first successes in clinical applications of mobility spectrometers.

FIGURE 7–2 Mobility spectra measured at delays after introduction of sample. Each spectrum is the average of three scans in a sample collected from a patient with bacterial vaginosis. Note the changes in the background spectrum (solid) as volatile trimethylamine reached the drift tube (dashed) and later as the semivolatile amines became dominant (dotted line). The reagent gas is nnonylamine and the amines are trimethylamine (TMA), putracine (PUT), and cadaverine (CAD).

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_277... 06/05/2007

Página 1 de 1

Page 278

FIGURE 7–3 Summary of the measurements of biogenic amines in vaginal discharge fluid, based on principal component analysis. The circles and triangles denote samples that were diagnosed as BV-positive and BV-negative, respectively, in the Amsel test.

7.3 Food Freshness and Odor Detection All foods that contain muscle tissue (meat, poultry, and fish) can form bio-genic amines through enzymatic and bacterial action on proteins and amino acids, as described earlier for vaginal infections. For example, TMA is formed by a stepwise degradation of choline to betaine and then to TMA, according to Equation 7–3: (7–3) The unpleasant odor of decaying fish (the same as in bacterial vaginosis) has been attributed mainly to TMA,14 although other compounds may also contribute to this odor. A handheld GC/IMS analyzer was used to determine the odors released from aging fish, and TMA was observed in the odors.15 However, this was not pursued further until a systematic investigation of meat, poultry, and fish was undertaken by Karpas et al.16 Biogenic amines were determined as a function of the storage temperature and time in different types of muscle food. Amine concentrations were correlated to the levels of six types of microorganisms counted by standard culture growth techniques (Table 7–1). In order to enhance the sensitivity toward semivolatile compounds, the sample introduction method described earlier for diagnosis of vaginal infections was used. In Figure 7–4, mobility spectra are shown for different types of muscle food after a few days at room temperature. The amount of

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_278... 06/05/2007

Página 1 de 1

Page 279 TABLE 7–1 The IMS Results for TMA and Total Amines, and the Microorganism Cultures Count for Samples of Ground Chicken When Fresh and after 1, 2, and 4 d at Room Temperature

Sample

TMA (ng/g)

Total Amines (arb.units)

1

2

3

4

5

6

Fresh

1.1±0.3

1.2±0.1

3.8E3

8.4E3

2.3E2

2.3E2

4.5E2

4.5E2

1d

3.7±1.0

2.1±0.5

7.4E6

1.0E7

1.3E5

8.6E3

1.5E5

1.9E5

2d

32.8±8

11.3±3

3.9E8

4.7E8

6.6E5

4.2E6

4.7E6

6.3E6

4d

41.0±8

14.3±4

1.4E9

1.9E9

5.9E4

1.0E7

7.8E6

7.5E6

Note: IMS results from “broth” diluted 10-fold (DF 10); TMA: amount (ng) for 0.1 ml. Original broth is a 10-g sample in 90 ml buffer. Dilution factor is 10. (1) pseudomonas, (2) aerobic, (3) enterobacteria, (4) staphylococcus, (5) fungi, and (6) lactoacidbacter.

biogenic amines formed depends on the type of meat, the storage period, and the temperature of storage. Storage of food in a deep freeze (−18°C) slowed the degradation processes and practically arrested decomposition for periods of several months. In meat at room temperature, biogenic amines were produced rapidly, and large concentrations of these amines were evident within 24 h. The results show that IMS detectors can be used as a diagnostic tool for food freshness or for the determination of food spoilage.

FIGURE 7–4 Mobility spectra showing the formation of volatile amines from the spoilage of pork, turkey, beef, and chicken during storage at room temperature for 1 d. Calibration was with 2 ng of TMA. Trimethylamine (TMA) and cadaverine (CAD) are apparent in the mobility spectra for each muscle food. The reagent gas was n-nonylamine.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_279... 06/05/2007

Página 1 de 1

Page 280 Another approach to screening bacteria is based upon extracellular enzymes that are secreted by bacteria and found on the outside surface of the cell wall. When these enzymes are allowed to react with a suitable substrate, a volatile substance is produced and is able to be detected using a mobility spectrometer. This approach was tested by Strachan and co-workers17,18 and adapted with a sampling system to automatically screen food for bacteria that are associated with food safety Although this method is effective, fast, and sensitive, the level of extracellular enzymes is governed in part by the health and recent history of the bacteria. A related method with improved reliability and specificity is based on immunoassay methods as discussed in Section 7.5. In an enclosed atmosphere, molds and fungi may flourish, particularly in dark, moist locations. These organisms or the chemicals released by their metabolism may cause headaches, allergic reactions, irritation of the respiratory tract and skin, as well as other health problems principally among asthmatic or sensitive subjects.19–20 Some of the chemicals that have been considered as metabolic by-products of molds and fungi, mainly ketones and alcohols, have been characterized by IMS devices with a photo-discharge UV lamp and with a 63Ni ion source.19 Minimum detection levels with the 63Ni ion source were about an order of magnitude better than those with the UV discharge lamp. It was demonstrated that an IMS analyzer was capable of monitoring the level of these chemicals in indoor air at subclinical concentrations. The chemicals emitted from bread mold cultures were directly measured with an 63Ni-based IMS analyzer and then with a GC/UV/IMS instrument after solid-phase microextraction. The direct measurements of the mixture by an IMS analyzer alone were difficult to interpret as several chemicals were present in the sample and the mobility spectrum was further complicated by the extensive fragmentation of the alcohol ions. The use of Solid-Phase Microextraction (SPME) for sampling headspace vapors and the thermal desorption for chromatographic prefractionation with an IMS detector provided improved measurements.20 The concept of using chemical instrumentation to replace olfactometry expert panels, employed in the perfume and flavors industries, has long been sought, and simple, inexpensive sensor arrays have recently been promoted as “artificial noses” or “electronic noses” (e-nose). However, these devices lack specificity in response because the signal is based on principles of solubility of vapors in polymers. Ion mobility spectrometers have been used to detect odors (pleasant like perfume or unpleasant like decaying meat) since the advent of the technique in the 1960s. Recently, perfumes, flowers, and sagebrush species have been differentiated using mobility spectra combined with principal component analysis (PCA) for interpretation of the spectra21 In another test, the vapors emanating from aluminum films that were coated with white and orange color prints were analyzed using a GC/IMS analyzer equipped with a photo-discharge lamp as the ion source.22 Distinct patterns were observed for each coating. Comparison of the mobility spectra of coated and uncoated transparent films

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_280... 06/05/2007

Página 1 de 1

Page 281 (one burdened with malodor) used in food wrapping showed differences in the response, and the malodorous film was readily detected. The use of GC for preseparation was important as the mobility constants of many of the compounds studied had similar values. Once again, the potential of IMS methods and particularly GC/IMS instruments for such applications is only beginning to emerge.

7.4 Proteins, Peptides, Amino Acids, and Other Large Biomolecules and Biopolymers The separation, detection, or identification of proteins, peptides, and amino acids using IMS has been explored by a handful of research groups in the U.S., namely, those of Hill, Bowers, Russell, Jarrold, Clemmer, and Guvremont (in Canada). Common to all these groups is the method of sample handling; samples are introduced into a drift tube as liquids via electrospray ionization (ESI) or as solids with matrix-assisted laser desorption and ionization. In most cases, a mass spectrometer is used for detecting and identifying the ions, and a mobility spectrometer is used to preseparate components and to obtain reduced mobility coefficients (see Chapter 5 for details on instruments). These studies involve mainly the estimation of the conformation of the gas-phase ions formed by these large compounds, and these are derived from molecular models and the collision cross sections obtained from Ko values. One facet of forming gas-phase ions from large molecules with several functional groups is that multiply-charged ions are commonly observed at ambient pressure with electrospray ionization. Thus, mobility spectra for a single compound from substances such as proteins and oligonucleotides (and synthetic polymers) will exhibit peaks for these various charge states. Even a simple mixture of biological molecules will be very complex with poorly resolved or unresolved peaks. Mass spectra of such mixtures will also be too complex for interpretation. Examples of the instrumentation used in these research projects include high-field asymmetric waveform ion mobility/mass spectrometry23– 25 and an IMS with time-of-flight mass spectrometer for multidimensional separation of complex mixtures.26–31,42 Among the compounds studied are those that are formed in a mixture from tryptic digestion of peptides32 and peptide libraries,26,31,33–37 bradykinin,24,38 polyglycine and polyalanine,34 carbohydrates,28 ubiquitin,39 chemically modified DNA oligonucleotides with up to eight bases in length,30 proteins,40 valinomycin,41 and others.45 The use of mobility spectrometers for studies of these biomolecule can be categorized into three groups: single-compound structural studies, the use of an IMS/MS instrument for characterizing ion mixtures, and the use of a mobility spectrometer as an ion filter before a mass spectrometer. In some

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_281... 06/05/2007

Página 1 de 1

Page 282 of these, the mobility spectrometers are operated at reduced pressures of less than 100 torr with a drift gas of helium or argon. Although these are not strictly analytical applications of IMS, the use of a mobility measurement in combination with other techniques illustrates an instance where similar information is difficult to obtain by other means. Mobility measurements used to obtain structural information on large biomolecules rely on drift tubes that operate at reduced pressure, usually with helium as the drift gas. Although these are usually not considered as “classic” analytical IMS instruments, the principles of ion mobility are central to the studies and represent a type of refined mobility experiment.

7.4.1 Conformation Studies As an ion travels through the drift gas, there are several types of interactions between the ion and the neutral molecules of the drift gas (see Chapter 2). These interactions depend on the ion (size, total charge, charge distribution within the ion, and shape) and the drift-gas molecule (size, dipole moment, quadrupole moment, and polarizability). The magnitude of these interactions determines the drift velocity of the ion, namely, the mobility. This is the basis for elucidation of ion structure or conformation from mobility measurements and has been applied to elucidate the structure of small ions like anilines43 and diamines44 as well as that of large ions like protonated polyglycine and polyalanine.34 The work of Clemmer’s group demonstrates the unique advantage of using IMS to examine the influence of solvent composition and capillary temperature on the gas-phase conformations of ubiquitin ions (+6 to +13) formed during electrospray ionization.39 Three general conformer types were observed: compact forms (favored for the +6 and +7 charge states); partially folded conformers (favored for the +8 and +9 ions); and unfolded conformers (favored for the +10 to +13 charge states). The population distribution of different conformers was highly sensitive to solvent composition and the capillary temperature used for electrospray ionization. The differences in mobility, and, therefore, ion cross section, were associated with the number of charges on an ion, where increases in charge state caused increases in coulombic repulsion and unfolding of the ion. Naturally, folded compact conformers will exhibit shorter drift times than the unfolded conformers with large cross sections of collision. Cross sections of ions in the gas phase are difficult to obtain by any other measurement techniques besides IMS.

7.4.2 Alkali Ions of Biomolecules In addition to protonated species, other ions such as Na+ and Li+ may be attached to biomolecules and may be readily observed in ESI/IMS/MS experiments. Hill’s group derived structural information about the most probable location of the charge in protonated and sodiated bradykinin

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_282... 06/05/2007

Página 1 de 1

Page 283 and kemptide (where sodium replaced a proton in singly and doubly charged peptides). The structures were derived from the cross sections determined with a mobility spectrometer37 A mobility difference between protonated and sodiated species was observed, and it appeared as if the doubly charged sodiated peptides had a smaller collision cross section than the doubly charged protonated ones, leading to the conclusion that the gas-phase conformations of these ions are different with respect to intramolecular interactions. Results are shown in Figure 7–5 for penta- and hexapeptides. Protonated and sodiated ions may also be formed in MALDI experiments as shown by Bowers’ group.38 They found that, in bradykinin (BK), several cationized species were generated in the gas phase, including the protonated form (BKH+), the sodiated form (BKNa+), and (BK-H+2Na)+. All three

FIGURE 7–5 Ion mobility spectra of mixtures of isomeric peptides. The top frame is for pentapeptides with inverse amino acid sequences and the bottom frame shows hexapeptides differing by N-terminal amino acid and the fourth amino acid. The spectra were obtained with nitrogen as a drift gas at 250°C and ambient pressure. The mass spectrometer was operated in the single-ion monitoring mode at mass-to-charge ratios of 246 (top) and 302 (bottom), showing only the doubly charged gas-phase ions of the peptides.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_283... 06/05/2007

Página 1 de 1

Page 284 species had similar cross sections of 245±3Å2, independent of temperature from 300 to 600 K. They concluded that BK is wrapped around the charge centers in a globular shape where the dynamics of structural variation exhibit little change in time-averaged sizes up to 600 K. In another publication, the collision cross sections of valinomycin-alkali ion complexes (Li, Na, K, Rb, and Cs) were derived from the measured mobility of the ions in 3 torr of helium in a short drift tube with a MALDI source.41 The systematic increase in cross section with ion size indicated that the backbone folding of the cyclic valinomycin molecule was dependent on the size of the alkali ion.

7.4.3 Further Studies of Biocompounds Cook’s group noticed in 1994 that positively charged horse heart apomyoglobin (mol wt 16,951 Da) appeared in two forms in an ESI/tandem MS instrument: a high-charge state with distribution centered around (M+20H20+ and a second distribution centered around (M+10H)10+, which were predominant at low and high target gas pressures, respectively.45 These two distinct charge-state distributions were interpreted as an open conformational form for the high charge state and a folded form for the low charge state. Preferential charge selection was dependent on the nature and pressure of the target gas as well as the nature of the protein, and conformers could be selected by control of the collision gas pressure, favoring one form over the other. It was noted that bimodal distributions were observed at intermediate pressures, but that charge states between the two distributions were not effectively populated under most of the conditions examined. Hard-sphere collision calculations showed large differences in collision frequencies and in the corresponding kinetic energy losses of the two conformational states and demonstrated that the observed charge-state selectivity could be explained through elastic collisions.45

7.5 Detection and Determination of Bacteria 7.5.1 Pyrolysis GC/IMS Methods The widespread use and acceptance of mobility spectrometers in the armed forces created interest in exploring the possibility of using handheld chemical agent monitors (CAM) to detect bacteria with extracellular enzymes46 and later by the immunoassay methods47 described in the following text. These methods were not suitable for continuous monitoring of air without dispensable reagents, so alternate approaches were sought using methods of pyrolysis. Nonetheless, the concept of retaining a

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_284... 06/05/2007

Página 1 de 1

Page 285 common analytical method for detection of both chemical and biological agents was appealing, and a method of pyrolysis coupled to a gas chromatograph/ion mobility spectrometer has been pioneered by the U.S. Army team of A.R Snyder and colleagues. Although the name of the organization has changed through the years (i.e., Edgewood Research, Development and Engineering Center, Edgewood Research Center for Development and Engineering, and Edgewood Chemical and Biological Center), the team and mission has remained fairly constant. The technology was a derivative of the CAM drift tube as described in Chapter 5,48 and was modified further with a pyrolysis inlet49 that could accept a sample of air containing bacteria. In order to detect bacterial aerosols in ambient air, a large volume of air was passed through a glass tube. After the collection step, the sample was rapidly heated to 300°C (i.e., pyrolyzed). The pyrolysis products were characterized using a high-speed GC column with IMS detector (Py-GC/ IMS).50–52 The findings were supported by experiments made in parallel using a GC/MS instrument.53 The presence of compounds that arise from the pyrolysis of spores and bacteria was determined at specific GC retention times and drift times. Certain compounds are characteristic of bacillus spores and are regarded as biomarkers, i.e., chemicals that are always present in an organism regardless of history or life stage of the bacteria. For example, Gram-positive spores such as Bacillus subtilis, var. globigii (BG) spores contain 5 to 15% by weight of calcium dipicolinate that can be pyrolyzed to dipicolinic acid (DPA), picolinic acid (PA), and pyridine. Picolinic acid has a high proton affinity, and it is detected in a sensitive fashion and identified by a conventional IMS analyzer. Picolinic acid occupies a unique region in the GC/IMS data domain with respect to other bacterial pyrolysis products. A 1000 to 1, air-to-air aerosol concentrator was interfaced to the PyGC/IMS instrument to test controlled aerial releases of the spores. In the 21 BG trials, the PyGC/IMS instrument experienced two true negatives and no falsepositives, and developed a software failure in one trial. The remaining 18 trials gave true positive results for the presence of BG aerosol in ambient air after a biorelease. The limit of detection for the Py-GC/IMS instrument was estimated at approximately 3300 BG spore-containing particles.51 In a later publication, the application of the Py-GC/IMS method was shown to be able to discriminate between aerosols of a Gram-positive spore (BG), a Gram-negative bacterium (Erwinia herbicola, EH), and a protein (ovalbumin) shown in Figure 7–6.52 Recent extensions of Py-GC/IMS have been made using pyrolysis with a gas chromatograph/differential mobility spectrometer, employing methods that generally paralleled those of Snyder et al. In this work by Eiceman et al.,54 the DMS detector provided differential mobility spectra simultaneously in positive and negative polarity. Three-dimensional profiles of ion intensity, compensation voltage, and retention time as shown in Figure 7–7 exhibited distinctive patterns that allowed the categorization of Gram-positive, Gram-negative, and spore forms of bacteria. The attraction

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_285... 06/05/2007

Página 1 de 1

Page 286

FIGURE 7–6 Results from monitoring by py-GC/IMS of ambient air following a release of biological aerosols. The upper frames show instrument response to spores of Gram-positive Bacillus subtilis var. globigii (BG), Gram-negative Erwinia herbicola (EH), and ovalbumin protein (OV). The dashed vertical lines represent the time boundaries of the aerosol releases. The lower frame shows results from samples collected and analyzed by the agar petri dish bacterial growth in units of ACPLA (viable bacteria component of the aerosol).

of the Py-GC/DMS instrument was the advantage of the micro-fabricated drift tube, including practical aspects of size, weight, and power, and the fundamental measurement facet of continuous ion analysis. Quantitative studies were made on the method, with direct application of bacteria in solution to a pyrolysis ribbon using a microliter syringe; 6000 bacteria were detected with an unoptimized inlet, and quantitative precision was ~10% relative standard deviation. Another approach to thermal processing of bacteria was demonstrated using a commercial mobility spectrometer where microgram quantities of whole bacterial cells were thermally desorbed in a heated anvil, producing complex patterns of positive and negative ion mobility spectra in a handheld IMS analyzer.55 The spectra differed reproducibly for different strains and species and for different conditions of growth, and can be used for the classification and differentiation of specific strains and species of bacteria, including pathogens. This provided a means to detect specific components of bacterial cells and to identify and classify bacteria within a minute without specialized test kits or reagents. Methods for improved ion-peak detection were also described for sequential sample desorption at stepped increases in temperature (programmed temperature ramping).

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_286... 06/05/2007

Página 1 de 1

Page 287

FIGURE 7–7 Topographic plots of compensation voltage versus retention time from the py-GC-DMS characterization of positive ions for E. coli (A), M. luteus (B), and B. megaterium (C). The intensity scale ranges from 0.9 V (white) to 2.5 V (black) in equal steps of 0.1 V.

7.5.2 Enzyme-Based Immunoassay IMS Ion mobility spectrometers have been employed as detectors for well-established methods for the determination of bacteria and enzyme-linked immunoassays (ELISA). In ELISA methods, primary antibodies attach to epitopes on the bacterial wall. Each antibody has a structure containing numerous epitopes that can be associated with a secondary antibody. The secondary antibody has also a region with enzymatic activity. This enzymatic region is able to react with a substrate to cleave a product that either is colored and can be determined by a spectrophotometer or is volatile and can be determined by headspace analysis. In the method developed by Snyder et al.47 and quantitatively explored by Smith et al.,56 the final product exhibited a distinctive negative production peak as observed with a mobility spectrometer. This was accomplished using a widely deployed military-grade CAM that was used without modification and suggested that it could serve as a potential bacteria analyzer provided reagent kits and an inlet adaptor were also distributed.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_287... 06/05/2007

Página 1 de 1

Page 288 In practice, a sample containing Bacillus cereus was placed in a vial where bacteria would adhere to the inner walls of the vial and the sample could be washed to remove impurities. In stages, reagents were added and sample washed to build the primary and secondary antibody structure. In the last step, ortho-nitrophenyl-beta-D-galactoside (ONPG) was added to the sample, where the ONPG reacted with a beta-galactosidase of the secondary antibody. The product of this reaction was ONP, a volatile compound that can be thermally desorbed into an IMS analyzer. Responses were compared to those of the conventional spectrophotometric assay. Both detection techniques produced a sigmoid-shape curve characteristic of immunoassay experiments. The bacterial detection limit with the IMS technique was estimated at below 1000 cells for an 8-min assay time. The only advantage of the ELISA-coupled IMS method is that the mobility spectrometer is capable of better detection limits than an optical spectroscopic method. This will have the benefit of reduced times for an assay or improved detection limits at fixed reaction times. The method has not been accepted by ELISA suppliers, perhaps because the existing ELISA methods are seen as sufficient to meet existing needs.

7.6 Conclusion Although mobility spectrometers are often defined as vapor analyzers, the record of the past decade demonstrates that biomolecules of high molecular mass and low volatility can also be characterized for gas-phase mobility. Such measurements do not simply replace a mass spectrometer but provide details about the molecule, now an ion, in ways not possible using any other methods or principles of measurement. Thus, a mobility spectrometer can uniquely offer insights into the structure and properties of the biomolecule. These methods are currently research tools only. In contrast, the possibility of using a mobility spectrometer, specifically a field-dependent mobility analyzer, as a filter before a mass spectrometer to reduce chemical noise in MS determinations of biomolecules with an electrospray ion source is being offered commercially as an inlet for a mass spectrometer. Whether this method or any of the biological initiatives with mobility spectrometry will endure or grow is wholly speculative. Practical analytical methods such as pyrolysis GC/IMS or GC/DMS and ELISA-coupled IMS have been demonstrated and await further developments by manufacturers or those needing such tools. The clinical uses of IMS analyzers for diagnosing bacterial infections such as bacterial vaginosis may become a large market use, unlike any previous use of IMS technology in medicine. This remains to be developed commercially and will succeed or fail, probably in the next decade.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_288... 06/05/2007

Página 1 de 1

Page 289

References 1. Fenn, J.B., Electrospray wings for molecular elephants (Nobel lecture), Angew. Chem. 2003, 42, 3871–3894. 2. Tanaka, K., The origin of macromolecule ionization by laser irradiation (Nobel lecture), Angew. Chem. 2003, 42, 3861–3870. 3. Eiceman, G.A.; Shoff, D.B.; Harden, C.S.; Snyder, A.R; Martinez, P.M.; Fleischer, M.E; Watkins, M.L., Ion mobility spectrometry of halothane, enflurane and isoflurane anesthetics in air and respired gases, Anal. Chem. 1989, 61, 1093–1099. 4. Martinez-Sandoval, P., Atmospheric Pressure Ion Chemistry and Physiological Respiration of Halogenated Anesthetic Gases, M.S. thesis, New Mexico State University, Las Cruces, NM, May 1991. 5. Ruzsanyi, V.; Sielemann, S.; Baumbach, J.I., Determination of VOCs in human breath using IMS, Int. J. Ion Mobility Spectrom. 2002, 5, 45–48. 6. Xie, Z.; Sielemann, S.; Schmidt, H.; Li, F.; Baumbach, J.I., Determination of acetone, 2-butanone, diethyl ketone and BTX using HSCC-UV-IMS, Anal. Bio-anal. Chem. 2002, 372, 606–610. 7. Lord, H.; Yu, Y.F.; Segal, A.; Pawliszyn, J., Breath analysis and monitoring by membrane extraction with sorbent interface, Anal. Chem. 2002, 74, 5650–5657. 8. Miekisch, W.; Schubert, J.K.; Vagts, D.A.; Geiger, K., Analysis of volatile disease markers in blood, Clin. Chem. 2001, 47, 1053–1060. 9. Sobel, J.D., Vaginitis, New Engl. J. Med. 1997, 37, 1896–1903. 10. Karpas, Z.; Chaim, W.; Tilman, B.; Gdalevsky, R.; Lorber, A., Diagnosis of vaginal infections by ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2002, 5, 49–54. 11. Karpas, Z.; Chaim, W.; Gdalevsky, R.; Tilman, B.; Lorber, A., A novel application for ion mobility spectrometry: diagnosing vaginal infections, Anal. Chim. Acta 2002, 474, 115–123. 12. Chaim, W.; Karpas, Z., Lorber, A., New technology for diagnosis of bacterial vaginosis, European J. Obst. Gynec. Reprod. Biol. 2003, 111, 83–87. 13. Amsel, R., Nonspecific vaginitis: diagnostic criteria and microbial and epidemiological associations. Am. J. Med. 1983, 74, 14–22. 14. Brand, J.M.; Galask, R.P., Trimethylamine: the substance mainly responsible for the fishy odor often associated with bacterial vaginosis, Obstet. Gynec. 1986, 63, 682–685. 15. Snyder, A.P.; Harden, C.S.; Davis, D.M.; Shoff, D.B.; Maswadeh, W.M., Hand-portable gaschromatography ion mobility spectrometer for the determination of the freshness of fish, 3rd International Workshop on Ion Mobility Spectrometry, Galveston, Texas, October 16–19, 1994. pp. 146–166. 16. Karpas, Z.; Tilman, B.; Gdalevsky, R.; Lorber, A., Determination of volatile biogenic amines in muscle food by ion mobility spectrometry (IMS), Anal. Chim. Acta 2002, 463, 155–163. 17. Strachan, N.J.C.; Nicholson, F.J.; Ogden, I.D., An automated sampling system using ion mobility spectrometry for the rapid detection of bacteria, Anal. Chim. Acta 1995, 313, 63–67. 18. Ogden, I.D.; Strachan, N.J.C., Applications of ion mobility spectrometry for food analysis, Special Publication—Royal Society of Chemistry, Biosensors for Food Analysis, 1998, 167–162.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_289... 06/05/2007

Página 1 de 1

Page 290 19. Ruzsanyi, V.; Sielemann, S.; Baumbach, J.I., Determination of microbial volatile organic compounds MVOC using IMS with different ionization sources, Int. J. Ion Mobility Spectrom. 2002, 5, 138–142. 20. Ruzsanyi, V.; Baumbach, J.I.; Eiceman, G.A., Detection of the mold markers using ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 2003, 6, 53–57 21. Clark, J. M; Daum, K.A.; Kalivas, J.H., Demonstrated potential of ion mobility spectrometry for detection of adulterated perfumes and plant speciation, Anal. Lett. 2003, 36, 215–244. 22. Vautz, W.; Sielemann, S.; Baumbach, J.I., Qualitative detection of odours using ion mobility spectrometry, 12th International Conference on Ion Mobility Spectrometry, Umeå, Sweden, July, 27–31, 2003. 23. Purves, R.W.; Guevremont, R., Electrospray-ionization high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal. Chem. 1999, 71, 2346–2357. 24. Purves, R.W.; Barnett, D.A.; Ells, B.; Guevremont, R., Gas-phase conformers of the [M+2H]2+ ion of Bradykinin investigated by combining high-field asymmetric wave-form ion mobility spectrometry, hydrogen/deuterium exchange, and energy-loss measurements, Rapid Commun. Mass Spectrom. 2001, 15, 1453–1456. 25. Gabryelski, W.; Froese, K.L., Rapid and sensitive differentiation of anomers, linkage, and position isomers of disaccharides using high-field asymmetric waveform ion mobility spectrometry FAIMS, J. Am. Soc. Mass Spectrom. 2003, 14, 265–277. 26. Valentine, S.J.; Kulchania, M.; Barnes, C.A.S.; Clemmer, D.E., Multidimensional separations of complex peptide mixtures—a combined high-performance liquid chromatography/ion mobility/time-of-flight massspectrometry approach, Int. J. Mass Spectrom. 2001, 212, 97–109. 27. Hoaglund, C.S.; Valentine, S.J.; Clemmer, D.E., An ion-trap interface for ESI-ion mobility experiments, Anal. Chem. 1997, 69, 4156–4161. 28. Lee, D.S.; Wu, C.; Hill, H.H., Detection of carbohydrates by electrospray-ionization ion mobility spectrometry following microbore high-performance liquid-chromatography, J. Chromatogr. A 1998, 822, 1–9. 29. Hoaglund, C.S.; Valentine, S.J.; Sporleder, C.R.; Reilly, J.P.; Clemmer, D.E., 3-Dimensional ion mobility TOFMS analysis of electrosprayed biomolecules, Anal. Chem. 1998, 70, 2236–2242. 30. Koomen, J.M.; Ruotolo, B.T.; Gillig, K.J.; McLean, J.A.; Russell, D.H.; Kang, M.J.; Dunbar, K.R.; Fuhrer, K.; Gonin, M.; Schultz, J.A., Oligonucleotide analysis with MALDI-Ion-Mobility-TOFMS, Anal. Bioanal. Chem. 2002, 373, 612–617. 31. Srebalus, C.A.; Clemmer, D.E., Assessment of purity and screening of peptide libraries by nested ion mobility TOFMS—Identification of RNase S-protein binders, Anal. Chem. 2001, 73, 424–433. 32. Counterman, A.E.; Clemmer, D.E., Cis-trans signatures of proline-containing tryptic peptides in the gasphase, Anal. Chem. 2002, 74, 1946–1951. 33. Srebalus, C.A.; Li, J.W.; Marshall, W.S.; Clemmer, D.E., Gas-phase separations of electrosprayed peptide libraries, Anal. Chem. 1999, 71, 3918–3927. 34. Hudgins, R.R., Conformations of GlynH + and ALAnH + peptides in the gas-phase, Biophys. J. 1999, 76, 1591–1597. 35. Beegle, L.W.; Kanik, L; Matz, L.; Hill, H.H., Effects of drift-gas polarizability on glycine peptides in ion mobility spectrometry, Int. J. Mass Spectrom. 2002, 216, 257–268.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_290... 06/05/2007

Página 1 de 1

Page 291 36. Wu, C; Siems, W.F.; Klasmeier, J.; Hill, H.H., Separation of isomeric peptides using electrospray ionization/high-resolution ion mobility spectrometry, Anal. Chem. 2000, 72, 391–395. 37. Wu, C; Klasmeier, J.; Hill, H.H., Atmospheric-pressure ion mobility spectrometry of protonated and sodiated peptides, Rapid Commun. Mass Spectrom. 1999, 13, 1138–1142. 38. Wyttenbach, T.; Vonhelden, G.; Bowers, M.T., Gas-phase conformation of bio-logical molecules— Bradykinin, J. Am. Chem. Soc. 1996, 118, 8355–8364. 39. Li, J.W.; Taraszka, J.A.; Counterman, A.E.; Clemmer, D.E., Influence of solvent composition and capillary temperature on the conformations of electrosprayed ions—unfolding of compact ubiquitin conformers from pseudonative and denatured solutions, Int. J. Mass Spectrom. 1999, 187, 37–47. 40. Hudgins, R.R.; Woenckhaus, J.; Jarrold, M.F., High-resolution ion mobility measurements for gas-phase proteins—correlation between solution-phase and gas-phase conformations, Int. J. Mass Spectrom. 1997, 165, 497–507. 41. Wyttenbach, T.; Batka, J.J.; Gidden, J.; Bowers, M.T., Host/guest conformation of biological systems: valinomycin/alkali ions, Int. J. Mass Spectrom. 1999, 193, 143–152. 42. Wyttenbach, T.; Kemper, P.R.; Bowers, M.T., Design of a new electrospray ion mobility mass spectrometer, Int. J. Mass Spectrom. 2001, 212, 13–23. 43. Karpas, Z.; Berant, Z.; Stimac, R.M., An IMS/MS study of the site of protonation in anilines, Struct. Chem. 1990,1, 201–204. 44. Karpas, Z., Evidence for proton-induced cyclization in αw diamines from mobility measurements, Int. J. Mass Spectrom. Ion Proc. 1989, 93, 237–242. 45. Cox, K.A.; Julian, R.K.; Cooks, R.G.; Kaiser, R.E., Conformer selection of protein ions by ion mobility in a triple quadrupole mass spectrometer, J. Am. Soc. Mass Spectrom. 1994, 5, 127–136. 46. Snyder, A.P.; Shoff, D.B.; Eiceman, G.A.; Blyth, D.A.; Parsons, J.A., Detection of bacteria by ion mobility spectrometry, Anal. Chem. 1991, 63, 526–529. 47. Snyder, A.P.; Blyth, D.A.; Parsons, J.A., Ion mobility spectrometry as an immunoassay detection technique, J. Microbiol. Methods 1996, 27, 81–88. 48. Snyder, A.P.; Harden, C.S.; Brittain, A.H.; Kim, M.G.; Arnold, N.S.; Meuzelaar, H.L.C., Portable handheld gas chromatography/ion mobility spectrometry device, Anal. Chem. 1993, 65, 299–306. 49. Tripathi, A.; Maswadeh, W.M.; Snyder, A.P., Optimization of quartz tube pyrolysis atmospheric-pressure ionization mass spectrometry for the generation of bacterial biomarkers, Rapid Commun. Mass Spectrom. 2001, 15, 1672–1680. 50. Snyder, A.P.; Maswadeh, W.M.; Parsons, J.A.; Tripathi, A.; Meuzelaar, H.L.C., Dworzanski; J.P.; Kim, M.G., Field detection of bacillus spore aerosols with stand-alone pyrolysis-gas chromatography-ion mobility spectrometry, Field Anal. Chem. Technol. 1999, 3, 315–326. 51. Snyder, A.P.; Tripathi, A.; Maswadeh, W.M.; Ho, J.; Spence, M., Field detection and identification of a bioaerosol suite by pyrolysis-gas chromatography-ion mobility spectrometry, Field Anal Chem. Technol. 2001, 5, 190–204. 52. Snyder, A.P.; Maswadeh, W.M.; Tripathi, A.; Dworzanski, J.P., Detection of Gram-negative ErwiniaHerbicola outdoor aerosols with pyrolysis-gas chromatography-ion mobility spectrometry, Field Anal. Chem. Technol. 2000, 4, 111–126.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_291... 06/05/2007

Página 1 de 2

Page 292 53. Snyder, A.R; Maswadeh, W.M.; Tripathi, A.; Eversole, J.; Ho, J.; Spence, M., Orthogonal analysis of mass- and spectral-based technologies for the field detection of bioaerosols, Anal. Chim. Acta, 2004, 513, 365–377. 54. Schmidt, H.; Tadjimukhamedov, F.; Mohrentz, I.V.; Smith, G.B.; Eiceman, G.A., Microfabricated differential mobility spectrometry with pryrolysis gas chromatography for chemical characterization of bacteria, Anal. Chem. 2004, 76, 5208–5217. 55. Vinopal, R.T.; Jadamec, J.R.; Defur, P.; Demars, A.L.; Jakubielski, S.; Green, C.; Anderson, C.P.; Dugas, J.E.; Debono, R.F., Fingerprinting bacterial strains using ion mobility spectrometry, Anal. Chim. Acta 2002, 457, 83–95. 56. Smith, G.B.; Eiceman, G.A.; Walsh, M.K.; Critz, S.A.; Andazola, E.; Ortega, E.; Cadena, F., Detection of Salmonella typhimurium by hand-held ion mobility spectrometer: a quantitative assessment of response characteristics, Field Anal. Chem. Technol. 1997, 4, 213–226.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_292... 06/05/2007

Página 1 de 1

Page 293

8 Developed Applications in Industrial and Environmental Monitoring 8.1 Introduction Ion mobility spectrometry (IMS) has been utilized for chemical measurements in several industrial and environmental applications in which the understanding of IMS behavior may be regarded as reasonably complete. However, these applications are generally limited in scope or acceptance and do not involve large numbers of analyzers, unlike the military and security applications of IMS in which tens of thousands of instruments are in use. The discussion of these uses, which are called developed applications here, serves to highlight the value and possibilities of IMS technology. Nonetheless, the impact of IMS on measurement science or technology in any of these developed applications is, as yet, limited. There are several reasons for this, including resistance to new technology, satisfaction with existing and competing methods, and the limited availability of instruments or software for mobility spectrometers. In other instances, the principles of IMS favor development of an application, and a few limited studies have demonstrated the utility or capability of IMS for an application. These are termed feasible applications and are discussed in Chapter 9. The category of developed applications may be divided into two distinct groups: environmental and industrial. In both categories, IMS analyzers are used mainly as monitors of concentrations of specific chemicals or simple chemical mixtures in ambient air, water, and soil. A common feature is that in unique challenges mobility spectrometers are able to meet the measurement needs that cannot be satisfied by other means.

8.2 Acidic and Corrosive Gases Acid gases—in particular, hydrogen chloride (HCl) and hydrogen fluoride (HF)—are emitted in several industrial processes including aluminum refining, ceramics manufacture, coal combustion, and plastic waste incineration,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_293... 06/05/2007

Página 1 de 1

Page 294 as well as other chemical processes. Due to the toxicity and corrosiveness of these acid gases, a safety hazard exists within the plant, and environmental pollution is a concern outside the perimeter. Monitoring the source of emission for HCl and HF is required by federal, state, and often also by local agencies, and regulations from the Environmental Protection Agency specify permissible concentrations. The reactive nature of these acid gases, the environmental conditions that exist in a stack, and the low detection limits to be monitored (ppm to ppb range) make continuous monitoring a formidable measurement task. The determination of HF using IMS should be favored by the electron affinity of HF but is complicated by the overlap between the product-ion peak for HF and the peak for the reactant ion, commonly O2−. The problem with resolution or ion separation was solved by a team at the Environmental Technology Group, now reformed as Molecular Analytics in Sparks, MD, and is based on the addition of methyl salicylate into the internal atmosphere of the drift tube. Methyl salicylate forms an adduct ion with O2− so that the peak for oxygen is replaced with a peak for the adduct ion, which is located at long drift times well displaced from that for HF. The product ion is described as (HF)3F− and is separated from the adduct ion of methyl salicylate* O2− as shown in Figure 8–1. The air sample is drawn over a semipermeable membrane that serves to protect the interior of the cell from particles and high moisture levels, and provides a degree of selectivity through differences in permeation rates. The molecules of interest permeate through the membrane and are carried into the drift tube by a carrier gas flow that sweeps the interior side of the membrane. Vapors are then ionized and separated according to their drift time and detected. Interest in monitoring HF arise not only from occupational hygiene and environmental concerns but also from the interest in hydrogen fluoride as a precursor in the production of chemical warfare agents. Thus, HF is considered a dual-use chemical and is important within the sphere of counterterrorism. Another interest in monitoring the levels of HF and HCl in industrial processes with gases such as hydrocarbons, vinyl chloride, chlorofluorocarbons, and other chemicals is the deleterious effect of the acid gases on the industrial catalysts. The apparatus described earlier may be used to continuously monitor HCl and HF and help prevent corrosion and the poisoning of catalysts.1 The presence of halogenated compounds such as chlorine dioxide (ClO2), together with chlorine in air, creates interferences and inaccurate results with electrochemical detectors that lack specificity in response.2 Indeed, both of these gases have a similar odor, but ClO2 is more toxic and much more soluble in solutions. Gordon et al. used an IMS analyzer to study the rate of emission and volatilization of ClO2 and predict exposure to levels of ClO2 in ambient air.2 Bromine and some brominated compounds have also been monitored in ambient air at a chemical plant in Israel.3 In the IMS drift tube, bromide ions (Br−) are usually the only stable ions of negative polarity that are formed from organobromine compounds. However, Br3− was formed from molecular

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_294... 06/05/2007

Página 1 de 1

Page 295

FIGURE 8–1 Mobility spectra in negative polarity for two concentrations of hydrogen fluoride (HF) with methyl salicylate as the reagent gas and a control sample with no HF. The dotted line is placed on the reactant ion, an O 2− adduct of methyl salicylate.

bromine as an adduct ion of Br2 and Br− when bromine levels were at an elevated concentration. In this way, at high vapor concentrations, molecular bromine (Br2) can be distinguished from HBr and other bromine-containing compounds. Chemical reactions between corrosive acid gases and the metallic radioactive foil (63Ni) may lead to the formation of nickel halides on the surface of the foil. Such salts are mechanically unstable and could become aero-solized and spread throughout the drift tube. Their particles may also be released into ambient environments with unfiltered exhaust flows from the drift tube. To prevent such releases, the drift tubes for acid gas monitoring should be designed to prevent contact between the acid gases and the radioactive metal foil. In one approach, the ion source is configured as a short region in which reactant ions are formed and which is isolated from the reaction region where these ions are brought into contact with the corrosive gas sample. In the ion source and the region between the source and the reaction region, only reactant ions exist because this volume is continuously purged with clean air without sample vapors. Ions are moved

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_295... 06/05/2007

Página 1 de 1

Page 296 in the clean flow by the voltage gradient toward the reaction region where they interact with the sample gas and form product ions. All vapors are vented from the drift tube without entering the drift region or contacting the ion source. Product ions formed in the reaction region are moved under the influence of the electric field into the drift region for mobility analysis. Thus, acid gases are prevented from making contact with the radioactive source, maintaining safe operation of the analyzer.

8.3 Volatile Organic Compounds and Halocarbons Volatile organic compounds (VOCs) include a large variety of aliphatic and aromatic hydrocarbons, halocarbons, and oxygenated hydrocarbons such as ketones, aldehydes, esters, and alcohols. In ambient air, some of these are potentially explosive or flammable, and the vapors of several common compounds such as benzene, chloroform, and carbon tetrachloride are considered as potential carcinogens or health hazards through inhalation exposure. The presence of some VOCs in soil, rivers, or reservoirs is seen as hazardous, endangering the quality of drinking water and ecosystems. Thus, environmental monitoring of VOCs may involve sampling air, water, or soil. A stand-alone IMS instrument would not be able to simultaneously monitor the constituents of a mixture that would be present in an actual environmental sample due to the mutual interferences from competitive ionization (see Chapter 3). An effective analytical strategy is to combine an ion mobility spectrometer with fast gas chromatograph for separation of the components of the sample before the IMS drift tube, which, together, serve as a “smart” detector, as described in Chapter 5. In the following sections, a few examples are described to demonstrate the capabilities of IMS for the monitoring of VOCs in environmental venues. An example of a fast chromatographic separation before an IMS detector can be seen in the combination of a multicapillary column (MCC) and drift tube equipped with a 10.6-eV photodischarge lamp. This combination enhanced the mobility analysis such that overlapping peaks in mobility spectra could be resolved and intermolecular charge-transfer reactions, which complicate the interpretation of the spectra, could be reduced. This was demonstrated in the analysis of a mixture of halogenated compounds (trans-1,2-dichloroethene, trichloroethene, and tetrachloroethene), that were efficiently separated at ambient temperature in approximately 1 min.4 A similar technique was used to detect methyl tertbutyl ether (MTBE), which is a gasoline additive, and benzene, toluene, and m-xylene (BTX) in water and nitrogen.5,6 In one study,5 two ionization sources were utilized: a radioactive 63Ni source and UV photoionization source. The compounds permeated through a silicone membrane and were thus extracted from the water sample. These substances were clearly separated according to their differences in retention times and

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_296... 06/05/2007

Página 1 de 1

Page 297 drift times in two-dimensional data analyses (see Chapter 5). Detection limits for MTBE from the entire method were 2 μg/l with the UV ion source and 30 pg/l with the 63Ni ion source in nitrogen for gas samples, and 20 mg/l (UV) and 1 μg/l (63Ni) in water with relative standard deviations of between 2.9 and 9%. The method was proven to be suitable for near-real-time monitoring as the total analysis time was <90 sec. Aliphatic ketones such as acetone, 2-butanone, diethyl ketone, and volatile aromatics including benzene, toluene, and xylene were monitored in air and exhaled breath (Chapter 7) using a high-speed capillary column7. Method detection limits were around 3 μg/l for these ketones in nitrogen, and the linear dynamic range was 4 to 320 μg/l. The reduced mobility values of the ions derived from several halogenated compounds were reported by Leonhardt et al.8 A different approach to on-site detection of MTBE in water using IMS without a GC column was reported by Stach et al.9 The peak of the MTBE proton-bound dimer, with a reduced mobility constant Ko=1.50 cm2 V−1sec−1, was used to measure MTBE in water in the range >30 μg/l. The method was based on binding the water vapor that is present in the headspace vapor to an adsorbent (EXtrelut) and injecting the remaining VOCs into a handheld ion mobility spectrometer. Interferences by hydrocarbons (alkanes, alkenes, and cycloalkanes) can be neglected due to their low proton affinity relative to that of MTBE. The time required for the determination of MTBE, including sample preparation, was approximately 5 min. The interesting aspects of this preparation method are that the major constituent of the matrix and the main interferent (water vapor) is removed and that MTBE, with a high proton affinity compared to the remaining organic compounds, is preferentially ionized. Detection of trace amounts of perfluorocarbons (PFCs, C5F12 to C9F20) in air is of interest in industrial and environmental applications.10 The mobility spectra of these compounds do not exhibit a single distinct product-ion peak, as usually observed in measurements of halogenated or nitro compounds, and a number of peaks that partially overlap are seen in the mobility spectra. The number of product-ion peaks varied from three (for C6F14) to nine (for C7F16). This phenomenon is not due to mixtures or multicomponent samples but to the ionization chemistry of these PFCs. To simplify the interpretation of the mobility spectra, a method of subtracting the mobility spectrum of the sample from that of the background was used and is shown in Figure 8–2. The difference mobility spectrum helps to reduce interferences in spectral interpretation and improves the detection limits that were reported to be in the upper ng/l range.10 A novel photoemissive electron source that serves as a selective ionization source for compounds that can capture electrons with quasi-thermal energy was developed for use in IMS by Walls et al.11 Pulsed UV irradiation (from a flashlamp or laser) impinging on a metal layer causes emission of electrons in distinct packages such that gating is not required. These electrons are captured first by oxygen molecules in air, forming O2− ions that can subsequently transfer an electron to molecules with high electron affinity, such as halocarbons and nitro compounds. The negative ions were characterized by the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_297... 06/05/2007

Página 1 de 1

Page 298

FIGURE 8–2 Mobility spectrum of perfluorooctane in air with negative polarity (top traces) and with methods of background subtraction (lower trace).

IMS, and real-time detection limits of low ppmv were obtained for chlorinated species. Vinyl chloride and other chlorinated and aromatic compounds were determined in air using GC/IMS by Simpson et al.12 A detection limit of 2 ppbv was reported for continuous monitoring of vinyl chloride, and this could be improved by two orders of magnitude with preconcentration of sample on an adsorbent trap. Trace amounts of benzene, toluene, ethylbenzene, and xylene (BTEX) were detected in water using a tubular silicone membrane interface with a portable IMS.13 Toluene was determined at 0.101 mg/l in purified water and corresponded to a headspace concentration of 2.75 μg/m3 with static sampling. Toluene at this level was not detected by headspace sampling without the membrane interface. These methods provided the capability to monitor trace concentrations of gasoline components in river water with a response time of several seconds. In principle, IMS can be used to measure hydrocarbon compounds despite their low proton affinity, provided that moisture and other interfering compounds are removed from the sample. Borsdorf et al. have shown that different normal and branched alkanes, potential contaminants from the petrochemical industry, can be measured using an IMS analyzer with a corona discharge ionization source.14 Polychlorinated biphenyls (PCBs) used mainly in transformer oil are environmentally important due to their suspected toxicity Ritchie and Rudolph15

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_298... 06/05/2007

Página 1 de 1

Page 299 used an IMS detector in the negative polarity to determine PCB isomers (congeners) having five or more chlorine atoms at detection levels of 35 ng by extracting them into an isooctane solution. Mixtures of up to four PCB congeners showed characteristic multiple peaks, although detection of Aroclors in transformer oil was suppressed by the presence of the antioxidant BHT (2,6-di-t-butyl-4methylphenol) in the oil. The wood preservative pentachlorophenol was easily detected in recycled wood shavings at levels of 52 ppm with the extrapolated minimum detectable level in the low-ppm region.15 IMS is ideally suited for monitoring volatile organic compounds such as amines, ethers, alcohols, and ketones that have proton affinity values higher than that of water. Under these conditions, product ions are readily formed in positive polarity, and mobility spectra are distinctive. The complications due to intricate ionization chemistry when a mixture is present in the drift tube can be resolved today with high-speed gas chromatography, which serves to prefractionate the sample. When chemicals such as alkanes and aromatic hydrocarbons are target analytes, ionization sources such as photo-discharge lamps may provide necessary ionization methods replacing the traditional source, 63Ni. Volatile organic compounds that contain halogen atoms, particularly chlorine or bromine, generally form halide fragment ions in negative polarity. In some cases, molecular or quasimolecular ions may also be formed, such as Br3− in elemental bromine. Fluorocarbons do not form F− ions that live long enough to be observed in mobility spectra, but they were reported to form several ions whose identity has not been established by MS. Processing the mobility spectra by mathematical procedures was required to derive quantitative data from the measurements. The versatility of IMS for these applications is manifested by the ability to monitor the concentration of VOCs in samples of air, water, and soil.

8.4 Ammonia in Water, Air, Clean Rooms, and Process Streams Levels of dissolved ammonia in water have been determined, using mobility spectrometers, by thermally extracting the ammonia through a silicon membrane and passing the vapors into the analyzer.16 This approach to sampling provided a limit of detection of 1.2 mg/l, although lower concentrations were attainable. The importance of accurately defining the chemical state of the analyte was emphasized, and the measurement was affected by the pH of the sample. The system was said to be capable of monitoring accurately changes in aqueous ammonia concentration over a period of approximately 24 h and artifacts attributable to memory effects within the membrane were

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_299... 06/05/2007

Página 1 de 1

Page 300 not observed.16 The instrumentation and method handling provided ammonia determination even in biologically active and polluted waters. As might be expected, sample processing required consumables to adjust pH and to prevent biofouling of tubing and the membrane. The entire apparatus was housed in a cabinet and required periodical recharging of reagents. Although this approach does not illustrate a high portability of IMS technology, the method is possible where others fail because of the specificity of ionization to ammonia. Ammonia can be determined in air by IMS, and its detection has also been described in conjunction with hydrazine and hydrazine derivatives.17 In that study, the presence of ammonia in the sampled air was considered as interfering with the measurement of hydrazines through similarities in drift times for the protonated monomers of each hydrazine. The addition of 5nonanone throughout the internal atmosphere of the drift tube led to the formation of adduct ions between the nitrogen bases, and the ketone and product ions were resolvable. Ammonia was clustered by the highest number of ketone molecules and consequently appeared at the longest drift time. These measurements were made using a handheld ion mobility spectrometer that was equipped with a membrane inlet, an 63Ni ion source, and reagent-gas reservoir, and the drift tube was operated at ambient temperature. Peak shape and drift times for each ion were influenced by the concentration of ketone vapor levels in the drift tube, and elevated levels of ketone vapor were necessary to form and retain the cluster ions. The lowest level of ketone in the drift gas necessary to observe resolution between the product ions was about 1 ppm. Other ketones were examined as modifiers of the atmosphere but showed inferior performance compared to 5-nonanone. Though some steric contribution to these differences was expected, the main difference between the ketones was thought to be the practical issue of vapor pressure and concentrations.18 Mass spectra that were obtained in a later follow-up study demonstrated that the number of ketone adducts was governed by the number of protons on the central nitrogen atom and decreased in the order: ammonia>hydrazine>monomethylhydrazine.19 A handheld IMS analyzer was flown on a space shuttle to determine hydrazine contamination on spacesuits without ammonia interferences.17 Low levels of ammonia in ethylene, hydrogen, and other light hydrocarbons in industrial processes can cause downstream process problems due to catalyst poisoning and other factors. An IMS system was reported to monitor NH3 in hydrocarbons with a limit of detection of 1 ppb.1 The analyzer was used in a wide variety of different process streams without any evident interference from coexisting compounds. In another report, NH3 in ethylene was measured by IMS with a polymer membrane inlet, 63Ni ion source, H+(H2O)n reactant ion, and nitrogen as the drift and source gas.20 Because ethylene has no noticeable effect on the analytical results, preconcentration or preseparation were unnecessary. Ethylene’s flammability was made inconsequential by the nitrogen atmosphere inside the spectrometer. Response to

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_300... 06/05/2007

Página 1 de 1

Page 301 NH3 concentrations between 200 and 1500 ppb was nearly linear, and the calculated minimum detection limit was 25 ppb.20 The levels of ammonia in the ambient air inside semiconductor plants, particularly in “clean areas” used for deep ultraviolet (DUV) photoresist and lithographic processes, are important because ammonia and 1-methyl-2-pyrrolidone can affect the acid catalysis of photoresist formulations. IMS can provide the necessary analytical capabilities to monitor both compounds at low ppb levels. Compared with GC, the time required for a measurement with IMS was significantly shorter than the 10 min required for thermal desorption/GC.21,22 Further discussion of IMS applications for other gases in the semiconductor production industry is given in Section 8.7.

8.5 Gas Purity and Trap Efficiency Argon and helium, as inert gases, are transparent with conventional analytical IMS analyzers in both positive- and negative-polarity, and the appearance of any ion peak in the mobility spectrum would indicate the presence of impurities in these important industrial gases. This feature has been utilized for detection of trace amounts of oxygen in nitrogen by using argon as the drift gas.23 Normally, the detection limit and accuracy for determining O2 in N2 by IMS suffer from poor resolution. However, product ions for oxygen and nitrogen formed by ionization chemistry in an argon atmosphere can be quantitatively determined in mobility spectra. A few patent applications have been submitted for detection of impurities in helium, nitrogen, hydrogen, and oxygen.24,25 The proposed method is for the quantitative analysis of the impurities in these gases, in which a pure gas (usually argon) is mixed, as a counterflow gas in the drift region of the IMS, with the gas to be analyzed. Impurities form ions that are not present in the pure gases, and can be identified and quantified on the basis of drift times and peak areas. The technique of monitoring impurities in gases by IMS has implications for other challenges within industrial processes. A problem in the use of absorbent traps for purifying gases in industry and for preconcentration in analytical chemistry is sample breakthrough, a situation in which a trap no longer retains impurities or analyte. Knowledge of this change in trap efficiency is usually expressed as the volume of gas when retention is changed. An IMS analyzer can be used to continuously monitor for contaminants in gas flow downstream from a trap and to test the effectiveness of different desorption methods (thermal desorption and solvent extraction or solvent vapor extraction). An IMS analyzer was used to follow the desorption of pentanone from a Tenax absorbent trap and to test the validity of a computer simulation of the process.26 Desorption of retained compounds was affected

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=Page_30...

06/05/2007

Página 1 de 1

Page 302 by competitive adsorption with a concentrated solvent vapor. Ion mobility spectra disclosed that pentanone was rapidly released from the adsorbent trap with an accompanying sixfold enrichment; moreover, desorption by the solvent vapor was more efficient than conventional methods of thermal desorption.

8.6 SF6 Purity in Electrical Switches One compelling application of IMS, in which no other technology is able to complete the measurement as well and as cost effectively as a drift tube, is the monitoring of gas purity of gasinsulated switches (GIS) in high-voltage substations.27–29 Sulfur hexafluoride (SF6) is used as an insulating gas inside switches, and may undergo chemical decomposition due to partial arcing and electrical discharge processes. This will lead to the formation of corrosive and toxic by-products. Accumulation of these impurities over time may result in the eventual and unpredictable failure of a switch; sudden failure of these switches may result in damage to the switch, power substation, and economic well-being of customers. Consequently, knowledge of the concentrations of the decomposition products inside a switch can provide a timely warning about the deteriorating performance of the switch. Mobility spectrometers have been developed to reside inside a switch and provide routine measures of gas purity. In this application, ions of the sample and impurities are made together and characterized in the same atmosphere, which is the sample. This leads to compromises in the resolution of the IMS, and changes in the gas composition are seen as shifts in the position of the main peak of the mobility spectrum (Figure 8–3). Such changes are correlated to the concentration of decomposition products formed in SF6, and a trend of drift time vs. calendar time can alert maintenance staff of an impending failure of the switch. Corrective action can be taken in time, and overall damage to the substation may be prevented. Comparatively simple low-resolution IMS analyzers permit each switch to be equipped with a drift tube. In a comparative test, the IMS detectors performed better than the standard colorimetric detector tubes by a factor of 2 to 4 and provided automated, highspeed response with MDL values of 20 ppm, sufficient to detect the impurities formed in a GIS. All measurements were validated using IR spectrometry (FTIR) as an independent method to determine the decomposition products.27 The gas composition in circuit breakers in gas-insulated substations during operation and the impurities in the reclaimed gas after a recycling procedure were examined. Results from the investigation of 36 different circuit breakers in an operating substation were presented, as well as ideas for new methods to check the fill-gas quality.28

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_302... 06/05/2007

Página 1 de 1

Page 303

FIGURE 8–3 Mobility spectra of gas inside circuit breakers from the three phases of the breaker (top frame). The bottom frame shows the classification of a circuit breaker based upon peak shifts in the mobility spectra of gases inside a breaker.28

8.7 Semiconductor Manufacturing The usefulness of IMS, or plasma chromatography as it was then called, to monitor contaminants in the manufacture of semiconductors was first explored by Carr at IBM in the 1970s and is summarized in the monograph he edited, titled Plasma Chromatography.30 At that time, several techniques were available within the electronics industry to determine impurities through elemental analysis, but methods were not available to measure

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_303... 06/05/2007

Página 1 de 1

Page 304 organic contaminants at ultra-trace levels on the surfaces of silicon wafers and other materials used in the production of semiconductors. A plasma chromatograph/mass spectrometer could meet the required detection limits because volatile contaminants were retained in the ambient pressure inlet where the sample was introduced for analysis. In contrast, when vacuum-based technologies were employed, important chemical information was lost through volatilization of contaminants. Both positive and negative ions were measured for as complete an evaluation of the sample as possible. The cleanliness of surfaces between steps in the manufacturing process could be examined as an assurance of production quality. The technique was also used to analyze headspace vapors contained inside sealed electronic packages that were used for storing components and transporting the products.30 Applications of IMS in semiconductor technology has also included the analysis of headspace vapors in sealed packages and failures in subcomponents, checks on the efficiency of cleaning and etching, characterization of process media, and surveys of the atmosphere of clean rooms. By the late 1990s, IMS technology was applied for other uses in the semi-conductor industry as reported mainly by Budde and colleagues at Siemens31–37 and by other groups in Germany.38–40 In one study, an IMS/MS instrument was used to detect volatile materials that were emitted or outgassed from wafer storage and transport boxes. These vapors were identified as plasticizers and were attributed to the polymer additives in the box material.31 Large variations of up to four orders of magnitude were found in the quantity of vapors emitted by various storage boxes with clear implications for quality control in the manufacturing of electronic components. Furthermore, photoresist solvents and other vapors originated from used containers and process media. In followup studies, enhanced stress testing was performed at elevated temperatures for polypropylene, polycarbonate, polytetrafluoro-ethylene, perfluoroalkoxy polymer, polyvinylidene fluoride, and acrylonitrile-butadiene-styrene copolymer.33 Contamination can arise from several sources in the manufacturing of electronic components, dictating a need for routine measurements. This was demonstrated by Budde’s study of organic contamination from wafer boxes, wafer carriers, pods, clean room air filters and filter frames, clean room paper, sealing foils, and other polymeric materials.34 As semiconductors are miniaturized further, the importance of impurities and the need for higher standards of cleanliness may be expected and could be satisfied by IMS analyzers as illustrated from manufacturing minienvironments.35 Most of the work described in the preceding text has been presented at conferences that were oriented toward the semiconductor industry, and a summary of these can be found in a journal article.36 The contamination of surfaces using mobility spectrometers was also described by Seng who was concerned with quality control in curing of surface films.38,39 An IMS analyzer was used to detect trace amounts of coating and ink components that remained unlinked during the curing processes. Photolithography is affected by airborne molecular contaminants that may be present in the manufacturing facility Kishkovich et al.41 presented

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_304... 06/05/2007

Página 1 de 1

Page 305 an overview of real-time monitoring methodologies, among them IMS, that can detect and measure these contaminants at low concentrations. Different monitoring strategies and monitoring technologies (ion chromatography, chemiluminescence, and IMS) were compared, with an emphasis on identification of critical points to monitor in tracks, steppers, clean rooms, and air filtration systems. In their conclusions, the authors appear to favor chemiluminescence over IMS for the purpose of control in photolithography.

8.8 Recirculated or Controlled Atmospheres Recirculated or controlled atmospheres here refer to arrangements in which the turnover of an air supply is limited and circulation of air is restricted by external conditions as with the space shuttle or a submarine or through operational constrictions in a clean room. In these circumstances, the level of some components of the ambient atmosphere must be monitored and controlled to prevent deleterious health effects to personnel and damage to equipment or manufacturing processes. Ion mobility spectrometers were found to be ideally suited for such applications, provided that the target compounds belong to a class that shows favorable ion chemistry in air at ambient pressure, i.e., good response in the analyzer. In this section, the versatility of IMS as a measurement technique is illustrated in a few examples. A unique example of an IMS application, air-quality monitoring on board the International Space Station by GC/IMS, is treated separately in Section 8.9. On a submarine, monoethanolamine (MEA) is used in the air purification system to remove carbon dioxide from the enclosed atmosphere. However, the level of MEA and other pollutants must be kept below certain well-defined exposure limits, and this necessitates continuous monitoring of vapor concentrations of MEA. Instrumentation is used on board for monitoring, and such analyzers must be convenient to operate, calibrate, and maintain. Moreover, demands on operations from consumables, size, weight, and cost should be minimized. Finally, instrumentation must resist, in both analytical response and reliability of construction, the high humidity levels present in submarine atmospheres. The group led by Bollan42 developed an IMS application for monitoring MEA near the maximum permissible concentration limit for a continuous 90-day period. The instrument was evaluated for stability, reproducibility, response time, recovery characteristics, and the effects of temperature and humidity in regard to the detection of MEA in an expected cocktail of probable contaminants. Ketones were used as reagent gases and modifiers of the drift gas17–19,43 to improve the detection of MEA in an atmosphere containing diesel fumes, Freon 22, and ammonia. The standard reagent gas for military handheld analyzers, acetone, did not show favorable discrimination for MEA. However, 4-heptanone in the internal atmosphere of the drift tube resulted in the formation of adduct ions

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_305... 06/05/2007

Página 1 de 1

Page 306

FIGURE 8–4 Mobility spectra for the detection of monoethanolamine (MEA) at 0.25 ppm in air with a trace of ammonia with 4-heptanone reagent gas (top frame) and in the presence of Freon-22 and diesel interferences (bottom frame).

between MEA and 4-heptanone and shifts in peak drift times. Changes in drift times for these adduct ions provide improved peak separation between MEA and interferences, and allowed MDLs of 5 ppb of MEA and several other alkanolamines. This is shown in Figure 8–4, in which spectra are shown for MEA in the presence of interferences. After the terrorist attack on the subway system in Japan in 1995, awareness grew over the threat to public safety by accidental or intentional releases of toxic vapors in ventilation systems of public facilities. One response to this has been the Environics EnviScreen Safety and Environment Control Systems.44 In this system, an array of detectors based upon the aspirator design found in the IMCELL technology is distributed throughout a protected area. The array is under the control of software that supports gas dispersion modeling for heavy and buoyant gases with multiple sensors. Moreover, environmental factors such as wind direction, air pressure, humidity, and additional parameters are blended into a dynamic and predictive capability. Smiths Detection has proposed a security or protection system called the Centurion, based on traditional drift tubes.45 The Centurion can be installed in heating,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_306... 06/05/2007

Página 1 de 1

Page 307 ventilation, and air conditioning (HVAC) systems and provides continuous monitoring of intake air for chemical warfare agents and toxic industrial chemicals. Although the sensitive nature of these products and the applications for public safety limit the disclosure of technical specifications, examples of continuous air monitoring with IMS analyzers can be found in Chapter 9.

8.9 VOCs in the Air of the International Space Station The atmosphere of the International Space Station is recycled, and air quality is maintained through the environmental control and life support system (ECLSS), which scrubs nominal contaminants from the recirculated atmosphere. However, concerns exist over crew exposure to pollutants that might arise from leaks, spills, and small fires. Some of these cannot be immediately or completely removed by ECLSS, and continuous exposure, even to low levels of some pollutants, are regarded as health hazards. Thus, a GC/IMS system known as the volatile organic analyzer (VOA) was developed in the 1990s to provide identification and quantitative determinations of a suite of ~25 VOCs including alcohols, aldehydes, aromatic compounds, and halocarbons.46–48 The VOA is comprised of two separate GC/IMS analyzers with integrated electronics and flow control. A sample of air is drawn with a pump through a preconcentrator trap in which contaminants are enriched. The trap contents are thermally desorbed into one of the two GC columns with a traditional IMS drift tube. The drift tube is a high-temperature analyzer from Graseby Dynamics (now Smiths Detection), that was developed in the three stages of a breadboard unit, a preflight demonstrator, and the in-flight instrument. An onboard computer controls instrument parameters and reduces the data to a simple output of concentrations for the crew and ground personnel. The VOA was installed on the International Space Station in September 2001 and operated until 2003 when a minor technical fault (blown fuse) halted operations. Prior to this time, sufficient experience was gained to demonstrate that this GC/IMS system was stable, reliable, and accurate. Moreover, the unit was mechanically robust. Examples of its performance with regard to analytical data are shown in Table 8–1, in which results are given for 15 chemicals by two independent methods. The first was onboard analysis by VOA (020402D), and the second was the standard method with a grab sampling cartridge (GSC) and on-earth laboratory measurement by GC/MS. The last column shows the lower quantitative limits (LQL) of the VOA. Comparison of the results illustrates a remarkable similarity in the measurements even though the samples were acquired in different locations (airlock vent vs. lab module). Moreover, concentrations of the compounds were near the detection limit of one or both the methods. The reproducibility of the VOA is shown in Table 8–2 for triplicate measurements on a single day using two GC/IMS analyzers (Part 1 and Part 2 in Table 8–2).

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_307... 06/05/2007

Página 1 de 1

Page 308 TABLE 8–1 Comparison of On-Board Analysis of Air in the International Space Station Using VOA vs. Earth-Based Analyses by GC/MS with a Grab Sampling Container (GSC)

T=2.5 Target Compound

LQL

VOA 020402D

GSC SN1040 4/2:1340

VOA LQL

mg/m3

mg/m3

mg/m3

Methanol

0.23

0.29

0.04

2-Propanol

0.13

0.10

0.07

2-Methyl-2-propanol

0.05

Trace

0.04

Ethanal

0.06

0.16

0.03

Ethanol

~2.3

2.50

0.09

1-Butanol

0.11

0.09

0.09

m-, p-Xylene

Trace

Trace

0.42

o-Xylene

0.12

0.23

0.07

Toluene

Trace

Trace

0.07

Isoprene

Trace

0.00

0.12

2-Butanone

Trace

Trace

0.04

Ethyl ethanoate

Trace

Trace

0.06

Propanone

0.09

0.20

0.03

Benzene

0.00

Trace

0.04

Dichloromethane

0.10

0.20

0.03

TABLE 8–2 Reproducibility of VOA Determination of Triplicate Air Samples On Board the International Space Station with Each of Two Independent GC/IMS Analyzers (Part 1 and Part 2)

020508C

020508D

020508E

mg/m3

mg/m3

mg/m3

Compound

Rel. Std. Dev. %

Part 1 Ethanol

0.88

1.0

1.5

29

Ethanal

0.13

0.13

0.12

4.6

1-Butanol

0.094

0.11

0.095

9.0

Propanone

0.072

0.077

0.071

4.4

Dichloromethane

0.056

0.052

0.053

3.9

Ethanol

1.8

1.85

1.85

1.6

Methanol

0.13

0.071

0.51

100

1-Butanol

0.15

0.14

0.12

11.2

2-Propanol

0.12

0.12

0.13

4.7

Propanone

0.15

0.16

0.084

31

Part 2

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_308... 06/05/2007

Página 1 de 1

Page 309

8.10 The Pharmaceutical Industry: Cleaning Verification Ion mobility spectrometry has found a novel application in the pharmaceutical industry—cleaning verification. The good manufacturing practice (GMP) requires that after the production of each batch of pharmaceuticals the machinery and production area must be meticulously cleaned in order to prevent cross-contamination of the next batch. The standard procedure is based on analysis of swipe samples collected from the production area by HPLC. As most pharmaceutical products contain at least one compound that is readily detected by IMS, either in positive or negative polarity, the technique can be deployed for verification that all traces of the previous product have been removed. Smith Detection has modified its standard Ionscan 400 instrument and added an auto-sampler (Figure 8–5A) as well as an innovative inlet system, shown in Figure 8–5B. The company claims that several pharmaceutical companies, including seven of the dozen market leaders, have installed IMS instruments for this purpose.49–51 According to one report, Cardinal Health evaluated IMS for use in cleaning verification with a protocol used to determine residual diphenhydramine HCl on stainless steel surfaces using a swab technique.51 The study showed that there was no interference from excipients, swabs, or solvents; a precision of 1.8% was obtained for six replicate injections at the action level; linearity with R2= 0.9955 was obtained; and all samples were analyzed within the linear range established. Overall, the IMS met all requirements of the protocol as did the

FIGURE 8–5 (a) The Ionscan-LS (Smiths Detection) equipped with an auto-sampler for cleaning verification in the pharmaceutical industry. (b) The high performance injection (HPI) feature used to prevent sample overloading and extend the dynamic range of the IMS device.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_309... 06/05/2007

Página 1 de 1

Page 310 TABLE 8–3 Diphenhydramine HCl Study Results

HPLC

IMS

Consumables

Mobile phase: Potassium phosphate pH 2.5 None and acetonitrile (70:30), flow rate 1.2 mL/min ≈1.2 liter eluting solution

Equipment setup

Prepare mobile phase: 60 min, establish mobile phase flow and column equilibration: 180 min Total: 240 min

Analysis/sample time Analysis/total 9 minutes 882 minutes (~17–3/4 hours) for all analyses associated with method validation

Warm up (from standby): 45 min Total: 45 min 1 minute 162 minutes (~2–3/4 hours)

HPLC method, but the IMS was superior as far as costs of consumables, instrument setup time, and sample throughput. As shown in Table 8–3, the overall time required for cleaning verification was reduced from 17.7 hours with HPLC analysis to 2.7 hours with IMS instrumentation, thus considerably shortening the downtime of the production equipment.

8.11 Conclusion As demonstrated in this chapter, IMS has been used to solve problems in a variety of venues and measurement challenges. The versatility and capability of IMS have been proven in the range of chemicals monitored, the complex matrices of some samples, and the demands on instruments from difficult environmental controls. The advantages of IMS include reliability and quantitative performance as shown in VOA, speed of response as utilized in the monitoring of ventilation systems, durability in accepting samples of corrosive gases, and high sensitivity in the detection of contaminants in purified gases and microelectronics components. Operations at ambient pressure permit the sampling of off-gas vapors without altering or damaging the sample during analysis. These form the basis of a harbinger of possibilities for IMS in solving analytical problems.

References 1. Bacon, T.; Weber, K., PPB Level Process Monitoring by Ion Mobility Spectroscopy (IMS), and Hydrogen Chloride and Hydrogen Fluoride Continuous Emission Monitoring, Brochure by Molecular Analytics, 14550-A York Road, Sparks, MD 21152, U.S. Web site: www.ionpro.com

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_310... 06/05/2007

Página 1 de 1

Page 311 2. Gordon, G.; Pacey, G.; Bubnis, B.; Laszewski, S.; Gaines, J., Safety in the work-place: ambient chlorine dioxide measurements in the presence of chlorine, Chemical Oxidation 1997, 4, 23–30. 3. Karpas, Z.; Pollevoy, Y.; Melloul, S., Determination of bromine in air by ion mobility spectrometry, Anal. Chim. Acta 1991, 249, 503–507. 4. Sielemann, S.; Baumbach, J.I.; Pilzecker, P.; Walendzik, G., Detection of trans1,2-dichloroethene, trichloroethene and tetrachloroethene using multicapillary columns coupled to ion mobility spectrometers with UV-ionisation sources, Int. J. Ion Mobility Spectrom. 1999, 2, 15–21. 5. Sielemann, S.; Baumbach, J.I.; Schmidt, H.; Pilzecker, P, Quantitative analysis of benzene, toluene, and mxylene with the use of a UV—ion mobility spectrometer, Field Anal. Chem. Technol. 2000, 4, 157–169. 6. Baumbach, J.I.; Sielemann, S.; Xie, Z.; Schmidt, H., Detection of the gasoline components methyl tert-butyl ether, benzene, toluene, and m-xylene using ion mobility spectrometers with a radioactive and UV ionization source, Anal. Chem. 2003, 75, 1483–1490. 7. Xie, Z.; Sielemann, S.; Schmidt, H.; Li, F.; Baumbach, J.I., Determination of acetone, 2-butanone, diethyl ketone and BTX using HSCC-UV-IMS, Anal. Bio-anal. Chem. 2002, 372, 606–610. 8. Leonhardt, M.J.; Leonhardt, J.W.; Bensch, H., Mobilities of halogenated compounds, Int. J. Ion Mobility Spectrom. 2002, 5, 43–46. 9. Stach, J.; Arthen-Engeland, T.; Flachowsky, J.; Borsdorf, H., A simple field method for determination of MTBE in water using hand-held ion mobility (IMS), Int. J. Ion Mobility Spectrom. 2002, 5, 82–86. 10. Schmidt, H.; Baumbach, J.I.; Klockow, D., Detection of perfluorocarbons using ion mobility spectrometry, Anal. Chim. Acta 2003, 484, 63–74. 11. Walls, C.J.; Swenson, O.F.; Gillispie, G.D., Real-time monitoring of chlorinated aliphatic compounds in air using ion mobility spectrometry with photoemissive electron sources, Proceedings of the SPIE— International Society for Optical Engineering 1999, 3534, pp. 290–298. 12. Simpson, G.; Klasmeier, M; Hill, H.H.; Atkinson, D.; Radolovich, G.; Lopez-Avila, V.; Jones, T.L., Evaluation of gas chromatography coupled with ion mobility spectrometry for monitoring vinyl chloride and other chlorinated and aromatic compounds in air samples, J. High Resolut. Chromatogr. 1996, 19, 301–312. 13. Wan, C; Harrington, P.B.; Davis, D.M., Trace analysis of BTEX compounds in water with a membrane interfaced ion mobility spectrometer, Talanta 1998, 46, 1169–1179. 14. Borsdorf, H.; Schelhorn, H.; Flachowsky, J.; Doring, H.R.; Stach, J., Determination of n-alkanes and branched-chain alkanes by Corona discharge ion mobility spectrometry, Int. J. Ion Mobility Spectrom. 1999, 2, 9–14. 15. Ritchie, R.K.; Rudolph, A., Environmental applications for ion mobility spectrometry, Proceedings of the 3rd International Workshop on Ion Mobility Spectrometry, 1994, 3301, pp. 193–208. 16. Przybylko, A.R.M.; Thomas, C.L.R; Anstice, P.J.; Fielden, P.R.; Brokenshire, J.; Irons, F., The determination of aqueous ammonia by ion mobility spectrometry, Anal. Chim. Acta 1995, 311, 77–83. 17. Eiceman, G.A.; Salazar, M.R.; Rodriguez, J.; Limero, T.F.; Beck, S.W.; Cross, J.H.; Young, R.; James, J.T., Ion mobility spectrometry of hydrazine, monomethyl-hydrazine, and ammonia in air with 5-nonanone reagent gas, Anal. Chem. 1995, 65, 1696–1702.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_311... 06/05/2007

Página 1 de 1

Page 312 18. Bollan, H., The Detection of Hydrazine and Related Materials by Ion Mobility Spectrometry, Ph.D. dissertation, Sheffield Hallum University, March 1998. 19. Bollan, H.; Stone, J.A., Brokenshire, J.; Rodriguez, J.A.; Eiceman, G.A., submitted 2005. 20. Cross, J.H.; Limero, T.F.; Lane, J.L.; Wang, E, Determination of ammonia in ethylene using ion mobility spectrometry, Talanta 1997, 45, 19–23. 21. Dean, K.R.; Carpio, R.A., Real-time detection of airborne contaminants in DUV lithographic processing environments, Proceedings of the Institute of Environmental Sciences 41st Contamination Control 1995, 9– 16. 22. Vigil, J.C.; Barrick, M.W.; Grafe, T.H., Contamination control for processing DUV chemically amplified photoresists, Proceedings of the SPIE—International Society for Optical Engineering, 1995, 2438, Advances in Resist Technology and Processing XII, 626–643. 23. Dheandhanoo, S.; Ketkar, S.N., Improvement in analysis of O2 in N2 by using air drift gas in an ion mobility spectrometer, Anal. Chem. 2003, 75, 698–700. 24. Pusterla, L.; Succi, M.; Bonucci, A.; Stimac, R., A method for measuring the concentration of impurities in helium by ion mobility spectrometry, PCT Int. Appl 2002. 25. Pusterla, L.; Succi, M.; Bonucci, A.; Stimac, R., A method for measuring the concentration of impurities in nitrogen, hydrogen and oxygen by means of ion mobility spectrometry, PCT Int. Appl. 2002. 26. Huxham, M.E.; Thomas, C.L.P., Cold vapour desorption of volatile organic compounds from an adsorbent trap, Anal. Commun. 1999, 36, 317–320. 27. Pilzecker, P; Baumbach, J.I.; Kurte, R., Detection of decomposition products in SF6: a comparison of colorimetric detector tubes and ion mobility spectrometry, Conference on Electrical Insulation and Dielectric Phenomena 2002, 865–868. 28. Baumbach, J.I.; Pilzecker, P.; Trindade, E., Monitoring of circuit breakers using ion mobility spectrometry to detect SF6-decomposition, Int. J. Ion Mobility Spectrom. 1999, 2, 35–39. 29. Soppart, O.; Pilzecker; P., Baumbach, J.I.; Klockow, D.; Trindade, E., Ion mobility spectrometry for on-site sensing of SF6 decomposition, IEEE Transactions on Dielectrics and Electrical Insulation 2000, 7, 229– 233. 30. Carr, T.W., Analysis of semiconductor devices and microelectronic packages by plasma chromatography/mass spectrometry, in Plasma Chromatography, T.W. Carr, Ed., Plenum Press, New York, 1984, chap. 7, 186–189. 31. Budde, K.J., Application of ion mobility spectrometry to semiconductor technology, Proceedings— Electrochemical Society 90–11, Anal Tech. Semicond. Mater. Process Charact. 1990, pp. 215–226. 32. Budde, K.J.; Holzapfel, W.J.; Beyer, M.M., Detection of volatile organic contaminants in semiconductor technology—a comparison of investigations by gas chromatography and by ion mobility, Proceedings— Institute of Environmental Sciences, 39th (Vol. 1), 1993, 366–372. 33. Budde, K.J., Determination of organic contamination from polymeric construction materials for semiconductor technology, materials research society symposium, Proceedings Ultraclean Semiconductor Processing Technology and Surface Chemical Cleaning and Passivation, 1995, 386, 165–176. 34. Budde, K.J., Organic surface analysis in semiconductor technology by ion mobility spectrometry, Proceedings—Electrochemical Society (Analytical Techniques for Semiconductor Materials and Process Characterization II), 1995, 95–30, 281–296.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_312... 06/05/2007

Página 1 de 2

Page 313 35. Budde, K.J.; Holzapfel, W.J.; Beyer, M.M., Determination of organic contamination from minienvironments according to SEMI document #2238, First results, Proceedings—Institute of Environmental Sciences, 1995, 41st Contamination Control, 155–162. 36. Budde, K.J.; Holzapfel, W.J.; Beyer, M.M., Application of ion mobility spectrometry to semiconductor technology: outgassings of advanced polymers under thermal stress, J. Electrochem. Soc. 1995, 142, 888– 897. 37. Budde, K.J.; Holzapfel, W.J., Organic contamination analysis in semiconductor silicon technology. Detrimental “cleanliness” in cleanrooms, Proceedings—Electrochemical Society 98–1 Silicon Materials Science and Technology, 1998, 2, 1496–1510. 38. Seng, H.P.; Mehnert, R.; Doering, H.R., Quality control in the radiochemical curing of surface films, Eur. Pat. Appl. 1995. 39. Seng, H.P., Controlling method for UV-curing processes. A method for the end user, European Coatings J. 1998,11, 838–841. 40. Dean, K.R.; Miller, D.A.; Carpio, R.A.; Petersen, J.S.; Rich, G.K., Effects of airborne molecular contamination on DUV photoresists, Photopolymer Sci. Technol. 1997, 10, 425–144. 41. Kishkovich, O.P.; Kinkead, D.; Higley, J.; Kirwin, R.; Piatt, J., Real-time methodologies for monitoring airborne molecular contamination in modern DUV photolithography facilities, Proceedings of the SPIE— The International Society for Optical Engineering Pt. 1, Metrology, Inspection, and Process Control for Microlithography XIII, 1999, 3677, 348–376. 42. Bollan, H.R.; West, D.J.; Brokenshire, J.L., Assessment of ion mobility spectrometry for monitoring monoethanolamine in recycled atmospheres, Int. J. Ion Mobility Spectrom. 1998,1, 48–53. 43. Gan, T.H.; Corino, G., Selective detection of alkanolamine vapors by ion mobility spectrometry with ketone reagent gases, Anal. Chem. 2000, 72, 807–815. 44. www.environics.fi/index.php?page=facility, Facility Protection. 45. http://63.89.158.169/products/Default.asp?Product=9 46. Limero, T.F.; James, J.T.; Reese, E.; Trowbridge, J.; Hohmann, R., The Volatile Organic Analyzer (VOA) aboard the International Space Station, SAE Technical Paper Series 2002–01–2407, 32nd International Conference on Environmental Systems, San Antonio, Texas, July 2002. 47. Brittain, A.; Bass, P; Breach, J.; Limero, T.F., Instrumentation for analyzing volatile organic compounds in inhabited enclosed environments, SAE Technical Paper Series 2000–01–2434, 30th International Conference on Environmental Systems, Toulouse, France, July 2000. 48. Limero, T.F.; Reese, E.; Trowbridge, J.; Hohmann, R.; James, J.T., Validation of the Volatile Organic Analyzer (VOA) aboard the International Space Station, SAE Technical Paper Series 2003–01–2646, 33rd International Conference on Environmental Systems, Vancouver, British Columbia, Canada, July 2003. 49. Walia, G.; Davis, M.; Stefanou, S.; DeBono, R., Using Ion Mobility Spectrometry for Cleaning Verification in Pharmaceutical Manufacturing, Pharmaceutical Technology, April 2002, 72–78. 50. Tan, Y.; DeBono, R., IMS for drugmaking: Ion mobility spectrometry can improve pharmaceutical companies’ production efficiency, Today’s Chemist at Work, November 2004, 15–16. 51. Payne, K.; Fawber, W.; Faria, J.; Buaron, J.; DeBono, R.; Mahmood, A., IMS for Cleaning Verification, Spectroscopy: Process Analytical Technologies Supplement, 1/2005.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_313... 06/05/2007

Página 1 de 1

Page 315

9 Feasible Applications 9.1 Introduction The word feasible, as used in the title of this chapter, is meant to suggest an underdeveloped application in IMS rather than an application that is based on speculative science or technology. In each instance described in the following text, measurements by mobility spectrometers were demonstrated, without any standing requirements by any government agency, organization, or industry. In a few instances, substantial development of a concept or method will be needed because, even though the measurements were proven and the application found to be feasible, a complete methodology is missing. In one or two other instances a measurement has been reported in nearly complete form, but either there are no obvious applications or existing methods are more than adequate. In all cases, the feasible applications illustrate that mobility can be seen as a general measurement concept; if gas-phase ions with lifetimes greater than 10 msec can be made from a sample, those ions can be characterized in a mobility spectrometer. Analytical information may be gleaned from the mobility spectrum. In moving a method from the category of feasible to proven or accepted, essential questions will need to be answered on the usefulness of the information, availability of instruments, validation of the measurement, and finally, whether it provides high value for cost. In the applications listed in this chapter, one or more of these questions are unanswered or answered unsatisfactorily.

9.2 Occupational Hygiene and Air Quality The composition of airborne vapors in industries is of interest in occupational hygiene because of the occurrence of inhalation exposures to chemicals that arise from solvents, coatings, industrial processes, and other sources. A similar problem exists in office buildings and homes, where emissions from synthetic materials or microorganisms may be responsible

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_315... 06/05/2007

Página 1 de 1

Page 316 for the sick-building syndrome. In such situations, the advantage of IMS technology is demonstrated by the use of ordinary drift tubes with inlets designed for vapor sampling. That is, instruments already developed for chemical weapons detection are immediately acceptable in technical configuration, though not yet commercially affordable. In the discussion below, the usefulness of IMS drift tubes for ambient air monitoring in several scenarios is illustrated.

9.2.1 University Stockroom Study In one monitoring study, each of three chemical agent monitors (CAMs) was equipped with either water, acetone, or dimethylsulfoxide (DMSO) as a reagent gas.1 The CAMs were not otherwise modified and were operated with positive ion polarity and in parallel. The air inside a university work site, the stockroom of the Department of Chemistry and Biochemistry at NMSU, was sampled continuously for several hours. Undergraduate laboratories were in session during the afternoon hours when activity increased in the stockroom with the dispensing of materiel and chemicals. Each analyzer was equipped with a specific reagent gas so response to vapors was associated with the proton affinity of the gas. The order of response from general to specific (or from general response to most chemicals to selective response to only amines and other high proton affinity chemicals) is water>acetone>dimethylsulfoxide. Thus, ionization chemistry can by itself bracket a chemical family, e.g., alcohols should be detectable in the CAM with water as the reagent gas but not in the CAMs containing acetone or DMSO. Similarly, ketones and esters should be observable in the CAMs with water and acetone but not in the CAM with DMSO. In practice, the reactant-ion peak intensity was recorded against clock time, though the mobility spectra provided additional information on the Ko values and concentrations of individual analytes. Results from these monitoring trials (Figure 9–1) demonstrated that vapors were released during the peak hours of activity in the department and that they were principally alcohols and other low proton affinity chemicals. This was seen as a negligible response with the DMSO-based CAM, some response with the acetone-based CAM, and a strong response with the water-based CAM. During the morning hours when the laboratories were closed, the ambient air in the stockroom was largely free of detectable vapors; this was a type of negative control. Positive controls on the experiment were made by vapor challenges of authentic standards. The concept was deemed acceptable from such controls. Nonetheless, the method and concept are impractical with CAM analyzers that are manufactured to military specifications and, therefore, too expensive for civilian uses. This concept can be revitalized with the development of miniature drift tubes and the planar micromachined differential mobility spectrometer, which can be manufactured inexpensively using methods of mass production.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_316... 06/05/2007

Página 1 de 1

Page 317

FIGURE 9–1 (A) Plots of RIP intensity vs. clock time for three IMS analyzers with different reagent ion chemistry during the designed releases of vapors into laboratory air. Traces corresponded to response from IMS analyzers containing reagent gases of water, acetone, and DMSO. All of the three IMS analyzers were exposed to acetone, acetophenone, and 2,4-lutidine for 30 to 50 min at 10:15 AM, 12:45 PM, and 2:20 PM, respectively (B) Intensity for the RIP vs. clock (top frame) during the morning and afternoon when laboratory was active using IMS analyzers based on acetone reagent gas (upper trace) and water reagent gas (lower trace).

9.2.2 Nicotine Exposure during Production of Skin Patches Another example that illustrates the value of a point analyzer with low detection limits and specificity from mobility analysis, in combination with favorable ionization chemistry, was the evaluation of nicotine emissions at a production site for transdermal systems, i.e., skin patches, for treating nicotine withdrawal. Nicotine patches are produced by depositing drops of nicotine onto an adsorbent layer that is drawn as a continuous sheet through

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_317... 06/05/2007

Página 1 de 1

Page 318 a machine in which barrier films and adhesive layers are laminated with the nicotine layer and punched to size. Operators of the machine were subjected to inhalation exposures to nicotine vapors during production of the patches, though neither the exact source of the nicotine emissions nor any fluctuations during the manufacturing day were determinable using the existing method of air sampling and analysis. The Occupational Safety and Health Administration (OSHA) exposure standard at that time for nicotine was 0.5 mg/m3 for continuous exposure, and these levels could be detected only when the sample was preenriched using a sorbent trap and an air sampling pump. The timespan of the sampling caused an integration of sample, eliminating any spikes in concentrations and preventing rapid surveys of the production site. A CAM equipped with water reagent gas was placed along with a computer on a wheeled cart that was positioned at the production site. In one test, the CAM was operated continuously during the workday near an operator’s station on the production machine. In another, parts and surfaces of the production machine were scanned rapidly for nicotine vapors by placing the CAM at these locations for several seconds. The entire machine was mapped for nicotine emissions within 15 min. The mobility spectra in positive polarity for nicotine showed characteristic product ions that were mass-identified as the protonated monomer and proton-bound dimer (Figure 9–2, bottom). The IMS exhibited near-instantaneous response, detection limits of 0.006 mg/m3, and median relative standard deviations of 3.1% for vapor levels of 0.01 to 0.25 mg/m3. During continuous monitoring of air near the machine, short-lived and elevated concentrations of isopropanol were detected when alcoholwetted cloth was used to clean the machine after contamination accumulated on particular surfaces. Concentrations of isopropanol in the air rose and fell rapidly with the cleaning procedure (Figure 9– 2, top); in contrast, the concentration of nicotine exhibited a gradual increase with time until reaching a plateau (Figure 9–2, top). This meant that the emission of nicotine was continuous and without rapid fluctuations. However, four sources emitting high levels of nicotine vapors were located in the production equipment using the CAM, and these corresponded to sites where liquid nicotine was placed on the adsorbent layer and where the waste film had accumulated. Thus, spikes in exposure could occur as operators moved near these parts of the machine. This experience demonstrated the value of a point sensor: a broad understanding of occupational exposures could be obtained after a few days of study using a CAM.

9.2.3 Air Quality in a University Research Laboratory The ambient air inside a well-ventilated laboratory in the Department of Chemistry and Biochemistry at NMSU was monitored continuously for VOCs using a differential mobility spectrometer.3 This monitoring exercise showed that elevated levels of solvent vapors were detected daily inside the laboratory during working hours (Figure 9–3, bottom trace). The concentrations of these vapors at times reached levels of ~1 ppm, corresponding to

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_318... 06/05/2007

Página 1 de 1

Page 319

FIGURE 9–2 Continuous monitoring of air by IMS. In top frame, signal intensities are shown for isopropanol (upper trace) and nicotine (lower trace) during work hours. Note that the spikes in the ion intensity of the alcohol correspond to cleaning of equipment, whereas nicotine concentration increases gradually due to accumulation of fugitive emissions in the workplace atmosphere. The mobility spectrum (bottom frame) for nicotine with hydrated proton-reactant ion showed a protonated monomer and proton-bound dimer at 0.25 mg/m 3.

the inadvertent release of vapors of the solvents used in the laboratory, i.e., toluene, methylene chloride, chloroform, and acetone. These releases occurred through the ordinary practices of handling solvents; no effort was made to alter the customary practices followed in the laboratory. During the evening hours and throughout the weekend, low levels of VOCs were detected in the laboratory (Figure 9–3, top trace). During these hours, the total concentration of the VOCs was calculated as 1 to 5 ppb, or only ~0.1% of the maximum concentration during laboratory work hours. Nonetheless, the patterns of concentration variations during these off hours

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_319... 06/05/2007

Página 1 de 1

Page 320

FIGURE 9–3 Differential mobility measurements of the concentration of various substances inside a chemical laboratory. Note that the changes in concentration correspond to the activity hours of the laboratory.

were repeatable and well above the limits of detection or noise of the analyzer. These fluctuations of VOCs were associated with the emissions from vehicles on nearby University Avenue and could be attributed to the rapid turnover of laboratory air with high-flow ventilation. Outside air was drawn into the building and conditioned for particulate matter and temperature before introduction into the laboratory. The chemical composition of air near the intake louvers was quickly mirrored by the chemical composition of the ambient air in the laboratory When the indoor monitoring experiments were completed, the DMS analyzer was relocated near an interstate highway Interstate 1–10, near Las Cruces, NM, for an accurate measurement of vehicular emissions dispersed in air. The analyzer exhibited differential mobility spectra that were similar to those from the suspected vehicular emissions inside the laboratory. The principal constituents observed in the differential mobility spectra, with a photo-discharge lamp as the ion source, were attributed to benzene, toluene, xylene, and MTBE. The intensity of the response with photoionization could be linked to the flow of traffic on the well-traveled interstate highway (Figure 9–4). However, the levels near the interstate highway were higher than those in the laboratory, reflecting the dilution of vapors during transfer from University Avenue to the Chemistry and Biochemistry Building, a distance of about 100 m. This study demonstrated that in some ventilation systems, distinctions between outside and inside cannot be made exclusively. These studies were made without a chromatographic

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_320... 06/05/2007

Página 1 de 1

Page 321

FIGURE 9–4 Differential mobility measurements of outdoor air near an interstate highway. The solid line is signal intensity for product ions from the DMS analyzer and the dots are counts of vehicle traffic. Note that the changes in concentration correspond to the traffic flow patterns on the highway.

inlet and indeed without any pretreatment of sample before analysis with the DMS instrument, indicating that air monitoring is already possible with the existing DMS technology. Additional refinements in methodology, calibration, and data processing will be necessary to make this application routine.

9.3 Fugitive Emissions from Industrial Activity Fugitive emissions refer to the airborne vapors that are released into ambient atmospheres at an industrial facility or commercial interest such as a fuel storage tank farm. Such emissions may arise from flaws in methods, faults in hardware, or poor control of inventory or wastes. A concern with fugitive emissions is that vapors can migrate off properties and into adjacent environments or neighborhoods. Thus, a need exists for perimeter monitoring at chemical industries or at any manufacturing site that handles or stores chemicals. Authentic instances of releases or false complaints about fugitive emissions as a means to financial harassment can be technically or legally expensive for an industry. Continuous monitoring of airborne vapors at fence lines should be a welcome tool for industries or companies, and IMS analyzers provide affordable solutions; however, this is an application of mobility spectrometers that is largely unused. One field demonstration of the concept of an emissions monitor was accomplished with a prototype IMS analyzer (PTIMS) that was positioned ~0.5 km from a chemical plant where herbicides and other chemicals were manufactured.4 Airborne vapors were monitored continuously for six aliphatic

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_321... 06/05/2007

Página 1 de 1

Page 322 amines used in the manufacturing processes. The monitoring task was initiated in 1991 and has been in operation for over 13 years, probably making it one of the longest continuous operations of an IMS monitor. Over the years, regular upgrades of the sampling system, data processing capabilities, and communications were made. Concentrations of the amines were determined using PTIMS with ammonia as the reagent gas. The minimal detectable levels (MDL) of the amines was below 5 ppb in air without pretreatment. The results were recorded on the local computer and transferred by modem and cellular phone to the control center in the plant. The findings of these studies are not open source data; nonetheless, the extended use of the IMS analyzer is evidence of its reliability and stability.

9.4 Smoke Alarm with Identification of Combustion Sources Smoke alarms detect particles from combustion in the air, using either ionization or photoelectric (light scattering) principles. The former is particularly good for detecting fires from fast-moving, intense flames, whereas the latter is preferred for slow-burning or smoldering fires. However, neither of these smoke alarms can distinguish the type of material burning nor provide clues to help locate the source of the fire. Identification of source is critical in some instances during the smoldering phase of combustion. False alarms may also trigger automatic sprinklers and cause economic losses and damage to equipment. Smoke alarms that are source specific would be valuable in such instances; for example, in a cotton warehouse, tobacco smoke would not trigger an alarm but burning cotton would be reason to start the sprinklers. Because combustion is never perfectly efficient, residual VOCs may be available in samples of gases or particulate matter. These VOCs may be characteristic of a particular fuel such as cotton, gasoline, or cigarettes, and chemical analysis of air may be a means to disclose a fire and its origin. Because VOCs in combustion effluents may be complex mixtures, prefractionation of the sample should be necessary. An exploratory study was made to determine if VOCs from combustion could be sampled and determined with a mobility spectrometer equipped with a gas chromatograph as inlet. Samples were collected using solid-phase microextraction (SPME) methods; vapors through partition and trap particulates by adhesion. A differential mobility spectrometer was chosen as the detector and SPME fibers were used for sampling, regardless of mechanism of retention, for simplicity of collecting vapors and particulates and for convenience of analysis with a gas chromatograph. Combustible materials including cotton, paper, grass, cigarettes, and gasoline were burned in an ambient atmosphere, and the gas plume from the fire was sampled for only 30 sec with an SPME fiber.5 Topographic plots of ion intensity vs. retention time and compensation voltage for fumes sampled from four materials are shown in Figure 9–5.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_322... 06/05/2007

Página 1 de 1

Page 323

FIGURE 9–5 Topographic plots from GC/DMS characterization of emissions of organic compounds collected on SPME fibers from combustion plumes of (A) cotton, (B) paper, (C) grass, and (D) gasoline.

Results show that a DMS analyzer can provide a second dimension to information obtained through chromatographic retention for the chemical analysis of combustion products. The plots demonstrate that the chemical composition of the combustion vapors were sufficiently distinct to allow identification of each combustion source. These differences were particularly evident in plots of ion intensity at a particular compensation voltage vs. retention time, suggesting that separation times could be made smaller and losses in separation efficiency would be tolerable. Though these results support the use of a GC/DMS as a sophisticated fire alarm that can identify the combustion source, the concept of identifying combustion sources by VOC patterns is not yet fully developed. Concern exists that the influence of combustion conditions on the composition of VOCs may be too varied to make a reliable alarm. This problem was recognized earlier and a patent application was submitted for a traditional mobility spectrometer.6 The application of GC/DMS as a fire alarm is justifiably categorized here as feasible because uncertainties regarding sampling and sample treatment remain to be clarified even though the technology may be well developed.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_323... 06/05/2007

Página 1 de 1

Page 324

9.5 Surface Analysis and Adsorbed Layers An early and successful use of IMS for surface analysis was in the determination of impurities on silicon wafers in microelectronics production as described by Carr and discussed in Chapter 8. This application was supported later by Budde’s findings (see Section 8.7). The target contaminants were volatile and readily released from surfaces when the sample was warmed. In the new generation of methods for surface analysis and adsorbed layers, the molecules of interest were not as volatile, and thermal desorption was replaced by laser ablation; the vapors released from the samples were, therefore, a complex mixture of several components.

9.5.1 Thermal Desorption of Natural Polymers An example of thermal release from a solid that is chemically complex is the speciation of wood fibers using mobility spectra.7,8 When slivers of wood are rapidly heated to 200°C, vapors are released into a drift tube for ionization and characterization using mobility. The mobility spectra showed a consistent and distinct pattern of differences between normal red oak heartwood and red oak wetwood. The major difference was attributed to pyrogallol and resorcinol that were desorbed from wetwood and not detected in normal heartwood. Thus, valuable lumber or wood supplies could be distinguished from more common and less expensive species. These findings and conclusions were supported by GC/MS analysis of methanol extracts of the wood samples. A comparable method useful for large, flat wood products involved a specialized sampling apparatus to extract vapors from tables or floors.9 Wood samples were examined using IMS analyzers for traces of preservatives whose use has been banned or strongly regulated.10–12 Mobility spectrometers have also been used in wood processing, unrelated to surface analysis, for the determination of moisture content in wood.13 A control system fitted with an IMS analyzer provides continuous measurements of the physical and chemical characteristics of the wood chip feedstock entering a pulp digester. Results from the IMS analysis are factored into the control system for optimization of steam flow and chemicals.

9.5.2 Adsorbates and Synthetic Polymers Where thermal desorption is inadequate to remove an analyte from the surface of a solid, films or adsorbates can be ablated using a laser beam.14 A laser beam can be directed—focused, or unfocused—against a solid, and compounds on its surface can be vaporized and ionized in air at ambient pressure. In these measurements, the wavelength of 1064 nm from an Nd: YAG laser was frequency-quadrupled to 266 nm, and compounds were

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_324... 06/05/2007

Página 1 de 1

Page 325 ionized by resonant multiphoton ionization with an unfocused beam. Though laser desorption and ionization (LDI) was used earlier in mass spectrometry under vacuum, no obstacles were encountered in determining large polycyclic aromatic hydrocarbons and carboxylic acid salts on a set of metals or inorganic solid by LDI/IMS at ambient pressure. In this and supporting studies,15,16 the mobility spectra of polycyclic aromatic hydrocarbons from LDI/IMS of solids suggested that the desorption and ionization events produced molecular ions; a possible limitation was the fragmentation observed for compounds substituted with polar moieties. When the beam is focused on the sample, the substrate can be ablated and ionized, forming mobility spectra characteristic of solids such as Teflon®, Plexiglas®, borosilicate and others (Figure 9–6). Methods based upon LDI/IMS have been applied to soils contaminated with petroleum products.17 Ions derived from polycyclic aromatic compounds are used to directly characterize the

FIGURE 9–6 Ion mobility spectra for solids after laser ablation and ionization. (FromYoung, Baumbach, and Eiceman, unpublished.)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_325... 06/05/2007

Página 1 de 1

Page 326 pollutants in the solid sample. In a final application, Simpson et al. described the use of a neural network for pattern recognition of ion mobility spectra obtained from laser ablation of polymeric materials.18 They reported that several polymers were accurately identified by the analysis of the patterns of the mobility spectra in negative polarity

9.6 Metal and Inorganic Ions Ion mobility spectrometry has long been associated with the characterization of organic compounds such as explosives and pollutants. However, among the earliest mobility coefficients measured were those for Li+, Na+, and K+ gas-phase metal cations. As recently as the 1960s, the mobilities of these and other ions were determined as a function of electric field strength and composition of the supporting atmosphere. In view of the availability of sensitive analytical methods such as flame photometry and others that are suitable for determining alkali and alkaline earth metals in solution, there was little motivation to develop IMS for metal ion determinations. Indeed, the only article concerning metals in the first 20 years of exploration with analytical IMS was a report on IMS and nickel carbonyl.19 This volatile and highly toxic chelate of nickel, Ni(CO)42+, was characterized as part of the search for the cause of Legionnaires’ Disease. Little more was published on metal ion determinations by analytical IMS until 1986. In that year, laser desorption of solids appeared to yield mobility spectra that were distinctive for various metals, though the results were not supported by mass spectral identification.14 Nonetheless, the work was reproducible and later extended to other metals (Figure 9–6). In the negative mode, an ion with mobility close to O2− was produced by laser ablation of metals, suggesting that the ablation and ionization steps released an electron into the atmosphere of the source region, leading to the formation of a negative ion. During the past decade electrospray ionization has been used to make ions of metals and inorganic species and mobility spectra for these have been documented. Metal ions including salts of A1, La, Sr, U, and Zn were determined by Dion et al.20 using ESI/IMS. The drift times were suggestive of the formation of metal solvent or metal-solvent anion clusters, and MDLs were in the range of 0.3 to 25 ppm. Anions in negative polarity were also measured by IMS using a traditional drift tube.21,22 Although reduced mobilities were determined for all anions, only nitrate and nitrite were determined in an actual water sample with MDLs of 10 ppb and 40 ppb, respectively. Significantly more work was reported by Canadian teams using field-dependent mobility analyzers to determine chlorate, bromate, perchlorate, and some organic anions in drinking water.23–26 A cylindrical FAIMS analyzer provided enhanced signal-to-noise ratio for the MS detector at concentrations of 62 μmol of the salts. Tabrizchi reported the mobility of ions

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_326... 06/05/2007

Página 1 de 1

Page 327 of sodium and potassium halides formed from a nichrome filament, and these included positive polarity peaks for Na+, K+, and, apparently, forms of MX+.27 In negative polarity, the mobility spectra exhibited halide ions and what were believed to be cluster of MXn+ hydrates. Potassium impurities in sodium bromide could be seen with MDLs of 100 ppm.

9.7 Aerosols and Electric Mobility Analyzers The next increase in size or weight of substances after biomolecules stretches the definition of mobility spectrometry and extends to species of aerosols with diameters from 1 nm to 100 μm. Aerosols have historically been excluded from discussions of analytical mobility spectrometry, although the principles and the aspirator-style IMS drift tubes are nearly identical to those accepted in the world of aerosol science.28,29 The principle of a differential mobility analyzer for size characterization of an aerosol is shown in Figure 9–7 (top). Charged particles are carried in a gas flow between concentric electrodes with an electric field; the center electrode has a negative DC potential of 0 to 10 kV, and ions that have a positive charge are attracted to the center electrode. Negative or neutral particles are swept through and out of the analyzer. All positively charged particles will move toward the center electrode at velocities related to mobility or size. Ions that have trajectories that place the particle near a sample orifice are removed and measured using an optical detector. A sweep of the DC voltage of the center rod brings particles of different sizes to the exit aperture and provides a measure of particle size (Figure 9–7, bottom). This is a vast topic with an enormous amount of activity in research and application. Aerosol studies could be a natural expansion of analytical IMS.

9.8 Summary The discussion in this chapter shows that the principle of a mobility measurement is a general concept and that the boundaries normally associated with analytical IMS should be considered elastic. Striking features of the applications described earlier are the scope and variety of measurements: detection of trace organic vapor in ambient air, identification of soil contaminants, composition of solids, and detection of inorganic ions in aqueous solution and also ions large enough to be charged particles. These applications demonstrate that the measurement of mobility of charged species in electric fields is a general concept and is not limited to the classic or historical uses of analytical IMS. Rather, it is the use of advanced sample introduction methods and preparation of sample by laser, electrospray,

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_327... 06/05/2007

Página 1 de 1

Page 328

FIGURE 9–7 Schematic of an electric mobility analyzer for aerosol particles (top) and results of measurement as a plot of distributions of size of aerosols in a sample (bottom). The principles of separation are based upon mobility of charged aerosols in air at ambient pressure. (From www.tsi.com.)

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_328... 06/05/2007

Página 1 de 1

Page 329 heated filaments, etc., that establishes the practical boundaries or constraints of the method. Innovations in ion formation will provide the first and most essential gate through which an IMSbased measurement must pass. Naturally, there are no guarantees for any technology, even for IMS, beyond this first gate.

References 1. Meng, Q.; Karpas, Z.; Eiceman, G.A., Monitoring indoor ambient atmospheres for VOCs using an ion mobility analyzer array with selective chemical ionization, Int. J. Environ. Anal. Chem. 1995, 61, 81–94. 2. Eiceman, G.A.; Sowa, S.; Lin, S.; Bell, S.E., Ion mobility spectrometry for continuous on-site monitoring of nicotine vapors in air during the manufacture of transdermal systems, J. Hazard. Mater. 1995, 43, 13–30. 3. Eiceman, G.A.; Nazarov, E.G.; Tadjikov, B.; Miller, R.A., Monitoring volatile organic compounds in ambient air inside and outside buildings with the use of a radio-frequency-based ion-mobility analyzer with a micromachined drift tube, Field Anal. Chem. Technol. 2000, 4, 297–308. 4. Karpas, Z., Operational experience with IMS for process control and environmental monitoring, Proceedings of the 5th International Workshop on Ion Mobility Spectrometry, Jackson Hole, WY, 1996, pp. 139–148. 5. Eiceman, G.A.; Tarassov, A.; Miller, R.A.; Nazarov, E.G.; Hughes, E.; Funk, P., Discrimination of combustion fuel sources using gas chromatography-planar field asymmetry ion mobility spectrometry, J. Sep. Sci. 2003, 26, 585–593. 6. Foulger, B.; Riches, J.; Bollan, H.R., PCT Int. Appl. 2000, 17 pp. 7. Lawrence, A.H., Rapid characterization of wood species by ion mobility spectrometry, J. Pulp Paper Sci. 1989, 15, J196–J199. 8. Pettersen, R.C.; Ward, J.C.; Lawrence, A.H., Detection of northern red oak wetwood by fast heating and ion mobility spectrometric analysis, Holzforschung 1993, 47, 513–522. 9. Baumbach, J.I.; Vautz, W., Poster, International Conference on Ion Mobility Spectrometry, Umeå, Sweden, 2004. 10. Matz, G.; Schroder, G., Fast detection of wood preservatives on waste wood with GC/MS, GC/ECD and ion mobility spectrometry, Field analytical methods for hazardous wastes and toxic chemicals, Proceedings of a Specialty Conference, Las Vegas, NV, January 29–31, 1997, 793–801. 11. Schroder, W.; Matz, G., Wood preservatives on waste wood: fast detection with GC/MS, GC/ECD and ion mobility spectrometry (IMS), Field screening Europe, Proceedings of the International Conference on Strategies and Techniques for the Investigation and Monitoring of Contaminated Sites, 1st, Karlsruhe, September 29–October 1, 1997, 347–350. 12. Schroder, W.; Matz, G.; Kubler, J., Fast detection of preservatives on waste wood with GC/MS, GC-ECD and ion mobility spectrometry, Field Anal. Chem. Technol. 1998 2, 287–297.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_329... 06/05/2007

Página 1 de 2

Page 330 13. Warren, S.R.; McKenzie, D.J., Process for Controlling a Digester Using Real Time Measurement of Moisture Content and Species of Wood, U.S. Patent Appl. Publ. 2002, 10 pp. 14. Eiceman, G.A.; Anderson, G.K.; Danen, W.C.; Ferris, M.J.; Tiee, J.J., Laser desorption and ionization of solid polycyclic aromatic hydrocarbons in air with analysis by ion mobility spectrometry, Anal. Lett. 1988, 21, 539–552. 15. Young, D.; Douglas, K.M.; Eiceman, G.A.; Lake, D.A.; Johnston, M.V., Laser desorption-ionization of polycyclic aromatic hydrocarbons on borosilicate glass at ambient pressure in air with ion mobility spectrometry, Anal. Chim Acta. 2002, 453, 231–243. 16. Eiceman, G.A.; Douglas, K.M.; Porter, L.L.; Young, D.; Lake, D.A.; Johnston, M.V.; Laser desorptionionization ion mobility spectrometry of substituted polycyclic aromatic compounds, 2005, in preparation. 17. Roch, T.; Baumbach, J.I., Laser-based ion mobility spectrometry as an analytical tool for soil analysis, Int. J. Ion Mobility Spectrom. 1998,1, 43–47. 18. Simpson, M; Anderson, D.R.; McLeod, C.W.; Cooke, M.; Saatchi, R., Use of pattern recognition for signatures generated by laser desorption-ion mobility spectrometry of polymeric materials, Analyst 1993, 118, 1293–1298. 19. Watson, W.M.; Kohler, C.F., Continuous environmental monitoring of nickel carbonyl by fourier transform infrared spectrometry and plasma chromatogrpahy, Environ. Sci. Technol. 1979, 13, 1241–1243. 20. Dion, H.M.; Ackerman, L.K.; Hill, H.H., Initial study of electrospray ionization-ion mobility spectrometry for the detection of metal cations, Int. J. Ion Mobility Spectrom. 2001, 4, 31–33. 21. Dion, H.M.; Ackerman, L.K.; Hill, H.H., Detection of inorganic ions from water by electrospray ionization-ion mobility spectrometry, Talanta 2002, 57, 1161–1171. 22. Dwivedi, P.; Matz, L.M.; Atkinson, D.A.; Hill, H.H., Jr., Electrospray ionization ion mobility spectrometry: a rapid analytical method for aqueous nitrate and nitrite analysis, Analyst 2004, 129, 139–144. 23. Barnet, D.A.; Guevremont, R.; Purves, R.W., Determination of parts-per-trillion levels of chlorate, bromate, and iodate by electrospray ionization/high-field asymmetric waveform ion mobility spectrometry/mass spectrometry, Appl. Spectrosc. 1999, 53, 1367–1374. 24. Ells, B.; Barnett, D.A.; Froese, K.; Purves, R.W.; Hrudey, S.E.; Guevremont, R., Detection of chlorinated and brominated byproducts of drinking water disinfection using electrospray ionization—high-field asymmetric waveform ion mobility spectrometry-mass spectrometry, Anal. Chem. 1999, 71, 4747–4752. 25. Handy, R.; Barnett, D.A.; Purves, R.W.; Horlick, G.; Guevremont, R., Determination of nanomolar levels of perchlorate in water using ESI-FAIMS-MS. J. Anal. At. Spectrom. 2000, 15, 907–911. 26. Ells, B.; Barnett, D.A.; Purves, R.W.; Guevremont, R., Detection of nine chlorinated and brominated haloacetic acids at part-per-trillion levels using ESI-FAIMS-MS, Anal. Chem., 2000, 72, 4555–4559. 27. Tabrizchi, M, Thermal ionization ion mobility spectrometry of alkali salts, Anal. Chem. 2003, 75, 3101– 3106. 28. http://www.tsi.com/particle/downloads/brochures/3080.pdf 29. Wang, S.C.; Flagan, R.C., Scanning electric mobility spectrometer, Aerosol. Sci. Technol. 1990, 13, 230– 240.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_330... 06/05/2007

Página 1 de 1

Page 331

10 Present Conditions, Barriers to Advances, and Future Developments in Ion Mobility Spectrometry 10.1 State of the Science and Technology of Ion Mobility Spectrometry (IMS) During the past 10 years, IMS has become recognized and accepted within the larger community of analytical scientists, and this recognition can be seen in the number of symposia at national and international conferences and in the number of IMS articles in chemistry, biochemistry, and physics journals. Acceptance is also apparent in invited articles or reviews in prominent analytical chemistry journals and in A-page articles in Analytical Chemistry. Such high visibility is suggestive of a level of interest in IMS that had not existed before the year 2000. This can be attributed to a growing awareness that IMS analyzers are the core technology to counter increased civilian vulnerability to unconventional warfare through their use in military preparedness, aviation security, and explosives detection. Awareness and interest in IMS have grown in increasing measure following the tragic attacks in New York City and Washington, DC, on September 11,2001, although neither explosives nor other contraband were involved in these assaults. These disasters and the emergence of public awareness of terrorism in the U.S., Israel, Kenya, and Spain, for example, and the subsequent wars in Afghanistan and Iraq have all elevated national interest in domestic protection against terrorism. A significant change during the past decade has been a growing appreciation in the field of mass spectrometry of the value of mobility in characterizing biomolecules. Although drift tubes used for probing the conformation of biomolecules differ in size and operating conditions from analytical IMS analyzers, the principles of ion mobility involved are identical. The association between mobility studies of biomolecules and analytical IMS have also been beneficial for the visibility and acceptance of IMS. During the 1980s, strong opinions existed against using ions in air at ambient pressure for chemical measurements, and these originated from the

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_331... 06/05/2007

Página 1 de 1

Page 332 errors that were made in the practice of IMS and in the interpretation of experimental results of the 1970s. This uneven beginning in the development of the science and fundamentals of a new method affected the reputation of the field well into the early 1990s. Today, the merits and weaknesses of IMS are understood for existing analyzers. This perhaps suggests that IMS has entered a stage of technical adolescence; the method is past its rough beginnings but is not yet a fully developed field. Today, IMS can still be regarded as a niche technology serving critical needs in narrow applications, and this condition is unlikely to change. Unless and until a renewed interest in onsite or field instruments for monitoring hazardous substances occurs, IMS will not become a common measurement tool. It is unlikely that IMS will appear in undergraduate textbooks on chemical instrumentation in the near future or that mobility measurements will be utilized the way that mass spectra are used today to characterize unknown samples. Despite these narrow opinions on the possibilities for IMS, developments in commercial instrumentation and the science and applications of IMS demonstrate that this measurement technique is far from stagnant. Rather, the topic of IMS is international, vibrant, and developing in ways that are optimistic for the future. During the past decade, mobility measurements have been complemented by the techniques of differential mobility, which was wholly unexpected in the early 1990s. This has added a layer of richness and possibilities in measurements and instrumentation. For example, two of the four new companies formed in the past decade, without prior history as a derivative or amalgamation of existing companies, have differential mobility as the central principle of their analyzers. In addition to differential mobility as another method of characterizing ions, drift-tube designs and combinations of drift tubes with gas chromatographs or mass spectrometers today are varied, with specialized designs in inlets, drift tubes, manufacturing methods, and more. Indeed, mobility-based ion filters reminiscent of early military analyzers have recently appeared with GC inlets from Ameritest, Inc, with the Sky-shield PSU-03. In broad analytical considerations, the operation of drift tubes has not coalesced into a common set of experimental parameters, and mobility spectrometers now function from ~100 torr to ambient pressure and through wide ranges of moisture and temperature. This high variability in designs and methods can be seen as a strength or a weakness (see the following discussion). A notable trend in instrumentation has been the miniaturization of drift tubes, and this has been motivated in large part by the desire to make IMS analyzers not just portable but pocket sized, unobtrusive, and lightweight. Examples of this include the lightweight chemical detector (LCD) from Smiths Detection, the microfabricated drift tube of the micro-DMx technology from Sionex Corporation, the micro-IMS from G.A.S., and other designs from research teams. Further miniaturization may be expected in the next decade, and integration of analyzers into the Internet or other methods of mass communication could occur. Another facet of instrumentation has been the appearance of viable alternatives to radioactive sources such as miniature

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_332... 06/05/2007

Página 1 de 1

Page 333 pulsed corona discharges. In 1994, there was no comprehensive scientific model for the ion chemistry or ion behavior in drift tubes. Today, the effects of kinetics and ion behavior on the appearance of mobility spectra are described, although not exhaustively documented. Spectral libraries are now available, although perhaps not well publicized or available in searchable and userfriendly software packages. The first assembly of researchers in IMS had been convened only a year or so before the manuscript for the first edition of Ion Mobility Spectrometry was completed. In the summer of 2004, an international conference on IMS was convened as the 13th annual meeting and was self-organized through the International Society for Ion Mobility Spectrometry. The Society has launched a journal that has provided a record of conference activities in a proceedings issue. The journal is abstracted and is expected to be a record and resource for IMS investigators and users.

10.2 Barriers to Advances in Performance and Uses Mobility spectrometers are rooted in principles of physics and chemistry that may or may not be compatible with the intentions of those wanting to make measurements. Commonly, an attempt to exploit a certain principle involves compromises driven by conflicting requirements. For example, ion drift velocity in an electric field is useful in characterizing an ion; however, band-widths oppose resolution of ions in such measurements by band broadening. In the history of instrument development from prior technologies, advances in techniques or alternate designs have been used in attempts to circumvent limiting principles. Some technical features of drift tubes or methodology now exist in IMS, and a few can be identified as genuine limitations on or barriers to performance or applications. Some of these extend beyond principles of chemistry and physics and involve the organization of information or practices of measurement. In some instances, such barriers are difficult to remove and require the consent of scientific and industrial communities using IMS. An example of this is the lack of any agreement on methods or even on the need to reach standards on control of moisture, temperature, and concentration of analyte because instrument manufacturers and researchers have developed their own preferred methodology, and there is no universally accepted mode of operation. Similarly, common components in drift tubes have no standard dimensions or designs.

10.2.1 Concepts and Practices in IMS In the early 1990s, there were no models for ion behavior or ion chemistry in a drift tube, but today such a model exists. The model, supported by experimental results, demonstrates that moisture, temperature, analyte

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_333... 06/05/2007

Página 1 de 1

Page 334 concentration, and residence time of ions (controlled by electric fields, drift-tube length, and pressure) are pivotal in the formation of ions and their behavior inside the drift tube. As of today, there is no recognized agreement on the importance of any of these parameters or any conformity to details. Although this is seen as a weakness, such conformity may not be helpful or advantageous at this stage in the development of IMS. There are applications of IMS with specific requirements that cannot be forced into a single instrument design or methodology The next stages of development in IMS will involve the availability of standards, spectral libraries, and methods to link reduced mobility values to ion structure. None of these will be possible until experimental parameters are brought under control. This does not exist now, and there is little promise of change. Standards have been proposed and explored: Will these or any others be incorporated into IMS instruments or methods? Will the role of ion residence time be widely recognized in the appearance of mobility spectra? In 2004, there are no models to link mobility and ion structure under specific experimental conditions. A next step in the science of IMS may include such models that are convenient, accurate, and intuitive. Some models do exist with research teams; however, these are neither convenient nor suitable for general applications. Though this is an issue of molecular modeling and the structure of ions, it involves software and graphic reduction; these, too, in 2004 are inadequate and need refinements. Such tools will be developed when a need for the capabilities motivates practitioners and researchers. Another explanation underlying the current condition of limited development with such tools is that IMS has been explored, until recently, by a relatively small number of investigators. This situation has changed in small but significant ways, in a few applications, in which IMS analyzers are the preferred measurement method. At present, possible new and novel uses for IMS may be harbingers of a vibrant future for IMS technology

10.2.2 Hardware and Instrumentation One of the most promising developments during the past decade is the emergence of several small instrument manufacturing companies producing IMS analyzers with designs suitable for general use. In two other instances, for different reasons, companies ceased production. Three other companies have survived more than a year, and three additional companies have emerged from countries that were part of the former USSR. An optimistic appraisal of this trend is that there is a growing interest in having IMS solutions to measurement problems. However, the demand for security and military preparedness has affected each of these companies and their approach to hardware development and marketing. Though some investigations can be made using commercial IMS instruments, most analyzers are still designed for a targeted application, with features uniquely suited to the application. Consequently there are only a few analyzers today that may be regarded as generally suitable for researchers or for someone seeking a general

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_334... 06/05/2007

Página 1 de 1

Page 335 measurement solution. This is unlike any other measurement technology in which a large number of refined and well-engineered instruments exist but cannot be used well for measurements in general. Some components such as ion shutters are unavailable and must be built by investigators. The ion shutters constitute a barrier to performance mainly because they seriously diminish the duty cycle. Improved methods of ion injection, possibly after accumulation and storage, must be developed. Similarly, the Faraday detector, which serves nicely for field instruments, is a barrier to the development of IMS as a general analytical method. Improved methods of ion injection and ion detection would aid the development of high-resolution IMS by allowing ion packets to be injected with pulse widths of 1 to 10 μsec and with good response through an improved detector. Components for this are unsatisfactory or unavailable. This is so also for electronics, which can be purchased as general use items, e.g., high voltage supplies and picoAmp current to voltage amplifiers. However, electronics specifically for IMS have not been available even within the IMS research community. This may be remedied in part with electronic designs for IMS provided in the CD accompanying this book. Finally, moisture is to IMS what vacuum is to mass spectrometry Thus, drift tubes must be fitted with sensors to monitor moisture and temperatures of the gas, rather than rely on drift-tube housing itself for control.

10.3 Future for IMS In the next decade, mobility spectrometers should be seen in extended use for detecting explosives, narcotics, and chemical weapons. This may be expected through improvements in drift tubes and some revolutionary developments including microfabricated drift tubes. Additionally, the large number of analyzers already deployed in the field will not be phased out soon. Of course, these trends cannot be guaranteed, and advances within a decade in miniaturization of mass spectrometers are being viewed as serious technical competition to existing on-site analyzers. One compelling application of IMS may be found in the fusion of IMS and MS, where a drift tube serves an ion filter for a mass spectrometer. Such a filter, based on the field asymmetric ion mobility spectrometry (FAIMS) analyzers by lonalytics can pass selected ions and remove other ions that are chemical noise in certain biological and medical measurements. Today, progress in IMS measurements is occurring in a range of bacterial, pharmaceutical, and medical diagnostic uses. These may or may not be refined and accepted commercially. Nonetheless, the growth in medical and clinical uses may be reasonably expected in all of analytical chemistry and also with IMS. The barriers in technology noted earlier are recognized in the IMS community, and efforts to improve various components may be expected. Based on

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_335... 06/05/2007

Página 1 de 2

Page 336 recent advances and known interests, these should appear within the next decade and should improve resolution and detection limits. Improvements in manufacturing techniques may be expected with each type of mobility spectrometer. Finally multiple dimensions of mobility through tandem IMS analyzers, particularly portable and laboratory-based GC/IMS instruments, and sophisticated combinations of ionization methods in a single instrument could be demonstrated and could improve analytical performance.

10.4 Final Thoughts In the early 1990s, IMS was considered an emerging technology It had been prematurely rejected by the academic community but had been accepted for chemical weapons detection and was on the verge of being accepted for explosives detection. The concern then was that the principles of mobility measurement could not be broadly extended from these uses into other measurement needs. One objective of the present monograph was to demonstrate that applications have arisen in some unexpected areas with IMS, demonstrating the value of mobility for chemical measurements. The principle of chemical characterizations using ions at ambient pressure was an appealing concept at the time of the inception of modern analytical IMS and remains so today. If the past decade is an indicator of the directions of the near future, this opinion will be shared by a growing number of measurement scientist and analytical chemists for the benefit of many.

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_336... 06/05/2007

Página 1 de 2

Page 337

Index A Ablation, 134, 326–328; see also laser ablation Absorbent, 301–302 Ac2H, see acetone ACADA, 22, 26, 30, 245 Accelerants, 265 Acetate fragmentation of butylacetate 225–226 mobility values of acetates and esters, 204 Acetone as alternate reagent gas, 105–106, 202–206, 245–248 in ambient air, 316–317 in breath, 275–276, 297 Acidic gases, determination by IMS, 293–296 Acids amino acids, 62, 250, 278, 281 haloacetic acids, 330 Acquisition, data by GC IMS, 194–195 data by IMS MS 224–225 signal acquisition, 165–166 software, 30, 168, 293, 306, 333 Acrylonitrile-butadiene-styrene, 304 Adducts effects on Ko values, 186 formation of cluster ions, 95 ketone adducts of hydrazines, 300 negative ions, 90 water adducts with ions, 6–7, 63 Adhesives materials in IMS, 172 nicotine patches, 317 Adsorbate determination by laser ablation, 324 Adsorption surface adsorption in GC interface, 206 tenax trap, 302 Aerial releases of spores, 285 Aerodynamics, Use in portal sampling, 254 Agents biological agents, 285–286 chemical warfare agents, 22, 26, 206, 243–249, 294, 307 Air aerosols of bacteria in air, 217, 285–286 amine detection, 321–322 bromine, 294–295 combustion vapors from burning 322–323 detection limits in air, 96–98 formation of ions in air, 5–10, 79–92 HF in air, 105–106, 293–296 mercaptans in air, determination by GC/DMS, 214–215 movement of ions in air, 3–5 monitoring clean rooms, 301–305 laboratory, 318–320 recirculated atmospheres, 305–308

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_337... 06/05/2007

Página 2 de 2

near highway, 320–321 nicotine in air, 317–318 sampling of air in portals, 253–254 vinyl chloride in air, 206 volatile compounds in respired air as measure of health, 297, 274–276 Volatile Organic Analyzer for air monitoring, 210–212 Aircraft or airplanes, see Unmanned airborne vehicles; see also Aviation Airport, monitoring for explosives, 254, 258; see also Aviation Albritton, D.L., 15, 35 Alcohol fragmentation of alcohol ions, 103, 280 mobility spectra and effect of temperature, 103 proton bound trimers ions of alcohols, 103 release into atmosphere, 316 retention times in GC IMS, 204 Aldehydes, retention times in GC IMS, 204 Alkali ions with biomolecules, 282–283, 284 Alkanes corona discharge ion source, 298 detection dry atmospheres, 208

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_337... 06/05/2007

Página 1 de 2

Page 338 formation by charge exchange reactions, 104 Alpha, also α alpha model instrument from Franklin GNO, 15, 18, 220, 224 correction factor in mobility equation, 44 field dependent equation, description of parameter in DMS, 60–82 Alternate reagent ion chemistry or reactant ions, 104–107 Aluminum building drift tubes, 170–172 clad capillary columns for GC IMS, 192–193 Ambient pressure chemical Ionization, see Atmospheric Pressure Chemical Ionization Amide groups on illicit drugs, 243 Amine(s) biogenic amines in medical testing, 276–278 elucidation of structure of diamines, 282 monitoring releases from a chemical plant, 321–322 proton affinity affects response, 121, 166 reduced mobilities, 48–54 vapors from decomposition of meat, 278–281 Ammonia, ambient air in semiconductor plants, 301 ammonia as alternate reagent gases, 105 determination in water, 300–301 inteferent for hydrazine detection, 106–107, 300 monitoring ammonia in hydrocarbon flow, 300 reagent gas for amine detection, 322 Amphetamine determination by field dependent IMS, 263 mobility value, 264 Amplifier, 164; see also CD for circuitry for amplifier Analog-to-digital converters, 21 Anesthetic gases in respired air, 274 Angiotensin, 281 Aniline detection in water, 98 site of protonation in, 227 structure, 281 Anion(s) solvent anion clusters, 326 O2− anion, 7 Annihilation of ions, 164 Anvil filter paper samples, 212–215, 254 thermal desorption of bacteria, 286 Aperture, 15 Aperture grid, 162–165, 170 APCI, see atmospheric pressure chemical ionization Arcing inside SF6 filled electrical high voltage switches, 302 Argon as drift gas, 48, 59, 282, 301 Army development of vapor detectors, 22 pyrolysis GC IMS for bacteria analysis, 216–217 Aroclors, determination in transformer oil by IMS, 299 Aromatic hydrocarbons airborne vapors, 168 laser ablation with IMS of polycyclic aromatic hydrocarbons, 325 Arson, 265 Aspirator, design of mobility spectrometer, 160–161 Asymmetric in differential mobility spectrometry (DMS), 60–66 waveform for field asymmetric field IMS (FAIMS), 27 Atmospheric Pressure Chemical Ionization (APCI) Ions in IMS, 136–144

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_338... 06/05/2007

Página 2 de 2

reactions in gas phase, 102–108 quantitative response with IMS, 92–98 Autolycus, 12 Averaging of spectra, 165–166 Aviation commercial aviation security, 9 explosives detectors for, 26, 249–254 B Bacillus subtilus biomarker detection by GC IMS, 285 pyrolysis GC IMS, 216–217, 285–286 pyrolysis GC DMS, 285–287 Bacteria determination by ELISA based method, 287–288 vapors from infection with bacteria, 276–278 Bacterial vaginosis (BV), 276–278 Baim, M.L., 25 Band-broadening, or bandwidths barrier to advances, 333

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_338... 06/05/2007

Página 1 de 2

Page 339 diffusion of ion swarm, 43 gas chromatography, 25, 190 ion shutter effect on peak width, 96 laser generated peak, 164 Barbiturates, 261, Barringer Research Corporation, 215, 257, 262 Baumbach, J.I., 163, 209 Beer, ethanol monitoring, 161 Bell, S.E., 102 Bendix Corporation, 163 Benzene gasoline, detected with photoionization source, 296 measurement using multicapillary columns, 209 target analyte in breath, 275 Beta emission, radioactive source into air, 12 Beta-VI description of drift tube, 18–19 karasek use of Beta-VI, 19 data acquisition by boxcar integration, 21 Bio-fouling with analysis of water and wastewater, 299–300 Biogenic amines, see Amines Biomarkers, see Bacteria Biomedical, see Clinical uses Bio-molecules, mobility spectrometry of, 229–230 Blanchard, W.C. 180 Bleed, from stationary phase column, 208 Blood, content of volatile compounds, 275 Blyth, D.A., 22 Boger, Z., 166 Bollan, H., 166 Bond-phase capillary column, benefit to IMS, 25, 208 Borsdorf, H., 298 Bowers, M.T., 281 Boxcar integrator, use in dual ion shutter drift tube, 224 Bromides, Br-, detection in air, 294 Br2, bromine, in air, 294 Bradbury-Nielson ion shutter design, 11 Bradykinin ESI-FAIMS, 281 MALDI IMS, 283 Brokenshire, J., 170 Bromate in water by FAIMS, 326 Bromide association with bromine 294 Bruker-Saxonia, 33, 245 BTEX monitoring in ambient air, 168 photoionization, 298 Budde, K.J., 304, 324 Buryakov, I.A., 37, 181, 209 Butane, detection with IMS at low moisture, 207 Butylacetate, fragmentation in drift region, 225–226 BV, see bacterial vaginosis C Cadaverine formation from lysine, 276 vaginal infections, 276–278 Calibration curves, 100, 102, 207 explosive response, 257 internal calibrant, 99, standard addition calibration, 111 stability, 100

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_339... 06/05/2007

Página 2 de 2

CAM, see Chemical agent monitor Cambridge, England, 28, 34, 180 Cancer, vapor markers in breath, 275–276 Candidiasis, yeast infection, 277 Capillary column, see Bonded phase capillary column Capsule, analysis of pharmaceutical capsules, 262 Carbohydrates, 281, 290 Carboxylic acids, laser desorption and IMS determination, 325 Carr, T.W., 36, 86, 173, 303, 312, 324 Cavendish Laboratory, 12 Centurion, IMS monitor (Smiths Detection) 265 Ceramic drift tube for miniature IMS, 163 insulating rings, 171 plates for DMS, 159 support for ion shutter, 170–171 Charge-transfer in gas phase ion molecule reactions, 21, 83, 104, 296 Chemical agent monitor (CAM), 31, 167, 169, 171, 210, 217, 245, 284, 287, 316 Chemical warfare agents mobility coefficients, 249 mobility spectrometers to detect, 23, 246 detection, 26, 206 Chlorate, detection in water, 326 Chloride alternate reagent ion, 97, 252 cluster formation, 85 field dependent mobility, 62 Chlorinated organic compounds, GC/IMS, 236, 311 ion mobility spectrometry, 298, 311 Chlorofluorocarbons, 216

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_339... 06/05/2007

Página 1 de 2

Page 340 Chloroform, 263, 296, 318 Chromatography, gas data from GC IMS, 184–201 development of modern IMS with GC/IMS, 25 elution profiles to make calibrationcurves, 207 enhanced IMS response, 111, 185 GC IONSCAN, 215 general aspects, 186–194 hand-held GC IMS, 291 improvement for ionization chemistry, 299, 202–205 mobility spectrometer for pyrolysis of bacteria, 216, 285, 291 multicapillary columns for fast separations, 296 smoke alarm based GC DMS, 322 Varian micro differential mobility detector, 215 Volatile organic analyzer, 28 Chromatography, liquid chromatography with mobility spectrometer, 218–219 Cigarette combustion vapors, 322–323 smoke as interference, 263 Cl−, see Chloride ion Cladding, removal of aluminum cladding from wall of GC capillary column, 193 Classification of mobility spectra by chemical family, 5, 33, 183 Clemmer, D.E., 38, 179, 229, 231, 239, 281, 290 Clinical uses of IMS bacterial vaginosis, 276–279 lung disease, 276 Cluster ions alternate reagent gases and clusters, 104–105 behavior in field dependent mobility, 62–64 effect of temperature, 53, 57 elimination from drift region for improved performance, 25 explosives, 251 formation with primary amines, 51 formation with water, 81–84 interface in IMS/MS, 222 intermediate in formation of product ions, 6 metal halide clusters, 327 protonated monomer and proton bound dimer formation, 95 separation of ions by formation of cluster ions, 107, 300 CN, tear gas, 265 CO2, carbon dioxide adducts with ions, 223, 225 drift gas, 48 removal from enclosed atmosphere, 305 mobile phase in SFC, 238 Cocaine, 27, 210, 214, 238, 259, 261, 264, 270 Codeine, 264, 271 Coefficient of mobility definition of mobility coefficient, 3, 5 dependence upon E/N, 42, 60–66 formula for coefficient of mobility, 8, 45 relationship to diffusion, 41 Cohen, M.J., 15, 17, 35–37, 180, 234, 235, 268 Collisions field dependent mobility, 60–65 free mean path, 42–43, 80 gedanken experiment, 59 ion formation, 80–88 Combustion, vapors determined by GC/DMS, 322–323 Commercial instrumentation, 18, 24, 30–33, 159, 160, 209, 233 Competition charge completion in electrospray ionization, 263 selectivity by competitive ionization, 105–106

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_340... 06/05/2007

Página 2 de 2

Concentrator, see also preconcentration explosive vapor collection, 257 bacteria concentrator for pyGC/IMS, 285 Conductors electrical conduction in drift tubes, 170–171 heat conduction in materials, 172 Conformation of biomolecules, 229–231, 281–282, 291, 331 Contaminants air of Spacestation by GC/IMS, 211, 307–308 drift tube, 107 surface contamination on semiconductors, 221, 238, 303–305 Contraband, see Explosives; Drugs of abuse Cooks, R.G., 176, 291 Corona discharges, 10, 14, 30, 35, 81, 113, 138, 177, 178, 222, 228, 239, 299, 311, 333 Corrosive formation of corrosive gases in electric switches, 302 determination of corrosive gases by IMS, 293–295 Cotton, vapors from burning, 322–323 CP-4900, a commerical GC-DMS analyzer, 32, 214–216 Crime scene analysis, see Forensics

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_340... 06/05/2007

Página 1 de 2

Page 341 Cross-section biomolecules, 228–231 role in mobility, 8, 9, 42–57, 59–62, 69, 72 D Dam, R., 36, 100, 111 Decane in respired air for lung disease, 276 Decaying flesh, detection of vapors by IMS, 278–281 Declustering of ions between IMS and MS, 227 effect on drift time, 102 model of field dependence of mobility, 60–66 Decomposition, of gases inside electrical insulating switches, 302 Decomposition, of ions between IMS and MS, 225 in drift tube, 109–110, 227 Designs of drift tubes. 145–146 Desolvation of ions after formation in electrospray ionization, 229 in drift tube, 62, Desorber as inlet for IMS, 213 Desorption explosives with heated anvil, 213 IMS to monitor desorption process, 301–302 laser in MALDI, 230–231, 273, 281 particles from luggage, 258 quantitative response with thermal desorption, 98–102 SPME fibers, 280 thermal desorption of polymers, 324–325 vapors from filter paper sample, 262 Detector differential mobility detector, 197–200, 214–215 Faraday detector, 4–5, 15, 164–165, 335 IMS as detector for GC, 25, 187–194 lightweight chemical detector, 163, 332 Lovelock ionization detector, 12, 15 mass spectrometer as detector, 21 Dialkylphthalates quantitative response, 98–99 Diamines bacterial vaginosis, 276 conformation studies with mobility, 282 DICE, 22 2,4-Dichlorophenylacetic acid, 97 Diesel, 12, 306 Differential mobility spectrometer (DMS) air monitoring, 320–321 drift tube description, 159–164 GC detector, 168, 197–200, 206, 323 inlet for mass spectrometer, 231 principles, 60–66 Diffusion of ions, 39–41 Dimer ion, see proton bound dimer Dimethylhydrazine, separation from ammonia with modified drift tube a tmospheres, 106–107 Dimethylsulfoxide (DMSO), reagent gas to control selectivity, 202, 316 Dinitrotoluene (DNT), 209, 250, 260 Dipicolinic acid (DPA) from pyrolysis of bacteria, 285 Dipole moment, role in mobility of ions, 41–43, 69, 73 Dissociative charge transfer, 104, 252 electron capture, 110 DMMP, dimethylmethylphosponate, 245 DMS, see Differential mobility spectrometer Drugs of abuse, 169, 209, 212, 243–244, 259–263

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_341... 06/05/2007

Página 2 de 2

E Ecgonidine, 213 ECLSS, 307 Efficiency gas chromatography separating efficiency, 190, 208 ionization, 96 membrane efficiency, 97 Effluent monitoring of stack emissions, 100, 294 EGDN, Ethylene glycol dinitrate, 97, 250, 253, 255, 260 Eiceman, G.A. 86, 233, 262, 274, 285 Einstein Equation, 41–42 Elastomers, 173 Electric fields dependence of mobility on electric field, 60–63 electric field with ion mobility, 41–48 role of electric field in aspirator design, 160–161 e−, the electron reactions in air, 43, 80–85 sources, 6–10, 297 Electron affinity, 84, 244, 294, 298 ECD, electron capture detector, 5, 12, 137 Electronegative, electronegativity, 243, 250, 252, 265

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_341... 06/05/2007

Página 1 de 2

Page 342 Electrospray ionization (ESI) inlet for drift tube, 229–231 ionization of biomolecules, 281–282, 284 metal ions from ESI, 326 study of charge competition within ESI, 263 ELISA, 287–288 Emission electron from thin film, 297 ionization of air by electron, 81 nicotine in air, 318 Enflurane, 274–275, 289 Enthalpy of gas phase ion molecule reactions, 89, 105 Entropy of gas phase ion molecule reactions, 89 Environics Oy, Finland, 32, 33, 160–161, 246, 265, 306 Environmental Technology Group (ETG), 245, 249 Environmental vapor monitor (EVM), 210 EnviScreen, 32, 306 Enzyme use in ELISA-IMS for bacteria detection, see ELISA Erwinia Herbicola, detection in air by pyrolysis GC IMS, 217, 237, 285–286, 291 ESI, see Electrospray ionization Esters, 8, 204, 296 Ethanol, 161, 182, 204, 308 Ethers, 189 Ethylbenzene, 298 Exhaled breath, volatile compounds, 274–276, 297 Exhaust, 12, 295 Explosives, application of IMS, 23, 26, 249–251 data base for explosives, 259–260 measurement by IMS, 252–259 Exponential dilution for sample preparation, 127 Extracellular enzymes, basis for detection of bacteria, see ELISA EXtrelut, 297 F FAIMS, see Field asymmetric ion mobility spectrometry False positive response, 261, 262 Faraday plate, 15, 30, 38, 145–146, 150, 157, 164, 224, 335 Federal Aviation Administration (FAA), 37, 253 Federal Bureau of Investigations (FBI), 262 Femotscan, 32 Fenn, J.B., 273 Fetterolf, D.D., 262 Fick’s First law of diffusion, 40–41 FID, see Flame Ionization Detector, Fields, electric effects on spectra, 103, 223, 334 for field dependent mobility, 62–63, 93 in mobility spectrometer, 3, 144, 149–151 Field asymmetric ion mobility spectrometry (FAIMS) development and description, 155–158 principles, 27, 60–66 use for biological measurements, 231 Field ion spectrometry (FIS), see field asymmetric ion mobility spectrometry Films, see also surface contamination ablation of by laser desorption, 324 surface contamination with films, 304–305 Filter paper for sampling, 96, 98–102, 254, 257 sieves for purification of gases in filters, 127, 145–146 Filtering of noise by digital signal averaging orprocessing, 21, 146, 147, 165 Fish quality from measurement of odors, 182, 278, 289

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_342... 06/05/2007

Página 2 de 2

Flame as ion source, 10, 14, 34, 143 Flame ionization detector, 143–144, 159, 206, 275 Flask, exponential dilution, 126–128 Fluoride, see Hydrofluoric acid Fluorocarbons, 299, 311 Fluoropolymers, 171 Food freshness, 273–281 Forensic uses of IMS, 174, 243, 253, 263 Formation product ions, 88–92 reactant ions, 80–88 Fourier transform, 153 Fragmentation appearance in mobility spectra, 103, 172 ion fragmentation at ambient pressure, 88, 90, 91, 93, 100, 167, 208 Fragment ion, 103–104, 110, 123, 167, 227, 251–252, 261, 299 Franklin GNO, 15–18, 20, 224 Freon, 306 Frequency of electric fields in differential mobility spectrometry, 155 Fugitive emissions, 319, 321–322 Fungi volatile from, detection by GC IMS, 262–263, 279–280

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_342... 06/05/2007

Página 1 de 2

Page 343 G GA, nerve gas, 249 Gas chromatography (GC), see also Chromatography, gas combined with IMS, 234 description, 186–216 DMS detector, 197 high speed, 25, 298 pyrolysis inlet, 217 Gasoline, 236, 296, 298, 311, 322–323 Gas-phase chemistry ambient pressure, 14 chemical weapons, 244 lachrymators, 265 Gate, see ion shutter Gaussian Peak Shape, 83, 205, 207 GB, nerve gas, 206, 249 GC-Ionscan, 31, 212–215 GE-Interlogix, 24, 31, 254, 257, 258, 318 GID-3, 31, 245, 246 Globigii, 285, 286 GP-IMS, 249 Gram-negative bacteria, 217, 285, 286 Gram-positive bacteria, 217, 285, 286 Graseby Dynamics, Ltd, UK, 22, 30, 151, 167, 170, 210, 211, 245, 247 Grass, vapor from burning, 322–323 Grid, see Aperture grid; Ion shutter Gunpowder, 258, 260 Guevremont, R., 68, 181, 239, 240, 271, 290, 330 H Halide ions, 107, 110, 327 Haloalkanes, 110, Halocarbon(s), 12, 21, 84, 85, 197, 199, 200, 296, 298, 307 Halothane, 237, 274, 289 Hand-held, 123, 126, 127, 163, 167, 210, 212, 263, 267, 297, 300, 306 Harden, C.S., 22, 114–115, 235–239, 267, 286, 289 Hard-sphere model, 58, 284 Harrington, P., 175, 183, 263, 268, 311 HBr, see hydrobromic acid HCI, see hydrochloric acid HD, Mustard gas, 206, 245, 249 Headspace sampling, 277, 261–263, 297, 298, 304 Helium atmosphere for ion source 11, 48–59, 75, 148, 228–229, 282, 301 modification of drift gas in IMS Herbicides, 141, 219, 321 Heroin, 209, 238, 261, 264, 270 Hexaglycine, 281 hexamethylene tri diamine (HMTD), 260, 270 HF, see hydrofluoric acid, 105–106, 124, 175, 293–295 High-resolution capillary columns, 208 mobility spectrometer, 141, 150, 151, 170, 183, 244, 267, 269, 291, 335, High-volume sampling for collecting biological aerosols, 213, 217, 262 Hill, Jr., H.H., 28, 59, 97, 114, 147, 173–180, 190, 200, 234–239, 270–271, 281, 311 His Majesty’s Explosive (HMX), 213, 250, 255, 260 Histamine, 276 Histidine, 276 Histogram, 161 Honeywell, 22 Horning, E.C., 21, 34, 36

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_343... 06/05/2007

Página 2 de 2

Humidity, 259, 265, 305–306, Hydration, 34, 64, 82, 83, 85, 222 Hydrazine (HZ) detection limit, 98 separation from ammonia, 106, 115, 300, 311 Hydrobromic acid (HBr), 295 Hydrochloric acid (HCl), 293–294 Hydroelectric, 33, Hydrofluoric acid, 124, 130, 166, 283–296 Hysteresis, 127 I I-CAM, 247 Ion Shutter controller, 119, 121 materials, 170 pulsed, 120, 151–154, 193–194, 198, 224, 227 second or dual, 145, 147, 194, 199, 201, 224 Illicit drugs, see Drugs of abuse Illness, 275–276 IMCELL™, 161, 177, 246, 267, 306 Immediate danger to life and health (IDLH), 265 Imide, see Polyimide Immunoassay, combination with IMS, 280, 284, 287, 288, 291 Impurities gas, 125, 147, 175, 186, 221, 301, 302, 312

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_343... 06/05/2007

Página 1 de 2

Page 344 materials, 169 surface, 324 Incineration, 293 Infected, 277 Inhalation, 171, 274, 275, 296, 315, 317 Insulation electrical, 312 thermal, 190–192 Insulator material for in drift tube, 147, 153, 171, 172 Interface drift tube, 122, 127 mass spectrometers, 222, 223 Interferences 104–106, 111–112, 297 Internal atmosphere, recirculated gas supply, 125, 305, 307 International Society for Ion Mobility Spectrometry (IS-IMS), 27, 28, 258 Iodoaniline, 59 Ionanalytics, 27, 32 Ion-ion repulsion, 96, 154 Ionization ambient (atmospheric) pressure chemical ionization, 21, 79 chemical ionization mass spectrometry, 14, 20, 24, 88 detection limits, 96 gases and air, 12, ionization based military detectors, 22 ionization of explosives, 250–252 proton transfers to form product ions, 88 sources by x-ray, 10 spark discharge, 12 electrospray ionization, 244 Ion-molecule interaction, 12, 47 Ion-neutral interactions, 43 Ionscan, 23, 24, 26, 31, 133, 134, 212–215, 255, 257, 262, 309 IonTrack, 23, 31, 37 Irradiation, 297, 289 IS-IMS. See International Society for Ion Mobility Spectrometry, Isoflurane, 237, 274–275, 289 Isomers difference in mobility, 8, 59, 221 structural information, 228, 239, 290 Isoprene, 276, 308 Itemizer, 23, 24, 26, 255 IUPAC, 30, 114 J Jarrold, M.F., 179, 239, 281, 291 JCAMP-DX, 29, 38 K Karasek, F.W., 18–20, 34–36, 68, 97, 113–115, 145, 180, 200, 234, 237, 261 Karpas, Z., 68, 97, 113–115, 173, 180, 183, 236, 239, 278, 289 Kebarle, P., 14, 34, 35, 113, 141, 176 Keller, R., 25, 36, 114 Kinetics ion formation, 14, 93 ion dissociation or decomposition, 104, 110 Ko, see Reduced mobility coefficient L LABVIEW, 30 Lachrymator (s), 263, 265, 271, 272 Lactobacilli, 276

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_344... 06/05/2007

Página 2 de 2

laminar flow, 22, 160 lamp, discharge for ionization, 136, 140, 158, 178, 202, 203, 206, 275, 280, 296, 320 Langevin, P., 10, 34, 39, 66 Laser desorption and ionization (LDI), 122, 134, 135, 177, 326, 330 Lawrence, A.H., 97, 115, 170, 175, 183, 261, 264, 268, 329 Legionnaire’s Disease, 326 Leonhardt, J.W., 177, 183, 267, 297, 311 Liquid chromatography (LC), 20, 36, 121, 185, 218, 237, 238, 290 Light-weight chemical detector (LCD), 152, 163, 183, 248, 332 Limero, T.F., 38, 115, 176, 236, 311, 312, 313 limit of detection (LOD), 96, 262, 285, 299, 300 linear response, 128 Lovelock, J.E., 12–13, 15, 35, 37, 177 Luggage, 23, 26, 250, 253, 254, 257–261, 269 Lumber, 324 Lung disease, 274 Lysine, 276 M M-43A1, 22 M-8A1, 22 Macor®, 146, 171, 184 Mason, E.A., 14, 34, 35, 39, 40, 44–47, 70 Mass-analyzed, 229, 232 Mass-identified, 186, 221, 238, 270, 318 Mass-mobility correlation, 50–51 Materials, for construction of drift tube, 119, 121, 126, 133, 144, 169, 170, 172

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_344... 06/05/2007

Página 1 de 2

Page 345 Matrix assisted laser desorption and ionization (MALDI), 136, 142–143, 230–232, 283–284 Matrix interference, 122, 136, 158, 297 McDaniel, E.W., 14–15, 34, 39, 40 MDL, see Miniumum detectable level MEA, see monoethanolamine Medical uses of IMS, 188, 274–291 Membranes to isolate of drift tube with inlet, 125–127, 133, 173 Mercaptans detection in natural gas, 163, 215, 235 Mescalaro, New Mexico, 28 Metabolic, volatile products, 280 Metal ions by ESI IMS, 326 Metamphetamines, 262 Methane ionization at ambient pressure, 207 matrix for detection of odorants in natural gas, 215 reagent gas, 202 Methanol component of air on Space Station, 308 extracts of wood, 324 reagent gas, 123, 202 Methylene chloride reagent gas, 124 Methylpolysilicone, 173 Methylsalicylate as drift gas modifer, 124 Methyl-tertbutyl-ether (MTBE), 206, 296–297 Metro, M.M., 36, 234 MH+, see Protonated monomer Mica as insulator for drift tube, 153, 171 Microbial, see Microorganisms Micro-DMx™, 332 Micro-fabricated drift tubes, 235 Microorganisms, 236, 276, 278, 289, 290, 316 Microparticles, micro-particles, 253, 262 Miller, R.A., 37, 158, 178, 181, 233, 235, 237, 240 Miniature drift tubes, 150–152, 316 Miniumum detectable level (MDL) capillary GC/IMS, 206 chemical warfare agents, 249 continuous monitoring of amines, 322 definition, 96–97 impurities in electrical switches, 302 Mixtures charge competition in IMS response, 110–111, 208 separation of peptides by mobility, 283 MMH, see monomethylhydrazine MMS-160, 227–228 Modifiers in drift gas for improved separation of ions 300, 305 Moisture control or management of moisture in analyzer, 148, 294, 335 effect on mobility importance on ion formation, 89–90, 102, 104, 245, 298, 333 Mold, 280 Molybdenum in surface ionization source, 141 Monoethanolamine (MEA), 305–306 Monomer, see Protonated monomer Monomethylhydrazine (MMH), 98, 311, 330, Morphine, 261, 263 MSA, 27, 32 MTBE, see methyl-tertbutyl-ether Multi-capillary column, 203, 209, 296 Multiphoton ionization, 123, 325 Mustard gas, 245, 267 N

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_345... 06/05/2007

Página 2 de 2

N2, see Nitrogen Nafion, 133 Nanoelectrospray, 143 Naphthalene, 123 Narcotics, see Drugs of abuse NASA, 16, 28 NATO, 244, 248 Nazarov, E.G., 37, 158, 178, 181, 233, 235, 237, 240 Nd-YAG, 140 Nebulize, 131 Needle corona discharge needle, 138, 140 syringe needle, 131, 132 Nerve agent, see Chemical warfare agents Neural networks analysis and categorization of spectral, 166 use in chemical measurements, 183 Neutralization, 85, 199 NH3, see Ammonia Nichrome, 327 Nickel Nickel carbonyl, 97, 326 radioactive nickel ion source, 131, 137, 295 surfaces, 170 Nicotinamide, 122 Nicotine, 97, 263, 317–319 Nielson, R.A., 151, 153, 180 Nitrate, ammonium nitrate, 256 nitrated organic compounds, 250 nitrate anion, 251 urea nitrate, 256 Nitrite, 326, 330

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_345... 06/05/2007

Página 1 de 2

Page 346 Nitrogen, 65, 80, 81, 97, 123, 127, 130, 133, 138, 145, 148, 283, 301 Nitroglycerine, 250, 255 ortho-Nitrophenyl-beta-D-galactosidase, enzyme in ELISA based detection with IMS, 288 n-nonylamine, 277, 279 Noise chemical noise, 288, 335 signal to noise, 154, 165, 166, 199, 225 O O2, see Oxygen Odor detection of in natural gas (odorants), 163, 215, 235 decaying food, 278–279 fish, 161, 278 medical diagnosis, 274 Oil fuel oil, 256 interference, 299 Olfactometry, 280 Oligomers, 281 Oligonucleotides, 143, 281 Oligosaccharides, 143 Organophosphates, 62, 64, 110, 203, 243, 244, 248 Orifice, between IMS and MS, 222–223 Ortho-nitrophenol (ONP), 288 Ortho-nitrophenyl-beta-D-galactoside (ONPG), 288 Outgassing vapor from polymer, 36 Ovalbumin, 217, 285–286 Oxygen acidity of oxygen hydrogen bond, 91 ionization of by electron attachment, 113, 140, 297 ionization of air, 6–7, 84 oxygen-adduct ions, 7, 84 P Particles, 133, 217, 253–257, 262, 269, 285, 295, 322, 327 Particulate, 126, 137, 172, 253, 320, 322 Pathogenic, also, pathogens, 276, 286 PCA, see Principal component analysis PCBs, see Polychlorinated biphenyls PCP, Inc. 18, 19, 30, 31, 115, 146, 220, 224, 227, 233, 269 Peak-broadening, see band-broadening Pentachlorophenol, 299 Pentaerythritol nitrate (PETN), 209, 250, 253, 256, 258, 260 Pentaglycine, 281 Pentanone, 5, 162, 302 Peptides characterization by ESI IMS, 291 mobility spectra, 281, 283 structure mobility relationships, 290 Perchlorate, 326, 330 Perfluorocarbons, 297, 311 Perfumes, 280, 290 Permeation sources, 124, 175, 257 Peroxide, 276 Peroxide explosives, 256, 269, 270 Pesticides, 12, 210, 212, 266 Petroleum, 325 Pharmaceutical, 309, 313, 264, 270 Phenol(s) acidity of OH bond in substituted phenols, 91 fluorinated phenol, 91 2,4-dinitrophenol, 258

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_346... 06/05/2007

Página 2 de 2

2,6-di-t-butyl-4-methylphenol (BHT), 299 p-nitrophenol, 221 pentachlorophenol, 299 quantitative response to phenol, 99–102 Phosgene, 249 Photoionization (PID) GC IMS with photoionization source, 178 response, 123, 140, 178, 320, UV source, 136, 295 Phthalate esters, 98, 133 Picolinic acid, 217, 237, 285 Plasma chromatography, 15–20, 186, 219 Platinum, 122, 141 Plexiglas, 325 Polarizable, 39, 49, 58, 59, 73, 77 Polyalanine, 281, 282 Polychlorinated biphenyls (PCBs), 299 Polycyclic aromatic hydrocarbons, 134, 177, 325, 330 Polyglycine, 281, 282 Polyimide, 171, 192 Polypropylene Insulating rings, 171 membrane, 161, 173 release of vapors, 304 Polytetrafluoroethylene (PTFE), 171–172, 304 Portal(s) for screening explosives, 250, 253, 254, 256, 257–258, 269 Positive alpha function, 63–64 polarity ions, 5–6, 10, 15, 53, 80–90 Postdetonation, 250, 253 Potassium ions, 67, 310, 327

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_346... 06/05/2007

Página 1 de 2

Page 347 Potential electric field gradients, 149–151 ionization potential of air, 80 ionization potential of organic compounds, 140 interaction potentials, 14, 39–48 orifice potentials, 222, 223 potential differences in ion shutter, 154 potential energy surfaces, 70, 71 Poultry analysis, 278 Pre-amplifier, 150 Precision, 98, 99, 101, 286, 309 Preconcentration of sample, 23, 25, 111, 122, 129, 202, 206, 244, 298, Prefractionation, 185, 275, 280, 281, 322 Preheating of drift gas, 172 Preparation of sample, 97, 98, 128, 263, 265, 297 Prescription drugs, 261–264 Preseparation, see Prefractionation Preservative, 299, 324, 329 Pressure drop across GC column, 210 low pressure IMS, 229–230 subambient pressure, 210, 216, 274, 282 vapor pressure of compounds, 97, 122, 133, 214, 250, 253, 300 Primary amines, 50, 59, 70–74, 141 Principal component analysis (PCA), 278, 280 Propellants, 98 Propylbenzene, 276 Protein(s), conformation studies, 282–284 decomposition of proteins, 276 electrospray IMS of proteins, 131 general discussion, 29, 281 pyrolysis analysis, 217, 285–286 Protonated monomers dynamics with RIP and proton bound dimer, 93–95 formation of, 5–6, mobility spectra, 89, 156, 226, 319 Proton bound dimers concentration dependence, 93 temperature effects, 103 Proton bound trimers, formation at low temperature, 103, 245 Protons, hydrated protons in reactions, 81, 83, 91, 104, 105, 202 Psilocybin, 262, 271 PTFE, see Polytetrafluoroethylene PTIMS, Rotem Industries, Israel, 321–322 Pulse corona discharges, 239, 333 ion shutter, 153–154, 165 laser in MALDI, 142 photodischarge lamp, 140, 297 vapor generator, 269 Pump cryogenic pumped mass spectrometer, 223 differentially pumped region for IMS/MS, 223, 228 gas pump in IMS, 172, 210 Purification of drift gas, also, purity, 65, 84, 126, 148, 208, 210, 305 Putrescine, 276–277 Py-GC, see Pyrolysis Py-GC-IMS, see Pyrolysis Pyridine, 53, 54, 285 Pyrogallol, 324 Pyrolysis, 216–217, 237, 284–287, 291

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_347... 06/05/2007

Página 2 de 2

Q Q-Scent Ltd., 32, 276 Quadrupole mass analyzer, 15, 193, 198, 223, 227–229, 233, 282, 291 moments, 41, 43, 69 Quantitative, response to vapor concentration, 133, 188, 205–208 Quasimolecular ion, 261, 299 R Radioactive ion sources, 137 RAID, 31, 245, 247, 249 Rasulev, U.Kh. 37, 68, 141, 179, 181 RDX, 97, 213, 250, 253, 255, 258, 260 Reactant ion peak (RIP), 5, 6, 57, 89, 93, 106, 167, 192, 193, 196, 203, 212, 277 Reagent gas, 104–107, 122–125, 245, 252, 316–318, 322, Re-circulated flows in handheld IMS, see Internal atmosphere Reduced mobility coefficient, 8, 50, 85, 186, 281 Reduced pressure effect on mobility, 29, 60, 274, 282 Repeatability, 99–102 Reproducibility, see Repeatability Repulsion between ions, 40, 47, 96, 135, 154, 282 Residence time of ions in drift region, 25, 80, 93, 96, 159, 190, 334 Resistive coating for drift tube, 150 Resolution definitions and examples, 164, 194 high resolution IMS, 141, 149, 151, 170, 239, 335

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_347... 06/05/2007

Página 1 de 2

Page 348 high resolution gas chromatography, 28, 208 Resorcinol, 324 Respiration of vapors from skin, 270, 274 Retention time, 188, 194–197, 203, 205, 207, 212, 297, 323, RFIMS, see FAIMS or DMS Ribonuclease, 281 RIP, see Reactant ion peak Rocket propellant, 98 Rotem Prototype, from Rotem Industries, Israel, 149 Russell, D.H., 38, 179, 232, 239, 281 Rutherford, E., 10, 34 S Sabre, 31, 254, 262 Sapphire beads, as insulators in drift tube, 20, 145, 146, 171 Saturation of response, 89, 128, 243 Scale of proton affinities, 90, 91 Segmented design of drift tube, 147 Selectivity control in ionization, 106 enhanced, 202, 123 Self-modeling, in data processing, 166 Semi-permeable membrane, 133 Semtex, 256 Separation voltage or field, 158 Separation chromatographic separations, 111, 237, 238, 296 ion separations, 4, 5, 7, 15, 59, 96, 124, 136, 306 Servo inlet, 128, 192 SF6, see Sulfur hexafluoride Shahin, M.M., 14, 81 Shape of ion on mobility, 5, 8, 57–59, 284, 288 peak shape, 5, 84, 104, 164, 206, 207, 300 Sheath-flow in inlet for GC-IMS, 129 Shells, for drift tube, 149, 172, 191, 335 Shutter, see Ion shutter Shuttle, U.S. Space, 100, 300, 305 Sick-building syndrome, 316 Side flow design of drift tube, 131 Siegel, M.W., 87, 113, 177 Sieve, molecular sieves for drying drift gas, 127, 148 Signal processing, 146, 147, 198, 263 Signal-to-noise, 96, 165, 199, 225, 326, SIMPLISMA, software for processing data, 166, 271 Simulant, 244, 249 Sionex Corporation. 27, 30, 32, 156, 158, 159, 164, 332 Size-to-charge, 8 Skatole, 276 Skin, respiration of vapors with IMS detection, 275 Smith Detection, 23, 24, 30, 31, 149, 152, 164, 218, 245–248, 255–257, 265, 306, 309, 332 Smoke vapors, detection by IMS, 137, 263, 322 Snyder, A.P., 114, 116, 177, 216, 217, 236, 237, 267, 285, 291 Sodiated substances, 143, 179, 239, 283, 291 Sodium in peptides, 283 mobility of sodium ion, 326–327 sodium chloride adducts, 229 Software analysis of spectra, 166 acquiring signals, 30, 165, 306 molecular modeling, 334 spectral libraries, 38, 333 Soil analysis, 293, 296, 325, 330 Solid phase micro-extraction (SPME), 132 Solvation, 34, 62, 65 Solvent detection in air, 12, 318–319 Space station, air monitoring by GC/IMS, 28–29, 100, 210–213, 307–308

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_348... 06/05/2007

Página 2 de 2

Spangler, G.E., 37, 86, 113, 114, 116, 149, 175, 180, 181, 267–269 Spermidine, 276 Spermine, 276 Spores, 217, 285–286, Stability thermal, 188 ion stability, 109, 225 response, 29, 98, 100, 305 Stach, J., 174, 297, 311 Stainless steel, properties and use in drift tube, 170, 173 Standards, Chemical, 38, 112, 116, 175, 197, 334 Stone, J.A., 38, 113, 115, 116, 180 STS-37, 100 Styrene, 276 Submarine(s), 12, 305 Substations, monitoring switches in hydroelectric substations, 302 Subway attack in Japan, 306 Sulfur hexafluoride, 53, 113, 302 Supersonic expansion, of gases in interface between ambient pressure and high vacuum of mass spectrometer, 220, 222 Surface analysis, 21, 324–326 contamination, 309

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_348... 06/05/2007

Página 1 de 2

Page 349 energy or potential surfaces, 44, 70–71 ionization source, 141, Swarms of ions, 3, 4, 58, 113, 154 T Tablet, pharmaceutical, 262 Tabrizchi, M., 68, 178, 326, 330 Taggants, in explosives, 253 Tammet, H.F., 36, 161, 182 Tandem mobility spectrometer, 336 Tarver, E.E., 154, 180 TATP, see Triacetone triperoxide TDI, see Toluene diisocyanate Teflon®, 149, 170, 171, 173 Temperature gas in IMS drift tube, 148 mobility correlations, 52–57 mobility spectra, 102–103, 167 reduced mobility, 3 Tenax, 302, Terror, also terrorism, terrorist, 23, 250, 306, 331 Tetrachloroethene, 296 Tetraglycine, 281 Tetramines, 276 Thermalized ions, 42–44, 80–81, 229, Thomas, C.L.P. 129, 174, 175, 176, 228, 234–235, 239 Time-of-flight mass spectrometry (TOF-MS), 141, 157, 223, 230 TMA, see Trimethylamine TNT, see Trinitrotoluene Tobacco, 141, 322 Toluene, 100, 167, 215, 236, 296–298, 308, 311 Toluene diisocyanate (TDI), 167, 168 Townsend, J.S., equation, formula, 39–43, 60, 138 Tracers, atmospheric, 12, 35, Traffic, emissions from, 320–321 Transdermal, 317, 329 Traps ion trap mass spectrometer, 179, 219, 229–231, 239, 290 particle, 254, 322 vapors by adsorption, 126, 130, 206, 275, 298, 301–302, 307, 312, 318 Triacetone triperoxide (TATP), 256, 269 Triamines, 276 Triazine, 141, 219 Tributylamine, 206 Trichloroethene, 297, 311 Trichomoniasis, 277 Triglycine, 281 Trimers, see Proton bound trimers Trimethylamine (TMA), 276, 277, 289 Ttrinitrotoluene (TNT), 97, 115, 209, 250, 269 Trioctylamine, 53 Triperoxide, 256, 269 Tritium, as source of ions, 137, 177 Tuned ion mobility spectrum, 225–228, 236 Turbomolecular pump, 223, 227 Tyrosine, 276 U Ubiquitin, 229, 281, 282, 291 Ultraviolet laser, 164 Ultraviolet light, 10, 140, 235 Underwater, determination of explosives, 250 Unidirectional flow, of drift gas, 108, 147, 190–191

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_349... 06/05/2007

Página 2 de 2

Unmanned aerial vehicles (UAV) with IMS analyzers, 233 Unsymmetrical dimethyl hydrazine (UDMH), 106–107 Urea nitrate, 256 Urine, 240, 259, 260, 263, 271, 274 V Vaginal infections, vaginosis, 276–278 Valinomycin, 281 Vaporization, vaporize, 133, 135 VaporTracer, 31, 254 Varian, 32, 214–216, 218, 265 Velocity gas molecules, 80 ion swarm, 3, 7, 10, 41, 42, 45–46, 61, 155, 160, 282 Viehland, L., 39, 47, 66–68 Volatile organic analyzer (VOA), 192, 210, 213–214 Volatile organic chemicals (VOCs), 130, 174–176, 181, 214, 235–236, 296, 299, 329 VX, nerve agent, 182, 206, 249, 267 W Wafers, impurities detection on surface of silicon wafers, 21, 304, 324 Warfare agents, see Chemical warfare agents Water ammonia in water, 300

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_349... 06/05/2007

Página 1 de 2

Page 350 BETX in water, 298 dissociation of ions, 90 effect on alpha function, 64 extraction of analyte from water samples, 133 formation of ions, 81–82 ion water collisions, 65 MBTE in water, 297 membrane for protection against water, 127 reagent gas, 100, 202–206, 316 solvation of ions, 34, 62, 63, 83, 88, 235 water samples analyzed with ESI source, 141 Wavelength laser desorption, 325 photodischarge lamps, 140 Wavelet compression for data processing, 166, 175, 183, 268 Wire-to-wire discharge, in corona ion source, 153 Wood speciation, 324 preservative, 299 Workplace, monitoring of vapors, 319, 311 X Xenon pulsed lamp for photoionization, 140 X-ray, 10, 253, 254 Xylene, 100, 296–298, 308, 320 Y Yeast, 182, 277 Z Zinc (Zn), 326

http://www.netlibrary.com/nlreader/nlreader.dll?bookid=110950&filename=page_350... 06/05/2007