x-ray spectrometry: recent technological...

X-Ray Spectrometry:Recent Technological Advances

Edited by

Kouichi Tsuji

Department of Applied Chemistry, Osaka City University, Japan

Jasna Injuk

Micro and Trace Analysis Center, University of Antwerp, Belgium

René Van Grieken


Innodata0470020423.jpg


Edited by

Kouichi Tsuji

Department of Applied Chemistry, Osaka City University, Japan

Jasna Injuk


René Van Grieken


Copyright 2004 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,West Sussex PO19 8SQ, England

Telephone (+44) 1243 779777Email (for orders and customer service enquiries): [email protected] our Home Page on www.wileyeurope.com or www.wiley.com

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form orby any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright,Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 TottenhamCourt Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should beaddressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO198SQ, England, or emailed to [email protected], or faxed to (+44) 1243 770620.This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is soldon the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expertassistance is required, the services of a competent professional should be sought.

Other Wiley Editorial Offices

John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA

Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA

Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany

John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809

John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1

Wiley also publishes its books in a variety of electronic formats. Some content that appearsin print may not be available in electronic books.

Library of Congress Cataloging-in-Publication Data

X-ray spectrometry : recent technological advances / edited by Kouichi Tsuji, Jasna Injuk,René Van Grieken.

p. ; cm.Includes bibliographical references and index.ISBN 0-471-48640-X (Hbk. : alk. paper)1. X-ray spectroscopy. I. Tsuji, Kouichi. II. Injuk, Jasna. III. Grieken, R. van (René)

[DNLM: 1. Chemistry, Analytical. 2. Spectrometry, X-Ray Emission – instrumentation.3. Spectrometry, X-Ray Emission – methods. QD 96.X2 X87 2004]QD96.X2X228 2004543′.62 – dc22

2003057604

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN 0-471-48640-X

Typeset in 10/12pt Times by Laserwords Private Limited, Chennai, IndiaPrinted and bound in Great Britain by Antony Rowe Ltd, Chippenham, WiltshireThis book is printed on acid-free paper responsibly manufactured from sustainable forestryin which at least two trees are planted for each one used for paper production.

http://www.wileyeurope.comhttp://www.wiley.com

Contents

Contributors . . . . . . . . . . . . . . . . . . vii

Preface . . . . . . . . . . . . . . . . . . . . . . xi

1 Introduction . . . . . . . . . . . . . . . . . . 11.1 Considering the Role of X-ray

Spectrometry in Chemical Analysisand Outlining the Volume . . . . . 1

2 X-Ray Sources . . . . . . . . . . . . . . . . 132.1 Micro X-ray Sources . . . . . . . . . 132.2 New Synchrotron Radiation Sources 292.3 Laser-driven X-ray Sources . . . . . 49

3 X-Ray Optics . . . . . . . . . . . . . . . . . 633.1 Multilayers for Soft and Hard

X-rays . . . . . . . . . . . . . . . . . . 633.2 Single Capillaries X-ray Optics . . 793.3 Polycapillary X-ray Optics . . . . . 893.4 Parabolic Compound Refractive

X-ray Lenses . . . . . . . . . . . . . . 111

4 X-Ray Detectors . . . . . . . . . . . . . . . 1334.1 Semiconductor Detectors for

(Imaging) X-ray Spectroscopy . . . 1334.2 Gas Proportional Scintillation

Counters for X-raySpectrometry . . . . . . . . . . . . . . 195

4.3 Superconducting Tunnel Junctions 2174.4 Cryogenic Microcalorimeters . . . . 2294.5 Position Sensitive Semiconductor

Strip Detectors . . . . . . . . . . . . . 247

5 Special Configurations . . . . . . . . . . . 2775.1 Grazing-incidence X-ray

Spectrometry . . . . . . . . . . . . . . 277

5.2 Grazing-exit X-ray Spectrometry . 2935.3 Portable Equipment for X-ray

Fluorescence Analysis . . . . . . . . 3075.4 Synchrotron Radiation for

Microscopic X-ray FluorescenceAnalysis . . . . . . . . . . . . . . . . . 343

5.5 High-energy X-ray Fluorescence . . 3555.6 Low-energy Electron Probe

Microanalysis and ScanningElectron Microscopy . . . . . . . . . 373

5.7 Energy Dispersive X-rayMicroanalysis in Scanning andConventional Transmission ElectronMicroscopy . . . . . . . . . . . . . . . 387

5.8 X-Ray Absorption Techniques . . . 405

6 New Computerisation Methods . . . . . 4356.1 Monte Carlo Simulation for X-ray

Fluorescence Spectroscopy . . . . . 4356.2 Spectrum Evaluation . . . . . . . . . 463

7 New Applications . . . . . . . . . . . . . . 4877.1 X-Ray Fluorescence Analysis in

Medical Sciences . . . . . . . . . . . 4877.2 Total Reflection X-ray Fluorescence

for Semiconductors and Thin Films 5177.3 X-Ray Spectrometry in

Archaeometry . . . . . . . . . . . . . 5337.4 X-Ray Spectrometry in Forensic

Research . . . . . . . . . . . . . . . . . 5537.5 Speciation and Surface Analysis of

Single Particles UsingElectron-excited X-ray EmissionSpectrometry . . . . . . . . . . . . . . 569

Index . . . . . . . . . . . . . . . . . . . . . . . . 593

Contributors

F. AdamsDepartment of Chemistry, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp, Belgium

J. BörjessonDepartment of Diagnostic Radiology, CountryHospital, SE-301 85 Halmstad, Sweden

A. BrunettiDepartment of Mathematics and Physics,University of Sassari, Via Vienna 2,1–07100 Sassari, Italy

R. BythewayBEDE Scientific Instruments Ltd, BelmontBusiness Park, Durham DH1 1TW, UK

A. CastellanoDepartment of Materials Science, University ofLecce, I-73100 Lecce, Italy

R. CesareoDepartment of Mathematics and Physics,University of Sassari, Via Vienna 2, I-07100Sassari, Italy

C. A. CondePhysics Department, University of Coimbra,P-3004-0516 Coimbra, Portugal

W. Da̧browskiFaculty of Physics and Nuclear Techniques,AGH University of Science and Technology, Al.Mickiewicza 30, 30–059 Krakow, Poland

E. Figueroa-FelicianoNASA/Goddard Space Flight Centre, Code 662,Greenbelt, MD 20771, USA

M. GaleazziUniversity of Miami, Department of Physics, POBox 248046, Coral Gables, FL 33124, USA

N. GaoX-ray Optical Systems, Inc., 30 Corporate Circle,Albany, NY 12203, USA

P. GrybośFaculty of Physics and Nuclear Techniques, AGHUniversity of Science and Technology, Al.Mickiewicza 30, 30-059 Krakow, Poland

P. HollSemiconductor Lab., MPI Halbleiterlabor,SIEMENS – Gelaende, Otto-Hahn-Ring 6,D-81739 München, Germany

J. de HoogDepartment of Chemistry, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp, Belgium

Y. HosokawaX-ray Precision, Inc., Bld. #2, Kyoto ResearchPark 134, 17 Chudoji, Minami-machi,Shimogyo-ku, Kyoto 600–8813, Japan

G. IsoyamaThe Institute of Scientific andIndustrial Research, Osaka University, 8-1Mihagaoka, Ibaraki, Osaka Pref. 567-0047, Japan

K. JanssensDepartment of Chemistry, Universiteitsplein I,University of Antwerp, B-2610 Antwerp, Belgium

viii CONTRIBUTORS

J. KawaiDepartment of Materials Science andEngineering, Kyoto University, Sakyo-ku, Kyoto606–8501, Japan

M. KurakadoElectronics and Applied Physics, OsakaElectro-Communication University, 18-8,Hatsucho, Neyagawa, Japan

S. KuypersCentre for Materials Advice and Analysis,Materials Technology Group, VITO (FlemishInstitute for Technological Research),B-2400 Mol, Belgium

P. LechnerSemiconductor Lab., MPI Halbleiterlabor,SIEMENS–Gelaerde, Otto-Hahn-Ring 6,D-81739 München, Germany

P. LembergeDepartment of Chemistry, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp, Belgium

B. LengelerRWTH, Aachen University, D-52056 Aachen,Germany

G. LutzSemiconductor Lab., MPI Halbleiterlabor,SIEMENS – Gelaende, Otto-Hahn-Ring 6,D-81739 München, Germany

S. MattssonDepartment of Radiation Physics, LundUniversity, Malmö University Hospital, SE-20502 Malmö, Sweden

Y. MoriWacker-NSCE Corporation,3434 Shimata, Hikari, Yamaguchi 743-0063,Japan

I. NakaiDepartment of Applied Chemistry, ScienceUniversity of Tokyo, 1-3 Kagurazaka, Shinjuku,Tokyo 162–0825, Japan

T. NinomiyaForensic Science Laboratory, Hyogo PrefecturalPolice Headquarters, 5-4-1 Shimoyamate,Chuo-Ku, Kobe 650–8510, Japan

J. OsanKFKI Atomic Energy Research Institute,Department of Radiation andEnvironmental Physics, PO Box 49, H-1525Budapest, Hungary

C. RoDepartment of Chemistry, Hallym University,Chun Cheon, Kang WonDo 200–702, Korea

M. A. Rosales MedinaUniversity of ‘Las Americas’, Puebla, CP 72820,Mexico

K. SakuraiNational Institute for Materials Science, 1-2-1Sengen, Tsukuba, Ibaraki 305-0047, Japan

C. SchroerRWTH, Aachen University, D-52056 Aachen,Germany

A. SimionoviciID22, ESRF, BP 220, F-38043 Grenoble, France

H. SoltauSemiconductor Lab., MPI Halbleiterlabor,SIEMENS – Gelaende, Otto-Hahn-Ring 6,D-81739 München, Germany

C. SpielmannPhysikalisches Institut EP1,Universität Würzburg, Am Hubland, D-97074Würzburg, Germany

L. StruederSemiconductor Lab., MPI Halbleiterlabor,SIEMENS – Gelaende, Otto-Hahn-Ring 6,D-81739 München, Germany

I. SzalokiInstitute of Experimental Physics, University ofDebrecen, Bem tér 18/a, H-4026 Debrecen,Hungary

CONTRIBUTORS ix

B. K. TannerBEDE Scientific Instruments Ltd, BelmontBusiness Park, Durham DH1 1TW, UK

M. TaylorBEDE Scientific Instruments Ltd, BelmontBusiness Park, Durham DH1 1TW, UK

K. TsujiOsaka City University, 3-3-138 Sugimoto,Sumiyoshi-ku, Osaka 558-8585, Japan

E. Van CappellenFEI Company, 7451 N.W. Evergreen Parkway,Hillsboro, OR 97124-5830, USA

R. Van GriekenDepartment of Chemistry, University of Antwerp,Universiteitsplein I, B-2610 Antwerp, Belgium

B. VekemansDepartment of Chemistry, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp,Belgium

L. VinczeDepartment of Chemistry, University of Antwerp,Universiteitsplein 1, B-2610 Antwerp, Belgium

M. WatanabeInstitute of Multidisciplinary Research forAdvanced Materials, Tohoku University,2-1-1 Katahira, Aoba-ku, Sendai 980-8577,Japan

K. YamashitaDepartment of Physics, Nagoya University,Chikusa-ku, Nagoya 464-8602, Japan

M. YanagiharaInstitute of Multidisciplinary Research forAdvanced Materials, Tohoku University,2-1-1 Katahira, Aoba-ku, Sendai 980-8577,Japan

A. ZucchiattiInstituto Nazionale di Fisica Nucleare, Sezione diGenova, Via Dodecanesco 33, I-16146 Genova,Italy

Preface

During the last decade, remarkable and often spec-tacular progress has been made in the method-ological but even more in the instrumental aspectsof X-ray spectrometry. This progress includes, forexample, considerable improvements in the designand production technology of detectors and con-siderable advances in X-ray optics, special config-urations and computing approaches. All this hasresulted in improved analytical performance andnew applications, but even more in the perspectiveof further dramatic enhancements of the poten-tial of X-ray based analysis techniques in the verynear future. Although there exist many books onX-ray spectrometry and its analytical applications,the idea emerged to produce a special book thatwould cover only the most advanced and high-techaspects of the chemical analysis techniques basedon X-rays that would be as up-to-date as possi-ble. In principle, all references were supposed tobe less than five years old. Due to rapid changesand immense progress in the field, the timescalefor the book was set to be very short. A big effortwas made to cover as many sub-areas as possible,and certainly those in which progress has been thefastest. By its nature, this book cannot cover thefundamental, well-known and more routine aspectsof the technique; for this, reference is made to sev-eral existing handbooks and textbooks.

This book is a multi-authored effort. We believethat having scientists who are actively engagedin a particular technique to cover those areas forwhich they are particularly qualified, outweighsany advantages of uniformity and homogeneity

that characterize a single-author book. In the spe-cific case of this book, it would have been trulyimpossible for any single person to cover a signif-icant fraction of all the fundamental and appliedsub-fields of X-ray spectrometry in which thereare so many advances nowadays. The Editors werefortunate enough to have the cooperation of trulyeminent specialists in each of the sub-fields. Manychapters are written by Japanese scientists, and thisis a bonus because much of their intensive andinnovating research on X-ray methods is too littleknown outside Japan. The Editors wish to thankall the distinguished contributors for their consid-erable and timely efforts. It was, of course, neces-sary to have this book, on so many advanced andhot topics in X-ray spectrometry, produced withinan unusually short time, before it would becomeobsolete; still the resulting heavy time-pressureput on the authors may have been unpleasant attimes.

We hope that even experienced workers inthe field of X-ray analysis will find this bookuseful and instructive, and particularly up-to-datewhen it appears, and will benefit from the largeamount of readily accessible information availablein this compact form, some of it presented forthe first time. We believe there is hardly anyoverlap with existing published books, becauseof the highly advanced nature and actuality ofmost chapters. Being sure that the expert authorshave covered their subjects with sufficient depth,we hope that we have chosen the topics ofthe different chapters to be wide-ranging enough

xii PREFACE

to cover all the important and emerging fieldssufficiently well.

We do hope this book will help analyticalchemists and other users of X-ray spectrometryto fully exploit the capabilities of this set ofpowerful analytical tools and to further expandits applications in such fields as material and

environmental sciences, medicine, toxicology,forensics, archaeometry and many others.

K. TsujiJ. Injuk

R. Van Grieken

Osaka, Antwerp

Chapter 1

Introduction

1.1 Considering the Role of X-ray Spectrometry inChemical Analysis and Outlining the Volume

R. VAN GRIEKENUniversity of Antwerp, Antwerp, Belgium

1.1.1 RATIONALE

Basic X-ray spectrometry (XRS) is, of course,not a new technique. The milestone develop-ments that shaped the field all took place severaldecades ago. Soon after the discovery of X-raysin 1895 by Wilhelm Conrad Röntgen, the possibil-ity of wavelength-dispersive XRS (WDXRS) wasdemonstrated and Coolidge introduced the high-vacuum X-ray tube in 1913. There was quite atime gap then until Friedmann and Birks builtthe first modern commercial X-ray spectrome-ter in 1948. The fundamental Sherman equation,correlating the fluorescent X-ray intensity quan-titatively with the chemical composition of asample, dates back to 1953. The fundamentalparameter (FP) approach, in its earliest version,was independently developed by Criss and Birksand Shiraiwa and Fujino, in the 1960s. Alsovarious practical and popular influence coeffi-cient algorithms, like those by Lachanche–Traill,de Jongh, Claisse–Quintin, Rasberry–Heinrich,Rousseau and Lucas–Tooth–Pine all date back to1960–1970. The first electron microprobe anal-yser (EMPA) was successfully developed in 1951by Castaing, who also outlined the fundamental

aspects of qualitative electron microprobe anal-ysis. The first semiconductor Si(Li) detectors,which heralded the birth of energy-dispersiveXRS (EDXRS), were developed, mainly at theLawrence Berkeley Lab, around 1965. Just before1970, accelerator-based charged-particle inducedXRS or proton-induced X-ray emission (PIXE)analysis was elaborated; much of the credit wentto the University of Lund in Sweden. A descrip-tion of the setup for total-reflection X-ray fluo-rescence (TXRF) was first published by Yonedaand Horiuchi and the method was further pursuedby Wobrauschek and Aiginger, both in the early1970s. The advantages of polarised X-ray beamsfor trace analysis were pointed out in 1963 byChampion and Whittam and Ryon put this furtherinto practice in 1977. There have been demonstra-tions of the potential for micro-X-ray fluorescence(XRF) since 1928 (by Glockner and Schreiber) andChesley began with practical applications of glasscapillaries in 1947. Synchrotron-radiation (SR)XRS was introduced in the late 1970s and Sparksdeveloped the first micro version at the StanfordSynchrotron Radiation Laboratory in 1980.

So around 1990, there was a feeling thatradically new and stunning developments were

X-Ray Spectrometry: Recent Technological Advances. Edited by K. Tsuji, J. Injuk and R. Van Grieken 2004 John Wiley & Sons, Ltd ISBN: 0-471-48640-X

2 CONSIDERING THE ROLE OF X-RAY SPECTROMETRY IN CHEMICAL ANALYSIS AND OUTLINING THE VOLUME

lacking in XRS and scientists began to have someambivalent opinions regarding the future role ofXRS in analytical chemistry. One could wonderwhether, in spite of remarkably steady progress,both instrumental and methodological, XRS hadreached a state of saturation and consolidation,typical for a mature and routinely used analysistechnique.

In the meantime, XRF had indeed developedinto a well-established and mature multi-elementtechnique. There are several well-known key rea-sons for this success: XRF is a universal techniquefor metal, powder and liquid samples; it is nonde-structive; it is reliable; it can yield qualitative andquantitative results; it usually involves easy sam-ple preparation and handling; it has a high dynamicrange, from the ppm level to 100 % and it can, insome cases, cover most of the elements from fluo-rine to uranium. Accuracies of 1 % and better arepossible for most atomic numbers. Excellent datatreatment software is available allowing the rapidapplication of quantitative and semi-quantitativeprocedures. In the previous decades, somewhatnew forms of XRS, with e.g. better sensitivityand/or spectral resolution and/or spatial resolu-tion and/or portable character, had been developed.However, alternative and competitive more sensi-tive analytical techniques for trace analysis had, ofcourse, also been improved; we have seen the riseand subsequent fall of atomic absorption spectrom-etry and the success of inductively coupled plasmaatomic emission and mass spectrometry (ICP-AESand ICP-MS) in the last two decades.

Since 1990, however, there has been dramaticprogress in several sub-fields of XRS, and inmany aspects: X-ray sources, optics, detectorsand configurations, and in computerisation andapplications as well. The aim of the followingchapters in this book is precisely to treat the latestand often spectacular developments in each ofthese areas. In principle, all references will pertainto the last 5–6 years. Many of the chapters willhave a high relevance for the future role of XRSin analytical chemistry, but certainly also for manyother fields of science where X-rays are of greatimportance. The following sections in this chapterwill give a flavour of the trends in the position

of different sub-fields of XRS based e.g. on therecent literature and will present the outline ofthis volume.

1.1.2 THE ROLE AND POSITION OFXRS IN ANALYTICAL CHEMISTRY

An attempt has been made to assess the recenttrends in the role and position of XRS based on aliterature survey (see also Injuk and Van Grieken,2003) and partially on personal experience andviews. For the literature assessment, which cov-ered the period from January 1990 till the end ofDecember 2000, a computer literature search onXRS was done in Chemical Abstracts, in orderto exclude (partially) the large number of XRSpublications on astronomy, etc.; still, it revealedan enormous number of publications. Figure 1.1.1shows that the volume of the annual literature onXRS, cited in Chemical Abstracts, including allarticles having ‘X-ray spectrometry/spectroscopy’in their title, is still growing enormously and expo-nentially. During the last decade, the number ofpublications on XRS in general has nearly dou-bled; in 2000, some 5000 articles were published,versus 120 annually some 30 years ago. As seenin Figure 1.1.2, XRS in general seems more alivethan ever nowadays.

However, the growth of the literature on specif-ically XRF is much less pronounced: from about500 articles per year in 1990 to about 700 in 2000,still a growth of 40 % in the last decade. While in1990 it looked like XRF had reached a state of sat-uration and consolidation, newer developments inthe 1990s, e.g. the often-spectacular ones describedin the other chapters of this volume, have some-how countered such fears. It is a fact that WDXRFremains the method of choice for direct accuratemulti-element analysis in the worldwide mineraland metallurgy industry. For liquid samples, how-ever, the competition of ICP-AES and ICP-MSremains formidable. It is striking that, while thereare still many more WDXRF units in operationaround the world than EDXRF instruments, thenumber of publications dealing with WDXRF isabout five times lower. This clearly reflects thepredominant use of the more expensive WDXRF

THE ROLE AND POSITION OF XRS IN ANALYTICAL CHEMISTRY 3

5500

5000

4500

4000

3500

3000

2500

num

ber

of a

rtic

les

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000year

Figure 1.1.1 Total annual number of articles on X-ray emission spectrometry in the period 1990–2000 (source of data: ChemicalAbstracts). Reproduced by permission of John Wiley & Sons, Ltd

5000

4000

3000

2000

1000

0

6000

1970 1980 1990 2000

120

700

2770

5400

year

Num

ber

of a

rtic

les

Figure 1.1.2 Number of articles on X-ray emission spectrometry since 1970 (source of data: Chemical Abstracts). Reproducedby permission of John Wiley & Sons, Ltd

in routine industrial analysis, where publishing isnot common, while EDXRF is mostly present inacademic and research institutions; there are manyapplications in environment-related fields whereultimate accuracies are not so mandatory.

Also the number of publications on radioisotopeXRF has been increasing from 40 in 1990 to100 in 1998, reflecting the frequent use of thetechnique in many field and on-line applications.

In Australia alone, more than 2000 portable XRFare employed in the mining and mineral industry.It is expected that the radioisotope-based on-lineinstallations will gradually be replaced by systemsbased on small X-ray tubes.

The annual number of articles dealing with var-ious aspects of the PIXE technique (but exclud-ing micro-PIXE) is in the range of 30 to 70with very prominent peaks every 3 years. These


are obviously related to the publication of theproceedings of the tri-annual PIXE Conferences(Figure 1.1.3), with many short articles. It is clear,and not only from the literature, that, of all X-rayemission techniques considered, PIXE is thrivingthe least; there is no clear growth in the litera-ture, although PIXE might still be the method ofchoice for the trace analysis of large numbers ofrelatively small samples, like e.g. for particulateair pollution monitoring using impactor depositsof aerosols. The number of PIXE installations inthe world is probably decreasing, and the futureof PIXE seems to be exclusively in its micro ver-sion; some 30 institutes are active in this field atthe moment. The literature on micro-PIXE is stillgrowing; a search on Web of Science showed thatthe annual number of articles on micro-PIXE wasaround 10 at the beginning of the previous decadeand around 35 in the last few years.

Since the early 1970s, SR-XRS has been experi-encing remarkable growth, nowadays approachingalmost 350 articles per year, with a doubling seenover the last decade. Investment in SR facilitiescontinues to be strong and with the increasingavailability of SR X-ray beam lines, new researchfields and perspectives are open today. Most ofthe presently operational SR sources belong to theso-called second-generation facilities. A clear dis-tinction is made from the first generation, in whichthe SR was produced as a parasitic phenomenon inhigh-energy collision experiments with elementaryparticles. Of special interest for the future are new

third-generation storage rings, which are specifi-cally designed to obtain unique intensity and bril-liance. SR has a major impact on microprobe-typemethods with a high spatial resolution, like micro-XRF, and on X-ray absorption spectrometry (XAS)as well as on TXRF. For highly specific applica-tions, SR-XRS will continue to grow. The costsof SR-XRS are usually not calculated, since inmost countries, SR facilities are free of charge forthose who have passed some screening procedure.Of course, such applications cannot be consideredas routine.

There are nowadays some 100 publicationsannually on TXRF, and this number has more orless doubled since 1990. However, it may seemthat TXRF has stabilized as an analytical methodfor ultra-trace determination from solutions anddissolved solids due to the fierce competition fromICP-MS, in particular. There are now only a fewcompanies offering TXRF units. But mostly forsurface analysis directly on a flat solid sample,TXRF is still unique. SR-TXRF might be oneof the methods of choice in future wafer surfaceanalysis (in addition to e.g. secondary ion massspectrometry). In addition, by scanning around thetotal-reflection angle, TXRF allows measurementsof the density, roughness and layer thickness anddepth profiling, which are, of course, of muchinterest in material sciences. New possibilitiesfor improving the performance of TXRF are inusing polarised primary radiation. SR has almostideal features for employment in combination with

300

250

200

150

100

50

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000year

num

ber

of a

rtic

les

Figure 1.1.3 Annual number of articles on PIXE in the period 1990–2000 (source of data: Web of Science). Reproduced bypermission of John Wiley & Sons, Ltd

THE ROLE AND POSITION OF XRS IN ANALYTICAL CHEMISTRY 5

TXRF. It is several orders (8–12) of magnitudehigher in brightness compared to X-ray tubes, hasa natural collimation in the vertical plane andis linearly polarised in the plane of the orbit ofthe high energy (GeV) electron or positrons. Thespectral distribution is continuous, so by propermonochromatisation, the performance of selectiveexcitation at best conditions is possible. SR offers asignificant reduction in TXRF detection limits anda remarkable improvement has been achieved overthe past 20 years from nanogram level in 1975 toattogram level in 1998.

Until recently, evolution of XRF into the micro-analytical field was hampered because of the dif-ficulties involved in focusing a divergent X-raybeam from an X-ray tube into a spot of smalldimensions. However, the development of SRsources and the recent advances in X-ray focusinghave changed the situation. Contemporary micro-XRF applications started only some 10 years agoon a significant scale, and it appears today to beone of the best microprobe methods for inorganicanalysis of various materials: it operates at ambi-ent pressure and, in contrast to PIXE and EMPA,no charging occurs. In many instances, no sam-ple preparation is necessary. The field of micro-XRF is currently subject to a significant evolu-tion in instrumentation: lead-glass capillaries andpolycapillary X-ray lenses, air-cooled micro-focusX-ray tubes, compact ED detector systems witha good resolution even at a high-count rate andno longer requiring liquid-nitrogen cooling. Com-mercial laboratory instrumentation using capillaryoptics combined with rapid scanning and composi-tional mapping capability is expected to grow, andvarious systems are commercially available. Dur-ing the 1980s, SR facilities around the world beganto implement X-ray microbeam capabilities ontheir beam lines for localised elemental analysis.Recent trends in SR micro-XRF are towards opti-misation of optics and smaller beam sizes down tothe submicrometer size.

With respect to the general applications ofXRS, it appears that environmental, geological,biological and archaeological applications makeup a stunning 70 % contribution to the literature;

undoubtedly this is far above their relative contri-butions in actual number of analysis, since XRFis certainly still a working horse in many typesof industries, for all kinds of routine analyses,but the latter applications are published very sel-dom. The number of articles dealing with environ-mental applications of XRS, in the past decade,shows a steady growth. Interestingly, the rela-tive contributions for the different topics coveredin the environmental applications, like soils andgeological material (23 %), biological materials(19 %), water (19 %), air (17 %) and waste mate-rial (8 %) have not changed considerably duringthe last decade.

Table 1.1.1 shows the relative share of labo-ratories in different countries to the literature onXRS generated in 1998 (according to AnalyticalAbstracts) and the language in which the publica-tions were written. It appears that European coun-tries produce almost one half of the total number ofpublications, while, of the non-European countries,China and Russia are leading. The low contributionof the USA is striking. There might be several rea-sons for this. Apparently, XRS is considered moreas a routine technique by the US industry and thereare almost no US academic centers working in thisfield. It is also true that in 1998, no volume ofAdvances in X-ray Analysis appeared and this cov-ers the proceedings of the popular Denver X-ray

Table 1.1.1 The relative share of laboratories in differentcountries to the literature on XRS generated in the year 1998(as covered by Analytical Abstracts) and the language in whichthe publications were written

Country Relative contribution (%) Language

China 13.4 25 % English75 % Chinese

Russia 10.2 70 % English30 % Russian

Japan 8.0 55 % English45 % Japanese

Other Asian 4.8 100 % EnglishGermany 9.3 95 % EnglishItaly 6.3 100 % EnglishUK 4.8 100 % EnglishOther European 26.2 100 % EnglishUSA 5.4 100 % EnglishOther American 5.4 100 % EnglishAustralia 3.6 100 % EnglishAfrica 2.7 100 % English


Analysis Conference. Finally, the most advancedresearch (as described in the following chaptersin this volume) may still be published in physicsjournals rather than in journals covered by Analyt-ical Abstracts. It also appeared from our literaturesearch that about one fourth of the XRS literatureis written in less accessible languages like Russian,Chinese and Japanese.

In view of the enormous advances that are beingmade in XRS and that, hopefully, are coveredwell in the following chapters of this book, onecan expect that the applications of XRS willdramatically be changed over the next few years,and that, in the literature, the distribution overfundamental aspects (probably not fully reflectedyet in the literature covered by Chemical Abstractsand Analytical Abstracts discussed above) will beradically different as well.

1.1.3 VOLUME OUTLINE

All of the chapters of this volume have beenwritten by acknowledged research and applicationleaders, the best that the editors could find in eachof the sub-fields. A relatively large fraction of themare Japanese scientists, and this may be a bonus forreaders elsewhere in the world, since only abouthalf of the advanced XRS research in Japan ispublished in English and hence it is not alwayssufficiently widely known, e.g. in the West.

All the chapters or sets of subchapters covertopics in which remarkable progress has beenmade during the last decade and which offer goodperspective for drastically changing the power ofXRS in the near future.

Chapter 2 deals with X-ray sources, which havebecome more powerful and diverse in the last fewyears. Significant improvements have been madeto the design and performances of conventionalX-ray tubes, and in their miniaturisation (whichis treated in a later chapter), but most impressivehas been the progress in micro-X-ray sources, thedevelopment of new synchrotron sources and thefirst steps towards X-ray laser and laser-inducedplasma X-ray sources applicable to XRS. Subchap-ter 2.1 (by M. Taylor, R. Bytheway and B. K.

Tanner of Bede plc, Durham, UK) describes howelectromagnetic rather than conventional electro-static focusing, for shaping and steering the elec-tron beam in the X-ray tube, allows the X-raysource dimensions to be controlled much betterthan in the past, to achieve a higher brilliance with-out target damage, to tailor the X-ray spot dimen-sions for optimising the input coupling with sub-sequent grazing-incidence X-ray optical elements,like ellipsoidal mirrors and polycapillaries (treatedin a later chapter), and hence to deliver high bril-liance beams of small dimension to the sample.These high-brightness micro-focus sources havebeen used mostly in X-ray diffraction (XRD) sofar, but they are likely to have a major impacton XRS in the near future as well. Subchapter 2.2on new synchrotron radiation sources was writ-ten by M. Watanabe (Institute of MultidisciplinaryResearch for Advanced Materials, Tohoku Uni-versity, Japan) and G. Isoyama (Institute of Sci-entific and Industrial Research, Osaka University,Japan). In this subchapter, new synchrotron radi-ation sources are introduced and the characteris-tics of synchrotron radiation are summarised. Newaspects and typical properties of the synchrotronradiation flux at the sample position are describedfor users of third-generation sources and candi-dates for fourth-generation sources are discussed.In Subchapter 2.3, C. Spielman (PhysikalischesInstitut EP1, University of Würzburg, Germany)treats a novel generation of laser-driven X-raysources, which could produce femtosecond pulsesof soft to hard X-rays, synchronisable to otherevents, and very high intensities, from compactlaboratory X-ray sources. This section describesrecent progress in the development of laser sourcesrelevant for X-ray generation and reviews the gen-eration of laser-produced incoherent radiation, thedevelopment of X-ray lasers and high-harmonicgeneration. Applications of coherent laboratory X-ray sources are still in their infancy, but thesemight be intriguing in the future, in XRS, X-raymicroscopy, X-ray photoelectron spectroscopy andmaybe X-ray interferometry, all of which have hadto rely on large-scale synchrotron facilities thus far,and might open the way to attosecond science.

VOLUME OUTLINE 7

The third chapter is all about X-ray optics,another field that has seen an explosive growthin the last decade, in various ways, resulting innew commercial instruments and new applicationlines. In Subchapter 3.1, M. Yanagihara (Insti-tute of Multidisciplinary Research for AdvancedMaterials, Tohoku University, Japan) and K.Yamashita (Department of Physics, Nagoya Uni-versity, Japan) discuss advances in multilayer pro-duction technology, due to the progress in thin-filmtechnology and polishing of super-smooth sub-strates to the sub-nanometer level, and in theirperformance and applications. For soft X-rays, thelatter include focusing, microscopy and polarime-try; for hard X-rays, obtaining microbeams formicroscopy, X-ray telescopes and multilayer-coated gratings are discussed. In Subchapter 3.2,Y. Hosokawa (X-ray Precision, Inc., Kyoto, Japan)presents the state-of-the-art for single capillar-ies, which make use of multiple external totalreflections. He shows how a very bright andnarrow X-ray microbeam can be realised usingsingle capillaries (or X-ray guide tubes), andhow this leads to a tabletop X-ray analyticalmicroscope. Several applications are presented. InSubchapter 3.3, N. Gao (X-ray Optical Systems,Albany, NY, USA) and K. Janssens (Universityof Antwerp, Belgium) give a detailed treatment ofthe fundamentals of multi-fiber polycapillaries andthe recent fused and heat-shaped monolithic ver-sions. In the last decade, polycapillary optics havebecome widespread and successfully used as cru-cial components in commercial X-ray microanal-ysis and low-power compact instruments. Novelanalytical applications are situated in elementalmicroanalysis in laboratory scale, portable andsynchrotron systems, micro-X-ray absorption near-edge spectroscopy (XANES), EMPA, etc. Futuredevelopments in performance and spot size arediscussed. Finally, the new compound refractivelenses, first fabricated in 1996, are presented inSubchapter 3.4 by A. Simionovici (European Syn-chrotron Research Facility, Grenoble, France) andC. Schroer and B. Lengeler (Aachen University ofTechnology, Germany). Their theory, design andproperties are considered, as well as their use forimaging and microbeam production. Some focus

is on parabolic refractive X-ray lenses that canbe used in e.g. a new hard X-ray microscope thatallows sub-micrometer resolution and for e.g. com-bined fluorescence spectroscopy and tomography.Applications in the realms of biology, XRF com-puted micro-tomography, geochemistry and envi-ronmental research are given.

The most dramatic and spectacular progress hascertainly been made recently in the field of X-raydetector technology, and all this is covered inChapter 4. In Subchapter 4.1, L. Strüder, G. Lutz,P. Lechner, H. Soltau and P. Holl (Max-Planck-Institute for Physics and Extraterrestrial Physics,pnSensor and/or the Semiconductor Lab of theMax Planck Institute, Munich, Germany) treatadvances in silicon detectors. After an introduc-tion to the basic operation principles of semicon-ductors and the electronics used, some importantnew detectors are discussed in detail and importantapplications in XRS and imaging are reviewed.The detectors include Silicon Drift Detectors forX-ray detection, Controlled Drift Detectors (CDD),fully depleted backside illuminated pn-CCD andActive Pixel Sensors (APS) for XRS. All thesequite sophisticated detectors have left their ini-tial fields of applications in high-energy physics,astrophysics and SR research. They are now amature technology and open many new indus-trial applications. These detectors exhibit now ahigh quantum efficiency, excellent energy resolu-tion, high radiation tolerance, good position resolu-tion, high speed, homogeneous response of the fullbandwidth of radiation and high background rejec-tion efficiency. In Subchapter 4.2, C. A. N. Conde(Department of Physics, University of Coimbra,Portugal) treats the role of new gas proportionalscintillation counters (GPSC) for XRS, after con-sidering the physics of the absorption of X-rays ingases, the transport of electrons and the productionof electroluminescence in gases, and the basic con-cepts of different types of GPSC. Their energy res-olution is only 8 % for 5.9 keV X-rays, but they canbe built with very large windows and be useful forvery soft X-rays like the K-lines of C and O. Dif-ferent types of cryogenic detectors, operating nearthe liquid helium temperature (implying sophisti-cated cooling systems) and offering unseen energy


resolutions, are truly a major development ofrecent years. However, their commercial availabil-ity and price range is still somewhat unclear at themoment. Both superconducting tunneling junctions(STJ) and microcalorimeters are treated in detailin this volume. In Subchapter 4.3, M. Kurakado(Department of Electronics and Applied Physics,Osaka Electro-Communication University, Japan)explains the unique working principles of STJ,which usually consist of two superconductor lay-ers and a nanometer-thick insulator layer, which isa tunnel barrier between the superconductor layersthat can be passed by excited electrons or holes,i.e. quasiparticles, to give rise to a signal. Single-junction detectors and two other types of STJdetectors are discussed. Fantastic energy resolu-tions around 10 eV are possible. New applicationsare emerging, including one- and two-dimensionalimaging. Other equally promising cryogenic detec-tors are the cryogenic microcalorimeters, treatedin Subchapter 4.4 by M. Galeazzi (Department ofPhysics, University of Miami, Coral Gables, FL,USA) and E. Figueroa-Feliciano (NASA/GoddardSpace Flight Center, Greenbelt, MD, USA). Theidea of detecting the increase in temperature pro-duced by incident photons instead of the ionisa-tion of charged pairs, like in semiconductor detec-tors, was put forward almost 20 years ago, andthe operating principle is rather simple, but thepractical construction is quite challenging. Only inrecent years has the practical construction of ade-quate cryogenic microcalorimeters been realised.The required characteristics, parameters and non-ideal behavior of different components and types,including large arrays, detector multiplexing andposition-sensitive imaging detectors, are discussedin detail. Several expected future developmentsare outlined. In the last section of this chapteron detectors, W. Dabrowski and P. Gryboś (Fac-ulty of Physics and Nuclear Techniques, Univer-sity of Mining and Metallurgy, Krakow, Poland)treat position-sensitive semiconductor strip detec-tors, for which the manufacturing technologies andreadout electronics have matured recently. Sili-con strip detectors, of the same type as used fordetection of relativistic charged particles, can beapplied for the detection of low-energy X-rays,

up to 20 keV. Regardless of some drawbacks dueto limited efficiency, silicon strip detectors aremost widely used for low-energy X-rays. Single-sided, double-sided and edge-on silicon strip detec-tors and the associated electronics are treated ingreat detail.

There are many special configurations andinstrumental approaches in XRS, which have beenaround for a while or have recently been devel-oped. Eight of these are reviewed in Chapter 5.In Subchapter 5.1, K. Sakurai (National Insti-tute for Materials Science, Tsukuba, Japan) dealswith TXRF or grazing-incidence XRF (GI-XRF).Although TXRF may have been fading awaya bit recently for trace element analysis ofliquid or dissolved samples, there have stillbeen advances in combination with wavelength-dispersive spectrometers and for low atomic num-ber element determinations. But mostly, there haverecently been interesting developments in surfaceand interface analysis of layered materials by angu-lar and/or energy-resolved XRF measurements,and in their combination with X-ray reflectome-try. Micro-XRF imaging without scans is a recentinnovation in GI-XRF as well. Future develop-ments include e.g. combining GI-XRF with X-rayfree-electron laser sources. An approach that hasnot been used widely so far is grazing-exit XRS(GE-XRS), related in some ways to GI-XRF. GE-XRF is the subject of Subchapter 5.2, by K. Tsuji(Osaka City University, Japan). Since the X-rayemission from the sample is measured in GE-XRS,different types of excitation probes can be used,not only X-rays but also electrons and chargedparticles. In addition, the probes can be used toirradiate the sample at right angles. This subchapterdescribes the principles, methodological character-istics, GE-XRS instrumentation, and recent appli-cations of GE-XRF, as well as GE-EPMA andGE-PIXE. At the end of this subchapter, the futureof GE-XRS is discussed, which implies the useof more suitable detectors and synchrotron radi-ation excitation. One interesting aspect of XRFis the enormously increased recent (commercial)interest in portable EDXRF systems. This topicis treated in the next subchapter by R. Cesareoand A. Brunetti (Department of Mathematics and

VOLUME OUTLINE 9

Physics, University of Sassari, Italy), A. Castel-lano (Department of Materials Science, Universityof Lecce, Italy) and M.A. Rosales Medina (Uni-versity ‘Las Americas’, Puebla, Mexico). Only inthe last few years, has technological progress pro-duced miniature and dedicated X-ray tubes, ther-moelectrically cooled X-ray detectors of small sizeand weight, small size multichannel analysers anddedicated software, allowing the construction ofcompletely portable small size EDXRF systemsthat have similar capabilities as the more elabo-rate laboratory systems. Portable equipment maybe necessary when objects to be analysed cannotbe transported (typically works of art) or whenan area should be directly analysed (soil analysis,lead inspection testing, etc.) or when the mappingof the object would require too many samples.The advantages and limitations of different set-ups, including optics, are discussed. A focusedsubchapter on the important new technology ofmicroscopic XRF using SR radiation has been pro-duced by F. Adams, L. Vincze and B. Vekemans(Department of Chemistry, University of Antwerp,Belgium). It describes the actual status with respectto lateral resolution and achievable detection lim-its, for high-energy, third-generation storage rings(particularly the European Synchrotron RadiationFacility, Grenoble, France), previous generationsources and other sources of recent construction.Related methods of analysis based on absorptionedge phenomena such as X-ray absorption spec-troscopy (XAS), XANES, X-ray micro-computedtomography (MXCT) and XRD are briefly dis-cussed as well. Particular attention is paid to theaccuracy of the XRF analyses. Subchapter 5.5 by I.Nakai (Department of Applied Chemistry, ScienceUniversity of Tokyo, Japan) deals with high-energyXRF. It considers SR sources and laboratory equip-ment, in particular a commercial instrument forhigh-energy XRF that has only recently becomeavailable. The characteristics of the techniqueinclude improved detection limits, chiefly for highatomic number elements. This makes it particularlysuitable for the determination of e.g. rare earthsvia their X-lines. Other interesting applicationexamples pertain to environmental, archaeological,geochemical and forensic research. Low-energy

EMPA and scanning electron microscopy (SEM)are the topics of S. Kuypers (Flemish Institute forTechnological Research, Mol, Belgium). The fun-damental and practical possibilities and limitationsof using soft X-rays, as performed in the two sepa-rate instruments, are discussed. The potential of thetwo techniques is illustrated with recent examplesrelated to the development of ultra-light-elementbased coatings for sliding wear applications, mem-branes for ultrafiltration and packaging materialsfor meat. In Subchapter 5.7, E. Van Cappellen (FEICompany, Hillsboro, OR, USA) treats ED X-raymicroanalysis in transmission electron microscopy(TEM), for both the scanning and conventionalmode. The section describes how EDXRS in the(S)TEM can be made quantitative, accurate andprecise, and is nowadays an extremely powerfultechnique in materials science and has not van-ished in favor of electron energy loss spectrom-etry (EELS) as predicted 20 years ago. Severalexamples are given of quantitative chemical map-ping, quantitative analysis of ionic compounds andother real-world applications. Finally, J. Kawai(Department of Materials Science and Engineering,Kyoto University, Japan) discusses in detail theadvances in XAS or X-ray absorption fine struc-ture spectroscopy (XAFS), which include XANES(X-ray absorption near edge structure) and EXAFS(extended X-ray absorption fine structure). X-rayabsorption techniques are now used in commer-cially available film thickness process monitorsfor plating, printed circuit and magnetic disk pro-cesses, in various kinds of industries. But they areused, both in laboratories and synchrotron facili-ties, for basic science as well. The X-ray absorp-tion techniques, described extensively in this sub-chapter, differ in probe type (electrons and X-rays,sometimes polarised or totally reflected), detectedsignals (transmitted X-rays, XRF, electrons, elec-tric currents, and many others) and applicationfields (high temperature, high pressure, low tem-perature, in situ chemical reaction, strong magneticfield, applying an electric potential, short measure-ment time, and plasma states). One shortcoming ofXAS techniques, that absorption spectra of all theelements were not measurable using one beamline,


has been overcome in many synchrotron facili-ties nowadays.

Chapter 6 reviews some advances in computer-isation concerning XRS. The first subchapter, writ-ten by L. Vincze, K. Janssens, B. Vekemans andF. Adams (Department of Chemistry, Universityof Antwerp, Belgium) deals with modern MonteCarlo (MC) simulation as an aid for EDXRF.The use of MC simulation models is becomingmore and more viable due to the rapid increaseof inexpensive computing power and the avail-ability of accurate atomic data for photon-matterinteractions. An MC simulation of the completeresponse of an EDXRF spectrometer is interest-ing from various points of view. A significantadvantage of the MC simulation based quantifi-cation scheme compared to other methods, such asFP algorithms, is that the simulated spectrum canbe compared directly to the experimental data inits entirety, taking into account not only the flu-orescence line intensities, but also the scatteredbackground of the XRF spectra. This is linkedwith the fact that MC simulations are not lim-ited to first- or second-order approximations andto ideal geometries. Moreover, by considering thethree most important interaction types in the 1–100keV energy range (photoelectric effect followedby fluorescence emission, Compton and Rayleighscattering), such models can be used in a generalfashion to predict the achievable analytical char-acteristics of e.g. future (SR)XRF spectrometersand to aid the optimisation/calibration of existinginstruments. The code illustrated in this subchapterhas experimentally been verified by comparisonsof simulated and experimental spectral distribu-tions of various samples. With respect to the sim-ulation of heterogeneous samples, an example isgiven for the modeling of XRF tomography experi-ments. The simulation of such lengthy XRF imag-ing experiments is important for performing fea-sibility studies and optimisation before the actualmeasurement is performed. Subchapter 6.2 by P.Lemberghe (Department of Chemistry, Universityof Antwerp, Belgium) describes progress in spec-trum evaluation for EDXRF, where it remainsa crucial step, as important as sample prepara-tion and quantification. Because of the increased

count rate and hence better precision due to newdetectors, more details became apparent in thespectra; fortunately, the availability of inexpensiveand powerful PCs now enables the implementationof mature spectrum evaluation packages. In thissubchapter, the discussed mathematical techniquesgo from simple net peak area determinations, to themore robust least-squares fitting using referencespectra and to least-squares fitting using analyticalfunctions. The use of linear, exponential or orthog-onal polynomials for the continuum fitting, and of amodified Gaussian and Voigtian for the peak fittingis discussed. Most attention is paid to partial least-squares regression, and some illustrative analyticalexamples are presented.

The final chapter, Chapter 7, deals with fivegrowing application fields of XRS. J. Börjesson(Lund University, Malmö and the Department ofDiagnostic Radiology, County Hospital, Halmstad,Sweden) and S. Mattsson (Department of Diagnos-tic Radiology, County Hospital, Halmstad Sweden)focus on applications in the medical sciences since1995, i.e. on recent advances in in vivo XRF meth-ods and their applications, and on examples of invitro use of the technique. The latter deals mostlywith the determination of heavy metals in tis-sues, in well-established ways. But there have beensignificant developments lately in in vivo analysiswith respect to sources, geometry, use of polarisedexciting radiation, MC simulations and calibra-tion, and the analytical characteristics have beenimproved. Examples of novel in vivo determina-tions of Pb, Cd, Hg, Fe, I, Pt, Au and U arediscussed. The next subchapter deals with novelapplications for semiconductors, thin films and sur-faces, and is authored by Y. Mori (Wacker-NSCECorp., Hikari, Japan). Progress in the industrialapplication of TXRF in this field is first discussed.The use of TXRF for semiconductor analysis cameinto popular use in the 1990s; today, more than 300TXRF spectrometers are installed in this industryworldwide, meaning that almost all leading-edgesemiconductor factories have introduced TXRF.Since the main purpose of TXRF is trace contam-ination analysis, improvements in the elementalrange (including light elements), detection abil-ity (e.g. by preconcentration) and standardisation

REFERENCE 11

(versus other techniques) are discussed. In addi-tion, XRF and X-ray reflectivity analysers forthe characterisation of thin films made from newmaterials are introduced. A. Zucchiatti (IstitutoNazionale di Fisica Nucleare, Genova, Italy) wrotethe next subchapter on the important applicationof XRS in archaeometry, covering instrumenta-tion from portable units through PIXE and syn-chrotrons. Applications of the latter techniquesinclude e.g. the study of Renaissance glazed terra-cotta sculptures, flint tools and Egyptian cosmetics.Also the XRF and XANES micro-mapping of cor-roded glasses is described. Radiation damage isa constant major concern in this field. The muchlarger availability of facilities and several techno-logical advances have made archaeometry a verydynamic field for XRS today and an even greaterresearch opportunity for tomorrow. T. Ninomiya(Forensic Science Laboratory, Hyogo PrefecturalPolice Headquarters, Kobe, Japan) illustrates somerecent forensic applications of TXRF and SR-XRF.Trace element analysis by TXRF is used to fin-gerprint poisoned food, liquor at crime scenes,counterfeit materials, seal inks and drugs. Foren-sic applications of SR-XRF include identifica-tion of fluorescent compounds sometimes usedin Japan to trace criminals, different kinds ofdrugs, paint chips and gunshot residues. Sub-chapter 7.5 deals with developments in electron-induced XRS that have mainly an impact on envi-ronmental research, namely the speciation andsurface analysis of individual particles, and is writ-ten by I. Szaloki (Physics Department, Univer-sity of Debrecen, Hungary), C.-U. Ro (Department

of Chemistry, Hallym University, Chun Cheon,Korea), J. Osan (Atomic Energy Research Institute,Budapest, Hungary) and J. De Hoog (Departmentof Chemistry, University of Antwerp, Belgium).In e.g. atmospheric aerosols, it is of interest toknow the major elements that occur together inone particle, i.e. to carry out chemical speciationat the single particle level. These major elementsare often of low atomic number, like C, N andO. In EDXRS, ultrathin-window solid-state detec-tors can measure these elements but for such softX-rays, matrix effects are enormous and quantifica-tion becomes a problem. Therefore an inverse MCmethod has been developed which can determinelow atomic number elements with an unexpectedaccuracy. To reduce beam damage and volatili-sation of some environmental particles, the useof liquid-nitrogen cooling of the sample stage inthe electron microprobe has been studied. Finally,irradiations with different electron beam energies,i.e. with different penetration power, have beenapplied, in combination with the MC simulation,to study the surface and core of individual parti-cles separately and perform some depth profiling.The given examples pertain to water-insoluble ele-ments in/on individual so-called Asian dust aerosolparticles, nitrate enrichments in/on marine aerosolsand sediment particles from a contaminated river.

REFERENCE

J. Injuk and R. Van Grieken, X-Ray Spectrometry, 32 (2003)35–39.

Chapter 2

X-Ray Sources

2.1 Micro X-ray Sources

M. TAYLOR, R. BYTHEWAY and B. K. TANNERBede plc, Durham, UK

2.1.1 INTRODUCTION

A little over a century ago, X-rays were discoveredby Wilhelm Conrad Röntgen (Röntgen, 1995) asa result of the impact of a beam of electrons,accelerated through an electrostatic field, on ametallic target. Current commercial X-ray tubeswork on the self-same principle, heated cathodeshaving replaced the original cold cathodes andwater cooling enabling much higher power loadsto be sustained. Electrostatic focusing of theelectron beam is used in almost all tubes andof the incremental improvements in sealed tubeperformance over the past 50 years, the recentdevelopment of ceramic tubes (e.g. Bohler andStehle, 1998) by Philips is the only one of note.

A discrete step in performance occurred inthe late 1950s with the development (Daviesand Hukins, 1984; Furnas, 1990) of the rotatinganode generator. Through rapid rotation (severalthousand revolutions per minute) of the target,the heat load and hence X-ray emission, couldbe increased as the heated region is allowed tocool in the period when away from the electronbeam. Rotating anode generators are manufacturedby a number of companies and provide the highestoverall power output of any electron impact device.

It was recognized many years ago that forsome applications, in particular those involvingimaging, that a small source size was desirable.In the 1960s Hilger and Watts developed a de-mountable, continuously pumped X-ray tube thatfound use, for example, in X-ray diffractiontopography (Bowen and Tanner, 1998) of singlecrystals. One of the electron optical configurationsfor this generator gave a microfocus source butwith the demise of the company and the adventof synchrotron radiation, use of such very smallsources was unusual.

The limitation of electron impact sources liesprincipally in the ability to conduct heat away fromthe region of electron impact, hence limiting thepower density on the target. Heat flow in solidsis governed by the heat diffusion equation, firstderived by Fourier in 1822. This describes thetemperature T at any point x, y, z in the solidand at time t. Assuming that there is a heat sourcedescribed by the function f (x, y, z, t) we have

∂T /∂t = α2∇2T + f/cρ (2.1.1)where α = (K/cρ)1/2 and K is the thermal conduc-tivity, c is the heat capacity and ρ is the density.Thus under steady-state conditions, the key param-eter in determining the temperature distribution

X-Ray Spectrometry: Recent Technological Advances. Edited by K. Tsuji, J. Injuk and R. Van Grieken 2004 John Wiley & Sons, Ltd ISBN: 0-471-48640-X

14 MICRO X-RAY SOURCES

Table 2.1.1 Physical properties of target materials

Material Melting temperature,Tm (

◦C)Thermal conductivity,

K (W cm−1K−1)Heat capacity,c (J g−1 K−1)

Density,ρ (g cm−3)

Diffusivity,α (cm s−1/2)

Cu 1084 4.01 0.38 8.93 1.09Al 660 2.37 0.90 2.7 0.99Mo 2623 1.38 0.25 10.22 0.74W 3422 1.73 0.13 19.3 0.83Diamond(Type IIa)

3500 23.2 0.51 3.52 3.60

is the thermal conductivity. However, in transientconditions, it is the parameter α, often referred toas the diffusivity, that is the important in deter-mining the maximum temperature at any point.Clearly, to avoid target damage, the maximum tem-perature T must be significantly below the meltingpoint of the target material.

Reference to Table 2.1.1 shows that the choiceof copper as the anode material is governed bymore than its ease of working and relatively lowcost. In rotating anode generators, even when theactual target material is tungsten or molybdenum,these materials are plated or brazed onto a cop-per base. Calculations performed many years agoby Müller (Müller, 1931) and Oosterkamp (Oost-erkamp, 1948) showed that the maximum permis-sible power on a target was proportional to thediameter of the focal spot on the target. Relativelylittle is gained from such strategies as makingturbulent the flow of coolant on the rear surfaceof the target. As the power of the X-ray tube isincreased, there must be a corresponding increasein focal spot size and inspection of manufacturers’specifications will readily attest to this fundamen-tal limitation. Synchrotron radiation sources, wherethere is no such problem of heat conduction, haveproved the route past this obstacle.

2.1.2 INTER-RELATIONSHIPBETWEEN SOURCE AND OPTICS

Nevertheless, there are many applications where itis either impossible or impractical to travel to asynchrotron radiation source and thus, driven bythe spectacular developments at the synchrotronradiation sources, there has been strong pressureto improve laboratory-based sources. As it is clear

from the above impasse that increase in the rawpower output was not the solution, attention hasfocused on the exploitation of X-ray optics. Itwas realized that there was a prodigious waste ofX-ray photons associated with standard collimationtechniques and that the scientific community hadprogressed no further than the pinhole camera.(Of the photons emitted into 2π solid angle, acollimator diameter 1 mm placed 10 cm from thesource accepts only 8 × 10−5 steradians. Only0.0013 % of the photons are used.)

The huge developments in X-ray optics overthe past decade are described elsewhere in thisvolume. In this subchapter we confine ourselves todiscussion of only two optical elements, ellipsoidalmirrors and polycapillary optics. These devices aremirrors that rely on the total external reflectionof X-rays at very low incidence angles andthe figuring of the optic surfaces to achievefocusing. However, as is evident from Figure 2.1.1,

0

0.2

0.4

0.6

0.8

1.0

0 0.2 0.4 0.6 0.8 1.0

0 Å r.m.s.5 Å r.m.s.10 Å r.m.s.15 Å r.m.s.20 Å r.m.s.

1.54 Å wavelength

Incidence angle (degrees)

Ref

lect

ivity

Figure 2.1.1 Reflectivity of a gold surface as a function ofincidence angle and surface roughness (r.m.s. = root meansquare of the amplitude of surface displacement)

A MICROFOCUS GENERATOR WITH MAGNETIC FOCUSING 15

because the refractive index for materials in theX-ray region of the spectrum is only smallerthan unity by a few parts in 105, the rangeof total external reflection is very limited. As aconsequence of this grazing incidence limitationon total external reflection, to maximize the photoncollection, the optic needs to be placed very closeto the X-ray source. [Although parabolic multilayermirrors have been developed (Schuster and Göbel,1995; Gutman and Verman, 1996; Stommer et al.,1997) that significantly enhance the flux deliveredfrom rotating anode generators, the gains arerelatively modest due to the large source sizeand large source to optic distance. Nevertheless,the combination of high power and insertion gainresults in such devices delivering a huge intensityat the specimen.]

A small optic to source distance has animmediate consequence in that, if the beamdivergence (or crossfire) at the sample is to besmall, the optic to sample distance must be large.The magnification of the source is therefore highand to maintain both a small beam size at thesample and low aberrations, the source size mustbe very small. Thus, to achieve a high insertiongain from a grazing incidence optic, it is essentialthat a microfocus X-ray source be used.

2.1.3 A MICROFOCUS GENERATORWITH MAGNETIC FOCUSING

Despite the widespread use of X-ray fluorescenceanalysis from extremely small electron beam spotsin scanning electron microscopes, until recentlyall commercial X-ray generators used exclusivelyelectrostatic focusing. This is despite the factthat electron microscope manufacturers long agorealized that magnetic focusing was superiorin many ways. Electrostatically focused tubesgenerally exhibit side lobes to the electron beamspot and are relatively inefficient at deliveringelectrons to the target itself.

The first electromagnetically focused microfo-cus tube was described by Arndt, Long and Dun-cumb (Arndt et al., 1998a) in 1998. The designmaximized the solid angle of collection of the

emitted X-rays and thus, in association with anellipsoidal mirror, achieved a high intensity at thesample. The observed intensity was in reasonableagreement with that calculated and compared withthat achieved with non-focusing X-ray optics usedwith conventional X-ray tubes operated at a powermore than 100 times as great.

In the patented design (US Patent No. 6282263,2001), the electron beam, of circular cross-section,from the gun is focused by an axial magnetic lensand then drawn out by a quadrupole lens to forman elongated spot on the target. When viewed ata small take-off angle an elongated focus is seen,foreshortened to a diameter between 10 and 20 µm.Within the X-ray generator (Figure 2.1.2) the tubeis sealed and interchangeable. The electron opticsenable the beam to be steered and focused intoeither a spot or a line with a length to width ratioof 20:1. An electron mask of tungsten is includedto form an internal electron aperture. The electrongun consists of a Wehnelt electrode and cathode thatcan be either a rhenium–tungsten hairpin filamentor an indirectly heated activated dispenser cathode.The advantages of the dispenser cathode is that it ismechanically stable and, due to the lower powerconsumption and operating temperature, it has agreater lifetime than heated filament cathodes. It isalso simpler to align in the Wehnelt electrode. Thetube is run in a space-charge limited condition (asopposed to the conventional saturated, temperaturelimited condition), with the filament maintained at aconstant temperature. As a result, the tube current isdetermined almost exclusively by the bias voltagebetween the filament and the Wehnelt electrode.The electrons are accelerated from the cathode, heldat a high negative potential, towards the groundedanode. They pass through a hole in the anode beforeentering a long cylinder and subsequently collidingwith the target. An electron cross-over is formedbetween the Wehnelt and anode apertures and thisis imaged onto the target by the iron-cored axialsolenoid. So far Cu, Mo and Rh target tubes havebeen run successfully. The power loading that canbe achieved is such that a small amount of water-cooling proves essential.

The ability to control the electron beam spotsize and shape by adjustment of the current in

16 MICRO X-RAY SOURCES

H

G

F

E

D

C

B

A

Figure 2.1.2 Schematic drawing of an electromagneticallyfocused microfocus X-ray tube. (A) Cathode, (B) Wehneltgrid electrode, (C) anode, (D) electromagnetic axial focusinglens, (E) electromagnetic quadrupole lens, (F) target, (G) X-rayshutter, (H) direction of X-ray beam

the quadrupole stigmator coils is critical to theoptimum performance of the microfocus tube.Figure 2.1.3 shows the variation of the sourcedimensions as a function of the current in thestigmator coils. We note that there is a significantrange, close to the minimum in source area, whichis almost independent of the stigmator current andwhere the source is approximately equiaxed.

As the tube accelerating voltage is increased,the value of the current in the focusing coils

0

20

40

60

80

−100 −50 0 50 100

Vertical dimensionHorizontal dimension

% of maximum stigmator current

Sou

rce

dim

ensi

on (

µm)

Figure 2.1.3 Source dimension as a function of the maximumcurrent through the stigmator coils

needs to be increased to achieve the minimum spotsize. Increasing the tube current usually increasesthe spot size, due to space charge effects. Itis our standard practice to set up the focusingcoils with the stigmator coil current chosen toachieve an equiaxed beam. However, by adjustingthe stigmator coil current to draw the source outinto a line, a significant gain in intensity can beobtained without compromising on the couplinginto subsequent X-ray optical elements. The outputis limited by the maximum power at which targetdamage does not occur and Figure 2.1.4 shows

0.5

0.7

0.9

1.1

1.3

1.5

30 40 50Tube voltage (kV)

Pow

er/[s

pot a

rea]

½ (

W µ

m−1

)

Figure 2.1.4 Power divided by the square root of the spot areaunder conditions for achieving maximum output intensity atthree settings of tube voltage. The line is a least-squares linearfit to the data

x-ray spectrometry: recent technological...

Documents